D at a S h eet , D S 2, Feb. 20 0 1
C165UTAH
Embedded C166 with USB,
IOM-2 and HDLC Support
Version 1.3
Wired
Communications
N e v e r
s t o p
t h i n k i n g .
Edition 2001-02-23
Published by Infineon Technologies AG,
St.-Martin-Strasse 53,
D-81541 München, Germany
© Infineon Technologies AG 2001.
All Rights Reserved.
Attention please!
The information herein is given to describe certain components and shall not be considered as warranted
characteristics.
Terms of delivery and rights to technical change reserved.
We hereby disclaim any and all warranties, including but not limited to warranties of non-infringement, regarding
circuits, descriptions and charts stated herein.
Infineon Technologies is an approved CECC manufacturer.
Information
For further information on technology, delivery terms and conditions and prices please contact your nearest
Infineon Technologies Office in Germany or our Infineon Technologies Representatives worldwide (see address
list).
Warnings
Due to technical requirements components may contain dangerous substances. For information on the types in
question please contact your nearest Infineon Technologies Office.
Infineon Technologies Components may only be used in life-support devices or systems with the express written
approval of Infineon Technologies, if a failure of such components can reasonably be expected to cause the failure
of that life-support device or system, or to affect the safety or effectiveness of that device or system. Life support
devices or systems are intended to be implanted in the human body, or to support and/or maintain and sustain
and/or protect human life. If they fail, it is reasonable to assume that the health of the user or other persons may
be endangered.
D at a S h eet , D S 2, Feb. 20 0 1
C165UTAH
Embedded C166 with USB,
IOM-2 and HDLC Support
Version 1.3
Wired
Communications
N e v e r
s t o p
t h i n k i n g .
C165UTAH
Revision History:
Previous Version:
Page
2001-02-23
DS 2
Data Sheet, 10.00, DS11)
Subjects (major changes since last revision)
91
Correction of the PEC Control Register:
The correct channel numbers are PEC channels 0 and 2.
The name of the corresponding register is PECXC2 and not PECXC1.
330
Correction: AS2 in was removed
529
Correction of the Parameter t14.
Improved formatting (text, figures, tables)
1)
All previous distributed versions are preliminary. They have been replaced by this version.
For questions on technology, delivery and prices please contact the Infineon
Technologies Offices in Germany or the Infineon Technologies Companies and
Representatives worldwide: see our webpage at http://www.infineon.com
C165UTAH
Table of Contents
Page
1
1.1
1.2
1.3
1.4
1.4.1
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Key Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Logic Symbol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Pinning Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Typical Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ISDN NT and PBX Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2
2.1
2.2
Pin Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
C165UTAH Pin Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
C165UTAH Pin Definitions and Functions . . . . . . . . . . . . . . . . . . . . . . . . . 19
3
3.1
3.2
3.3
3.4
3.5
Architectural Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Basic CPU Concepts and Optimizations . . . . . . . . . . . . . . . . . . . . . . . . . .
On-Chip System Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Clock Generation Concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
On-Chip Peripheral Blocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Protected Bits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
28
29
34
36
41
45
4
4.1
4.2
4.3
Memory Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Internal RAM and SFR Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
External Memory Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Crossing Memory Boundaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
46
48
53
54
5
5.1
5.2
5.3
5.4
5.5
Central Processor Unit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Instruction Pipelining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Bit-Handling and Bit-Protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Instruction State Times . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CPU Special Function Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
PEC - Extension of Functionality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
55
57
63
64
65
85
6
6.1
6.2
6.3
6.4
6.5
DMA - External PEC (EPEC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
EPEC Functionality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
EPEC Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
EPEC Register Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
EPEC Transfer Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
Implementation of EPEC Interrupt Generation Unit . . . . . . . . . . . . . . . . . 101
7
7.1
7.2
7.3
7.4
7.5
7.6
7.7
Interrupt and Trap Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Interrupt System Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Interrupt Control Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Operation of the PEC Channels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Prioritization of Interrupt and PEC Service Requests . . . . . . . . . . . . . . .
Saving the Status during Interrupt Service . . . . . . . . . . . . . . . . . . . . . . . .
Interrupt Response Times . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
PEC Response Times . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Data Sheet
11
12
15
16
17
17
102
103
108
113
115
118
119
122
2001-02-23
C165UTAH
Table of Contents
Page
7.8
7.8.1
7.8.2
7.8.3
7.8.4
7.9
External Interrupts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .124
Fast External Interrupts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .125
External Interrupt Source Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . .126
Interrupt Subnode Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .127
The Interrupt Control Register . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .128
Trap Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .129
8
8.1
8.1.1
8.2
8.2.1
8.3
8.3.1
8.4
8.4.1
8.5
8.5.1
8.6
8.6.1
8.7
Parallel Ports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .134
PORT0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .138
Alternate Functions of PORT0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .142
PORT1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .144
Alternate Functions of PORT1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .149
PORT2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .150
Alternate Functions of PORT2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .153
PORT3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .155
Alternate Functions of PORT3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .157
PORT4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .161
Alternate Functions of PORT4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .164
PORT6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .165
Alternate Functions of PORT6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .168
PORT7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .171
9
Dedicated Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .174
10
10.1
10.2
10.3
10.4
10.5
10.6
10.7
10.8
External Bus Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .177
External Bus Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
Programmable Bus Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .187
READY Controlled Bus Cycles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .192
Controlling the External Bus Controller . . . . . . . . . . . . . . . . . . . . . . . . . . .193
EBC Idle State . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .205
External Bus Arbitration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .206
XBUS Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .209
Initialization of the C165UTAH’s X-peripherals . . . . . . . . . . . . . . . . . . . . . 210
11
11.1
11.1.1
11.1.2
11.1.3
General Purpose Timer Unit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .212
Kernel Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .213
Functional Description of Timer Block 1 . . . . . . . . . . . . . . . . . . . . . . . .213
Functional Description of Timer Block 2 . . . . . . . . . . . . . . . . . . . . . . . .230
GPT Register Set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
12
12.1
12.1.1
12.1.2
12.1.3
Asynchronous/Synchr. Serial Interface . . . . . . . . . . . . . . . . . . . . . . . .250
Functional Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .250
Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .250
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .252
Register Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .253
Data Sheet
2001-02-23
C165UTAH
Table of Contents
Page
12.1.4
12.1.5
12.1.6
12.1.7
12.1.8
12.1.9
12.1.10
General Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Asynchronous Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Synchronous Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Baudrate Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Autobaud Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Hardware Error Detection Capabilities . . . . . . . . . . . . . . . . . . . . . . . .
Interrupts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
264
265
273
275
281
287
288
13
13.1
13.1.1
13.1.2
13.2
13.2.1
13.2.2
13.2.3
13.2.4
13.2.5
13.2.6
13.2.7
13.2.8
13.2.9
13.2.10
13.2.11
13.2.12
Real Time Clock (RTC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Function Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
RTC Block Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
RTC Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
System Clock Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Cyclic Interrupt Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Alarm Interrupt Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
48-bit Timer Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Defining the RTC Time Base . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Increased RTC Accuracy through Software Correction . . . . . . . . . . . .
Hardware dependend RTC Accuracy . . . . . . . . . . . . . . . . . . . . . . . . .
Interrupt Sub Node RTCISNC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
RTC Disable Functionality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Register Definition of RTC module . . . . . . . . . . . . . . . . . . . . . . . . . . . .
290
290
290
290
290
291
292
292
292
292
293
293
295
295
295
296
297
14
14.1
14.2
14.3
14.4
14.5
High-Speed Synchronous Serial Interface . . . . . . . . . . . . . . . . . . . . .
Full-Duplex Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Half Duplex Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Baud Rate Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Error Detection Mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SSC Interrupt Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
302
307
310
312
313
315
15
15.1
15.2
15.3
15.4
15.5
15.6
15.7
15.7.1
15.7.2
15.7.3
USB Interface Controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
USB Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
USB Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
USB Endpoints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
USB Interface Controller (USBD) Architecture . . . . . . . . . . . . . . . . . . . . .
Endpoint Info Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
USB Microprocessor Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Programmers Guidlines: Using USB and EPEC . . . . . . . . . . . . . . . . . . .
Writing the configuration-value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
In-Transfer (Transmit) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Out-Transfer (Receive) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
318
318
318
319
328
328
332
342
342
343
343
Data Sheet
2001-02-23
C165UTAH
Table of Contents
15.7.4
15.7.5
15.7.6
15.7.7
15.7.8
15.7.9
15.7.10
16
16.1
16.1.1
16.1.2
16.1.3
16.1.4
16.2
16.3
16.3.1
16.3.2
16.3.3
16.4
16.4.1
16.4.2
16.4.3
16.4.4
16.5
16.5.1
16.5.2
16.5.3
16.5.4
16.6
16.6.1
16.6.2
16.6.3
16.6.4
16.7
16.7.1
16.7.2
16.7.3
16.7.4
16.7.5
16.7.6
16.8
16.8.1
Data Sheet
Page
Reading out Setup-Packets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .344
Special case: Setup-Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .344
Setting of configuration and alternate settings of interfaces . . . . . . . . .345
Stalling Endpoints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345
Start of Frame . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .345
Suspend and Suspendoff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .346
Device disconnecting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .346
IOM-2 Interface Controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .347
IOM-2 and PCM Frame Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .347
PCM Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .348
Terminal Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .348
Linecard Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .350
IOM-2 Handler . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .350
IOM-2 Monitor Handler . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .352
Handshake Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .352
Abort Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .356
MONITOR Interrupt Logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .358
C/I Channel Handler . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 359
C/I0 - Command/Indication 0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .359
C/I1 - Command/Indication 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .359
CIC Interrupt Logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .360
D-Channel Access Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .360
Controller Data Access Handler . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .362
Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .363
Looping and Shifting Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .364
Monitoring Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .366
Synchronous Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .367
Bus Activation / Deactivation (Power Down) . . . . . . . . . . . . . . . . . . . . . . .370
Deactivation Request, Downstream (C165UTAH) to Upstream . . . . . .371
Deactivation, Upstream to Downstream (C165UTAH) . . . . . . . . . . . . .372
Activation Request, Downstream (C165UTAH) to Upstream . . . . . . . .372
Activation, Upstream to Downstream (C165UTAH) . . . . . . . . . . . . . . . 372
HDLC Controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .373
Message Transfer Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .374
Data Reception . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .375
Data Transmission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .378
General Access to IOM-2 Channels . . . . . . . . . . . . . . . . . . . . . . . . . . .381
Extended Transparent Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .381
HDLC Controller Interrupts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .382
IOM-2/HDLC Controller Register Set . . . . . . . . . . . . . . . . . . . . . . . . . . . .384
Register Description Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .384
2001-02-23
C165UTAH
Table of Contents
16.8.2
16.8.3
16.8.4
Page
Register Table ordered by Address . . . . . . . . . . . . . . . . . . . . . . . . . . . 385
Detailed Register Description ordered by Address . . . . . . . . . . . . . . . 390
HDLC-Channel Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413
17
17.1
Watchdog Timer (WDT) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 430
Operation of the Watchdog Timer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 431
18
Bootstrap Loader . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 434
19
19.1
System Reset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 439
System Startup Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 446
20
20.1
20.2
20.3
20.4
20.4.1
Power Reduction Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Idle Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Power Down Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Status of Output Pins during Idle and Power Down Mode . . . . . . . . . . . .
Extended Power Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Sleep Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
451
451
453
453
455
456
21
21.1
21.2
21.2.1
21.3
21.4
21.4.1
21.4.2
21.5
21.6
21.6.1
21.6.2
System Control Unit (CSCU) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Operational Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Overview of CSCU submodules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
XBUS Peripheral Configuration Block . . . . . . . . . . . . . . . . . . . . . . . . . . .
System Control Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Register Write Protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Clock Output Frequency Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Peripheral Management Module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Identification Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ID Register Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
460
460
460
460
462
463
463
466
470
471
471
472
22
22.1
22.2
22.3
22.4
22.5
22.6
22.7
22.8
22.9
22.10
System Programming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Stack Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Register Banking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Procedure Call Entry and Exit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Table Searching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Peripheral Control and Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Floating Point Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Trap/Interrupt Entry and Exit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Unseparable Instruction Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . .
Overriding the DPP Addressing Mechanism . . . . . . . . . . . . . . . . . . . . .
Pits, Traps and Mines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
474
476
481
482
484
485
485
485
486
486
488
23
23.1
Register Set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 489
Register Description Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 489
Data Sheet
2001-02-23
C165UTAH
Table of Contents
Page
23.2
23.3
23.4
23.5
CPU General Purpose Registers (GPRs) . . . . . . . . . . . . . . . . . . . . . . . . .490
Special Function Registers ordered by Address . . . . . . . . . . . . . . . . . . .491
Special Function Registers ordered by Name . . . . . . . . . . . . . . . . . . . . . .499
Special Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .507
24
Instruction Set Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 508
25
25.1
25.2
25.3
25.4
25.5
25.6
25.7
25.7.1
25.7.2
25.7.3
25.7.4
25.7.5
25.8
25.8.1
AC/DC Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .512
Absolute Maximum Ratings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .512
Recommended Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . .512
DC Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .512
USB Full-speed (12 Mbit/s) Driver Characteristics . . . . . . . . . . . . . . . . . .514
Failsafe operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .514
Testing Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .515
AC Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .515
Definition of Internal Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .516
System Reset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .517
External Clock Drive XTAL1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .518
IOM-2 Interface Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .519
JTAG Interface Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .521
Asynchronous Bus Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .522
Memory Cycle Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 522
26
Package Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .538
Data Sheet
2001-02-23
Embedded C166 with USB, IOM-2 and HDLC Support
C165UTAH
Version 1.3
1
C165UTAH
CMOS
Overview
The C165UTAH is a new low cost member of the
Infineon Communication Controller family using low
power CMOS technology. The device combines the
successful Infineon C166 16-bit full-static core with a
full-speed Universal Serial Bus (USB) interface, four
independent HDLC controllers, IOM-2 interface and
3-kbyte of Dual-Port on-chip RAM to a USB type
Intelligent Terminal Adapter with HDLC support.
P-TQFP-144
The C165UTAH adresses all high feature ISDN TA, Intelligent NT or SOHO PBX
designs, offering up to 18 MIPS along with legacy peripherals such as USART, SCI and
Timers.
The USB device core has a built-in DMA, which provides maximum flexibility and
performance. Off-loading the CPU in such a manner allows the user to implement value
add software features enabling product differentiation.
The C165UTAH provides:
• On-Chip full-static C166 Core supporting a 16- or 8-bit C16x Family System running
up to 36 MHz
• ISDN BRI supporting data rates of 56 kbit/s, 64 kbit/s, 128 kbit/s and 144 kbit/s
• 12 Mbit/s Full-Speed USB Interface Vers. 1.1 compliant
– Support for Audio, Data and Communication Device Classes
• IOM-2/PCM Interface
– Terminal Mode Type Interface to CODEC and S/U Transceiver
– Linecard Mode Type Interface up to 8 IOM-2 channels or 32 PCM channels
– 1536/786 kHz and 1536 kHz...4096 kHz in 512 kHz steps
– Access to two Intercommunication channels (IC1, IC2)
– Access to two MON channels (MON0, MON1)
– Access to two C/I channels (CI0, CI1)
Type
Package
C165UTAH
P-TQFP-144
Data Sheet
11
2001-02-23
C165UTAH
Overview
– S/G access support
• Four On-Chip Independent Full-Duplex HDLC Formatters
– 8 independent 8-byte FIFOs for each transmit and receive channel
• USART Interface with AutoBaud Support (1,200 bit/s - 230,400 bit/s)
– AT-Command sensitive AutoBaud Detection
1.1
Key Features
The C165UTAH is a new low-cost member of the Infineon Communication Controller
family. The device has the following features:
• C166 Static Core with Peripherals including:
– Full-static core up to 18 MIPS (@36 MHz)
– Peripheral Event Controller (PEC) for 8 independent DMA channels
– 16 Dynamically Programmable Priority-Level Interrupt System
– Eight External Interrupts
– Up to 72 SW-configurative Input/Output (I/O) Ports, some with Interrupt Capabilities
– 8-bit or 16-bit External Data Bus
– Multiplexed or Demultiplexed Address/Data Bus
– Up to 8-Mbyte Linear Address Space for Code and Data
– Five Programmable Chip-Select Lines with Wait-State Generator Logic
– On-Chip 3,072-Byte Dual-Port SRAM for user applications
– On-Chip 1,024-Byte Special Function Register Area
– On-Chip PLL with Output Clock Signal
– Five Multimode General Purpose Timers
– On-Chip Programmable Watchdog Timer
– Glueless Interface to EPROM, Flash EPROM and SRAM
– Low-Power Management Supporting Idle-, Power-Down- and Sleep-Mode and
additional CPU clock slow-down mode with mode control for each peripheral
– USART interface with Auto Baud Rate detection up to 230,400 kbit/s
– USART Baud Rate generation in asynchronous mode up to 2.25 MBaud @ 36 MHz
– USART Baud Rate generation in synchronous mode up to 4.5 MBaud @ 36 MHz
– USART standard Baud Rates generation with very small deviation (230.4 kBaud
< 0.01%, 460.8 kBaud < 0.15 %, 691.2 kBaud < 0.04 %, 921.6 kBaud < 0.15 % ) @
36 MHz
– High speed Serial Synchronous Channel Interface (SSC) with ALIS-3.0 and AC97
compatibility up to 18 MBaud in SSC Master Mode and up to 9 MBaud in SSC Slave
Mode @ 36 MHz
Data Sheet
12
2001-02-23
C165UTAH
Overview
• USB Interface including:
– USB Specification 1.1 Compliant
– 12 Mbit/s Full-Speed Mode
– 7SW-configurable Endpoints, in addition to the bi-directional Control Endpoint 0
– 3 Configurations with 3 alternate settings and 4 interfaces supported
– Each non-Control Endpoint can be either Isochronous, Bulk or Interrupt
– Autonomous DMA Transfer by on-chip DMA for 8 USB endpoints
• ISDN Terminal Adapter Features including:
– Four Independent Full-Duplex HDLC Controllers
– IOM-2/PCM interface supporting TE, LT and PCM mode
– MON and CI1/CI2-Handler
– Two D-Channels
– Two B-Channels Supported
– Concatenated 2B+D channel Support
– Two Intercommunication Channels IC1, IC2
– D-Channel Access Control to first IOM channel-0 by S/G bit
– CDA Channel Access to individual IOM-2/PCM channels by SW
• On-Chip PLL for CPU and USB clock generation
• External crystal and direct driven input clock of 8 MHz when USB interface is used. In
applications without USB, the input clock frequency can vary between 4 and 20 MHz
dependent on the CPU target clock frequency.
• Single and variable crystal clock input frequency (using USB 8 MHz only)
• Bootstrap Loader support via USART interface
• On-Chip Debug Support (OCDS)
• JTAG Boundary Scan Test support according to IEEE 1149.1
• 3.3 V single supply voltage
• 5 V (TTL-) tolerant I/Os
• C165UTAH is available in 144-Pin P-TQFP package
Data Sheet
13
2001-02-23
C165UTAH
Overview
Power Management
Besides the basic power-save (power-reduction) modes Idle mode and Power down
mode, the C165UTAH offers a number of additional power management features, which
can be selectively used for effective power reduction. Refer to Table 1.
Table 1
Mode
Overview of Power Management Modes
Description
CPU Wake-up
Running The system is fully operational. All clocks and ---mode
peripherals are set and enabled, as determined
by software. Full power consumption.
Slow
down
mode
The CPU runs slower. The oscillator runs at a
lower frequency; the clock is divided by a
programmable factor (1...32). Peripherals
management is possible; incl. PLL On/Off.
Refer to register SYSCON 2.
Controlled by software.
Idle
mode
When the processor has no active tasks to
• Any interrupt
perform, it enters Idle mode by the IDLE
• Reset
command. All peripherals remain powered and
clocked, however, peripherals management is
possible. For detailed description see
Chapter 20.1, "Idle Mode".
Sleep
mode
The program stops execution and turns off the
clocks for:
• almost the entire chip, but RTC, or
• the entire chip.
The whole clock system is stopped.
Refer to register SYSCON 1.
• All enabled external
interrupts
• NMI
• RTC timer
(in asynchronous mode)
• PEC requests
• ASC interface
• SSC interface
Power
down
mode
The program stops execution (instruction
PWRDN) and turns off the clocks for the CPU
and for all peripherals; ports optionally.
• Reset
Note: Peripherals Management enables the user to control (via software) the clock of
selected peripherals. Refer to register SYSCON 3.
• C165UTAH power requirement
Characteristics, Table 105
Data Sheet
in individual modes
14
is
described
in
DC
2001-02-23
C165UTAH
Overview
1.2
Logic Symbol
The C165UTAH logic symbol is shown in Figure 1 below.
P-TQFP-144
Full Speed
USB
C165UTAH
Address/Data
Bus
General Purpose
I/O
IOM-2/PCM
USART
SSC/SCI
Clock
Figure 1
Data Sheet
C165UTAH Logic Symbol
15
2001-02-23
C165UTAH
Overview
1.3
Pinning Diagram
unconnected
unconnected
unconnected
VDD
VSS
P1H.4
P1H.3
P1H.2
P1H.1
P1H.0
VDD
VSS
P1L.7
P1L.6
P1L.5
P1L.4
VDD
VSS
P1L.3
P1L.2
P1L.1
P1L.0
VDD
VSS
P0H.7
P0H.6
P0H.5
P0H.4
VDD
VSS
P0H.3
P0H.2
VDD
VSS
unconnected
unconnected
Figure 2 shows the pinning diagram of the C165UTAH.
108
109
104
100
96
92
88
84
80
76
73
72
112
68
116
64
C165 UTAH
120
60
124
56
SAF C165UTAH - LF
128
52
132
P-TQFP-144
48
136
44
140
40
144
1
4
8
12
16
20
24
28
32
37
36
unconnected
P0H.1
P0H.0
P0L.7
P0L.6
P0L.5
P0L.4
VDD
VSS
POL.3
POL.2
POL.1
P0L.0
VDD
VSS
EA
ALE
READY
WR / WRL
RD
VDD
VSS
P7.5
P7.4
P7.3
P7.2
P7.1
P7.0
VDD
VSS
P4.6
P4.5
P4.4
P4.3
P4.2
P4.1
unconnected
unconnected
TCK
TDI
TDO
TMS
VSS
VDD
DU
DD
DCL
FSC
CLKMODE
VSSA
XTAL1
XTAL2
VDDA
P3.3
P3.5
P3.6
P3.7
VSS
VDD
P3.8
P3.9
P3.10
P3.11
P3.12
P3.13
P3.15
P4.0
VSS
VDD
BRKIN
BRKOUT
unconnected
P1H.5
P1H.6
P1H.7
VSS
VDD
RSTIN
RSTOUT
NMI
P6.0
P6.1
P6.2
P6.3
P6.4
VSS
VDD
P6.5
P6.6
P6.7
VSSU
DMNS
DPLS
VDDU
P2.0
P2.1
P2.2
P2.3
VSS
VDD
P2.4
P2.5
P2.6
P2.7
VSS
VDD
TEST
TRST
144-T_UTAH
Figure 2
Data Sheet
Pinning Diagram of the C165UTAH
16
2001-02-23
C165UTAH
Overview
1.4
Typical Applications
1.4.1
ISDN NT and PBX Applications
The C165UTAH is designed to manage control message and data flow between the
ISDN S/U transceiver and a Personal Computer. Data and message transfer is possible
between either two of the following physical interfaces: IOM-2 interface, USB interface
or local memory via 16-bit µP-interface. Since the IOM-2 is transparently accessible,
additional IOM-2 devices such as CODEC, Voice-Encrypter or Voice-Codec devices can
be accessed e.g. via the second IOM-2 channel (IC1/IC2).
Figure 3 gives a general overview of the ISDN NT/PBX application for C165UTAH.
S/T-Interface
U-Interface
S
INTC-Q2 NH
PEF81912 NH
U
POTS-Interface
C165UTAH
SLICOFI-2
SRAM
Flash
external BUS
HV-SLIC
EBC
HDLC
IOM-2
HV-SLIC
C165
Core
XBUS
RAM
V.24-Interface
V.24
PEC
USB
EPEC
USB Host Interface
Figure 3
C165UTAH in High Feature Intelligent Network Terminations
Note: In IOM-2 LT mode for PBX systems, when additional external D-channel
controllers are used, the DRDY signal to control the access has to be connected
to an external fast interrupt. Within the terminal mode, this arbitration can be done
with the Stop/Go bit on IOM-2.
Data Sheet
17
2001-02-23
C165UTAH
Pin Descriptions
2
Pin Descriptions
2.1
C165UTAH Pin Diagram
P-TQFP-144
C165UTAH
RSTIN
RSTOUT
NMI
CLKMODE
XTAL1
XTAL2
GPIO
Port
Fast
External
Interrupts
P7(5:0)
EXnINT(7:0)
or I/O
GPT1/2
Port
P2(7:0)
P3(8)
P3(9)
P3(13)
USART/ASC I/O or TxD
Port
I/O or RxD
P3(10)
P3(11)
OCDS
IOM-2/PCM
Port
Figure 4
Data Sheet
P0L(7:0)
I/O or AD(7:0)
P0H(7:0)
I/O or AD(15:8)
P1L(7:0)
I/O or A(7:0)
P1H(7:0)
I/O or A(15:8)
I/O or A(22:16)
ALE
RD
WR/WRL
READY
P3(12)
MicroController
Bus
I/O or BHE or WRH
EA
P6(4:0)
BRKOUT
BRKIN
P6(5)
P6(6)
P6(7)
FSC
DCL
TRST
TDI
TMS
TCK
TDO
DU
DD
USB
Port
I/O or CLKOUT
P4(6:0)
P3(7:5,3)
I/O or MRST
I/O or MTSR
I/O or SCLK
SSC/SCI
Serial
Port
P3(15)
DPLS
DMNS
I/O or CS(4:0)
I/O or HOLD
I/O or HLDA
I/O or BREQ
JTAG
Port
TEST
C165UTAH Pin Configuration
18
2001-02-23
C165UTAH
Pin Descriptions
2.2
Table 2
Pin No.
C165UTAH Pin Definitions and Functions
Microprocessor Bus and Control Signals
Input (I)
Output (O)
Function
PORT0:
60-63, 66-69, P0L070-71, 77-78, P0L7,
81-84
P0H0P0H7
I/O
PORT0 consists of the two 8-bit bidirectional
I/O ports P0L and P0H. It is bitwise
programmable for input or output via direction
bits. For a pin configured as input, the output
driver is put into high-impedance.
In case of an external bus configuration,
PORT0 serves as the address (A) and
address/data (AD) bus in demultiplexed bus
modes.
Demultiplexed bus modes:
Data Path Width: 8-bit
16-bit
P0L0-P0L7:
D0-D7
D0-D7
P0H0-P0H7:
I/O
D8-D15
Multiplexed bus modes:
Data Path Width: 8-bit
16-bit
P0L0-P0L7:
AD0-AD7 AD0-AD7
P0H0-P0H7:
A8-A15
AD8-AD15
PORT1:
87-90, 93-96, P1L099-103, 109- P1L7,
111
P1H0P1H7
I/O
PORT1 consists of the two 8-bit bidirectional
I/O ports P1L and P1H. It is bitwise
programmable for input or output via direction
bits. For a pin configured as input, the output
driver is put into high-impedance.
PORT1 is used as the 16-bit address bus (A)
in demultiplexed bus modes and also after
switching from a demultiplexed bus mode to a
multiplexed bus mode (see Chapter 8.3).
Data Sheet
Symbol
19
2001-02-23
C165UTAH
Pin Descriptions
Table 2
Microprocessor Bus and Control Signals
Pin No.
Symbol
Input (I)
Output (O)
Function
31, 37-42
P4.0 P4.6
I/O
PORT4 is an 7-bit bidirectional I/O port. It is
bit-wise programmable for input or output via
direction bits. For a pin configured as input, the
output driver is put into high-impedance state.
In case of an external bus configuration, Port4
can be used to output the segment address
lines:
P40
A16
Least Significant Segment
Address Line
...
...
...
P4.6
A22
Most Significant Segment
Address Line
O
O
114
RSTIN
115
RSTOUT O
116
NMI
Data Sheet
I
I
Reset Input with Schmitt-Trigger
characteristics. A low level at this pin for a
specified duration while the oscillator is
running resets the device. An internal pull-up
resistor permits power-on reset using only a
capacitor connected to VSS.
Internal Reset Indication Output. This pin is set
to a low level when the C165UTAH is
executing either a hardware-, software- or a
watchdog timer reset. RSTOUT remains low
until the C165UTAH has initialized itself.
Non-Maskable Interrupt Input. A high to low
transition at this pin causes the CPU to vector
to the NMI trap routine. When the PWRDN
(power down) instruction is executed, the NMI
pin must be low in order to force the CPU to go
into power down mode. If NMI is high, when
PWRDN is executed, the device will continue
to run in normal mode. If not used, pin NMI
should be pulled high externally.
20
2001-02-23
C165UTAH
Pin Descriptions
Table 2
Microprocessor Bus and Control Signals
Pin No.
Symbol
Input (I)
Output (O)
Function
117-121,
124-126
P6.0P6.7
O
I/O
Port6 is an 8-bit bidirectional I/O port. It is bitwise programmable for input or output via
direction bits. For a pin configured as input, the
output driver is put into high-impedance state.
Port6 outputs can be configured as push/pull
or open-drain drivers.
P6.0 CS0
Chip Select 0 Output
...
...
...
P6.4 CS4
Chip Select 4 Output
P6.5
HOLD
External Master Hold
Request Input
P6.6
HLDA
Hold Acknowledge Output
P6.7
BREQ
Bus Request Output
O
...
O
I
O
O
131-134,
137-140
P2.0P2.7
I/O
I
I
PORT2 is an 8-bit bidirectional I/O port. It is
bit-wise programmable for input or output via
direction bits. For a pin configured as input, the
output driver is put into high-impedance state.
Port2 outputs can be configured as push/pull
or open-drain drivers.
P2.0
EX0IN
Fast External Interrupt 0
Input
P2.7
EX7IN
Fast External Interrupt 7
Input
53
RD
O
External Memory Read Strobe. RD is
activated for every external instruction or data
read access.
54
WR/WRL O
External Memory Write Strobe. In WR mode
this pin is activated for every external data
write access. In WRL mode this pin is
activated for low byte data write accesses on a
16-bit bus, and for every data write access on
an 8-bit bus. See WRCFG in register
SYSCON for mode selection.
56
ALE
Address Latch Enable Output. Can be used for
latching the address into external memory or
an address latch in the multiplexed bus
modes.
Data Sheet
O
21
2001-02-23
C165UTAH
Pin Descriptions
Table 2
Microprocessor Bus and Control Signals
Pin No.
Symbol
Input (I)
Output (O)
Function
55
READY
I
Ready Input. When the ready function is
enabled, a high level at this pin during an
external memory access will force the
insertion of memory cycle time waitstates until
the pin returns to an low level.
57
EA
I
External Access Enable pin. A low level at this
pin during and after Reset forces the CPU to
begin instruction execution out of external
memory.
Note: This pin must always be set to ’0’.
45-50
Data Sheet
P7.0P7.5
I/O
PORT7 is an 6-bit bidirectional I/O port. It is
bit-wise programmable for input or output via
direction bits. For a pin configured as input, the
output driver is put into high-impedance state.
Port7 outputs are push/pull drivers.
P7.0
GPIO0
...
P7.5
GPIO5
22
2001-02-23
C165UTAH
Pin Descriptions
Table 3
Pin No.
General Purpose I/O and Control Signals
Symbol
18-21, 24-30 P3.3,
P3.5P3.13,
P3.15
Input (I)
Output (O)
Function
I/O
I/O
I/O
PORT3 is a 11-bit bidirectional I/O port. It is
bit-wise programmable for input or output via
direction bits. For a pin configured as input, the
output driver is put into high-impedance state.
Port3 outputs can be configured as push/pull
or open-drain drivers.
The following PORT3 pins also serve for
alternate functions:
P3.3 T3OUT GPT1 Timer T3 Toggle Latch
Output
P3.5 T4IN
GPT1 Timer T4 Input for
Count/Gate/Reload/Capture
Input for Timer 3 T3EUD
Input for Timer 2 T2EUD
P3.6 T3IN
GPT1 Timer T3 Count/Gate/
Input
P3.7 T2IN
GPT1 Timer T2 Input for
Count/Gate/Reload/Capture
P3.8 MRST SSC Master-Rec./SlaveTransmit I/O
P3.9 MTSR SSC Master-Transmit/SlaveRec. O/I
P3.10 TxD0
ASC Clock/Data Output
(Async./Sync.)
P3.11 RxD0 ASC Data Input (Async.) or
I/O (Sync.)
External Memory High Byte
P3.12 BHE
Enable Signal
WRH
External Memory High Byte
Write Strobe
P3.13 SCLK SSC Master Clock Output/
Slave Clock Input (CPU
Clock)
P3.15 CLKOUT System Clock Output
(CPU Clock)
O
I
I
I
I/O
I/O
O
I/O
O
O
I/O
O
Data Sheet
23
2001-02-23
C165UTAH
Pin Descriptions
Table 4
USB Interface Signals
Pin No.
Symbol
Input (I)
Output (O)
Function
129
DPLS
I/O
USB Data+ input/output signal.
128
DMNS
I/O
USB Data- input/output signal.
Table 5
IOM-2 Interface Signals
Pin No.
Symbol
Input (I)
Output (O)
Function
9
DU
I/O, OD
IOM-2 Data Upstream Signal pin. (From
subscriber to network). For the pin configured
as input, the output driver is put into highimpedance state. Open-drain.
10
DD
I/O, OD
IOM-2 Data Downstream Signal pin. (From
network to subscriber). For the pin configured
as input, the output driver is put into highimpedance state. Open-drain.
11
DCL
I
IOM-2 Data Clock Signal Input pin
12
FSC
I
IOM-2 Frame Sync. Clock Signal Input pin
Table 6
Clock Interface Signals
Pin No.
Symbol
Input (I)
Output (O)
Function
15
XTAL1
I
External crystal input to the on-chip
oscillator. Clock input for direct driven clock
without using an external crystal. Function is
determined by the CLKMODE pin.
16
XTAL2
O
Output from the oscillator amplifier circuit. To
clock the C165UTAH from an external
source, drive XTAL1, while XTAL2 leaving
unconnected. Minimum and maximum high/
low and rise/fall times specified in the AC
characteristics must be observed.
13
CLKMODE I
Data Sheet
Clock Mode Select pin. CLKMODE must be
set to LOW if an external crystal is used. Set
to HIGH signal enables the direct clock input
path and switches the internal oscillator in
power down mode.
24
2001-02-23
C165UTAH
Pin Descriptions
Table 7
Boundary Scan / JTAG / Test Interface Signals/OCDS
Pin No.
Symbol
Input (I)
Output (O)
Function
3
TCK
I
Boundary Scan Test Clock Input. There is no
internal pull device implemented. During
normal operation, it is recommended to
connect TCK to VSS.
4
TDI
I
Boundary Scan Test Data Input. An internal
pull-up device is connected to TDI. During
normal operation, TDI can be left open.
5
TDO
O
Boundary Scan Test Data Output. During
normal operation, the output TDO can be left
open.
6
TMS
I
Boundary Scan Test Mode Select Input. An
internal pull-up device is connected to TMS.
During normal operation, TMS can be left
open.
144
TRST
I
Boundary Scan Test Reset. There is an
internal pull-up device implemented. TRST is
low active, which means the boundary scan
tap controller resets while TRST = ’0’. During
normal operation, TRST must be connected to
LOW signal (using ’0’ signal or external pulldown device), since the tap controller needs to
be in reset mode in normal operation.
In boundary scan test mode, TRST can be left
open, since the internal pull-up device
provides the necessary HIGH signal.
143
TEST
I
Test Mode Enable Pin.
HIGH signal enables the chip internal test
mode.
Note: In normal operation, TEST must be
connected to VSS (LOW signal) since
no internal Pull-Down resistor is
provided.
Data Sheet
25
2001-02-23
C165UTAH
Pin Descriptions
Table 7
Boundary Scan / JTAG / Test Interface Signals/OCDS
Pin No.
Symbol
Input (I)
Output (O)
Function
34
BRKIN
I
In OCDS mode, a falling edge from HIGH to
LOW signal on brkin forces the system to stop.
An internal pull-up resistor is provided.
35
BRKOUT O
Table 8
Pin No.
In OCDS mode, a falling edge on brkout
indicates the trigger of a pre-selected OCDS
event.
Power/Ground Signals
Input (I)
Output (O)
Function
8, 23, 33, 44, VDD
52, 59, 65, 76,
80, 86, 92, 98,
105, 113, 123,
136, 142
-
Digital Supply Voltage
7, 22, 32, 43, VSS
51, 58, 64, 75,
79, 85, 91, 97,
104, 112, 122,
135,141
-
17
VDDAX
-
Analog Supply Voltage:
VDDAX supplies the oscillator circuitry only,
and is internal not connected to VDD in order
to separate possible noise influence from the
noise sensitive part. External, on board level,
the VDDAX can be connected to the same
power supply as the VDD.
14
VSSAX
-
Analog Ground
VSSAX is connected to the oscillator circuitry
Gound only, in order to separate possible
noise influence. External, on board level, the
VSSA can be connected to the same Ground
as the VSS.
Data Sheet
Symbol
Note: All pins must be connected to VDD.
Digital Ground
Note: All pins must be connected to VSS.
26
2001-02-23
C165UTAH
Pin Descriptions
Table 8
Power/Ground Signals
Pin No.
Symbol
Input (I)
Output (O)
Function
130
VDDU
-
Digital Supply Voltage for USB Transceiver
VDDU supplies the USB transceiver only and
is internally not connected to VDD in order to
separate possible noise influence. External,
on board level, the VDDU can be connected to
the same power supply as VDD.
127
VSSU
-
Digital Ground for USB Transceiver
VSSU is connected to the USB transceiver
only and is internally not connected to the
common ground in order to separate possible
noise influence. External, on board level, the
VSSU can be connected to the same ground
as VSS.
Table 9
Unconnected Pins
Pin No.
Symbol
Input (I)
Output (O)
Function
1, 2, 36, 72,
73, 74, 106,
107, 108
unconnected
-
These pins are unconnected - no
function. Reserved for future use.
Data Sheet
27
2001-02-23
C165UTAH
Architectural Overview
3
Architectural Overview
The architecture of the C165UTAH combines the advantages of both RISC and CISC
processors in a very well-balanced way. The sum of the features which are combined
result in a high performance microcontroller, which is the right choice not only for today's
applications, but also for future engineering challenges. The C165UTAH not only
integrates a powerful CPU core and a set of peripheral units into one chip, but also
connects the units in a very efficient way. One of the four buses used concurrently on the
C165UTAH is the XBUS, an internal representation of the external bus interface. This
bus provides a standardized method of integrating application-specific peripherals to
produce derivates of the standard C165UTAH.
C165UTAH
XBUS Module
CPU
CORE
ROM
Interrupt Controller
Internal
RAM
PEC
OSC
JTAG
BUS
CTL
EPEC
OCDS
PLL
P0
P1
Figure 5
Data Sheet
WDT
SSC
IOM-2
USB
ASC
GPT1
GPT2
P2
P3
P4
P7
P6
C165UTAH Functional Block Diagram
28
2001-02-23
C165UTAH
Architectural Overview
3.1
Basic CPU Concepts and Optimizations
The main core of the CPU consists of a 4-stage instruction pipeline, a 16-bit arithmetic
and logic unit (ALU) and dedicated SFRs. Additional hardware is provided for a separate
multiply and divide unit, a bit-mask generator and a barrel shifter.
CPU
SP
STKOV
STKUN
MDH
MDL
Exec. Unit
Instr. Ptr.
Instr. Reg.
Mul/Div-HW
Bit-Mask Gen
ROM
4-Stage
Pipeline
Internal
RAM
R15
R15
General
ALU
32
16
(16-bit)
Purpose
Barrel - Shifter
Registers
R0
PSW
SYSCON
Context Ptr.
BUSCON 0
BUSCON 1
BUSCON 2
BUSCON 3
BUSCON 4
ADDRSEL 1
ADDRSEL 2
ADDRSEL 3
ADDRSEL 4
Data Page Ptr.
Code Seg. Ptr.
R0
16
MCB02147
Figure 6
CPU Block Diagram
To meet the demand for greater performance and flexibility, a number of areas has been
optimized in the processor core. Functional blocks in the CPU core are controlled by
signals from the instruction decode logic. These are summarized below, and described
in detail in the following sections:
1) High Instruction Bandwidth / Fast Execution
2) High Function 8-bit and 16-bit Arithmetic and Logic Unit
3) Extended bit Processing and Peripheral Control
4) High Performance Branch-, Call-, and Loop Processing
5) Consistent and Optimized Instruction Formats
6) Programmable Multiple Priority Interrupt Structure
Data Sheet
29
2001-02-23
C165UTAH
Architectural Overview
High Instruction Bandwidth / Fast Execution
Based on the hardware provisions, most of the C165UTAH's instructions can be
executed in just one machine cycle, which requires 55.6 ns at 36 MHz CPU clock. For
example, shift and rotate instructions are always processed within one machine cycle,
independent of the number of bits to be shifted.
Branch-, multiply- and divide instructions normally take more than one machine cycle.
These instructions, however, have also been optimized. For example, branch
instructions only require an additional machine cycle, when a branch is taken, and most
branches taken in loops require no additional machine cycles at all, due to the so-called
‘Jump Cache’.
A 32-bit / 16-bit division takes 1µs, a 16-bit * 16-bit multiplication takes 0.5 µs.
The instruction cycle time has been dramatically reduced through the use of instruction
pipelining. This technique allows the core CPU to process portions of multiple sequential
instruction stages in parallel. The following four stage pipeline provides the optimum
balancing for the CPU core:
FETCH: In this stage, an instruction is fetched from the RAM or from the external
memory, based on the current IP value.
DECODE: In this stage, the previously fetched instruction is decoded and the required
operands are fetched.
EXECUTE: In this stage, the specified operation is performed on the previously fetched
operands.
WRITE BACK: In this stage, the result is written to the specified location.
If this technique were not used, each instruction would require four machine cycles. This
increased performance allows a greater number of tasks and interrupts to be processed.
Instruction Decoder
Instruction decoding is primarily generated from PLA outputs based on the selected
opcode. No microcode is used and each pipeline stage receives control signals staged
in control registers from the decode stage PLAs. Pipeline holds are primarily caused by
wait states for external memory accesses and cause the holding of signals in the control
registers. Multiple-cycle instructions are performed through instruction injection and
simple internal state machines which modify required control signals.
High Function 8-bit and 16-bit Arithmetic and Logic Unit
All standard arithmetic and logical operations are performed in a 16-bit ALU. In addition,
for byte operations, signals are provided from bits six and seven of the ALU result to
correctly set the condition flags. Multiple precision arithmetic is provided through a
'CARRY-IN' signal to the ALU from previously calculated portions of the desired
operation. Most internal execution blocks have been optimized to perform operations on
either 8-bit or 16-bit quantities. Once the pipeline has been filled, one instruction is
Data Sheet
30
2001-02-23
C165UTAH
Architectural Overview
completed per machine cycle, except for multiply and divide. An advanced Booth
algorithm has been incorporated to allow four bits to be multiplied and two bits to be
divided per machine cycle. Thus, these operations use two coupled 16-bit registers, MDL
and MDH, and require four and nine machine cycles, respectively, to perform a 16-bit by
16-bit (or 32-bit by 16-bit) calculation plus one machine cycle to setup and adjust the
operands and the result. Even these longer multiply and divide instructions can be
interrupted during their execution to allow for very fast interrupt response. Instructions
have also been provided to allow byte packing in memory while providing sign extension
of bytes for word wide arithmetic operations. The internal bus structure also allows
transfers of bytes or words to or from peripherals based on the peripheral requirements.
A set of consistent flags is automatically updated in the PSW after each arithmetic,
logical, shift, or movement operation. These flags allow branching on specific conditions.
Support for both signed and unsigned arithmetic is provided through user-specifiable
branch tests. These flags are also preserved automatically by the CPU upon entry into
an interrupt or trap routine.
All targets for branch calculations are also computed in the central ALU.
A 16-bit barrel shifter provides multiple bit shifts in a single cycle. Rotates and arithmetic
shifts are also supported.
Extended Bit Processing and Peripheral Control
A large number of instructions has been dedicated to bit processing. These instructions
provide efficient control and testing of peripherals while enhancing data manipulation.
Unlike other microcontrollers, these instructions provide direct access to two operands
in the bit-addressable space without requiring to move them into temporary flags.
The same logical instructions available for words and bytes are also supported for bits.
This allows the user to compare and modify a control bit for a peripheral in one
instruction. Multiple bit shift instructions have been included to avoid long instruction
streams of single bit shift operations. These are also performed in a single machine
cycle.
In addition, bit field instructions have been provided, which allow the modification of
multiple bits from one operand in a single instruction.
High Performance Branch-, Call-, and Loop Processing
Due to the high percentage of branching in controller applications, branch instructions
have been optimized to require one extra machine cycle only when a branch is taken.
This is implemented by precalculating the target address while decoding the instruction.
To decrease loop execution overhead, three enhancements have been provided:
• he first solution provides single cycle branch execution after the first iteration of a loop.
Thus, only one machine cycle is lost during the execution of the entire loop. In loops
which fall through upon completion, no machine cycles are lost when exiting the loop.
Data Sheet
31
2001-02-23
C165UTAH
Architectural Overview
No special instructions are required to perform loops, and loops are automatically
detected during execution of branch instructions.
• The second loop enhancement allows the detection of the end of a table and avoids
the use of two compare instructions embedded in loops. One simply places the lowest
negative number at the end of the specific table, and specifies branching if neither this
value nor the compared value have been found. Otherwise the loop is terminated if
either condition has been met. The terminating condition can then be tested.
• The third loop enhancement provides a more flexible solution than the Decrement and
Skip on Zero instruction which is found in other microcontrollers. Through the use of
Compare and Increment or Decrement instructions, the user can make comparisons
to any value. This allows loop counters to cover any range. This is particularly
advantageous in table searching.
Saving of system state is automatically performed on the internal system stack avoiding
the use of instructions to preserve state upon entry and exit of interrupt or trap routines.
Call instructions push the value of the IP on the system stack, and require the same
execution time as branch instructions.
Instructions have also been provided to support indirect branch and call instructions.
This supports implementation of multiple CASE statement branching in assembler
macros and high level languages.
Consistent and Optimized Instruction Formats
To obtain optimum performance in a pipelined design, an instruction set has been
designed which incorporates concepts from Reduced Instruction Set Computing (RISC).
These concepts primarily allow fast decoding of the instructions and operands while
reducing pipeline holds. These concepts, however, do not preclude the use of complex
instructions, which are required by microcontroller users. The following goals were used
to design the instruction set:
1. Provide powerful instructions to perform operations which currently require
sequences of instructions and are frequently used. Avoid transfer into and out of
temporary registers such as accumulators and carry bits. Perform tasks in parallel
such as saving state upon entry into interrupt routines or subroutines.
2. Avoid complex encoding schemes by placing operands in consistent fields for each
instruction. Also avoid complex addressing modes which are not frequently used. This
decreases the instruction decode time while also simplifying the development of
compilers and assemblers.
3. Provide most frequently used instructions with one-word instruction formats. All other
instructions are placed into two-word formats. This allows all instructions to be placed
on word boundaries, which alleviates the need for complex alignment hardware. It
also has the benefit of increasing the range for relative branching instructions.
Data Sheet
32
2001-02-23
C165UTAH
Architectural Overview
The high performance offered by the hardware implementation of the CPU can efficiently
be utilized by a programmer via the highly functional C165UTAH instruction set which
includes the following instruction classes:
•
•
•
•
•
•
•
•
•
•
•
•
Arithmetic Instructions
Logical Instructions
Boolean Bit Manipulation Instructions
Compare and Loop Control Instructions
Shift and Rotate Instructions
Prioritize Instruction
Data Movement Instructions
System Stack Instructions
Jump and Call Instructions
Return Instructions
System Control Instructions
Miscellaneous Instructions
Possible operand types are bits, bytes and words. Specific instruction support the
conversion (extension) of bytes to words. A variety of direct, indirect or immediate
addressing modes are provided to specify the required operands.
Programmable Multiple Priority Interrupt System
The following enhancements have been included to allow processing of a large number
of interrupt sources:
1. Peripheral Event Controller (PEC): This processor is used to off-load many interrupt
requests from the CPU. It avoids the overhead of entering and exiting interrupt or trap
routines by performing single-cycle interrupt-driven byte or word data transfers with
an optional increment of either the PEC source or the destination pointer. Just one
cycle is 'stolen' from the current CPU activity to perform a PEC service.
2. Multiple Priority Interrupt Controller: This controller allows all interrupts to be placed at
any specified priority. Interrupts may also be grouped, which provides the user with
the ability to prevent similar priority tasks from interrupting each other. For each of the
possible interrupt sources there is a separate control register, which contains an
interrupt request flag, an interrupt enable flag and an interrupt priority bitfield. Once
having been accepted by the CPU, an interrupt service can only be interrupted by a
higher prioritized service request. For standard interrupt processing, each of the
possible interrupt sources has a dedicated vector location.
3. Multiple Register Banks: This feature allows the user to specify up to sixteen general
purpose registers located anywhere in the internal RAM. A single one-machine-cycle
instruction allows to switch register banks from one task to another.
4. Interruptable Multiple Cycle Instructions: Reduced interrupt latency is provided by
allowing multiple-cycle instructions (multiply, divide) to be interruptable.
Data Sheet
33
2001-02-23
C165UTAH
Architectural Overview
With an interrupt response time within a range from just 140 ns to 280 ns (in case of
internal program execution), the C165UTAH is capable of reacting very fast on nondeterministic events.
Its fast external interrupt inputs are sampled every 28 ns and allow to recognize even
very short external signals.
The C165UTAH also provides an excellent mechanism to identify and to process
exceptions or error conditions that arise during run-time, so called 'Hardware Traps'.
Hardware traps cause an immediate non-maskable system reaction which is similiar to
a standard interrupt service (branching to a dedicated vector table location). The
occurrence of a hardware trap is additionally signified by an individual bit in the trap flag
register (TFR). Except for another higher prioritized trap service being in progress, a
hardware trap will interrupt any current program execution. In turn, hardware trap
services can normally not be interrupted by standard or PEC interrupts.
Software interrupts are supported by means of the 'TRAP' instruction in combination with
an individual trap (interrupt) number.
3.2
On-Chip System Resources
The C165UTAH controllers provide a number of powerful system resources designed
around the CPU. The combination of CPU and these resources results in the high
performance of the members of this controller family.
Peripheral Event Controller (PEC) and Interrupt Control
The Peripheral Event Controller allows to respond to an interrupt request with a single
data transfer (word or byte) which only consumes one instruction cycle and does not
require to save and restore the machine status. Each interrupt source is prioritized every
machine cycle in the interrupt control block. If PEC service is selected, a PEC transfer is
started. If CPU interrupt service is requested, the current CPU priority level stored in the
PSW register is tested to determine whether a higher priority interrupt is currently being
serviced. When an interrupt is acknowledged, the current state of the machine is saved
on the internal system stack and the CPU branches to the system specific vector for the
peripheral.
The PEC contains a set of SFRs which store the count value and control bits for eight
data transfer channels. In addition, the PEC uses a dedicated area of RAM which
contains the source and destination addresses. The PEC is controlled similar to any
other peripheral through SFRs containing the desired configuration of each channel.
An individual PEC transfer counter is implicitly decremented for each PEC service
except forming in the continuous transfer mode. When this counter reaches zero, a
standard interrupt is performed to the vector location related to the corresponding
source. PEC services are very well suited, for example, to move register contents to/from
Data Sheet
34
2001-02-23
C165UTAH
Architectural Overview
a memory table. The C165UTAH has 8 PEC channels each of which offers such fast
interrupt-driven data transfer capabilities.
Memory Areas
The memory space of the C165UTAH is configured in a Von Neumann architecture
which means that code memory, data memory, registers and I/O ports are organized
within the same linear address space which covers up to 8 MBytes. The entire memory
space can be accessed bytewise or wordwise. Particular portions of the on-chip memory
have additionally been made directly bit addressable.
A 16-bit wide internal RAM (IRAM) provides fast access to General Purpose Registers
(GPRs), user data (variables) and system stack. The internal RAM may also be used for
code. A unique decoding scheme provides flexible user register banks in the internal
memory while optimizing the remaining RAM for user data. The size of the internal RAM
is 3 KByte.
The CPU disposes of an actual register context consisting of up to 16 wordwide and/or
bytewide GPRs, which are physically located within the on-chip RAM area. A Context
Pointer (CP) register determines the base address of the active register bank to be
accessed by the CPU at a time. The number of register banks is only restricted by the
available internal RAM space. For easy parameter passing, a register bank may overlap
others.
A system stack of up to 1024 words is provided as a storage for temporary data. The
system stack is also located within the on-chip RAM area, and it is accessed by the CPU
via the stack pointer (SP) register. Two separate SFRs, STKOV and STKUN, are
implicitly compared against the stack pointer value upon each stack access for the
detection of a stack overflow or underflow.
Hardware detection of the selected memory space is placed at the internal memory
decoders and allows the user to specify any address directly or indirectly and obtain the
desired data without using temporary registers or special instructions.
For Special Function Registers 1024 Bytes of the address space are reserved. The
standard Special Function Register area (SFR) uses 512 bytes, while the Extended
Special Function Register area (ESFR) uses the other 512 bytes. (E)SFRs are wordwide
registers which are used for controlling and monitoring functions of the different on-chip
units. Unused (E)SFR addresses are reserved for future members of the C165UTAH
family with enhanced functionality.
External Bus Interface
In order to meet the needs of designs where more memory is required than is provided
on chip, up to 8 MBytes of external RAM and/or ROM can be connected to the
microcontroller via its external bus interface. The integrated External Bus Controller
(EBC) allows to access external memory and/or peripheral resources in a very flexible
Data Sheet
35
2001-02-23
C165UTAH
Architectural Overview
way. For up to five address areas the bus mode (multiplexed / demultiplexed), the data
bus width (8-bit / 16-bit) and even the length of a bus cycle (waitstates, signal delays)
can be selected independently. This allows to access a variety of memory and peripheral
components directly and with maximum efficiency. If the device does not run in Single
Chip Mode, where no external memory is required, the EBC can control external
accesses in one of the following four different external access modes:
16-/18-/20-/24-bit Addresses, 16-bit Data, Demultiplexed
16-/18-/20-/24-bit Addresses, 8-bit Data, Demultiplexed
16-/18-/20-/24-bit Addresses, 16-bit Data, Multiplexed
16-/18-/20-/24-bit Addresses, 8-bit Data, Multiplexed
The demultiplexed bus modes use PORT1 for addresses and PORT0 for data input/
output. The multiplexed bus modes use PORT0 for both addresses and data input/
output. All modes use Port 4 for the upper address lines (A16...) if selected.
Important timing characteristics of the external bus interface (waitstates, ALE length and
Read/Write Delay) have been made programmable to allow the user the adaption of a
wide range of different types of memories and/or peripherals. Access to very slow
memories or peripherals is supported via a particular 'Ready' function.
For applications which require less than 64 KBytes of address space, a non-segmented
memory model can be selected, where all locations can be addressed by 16 bits, and
thus Port 4 is not needed as an output for the upper address bits (A22/A19/A17...A16),
as is the case when using the segmented memory model.
The on-chip XBUS is an internal representation of the external bus and allows to access
integrated application-specific peripherals/modules in the same way as external
components. It provides a defined interface for these customized peripherals.
3.3
Clock Generation Concept
The on-chip clock generator provides the C165UTAH with its basic clock signal that
controls all activities of the controller hardware. Its oscillator can either run with an
external crystal and appropriate oscillator circuitry (see also recommendations in
chapter „Dedicated Pins“) or it can be driven by an external oscillator. The oscillator
either directly feeds the external clock signal to the controller hardware (through buffers),
divides the external clock frequency by 2 or 4, or feeds an on-chip phase locked loop
(PLL) which multiplies the input frequency by a selectable factor F. This resulting internal
clock signal is also referred to as “CPU clock”. Two separated clock signals are
generated for the CPU itself and the peripheral part of the chip. While the CPU clock is
stopped during the idle mode, the peripheral clock keeps running. Both clocks are
switched off, when the power down mode is entered.
Data Sheet
36
2001-02-23
C165UTAH
Architectural Overview
Note: Pin13 CLKMODE must be connected to LOW signal if an external crystal is used.
Pin cockmode connected to HIGH signal enables the direct input path and
switches the oscillator circuit in power down mode.
The on-chip PLL circuit allows operation of the C165UTAH on a low frequency external
clock while still providing maximum performance. The PLL generates a CPU clock signal
with 50% duty cycle. The PLL also provides fail safe mechanisms which allow the
detection of frequency deviations and the execution of emergency actions in case of an
external clock failure.
In addition to the CPU clock, the PLL generates the USB clock which is used for the USB
module only. Figure 7 shows the general clock generation concept of the C165UTAH.
Note: If the USB interface of the C165UTAH is used, an USB clock of 48 MHz is
mandatory. In addition, a CPU clock equal or greater than 20 MHz is required in
order to guarantee the full USB functionality. According to Table 10, the only
possible input clock frequency when operating the USB interface is 8 MHz, either
using an external crystal or an direct input clock.
The following constrains must be taken into account when considering the clock concept:
1. The USB clock must be 48 MHz, see the note above. Since there is a fixed PLL
prescaler of 1/6, the XTAL1 frequency must be 8 MHz.
2. If running the USB interface, the CPU clock must be equal or greater than 20 MHz
3. The maximum CPU clock frequency is 36 MHz (18 MIPS).
4. All AC and DC specifications described in Chapter 25, "AC/DC Characteristics" must
be fulfilled.
5. The input frequency of the internal oscillator circuit is 4 MHz up to 20 MHz. This
applies only, if an external crystal is used.
6. In direct drive mode, the internal oscillator circuit is bypassed. The clock input
frequency range is 4 MHz up to 36 MHz.
7. The IOM-2 module requires a minimum CPU clock of 5 times bitrate clock (BCL) on
pin DCL. In PCM or LT mode the resulting clock is 5 * 2 MHz = 10 MHz. In TE mode
the minimum CPU clock frequency is 5 * 0.7 MHz ≈ 4 MHz.
Data Sheet
37
2001-02-23
C165UTAH
Architectural Overview
SYSCON2.10:14
Div2
Pin 30
clockout
%2
Reload
Mux
Slow
Down
Divider
Main CPU_Clock
Div1
Pin 16
XTAL2
Mux
fXTAL
fOSC
fCPU = fIN x F
%2
Mux
Mux
Mux
Pin 15
XTAL1
OSC
SYSCON2.7
OWD
control signals
Pin 13
clkmode
CLKCFG
(SYSCON: P0H7:5)
Clock
Control
Unit
CLOCK UNIT
common mode feedback
Phase
Detector
Charge
Loop
Pump
Filter
VCO
(288 MHz
@ 8 MHz)
Lock
CPU Divider
1/(6,8,12,24)
USB Divider
1/6 (fixed)
Detection
USB_Clock
(must be 48 MHz)
fUSB = fIN x 6
Feedback
Divider 1/6
PLL UNIT
Figure 7
C165UTAH General Clock Concept
Note: All supported clock modes for the C165UTAH are shown in Table 10. Because of
the limited size of the register, there are not all combinations adjustable, which can
be derived theoretical from Figure 7.
Data Sheet
38
2001-02-23
C165UTAH
Architectural Overview
Table 10
C165UTAH Clock Generation Modes
P0H.7-P0H.5 Frequency
Divider Activation
USB Interface is NOT used
0
0
1
fXTAL * 0.5
direct drive, D1 not active, D2 active, PLL free running (2..5 MHZ)
Note: The PLL can be switched off completely by setting bit
PLLDIS = ’1’ (SYSCON3.13, see page 470).
0
1
0
fXTAL * 1.5
D1 not active, D2 not active, F = 1.5
0
1
1
fXTAL * 1.0
direct drive, D1 not active, D2 not active, PLL free running (2..5
MHz)
Note: The PLL can be switched off completely by setting bit
PLLDIS = ’1’ (SYSCON3.13, see page 470).
fXTAL * 6.0
1
0
0
1
0
1
1
1
0
fXTAL * 3.0
D1 not active, D2 not active, F = 3.0
1
1
1
fXTAL * 4.5
D1 not active, D2 not active, F = 4.5, Default Mode
0
0
0
D1 not active, D2 not active, F = 6.0
fXTAL * 1.125 D1 active, D2 active, F = 1.125
fXTAL * 0.375 D1 active, D2 active, F = 0.375
USB Interface is used (USB clock must be 48 MHz)
1
1
1
1
0
1
fXTAL * 3
fXTAL * 4.5
D1 not active, D2 not active, F = 3.0
fUSB =def 48 MHz
Τ fXTAL = 8 MHz
Τ fCPU = 24 MHz
D1 not active, D2 not active, F = 4.5, Default Mode
fUSB =def 48 MHz
Τ fXTAL = 8 MHz
Τ fCPU = 36 MHz
PLL Operation
On power-up the PLL provides a stable clock signal within ca. 1 ms after VDD has
reached 3.3 V±10%, even if there is no external clock signal (in this case the PLL will run
on its basic frequency of 2...5 MHz). The PLL starts synchronizing with the external clock
signal as soon as it is available. Within ca. 1 ms after stable oscillations of the external
clock within the specified frequency range the PLL will be synchronous with this clock at
a frequency of F * fOSC, ie. the PLL locks to the external clock.
Note: If the C165UTAH is required to operate on the desired CPU clock directly after
reset make sure that RSTIN remains active until the PLL has locked (ca. 1 ms).
When PLL operation is selected the CPU clock is a selectable multiple of the oscillator
frequency, ie. the input frequency. The table above lists the possible selections.
Data Sheet
39
2001-02-23
C165UTAH
Architectural Overview
The PLL constantly synchronizes to the external clock signal. Due to the fact that the
external frequency is 1/F’th of the PLL output frequency the output frequency may be
slightly higher or lower than the desired frequency. This jitter is irrelevant for longer time
periods. For short periods (1...4 CPU clock cycles) it remains below 4%.
When the PLL detects a missing input clock signal it generates an interrupt request. This
warning interrupt indicates that the PLL frequency is no more locked, ie. no more stable.
This occurs when the input clock is unstable and especially when the input clock fails
completely, eg. due to a broken crystal. In this case the synchronization mechanism will
reduce the PLL output frequency down to the PLL’s basic frequency (2...5 MHz). The
basic frequency is still generated and allows the CPU to execute emergency actions in
case of a loss of the external clock.
Prescaler Operation
When pins P0.15-13 (P0H.7-5) are equal ’001’ during reset the CPU clock is derived
from the internal oscillator (input clock signal) by a 2:1 prescaler (see Table 10).
The frequency of fCPU is half the frequency of fXTAL and the high and low time of fCPU (ie.
the duration of an individual TCL) is defined by the period of the input clock fXTAL.
The timings listed in the ’AC Characteristics’ of the data sheet that refer to TCLs
therefore can be calculated using the period of fXTAL for any TCL.
Direct Drive
When pins P0.15-13 (P0H.7-5) equal ’011’ during reset the clock system is directly
driven from the internal oscillator with the input clock signal, ie. fOSC = fCPU.
The maximum input clock frequency depends on the clock signal’s duty cycle, because
the minimum values for the clock phases (TCLs) must be respected.
Oscillator Watchdog
The C165UTAH provides an Oscillator Watchdog (OWD) which monitors the clock signal
generated by the on-chip oscillator (either with a crystal or via external clock drive) in
prescaler or direct drive mode. For this operation the PLL provides a clock signal which
is used to supervise transitions on the oscillator clock. This PLL clock is independent
from the XTAL1 clock. When the expected oscillator clock transitions are missing the
OWD activates the PLL Unlock / OWD interrupt node and supplies the CPU with the PLL
clock signal. Under these circumstances the PLL will oscillate with its basic frequency.
The OWD’s interrupt output can be disabled by setting bit OSCENBL = '0' (default after
reset) in SYSCON register. In this case no oscillator watchdog interrupt request is
generated and the CPU clock signal is derived from the oscillator clock in any case
Note: The CPU clock source is only switched back to the oscillator clock after a
hardware reset.
Data Sheet
40
2001-02-23
C165UTAH
Architectural Overview
3.4
On-Chip Peripheral Blocks
The C165UTAH clearly separates peripherals from the core. This structure permits the
maximum number of operations to be performed in parallel and allows peripherals to be
added or deleted from family members without modifications to the core. Each functional
block processes data independently and communicates information over common
buses. Peripherals are controlled by data written to the respective Special Function
Registers (SFRs). These SFRs are located either within the standard SFR area
(00’FE00H...00’FFFFH) or within the extended ESFR area (00’F000H...00’F1FFH).
These built in peripherals either allow the CPU to interface with the external world, or
provide functions on-chip that otherwise were to be added externally in the respective
system.
The C165UTAH peripherals are:
• Two General Purpose Timer Blocks (GPT1 and GPT2)
• An Asynchronous/Synchronous Serial Interface (ASC)
• A High-Speed Synchronous Serial Interface (SSC)
• An Universal Serial Bus Interface (USB)
• An IOM-2 Interface (IOM-2)
• A Watchdog Timer (WDT)
• Seven I/O ports with a total of 72 I/O lines
Each peripheral also contains a set of Special Function Registers (SFRs), which control
the functionality of the peripheral and temporarily store intermediate data results. Each
peripheral has an associated set of status flags. Individually selected clock signals are
generated for each peripheral from binary multiples of the CPU clock.
Peripheral Interfaces
The on-chip peripherals generally have two different types of interfaces, an interface to
the CPU and an interface to external hardware. Communication between CPU and
peripherals is performed through Special Function Registers (SFRs) and interrupts. The
SFRs serve as control/status and data registers for the peripherals. Interrupt requests
are generated by the peripherals based on specific events which occur during their
operation (eg. operation complete, error, etc.).
For interfacing with external hardware, specific pins of the parallel ports are used, when
an input or output function has been selected for a peripheral. During this time, the port
pins are controlled by the peripheral (when used as outputs) or by the external hardware
which controls the peripheral (when used as inputs). This is called the 'alternate (input
or output) function' of a port pin, in contrast to its function as a general purpose I/O pin.
Peripheral Timing
Internal operation of CPU and peripherals is based on the CPU clock (fCPU). The on-chip
oscillator derives the CPU clock from the crystal or from the external clock signal. The
Data Sheet
41
2001-02-23
C165UTAH
Architectural Overview
clock signal which is gated to the peripherals is independent from the clock signal which
feeds the CPU. During Idle mode the CPU’s clock is stopped while the peripherals
continue their operation. Peripheral SFRs may be accessed by the CPU once per state.
When an SFR is written to by software in the same state where it is also to be modified
by the peripheral, the software write operation has priority. Further details on peripheral
timing are included in the specific sections about each peripheral.
Programming Hints
Access to SFRs
All SFRs reside in data page 3 of the memory space. The following addressing
mechanisms allow to access the SFRs:
• ndirect or direct addressing with 16-bit (mem) addresses it must be guaranteed that
the used data page pointer (DPP0...DPP3) selects data page 3.
• accesses via the Peripheral Event Controller (PEC) use the SRCPx and DSTPx
pointers instead of the data page pointers
• short 8-bit (reg) addresses to the standard SFR area do not use the data page
pointers but directly access the registers within this 512 Byte area.
• short 8-bit (reg) addresses to the extended ESFR area require switching to the 512
Byte extended SFR area. This is done via the EXTension instructions EXTR,
EXTP(R), EXTS(R).
Byte write operations to word wide SFRs via indirect or direct 16-bit (mem) addressing
or byte transfers via the PEC force zeros in the non-addressed byte. Byte write
operations via short 8-bit (reg) addressing can only access the low byte of an SFR and
force zeros in the high byte. It is therefore recommended, to use the bit field instructions
(BFLDL and BFLDH) to write to any number of bits in either byte of an SFR without
disturbing the non-addressed byte and the unselected bits.
Reserved Bits
Some of the bits which are contained in the C165UTAH's SFRs are marked as
'Reserved'. User software should never write '1's to reserved bits. These bits are
currently not implemented and may be used in future products to invoke new functions.
In this case, the active state for these functions will be '1', and the inactive state will be
'0'. Therefore writing only ‘0’s to reserved locations provides portability of the current
software to future devices. Read accesses to reserved bits return ‘0’s.
Parallel Ports
The C165UTAH provides up to 72 I/O lines which are organized into seven input/output
ports. All port lines are bit-addressable, and all input/output lines are individually (bitwise) programmable as inputs or outputs via direction registers. The I/O ports are true
bidirectional ports which are switched to high impedance state when configured as
inputs. The output drivers of three I/O ports can be configured (pin by pin) for push/pull
Data Sheet
42
2001-02-23
C165UTAH
Architectural Overview
operation or open-drain operation via control registers. During the internal reset, all port
pins are configured as inputs.
All port lines have programmable alternate input or output functions associated with
them. PORT0 and PORT1 may be used as address and data lines when accessing
external memory, while Port 4 outputs the additional segment address bits A22/A19/
A17...A16 in systems where segmentation is used to access more than 64 KBytes of
memory. Port 6 provides optional bus arbitration signals (BREQ, HLDA, HOLD) and chip
select signals. Port 2 accepts the fast external interrupt inputs. Port 3 includes alternate
functions of timers, serial interfaces, the optional bus control signal BHE and the system
clock output (CLKOUT). Port 7 is used for general purpose I/Os. All port lines that are
not used for these alternate functions may be used as general purpose I/O lines.
Serial Channels
Serial communication with other microcontrollers, processors, terminals or external
peripheral components is provided by two serial interfaces with different functionality, an
Asynchronous/Synchronous Serial Channel (ASC) and a High-Speed Synchronous
Serial Channel (SSC).
ASC is upward compatible with the serial ports of the Infineon 8-bit microcontroller
families and supports full-duplex asynchronous communication at up to 2.25 MBaud and
half-duplex synchronous communication at up to 4.5 MBaud @ 36 MHz CPU clock.
A dedicated baud rate generator allows to set up all standard baud rates without
oscillator tuning. For transmission, reception and error handling 4 separate interrupt
vectors are provided. In asynchronous mode, 8- or 9-bit data frames are transmitted or
received, preceded by a start bit and terminated by one or two stop bits. For
multiprocessor communication, a mechanism to distinguish address from data bytes has
been included (8-bit data plus wake up bit mode).
In synchronous mode, the ASC transmits or receives bytes (8 bits) synchronously to a
shift clock which is generated by the ASC. The ASC always shifts the LSB first. A loop
back option is available for testing purposes.
A number of optional hardware error detection capabilities has been included to increase
the reliability of data transfers. A parity bit can automatically be generated on
transmission or be checked on reception. Framing error detection allows to recognize
data frames with missing stop bits. An overrun error will be generated, if the last
character received has not been read out of the receive buffer register at the time the
reception of a new character is complete.
SSC supports full-duplex synchronous communication at up to 18 Mbaud @ 36 MHz
CPU clock in SSC master mode and up to 9 MBaud @ 36 MHz in SSC slave mode. It
may be configured so it interfaces with serially linked peripheral components. A
dedicated baud rate generator allows to set up all standard baud rates without oscillator
tuning. For transmission, reception and error handling 3 separate interrupt vectors are
provided.
Data Sheet
43
2001-02-23
C165UTAH
Architectural Overview
The SSC transmits or receives characters of 2...16 bits length synchronously to a shift
clock which can be generated by the SSC (master mode) or by an external master (slave
mode). The SSC can start shifting with the LSB or with the MSB and allows the selection
of shifting and latching clock edges as well as the clock polarity. A number of optional
hardware error detection capabilities has been included to increase the reliability of data
transfers. Transmit and receive error supervise the correct handling of the data buffer.
Phase and baudrate error detect incorrect serial data.
General Purpose Timer (GPT) Unit
The GPT units represent a very flexible multifunctional timer/counter structure which
may be used for many different time related tasks such as event timing and counting,
pulse width and duty cycle measurements, pulse generation, or pulse multiplication.
The five 16-bit timers are organized in two separate modules, GPT1 and GPT2. Each
timer in each module may operate independently in a number of different modes, or may
be concatenated with another timer of the same module.
Each timer can be configured individually for one of three basic modes of operation,
which are Timer, Gated Timer, and Counter Mode. In Timer Mode the input clock for a
timer is derived from the internal CPU clock divided by a programmable prescaler, while
Counter Mode allows a timer to be clocked in reference to external events (via TxIN).
Pulse width or duty cycle measurement is supported in Gated Timer Mode where the
operation of a timer is controlled by the ‘gate’ level on its external input pin TxIN.
The count direction (up/down) for each timer is programmable by software or may
additionally be altered dynamically by an external signal (TxEUD) to facilitate eg. position
tracking.
The core timers T3 and T6 have output toggle latches (TxOTL) which change their state
on each timer over-flow/underflow. The state of these latches may be output on port pins
(TxOUT) or may be used internally to concatenate the core timers with the respective
auxiliary timers resulting in 32/33-bit timers/counters for measuring long time periods
with high resolution.
Various reload or capture functions can be selected to reload timers or capture a timer’s
contents triggered by an external signal or a selectable transition of toggle latch TxOTL.
Watchdog Timer
The Watchdog Timer represents one of the fail-safe mechanisms which have been
implemented to prevent the controller from malfunctioning for longer periods of time.
The Watchdog Timer is always enabled after a reset of the chip, and can only be
disabled in the time interval until the EINIT (end of initialization) instruction has been
executed. Thus, the chip’s start-up procedure is always monitored. The software has to
be designed to service the Watchdog Timer before it overflows. If, due to hardware or
software related failures, the software fails to do so, the Watchdog Timer overflows and
Data Sheet
44
2001-02-23
C165UTAH
Architectural Overview
generates an internal hardware reset and pulls the RSTOUT pin low in order to allow
external hardware components to reset.
The Watchdog Timer is a 16-bit timer, clocked with the CPU clock divided either by 2 or
by 128. The high byte of the Watchdog Timer register can be set to a prespecified reload
value (stored in WDTREL) in order to allow further variation of the monitored time
interval. Each time it is serviced by the application software, the high byte of the
Watchdog Timer is reloaded.
3.5
Protected Bits
The C165UTAH provides a special mechanism to protect bits which can be modified by
the on-chip hardware from being changed unintentionally by software accesses to
related bits (see also chapter “The Central Processing Unit”).
The following bits are protected:
Register
Bit Name
Notes
T2IC, T3IC, T4IC
T2IR, T3IR, T4IR
GPT1 timer interrupt request flags
T5IC, T6IC
T5IR, T6IR
GPT2 timer interrupt request flags
CRIC
CRIR
GPT2 CAPREL interrupt request flag
T3CON, T6CON
T3OTL, T6OTL
GPTx timer output toggle latches
S0TIC, S0TBIC
S0TIR, S0TBIR
ASC transmit(buffer) interrupt request flags
S0RIC, S0EIC
S0RIR, S0EIR
ASC receive/error interrupt request flags
S0CON
S0REN
ASC receiver enable flag
SSCTIC, SSCRIC
SSCTIR, SSCRIR
SSC transmit/receive interrupt request flags
SSCEIC
SSCEIR
SSC error interrupt request flag
SSCCON
SSCBSY
SSC busy flag
SSCCON
SSCBE, SSCPE
SSC error flags
SSCCON
SSCRE, SSCTE
SSC error flags
TFR
TFR.15,14,13
Class A trap flags
TFR
TFR.7,3,2,1,0
Class B trap flags
XPyIC (y=3...0)
XPyIR (y=3...0)
X-Peripheral y interrupt request flag
Σ = 33 protected bits in the C165UTAH
Data Sheet
45
2001-02-23
C165UTAH
Memory Organization
4
Memory Organization
The memory space of the C165UTAH is configured in a “Von Neumann” architecture.
This means that code and data are accessed within the same linear address space. All
of the physically separated memory areas, internal RAM, the internal Special Function
Register Areas (SFRs and ESFRs), the address areas for integrated XBUS peripherals
and external memory are mapped into one common address space.
The C165UTAH provides a total addressable memory space of 8 MBytes. This address
space is arranged as 128 segments of 64 KBytes each, and each segment is again
subdivided into four data pages of 16 KBytes each (see Figure 8).
SEGMENT 1
SEGMENT 1
01’0000H
IRAM/SFR
XPER/XRAM
00’FFFFH
SFR Area
XRAM Extension
00’FFFFH
00’FE00H
00’E000H
Reserved
for
XRAM
IRAM
up to 3 KByte
00’A000H
Reserved
for
XFLASH
00’F200H
ESFR Area
00’8000H
00’F000H
Reserved
for XPERs
Internal
Program
Memory
Area
00’E800H
Reserved for
Compatible
XRAM
2 KB
00’0000H
00’E000H
Segment 0
Figure 8
Data Sheet
8 KByte
Memory Areas and Address Space
46
2001-02-23
C165UTAH
Memory Organization
Bytes are stored at even or odd byte addresses. Words are stored in ascending memory
locations with the low byte at an even byte address being followed by the high byte at
the next odd byte address. Double words (code only) are stored in ascending memory
locations as two subsequent words. Single bits are always stored in the specified bit
position at a word address. bit position 0 is the least significant bit of the byte at an even
byte address, and bit position 15 is the most significant bit of the byte at the next odd
byte address. bit addressing is supported for a part of the Special Function Registers, a
part of the internal RAM and for the General Purpose Registers.
xxxx6 H
15
14
Bits
8
xxxx5 H
7
6
Bits
0
xxxx4 H
Byte
xxxx3 H
Byte
xxxx2 H
Word (High Byte)
xxxx1 H
Word (Low Byte)
xxxx0 H
xxxxF H
MCD01996
Figure 9
Storage of Words, Byte and Bits in a Byte Organized Memory
Note: Byte units forming a single word or a double word must always be stored within
the same physical (internal, external, RAM) and organizational (page, segment)
memory area.
Data Sheet
47
2001-02-23
C165UTAH
Memory Organization
4.1
Internal RAM and SFR Area
The RAM/SFR area is located within data page 3 and provides access to the internal
RAM (IRAM, organized as X*16) and to two 512 Byte blocks of Special Function
Registers (SFRs). The C165UTAH provides 3 KByte of IRAM, see Figure 10.
The internal RAM serves for several purposes:
•
•
•
•
•
System Stack (programmable size)
General Purpose Register Banks (GPRs)
Source and destination pointers for the Peripheral Event Controller (PEC)
Variable and other data storage, or
Code storage.
RAM/SFR Area
00 ’ FFFF H
00 ’ FFFF H
00 ’ F000 H
SFR Area
00 ’ FE00 H
Data Page 3
00 ’ E000 H
External
Memory
00 ’ C000 H
Data Page 2
00 ’ 8000 H
Internal
RAM
Data Page 1
Internal
ROM
Area
00 ’ 4000 H
00 ’ F200 H
Data Page 0
ESFR Area
System Segment 0
64 KByte
Figure 10
Data Sheet
00 ’ 0000 H
RAM/SFR Area
4 KByte
00 ’ F000 H
MCD02233
Internal RAM Area and SFR Areas
48
2001-02-23
C165UTAH
Memory Organization
Note: The upper 256 Bytes of SFR area, ESFR area and internal RAM are bitaddressable (see shaded blocks in Figure 10).
Code accesses are always made on even byte addresses. The highest possible code
storage location in the internal RAM is either 00’FDFEH for single word instructions or
00’FDFCH for double word instructions. The respective location must contain a branch
instruction (unconditional), because sequential boundary crossing from internal RAM to
the SFR area is not supported and causes erroneous results.
Any word and byte data in the internal RAM can be accessed via indirect or long 16-bit
addressing modes, if the selected DPP register points to data page 3. Any word data
access is made on an even byte address. The highest possible word data storage
location in the internal RAM is 00’FDFEH. For PEC data transfers, the internal RAM can
be accessed independent of the contents of the DPP registers via the PEC source and
destination pointers.
The upper 256 Byte of the internal RAM (00’FD00H through 00’FDFFH) and the GPRs of
the current bank are provided for single bit storage, and thus they are bit addressable.
System Stack
The system stack may be defined within the internal RAM. The size of the system stack
is controlled by bitfield STKSZ in register SYSCON (see table below).
Stack Size (Words)
Internal RAM Addresses (Words)
000B
256
00’FBFEH...00’FA00H (Default after Reset)
001B
128
00’FBFEH...00’FB00H
010B
64
00’FBFEH...00’FB80H
011B
32
00’FBFEH...00’FBC0H
100B
512
00’FBFEH...00’F800H
101B
---
Reserved. Do not use this combination.
110B
---
Reserved. Do not use this combination.
111B
1024
00’FDFEH...00’F600H (Note: No circular stack)
For all system stack operations the on-chip RAM is accessed via the Stack Pointer (SP)
register. The stack grows downward from higher towards lower RAM address locations.
Only word accesses are supported to the system stack. A stack overflow (STKOV) and
a stack underflow (STKUN) register are provided to control the lower and upper limits of
the selected stack area. These two stack boundary registers can be used not only for
protection against data destruction, but also allow to implement a circular stack with
hardware supported system stack flushing and filling (except for option ’111’).
The technique of implementing this circular stack is described in chapter “System
Programming”.
Data Sheet
49
2001-02-23
C165UTAH
Memory Organization
General Purpose Registers
The General Purpose Registers (GPRs) use a block of 16 consecutive words within the
internal RAM. The Context Pointer (CP) register determines the base address of the
currently active register bank. This register bank may consist of up to 16 word GPRs (R0,
R1, ..., R15) and/or of up to 16 byte GPRs (RL0, RH0, ..., RL7, RH7). The sixteen byte
GPRs are mapped onto the first eight word GPRs (see table below).
In contrast to the system stack, a register bank grows from lower towards higher address
locations and occupies a maximum space of 32 Byte. The GPRs are accessed via short
2-, 4- or 8-bit addressing modes using the Context Pointer (CP) register as base address
(independent of the current DPP register contents). Additionally, each bit in the currently
active register bank can be accessed individually.
Mapping of General Purpose Registers to RAM Addresses
Internal RAM Address
Byte Registers
Word Register
+ 1EH
---
R15
+ 1CH
---
R14
+ 1AH
---
R13
+ 18H
---
R12
+ 16H
---
R11
+ 14H
---
R10
+ 12H
---
R9
+ 10H
---
R8
+ 0EH
RH7
RL7
R7
+ 0CH
RH6
RL6
R6
+ 0AH
RH5
RL5
R5
+ 08H
RH4
RL4
R4
+ 06H
RH3
RL3
R3
+ 04H
RH2
RL2
R2
+ 02H
RH1
RL1
R1
+ 00H
RH0
RL0
R0
C165UTAH supports fast register bank (context) switching. Multiple register banks can
physically exist within the internal RAM at the same time. Only the register bank selected
by the Context Pointer register (CP) is active at a given time, however. Selecting a new
active register bank is simply done by updating the CP register. A particular Switch
Context (SCXT) instruction performs register bank switching and an automatic saving of
the previous context. The number of implemented register banks (arbitrary sizes) is only
limited by the size of the available internal RAM.
Data Sheet
50
2001-02-23
C165UTAH
Memory Organization
Details on using, switching and overlapping register banks are described in chapter
“System Programming”.
PEC Source and Destination Pointers
The 16 word locations in the internal RAM from 00’FCE0H to 00’FCFEH (just below the
bit-addressable section) are provided as source and destination address pointers for
data transfers on the eight PEC channels. Each channel uses a pair of pointers stored
in two subsequent word locations with the source pointer (SRCPx) on the lower and the
destination pointer (DSTPx) on the higher word address (x = 7...0).
00’FD00 H
00’FCFE H
DSTP7
00’FCFE H
00’FCFC H
SRCP7
00’FCE0 H
00’FCDE H
PEC
Source
and
Destination
Pointers
Internal
RAM
00’FCE2 H
DSTP0
00’F600 H
00’FCE0 H
SRCP0
00’F5FE H
MCD03903
Figure 11
Location of the PEC Pointers
Whenever a PEC data transfer is performed, the pair of source and destination pointers,
which is selected by the specified PEC channel number, is accessed independent of the
current DPP register contents and also the locations referred to by these pointers are
accessed independent of the current DPP register contents. If a PEC channel is not
used, the corresponding pointer locations area available and can be used for word or
byte data storage.
For more details about the use of the source and destination pointers for PEC data
transfers see section “Interrupt and Trap Functions”.
Data Sheet
51
2001-02-23
C165UTAH
Memory Organization
Special Function Registers
The functions of the CPU, the bus interface, the I/O ports and the on-chip peripherals of
the C165UTAH are controlled via a number of so-called Special Function Registers
(SFRs). These SFRs are arranged within two areas of 512 Byte size each. The first
register block, the SFR area, is located in the 512 Bytes above the internal RAM
(00’FFFFH...00’FE00H), the second register block, the Extended SFR (ESFR) area, is
located in the 512 Bytes below the internal RAM (00’F1FFH...00’F000H).
Special function registers can be addressed via indirect and long 16-bit addressing
modes. Using an 8-bit offset together with an implicit base address allows to address
word SFRs and their respective low bytes. However, this does not work for the
respective high bytes!
Note: Writing to any byte of an SFR causes the non-addressed complementary byte to
be cleared!
The upper half of each register block is bit-addressable, so the respective control/status
bits can directly be modified or checked using bit addressing.
When accessing registers in the ESFR area using 8-bit addresses or direct bit
addressing, an Extend Register (EXTR) instruction is required before, to switch the short
addressing mechanism from the standard SFR area to the Extended SFR area. This is
not required for 16-bit and indirect addresses. The GPRs R15...R0 are duplicated, ie.
they are accessible within both register blocks via short 2-, 4- or 8-bit addresses without
switching.
ESFR_SWITCH_EXAMPLE:
EXTR
#4
;Switch to ESFR area for next 4 instr.
MOV
ODP2, #data16
;ODP2 uses 8-bit reg addressing
BFLDL
DP6, #mask, #data8
;Bit addressing for bit fields
BSET
DP1H.7
;Bit addressing for single bits
MOV
T8REL, R1
;T8REL uses 16-bit mem address,
;R1 is duplicated into the ESFR space
;(EXTR is not required for this access)
;----
;-------------------
;The scope of the EXTR #4 instruction...
;...ends here!
MOV
T8REL, R1
;T8REL uses 16-bit mem address,
;R1 is accessed via the SFR space
Data Sheet
52
2001-02-23
C165UTAH
Memory Organization
In order to minimize the use of the EXTR instructions the ESFR area mostly holds
registers which are mainly required for initialization and mode selection. Registers that
need to be accessed frequently are allocated to the standard SFR area, wherever
possible.
Note: The tools are equipped to monitor accesses to the ESFR area and will
automatically insert EXTR instructions, or issue a warning in case of missing or
excessive EXTR instructions.
4.2
External Memory Space
The C165UTAH is capable of using an address space of up to 8 MByte. Only parts of
this address space are occupied by internal memory areas. All addresses which are not
used for on-chip memory (RAM) or for registers may reference external memory
locations. This external memory is accessed via the C165UTAH’s external bus interface.
Four memory bank sizes are supported:
• Non-segmented mode: 64 KByte with A15...A0 on PORT0 or PORT1
• 2-bit segmented mode: 256 KByte with A17...A16 on Port 4 and
A15...A0 on PORT0 or PORT1
• 4-bit segmented mode: 1 MByte with A19...A16 on Port 4 and
A15...A0 on PORT0 or PORT1
• 8-bit segmented mode: 8 MByte with A22...A16 on Port 4 and A15...A0 on PORT0 or
PORT1
Each bank can be directly addressed via the address bus, while the programmable chip
select signals can be used to select various memory banks.
The C165UTAH also supports four different bus types:
•
•
•
•
Multiplexed 16-bit Buswith address and data on PORT0 (Default after Reset)
Multiplexed 8-bit Buswith address and data on PORT0/P0L
Demultiplexed 16-bit Buswith address on PORT1 and data on PORT0
Demultiplexed 8-bit Buswith address on PORT1 and data on P0L
Memory model and bus mode are selected during reset by pin EA and PORT0 pins. For
further details about the external bus configuration and control please refer to chapter
"The External Bus Interface".
External word and byte data can only be accessed via indirect or long 16-bit addressing
modes using one of the four DPP registers. There is no short addressing mode for
external operands. Any word data access is made to an even byte address.
For PEC data transfers the external memory can be accessed independent of the
contents of the DPP registers via the PEC source and destination pointers.
Data Sheet
53
2001-02-23
C165UTAH
Memory Organization
The external memory is not provided for single bit storage and therefore it is not bit
addressable.
4.3
Crossing Memory Boundaries
The address space of the C165UTAH is implicitly divided into equally sized blocks of
different granularity and into logical memory areas. Crossing the boundaries between
these blocks (code or data) or areas requires special attention to ensure that the
controller executes the desired operations.
Memory Areas are partitions of the address space that represent different kinds of
memory (if provided at all). These memory areas are the internal RAM/SFR area, the onchip X-Peripherals and the external memory.
Accessing subsequent data locations that belong to different memory areas is no
problem. However, when executing code, the different memory areas must be switched
explicitly via branch instructions. Sequential boundary crossing is not supported and
leads to erroneous results.
Note: Changing from the external memory area to the internal RAM/SFR area takes
place within segment 0.
Segments are contiguous blocks of 64 KByte each. They are referenced via the code
segment pointer CSP for code fetches and via an explicit segment number for data
accesses overriding the standard DPP scheme.
During code fetching segments are not changed automatically, but rather must be
switched explicitly. The instructions JMPS, CALLS and RETS will do this.
In larger sequential programs make sure that the highest used code location of a
segment contains an unconditional branch instruction to the respective following
segment, to prevent the prefetcher from trying to leave the current segment.
Data Pages are contiguous blocks of 16 KByte each. They are referenced via the data
page pointers DPP3...0 and via an explicit data page number for data accesses
overriding the standard DPP scheme. Each DPP register can select one of the possible
1024 data pages. The DPP register that is used for the current access is selected via the
two upper bits of the 16-bit data address. Subsequent 16-bit data addresses that cross
the 16 KByte data page boundaries therefore will use different data page pointers, while
the physical locations need not be subsequent within memory.
Data Sheet
54
2001-02-23
C165UTAH
Central Processor Unit
5
Central Processor Unit
Basic tasks of the CPU are to fetch and decode instructions, to supply operands for the
arithmetic and logic unit (ALU), to perform operations on these operands in the ALU, and
to store the previously calculated results. As the CPU is the main engine of the
C165UTAH controller, it is also affected by certain actions of the peripheral subsystem.
Since a four stage pipeline is implemented in the C165UTAH, up to four instructions can
be processed in parallel. Most instructions of the C165UTAH are executed in one
machine cycle (2 CPU clock cycles) due to this parallelism. This chapter describes how
the pipeline works for sequential and branch instructions in general, and which hardware
provisions have been made to speed the execution of jump instructions in particular. The
general instruction timing is described including standard and exceptional timing.
While internal memory accesses are normally performed by the CPU itself, external
peripheral or memory accesses are performed by a particular on-chip External Bus
Controller (EBC), which is automatically invoked by the CPU whenever a code or data
address refers to the external address space. If possible, the CPU continues operating
while an external memory access is in progress. If external data are required but are not
yet available, or if a new external memory access is requested by the CPU, before a
previous access has been completed, the CPU will be held by the EBC until the request
can be satisfied. The EBC is described in a dedicated chapter.
CPU
SP
STKOV
STKUN
MDH
MDL
Exec. Unit
Instr. Ptr.
Instr. Reg.
Mul/Div-HW
Bit-Mask Gen
ROM
4-Stage
Pipeline
Internal
RAM
R15
R15
General
ALU
32
16
(16-bit)
Purpose
Barrel - Shifter
Registers
R0
PSW
SYSCON
Context Ptr.
BUSCON 0
BUSCON 1
BUSCON 2
BUSCON 3
BUSCON 4
ADDRSEL 1
ADDRSEL 2
ADDRSEL 3
ADDRSEL 4
Data Page Ptr.
Code Seg. Ptr.
R0
16
MCB02147
Figure 12
Data Sheet
CPU Block Diagram
55
2001-02-23
C165UTAH
Central Processor Unit
The on-chip peripheral units of the C165UTAH work nearly independent of the CPU with
a separate clock generator. Data and control information is interchanged between the
CPU and these peripherals via Special Function Registers (SFRs). Whenever
peripherals need a non-deterministic CPU action, an on-chip Interrupt Controller
compares all pending peripheral service requests against each other and prioritizes one
of them. If the priority of the current CPU operation is lower than the priority of the
selected peripheral request, an interrupt will occur.
Basically, there are two types of interrupt processing:
• Standard interrupt processing forces the CPU to save the current program status
and the return address on the stack before branching to the interrupt vector jump
table.
• PEC interrupt processing steals just one machine cycle from the current CPU
activity to perform a single data transfer via the on-chip Peripheral Event Controller
(PEC).
System errors detected during program execution (socalled hardware traps) or an
external non-maskable interrupt are also processed as standard interrupts with a very
high priority.
In contrast to other on-chip peripherals, there is a closer conjunction between the
watchdog timer and the CPU. If enabled, the watchdog timer expects to be serviced by
the CPU within a programmable period of time, otherwise it will reset the chip. Thus, the
watchdog timer is able to prevent the CPU from going totally astray when executing
erroneous code. After reset, the watchdog timer starts counting automatically, but it can
be disabled via software, if desired.
Beside its normal operation there are the following particular CPU states:
• Reset state: Any reset (hardware, software, watchdog) forces the CPU into a
predefined active state.
• IDLE state: The clock signal to the CPU itself is switched off, while the clocks for the
on-chip peripherals keep running.
• POWER DOWN state: All of the on-chip clocks are switched off.
A transition into an active CPU state is forced by an interrupt (if being IDLE) or by a reset
(if being in POWER DOWN mode).
The IDLE, POWER DOWN and RESET states can be entered by particular C165UTAH
system control instructions.
A set of Special Function Registers is dedicated to the functions of the CPU core:
•
•
•
•
•
•
General System Configuration: SYSCON (RP0H)
CPU Status Indication and Control: PSW
Code Access Control: IP, CSP
Data Paging Control: DPP0, DPP1, DPP2, DPP3
GPRs Access Control: CP
System Stack Access Control: SP, STKUN, STKOV
Data Sheet
56
2001-02-23
C165UTAH
Central Processor Unit
• Multiply and Divide Support: MDL, MDH, MDC
• ALU Constants Support: ZEROS, ONES
5.1
Instruction Pipelining
The instruction pipeline of the C165UTAH partitiones instruction processing into four
stages of which each one has its individual task:
1st –>FETCH:
In this stage the instruction selected by the Instruction Pointer (IP) and the Code
Segment Pointer (CSP) is fetched from either the internal RAM, or external memory.
2nd –>DECODE:
In this stage the instructions are decoded and, if required, the operand addresses are
calculated and the respective operands are fetched. For all instructions, which implicitly
access the system stack, the SP register is either decremented or incremented, as
specified. For branch instructions the Instruction Pointer and the Code Segment Pointer
are updated with the desired branch target address (provided that the branch is taken).
3rd –>EXECUTE:
In this stage an operation is performed on the previously fetched operands in the ALU.
Additionally, the condition flags in the PSW register are updated as specified by the
instruction. All explicit writes to the SFR memory space and all auto-increment or autodecrement writes to GPRs used as indirect address pointers are performed during the
execute stage of an instruction, too.
4th –>WRITE BACK:
In this stage all external operands and the remaining operands within the internal RAM
space are written back.
A particularity of the C165UTAH are the so-called injected instructions. These injected
instructions are generated internally by the machine to provide the time needed to
process instructions, which cannot be processed within one machine cycle. They are
automatically injected into the decode stage of the pipeline, and then they pass through
the remaining stages like every standard instruction. Program interrupts are performed
by means of injected instructions, too. Although these internally injected instructions will
not be noticed in reality, they are introduced here to ease the explanation of the pipeline
in the following.
Sequential Instruction Processing
Each single instruction has to pass through each of the four pipeline stages regardless
of whether all possible stage operations are really performed or not. Since passing
through one pipeline stage takes at least one machine cycle, any isolated instruction
takes at least four machine cycles to be completed. Pipelining, however, allows parallel
(ie. simultaneous) processing of up to four instructions. Thus, most of the instructions
Data Sheet
57
2001-02-23
C165UTAH
Central Processor Unit
seem to be processed during one machine cycle as soon as the pipeline has been filled
once after reset (see figure below).
Instruction pipelining increases the average instruction throughput considered over a
certain period of time. In the following, any execution time specification of an instruction
always refers to the average execution time due to pipelined parallel instruction
processing.
1 Machine
Cycle
FETCH
I1
DECODE
I2
I3
I4
I5
I6
I1
I2
I3
I4
I5
I1
I2
I3
I4
I1
I2
I3
EXECUTE
WRITEBACK
time
Figure 13
Sequential Instruction Pipelining
Standard Branch Instruction Processing
Instruction pipelining helps to speed sequential program processing. In the case that a
branch is taken, the instruction which has already been fetched providently is mostly not
the instruction which must be decoded next. Thus, at least one additional machine cycle
is normally required to fetch the branch target instruction. This extra machine cycle is
provided by means of an injected instruction (see Figure 14).
Injection
1 Machine
Cycle
FETCH
BRANCH
In+2
ITARGET
ITARGET+1
ITARGET+2
ITARGET+3
DECODE
In
BRANCH
(IINJECT)
ITARGET
ITARGET+1
ITARGET+2
EXECUTE
...
In
BRANCH
(IINJECT)
ITARGET
ITARGET+1
WRITEBACK
...
...
In
BRANCH
(IINJECT)
ITARGET
time
Figure 14
Data Sheet
Standard Branch Instruction Pipelining
58
2001-02-23
C165UTAH
Central Processor Unit
If a conditional branch is not taken, there is no deviation from the sequential program
flow, and thus no extra time is required. In this case the instruction after the branch
instruction will enter the decode stage of the pipeline at the beginning of the next
machine cycle after decode of the conditional branch instruction.
Cache Jump Instruction Processing
The C165UTAH incorporates a jump cache to optimize conditional jumps, which are
processed repeatedly within a loop. Whenever a jump on cache is taken, the extra time
to fetch the branch target instruction can be saved and thus the corresponding cache
jump instruction in most cases takes only one machine cycle.
This performance is achieved by the following mechanism:
Whenever a cache jump instruction passes through the decode stage of the pipeline for
the first time (and provided that the jump condition is met), the jump target instruction is
fetched as usual, causing a time delay of one machine cycle. In contrast to standard
branch instructions, however, the target instruction of a cache jump instruction (JMPA,
JMPR, JB, JBC, JNB, JNBS) is additionally stored in the cache after having been
fetched.
After each repeatedly following execution of the same cache jump instruction, the jump
target instruction is not fetched from progam memory but taken from the cache and
immediatly injected into the decode stage of the pipeline (see Figure 15).
A time saving jump on cache is always taken after the second and any further occurrence
of the same cache jump instruction, unless an instruction which, has the fundamental
capability of changing the CSP register contents (JMPS, CALLS, RETS, TRAP, RETI),
or any standard interrupt has been processed during the period of time between two
following occurrences of the same cache jump instruction.
1 Machine
Cycle
Injection of cached
Target Instruction
Injection
FETCH
In+2
ITARGET
ITARGET+1
In+2
ITARGET+1
ITARGET+2
DECODE
Cache Jmp
(IINJECT)
ITARGET
Cache Jmp
ITARGET
ITARGET+1
EXECUTE
In
Cache Jmp
(IINJECT)
In
Cache Jmp
ITARGET
WRITEBACK
...
In
Cache Jmp
...
In
Cache Jmp
1st loop iteration
Figure 15
Data Sheet
Repeated loop iteration
Cache Jump Instruction Pipelining
59
2001-02-23
C165UTAH
Central Processor Unit
Particular Pipeline Effects
Since up to four different instructions are processed simultaneously, additional hardware
has been spent in the C165UTAH to consider all causal dependencies which may exist
on instructions in different pipeline stages without a loss of performance. This extra
hardware (ie. for 'forwarding' operand read and write values) resolves most of the
possible conflicts (eg. multiple usage of buses) in a time optimized way and thus avoids
that the pipeline becomes noticeable for the user in most cases. However, there are
some very rare cases, where the circumstance that the C165UTAH is a pipelined
machine requires attention by the programmer. In these cases the delays caused by
pipeline conflicts can be used for other instructions in order to optimize performance.
• Context Pointer Updating
An instruction, which calculates a physical GPR operand address via the CP register, is
mostly not capable of using a new CP value, which is to be updated by an immediately
preceding instruction. Thus, to make sure that the new CP value is used, at least one
instruction must be inserted between a CP-changing and a subsequent GPR-using
instruction, as shown in the following example:
In
In+1
In+2
: SCXT CP, #0FC00h
: ....
: MOV R0, #dataX
; select a new context
; must not be an instruction using a GPR
; write to GPR 0 in the new context
• Data Page Pointer Updating
An instruction, which calculates a physical operand address via a particular DPPn (n=0
to 3) register, is mostly not capable of using a new DPPn register value, which is to be
updated by an immediately preceding instruction. Thus, to make sure that the new DPPn
register value is used, at least one instruction must be inserted between a DPPnchanging instruction and a subsequent instruction which implicitly uses DPPn via a long
or indirect addressing mode, as shown in the following example:
In
In+1
In+2
: MOV DPP0, #4
: ....
: MOV DPP0:0000H, R1
; select data page 4 via DPP0
; must not be an instruction using DPP0
; move contents of R1 to address location 01’0000H
; (in data page 4) supposed segmentation is enabled
• Explicit Stack Pointer Updating
None of the RET, RETI, RETS, RETP or POP instructions is capable of correctly using
a new SP register value, which is to be updated by an immediately preceding instruction.
Thus, in order to use the new SP register value without erroneously performed stack
accesses, at least one instruction must be inserted between an explicitly SP-writing and
any subsequent of the just mentioned implicitly SP-using instructions, as shown in the
following example:
In
: MOV SP, #0FA40H
Data Sheet
; select a new top of stack
60
2001-02-23
C165UTAH
Central Processor Unit
In+1
: ....
In+2
: POP R0
; must not be an instruction popping operands
; from the system stack
; pop word value from new top of stack into R0
Note: Conflicts with instructions writing to the stack (PUSH, CALL, SCXT) are solved
internally by the CPU logic.
• External Memory Access Sequences
The effect described here will only become noticeable, when watching the external
memory access sequences on the external bus (eg. by means of a Logic Analyzer).
Different pipeline stages can simultaneously put a request on the External Bus Controller
(EBC). The sequence of instructions processed by the CPU may diverge from the
sequence of the corresponding external memory accesses performed by the EBC, due
to the predefined priority of external memory accesses:
1st
2nd
3rd
Write Data
Fetch Code
Read Data.
• Controlling Interrupts
Software modifications (implicit or explicit) of the PSW are done in the execute phase of
the respective instructions. In order to maintain fast interrupt responses, however, the
current interrupt prioritization round does not consider these changes, ie. an interrupt
request may be acknowledged after the instruction that disables interrupts via IEN or
ILVL or after the following instructions. Timecritical instruction sequences therefore
should not begin directly after the instruction disabling interrupts, as shown in the
following example:
INT_OFF:
BCLR IEN
IN-1
CRIT_1ST: IN
...
CRIT_LAST: IN+x
INT_ON:
BSET IEN
; globally disable interrupts
; non-critical instruction
; begin of uninterruptable critical sequence
; end of uninterruptable critical sequence
; globally re-enable interrupts
Note: The described delay of 1 instruction also applies for enabling the interrupts system
ie. no interrupt requests are acknowledged until the instruction following the
enabling instruction.
• Initialization of Port Pins
Modifications of the direction of port pins (input or output) become effective only after the
instruction following the modifying instruction. As bit instructions (BSET, BCLR) use
internal read-modify-write sequences accessing the whole port, instructions modifying
Data Sheet
61
2001-02-23
C165UTAH
Central Processor Unit
the port direction should be followed by an instruction that does not access the same port
(see example below).
WRONG:
BSET DP3.13
BSET P3.5
; change direction of P3.13 to output
; P3.13 is still input, the rd-mod-wr reads pin P3.13
RIGHT:
BSET DP3.13
NOP
BSET P3.5
; change direction of P3.13 to output
; any instruction not accessing port 3
; P3.13 is now output,
; the rd-mod-wr reads the P3.13 output latch
• Changing the System Configuration
The instruction following an instruction that changes the system configuration via register
SYSCON (eg. segmentation, stack size) cannot use the new resources (eg. stack). In
these cases an instruction that does not access these resources should be inserted.
• BUSCON/ADDRSEL
The instruction following an instruction that changes the properties of an external
address area cannot access operands within the new area. In these cases an instruction
that does not access this address area should be inserted. Code accesses to the new
address area should be made after an absolute branch to this area.
Note: As a rule, instructions that change external bus properties should not be executed
from the respective external memory area.
• Timing
Instruction pipelining reduces the average instruction processing time in a wide scale
(from four to one machine cycles, mostly). However, there are some rare cases, where
a particular pipeline situation causes the processing time for a single instruction to be
extended either by a half or by one machine cycle. Although this additional time
represents only a tiny part of the total program execution time, it might be of interest to
avoid these pipeline-caused time delays in time critical program modules.
Besides a general execution time description, the following section provides some hints
on how to optimize time-critical program parts with regard to such pipeline-caused timing
particularities.
Data Sheet
62
2001-02-23
C165UTAH
Central Processor Unit
5.2
Bit-Handling and Bit-Protection
The C165UTAH provides several mechanisms to manipulate bits. These mechanisms
either manipulate software flags within the internal RAM, control on-chip peripherals via
control bits in their respective SFRs or control I/O functions via port pins.
The instructions BSET, BCLR, BAND, BOR, BXOR, BMOV, BMOVN explicitly set or
clear specific bits. The instructions BFLDL and BFLDH allow to manipulate up to 8 bits
of a specific byte at one time. The instructions JBC and JNBS implicitly clear or set the
specified bit when the jump is taken. The instructions JB and JNB (also conditional jump
instructions that refer to flags) evaluate the specified bit to determine if the jump is to be
taken.
Note: Bit operations on undefined bit locations will always read a bit value of ‘0’, while
the write access will not effect the respective bit location.
All instructions that manipulate single bits or bit groups internally use a read-modify-write
sequence that accesses the whole word, which contains the specified bit(s).
This method has several consequences:
• Bits can only be modified within the internal address areas, ie. internal RAM and SFRs.
External locations cannot be used with bit instructions.
The upper 256 bytes of the SFR area, the ESFR area and the internal RAM are bitaddressable (see chapter “Memory Organization”), ie. those register bits located within
the respective sections can be directly manipulated using bit instructions. The other
SFRs must be accessed byte/word wise.
Note: All GPRs are bit-addressable independent of the allocation of the register bank via
the context pointer CP. Even GPRs which are allocated to not bit-addressable
RAM locations provide this feature.
• The read-modify-write approach may be critical with hardware-effected bits. In these
cases the hardware may change specific bits while the read-modify-write operation is in
progress, where the writeback would overwrite the new bit value generated by the
hardware. The solution is either the implemented hardware protection (see below) or
realized through special programming (see “Particular Pipeline Effects”).
Protected bits are not changed during the read-modify-write sequence, ie. when
hardware sets eg. an interrupt request flag between the read and the write of the readmodify-write sequence. The hardware protection logic guarantees that only the intended
bit(s) is/are effected by the write-back operation.
Note: If a conflict occurs between a bit manipulation generated by hardware and an
intended software access the software access has priority and determines the
final value of the respective bit.
A summary of the protected bits implemented in the C165UTAH can be found at the end
of chapter “Architectural Overview”.
Data Sheet
63
2001-02-23
C165UTAH
Central Processor Unit
5.3
Instruction State Times
Basically, the time to execute an instruction depends on where the instruction is fetched
from, and where possible operands are read from or written to. The fastest processing
mode of the C165UTAH is to execute a program fetched from the internal code memory.
In that case most of the instructions can be processed within just one machine cycle,
which is also the general minimum execution time.
All external memory accesses are performed by the C165UTAH’s on-chip External Bus
Controller (EBC), which works in parallel with the CPU.
This section summarizes the execution times in a very condensed way. A detailled
description of the execution times for the various instructions and the specific exceptions
can be found in the “C16x Family Instruction Set Manual”.
The table below shows the minimum execution times required to process a C165UTAH
instruction fetched from the internal RAM or from external memory. These execution
times apply to most of the C165UTAH instructions - except some of the branches, the
multiplication, the division and a special move instruction. The numbers in the table are
in units of [ns], refer to a CPU clock of 20 MHz and assume no waitstates.
Table 11
Minimum Execution Times
Instruction Fetch
Word Operand Access
Memory Area
Word
Instruction
Doubleword
Instruction
Read from
Write to
Internal RAM
6
8
0/1
0
16-bit Demux Bus
2
4
2
2
16-bit Mux Bus
3
6
3
3
8-bit Demux Bus
4
8
4
4
8-bit Mux Bus
6
12
6
6
Execution from the internal RAM provides flexibility in terms of loadable and modifyable
code on the account of execution time.
Execution from external memory strongly depends on the selected bus mode and the
programming of the bus cycles (waitstates).
The operand and instruction accesses listed below can extend the execution time of an
instruction:
• Internal RAM operand reads via indirect addressing modes
• Internal SFR operand reads immediately after writing
• External operand reads
• External operand writes
Data Sheet
64
2001-02-23
C165UTAH
Central Processor Unit
• Jumps to non-aligned double word instructions in the internal ROM space
• Testing Branch Conditions immediately after PSW writes
5.4
CPU Special Function Registers
The core CPU requires a set of Special Function Registers (SFRs) to maintain the
system state information, to supply the ALU with register-addressable constants and to
control system and bus configuration, multiply and divide ALU operations, code memory
segmentation, data memory paging, and accesses to the General Purpose Registers
and the System Stack.
The access mechanism for these SFRs in the CPU core is identical to the access
mechanism for any other SFR. Since all SFRs can simply be controlled by means of any
instruction, which is capable of addressing the SFR memory space, a lot of flexibility has
been gained, without the need to create a set of system-specific instructions.
Note, however, that there are user access restrictions for some of the CPU core SFRs
to ensure proper processor operations. The instruction pointer IP and code segment
pointer CSP cannot be accessed directly at all. They can only be changed indirectly via
branch instructions.
The PSW, SP, and MDC registers can be modified not only explicitly by the programmer,
but also implicitly by the CPU during normal instruction processing. Note that any explicit
write request (via software) to an SFR supersedes a simultaneous modification by
hardware of the same register.
Note: Any write operation to a single byte of an SFR clears the non-addressed
complementary byte within the specified SFR.
Non-implemented (reserved) SFR bits cannot be modified, and will always supply
a read value of '0'.
The System Configuration Register SYSCON
This bit-addressable register provides general system configuration and control
functions. The reset value for register SYSCON depends on the state of the PORT0 pins
during reset (see hardware effectable bits).
Data Sheet
65
2001-02-23
C165UTAH
Central Processor Unit
SYSCON (FF12H / 89H)
15
14
STKSZ
rw
Bit
13
12
SFR
11
10
9
ROM SGT ROM BYT
DIS
S1
DIS
EN
rw
rw
rw
rw
8
Reset Value: 0XX0H
7
6
CLK WR CS
EN CFG CFG
rw
rw
rw
5
4
3
-
OSC
ENBL
-
-
rw
-
2
1
0
VISI XPERXPEN BLE SHARE
rw
rw
rw
Function
XPER-SHARE Reserved
The XPER-SHARE mode, known from other C16x Infineon derivatives, is not
supported in the C165UTAH. This bit must be set to ’0’ signal.
VISIBLE
Visible Mode Control
’0’: Accesses to XBUS peripherals are done internally
’1’: XBUS peripheral accesses are made visible on the external pins
XPEN
XBUS Peripheral Enable Bit
’0’: Accesses to the on-chip X-Peripherals and their functions are disabled
’1’: The on-chip X-Peripherals are enabled and can be accessed
Note: This bit is valid only for derivatives that contain X-Peripherals.
OSCENBL
Oscillator Watchdog Enable Bit
‘0’: The oscillator watchdog is disabled. Default configuration.
‘1’: The oscillator watchdog is enabled.
CSCFG
Chip Select Configuration Control
‘0’: Latched CS mode. The CS signals are latched internally and driven to the
enabled port pins synchronously.
‘1’: Unlatched CS mode. The CS signals are directly derived from the address
and driven to the enabled port pins.
WRCFG
Write Configuration Control (Set according to pin P0H.0 during reset)
’0’: Pins WR and BHE retain their normal function
’1’: Pin WR acts as WRL, pin BHE acts as WRH
CLKEN
System Clock Output Enable (CLKOUT)
’0’: CLKOUT disabled: pin may be used for general purpose I/O
’1’: CLKOUT enabled: pin outputs the system clock signal
BYTDIS
Disable/Enable Control for Pin BHE (Set according to data bus width)
’0’: Pin BHE enabled
’1’: Pin BHE disabled, pin may be used for general purpose I/O
ROMEN
Internal Boot-ROM Enable
’0’: Internal Boot-ROM is disabled. Access of the lower 32k address space will
be linked to external memory. During normal operation, bit ROMEN must
always be set to ’0’ signal
’1’: Internal Boot-ROM is enabled. This bit is only set in BSL mode.
Note: During BSL mode, if the lowest 32k of external memory needs to be
programmed, bit ROMEN must be set to ’0’ signal.
After BSL mode, make sure that bit ROMEN is cleared.
Data Sheet
66
2001-02-23
C165UTAH
Central Processor Unit
Bit
Function
SGTDIS
Segmentation Disable/Enable Control
’0’: Segmentation enabled
(CSP and IP are saved/restored during interrupt entry/exit)
’1’: Segmentation disabled (Only IP is saved/restored)
ROMS1
Reserved
The ROMS1, known from other C16x Infineon derivatives, is not supported in
the C165UTAH. This bit must be set to ’0’ signal.
STKSZ
System Stack Size
Selects the size of the system stack (in the internal RAM) from 32 to 1024 words
Note: Register SYSCON cannot be changed after execution of the EINIT instruction.
System Clock Output Enable (CLKEN)
The system clock output function is enabled by setting bit CLKEN in register SYSCON
to '1'. If enabled, port pin P3.15 takes on its alternate function as CLKOUT output pin.
The clock output is a 50 % duty cycle clock whose frequency equals the CPU operating
frequency (fOUT = fCPU).
Note: The output driver of port pin P3.15 is switched on automatically, when the
CLKOUT function is enabled. The port direction bit is disregarded.
After reset, the clock output function is disabled (CLKEN = ‘0’).
Segmentation Disable/Enable Control (SGTDIS)
Bit SGTDIS allows to select either the segmented or non-segmented memory mode.
In non-segmented memory mode (SGTDIS='1') it is assumed that the code address
space is restricted to 64 KBytes (segment 0) and thus 16 bits are sufficient to represent
all code addresses. For implicit stack operations (CALL or RET) the CSP register is
totally ignored and only the IP is saved to and restored from the stack.
In segmented memory mode (SGTDIS='0') it is assumed that the whole address space
is available for instructions. For implicit stack operations (CALL or RET) the CSP register
and the IP are saved to and restored from the stack. After reset the segmented memory
mode is selected.
Note: Bit SGTDIS controls if the CSP register is pushed onto the system stack in addition
to the IP register before an interrupt service routine is entered, and it is repopped
when the interrupt service routine is left again.
Data Sheet
67
2001-02-23
C165UTAH
Central Processor Unit
System Stack Size (STKSZ)
This bitfield defines the size of the physical system stack, which is located in the internal
RAM of the C165UTAH. An area of 32...512 words or all of the internal RAM may be
dedicated to the system stack. A so-called “circular stack” mechanism allows to use a
bigger virtual stack than this dedicated RAM area.
These techniques as well as the encoding of bitfield STKSZ are described in more detail
in chapter “System Programming”.
The Processor Status Word PSW
This bit-addressable register reflects the current state of the microcontroller. Two groups
of bits represent the current ALU status, and the current CPU interrupt status. A separate
bit (USR0) within register PSW is provided as a general purpose user flag.
Data Sheet
68
2001-02-23
C165UTAH
Central Processor Unit
PSW (FF10H / 88H)
15
14
13
SFR
12
Reset Value: 0000H
11
10
9
8
7
6
5
4
3
2
1
0
ILVL
IEN
HLD
EN
-
-
-
USR0
MUL
IP
E
Z
V
C
N
rw
rw
rw
-
-
-
rw
rw
rw
rw
rw
rw
rw
Bit
Function
N
Negative Result
Set, when the result of an ALU operation is negative.
C
Carry Flag
Set, when the result of an ALU operation produces a carry bit.
V
Overflow Result
Set, when the result of an ALU operation produces an overflow.
Z
Zero Flag
Set, when the result of an ALU operation is zero.
E
End of Table Flag
Set, when the source operand of an instruction is 8000H or 80H.
MULIP
Multiplication/Division In Progress
‘0’: There is no multiplication/division in progress.
‘1’: A multiplication/division has been interrupted.
USR0
User General Purpose Flag
May be used by the application software.
HLDEN,
ILVL, IEN
Interrupt and EBC Control Fields
Define the response to interrupt requests and enable external bus arbitration.
(Described in section “Interrupt and Trap Functions”)
ALU Status (N, C, V, Z, E, MULIP)
The condition flags (N, C, V, Z, E) within the PSW indicate the ALU status due to the last
recently performed ALU operation. They are set by most of the instructions due to
specific rules, which depend on the ALU or data movement operation performed by an
instruction.
After execution of an instruction which explicitly updates the PSW register, the condition
flags cannot be interpreted as described in the following, because any explicit write to
the PSW register supersedes the condition flag values, which are implicitly generated by
the CPU. Explicitly reading the PSW register supplies a read value which represents the
state of the PSW register after execution of the immediately preceding instruction.
Note: After reset, all of the ALU status bits are cleared.
• N-Flag: For most of the ALU operations, the N-flag is set to '1', if the most significant
bit of the result contains a '1', otherwise it is cleared. In the case of integer operations the
N-flag can be interpreted as the sign bit of the result (negative: N=’1’, positive: N=’0’).
Data Sheet
69
2001-02-23
C165UTAH
Central Processor Unit
Negative numbers are always represented as the 2's complement of the corresponding
positive number. The range of signed numbers extends from '–8000H' to '+7FFFH' for the
word data type, or from '–80H' to '+7FH' for the byte data type.For Boolean bit operations
with only one operand the N-flag represents the previous state of the specified bit. For
Boolean bit operations with two operands the N-flag represents the logical XORing of the
two specified bits.
• C-Flag: After an addition the C-flag indicates that a carry from the most significant bit
of the specified word or byte data type has been generated. After a subtraction or a
comparison the C-flag indicates a borrow, which represents the logical negation of a
carry for the addition.
This means that the C-flag is set to '1', if no carry from the most significant bit of the
specified word or byte data type has been generated during a subtraction, which is
performed internally by the ALU as a 2's complement addition, and the C-flag is cleared
when this complement addition caused a carry.
The C-flag is always cleared for logical, multiply and divide ALU operations, because
these operations cannot cause a carry anyhow.
For shift and rotate operations the C-flag represents the value of the bit shifted out last.
If a shift count of zero is specified, the C-flag will be cleared. The C-flag is also cleared
for a prioritize ALU operation, because a '1' is never shifted out of the MSB during the
normalization of an operand.
For Boolean bit operations with only one operand the C-flag is always cleared. For
Boolean bit operations with two operands the C-flag represents the logical ANDing of the
two specified bits.
• V-Flag: For addition, subtraction and 2's complementation the V-flag is always set to
'1', if the result overflows the maximum range of signed numbers, which are
representable by either 16 bits for word operations ('–8000H' to '+7FFFH'), or by 8 bits for
byte operations ('–80H' to '+7FH'), otherwise the V-flag is cleared. Note that the result of
an integer addition, integer subtraction, or 2's complement is not valid, if the V-flag
indicates an arithmetic overflow.
For multiplication and division the V-flag is set to '1', if the result cannot be represented
in a word data type, otherwise it is cleared. Note that a division by zero will always cause
an overflow. In contrast to the result of a division, the result of a multiplication is valid
regardless of whether the V-flag is set to '1' or not.
Since logical ALU operations cannot produce an invalid result, the V-flag is cleared by
these operations.
The V-flag is also used as 'Sticky Bit' for rotate right and shift right operations. With only
using the C-flag, a rounding error caused by a shift right operation can be estimated up
to a quantity of one half of the LSB of the result. In conjunction with the V-flag, the C-flag
allows evaluating the rounding error with a finer resolution (see table below).
For Boolean bit operations with only one operand the V-flag is always cleared. For
Data Sheet
70
2001-02-23
C165UTAH
Central Processor Unit
Boolean bit operations with two operands the V-flag represents the logical ORing of the
two specified bits.
Table 12
Shift Right Rounding Error Evaluation
C-Flag
V-Flag
Rounding Error Quantity
0
0
1
1
0
1
0
1
0<
No rounding error
Rounding error
Rounding error
Rounding error
< 1/2 LSB
= 1/2 LSB
> 1/2 LSB
• Z-Flag: The Z-flag is normally set to '1', if the result of an ALU operation equals zero,
otherwise it is cleared.
For the addition and subtraction with carry the Z-flag is only set to '1', if the Z-flag already
contains a '1' and the result of the current ALU operation additionally equals zero. This
mechanism is provided for the support of multiple precision calculations.
For Boolean bit operations with only one operand the Z-flag represents the logical
negation of the previous state of the specified bit. For Boolean bit operations with two
operands the Z-flag represents the logical NORing of the two specified bits. For the
prioritize ALU operation the Z-flag indicates, if the second operand was zero or not.
• E-Flag: The E-flag can be altered by instructions, which perform ALU or data
movement operations. The E-flag is cleared by those instructions which cannot be
reasonably used for table search operations. In all other cases the E-flag is set
depending on the value of the source operand to signify whether the end of a search
table is reached or not. If the value of the source operand of an instruction equals the
lowest negative number, which is representable by the data format of the corresponding
instruction ('8000H' for the word data type, or '80H' for the byte data type), the E-flag is
set to '1', otherwise it is cleared.
• MULIP-Flag: The MULIP-flag will be set to '1' by hardware upon the entrance into an
interrupt service routine, when a multiply or divide ALU operation was interrupted before
completion. Depending on the state of the MULIP bit, the hardware decides whether a
multiplication or division must be continued or not after the end of an interrupt service.
The MULIP bit is overwritten with the contents of the stacked MULIP-flag when the
return-from-interrupt-instruction (RETI) is executed. This normally means that the
MULIP-flag is cleared again after that.
Note: The MULIP flag is a part of the task environment! When the interrupting service
routine does not return to the interrupted multiply/divide instruction (ie. in case of
a task scheduler that switches between independent tasks), the MULIP flag must
be saved as part of the task environment and must be updated accordingly for the
new task before this task is entered.
Data Sheet
71
2001-02-23
C165UTAH
Central Processor Unit
CPU Interrupt Status (IEN, ILVL)
The Interrupt Enable bit allows to globally enable (IEN=’1’) or disable (IEN=’0’) interrupts.
The four-bit Interrupt Level field (ILVL) specifies the priority of the current CPU activity.
The interrupt level is updated by hardware upon entry into an interrupt service routine,
but it can also be modified via software to prevent other interrupts from being
acknowledged. In case an interrupt level '15' has been assigned to the CPU, it has the
highest possible priority, and thus the current CPU operation cannot be interrupted
except by hardware traps or external non-maskable interrupts. For details please refer
to chapter “Interrupt and Trap Functions”.
After reset all interrupts are globally disabled, and the lowest priority (ILVL=0) is
assigned to the initial CPU activity.
The Instruction Pointer IP
This register determines the 16-bit intra-segment address of the currently fetched
instruction within the code segment selected by the CSP register. The IP register is not
mapped into the C165UTAH's address space, and thus it is not directly accessable by
the programmer. The IP can, however, be modified indirectly via the stack by means of
a return instruction.
The IP register is implicitly updated by the CPU for branch instructions and after
instruction fetch operations.
IP (---- / --)
15
14
--13
12
11
10
9
8
Reset Value: 0000H
7
6
5
4
3
2
1
0
ip
(r)(w)
Bit
Function
ip
Specifies the intra segment offset, from where the current instruction is to be
fetched. IP refers to the current segment .
The Code Segment Pointer CSP
This non-bit addressable register selects the code segment being used at run-time to
access instructions. The lower 8 bits of register CSP select one of up to 256 segments
of 64 KBytes each, while the upper 8 bits are reserved for future use.
Data Sheet
72
2001-02-23
C165UTAH
Central Processor Unit
CSP (FE08H / 04H)
SFR
Reset Value: 0000H
15
14
13
12
11
10
9
8
7
6
5
4
3
-
-
-
-
-
-
-
-
SEGNR
-
-
-
-
-
-
-
-
r
2
1
0
Bit
Function
SEGNR
Segment Number
Specifies the code segment, from where the current instruction is to be fetched.
SEGNR is ignored, when segmentation is disabled.
Code memory addresses are generated by directly extending the 16-bit contents of the
IP register by the contents of the CSP register as shown in the figure below.
In case of the segmented memory mode the selected number of segment address bits
(via bitfield SALSEL) of register CSP is output on the respective segment address pins
of Port 4 for all external code accesses. For non-segmented memory mode or Single
Chip Mode the content of this register is not significant, because all code acccesses are
automatically restricted to segment 0.
Note: The CSP register can only be read but not written by data operations. It is,
however, modified either directly by means of the JMPS and CALLS instructions,
or indirectly via the stack by means of the RETS and RETI instructions.
Upon the acceptance of an interrupt or the execution of a software TRAP
instruction, the CSP register is automatically set to zero.
Data Sheet
73
2001-02-23
C165UTAH
Central Processor Unit
IP Register
CSP Register
Code Segment
FF’FFFF H
255
254
1
0
15
0
15
0
FE’0000 H
01’0000 H
24/20/18-Bit Physical Code Address
00’0000 H
MCA02265
Figure 16
Addressing via the Code Segment Pointer
Note: When segmentation is disabled, the IP value is used directly as the 16-bit address.
Data Page Pointers DPP0, DPP1, DPP2, DPP3
These four non-bit addressable registers select up to four different data pages being
active simultaneously at run-time. The lower 10 bits of each DPP register select one of
the 1024 possible 16-Kbyte data pages while the upper 6 bits are reserved for future use.
The DPP registers allow to access the entire memory space in pages of 16 Kbytes each.
The DPP registers are implicitly used, whenever data accesses to any memory location
are made via indirect or direct long 16-bit addressing modes (except for override
accesses via EXTended instructions and PEC data transfers). After reset, the Data Page
Pointers are initialized in a way that all indirect or direct long 16-bit addresses result in
identical 18-bit addresses. This allows to access data pages 3...0 within segment 0 as
shown in the figure below. If the user does not want to use any data paging, no further
action is required.
Data Sheet
74
2001-02-23
C165UTAH
Central Processor Unit
DPP0 (FE00H / 00H)
SFR
9
8
Reset Value: 0000H
15
14
13
12
11
10
-
-
-
-
-
-
DPP0PN
-
-
-
-
-
-
rw
DPP1 (FE02H / 01H)
7
6
5
4
SFR
9
8
14
13
12
11
10
-
-
-
-
-
-
DPP1PN
-
-
-
-
-
-
rw
7
6
5
4
SFR
9
8
14
13
12
11
10
-
-
-
-
-
-
DPP2PN
-
-
-
-
-
-
rw
7
6
5
4
SFR
9
8
1
0
3
2
1
0
Reset Value: 0002H
15
DPP3 (FE06H / 03H)
2
Reset Value: 0001H
15
DPP2 (FE04H / 02H)
3
3
2
1
0
Reset Value: 0003H
15
14
13
12
11
10
7
6
5
4
-
-
-
-
-
-
DPP3PN
-
-
-
-
-
-
rw
3
2
1
0
Bit
Function
DPPxPN
Data Page Number of DPPx
Specifies the data page selected via DPPx. Only the least significant two bits of
DPPx are significant, when segmentation is disabled.
Data paging is performed by concatenating the lower 14 bits of an indirect or direct long
16-bit address with the contents of the DPP register selected by the upper two bits of the
16-bit address. The content of the selected DPP register specifies one of the 1024
possible data pages. This data page base address together with the 14-bit page offset
forms the physical 24-bit address (selectable part is driven to the address pins).
In case of non-segmented memory mode, only the two least significant bits of the
implicitly selected DPP register are used to generate the physical address. Thus,
extreme care should be taken when changing the content of a DPP register, if a nonsegmented memory model is selected, because otherwise unexpected results could
occur.
Data Sheet
75
2001-02-23
C165UTAH
Central Processor Unit
In case of the segmented memory mode the selected number of segment address bits
(via bitfield SALSEL) of the respective DPP register is output on the respective segment
address pins of Port 4 for all external data accesses.
A DPP register can be updated via any instruction, which is capable of modifying an
SFR.
Note: Due to the internal instruction pipeline, a new DPP value is not yet usable for the
operand address calculation of the instruction immediately following the
instruction updating the DPP register.
16-Bit Data Address
15 14
0
Data Pages
1023
1022
1021
DPP Registers
3
DPP3-11
2
DPP2-10
1
DPP1-01
0
DPP0-00
14-Bit
Intra-Page Address
(concatenated with
content of DPPx).
Affer reset or with segmentation disabled the DPP registers select data pages 3...0.
All of the internal memory is accessible in these cases.
MCA02264
Figure 17
Addressing via the Data Page Pointers
Context Pointer CP
This non-bit addressable register is used to select the current register context. This
means that the CP register value determines the address of the first General Purpose
Register (GPR) within the current register bank of up to 16 wordwide and/or bytewide
GPRs.
Data Sheet
76
2001-02-23
C165UTAH
Central Processor Unit
CP (FE10H / 08H)
SFR
11
10
9
8
Reset Value: FC00H
15
14
13
12
7
6
5
4
3
2
1
0
1
1
1
1
cp
0
r
r
r
r
rw
r
Bit
Function
cp
Modifiable portion of register CP
Specifies the (word) base address of the current register bank.
When writing a value to register CP with bits CP.11...CP.9 = ‘000’, bits
CP.11...CP.10 are set to ‘11’ by hardware, in all other cases all bits of bit field
“cp” receive the written value.
Note: It is the user's responsibility that the physical GPR address specified via CP
register plus short GPR address must always be an internal RAM location. If this
condition is not met, unexpected results may occur.
• Do not set CP below the IRAM start address, ie. 00’FA00H/00’F600H/00’F200H
(1/2/3 KB)
• Do not set CP above 00’FDFEH
• Be careful using the upper GPRs with CP above 00’FDE0H
The CP register can be updated via any instruction which is capable of modifying an
SFR.
Note: Due to the internal instruction pipeline, a new CP value is not yet usable for GPR
address calculations of the instruction immediately following the instruction
updating the CP register.
The Switch Context instruction (SCXT) allows to save the content of register CP on the
stack and updating it with a new value in just one machine cycle.
Data Sheet
77
2001-02-23
C165UTAH
Central Processor Unit
Internal RAM
R15
(CP) + 30
R14
(CP) + 28
R13
R12
Context
Pointer
R11
R10
R9
R8
R7
.
.
.
R6
R5
R4
R3
R2
R1
(CP) + 2
R0
(CP)
MCD02003
Figure 18
Register Bank Selection via Register CP
Several addressing modes use register CP implicitly for address calculations. The
addressing modes mentioned below are described in chapter “Instruction Set
Summary”.
Short 4-Bit GPR Addresses (mnemonic: Rw or Rb) specify an address relative to the
memory location specified by the contents of the CP register, ie. the base of the current
register bank.
Depending on whether a relative word (Rw) or byte (Rb) GPR address is specified, the
short 4-bit GPR address is either multiplied by two or not before it is added to the content
of register CP (see figure below). Thus, both byte and word GPR accesses are possible
in this way.
GPRs used as indirect address pointers are always accessed wordwise. For some
instructions only the first four GPRs can be used as indirect address pointers. These
GPRs are specified via short 2-bit GPR addresses. The respective physical address
calculation is identical to that for the short 4-bit GPR addresses.
Short 8-Bit Register Addresses (mnemonic: reg or bitoff) within a range from F0H to
FFH interpret the four least significant bits as short 4-bit GPR address, while the four
Data Sheet
78
2001-02-23
C165UTAH
Central Processor Unit
most significant bits are ignored. The respective physical GPR address calculation is
identical to that for the short 4-bit GPR addresses. For single bit accesses on a GPR, the
GPR's word address is calculated as just described, but the position of the bit within the
word is specified by a separate additional 4-bit value.
Specified by reg or bitoff
Context
Pointer
1111
4-Bit GPR
Address
Internal
RAM
*2
Control
+
GPRs
For byte GPR
accesses
Figure 19
For word GPR
accesses
Must be
within the
internal
RAM area
MCD02005
Implicit CP Use by Short GPR Addressing Modes
Stack Pointer SP
This non-bit addressable register is used to point to the top of the internal system stack
(TOS). The SP register is pre-decremented whenever data is to be pushed onto the
stack, and it is post-incremented whenever data is to be popped from the stack. Thus,
the system stack grows from higher toward lower memory locations.
Since the least significant bit of register SP is tied to '0' and bits 15 through 12 are tied
to '1' by hardware, the SP register can only contain values from F000H to FFFEH. This
allows to access a physical stack within the internal RAM of the C165UTAH. A virtual
stack (usually bigger) can be realized via software. This mechanism is supported by
registers STKOV and STKUN (see respective descriptions below).
The SP register can be updated via any instruction, which is capable of modifying an
SFR.
Note: Due to the internal instruction pipeline, a POP or RETURN instruction must not
immediately follow an instruction updating the SP register.
Data Sheet
79
2001-02-23
C165UTAH
Central Processor Unit
SP (FE12H / 09H)
SFR
11
10
9
8
Reset Value: FC00H
15
14
13
12
7
6
1
1
1
1
sp
0
r
r
r
r
rw
r
Bit
Function
sp
Modifiable portion of register SP
Specifies the top of the internal system stack.
5
4
3
2
1
0
Stack Overflow Pointer STKOV
This non-bit addressable register is compared against the SP register after each
operation, which pushes data onto the system stack (eg. PUSH and CALL instructions
or interrupts) and after each subtraction from the SP register. If the content of the SP
register is less than the content of the STKOV register, a stack overflow hardware trap
will occur.
Since the least significant bit of register STKOV is tied to '0' and bits 15 through 12 are
tied to '1' by hardware, the STKOV register can only contain values from F000H to
FFFEH.
STKOV (FE14H / 0AH)
SFR
11
10
9
8
Reset Value: FA00H
15
14
13
12
7
6
5
4
1
1
1
1
stkov
0
r
r
r
r
rw
r
Bit
Function
stkov
Modifiable portion of register STKOV
Specifies the lower limit of the internal system stack.
3
2
1
0
The Stack Overflow Trap (entered when (SP) < (STKOV)) may be used in two different
ways:
• Fatal error indication treats the stack overflow as a system error through the
associated trap service routine. Under these circumstances data in the bottom of the
stack may have been overwritten by the status information stacked upon servicing the
stack overflow trap.
• Automatic system stack flushing allows to use the system stack as a 'Stack Cache'
for a bigger external user stack. In this case register STKOV should be initialized to a
Data Sheet
80
2001-02-23
C165UTAH
Central Processor Unit
value, which represents the desired lowest Top of Stack address plus 12 according to
the selected maximum stack size. This considers the worst case that will occur, when a
stack overflow condition is detected just during entry into an interrupt service routine.
Then, six additional stack word locations are required to push IP, PSW, and CSP for both
the interrupt service routine and the hardware trap service routine.
More details about the stack overflow trap service routine and virtual stack management
are given in chapter “System Programming”.
Stack Underflow Pointer STKUN
This non-bit addressable register is compared against the SP register after each
operation, which pops data from the system stack (eg. POP and RET instructions) and
after each addition to the SP register. If the content of the SP register is greater than the
the content of the STKUN register, a stack underflow hardware trap will occur.
Since the least significant bit of register STKUN is tied to '0' and bits 15 through 12 are
tied to '1' by hardware, the STKUN register can only contain values from F000H to
FFFEH.
STKUN (FE16H / 0BH)
SFR
11
10
9
8
Reset Value: FC00H
15
14
13
12
7
6
5
4
1
1
1
1
stkun
0
r
r
r
r
rw
r
Bit
Function
stkun
Modifiable portion of register STKUN
Specifies the upper limit of the internal system stack.
3
2
1
0
The Stack Underflow Trap (entered when (SP) > (STKUN)) may be used in two different
ways:
• Fatal error indication treats the stack underflow as a system error through the
associated trap service routine.
• Automatic system stack refilling allows to use the system stack as a 'Stack Cache'
for a bigger external user stack. In this case register STKUN should be initialized to a
value, which represents the desired highest Bottom of Stack address.
More details about the stack underflow trap service routine and virtual stack
management are given in chapter “System Programming”.
Data Sheet
81
2001-02-23
C165UTAH
Central Processor Unit
Scope of Stack Limit Control
The stack limit control realized by the register pair STKOV and STKUN detects cases
where the stack pointer SP is moved outside the defined stack area either by ADD or
SUB instructions or by PUSH or POP operations (explicit or implicit, ie. CALL or RET
instructions).
This control mechanism is not triggered, ie. no stack trap is generated, when
• the stack pointer SP is directly updated via MOV instructions
• the limits of the stack area (STKOV, STKUN) are changed, so that SP is outside of the
new limits.
Multiply/Divide High Register MDH
This register is a part of the 32-bit multiply/divide register, which is implicitly used by the
CPU, when it performs a multiplication or a division. After a multiplication, this non-bit
addressable register represents the high order 16 bits of the 32-bit result. For long
divisions, the MDH register must be loaded with the high order 16 bits of the 32-bit
dividend before the division is started. After any division, register MDH represents the
16-bit remainder.
MDH (FE0CH / 06H)
15
14
13
SFR
12
11
10
9
8
Reset Value: 0000H
7
6
5
4
3
2
1
0
mdh
rw
Bit
Function
mdh
Specifies the high order 16 bits of the 32-bit multiply and divide register MD.
Whenever this register is updated via software, the Multiply/Divide Register In Use
(MDRIU) flag in the Multiply/Divide Control register (MDC) is set to '1'.
When a multiplication or division is interrupted before its completion and when a new
multiply or divide operation is to be performed within the interrupt service routine, register
MDH must be saved along with registers MDL and MDC to avoid erroneous results.
A detailed description of how to use the MDH register for programming multiply and
divide algorithms can be found in chapter “System Programming”.
Data Sheet
82
2001-02-23
C165UTAH
Central Processor Unit
Multiply/Divide Low Register MDL
This register is a part of the 32-bit multiply/divide register, which is implicitly used by the
CPU, when it performs a multiplication or a division. After a multiplication, this non-bit
addressable register represents the low order 16 bits of the 32-bit result. For long
divisions, the MDL register must be loaded with the low order 16 bits of the 32-bit
dividend before the division is started. After any division, register MDL represents the 16bit quotient.
MDL (FE0EH / 07H)
15
14
13
SFR
12
11
10
9
8
Reset Value: 0000H
7
6
5
4
3
2
1
0
mdl
rw
Bit
Function
mdl
Specifies the low order 16 bits of the 32-bit multiply and divide register MD.
Whenever this register is updated via software, the Multiply/Divide Register In Use
(MDRIU) flag in the Multiply/Divide Control register (MDC) is set to '1'. The MDRIU flag
is cleared, whenever the MDL register is read via software.
When a multiplication or division is interrupted before its completion and when a new
multiply or divide operation is to be performed within the interrupt service routine, register
MDL must be saved along with registers MDH and MDC to avoid erroneous results.
A detailed description of how to use the MDL register for programming multiply and
divide algorithms can be found in chapter “System Programming”.
Multiply/Divide Control Register MDC
This bit addressable 16-bit register is implicitly used by the CPU, when it performs a
multiplication or a division. It is used to store the required control information for the
corresponding multiply or divide operation. Register MDC is updated by hardware during
each single cycle of a multiply or divide instruction.
MDC (FF0EH / 87H)
SFR
Reset Value: 0000H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
-
-
-
-
-
-
-
-
!!
!!
!!
MDR
IU
!!
!!
!!
!!
-
-
-
-
-
-
-
-
r(w)
r(w)
r(w)
r(w)
r(w)
r(w)
r(w)
r(w)
When a division or multiplication was interrupted before its completion and the multiply/
divide unit is required, the MDC register must first be saved along with registers MDH
Data Sheet
83
2001-02-23
C165UTAH
Central Processor Unit
Bit
Function
MDRIU
Multiply/Divide Register In Use
‘0’: Cleared, when register MDL is read via software.
‘1’: Set when register MDL or MDH is written via software, or when a multiply
or divide instruction is executed.
!!
Internal Machine Status
The multiply/divide unit uses these bits to control internal operations.
Never modify these bits without saving and restoring register MDC.
and MDL (to be able to restart the interrupted operation later), and then it must be
cleared prepare it for the new calculation. After completion of the new division or
multiplication, the state of the interrupted multiply or divide operation must be restored.
The MDRIU flag is the only portion of the MDC register which might be of interest for the
user. The remaining portions of the MDC register are reserved for dedicated use by the
hardware, and should never be modified by the user in another way than described
above. Otherwise, a correct continuation of an interrupted multiply or divide operation
cannot be guaranteed.
A detailed description of how to use the MDC register for programming multiply and
divide algorithms can be found in chapter “System Programming”.
Constant Zeros Register ZEROS
All bits of this bit-addressable register are fixed to '0' by hardware. This register can be
read only. Register ZEROS can be used as a register-addressable constant of all zeros,
ie. for bit manipulation or mask generation. It can be accessed via any instruction, which
is capable of addressing an SFR.
ZEROS (FF1CH / 8EH)
SFR
Reset Value: 0000H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
Constant Ones Register ONES
All bits of this bit-addressable register are fixed to '1' by hardware. This register can be
read only. Register ONES can be used as a register-addressable constant of all ones,
ie. for bit manipulation or mask generation. It can be accessed via any instruction, which
is capable of addressing an SFR.
Data Sheet
84
2001-02-23
C165UTAH
Central Processor Unit
ONES (FF1EH / 8FH)
SFR
Reset Value: FFFFH
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
r
5.5
PEC - Extension of Functionality
Introduction
Compared to existing C16x architecture, the PEC transfer function is enhanced by
extended functionality. The extended PEC function is a further step into DMA control
functionality. It especially supports integrated system design with XBUS as system bus.
Note: The device address decoding structure is always based on 24-bit addresses. But
due to the limited number of port P4 pins, only the address bits A22:A16 can be
made visible on the external X-Bus interface.
The extended PEC functions are defined as follows:
– Source pointer and destination pointer are extended to 24-bit pointer, thus enabling
PEC controlled data transfer between any two locations within the total address
space. Both 8-bit segment numbers of every source/destination pointer pair are
defined in one 16-bit SFR register; thus, 8 PEC segment number registers are
available for the 8 PEC channels.
– Two of the PEC channels are expanded by additional 16-bit transfer count registers;
when enabled, the original 8-bit bytecount in the control register serves as package
length count, thus defining the amount of bytes or words to be transferred with one
request. In C165UTAH the package size is always limited to one transfer.
– For always two channels a chaining feature is provided. When enabled in the PEC
control register, a termination interrupt of one channel will automatically switch
transfer control to the other channel of the channel pair.
24-bit Extension of Source and Destination Pointers
The source and destination pointers specify the locations between which the data is to
be moved. For each of the eight PEC channels the source and destination pointers are
specified by one SFR register and two IRAM memory locations. One SFR register stores
the 8-bit segment number of the source (PECSSN) and the 8-bit segment number of the
destination (PECDSN) location in a respective 16-bit PEC Segment Number register
(PECSNx). The respective segment offset of source and destination are stored in IRAM
memory location identical to the IRAM locations of SRCPx and DSTPx pointers of FullCustom C16x standard PEC channels - thus the extension is fully compatible. With the
segment number extension of source and destination, data can be transferred by a PEC
transfer between any two locations within the 8 MByte address space of the C165UTAH.
Data Sheet
85
2001-02-23
C165UTAH
Central Processor Unit
Note: The segment number extension of source and destination is provided for all 8 PEC
channels. After reset, all 8 segment number registers PECSNx are cleared,
providing full compatibility to FC-C16x PEC channels.
The PEC segment number registers PECSNx are defined as follows:
PECSNx (Addresses see table)
15
14
13
12
11
SFR
10
9
8
Reset Value: 0000H
7
6
5
4
3
PECDSN
PECSSN
rw
rw
2
1
0
Bit
Function
PECSSN
PEC Source Segment Number
8-bit Segment Number used for addressing the source of the respective PEC
transfer.
Note: Bits 6:0 can be used externally (address bits A22:A16). Due to the limited
number of pins, the upper bit 7 can not be used externally but can still be
used for chip select (CS) generation.
PECDSN
PEC Destination Segment Number
8-bit Segment Number used for addressing the destination of the respective
PEC transfer.
Note: Bits 14:8 can be used externally (address bits A22:A16). Due to the limited
number of pins, the upper bit 15 can not be used externally but can still be
used for chip select (CS) generation.
Table 13
PEC Segment Number Register Addresses
Register
Address
Reg. Space
Register
Address
Reg. Space
PECSN0
FED0H / 68H
SFR
PECSN4
FED8H / 6CH
SFR
PECSN1
FED2H / 69H
SFR
PECSN5
FEDAH / 6DH
SFR
PECSN2
FED4H / 6AH
SFR
PECSN6
FEDCH / 6EH
SFR
PECSN3
FED6H / 6BH
SFR
PECSN7
FEDEH / 6FH
SFR
Data Sheet
86
2001-02-23
C165UTAH
Central Processor Unit
Extended PEC Channel Control
The PEC control registers with the extended functionality and their application for new
PEC control are defined as follows:
PECCx
(Addresses: see table)
SFR
10
9
8
Reset Value: 0000H
15
14
13
12
11
7
6
5
4
3
PT
-
-
CLT
CL
INC
BWT
COUNT
rw
-
rw
rw
rw
rw
rw
rw
2
1
Bit
Function
COUNT
PEC Transfer Count
Counts PEC transfers (bytes or words) and influences the channel’s action
BWT
Byte / Word Transfer Selection
0: Transfer a Word
1: Transfer a Byte
INC
Increment Control (Modification of SRCPx or DSTPx)
0 0: Pointers are not modified
0 1: Increment DSTPx by 1 or 2 (BWT)
1 0: Increment SRCPx by 1 or 2 (BWT)
1 1: Reserved. Do not use this combination. (changed to 10 by hardware)
CL
Channel Link Control
0: PEC channels work independent
1: Pairs of channels are linked together
CLT
Channel Link Toggle State
0: Even numbered PEC channel of linked channels active
1: Odd numbered PEC channel of linked channels active
PT
Package Transfer
0: Single Transfer; extended Count2 not enabled
1: Package Transfer; extended Count2 enabled (only for channels 0 and 2)
0
Note: Package Transfer is only supported in PECC0 and PECC2
Data Sheet
87
2001-02-23
C165UTAH
Central Processor Unit
PEC Control Register Addresses
Register
Address
Reg. Space
Register
Address
Reg. Space
PECC0
FEC0H / 60H
SFR
PECC4
FEC8H / 64H
SFR
PECC1
FEC2H / 61H
SFR
PECC5
FECAH / 65H
SFR
PECC2
FEC4H / 62H
SFR
PECC6
FECCH / 66H SFR
PECC3
FEC6H / 63H
SFR
PECC7
FECEH / 67H
SFR
Byte/Word Transfer bit BWT controls, if a byte or a word is moved during a PEC service
cycle. This selection controls the transferred data size and the increment step for the
modified pointer.
Increment Control Field INC controls, if one of the PEC pointers is incremented after
the PEC transfer. It is not possible to increment both pointers, however. If the pointers
are not modified (INC=’00’), the respective channel will always move data from the same
source to the same destination.
Note: The reserved combination ‘11’ is changed to ‘10’ by hardware. However, it is not
recommended to use this combination.
The PEC Transfer Count Field COUNT controls the action of a respective PEC channel,
where the content of bit field COUNT at the time the request is activated selects the
action. COUNT may allow a specified number of PEC transfers, unlimited transfers or no
PEC service at all.
The table below summarizes, how the COUNT field itself, the interrupt requests flag IR
and the PEC channel action depends on the previous content of COUNT.
Previous Modified
COUNT
COUNT
IR after PEC Action of PEC Channel
service
and Comments
FFH
FFH
‘0’
Move a Byte / Word
Continuous transfer mode, ie. COUNT is not modified
FEH..02H
FDH..01H
‘0’
Move a Byte / Word and decrement COUNT
01H
00H
‘1’
Move a Byte / Word
Leave request flag set, which triggers another request
00H
00H
(‘1’)
No action!
Activate interrupt service routine rather than PEC channel.
The PEC transfer counter allows to service a specified number of requests by the
respective PEC channel, and then (when COUNT reaches 00H) activate the interrupt
service routine, which is associated with the priority level. After each PEC transfer the
COUNT field is decremented and the request flag is cleared to indicate that the request
has been serviced.
Data Sheet
88
2001-02-23
C165UTAH
Central Processor Unit
Continuous transfers are selected by the value FFH in bit field COUNT. In this case
COUNT is not modified and the respective PEC channel services any request until it is
disabled again.
When COUNT is decremented from 01H to 00H after a transfer, the request flag is not
cleared, which generates another request from the same source. When COUNT already
contains the value 00H, the respective PEC channel remains idle and the associated
interrupt service routine is activated instead. This allows to choose, if a level 15 or 14
request is to be serviced by the PEC or by the interrupt service routine.
Note: PEC transfers are only executed, if their priority level is higher than the CPU level,
ie. only PEC channels 7...4 are processed, while the CPU executes on level 14.
All interrupt request sources that are enabled and programmed for PEC service
should use different channels. Otherwise only one transfer will be performed for
all simultaneous requests. When COUNT is decremented to 00H, and the CPU is
to be interrupted, an incorrect interrupt vector will be generated.
Channel Link control bit CL controls the channel link mode. In this mode PEC
channels work by pair (channels 0 and 1, 2 and 3, 4 and 5, 6 and 7). The channel link
mode is enabled for one pair when the CL bit is set in any of the 2 PECCx registers. In
this case, the 2 channels handle PEC requests alternative to each other. The whole data
transfer is divided into several block transfers where each block is controlled by a PEC
channel. When a block transfer is completed, a channel link interrupt is generated and
the request processing is switched to the other PEC channel of the pair. This mechanism
allows to set up shadow and multiple buffers for PEC transfers by changing pointers and
count values of one channel when the other channel is active.
The very first transfer is always initiated with the even channel (called channel A, that is
channel 0, 2, 4 or 6). When the associated count field reaches 0 (COUNT or COUNT2
depending on the selected mode), the request service is transfered to the odd channel
(channel B, that is channel 1, 3, 5 or 7). If the CL bit of the "linked" channel is set and the
count field is different from 0, the next PEC requests will be serviced by this channel.
The channel link interrupts share one common interrupt node (Trap number 4CH - vector
location 00’0130H). This node is controlled by the Channel Link Interrupt Sub-Node
Control (CLISNC) register. It raises an interrupt request in case of one or more channel
link request flag and the respective enable control bit is set in CLISNC register. These
flags signal a PEC condition of the PEC linked channels which requires an action from
the CPU. The following conditions are possible:
1. in single transfer mode, a COUNT value change from 01H to 00H in a linked PEC
channel and the CL flag is set in the respective PEC control register,
2. in packet transfer mode, a COUNT2 value change from 0001H to 0000H in a linked
PEC channel and the CL flag is set in the respective PEC control register.
Data Sheet
89
2001-02-23
C165UTAH
Central Processor Unit
In these cases the CPU is requested to update the PEC control and pointer registers
while the next block transfer is executed. The last block transfer is determined by the
missing link bit in the linked PEC control register. If a service request hits a linked
channel with a COUNT field equal to 00H and the channel link flag disabled, a standard
interrupt is performed as known from standard PEC channels.
CLISNC (FFA8H / D4H)
SFR
Reset Value: 0000H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
-
-
C6
IR
C6
IE
-
-
C4
IR
C4
IE
-
-
C2
IR
C2
IE
-
-
C0
IR
C0
IE
-
-
rw
rw
-
-
rw
rw
-
-
rw
rw
-
-
rw
rw
Bit
Function
xxIE
Channel Link Interrupt Enable Bit (individual for each pair of linked channels)
’0’: Interrupt request disabled
’1’: Interrupt request enabled
xxIR
Channel Service Request Flag
’0’: No channel link service request pending
’1’: The channel pair has raised a request to service a PEC channel after
channel link
Packet Transfer control bit PT is implemented only in PECC0 and PECC2. When set
to ’1’, this bit enables the Packet Transfer mode. In this mode, each service request
initiates the transfer of an entire data packet of a fixed size. The COUNT field in the
PECCx register is used to define the size of the packet (in number of bytes or words
depending on the value of BWT). Therefore packets up to 256 bytes/words may be
transfered.
The register PECXC0/2 is then used to specify the number of requests to be serviced by
a PEC packet transfer before activating the interrupt service routine, which is associated
with the priority level. After each PEC packet transfer, the COUNT2 field is decremented
and the request flag is cleared, and then when COUNT2 reaches 0000H, an interrupt
request is generated to the corresponding interrupt vector.
Note: In the C165UTAH, the packet size is limited to 1. Packet transfers are not
supported, but the extended transfer count COUNT2 is used when PT bit is set.
Data Sheet
90
2001-02-23
C165UTAH
Central Processor Unit
PECXCx (FEFyH / 7zH, see table)
15
14
13
12
11
10
SFR
9
8
Reset Value: 0000H
7
6
5
4
3
2
1
0
COUNT2
rw
Bit
Function
COUNT2
Extended PEC Transfer Count
Counts PEC transfers and influences the channel’s action in Packet transfer
mode
PEC Extended Control Register Addresses
Register
Address
Reg. Space
Register
Address
Reg. Space
PECXC0
FEF0H / 78H
SFR
PECXC2
FEF2H / 79H
SFR
Source and destination pointers specifiy the locations between which the data is to be
moved. For PEC transfer description refer to Chapter 7.3, page 113, where the PEC
operation is descriped more in detail.
Channel Link Mode for Data Chaining
Data chaining with linked PEC channels is enabled, if the Channel Link Control Bit in
PECCx register is set to ’1’, either in one or both PEC channel control registers of a
channel pair. In this case, two PEC channels are linked together and handle chained
block transfers alternatively to each other. The whole data transfer is divided into several
block transfers where each block is controlled by one PEC channel of a channel pair.
When a data block is completely transferred a channel link interrupt is generated and
the PEC service request processing is automatically switched to the ’other’ PEC channel
of the channel-pair. Thus, PEC service requests addressed to a linked PEC channel are
either handled by linked PEC channel A or by linked PEC channel B. This channel toggle
allows to set up shadow and multiple buffers for PEC transfers by changing pointer and
count values of one channel while the other channel is active. The following table list the
channels that can be linked together and the channel numbers to address the linked
channels.
Data Sheet
91
2001-02-23
C165UTAH
Central Processor Unit
Table 14
PEC Channels which could be linked together
Linked PEC Channels
Linked PEC Channel
PEC Channel
A
PEC Channel
B
channel 0
channel 1
channel 0
channel 2
channel 3
channel 2
channel 4
channel 5
channel 4
channel 6
channel 7
channel 6
For each pair of linked channels, an internal channel flag, the Channel Link Toggle flag
CLT identifies which of the two PEC channels will serve the next PEC request. The CLT
flag is indicated in both PECCx registers of two linked PEC channels, where the CLT bit
in channel B always is inverse to the CLT bit in channel A. The very first transfer is
always started with the channel A if the CLT bit was not programmed otherwise before.
The CLT bit is only valid in case of linked PEC channels, indicated by the CL bits of linked
channels. If linking is not enabled, the CLT bit of both channels is always zero
(compatibility!).
The internal channel link flag CLT toggles, and the other channel begins service with the
next request if the "old" channel stops the service (COUNT=0 or COUNT2=0, dependent
on the mode), and if the new channel has in its PEC control register the CL flag enabled
and its transfer count is more than zero. Note: With the last transfer of a block transfer
(COUNT=0 or COUNT2=0), the channel link control flag CL of that channel is cleared in
its PECCx register. If the channel link flag CL of the new (chained) PEC control register
is found to be zero the whole data transfer is finished and the channel link interrupt is
coincidently a termination interrupt. The channel link mode is finished and the internal
channel toggle flag is cleared after the last transfer of the block, if the CL flags of both
pair channels are cleared.
Data Sheet
92
2001-02-23
C165UTAH
DMA - External PEC (EPEC)
6
DMA - External PEC (EPEC)
The EPEC provides fast and easy means to transfer single data between any memory
location within the address space by using the XBUS. The advantages are reduced
XBUS protocol handling and capability of addressing all system resources including
internal RAM and SFR.
6.1
EPEC Functionality
The EPEC provides a DMA controller for the USB device core to provide fast and flexible
data tranfer capability. The EPEC is implemented as a 16 channel controller with a 24bit source pointer, a 24-bit destination pointer and a 10-bit Transmit Byte Length Counter
with auto-increment support of two bytes (one word) per channel with Terminal Count
(TC) indication (Interrupt pulse valid for one clock cycle). After TC is reached, the counter
stops itself.
The EPEC is connected to the XBUS and to a proprietary 24-bit bus connected directly
to the C166 CBC. The EPEC has the highest priority among other interrupts and PECs
and does not participate in the interrupt priorization round. In case of an DPEC/EPEC
collision, the DPEC will get priority and one instruction cycle later the EPEC is
processed. The EPEC provides DMA like functionality by injecting a memory transfer
instruction (mov [dest], [src]) into the decode stage of the pipeline and thus only needs
one additional instruction cycle. Even in IDLE mode, the EPEC will be processed waking
up the CPU for one instruction cycle and immediately going back to IDLE state.
6.2
EPEC Implementation
The EPEC control block is located in the CBC core with its main purpose to synchronize
the external EPEC request to the internal T1-T4 states of the CPU and the priorization
between DPEC and EPEC. It also drives the externally provided 24-bit source and
destination pointer values on the internal memory address bus, thus controlling the
whole timing with respect to the CPU.
The EPEC Block diagram is shown in Figure 20 below.
Data Sheet
93
2001-02-23
C165UTAH
DMA - External PEC (EPEC)
epec_ack
Core
epec_req
Source Pointer#0
epec_ptr(23:0)
Destination Pointer#0
Source Pointer#7
epec_int
RX_DONE
smif/bpi
EPEC
MUX
PEC
TX_REQn
RX_REQn
TX_DONEn
Request Decoder
and Control
Destination Pointer#7
Start Register
10-bit TX_COUNTER#0
10-bit TX_COUNTER#7
XBUS
Figure 20
DMA/EPEC Block Diagram
The TX_REQn and RX_REQn shown in Figure 20 will be generated by the USBD to
request a word transfer over the XBUS from/to the FIFOs.
Besides the transfer request interrupts the EPEC provides a 10-bit transmit counter
register per channel, which will be written by SW. After the terminal count value is
reached, the counter stops and generates and TX_DONE pulse to the USBD. For each
endpoint an EPEC_CTRL register is provided to control the endpoint.
6.3
EPEC Register Description
The EPEC register description below shows the individual channel assignments
between requesting USB source interrupts and each individual EPEC channel. The
EPEC Register Base address is 00ED00H.
Table 15
EPEC Register Summary
00ED00H+
Name
Function
00H
EPECCLC
EPEC Clock Control Register
08H
EPECID
EPEC Identification Register
10H
EPEC_SPTR_IN_R00
16 LSBs of USB endpoint#0 source pointer IN
12H
EPEC_SPTR_IN_R01
8 MSBs of USB endpoint#0 source pointer IN
14H
EPEC_SPTR_OUT_R00
16 LSBs of USB endpoint#0 source pointer OUT
Data Sheet
94
2001-02-23
C165UTAH
DMA - External PEC (EPEC)
Table 15
EPEC Register Summary
00ED00H+
Name
Function
16H
EPEC_SPTR_OUT_R01
8 MSBs of USB endpoint#0 source pointer OUT
18H
EPEC_SPTR_REG10
16 LSBs of USB endpoint#1 source pointer
1AH
EPEC_SPTR_REG11
8 MSBs of USB endpoint#1 source pointer
1CH
EPEC_SPTR_REG20
16 LSBs of USB endpoint#2 source pointer
1EH
EPEC_SPTR_REG21
8 MSBs of USB endpoint#2 source pointer
20H
EPEC_SPTR_REG30
16 LSBs of USB endpoint#3 source pointer
22H
EPEC_SPTR_REG31
8 MSBs of USB endpoint#3 source pointer
24H
EPEC_SPTR_REG40
16 LSBs of USB endpoint#4 source pointer
26H
EPEC_SPTR_REG41
8 MSBs of USB endpoint#4 source pointer
28H
EPEC_SPTR_REG50
16 LSBs of USB endpoint#5 source pointer
2AH
EPEC_SPTR_REG51
8 MSBs of USB endpoint#5 source pointer
2CH
EPEC_SPTR_REG60
16 LSBs of USB endpoint#6 source pointer
2EH
EPEC_SPTR_REG61
8 MSBs of USB endpoint#6 source pointer
30H
EPEC_SPTR_REG70
16 LSBs of USB endpoint#7 source pointer
32H
EPEC_SPTR_REG71
8 MSBs of USB endpoint#7 source pointer
34H
EPEC_DPTR_IN_R00
16 LSBs of USB endpoint#0 destination pointer IN
36H
EPEC_DPTR_IN_R01
8 MSBs of USB endpoint#0 destination pointer IN
38H
EPEC_DPTR_OUT_R00
16 LSBs of USB endpoint#0 destination pointer OUT
3AH
EPEC_DPTR_OUT_R01
8 MSBs of USB endpoint#0 destination pointer OUT
3CH
EPEC_DPTR_REG10
16 LSBs of USB endpoint#1 destination pointer
3EH
EPEC_DPTR_REG11
8 MSBs of USB endpoint#1 destination pointer
40H
EPEC_DPTR_REG20
16 LSBs of USB endpoint#2 destination pointer
42H
EPEC_DPTR_REG21
8 MSBs of USB endpoint#2 destination pointer
44H
EPEC_DPTR_REG30
16 LSBs of USB endpoint#3 destination pointer
46H
EPEC_DPTR_REG31
8 MSBs of USB endpoint#3 destination pointer
48H
EPEC_DPTR_REG40
16 LSBs of USB endpoint#4 destination pointer
4AH
EPEC_DPTR_REG41
8 MSBs of USB endpoint#4 destination pointer
4CH
EPEC_DPTR_REG50
16 LSBs of USB endpoint#5 destination pointer
4EH
EPEC_DPTR_REG51
8 MSBs of USB endpoint#5 destination pointer
50H
EPEC_DPTR_REG60
16 LSBs of USB endpoint#6 destination pointer
52H
EPEC_DPTR_REG61
8 MSBs of USB endpoint#6 destination pointer
54H
EPEC_DPTR_REG70
16 LSBs of USB endpoint#7 destination pointer
56H
EPEC_DPTR_REG71
8 MSBs of USB endpoint#7 destination pointer
Data Sheet
95
2001-02-23
C165UTAH
DMA - External PEC (EPEC)
Table 15
EPEC Register Summary
00ED00H+
Name
Function
58H
EPEC_CTRL_IN_R0
Control and Status register for USB endpoint#0 IN
5AH
EPEC_CTRL_OUT_R0
Control and Status register for USB endpoint#0 OUT
5CH
EPEC_CTRL_REG1
Control and Status register for USB endpoint#1
5EH
EPEC_CTRL_REG2
Control and Status register for USB endpoint#2
60H
EPEC_CTRL_REG3
Control and Status register for USB endpoint#3
62H
EPEC_CTRL_REG4
Control and Status register for USB endpoint#4
64H
EPEC_CTRL_REG5
Control and Status register for USB endpoint#5
66H
EPEC_CTRL_REG6
Control and Status register for USB endpoint#6
68H
EPEC_CTRL_REG7
Control and Status register for USB endpoint#7
6AH
EPEC_INT_REG
EPEC Interrupt
6CH
EPEC_INTMSK_REG
EPEC Interrupt Mask Register
These registers are all reserved
6E..FFH
The detailed register description is shown below.
EPEC Clock Control Register
Address:
Name:
ED00H
EPECCLC
15 14 13 12 11 10
9
8
7
6
5
4
3
EPEC
EPEC
EX_DIS GPSEN
Reserved
Field
Bits
Type Value Description
EPECEX_DIS
3
R/W
Data Sheet
2
0
1
0
EPEC
DIS
EPEC
DISR
EPEC Controller Clock Disable
0: The clock of the EPEC interface
controller can be switched off using the
SYSCON register.
1: The clock of the EPEC interface
controller can NOT be switched off
using the SYSCON registers.
96
2001-02-23
C165UTAH
DMA - External PEC (EPEC)
Field
Bits
Type Value Description
EPECGPSEN
2
R/W
0
EPEC Controller Clock OCDS Disable
0: The clock of the EPEC interface
controller is enabled, normal operation.
1: The clock of the EPEC interface
controller is disabled during debugging
mode (OCDS)
EPECDIS
1
R
0
EPEC Controller Clock Status
0: The status of the EPEC interface
controller clock is ’enabled’.
1: The status of the EPEC interface
controller clock is ’disabled’.
EPECDISR
0
R/W
0
EPEC Controller Clock Disable
0: The clock of the EPEC interface
controller is enabled, normal operation.
1: The clock of the EPEC interface
controller is disabled.
RESERVED
15:4
-
0
These bits are reserved
The register EPECCLC is clocked with the bus clock to be able to switch the EPEC
interface controller clock on again, if it was off. If required, switching off the clock can be
prevented by the EPEC controller.
The state of the EPEC interface controller clock is controlled by the register bit
EPECDISR. The actual clock state will be shown by the state bit EPECDIS.
For on chip debugging support (OCDS) an additional bit EPECGPSEN is introduced to
stop the peripheral clock for arbitrary lengths of time during debugging if this function is
enabled. If debugging mode is active, the peripheral core rejects write access to
registers connected to the peripheral clock.
To be compatible with previous C16x products an EPECEX_DISR signal is provided to
disable the peripheral clock.
Data Sheet
97
2001-02-23
C165UTAH
DMA - External PEC (EPEC)
EPEC Identification Register
Address:
Name:
15
14
ED08H
EPECID
13
12
11
10
9
8
7
6
5
4
3
2
1
0
ID
Field
Bits
Type Value Description
ID
15:0
R
0
EPEC Identification Register.
Note: The value of the EPECID register is
hardwired to ZERO in the
C165UTAH.
EPEC_SPTR_REGx0 (x=7..0)
Table 16
Reset Value: 0000H
EPEC_SPTR_REGx0 Source Pointer Register
Bit No. Name
Function
15:0
16 LSBs of USB endpoint#x source pointer
SPTRx(15:0)
The EPEC source pointer registers (x0) provide the least significant 16-bits of the 24-bit
source pointer address for USB endpoints.
EPEC_SPTR_REGx1 (x=7..0)
Table 17
Reset Value: 0000H
EPEC_SPTR_REGx1 Source Pointer Register
Bit No. Name
Function
15:8
reserved
always 00H
7:0
SPTRx(23:16)
8 MSBs of USB endpoint#x source pointer
The EPEC source pointer registers (x1) provide the most significant 8-bits of the 24-bit
source pointer for USB endpoints.
EPEC_DPTR_REGx0 (x=7..0)
Table 18
Reset Value: 0000H
EPEC_DPTR_REGx0 Destination Pointer Register
Bit No. Name
Function
15:0
16 LSBs of USB endpoint#x destination pointer
Data Sheet
DPTRx(15:0)
98
2001-02-23
C165UTAH
DMA - External PEC (EPEC)
The EPEC destination pointer registers (x0) provide the least significant 16-bits of the
24-bit destination Pointer address for USB endpoints.
EPEC_DPTR_REGx1 (x=7..0)
Table 19
Reset Value: 0000H
EPEC_DPTR_REG01 Destination Pointer Register
Bit No. Name
Function
15:8
reserved
always 00H
7:0
DPTRx(23:16)
8 MSBs of USB endpoint#x source pointer
The EPEC destination pointer registers (x1) provide the most significant 8-bits of the 24bit destination pointer for USB endpoints.
EPEC_CTRL_REGx (x=7..0)
Table 20
Reset Value: 0000H
EPEC_CTRL_REGx Source Pointer Register
Bit No. Name
Function
15
TXR_ENAx
Transfer / Receive Enable control bit, set by SW and
cleared by EPEC after transfer complete
’1’: Transmitter / Receiver enabled
’0’: Transmitter / Receiver disabled
14
EXT_SRC
External Source
’1’: External source is selected
’0’: Reserved (program memory)
EXT_SRC must be set to ’1’
13:12
REQ_SRC
EPEC request sourceRX/TX Fifo
’10’: EPEC is connected to TX Fifo Request (IN)
’01’: EPEC is connected to RX Fifo Request (OUT)
’11’: EPEC is connected to RX and TX Fifo Request
(BI) for Control endpoint 0.
’00’: reserved
11
CNT_UP_DN
Byte Counter direction select
’1’: Rx
’0’: Tx
10
CLR
Clear EPEC channel
’1’: Clears EPEC channel settings into idle mode
’0’: no action
9:0
BYTE_CNT
Number of bytes to be transmitted
Data Sheet
99
2001-02-23
C165UTAH
DMA - External PEC (EPEC)
The EPEC Transmit Byte Length registers (x) provide the 10-bit Transmit bytes length of
the actual packet to be send to the USB endpoint#x and the related bit. Each EPEC
channel can be cleared by SW, if necessary.
All USB source and destination pointers will be used in either receive or transmit
direction, since the USB endpoint’s direction of data is SW-configurable as IN, OUT or
bidirectional (except endpoint 0) after USB device controller reset.
EPEC_INT_REG
Table 21
Reset Value: 0000H
EPEC_INT_REG Interrupt Register
Bit No. Name
Function
15:8
RxTxSTART
Rx / Tx Start
Signal ’1’ indicates channel has started transfering
data.
7:0
TXDONE_INTx (x=7..0)
TX packet transfer completed by EPEC
’1’: transfer complete
’0’: busy or idle
The EPEC interrupt register indicates the end of an TX-packet transfer for an USB
endpoint.
EPEC_INTMSK_REG
Table 22
Reset Value: 0000H
EPEC_INTMSK_REG Interrupt Register
Bit No. Name
Function
15:8
RxTxSTARTMSK
Rx / Tx Start Mask
’1’: masked
’0’: not masked
7:0
TXDONE_INTMSKx (x=7..0)
Mask interrupt TX packet transfer completed by
EPEC
’1’: masked
’0’: not masked
The EPEC interrupt mask register masks out the end of an TX-packet transfer interrupt
for an USB endpoint.
6.4
EPEC Transfer Example
The EPEC (external peripheral event controller, external in the sense that it is external
to the CPU block) controls the transfer of data between the USB block and the external
or internal RAM.
Data Sheet
100
2001-02-23
C165UTAH
DMA - External PEC (EPEC)
Note:
If the EPEC is not enabled, then no transfer is possible.
The sequence of operations is as follows:
1. The USB block will generate FIFO request signals as soon as the system reset was
deassert-ed. It thus signals to the EPEC that the USB FIFOs are ready to receive data.
2. Now the USB has to be configured. The EPEC channel that serves USB
endpoint_0_IN is setup with source and destination pointer, the EPEC control register
is programmed with the number of bytes that need to be transfered and the bit for
external/ internal source has to be set according to the application.
3. Now the EPEC channel for endpoint_0_IN can be activated by setting the enable bit
in EPEC control register for endpoint_0_IN.
4. The transfer of configuration data to the USB FIFO for endpoint_0_IN starts because
the USB FIFO has signaled that there is space available and the EPEC channel has
been enabled.
5. After the EPEC byte counter has reached the number of bytes that have to be
transfered, the EPEC channel disables itself and indicates the transmit end of data
condition in the EPEC interrupt register.
6. The indication of the transmit end of data condition triggers the generation of the
EPEC interrupt pulse to the CPU on the irq(40) line.
7. After the interrupt generation unit has generated the interrupt pulse it waits for a write
to the interrupt register. It then is ready to generate the next interrupt pulse to the CPU.
If no write to the interrupt register takes place, no new interrupt pulse can be asserted.
6.5
Implementation of EPEC Interrupt Generation Unit
Currently the EPEC interrupt controller implements a clear on write functionality.
This implies the following for the interrupt routine:
1. One of two conditions generates an entry into the EPEC interrupt register: either a
channel start or a channel transmit end of data (for endpoint 0 this functionality is
reduced to start for direction in, EPEC transmit; and end for direction out, EPEC
receive; otherwise we would have more than 16 interrupts). The generation of the
entries into the EPEC interrupt register can be controled by programming the EPEC
interrupt mask register.
2. The interrupt routine is triggered by the interrupt pulse that the EPEC interrupt
controller gernerates on irq(40).
3. The routine should then read the EPEC interrupt register to determine the source of
the interrupt.
4. The interrupt has to be acknowledged by writing a ‘1’ to the position of the interrupt
source in the EPEC interrupt register.
5. Now the interrupt controller is ready to generate the next interrupt pulse.
Data Sheet
101
2001-02-23
C165UTAH
Interrupt and Trap Functions
7
Interrupt and Trap Functions
The architecture of the C165UTAH supports several mechanisms for fast and flexible
response to service requests that can be generated from various sources internal or
external to the microcontroller. These mechanisms include:
Normal Interrupt Processing
The CPU temporarily suspends the current program execution and branches to an
interrupt service routine in order to service an interrupt requesting device. The current
program status (IP, PSW, in segmentation mode also CSP) is saved on the internal
system stack. A prioritization scheme with 16 priority levels allows the user to specify the
order in which multiple interrupt requests are to be handled.
Interrupt Processing via the Peripheral Event Controller (PEC)
A faster alternative to normal software controlled interrupt processing is servicing an
interrupt requesting device with the C165UTAH's integrated Peripheral Event Controller
(PEC). Triggered by an interrupt request, the PEC performs a single word or byte data
transfer between any two locations in segment 0 (data pages 0 through 3) through one
of eight programmable PEC Service Channels. During a PEC transfer the normal
program execution of the CPU is halted for just 1 instruction cycle. No internal program
status information needs to be saved. The same prioritization scheme is used for PEC
service as for normal interrupt processing. PEC transfers share the 2 highest priority
levels.
Trap Functions
Trap functions are activated in response to special conditions that occur during the
execution of instructions. A trap can also be caused externally by the Non-Maskable
Interrupt pin NMI. Several hardware trap functions are provided for handling erroneous
conditions and exceptions that arise during the execution of an instruction. Hardware
traps always have highest priority and cause immediate system reaction. The software
trap function is invoked by the TRAP instruction, which generates a software interrupt for
a specified interrupt vector. For all types of traps the current program status is saved on
the system stack.
External Interrupt Processing
Although the C165UTAH does not provide dedicated interrupt pins, it allows to connect
external interrupt sources and provides several mechanisms to react on external events,
including standard inputs, non-maskable interrupts and fast external interrupts. These
interrupt functions are alternate port functions, except for the non-maskable interrupt and
the reset input.
Data Sheet
102
2001-02-23
C165UTAH
Interrupt and Trap Functions
7.1
Interrupt System Structure
The C165UTAH provides up to 64 separate interrupt nodes that may be assigned to 16
priority levels. The 4 lowest nodes are reserved for the CPU - thus, up to 60 nodes are
available for all interrupts. In order to support modular and consistent software design
techniques, each source of an interrupt or PEC request is supplied with a separate
interrupt control register and interrupt vector. The control register contains the interrupt
request flag, the interrupt enable bit, and the interrupt priority of the associated source.
Each source request is activated by one specific event, depending on the selected
operating mode of the respective device. The only exceptions are the two serial channels
of the table, where an error interrupt request can be generated by different kinds of error,
and the two subnode interrupts controlled by the ISNC and CLISNC registers (see
Interrupt and PEC descriptions). However, specific status flags which identify the type of
error are implemented in the serial channels’ control registers.
The C165UTAH provides a vectored interrupt system. In this system specific vector
locations in the memory space are reserved for the reset, trap, and interrupt service
functions. Whenever a request occurs, the CPU branches to the location that is
associated with the respective interrupt source. This allows direct identification of the
source that caused the request. The only exceptions are the class B hardware traps,
which all share the same interrupt vector. The status flags in the Trap Flag Register
(TFR) can then be used to determine which exception caused the trap. For the special
software TRAP instruction, the vector address is specified by the operand field of the
instruction, which is a seven bit trap number.
The reserved vector locations build a jump table in the low end of the C165UTAH’s
address space (segment 0). The jump table is made up of the appropriate jump
instructions that transfer control to the interrupt or trap service routines, which may be
located anywhere within the address space. The entries of the jump table are located at
the lowest addresses in code segment 0 of the address space. Each entry occupies 2
words, except for the reset vector and the hardware trap vectors, which occupy 4 or 8
words.
The table below lists all sources that are capable of requesting interrupt or PEC service
in the C165UTAH, the associated interrupt vectors, their locations, their trap numbers
and the SFR addresses of associated interrupt control registers. It also lists the
mnemonics of the corresponding Interrupt Enable flags. The mnemonics are composed
of a part that specifies the respective source, followed by a part that specifies their
function (IE=Interrupt Enable flag). The same composition is used for the mnemonics of
according interrupt request flags (IR=Interrupt Request flag; example: CC0IR belongs to
interrupt source CC0INT) and for the names of according interrupt control registers
(IC=Interrupt Control; example: CC0IC) which are not included in Table 23.
Data Sheet
103
2001-02-23
C165UTAH
Interrupt and Trap Functions
Table 23
C165UTAH Interrupts and PEC Service Requests
Nr.
Source of
Interrupt or PEC
Service Request
Interrupt
Name
Enable
Flag
Vector
Trap
Location Number
SFR
hex
Addr
irq(0)
GPT Timer 2
T2INT
T2IE
00’0088H 22H / 34D
FF60
irq(1)
GPT Timer 3
T3INT
T3IE
00’008CH 23H / 35D
FF62
irq(2)
GPT Timer 4
T4INT
T4IE
00’0090H 24H / 36D
FF64
irq(3)
GPT Timer 5
T5INT
T5IE
00’0094H 25H / 37D
FF66
irq(4)
GPT Timer 6
T6INT
T6IE
00’0098H 26H / 38D
FF68
irq(5)
GPT CAPREL
Register
CRINT
CRIE
00’009CH 27H / 39D
FF6A
irq(6)
ASC Transmit
S0TINT
S0TIE
00’00A8H 2AH / 42D FF6C
irq(7)
ASC Receive
S0RINT
S0RIE
00’00ACH 2BH / 43D FF6E
irq(8)
ASC Error
S0EINT
S0EIE
00’00B0H 2CH / 44D FF70
irq(9)
ASC Transmit
Buffer
S0TBINT
S0TBIE
00’011CH 47H / 71D
irq(10) SSC Transmit
SSCTINT
SSCTIE
00’00B4H 2DH / 45D FF72
irq(11) SSC Receive
SSCRINT
SSCRIE
00’00B8H 2EH / 46D FF74
irq(12) SSC Error
SSCEINT
SSCEIE
00’00BCH 2FH / 47D FF76
irq(13) ASC Autobaud
Start
ABSTINT
ABSTIE
00’0118H 46H / 70D
F194
irq(14) ASC Autobaud End ABENDINT
ABENDIE
00’0114H 45H / 69D
F18C
irq(15) rRTC Interrupt
RTC_INT
RTCIE
00’0110H 44H / 68D
F184
irq(16) UDC SETUP
USETINT
USETIE
00’00F0H 3CH / 60D F178
irq(17) UDC Load Config
Done
ULCDINT
ULCDIE
00’00ECH 3BH / 59D F176
irq(18) UDC Suspend
USSINT
USSIE
00’00E8H 3AH / 58D F174
irq(19) UDC Suspend off
USSOINT
USSOIE
00’00E4H 39H / 57D
F172
irq(20) UDC Start of Frame USOFINT
USOFIE
00’00E0H 38H / 56D
F170
irq(21) UDC Config Val
UCFGVINT
UCFGVIE
00’00DCH 37H / 55D
F16E
irq(22) UDC TXWR
UTXRINT
UTXRIE
00’00D8H 36H / 54D
F16C
irq(23) UDC RXRR
URXRINT
URXRIE
00’00D4H 35H / 53D
F16A
irq(24) UDC TX Done7
UTD7INT
UTD7IE
00’00D0H 34H / 52D
F168
irq(25) UDC TX Done6
UTD6INT
UTD6IE
00’00CCH 33H / 51D
F166
Data Sheet
104
F19C
2001-02-23
C165UTAH
Interrupt and Trap Functions
Interrupt
Name
Enable
Flag
Vector
Trap
Location Number
SFR
hex
Addr
irq(26) UDC TX Done5
UTD5INT
UTD5IE
00’00C8H 32H / 50D
F164
irq(27) UDC TX Done4
UTD4INT
UTD4IE
00’00C4H 31H / 49D
F162
irq(28) UDC TX Done3
UTD3INT
UTD3IE
00’00C0H 30H / 48D
F160
irq(29) UDC TX Done2
UTD2INT
UTD2IE
00’005CH 17H / 23D
FF86
irq(30) UDC TX Done1
UTD1INT
UTD1IE
00’0058H 16H / 22D
FF84
irq(31) UDC TX Done0
UTD0INT
UTD0IE
00’0054H 15H / 21D
FF82
irq(32) UDC RX Done7
URD7INT
URD7IE
00’0050H 14H / 20D
FF80
irq(33) UDC RX Done6
URD6INT
URD6IE
00’004CH 13H / 19D
FF7E
irq(34) UDC RX Done5
URD5INT
URD5IE
00’0048H 12H / 18D
FF7C
irq(35) UDC RX Done4
URD4INT
URD4IE
00’0044H 11H / 17D
FF7A
irq(36) UDC RX Done3
URD3INT
URD3IE
00’0040H 10H / 16D
FF78
irq(37) UDC RX Done2
URD2INT
URD2IE
00’0080H 20H / 32D
FF9C
irq(38) UDC RX Done1
URD1INT
URD1IE
00’0084H 21H / 33D
FF9E
irq(39) UDC RX Done0
URD0INT
URD0IE
00’00F4H 3DH / 61D F17A
irq(40) EPEC
EPECINT
EPECIE
00’00F8H 3EH / 62D F17C
Nr.
Source of
Interrupt or PEC
Service Request
irq(41) reserved
irq(42) IOM-2 I/O
IOMIOINT
IOMIOIE
00’00A0H 28H / 40D
FF98
00’00A4H 29H / 41D
FF9A
irq(43) IOM-2 Channel0 TX IOMC0TINT IOMC0TIE 00’00FCH 3FH / 63D F17E
IOMC0RINT IOMC0RIE 00’0120H 48H / 72D
F182
irq(45) IOM-2 Channel1 TX IOMC1TINT IOMC1TIE 00’0124H 49H / 73D
F18A
irq(44) IOM-2 Channel0
RX
irq(46) IOM-2 Channel1
RX
IOMC1RINT IOMC1RIE 00’0128H 4AH / 74D F192
irq(47) reserved
00’012CH 4BH / 75D F19A
firq(0) Fast ext. Interrupt
EX0INT
EX0IE
00’0060H 18H / 24D
FF88
firq(1) Fast ext. Interrupt
EX1INT
EX1IE
00’0064H 19H / 25D
FF8A
firq(2) Fast ext. Interrupt
EX2INT
EX2IE
00’0068H 1AH / 26D FF8C
firq(3) Fast ext. Interrupt
EX3INT
EX3IE
00’006CH 1BH / 27D FF8E
firq(4) Fast ext. Interrupt
EX4INT
EX4IE
00’0070H 1CH / 28D FF90
firq(5) Fast ext. Interrupt
EX5INT
EX5IE
00’0074H 1DH / 29D FF92
Data Sheet
105
2001-02-23
C165UTAH
Interrupt and Trap Functions
Nr.
Source of
Interrupt or PEC
Service Request
Interrupt
Name
Enable
Flag
Vector
Trap
Location Number
firq(6) Fast ext. Interrupt
EX6INT
EX6IE
00’0078H 1EH / 30D FF94
firq(7) Fast ext. Interrupt
EX7INT
EX7IE
00’007CH 1FH / 31D FF96
xb(0)
UDC TXWR
UTXRINT
UTXRIE
00’0100H 40H / 64D
F186
xb(1)
EPEC
EPECINT
EPECIE
00’0104H 41H / 65D
F18E
xb(2)
IOM-2 IO
IOMIOINT
IOMIOIE
00’0108H 42H / 66D
F196
xb(3)
Internal PLL Lock / XP3INT
RTC
XP3IE
00’010CH 43H / 67D
F19E
CLISN Interrupt
CLISNIE
00’0130H 4CH / 76D FFA8
CLISNINT
SFR
hex
Addr
Note:
1. The X-Bus interrupts xb(0), xb(1) and xb(2), known from other C16x device’s, are
connected to the main interrupt node of the respective X-Bus peripheral: UTXRINT
(xb(0) and irq(22)), EPECINT (xb(1) and irq(40)) and IOMIOINT (xb(2) and irq(42)).
2. Each entry of the interrupt vector table provides space for two word instructions or one
doubleword instruction. The respective vector location results from multiplying the trap
number by 4 (4 bytes per entry).
3. One interrupt control register is provided for each interrupt node. All IC registers of the
C165UTAH can be found in the SFR list.
Table 24 lists the vector locations for hardware traps and the corresponding status flags
in register TFR. It also lists the priorities of trap service for cases, where more than one
trap condition might be detected within the same instruction. After any reset (hardware
reset, software reset instruction SRST, or reset by watchdog timer overflow) program
execution starts at the reset vector at location 00’0000H. Reset conditions have priority
over every other system activity and therefore have the highest priority (trap priority III).
Software traps may be initiated to any vector location between 00’0000H and 00’01FCH.
A service routine entered via a software TRAP instruction is always executed on the
current CPU priority level which is indicated in bit field ILVL in register PSW. This means
that routines entered via the software TRAP instruction can be interrupted by all
hardware traps or higher level interrupt requests.
Data Sheet
106
2001-02-23
C165UTAH
Interrupt and Trap Functions
Table 24
Hardware Traps and Vector Locations
Exception Condition
Trap
Flag
Trap
Vector
Vector
Location
Trap
Number
Trap
Priority
RESET
RESET
RESET
00’0000H
00’0000H
00’0000H
00H
00H
00H
III
III
III
NMITRAP
STOTRAP
STUTRAP
DEBTRAP
00’0008H
00’0010H
00’0018H
00’0020H
02H
04H
06H
08H
II
II
II
II
UNDOPC BTRAP
PRTFLT BTRAP
#
ILLOPA
BTRAP
00’0028H
00’0028H
0AH
0AH
I
I
00’0028H
0AH
I
ILLINA
ILLBUS
00’0028H
00’0028H
0AH
0AH
I
I
Reset Functions:
Hardware Reset
Software Reset
Watchdog Timer
Overflow
Class A Hardware Traps:
Non-Maskable Interrupt
Stack Overflow
Stack Underflow
Debug Trap
Class B Hardware Traps:
Undefined Opcode
Protected Instruction
Fault
Illegal Word Operand
Access
Illegal Instruction Access
Illegal External Bus
Access
NMI
STKOF
STKUF
DEBUG
BTRAP
BTRAP
Reserved
[2CH – 3CH] [0BH –
0FH]
Software Traps
TRAP Instruction
Any
Any
[00’0000H – [00H –
00’01FCH] 7FH]
in steps
of 4H
Current
CPU
Priority
Normal Interrupt Processing and PEC Service
During each instruction cycle one out of all sources which require PEC or interrupt
processing is selected according to its interrupt priority. This priority of interrupts and
PEC requests is programmable in two levels. Each requesting source can be assigned
to a specific priority. A second level (called “group priority”) allows to specify an internal
order for simultaneous requests from a group of different sources on the same priority
level. At the end of each instruction cycle the one source request with the highest current
priority will be determined by the interrupt system. This request will then be serviced, if
its priority is higher than the current CPU priority in register PSW.
Data Sheet
107
2001-02-23
C165UTAH
Interrupt and Trap Functions
Interrupt System Register Description
Interrupt processing is controlled globally by register PSW through a general interrupt
enable bit (IEN) and the CPU priority field (ILVL). Additionally the different interrupt
sources are controlled individually by their specific interrupt control registers (...IC).
Thus, the acceptance of requests by the CPU is determined by both the individual
interrupt control registers and the PSW. PEC services are controlled by the respective
PECCx register and the source and destination pointers, which specify the task of the
respective PEC service channel.
7.2
Interrupt Control Registers
All interrupt control registers are organized identically. The lower 8 bits of an interrupt
control register contain the complete interrupt status information of the associated
source, which is required during one round of prioritization, the upper 8 bits of the
respective register are reserved. All interrupt control registers are bit-addressable and all
bits can be read or written via software. This allows each interrupt source to be
programmed or modified with just one instruction. When accessing interrupt control
registers through instructions which operate on word data types, their upper 8 bits
(15...8) will return zeros, when read, and will discard written data.
The layout of the Interrupt Control registers shown below applies to each xxIC register,
where xx stands for the mnemonic for the respective source.
Data Sheet
108
2001-02-23
C165UTAH
Interrupt and Trap Functions
xxIC (yyyyH / zzH)
15
14
13
12
11
10
9
8
7
Reset Value: - - 00H
6
5
xxIR xxIE
-
-
Bit
-
-
-
-
-
-
rw
rw
4
3
2
1
0
ILVL
GLVL
rw
rw
Function
GLVL
Group Level
Defines the internal order for simultaneous requests of the same priority.
3: Highest group priority
0: Lowest group priority
ILVL
Interrupt Priority Level
Defines the priority level for the arbitration of requests.
FH: Highest priority level
0H: Lowest priority level
xxIE
Interrupt Enable Control Bit (individually enables/disables a specific
source)
‘0’: Interrupt request is disabled
‘1’: Interrupt Request is enabled
xxIR
Interrupt Request Flag
‘0’: No request pending
‘1’: This source has raised an interrupt request
The Interrupt Request Flag is set by hardware whenever a service request from the
respective source occurs. It is cleared automatically upon entry into the interrupt service
routine or upon a PEC service. In the case of PEC service the Interrupt Request flag
remains set, if the COUNT field in register PECCx of the selected PEC channel
decrements to zero. This allows a normal CPU interrupt to respond to a completed PEC
block transfer.
Note: Modifying the Interrupt Request flag via software causes the same effects as if it
had been set or cleared by hardware.
Interrupt Priority Level and Group Level
The four bits of bit field ILVL specify the priority level of a service request for the
arbitration of simultaneous requests. The priority increases with the numerical value of
ILVL, so 0000B is the lowest and 1111B is the highest priority level.
When more than one interrupt request on a specific level gets active at the same time,
the values in the respective bit fields GLVL are used for second level arbitration to select
Data Sheet
109
2001-02-23
C165UTAH
Interrupt and Trap Functions
one request for being serviced. Again the group priority increases with the numerical
value of GLVL, so 00B is the lowest and 11B is the highest group priority.
Note: All interrupt request sources that are enabled and programmed to the same
priority level must always be programmed to different group priorities. Otherwise
an incorrect interrupt vector will be generated.
Upon entry into the interrupt service routine, the priority level of the source that won the
arbitration and who’s priority level is higher than the current CPU level, is copied into bit
field ILVL of register PSW after pushing the old PSW contents on the stack.
The interrupt system of the C165UTAH allows nesting of up to 15 interrupt service
routines of different priority levels (level 0 cannot be arbitrated).
Interrupt requests that are programmed to priority levels 15 or 14 (ie, ILVL=111XB) will
be serviced by the PEC, unless the COUNT field of the associated PECC register
contains zero. In this case the request will instead be serviced by normal interrupt
processing. Interrupt requests that are programmed to priority levels 13 through 1 will
always be serviced by normal interrupt processing.
Note: Priority level 0000B is the default level of the CPU. Therefore a request on level 0
will never be serviced, because it can never interrupt the CPU. However, an
enabled interrupt request on level 0000B will terminate the C165UTAH’s Idle mode
and reactivate the CPU.
For interrupt requests which are to be serviced by the PEC, the associated PEC channel
number is derived from the respective ILVL (LSB) and GLVL (see figure below). So
programming a source to priority level 15 (ILVL=1111B) selects the PEC channel group
7...4, programming a source to priority level 14 (ILVL=1110B) selects the PEC channel
group 3...0. The actual PEC channel number is then determined by the group priority
field GLVL.
Interrupt
Control Register
ILVL
PEC Control
GLVL
PEC Channel
MCD02006
Figure 21
Data Sheet
Priority Levels and PEC Channels
110
2001-02-23
C165UTAH
Interrupt and Trap Functions
Simultaneous requests for PEC channels are prioritized according to the PEC channel
number, where channel 0 has lowest and channel 7 has highest priority.
Note: All sources that request PEC service must be programmed to different PEC
channels. Otherwise an incorrect PEC channel may be activated.
The table below shows in a few examples, which action is executed with a given
programming of an interrupt control register.
Table 25
Programming Example
Priority Level
Type of Service
ILVL
GLVL
COUNT = 00H
COUNT ≠ 00H
1111
11
CPU interrupt,
level 15, group priority 3
PEC service,
channel 7
1111
10
CPU interrupt,
level 15, group priority 2
PEC service,
channel 6
1110
10
CPU interrupt,
level 14, group priority 2
PEC service,
channel 2
1101
10
CPU interrupt,
level 13, group priority 2
CPU interrupt,
level 13, group priority 2
0001
11
CPU interrupt,
level 1, group priority 3
CPU interrupt,
level 1, group priority 3
0001
00
CPU interrupt,
level 1, group priority 0
CPU interrupt,
level 1, group priority 0
0000
XX
No service!
No service!
Note: All requests on levels 13...1 cannot initiate PEC transfers. They are always
serviced by an interrupt service routine. No PECC register is associated and no
COUNT field is checked.
Interrupt Control Functions in the PSW
The Processor Status Word (PSW) is functionally divided into 2 parts: the lower byte of
the PSW basically represents the arithmetic status of the CPU, the upper byte of the
PSW controls the interrupt system of the C165UTAH and the arbitration mechanism for
the external bus interface.
Note: Pipeline effects have to be considered when enabling/disabling interrupt requests
via modifications of register PSW (see chapter “The Central Processing Unit”).
Data Sheet
111
2001-02-23
C165UTAH
Interrupt and Trap Functions
PSW (FF10H / 88H)
15
14
13
SFR
12
Reset Value: 0000H
11
10
9
8
7
6
5
4
3
2
1
0
ILVL
IEN
HLD
EN
-
-
-
USR0
MUL
IP
E
Z
V
C
N
rw
rw
rw
-
-
-
rw
rw
rw
rw
rw
rw
rw
Bit
Function
N, C, V, Z, E,
MULIP, USR0
CPU status flags (Described in section “The Central Processing Unit”)
Define the current status of the CPU (ALU, multiplication unit).
HLDEN
HOLD Enable (Enables External Bus Arbitration)
0: Bus arbitration disabled, P6.7...P6.5 may be used for general purpose I/O
1: Bus arbitration enabled, P6.7...P6.5 serve as BREQ, HLDA, HOLD, resp.
ILVL
CPU Priority Level
Defines the current priority level for the CPU
FH: Highest priority level
0H: Lowest priority level
IEN
Interrupt Enable Control Bit (globally enables/disables interrupt requests)
‘0’: Interrupt requests are disabled
‘1’: Interrupt requests are enabled
CPU Priority ILVL defines the current level for the operation of the CPU. This bit field
reflects the priority level of the routine that is currently executed. Upon entry into an
interrupt service routine this bit field is updated with the priority level of the request that
is being serviced. The PSW is saved on the system stack before. The CPU level
determines the minimum interrupt priority level that will be serviced. Any request on the
same or a lower level will not be acknowledged.
The current CPU priority level may be adjusted via software to control which interrupt
request sources will be acknowledged.
PEC transfers do not really interrupt the CPU, but rather “steal” a single cycle, so PEC
services do not influence the ILVL field in the PSW.
Hardware traps switch the CPU level to maximum priority (ie. 15) so no interrupt or PEC
requests will be acknowledged while an exception trap service routine is executed.
Note: The TRAP instruction does not change the CPU level, so software invoked trap
service routines may be interrupted by higher requests.
Interrupt Enable bit IEN globally enables or disables PEC operation and the
acceptance of interrupts by the CPU. When IEN is cleared, no interrupt requests are
accepted by the CPU. When IEN is set to '1', all interrupt sources, which have been
individually enabled by the interrupt enable bits in their associated control registers, are
globally enabled.
Note: Traps are non-maskable and are therefore not affected by the IEN bit.
Data Sheet
112
2001-02-23
C165UTAH
Interrupt and Trap Functions
7.3
Operation of the PEC Channels
The C165UTAH's Peripheral Event Controller (PEC) provides 8 PEC service channels,
which move a single byte or word. This is the fastest possible interrupt response and in
many cases is sufficient to service the respective peripheral request (eg. serial channels,
etc.). Each channel is controlled by a dedicated PEC Channel Counter/Control register
(PECCx) and a pair of pointers for source (SRCPx) and destination (DSTPx) of the data
transfer.
The PECC registers control the action that is performed by the respective PEC channel.
Note: For the PECCx register description, please also refer to page 87 of SubChapter "Extended PEC Channel Control".
Byte/Word Transfer bit BWT controls, if a byte or a word is moved during a PEC service
cycle. This selection controls the transferred data size and the increment step for the
modified pointer.
Increment Control Field INC controls, if one of the PEC pointers is incremented after
the PEC transfer. It is not possible to increment both pointers, however. If the pointers
are not modified (INC=’00’), the respective channel will always move data from the same
source to the same destination.
Note: The reserved combination ‘11’ is changed to ‘10’ by hardware. However, it is not
recommended to use this combination.
The PEC Transfer Count Field COUNT controls the action of a respective PEC channel,
where the content of bit field COUNT at the time the request is activated selects the
action. COUNT may allow a specified number of PEC transfers, unlimited transfers or no
PEC service at all.
The table below summarizes, how the COUNT field itself, the interrupt requests flag IR
and the PEC channel action depends on the previous content of COUNT.
Previous Modified
COUNT
COUNT
IR after PEC Action of PEC Channel
service
and Comments
FFH
FFH
‘0’
Move a Byte / Word
Continuous transfer mode, ie. COUNT is not modified
FEH..02H
FDH..01H
‘0’
Move a Byte / Word and decrement COUNT
01H
00H
‘1’
Move a Byte / Word
Leave request flag set, which triggers another request
00H
00H
(‘1’)
No action!
Activate interrupt service routine rather than PEC channel.
Data Sheet
113
2001-02-23
C165UTAH
Interrupt and Trap Functions
The PEC transfer counter allows to service a specified number of requests by the
respective PEC channel, and then (when COUNT reaches 00H) activate the interrupt
service routine, which is associated with the priority level. After each PEC transfer the
COUNT field is decremented and the request flag is cleared to indicate that the request
has been serviced.
Continuous transfers are selected by the value FFH in bit field COUNT. In this case
COUNT is not modified and the respective PEC channel services any request until it is
disabled again.
When COUNT is decremented from 01H to 00H after a transfer, the request flag is not
cleared, which generates another request from the same source. When COUNT already
contains the value 00H, the respective PEC channel remains idle and the associated
interrupt service routine is activated instead. This allows to choose, if a level 15 or 14
request is to be serviced by the PEC or by the interrupt service routine.
Note: PEC transfers are only executed, if their priority level is higher than the CPU level,
ie. only PEC channels 7...4 are processed, while the CPU executes on level 14.
All interrupt request sources that are enabled and programmed for PEC service
should use different channels. Otherwise only one transfer will be performed for
all simultaneous requests. When COUNT is decremented to 00H, and the CPU is
to be interrupted, an incorrect interrupt vector will be generated.
The source and destination pointers specifiy the locations between which the data is
to be moved. A pair of pointers (SRCPx and DSTPx) is associated with each of the 8
PEC channels. These pointers do not reside in specific SFRs, but are mapped into the
internal RAM of the C165UTAH just below the bit-addressable area (see figure below).
Data Sheet
114
2001-02-23
C165UTAH
Interrupt and Trap Functions
Figure 22
DSTP7
00’FCFEH
DSTP3
00’FCEEH
SRCP7
00’FCFCH
SRCP3
00’FCECH
DSTP6
00’FCFAH
DSTP2
00’FCEAH
SRCP6
00’FCF8H
SRCP2
00’FCE8H
DSTP5
00’FCF6H
DSTP1
00’FCE6H
SRCP5
00’FCF4H
SRCP1
00’FCE4H
DSTP4
00’FCF2H
DSTP0
00’FCE2H
SRCP4
00’FCF0H
SRCP0
00’FCE0H
Mapping of PEC Pointers into the Internal RAM
PEC data transfers do not use the data page pointers DPP3...DPP0, see also
Chapter 5.5, "PEC - Extension of Functionality". The PEC source and destination
pointers are used as 16-bit intra-segment addresses within segment 0, so data can be
transferred between any two locations within the first four data pages 3...0.
The pointer locations for inactive PEC channels may be used for general data storage.
Only the required pointers occupy RAM locations.
Note: If word data transfer is selected for a specific PEC channel (ie. BWT=’0’), the
respective source and destination pointers must both contain a valid word address
which points to an even byte boundary. Otherwise the Illegal Word Access trap will
be invoked, when this channel is used.
7.4
Prioritization of Interrupt and PEC Service Requests
Interrupt and PEC service requests from all sources can be enabled, so they are
arbitrated and serviced (if they win), or they may be disabled, so their requests are
disregarded and not serviced.
Enabling and disabling interrupt requests may be done via three mechanisms:
Control Bits allow to switch each individual source “ON” or “OFF”, so it may generate a
request or not. The control bits (xxIE) are located in the respective interrupt control
registers. All interrupt requests may be enabled or disabled generally via bit IEN in
register PSW. This control bit is the “main switch” that selects, if requests from any
source are accepted or not.
Data Sheet
115
2001-02-23
C165UTAH
Interrupt and Trap Functions
For a specific request to be arbitrated the respective source’s enable bit and the global
enable bit must both be set.
The Priority Level automatically selects a certain group of interrupt requests that will be
acknowledged, disclosing all other requests. The priority level of the source that won the
arbitration is compared against the CPU’s current level and the source is only serviced,
if its level is higher than the current CPU level. Changing the CPU level to a specific value
via software blocks all requests on the same or a lower level. An interrupt source that is
assigned to level 0 will be disabled and never be serviced.
The ATOMIC and EXTend instructions automatically disable all interrupt requests for
the duration of the following 1...4 instructions. This is useful eg. for semaphore handling
and does not require to re-enable the interrupt system after the unseparable instruction
sequence (see chapter “System Programming”).
Interrupt Class Management
An interrupt class covers a set of interrupt sources with the same importance, ie. the
same priority from the system’s viewpoint. Interrupts of the same class must not interrupt
each other. The C165UTAH supports this function with two features:
Classes with up to 4 members can be established by using the same interrupt priority
(ILVL) and assigning a dedicated group level (GLVL) to each member. This functionality
is built-in and handled automatically by the interrupt controller.
Classes with more than 4 members can be established by using a number of adjacent
interrupt priorities (ILVL) and the respective group levels (4 per ILVL). Each interrupt
service routine within this class sets the CPU level to the highest interrupt priority within
the class. All requests from the same or any lower level are blocked now, ie. no request
of this class will be accepted.
The example below establishes 3 interrupt classes which cover 2 or 3 interrupt priorities,
depending on the number of members in a class. A level 6 interrupt disables all other
sources in class 2 by changing the current CPU level to 8, which is the highest priority
(ILVL) in class 2. Class 1 requests or PEC requests are still serviced in this case.
The 19 interrupt sources (excluding PEC requests) are so assigned to 3 classes of
priority rather than to 7 different levels, as the hardware support would do.
Data Sheet
116
2001-02-23
C165UTAH
Interrupt and Trap Functions
Table 26
Software controlled Interrupt Classes (Example)
ILVL
(Priority)
GLVL
3
2
1
Interpretation
0
15
PEC service on up to 8 channels
14
13
12
X
11
X
X
X
X
Interrupt Class 1
5 sources on 2 levels
Interrupt Class 2
9 sources on 3 levels
10
9
8
X
X
X
X
7
X
X
X
X
6
X
5
X
X
X
X
4
X
Interrupt Class 3
5 sources on 2 levels
3
2
1
0
Data Sheet
No service!
117
2001-02-23
C165UTAH
Interrupt and Trap Functions
7.5
Saving the Status during Interrupt Service
Before an interrupt request that has been arbitrated is actually serviced, the status of the
current task is automatically saved on the system stack. The CPU status (PSW) is saved
along with the location, where the execution of the interrupted task is to be resumed after
returning from the service routine. This return location is specified through the Instruction
Pointer (IP) and, in case of a segmented memory model, the Code Segment Pointer
(CSP). Bit SGTDIS in register SYSCON controls, how the return location is stored.
The system stack receives the PSW first, followed by the IP (unsegmented) or followed
by CSP and then IP (segmented mode). This optimizes the usage of the system stack,
if segmentation is disabled.
The CPU priority field (ILVL in PSW) is updated with the priority of the interrupt request
that is to be serviced, so the CPU now executes on the new level. If a multiplication or
division was in progress at the time the interrupt request was acknowledged, bit MULIP
in register PSW is set to ‘1’. In this case the return location that is saved on the stack is
not the next instruction in the instruction flow, but rather the multiply or divide instruction
itself, as this instruction has been interrupted and will be completed after returning from
the service routine.
High
Addresses
Status of
Interrupted
Task
SP
---
SP
--
PSW
PSW
IP
CSP
--
IP
SP
Low
Addresses
a) System Stack before
Interrupt Entry
b) System Stack after
Interrupt Entry
(Unsegmented)
c) System Stack after
Interrupt Entry
(Segmented)
MCD02226
Figure 23
Task Status saved on the System Stack
The interrupt request flag of the source that is being serviced is cleared. The IP is loaded
with the vector associated with the requesting source (the CSP is cleared in case of
segmentation) and the first instruction of the service routine is fetched from the
respective vector location, which is expected to branch to the service routine itself. The
data page pointers and the context pointer are not affected.
Data Sheet
118
2001-02-23
C165UTAH
Interrupt and Trap Functions
When the interrupt service routine is left (RETI is executed), the status information is
popped from the system stack in the reverse order, taking into account the value of bit
SGTDIS.
Context Switching
An interrupt service routine usually saves all the registers it uses on the stack, and
restores them before returning. The more registers a routine uses, the more time is
wasted with saving and restoring. The C165UTAH allows to switch the complete bank of
CPU registers (GPRs) with a single instruction, so the service routine executes within its
own, separate context.
The instruction “SCXT CP, #New_Bank” pushes the content of the context pointer (CP)
on the system stack and loads CP with the immediate value “New_Bank”, which selects
a new register bank. The service routine may now use its “own registers”. This register
bank is preserved, when the service routine terminates, ie. its contents are available on
the next call.
Before returning (RETI) the previous CP is simply POPped from the system stack, which
returns the registers to the original bank.
Note: The first instruction following the SCXT instruction must not use a GPR.
Resources that are used by the interrupting program must eventually be saved and
restored, eg. the DPPs and the registers of the MUL/DIV unit.
7.6
Interrupt Response Times
The interrupt response time defines the time from an interrupt request flag of an enabled
interrupt source being set until the first instruction (I1) being fetched from the interrupt
vector location. The basic interrupt response time for the C165UTAH is 3 instruction
cycles.
Data Sheet
119
2001-02-23
C165UTAH
Interrupt and Trap Functions
Pipeline Stage
Cycle 1
Cycle 2
Cycle 3
Cycle 4
FETCH
N
N+1
N+2
I1
DECODE
N-1
N
TRAP (1)
TRAP (2)
EXECUTE
N-2
N-1
N
TRAP
WRITEBACK
N-3
N-2
N-1
N
IR-Flag
1
0
Interrupt Response Time
Figure 24
Pipeline Diagram for Interrupt Response Time
All instructions in the pipeline including instruction N (during which the interrupt request
flag is set) are completed before entering the service routine. The actual execution time
for these instructions (eg. waitstates) therefore influences the interrupt response time.
In the figure above the respective interrupt request flag is set in cycle 1 (fetching of
instruction N). The indicated source wins the prioritization round (during cycle 2). In cycle
3 a TRAP instruction is injected into the decode stage of the pipeline, replacing
instruction N+1 and clearing the source's interrupt request flag to '0'. Cycle 4 completes
the injected TRAP instruction (save PSW, IP and CSP, if segmented mode) and fetches
the first instruction (I1) from the respective vector location.
All instructions that entered the pipeline after setting of the interrupt request flag (N+1,
N+2) will be executed after returning from the interrupt service routine.
The minimum interrupt response time is 5 states (10 TCL). This requires program
execution from the internal code memory, no external operand read requests and setting
the interrupt request flag during the last state of an instruction cycle. When the interrupt
request flag is set during the first state of an instruction cycle, the minimum interrupt
response time under these conditions is 6 state times (12 TCL).
The interrupt response time is increased by all delays of the instructions in the pipeline
that are executed before entering the service routine (including N).
• When internal hold conditions between instruction pairs N-2/N-1 or N-1/N occur, or
instruction N explicitly writes to the PSW or the SP, the minimum interrupt response
time may be extended by 1 state time for each of these conditions.
• When instruction N reads an operand from the internal code memory, or when N is a
call, return, trap, or MOV Rn, [Rm+ #data16] instruction, the minimum interrupt
Data Sheet
120
2001-02-23
C165UTAH
Interrupt and Trap Functions
response time may additionally be extended by 2 state times during internal code
memory program execution.
• In case instruction N reads the PSW and instruction N-1 has an effect on the condition
flags, the interrupt response time may additionally be extended by 2 state times.
The worst case interrupt response time during internal code memory program execution
adds to 12 state times (24 TCL).
Any reference to external locations increases the interrupt response time due to pipeline
related access priorities. The following conditions have to be considered:
• Instruction fetch from an external location
• Operand read from an external location
• Result write-back to an external location
Depending on where the instructions, source and destination operands are located,
there are a number of combinations. Note, however, that only access conflicts contribute
to the delay.
A few examples illustrate these delays:
• The worst case interrupt response time including external accesses will occur, when
instructions N, N+1 and N+2 are executed out of external memory, instructions N-1
and N require external operand read accesses, instructions N-3 through N write back
external operands, and the interrupt vector also points to an external location. In this
case the interrupt response time is the time to perform 9 word bus accesses, because
instruction I1 cannot be fetched via the external bus until all write, fetch and read
requests of preceding instructions in the pipeline are terminated.
• When the above example has the interrupt vector pointing into the internal code
memory, the interrupt response time is 7 word bus accesses plus 2 states, because
fetching of instruction I1 from internal code memory can start earlier.
• When instructions N, N+1 and N+2 are executed out of external memory and the
interrupt vector also points to an external location, but all operands for instructions N3 through N are in internal memory, then the interrupt response time is the time to
perform 3 word bus accesses.
• When the above example has the interrupt vector pointing into the internal code
memory, the interrupt response time is 1 word bus access plus 4 states.
After an interrupt service routine has been terminated by executing the RETI instruction,
and if further interrupts are pending, the next interrupt service routine will not be entered
until at least two instruction cycles have been executed of the program that was
interrupted. In most cases two instructions will be executed during this time. Only one
instruction will typically be executed, if the first instruction following the RETI instruction
is a branch instruction (without cache hit), or if it reads an operand from internal code
memory, or if it is executed out of the internal RAM.
Note: A bus access in this context includes all delays which can occur during an external
bus cycle.
Data Sheet
121
2001-02-23
C165UTAH
Interrupt and Trap Functions
7.7
PEC Response Times
The PEC response time defines the time from an interrupt request flag of an enabled
interrupt source being set until the PEC data transfer being started. The basic PEC
response time for the C165UTAH is 2 instruction cycles.
Pipeline Stage
Cycle 1
Cycle 2
Cycle 3
Cycle 4
FETCH
N
N+1
N+2
N+2
DECODE
N-1
N
PEC
N+1
EXECUTE
N-2
N-1
N
PEC
WRITEBACK
N-3
N-2
N-1
N
IR-Flag
1
0
PEC Response Time
Figure 25
Pipeline Diagram for PEC Response Time
In Figure 25 the respective interrupt request flag is set in cycle 1 (fetching of instruction
N). The indicated source wins the prioritization round (during cycle 2). In cycle 3 a PEC
transfer “instruction” is injected into the decode stage of the pipeline, suspending
instruction N+1 and clearing the source's interrupt request flag to '0'. Cycle 4 completes
the injected PEC transfer and resumes the execution of instruction N+1.
All instructions that entered the pipeline after setting of the interrupt request flag (N+1,
N+2) will be executed after the PEC data transfer.
Note: When instruction N reads any of the PEC control registers PECC7...PECC0, while
a PEC request wins the current round of prioritization, this round is repeated and
the PEC data transfer is started one cycle later.
The minimum PEC response time is 3 states (6 TCL). This requires program execution
from the internal code memory, no external operand read requests and setting the
interrupt request flag during the last state of an instruction cycle. When the interrupt
request flag is set during the first state of an instruction cycle, the minimum PEC
response time under these conditions is 4 state times (8 TCL).
The PEC response time is increased by all delays of the instructions in the pipeline that
are executed before starting the data transfer (including N).
Data Sheet
122
2001-02-23
C165UTAH
Interrupt and Trap Functions
• When internal hold conditions between instruction pairs N-2/N-1 or N-1/N occur, the
minimum PEC response time may be extended by 1 state time for each of these
conditions.
• When instruction N reads an operand from the internal code memory, or when N is a
call, return, trap, or MOV Rn, [Rm+ #data16] instruction, the minimum PEC response
time may additionally be extended by 2 state times during internal code memory
program execution.
• In case instruction N reads the PSW and instruction N-1 has an effect on the condition
flags, the PEC response time may additionally be extended by 2 state times.
• The worst case PEC response time during internal code memory program execution
adds to 9 state times (18 TCL).
Any reference to external locations increases the PEC response time due to pipeline
related access priorities. The following conditions have to be considered:
• Instruction fetch from an external location
• Operand read from an external location
• Result write-back to an external location
Depending on where the instructions, source and destination operands are located,
there are a number of combinations. Note, however, that only access conflicts contribute
to the delay.
A few examples illustrate these delays:
• The worst case interrupt response time including external accesses will occur, when
instructions N and N+1 are executed out of external memory, instructions N-1 and N
require external operand read accesses and instructions N-3, N-2 and N-1 write back
external operands. In this case the PEC response time is the time to perform 7 word
bus accesses.
• When instructions N and N+1 are executed out of external memory, but all operands
for instructions N-3 through N-1 are in internal memory, then the PEC response time
is the time to perform 1 word bus access plus 2 state times.
Once a request for PEC service has been acknowledged by the CPU, the execution of
the next instruction is delayed by 2 state times plus the additional time it might take to
fetch the source operand from internal code memory or external memory and to write the
destination operand over the external bus in an external program environment.
Note: A bus access in this context includes all delays which can occur during an external
bus cycle.
Data Sheet
123
2001-02-23
C165UTAH
Interrupt and Trap Functions
7.8
External Interrupts
Although the C165UTAH has no dedicated INTR input pins, it provides many possibilities
to react on external asynchronous events by using a number of I/O lines for interrupt
input. The interrupt function may either be combined with the pin’s main function or may
be used instead of it, ie. if the main pin function is not required.
Interrupt signals may be connected to:
• EX7IN...EX0IN, the fast external interrupt input pins,
• T4IN, T2IN, the timer input pins,
For each of these pins either a positive, a negative, or both a positive and a negative
external transition can be selected to cause an interrupt or PEC service request. The
edge selection is performed in the control register of the peripheral device associated
with the respective port pin. The peripheral must be programmed to a specific operating
mode to allow generation of an interrupt by the external signal. The priority of the
interrupt request is determined by the interrupt control register of the respective
peripheral interrupt source, and the interrupt vector of this source will be used to service
the external interrupt request.
Note: In order to use any of the listed pins as external interrupt input, it must be switched
to input mode via its direction control bit DPx.y in the respective port direction
control register DPx.
Table 27
Pins to be used as External Interrupt Inputs
Port Pin
Original Function
Control Register
P2.7-0/EX7-0IN
Fast external interrupt input pin
EXICON
P3.7/T2IN
Auxiliary timer T2 input pin
T2CON
P3.5/T4IN
Auxiliary timer T4 input pin
T4CON
Pins T2IN or T4IN can be used as external interrupt input pins when the associated
auxiliary timer T2 or T4 in block GPT1 is configured for capture mode. This mode is
selected by programming the mode control fields T2M or T4M in control registers
T2CON or T4CON to 101B. The active edge of the external input signal is determined by
bit fields T2I or T4I. When these fields are programmed to X01B, interrupt request flags
T2IR or T4IR in registers T2IC or T4IC will be set on a positive external transition at pins
T2IN or T4IN, respectively. When T2I or T4I are programmed to X10B, then a negative
external transition will set the corresponding request flag. When T2I or T4I are
programmed to X11B, both a positive and a negative transition will set the request flag.
In all three cases, the contents of the core timer T3 will be captured into the auxiliary
timer registers T2 or T4 based on the transition at pins T2IN or T4IN. When the interrupt
Data Sheet
124
2001-02-23
C165UTAH
Interrupt and Trap Functions
enable bits T2IE or T4IE are set, a PEC request or an interrupt request for vector T2INT
or T4INT will be generated.
Note: The non-maskable interrupt input pin NMI and the reset input RSTIN provide
another possibility for the CPU to react on an external input signal. NMI and RSTIN
are dedicated input pins, which cause hardware traps.
7.8.1
Fast External Interrupts
The input pins that may be used for external interrupts are sampled every 16 TCL, ie.
external events are scanned and detected in timeframes of 16 TCL. The C165UTAH
provides 8 external interrupt inputs that are sampled every 2 TCL, so external events are
captured faster than with standard interrupt inputs.
The pins of Port 2 (P2.7...P2.0) can individually be programmed to this fast interrupt
mode, where also the trigger transition (rising, falling or both) can be selected. The
External Interrupt Control register EXICON controls this feature for all 8 pins.
EXICON (F1C0H / E0H)
15
14
13
12
ESFR
11
10
9
8
Reset Value: 0000H
7
6
5
4
3
2
1
0
EXI7ES
EXI6ES
EXI5ES
EXI4ES
EXI3ES
EXI2ES
EXI1ES
EXI0ES
rw
rw
rw
rw
rw
rw
rw
rw
Bit
Function
EXIxES
External Interrupt x Edge Selection Field (x=7...0)
0 0: Fast external interrupts disabled: standard mode
0 1: Interrupt on positive edge (rising)
1 0: Interrupt on negative edge (falling)
1 1: Interrupt on any edge (rising or falling)
Note:
1. The fast external interrupt inputs are sampled every 2 TCL. The interrupt request
arbitration and processing, however, is executed every 8 TCL.
2. In Sleep mode, no clock is available. Therefore sampling is performed with
asynchronous structures.
3. In Sleep mode fast external interrupts as well as the NMI input are controlled for spike
suppression in the System Control Block. Input signals shorter than 10 ns are
suppressed, detection is guaranteed for minimum 150 ns input signals.
Data Sheet
125
2001-02-23
C165UTAH
Interrupt and Trap Functions
7.8.2
External Interrupt Source Control
Fast external interrupts may also have interrupt sources selected from other peripherals.
This function is very advantageous in Slow Down mode or in Sleep mode, if for example
the SSC interface shall be used to wake-up the system. The register EXISEL is used to
switch the receive inputs of the serial interfaces to the fast external interrupts, in order to
detect incomming messages in case of disabled serial interface modules.
The EXISEL register is defined as follows:
EXISEL (F1DAH / EDH)
15
14
13
12
ESFR-bReset Value: 0000H
11
10
9
8
7
6
5
4
3
2
1
0
01
EXI6SS
EXI5SS
EXI4SS
EXI3SS
EXI2SS
00
00
rw
rw
rw
rw
rw
rw
rw
rw
Bit
Function
EXI0SS
0 0: Input from default pin. Must be set to ’00’.
0 1: Not allowed.
1 0: Not allowed.
1 1: Not allowed.
EXI1SS
0 0: Input from default pin.Must be set to ’00’.
0 1: Not allowed.
1 0: Not allowed.
1 1: Not allowed.
EXI2SS
0 0: Input from default pin.
0 1: Input from alternate source ASC_RxD @ P3.11.
1 0: Input from default pin ORed with alternate source ASC_RxD @ P3.11.
1 1: Input from default pin ANDed with alternate source ASC_RxD @ P3.11.
EXI3SS
0 0: Input from default pin.
0 1: Input from alternate source SSC_RxD @ P3.9.
1 0: Input from default pin ORed with alternate source SSC_RxD @ P3.9.
1 1: Input from default pin ANDed with alternate source SSC_RxD @ P3.9.
EXI4SS
0 0: Input from default pin.
0 1: Input from alternate source SSC_SCLK @ P3.13.
1 0: Input from default pin ORed with alternate source SSC_SCLK @ P3.13.
1 1: Input from default pin ANDed with alternate source SSC_SCLK @ P3.13.
EXI5SS
0 0: Input from default pin.
0 1: Input from alternate source USB_suspend interrupt.
1 0: Input from default pin ORed with alternate source USB_suspend interrupt.
1 1: Input from default pin ANDed with alternate source USB_suspend interrupt.
Data Sheet
126
2001-02-23
C165UTAH
Interrupt and Trap Functions
Bit
Function
EXI6SS
0 0: Input from default pin.
0 1: Input from alternate source IOM-2 @ DCL.
1 0: Input from default pin ORed with alternate source IOM-2 @ DCL.
1 1: Input from default pin ANDed with alternate source IOM-2 @ DCL.
EXI7SS
0 0: Not allowed.
0 1: Input from source RTC_INT.
1 0: Not allowed.
1 1: Not allowed.
7.8.3
Interrupt Subnode Control
The RTC (Real Time Clock) interrupt T14INT and the PLL/OWD interrupt share one
interrupt node, the XPER3 interrupt node. In order to enable the interrupt handler to
determine the source of that shared interrupt request, the subnode interrupt control
register ISNC is provided. The separate interrupt request and enable flags of register
ISNC (see below) for the PLL (PLLIR, PLLIE) as well as for the RTC (T14IR, T14IE) are
used as shown in Figure 26.
ISNC
T14INT
PLLINT
Diff.
Circuit
Diff.
Circuit
T14
IR1)
&
T14
IE
1
PLL
IR1)
Pulse
Generation
INT
&
PLL
IE
Note: The Interrupt Service Routine must clear the IR flag in register ISNC manually.
Otherwise no further interrupts can be detected.
All request flags are bit protected.
Figure 26
Data Sheet
Interrupt Subnode Control for PLL / RTC Interrupts
127
2001-02-23
C165UTAH
Interrupt and Trap Functions
The ISNC register is defined as follows:
ISNC (F1DEH / EFH)
15
14
13
ESFR-bReset Value : 0000H
12
11
10
9
8
7
6
5
Bit
Function
T14IR
T14 Overflow Interrupt Request Flag
‘0’: No request pending
‘1’: This source has raised an interrupt request
T14IE
T14 Overflow Interrupt Enable Control Bit
‘0’: Interrupt request is disabled
‘1’: Interrupt request is enabled
PLLIR
PLL Interrupt Request Flag
‘0’: No request pending
‘1’: This source has raised an interrupt request
PLLIE
PLL Interrupt Enable Control Bit
‘0’: Interrupt request is disabled
‘1’: Interrupt request is enabled
7.8.4
The Interrupt Control Register
4
3
2
1
0
PLL
IE
PLL
IR
RTC
T14
IE
RTC
T14
IR
The interrupt control registers listed below (FEI7IC..FEI0IC) control the fast external
interrupts of the C165UTAH.
Data Sheet
128
2001-02-23
C165UTAH
Interrupt and Trap Functions
FEIxIC (See Table)
15
14
13
SFR
12
11
10
9
8
Reset Value: - - 00HH
7
6
FEIx FEIx
IR
IE
-
-
-
-
-
-
-
-
rw
rw
5
4
3
2
1
0
ILVL
GLVL
rw
rw
Note: Please refer to the general Interrupt Control Register description for an
explanation of the control fields.
Table 28
Fast External Interrupt Control Register Addresses
Register
Address
External Interrupt
FEI0IC
FF88H / C4H
EX0IN
FEI1IC
FF8AH / C5H
EX1IN
FEI2IC
FF8CH / C6H
EX2IN
FEI3IC
FF8EH / C7H
EX3IN
FEI4IC
FF90H / C8H
EX4IN
FEI5IC
FF92H / C9H
EX5IN
FEI6IC
FF94H / CAH
EX6IN
FEI7IC
FF96H / CBH
EX7IN
7.9
Trap Functions
Traps interrupt the current execution similar to standard interrupts. However, trap
functions offer the possibility to bypass the interrupt system's prioritization process in
cases where immediate system reaction is required. Trap functions are not maskable
and always have priority over interrupt requests on any priority level.
The C165UTAH provides two different kinds of trapping mechanisms. Hardware traps
are triggered by events that occur during program execution (eg. illegal access or
undefined opcode), software traps are initiated via an instruction within the current
execution flow.
Software Traps
The TRAP instruction is used to cause a software call to an interrupt service routine. The
trap number that is specified in the operand field of the trap instruction determines which
vector location in the address range from 00’0000H through 00’01FCH will be branched
to.
Executing a TRAP instruction causes a similar effect as if an interrupt at the same vector
had occurred. PSW, CSP (in segmentation mode), and IP are pushed on the internal
system stack and a jump is taken to the specified vector location. When segmentation is
enabled and a trap is executed, the CSP for the trap service routine is set to code
Data Sheet
129
2001-02-23
C165UTAH
Interrupt and Trap Functions
segment 0. No Interrupt Request flags are affected by the TRAP instruction. The
interrupt service routine called by a TRAP instruction must be terminated with a RETI
(return from interrupt) instruction to ensure correct operation.
Note: The CPU level in register PSW is not modified by the TRAP instruction, so the
service routine is executed on the same priority level from which it was invoked.
Therefore, the service routine entered by the TRAP instruction can be interrupted
by other traps or higher priority interrupts, other than when triggered by a
hardware trap.
Hardware Traps
Hardware traps are issued by faults or specific system states that occur during runtime
of a program (not identified at assembly time). A hardware trap may also be triggered
intentionally, eg. to emulate additional instructions by generating an Illegal Opcode trap.
The C165UTAH distinguishes eight different hardware trap functions. When a hardware
trap condition has been detected, the CPU branches to the trap vector location for the
respective trap condition. Depending on the trap condition, the instruction which caused
the trap is either completed or cancelled (ie. it has no effect on the system state) before
the trap handling routine is entered.
Hardware traps are non-maskable and always have priority over every other CPU
activity. If several hardware trap conditions are detected within the same instruction
cycle, the highest priority trap is serviced (see table in section “Interrupt System
Structure”).
PSW, CSP (in segmentation mode), and IP are pushed on the internal system stack and
the CPU level in register PSW is set to the highest possible priority level (ie. level 15),
disabling all interrupts. The CSP is set to code segment zero, if segmentation is enabled.
A trap service routine must be terminated with the RETI instruction.
The eight hardware trap functions of the C165UTAH are divided into two classes:
Class A traps are
• external Non-Maskable Interrupt (NMI)
• Stack Overflow
• Stack Underflow trap
These traps share the same trap priority, but have an individual vector address.
Class B traps are
• Undefined Opcode
• Protection Fault
• Illegal Word Operand Access
• Illegal Instruction Access
• Illegal External Bus Access Trap
These traps share the same trap priority, and the same vector address.
Data Sheet
130
2001-02-23
C165UTAH
Interrupt and Trap Functions
The bit-addressable Trap Flag Register (TFR) allows a trap service routine to identify the
kind of trap which caused the exception. Each trap function is indicated by a separate
request flag. When a hardware trap occurs, the corresponding request flag in register
TFR is set to '1'.
TFR (FFACH / D6H)
SFR
Reset Value: 0000H
15
14
13
12
11
10
9
8
7
6
5
4
NMI
STK
OF
STK
UF
-
-
-
-
-
UND
OPC
-
-
-
rw
rw
rw
-
-
-
-
-
rw
-
-
-
3
2
PRT ILL
FLT OPA
rw
1
0
ILL
INA
ILL
BUS
rw
rw
rw
Bit
Function
ILLBUS
Illegal External Bus Access Flag
An external access has been attempted with no external bus defined.
ILLINA
Illegal Instruction Access Flag
A branch to an odd address has been attempted.
ILLOPA
Illegal Word Operand Access Flag
A word operand access (read or write) to an odd address has been attempted.
PRTFLT
Protection Fault Flag
A protected instruction with an illegal format has been detected.
UNDOPC
Undefined Opcode Flag
The currently decoded instruction has no valid C165UTAH opcode.
STKUF
Stack Underflow Flag
The current stack pointer value exceeds the content of register STKUN.
STKOF
Stack Overflow Flag
The current stack pointer value falls below the content of register STKOV.
NMI
Non Maskable Interrupt Flag
A negative transition (falling edge) has been detected on pin NMI.
Note: The trap service routine must clear the respective trap flag, otherwise a new trap
will be requested after exiting the service routine. Setting a trap request flag by
software causes the same effects as if it had been set by hardware.
The reset functions (hardware, software, watchdog) may be regarded as a type of trap.
Reset functions have the highest system priority (trap priority III).
Class A traps have the second highest priority (trap priority II), on the 3rd rank are class
B traps, so a class A trap can interrupt a class B trap. If more than one class A trap occur
at a time, they are prioritized internally, with the NMI trap on the highest and the stack
underflow trap on the lowest priority.
All class B traps have the same trap priority (trap priority I). When several class B traps
get active at a time, the corresponding flags in the TFR register are set and the trap
service routine is entered. Since all class B traps have the same vector, the priority of
Data Sheet
131
2001-02-23
C165UTAH
Interrupt and Trap Functions
service of simultaneously occurring class B traps is determined by software in the trap
service routine.
A class A trap occurring during the execution of a class B trap service routine will be
serviced immediately. During the execution of a class A trap service routine, however,
any class B trap occurring will not be serviced until the class A trap service routine is
exited with a RETI instruction. In this case, the occurrence of the class B trap condition
is stored in the TFR register, but the IP value of the instruction which caused this trap is
lost.
In the case where e.g. an Undefined Opcode trap (class B) occurs simultaneously with
an NMI trap (class A), both the NMI and the UNDOPC flag is set, the IP of the instruction
with the undefined opcode is pushed onto the system stack, but the NMI trap is executed.
After return from the NMI service routine, the IP is popped from the stack and
immediately pushed again because of the pending UNDOPC trap.
External NMI Trap
Whenever a high to low transition on the dedicated external NMI pin (Non-Maskable
Interrupt) is detected, the NMI flag in register TFR is set and the CPU will enter the NMI
trap routine. The IP value pushed on the system stack is the address of the instruction
following the one after which normal processing was interrupted by the NMI trap.
Stack Overflow Trap
Whenever the stack pointer is decremented to a value which is less than the value in the
stack overflow register STKOV, the STKOF flag in register TFR is set and the CPU will
enter the stack overflow trap routine. Which IP value will be pushed onto the system
stack depends on which operation caused the decrement of the SP. When an implicit
decrement of the SP is made through a PUSH or CALL instruction, or upon interrupt or
trap entry, the IP value pushed is the address of the following instruction. When the SP
is decremented by a subtract instruction, the IP value pushed represents the address of
the instruction after the instruction following the subtract instruction.
For recovery from stack overflow it must be ensured that there is enough excess space
on the stack for saving the current system state (PSW, IP, in segmented mode also CSP)
twice. Otherwise, a system reset should be generated.
Stack Underflow Trap
Whenever the stack pointer is incremented to a value which is greater than the value in
the stack underflow register STKUN, the STKUF flag is set in register TFR and the CPU
will enter the stack underflow trap routine. Again, which IP value will be pushed onto the
system stack depends on which operation caused the increment of the SP. When an
implicit increment of the SP is made through a POP or return instruction, the IP value
pushed is the address of the following instruction. When the SP is incremented by an
Data Sheet
132
2001-02-23
C165UTAH
Interrupt and Trap Functions
add instruction, the pushed IP value represents the address of the instruction after the
instruction following the add instruction.
Undefined Opcode Trap
When the instruction currently decoded by the CPU does not contain a valid C165UTAH
opcode, the UNDOPC flag is set in register TFR and the CPU enters the undefined
opcode trap routine. The IP value pushed onto the system stack is the address of the
instruction that caused the trap.
This can be used to emulate unimplemented instructions. The trap service routine can
examine the faulting instruction to decode operands for unimplemented opcodes based
on the stacked IP. In order to resume processing, the stacked IP value must be
incremented by the size of the undefined instruction, which is determined by the user,
before a RETI instruction is executed.
Protection Fault Trap
Whenever one of the special protected instructions is executed where the opcode of that
instruction is not repeated twice in the second word of the instruction and the byte
following the opcode is not the complement of the opcode, the PRTFLT flag in register
TFR is set and the CPU enters the protection fault trap routine. The protected
instructions include DISWDT, EINIT, IDLE, PWRDN, SRST, and SRVWDT. The IP value
pushed onto the system stack for the protection fault trap is the address of the instruction
that caused the trap.
Illegal Word Operand Access Trap
Whenever a word operand read or write access is attempted to an odd byte address, the
ILLOPA flag in register TFR is set and the CPU enters the illegal word operand access
trap routine. The IP value pushed onto the system stack is the address of the instruction
following the one which caused the trap.
Illegal Instruction Access Trap
Whenever a branch is made to an odd byte address, the ILLINA flag in register TFR is
set and the CPU enters the illegal instruction access trap routine. The IP value pushed
onto the system stack is the illegal odd target address of the branch instruction.
Illegal External Bus Access Trap
Whenever the CPU requests an external instruction fetch, data read or data write, and
no external bus configuration has been specified, the ILLBUS flag in register TFR is set
and the CPU enters the illegal bus access trap routine. The IP value pushed onto the
system stack is the address of the instruction following the one which caused the trap.
Data Sheet
133
2001-02-23
C165UTAH
Parallel Ports
8
Parallel Ports
In order to accept or generate single external control signals or parallel data, the
C165UTAH provides up to 72 parallel I/O lines. The C165UTAH features Port 0
(inculdes 8 bit P0H and 8 bit P0L), Port 1 (8 bit P1H and 8 bit P1L), Port 2 (8 bit), Port 3
(11 bit), Port 4 (7 bit), Port 6 (8 bit) and Port 7 (6 bit).
These port lines may be used for general purpose Input/Output controlled via software
or may be used implicitly by C165UTAH’s integrated peripherals or the External Bus
Controller.
All port lines are bit addressable, and all input/output lines are individually (bit-wise)
programmable as inputs or outputs via direction registers. The I/O ports are true
bidirectional ports which are switched to high impedance state when configured as
inputs. The output drivers of the I/O ports (P0H, P1, P2, P3, P4, P6, P7) can be
configured (pin by pin) for push/pull operation or open-drain operation via control
registers. The logic level of a pin is clocked into the input latch once per state time,
regardless whether the port is configured for input or output.
A write operation to a port pin configured as an input (DPx.y = ’0’) causes the value to
be written into the port output latch, while a read operation returns the latched state of
the pin itself. A read-modify-write operation reads the value of the pin, modifies it, and
writes it back to the output latch.
Writing to a pin configured as an output (DPx.y=‘1’) causes the output latch and the pin
to have the written value, since the output buffer is enabled. Reading this pin returns the
value of the output latch. A read-modify-write operation reads the value of the output
latch, modifies it, and writes it back to the output latch, thus also modifying the level at
the pin.
Data Sheet
134
2001-02-23
C165UTAH
Parallel Ports
Data Input / Output
Registers
P0L
Direction Control
Registers
Open Drain Control
Registers
DP0L
Pull Up/Down Control
Registers
PxPUDSEL:
P0H
DP0H
ODP0H
P1L
DP1L
ODP1L
P1H
DP1H
ODP1H
P0L, P0H, P1L,
P1H, P2, P3,
P4, P6, P7
P2
DP2
ODP2
PxPUDEN:
P3
DP3
ODP3
P4
DP4
ODP4
P6
DP6
ODP6
P0L, P0H, P1L,
P1H, P2, P3,
P4, P6, P7
P7
DP7
ODP7
PxPHEN:
P0L, P0H, P1L,
P1H, P2, P3,
P4, P6, P7
Figure 27
SFRs and Pins associated with the Parallel Ports
In the C165UTAH certain ports provide Open Drain Control, which allows to switch the
output driver of a port pin from a push/pull configuration to an open drain configuration.
In push/pull mode a port output driver has an upper and a lower transistor, thus it can
actively drive the line either to a high or a low level. In open drain mode the upper
transistor is always switched off, and the output driver can only actively drive the line to
a low level. When writing a ‘1’ to the port latch, the lower transistor is switched off and
the output enters a high-impedance state. The high level must then be provided by an
external pullup device. With this feature, it is possible to connect several port pins
together to a Wired-AND configuration, saving external glue logic and/or additional
software overhead for enabling/disabling output signals.
This feature is implemented for all ports except P0L, and is controlled through the
respective Open Drain Control Registers ODPx. These registers allow the individual bitwise selection of the open drain mode for each port line. If the respective control bit
ODPx.y is ‘0’ (default after reset), the output driver is in the push/pull mode. If ODPx.y is
‘1’, the open drain configuration is selected. Note that all ODPx registers are located in
the ESFR space.
Data Sheet
135
2001-02-23
C165UTAH
Parallel Ports
External
Pullup
Pin
Pin
Q
Q
Push/Pull Output Driver
Open Drain Output Driver
MCS01975
Figure 28
Output Drivers in Push/Pull Mode and in Open Drain Mode
Alternate Port Functions
Each port line has one programmable alternate input or output function associated.
PORT0 and PORT1 may be used as the address and data lines when accessing
external memory.
Port 2 is used for fast external interrupt inputs.
Port 3 includes alternate input/output functions of timers, serial interfaces, the optional
bus control signal BHE/WRH and the system clock output (CLKOUT).
Port 4 outputs the additional segment address bits A22/A19/A17...A16 in systems where
more than 64 KBytes of memory are to be accessed directly.
Port 6 provides the optional chip select outputs and the bus arbitration lines.
Port 7 is used for general purpose I/Os.
If an alternate output function of a pin is to be used, the direction of this pin must be
programmed for output (DPx.y=‘1’), except for some signals that are used directly after
reset and are configured automatically. Otherwise the pin remains in the high-impedance
state and is not effected by the alternate output function. The respective port latch should
hold a ‘1’, because its output is ANDed with the alternate output data.
Note: DP0L and, if a 16 bit external XBus data bus is used, also DP0H must be ’0’ as
long as the XBUS is active.
If an alternate input function of a pin is used, the direction of the pin must be programmed
for input (DPx.y=‘0’) if an external device is driving the pin. The input direction is the
default after reset. If no external device is connected to the pin, however, one can also
set the direction for this pin to output. In this case, the pin reflects the state of the port
Data Sheet
136
2001-02-23
C165UTAH
Parallel Ports
output latch. Thus, the alternate input function reads the value stored in the port output
latch. This can be used for testing purposes to allow a software trigger of an alternate
input function by writing to the port output latch.
On most of the port lines, the user software is responsible for setting the proper direction
when using an alternate input or output function of a pin. This is done by setting or
clearing the direction control bit DPx.y of the pin before enabling the alternate function.
There are port lines, however, where the direction of the port line is switched
automatically. For instance, in the multiplexed external bus modes of PORT0, the
direction must be switched several times for an instruction fetch in order to output the
addresses and to input the data. Obviously, this cannot be done through instructions. In
these cases, the direction of the port line is switched automatically by hardware if the
alternate function of such a pin is enabled.
Note: In this case, make sure DP0 is set to ’0’ signal.
To determine the appropriate level of the port output latches, check how the alternate
data output is combined with the respective port latch output.
There is one basic structure for all port lines with only an alternate input function. Port
lines with only an alternate output function, however, have different structures due to the
way the direction of the pin is switched and depending on whether the pin is accessible
by the user software or not in the alternate function mode.
All port lines that are not used for these alternate functions may be used as general
purpose I/O lines. When using port pins for general purpose output, the initial output
value should be written to the port latch prior to enabling the output drivers, in order to
avoid undesired transitions on the output pins. This applies to single pins as well as to
pin groups (see examples below).
OUTPUT_ENABLE_SINGLE_PIN:
BSET
P4.0
;Initial output level is ’high’
BSET
DP4.0
;Switch on the output driver
OUTPUT_ENABLE_PIN_GROUP:
BFLDL
P4, #05H, #05H
;Initial output level is ’high’
BFLDL
DP4, #05H, #05H
;Switch on the output drivers
Each of these ports and the alternate input and output functions are described in detail
in the following subsections.
Data Sheet
137
2001-02-23
C165UTAH
Parallel Ports
8.1
PORT0
The two 8-bit ports P0H and P0L represent the higher and lower part of PORT0,
respectively. Both halfs of PORT0 can be written (eg. via a PEC transfer) without
effecting the other half.
If this port is used for general purpose I/O, the direction of each line can be configured
via the corresponding direction registers DP0H and DP0L.
Each port line of PORT0H can be switched into push/pull or open drain mode via the
open drain control register ODP0H.
For port pins configured as input (via DP0x or alternate function), an internal pull
transistor is connected to the pad if register P0xPUDEN = ’1’, no matter whether the
C165UTAH is in normal operation mode or in power down mode. Either pulldown
transistor or pullup transistor will be selected via P0xPUDSEL.
For port pins configured as output, the internal pull transistors are always disabled. The
output driver is disabled in power down mode unless P0xPHEN = ’1’.
After reset, P0xPUDEN and P0xPUDSEL are set to HIGH signal, thereby providing the
default reset configuration 1111H to the UTAH during reset.
Note: While this feature allows the user to start the UTAH after reset in default
configuration without external pull devices, the default configuration may be
overwritten by stronger external pulldown devices. In this case, Software should
disable the internal pull’s after reset (see also next chapter ’Alternate Function’).
Data Sheet
138
2001-02-23
C165UTAH
Parallel Ports
P0L (FF00H / 80H)
15
14
13
SFR
12
11
10
9
8
Reset Value: - - 00H
7
6
5
4
3
2
1
0
P0L.7 P0L.6 P0L.5 P0L.4 P0L.3 P0L.2 P0L.1 P0L.0
-
-
-
-
-
-
-
-
P0H (FF02H / 81H)
15
14
13
rw
rw
rw
rw
SFR
12
11
10
9
8
rw
rw
rw
rw
Reset Value: - - 00H
7
6
5
4
3
2
1
0
P0H.7 P0H.6 P0H.5 P0H.4 P0H.3 P0H.2 P0H.1 P0H.0
-
-
-
-
-
-
-
-
rw
Bit
Function
P0X.y
Port data register P0H or P0L bit y
DP0L (F100H / 80H)
15
14
13
rw
rw
rw
ESFR
12
11
10
9
8
7
rw
rw
rw
rw
Reset Value: - - 00H
6
5
4
3
2
1
0
DP0L DP0L DP0L DP0L DP0L DP0L DP0L DP0L
.0
.7
.6
.5
.4
.3
.2
.1
-
-
-
-
-
-
-
DP0H (F102H / 81H)
15
14
13
-
rw
rw
rw
rw
ESFR
12
11
10
9
8
7
rw
rw
rw
rw
Reset Value: - - 00H
6
5
4
3
2
1
0
DP0H DP0H DP0H DP0H DP0H DP0H DP0H DP0H
.7
.6
.5
.4
.3
.2
.1
.0
-
-
-
-
-
-
-
-
rw
rw
rw
rw
Bit
Function
DP0X.y
Port direction register DP0H or DP0L bit y
DP0X.y = 0: Port line P0X.y is an input (high-impedance)
DP0X.y = 1: Port line P0X.y is an output
Data Sheet
139
rw
rw
rw
rw
2001-02-23
C165UTAH
Parallel Ports
ODP0H (FE22H / 11H)
15
14
13
12
SFR
11
10
9
8
Reset Value: - - 00H
7
6
5
4
3
2
1
0
ODP0 ODP0 ODP0 ODP0 ODP0 ODP0 ODP0 ODP0
H.7 H.6
H.5 H.4 H.3 H.2
H.1 H.0
-
-
-
-
-
-
-
-
rw
rw
rw
rw
rw
Bit
Function
ODP0H.y
Port0H Open Drain control register bit y
ODP0H.y = 0: Port line P0H.y output driver in push/pull mode
ODP0H.y = 1: Port line P0H.y output driver in open drain mode
P0LPUDSEL (FE60H / 30H)
SFR
15
14
13
12
11
10
9
8
-
-
-
-
-
-
-
-
P0HPUDSEL (FE62H / 31H)
14
13
12
11
10
9
8
-
-
-
-
-
-
-
-
rw
7
6
5
4
3
2
1
0
P0L P0L P0L P0L P0L P0L P0L P0L
PUD PUD PUD PUD PUD PUD PUD PUD
SEL.7 SEL.6 SEL.5 SEL.4 SEL.3 SEL.2 SEL.1 SEL.0
rw
rw
rw
rw
rw
rw
rw
rw
Reset Value: - - FFH
7
6
5
4
3
2
1
0
P0H P0H P0H P0H P0H P0H P0H P0H
PUD PUD PUD PUD PUD PUD PUD PUD
SEL.7 SEL.6 SEL.5 SEL.4 SEL.3 SEL.2 SEL.1 SEL.0
rw
rw
rw
rw
rw
rw
rw
Bit
Function
P0xPUDSEL.y
Pulldown/Pullup Selection
P0xPUDSEL.y = 0: internal programmable pulldown transistor is selected
P0xPUDSEL.y = 1: internal programmable pullup transistor is selected
Data Sheet
rw
Reset Value: - - FFH
SFR
15
rw
140
rw
2001-02-23
C165UTAH
Parallel Ports
P0LPUDEN (FE64H / 32H)
SFR
15
14
13
12
11
10
9
8
-
-
-
-
-
-
-
-
P0HPUDEN (FE66H / 33H)
Reset Value: - - FFH
7
6
5
4
3
2
1
0
P0L P0L P0L P0L P0L P0L P0L P0L
PUD PUD PUD PUD PUD PUD PUD PUD
EN.7 EN.6 EN.5 EN.4 EN.3 EN.2 EN.1 EN.0
rw
rw
rw
rw
SFR
15
14
13
12
11
10
9
8
-
-
-
-
-
-
-
-
rw
rw
7
6
5
4
3
2
1
0
P0H P0H P0H P0H P0H P0H P0H P0H
PUD PUD PUD PUD PUD PUD PUD PUD
EN.7 EN.6 EN.5 EN.4 EN.3 EN.2 EN.1 EN.0
rw
rw
rw
rw
rw
rw
Function
P0xPUDEN.y
Pulldown/Pullup Enable
P0xPUDEN.y = 0: internal programmable pull transistor is disabled
P0xPUDEN.y = 1: internal programmable pull transistor is enabled
SFR
15
14
13
12
11
10
9
8
-
-
-
-
-
-
-
-
P0HPHEN (FE6AH / 35H)
14
13
12
11
10
9
8
-
-
-
-
-
-
-
-
rw
7
6
5
4
3
2
1
0
P0L P0L P0L P0L P0L P0L P0L P0L
PHEN PHEN PHEN PHEN PHEN PHEN PHEN PHEN
.6
.5
.4
.3
.2
.1
.0
.7
rw
rw
rw
rw
rw
rw
rw
rw
Reset Value: - - 00H
7
6
5
4
3
2
1
0
P0H P0H P0H P0H P0H P0H P0H P0H
PHEN PHEN PHEN PHEN PHEN PHEN PHEN PHEN
.7
.6
.5
.4
.3
.2
.1
.0
rw
rw
rw
rw
rw
Bit
Function
P0xPHEN.y
Output Driver Enable in Power Down Mode
P0xPHEN.y = 0: output driver is disabled in power down mode
P0xPHEN.y = 1: output driver is enabled in power down mode
Data Sheet
rw
Reset Value: - - 00H
SFR
15
rw
Reset Value: - - FFH
Bit
P0LPHEN (FE68H / 34H)
rw
141
rw
rw
rw
2001-02-23
C165UTAH
Parallel Ports
8.1.1
Alternate Functions of PORT0
When an external bus is enabled, PORT0 is used as data bus or address/data bus.
Note that an external 8-bit demultiplexed bus only uses P0L, while P0H is free for I/O
(provided that no other bus mode is enabled).
PORT0 is also used to select the system startup configuration. During reset, PORT0 is
configured to input, and each line is held high through an internal pullup device. Each
line can now be individually pulled to a low level (see DC-level specifications in the
respective Data Sheets) through an external pulldown device. A default configuration is
selected when the respective PORT0 lines are at a high level. Through pulling individual
lines to a low level, this default can be changed according to the needs of the
applications.
The internal pullup devices are designed such that an external pulldown resistors (see
specification) can be used to apply a correct low level. These external pulldown resistors
can remain connected to the PORT0 pins also during normal operation, however, care
has to be taken such that they do not disturb the normal function of PORT0 (this might
be the case, for example, if the external resistor is too strong).
With the end of reset, the selected bus configuration will be written to the BUSCON0
register. The configuration of the high byte of PORT0, will be copied into the special
register RP0H. This read-only register holds the selection for the number of chip selects
and segment addresses. Software can read this register in order to react according to
the selected configuration, if required.
Note: When the reset is terminated, the internal pullup devices must be switched off by
Software and PORT0 will be switched to the appropriate operating mode.
During external accesses in multiplexed bus modes PORT0 first outputs the 16-bit intrasegment address as an alternate output function. PORT0 is then switched to highimpedance input mode to read the incoming instruction or data. In 8-bit data bus mode,
two memory cycles are required for word accesses, the first for the low byte and the
second for the high byte of the word. During write cycles PORT0 outputs the data byte
or word after outputting the address.
During external accesses in demultiplexed bus modes PORT0 reads the incoming
instruction or data word or outputs the data byte or word.
Data Sheet
142
2001-02-23
C165UTAH
Parallel Ports
Alternate Function
a)
b)
P0H.7
P0H.6
P0H.5
P0H.4
P0H.3
P0H.2
P0H.1
P0H.0
P0L.7
P0L.6
P0L.5
P0L.4
P0L.3
P0L.2
P0L.1
P0L.0
D7
D6
D5
D4
D3
D2
D1
D0
General Purpose
Input/Output
8-bit
Demux Bus
P0H
PORT0
P0L
Figure 29
c)
D15
D14
D13
D12
D11
D10
D9
D8
D7
D6
D5
D4
D3
D2
D1
D0
16-bit
Demux Bus
d)
AD15
AD14
AD13
AD12
AD11
AD10
AD9
AD8
AD7
AD6
AD5
AD4
AD3
AD2
AD1
AD0
8-bit
MUX Bus
AD15
AD14
AD13
AD12
AD11
AD10
AD9
AD8
AD7
AD6
AD5
AD4
AD3
AD2
AD1
AD0
16-bit
MUX Bus
PORT0 I/O and Alternate Functions
When an external bus mode is enabled, the direction of the port pin and the loading of
data into the port output latch are controlled by the bus controller hardware. The input of
the port output latch is disconnected from the internal bus and is switched to the line
labeled “Alternate Data Output” via a multiplexer. The alternate data can be the 16-bit
intrasegment address or the 8/16-bit data information. The incoming data on PORT0 is
read on the line “Alternate Data Input”. While an external bus mode is enabled, the user
software should not write to the port output latch, otherwise unpredictable results may
occur. When the external bus modes are disabled, the contents of the direction register
last written by the user becomes active.
Figure 30 shows the structure of a PORT0 pin.
Data Sheet
143
2001-02-23
C165UTAH
Parallel Ports
Write DP0H.y/DP0L.y
Alternate
Direction
1
MUX
0
Direction
Latch
Read DP0H.y/DP0L.y
Ι
n
t
e
r
n
a
l
Write P0H.y/P0L.y
Alternate
Function
Enable
Alternate
Data
Output
1
MUX
0
Port Output
Latch
B
u
s
Output
Buffer
P0H.y
P0L.y
Read P0H.y/P0L.y
Clock
1
MUX
0
Input
Latch
MCB02231
y = 7...0
Figure 30
8.2
Block Diagram of a PORT0 Pin
PORT1
The two 8-bit ports P1H and P1L represent the higher and lower part of PORT1,
respectively. Both halfs of PORT1 can be written (eg. via a PEC transfer) without
effecting the other half.
If this port is used for general purpose I/O, the direction of each line can be configured
via the corresponding direction registers DP1H and DP1L.
Each port line can be switched into push/pull or open drain mode via the open drain
control register ODP1L and ODP1H.
For port pins configured as input (via DP1x or alternate function), an internal pull
transistor is connected to the pad if register P1xPUDEN = ’1’, no matter wheter the
C165UTAH is in normal operation mode or in power down mode. Either pulldown
transistor or pullup transistor will be selected via P1xPUDSEL.
Data Sheet
144
2001-02-23
C165UTAH
Parallel Ports
For port pins configured as output, the internal pull transistors are always disabled. The
output driver is disabled in power down mode unless P1xPHEN = ’1’.
After reset, P1xPUDEN and P1xPUDSEL are set to LOW signal.
Data Sheet
145
2001-02-23
C165UTAH
Parallel Ports
P1L (FF04H / 82H)
15
14
13
SFR
12
11
10
9
8
Reset Value: - - 00H
7
6
5
4
3
2
1
0
P1L.7 P1L.6 P1L.5 P1L.4 P1L.3 P1L.2 P1L.1 P1L.0
-
-
-
-
-
-
-
-
P1H (FF06H / 83H)
15
14
13
rw
rw
rw
rw
SFR
12
11
10
9
8
rw
rw
rw
rw
Reset Value: - - 00H
7
6
5
4
3
2
1
0
P1H.7 P1H.6 P1H.5 P1H.4 P1H.3 P1H.2 P1H.1 P1H.0
-
-
-
-
-
-
-
-
rw
Bit
Function
P1X.y
Port data register P1H or P1L bit y
DP1L (F104H / 82H)
15
14
13
rw
rw
rw
ESFR
12
11
10
9
8
7
rw
rw
rw
rw
Reset Value: - - 00H
6
5
4
3
2
1
0
DP1L DP1L DP1L DP1L DP1L DP1L DP1L DP1L
.7
.6
.5
.4
.3
.2
.1
.0
-
-
-
-
-
-
-
DP1H (F106H / 83H)
15
14
13
-
rw
rw
rw
rw
ESFR
12
11
10
9
8
7
rw
rw
rw
rw
Reset Value: - - 00H
6
5
4
3
2
1
0
DP1H DP1H DP1H DP1H DP1H DP1H DP1H DP1H
.7
.6
.5
.4
.3
.2
.1
.0
-
-
-
-
-
-
-
-
rw
rw
rw
rw
Bit
Function
DP1X.y
Port direction register DP1H or DP1L bit y
DP1X.y = 0: Port line P1X.y is an input (high-impedance)
DP1X.y = 1: Port line P1X.y is an output
ODP1L (FE24H / 12H)
15
14
13
12
SFR
11
10
9
8
rw
rw
rw
rw
Reset Value: - - 00H
7
6
5
4
3
2
1
0
ODP1 ODP1 ODP1 ODP1 ODP1 ODP1 ODP1 ODP1
L.7
L.6
L.5
L.4
L.3
L.2
L.1
L.0
-
-
Data Sheet
-
-
-
-
-
-
rw
146
rw
rw
rw
rw
rw
rw
rw
2001-02-23
C165UTAH
Parallel Ports
ODP1H (FE26H / 13H)
15
14
13
12
SFR
11
10
9
8
Reset Value: - - 00H
7
6
5
4
3
2
1
0
ODP1 ODP1 ODP1 ODP1 ODP1 ODP1 ODP1 ODP1
H.7 H.6
H.5 H.4 H.3 H.2
H.1 H.0
-
-
-
-
-
-
-
-
rw
rw
rw
rw
rw
Bit
Function
ODP1x.y
Port1x Open Drain control register bit y
ODP1x.y = 0: Port line P1x.y output driver in push/pull mode
ODP1x.y = 1: Port line P1x.y output driver in open drain mode
P1LPUDSEL (FE6CH / 36H)
SFR
15
14
13
12
11
10
9
8
-
-
-
-
-
-
-
-
P1HPUDSEL (FE6EH / 37H)
14
13
12
11
10
9
8
-
-
-
-
-
-
-
-
rw
7
6
5
4
3
2
1
0
P1L P1L P1L P1L P1L P1L P1L P1L
PUD PUD PUD PUD PUD PUD PUD PUD
SEL.7 SEL.6 SEL.5 SEL.4 SEL.3 SEL.2 SEL.1 SEL.0
rw
rw
rw
rw
rw
rw
rw
rw
Reset Value: - - 00H
7
6
5
4
3
2
1
0
P1H P1H P1H P1H P1H P1H P1H P1H
PUD PUD PUD PUD PUD PUD PUD PUD
SEL.7 SEL.6 SEL.5 SEL.4 SEL.3 SEL.2 SEL.1 SEL.0
rw
rw
rw
rw
rw
rw
rw
Bit
Function
P1xPUDSEL.y
Pulldown/Pullup Selection
P1xPUDSEL.y = 0: internal programmable pulldown transistor is selected
P1xPUDSEL.y = 1: internal programmable pullup transistor is selected
Data Sheet
rw
Reset Value: - - 00H
SFR
15
rw
147
rw
2001-02-23
C165UTAH
Parallel Ports
P1LPUDEN (FE70H / 38H)
SFR
15
14
13
12
11
10
9
8
-
-
-
-
-
-
-
-
P1HPUDEN (FE72H / 39H)
Reset Value: - - 00H
7
6
5
4
3
2
1
0
P1L P1L P1L P1L P1L P1L P1L P1L
PUD PUD PUD PUD PUD PUD PUD PUD
EN.7 EN.6 EN.5 EN.4 EN.3 EN.2 EN.1 EN.0
rw
rw
rw
rw
SFR
15
14
13
12
11
10
9
8
-
-
-
-
-
-
-
-
rw
rw
7
6
5
4
3
2
1
0
P1H P1H P1H P1H P1H P1H P1H P1H
PUD PUD PUD PUD PUD PUD PUD PUD
EN.7 EN.6 EN.5 EN.4 EN.3 EN.2 EN.1 EN.0
rw
rw
rw
rw
rw
rw
Function
P1xPUDEN.y
Pulldown/Pullup Enable
P1xPUDEN.y = 0: internal programmable pull transistor is disabled
P1xPUDEN.y = 1: internal programmable pull transistor is enabled
SFR
15
14
13
12
11
10
9
8
-
-
-
-
-
-
-
-
P1HPHEN (FE76H / 3BH)
14
13
12
11
10
9
8
-
-
-
-
-
-
-
-
rw
rw
Reset Value: - - 00H
7
6
5
4
3
2
1
0
P1L P1L P1L P1L P1L P1L P1L P1L
PHEN PHEN PHEN PHEN PHEN PHEN PHEN PHEN
.6
.5
.4
.3
.2
.1
.0
.7
rw
rw
rw
rw
SFR
15
rw
Reset Value: - - 00H
Bit
P1LPHEN (FE74H / 3AH)
rw
rw
rw
rw
rw
Reset Value: - - 00H
7
6
5
4
3
2
1
0
P1H P1H P1H P1H P1H P1H P1H P1H
PHEN PHEN PHEN PHEN PHEN PHEN PHEN PHEN
.7
.6
.5
.4
.3
.2
.1
.0
rw
rw
rw
rw
rw
Bit
Function
P1xPHEN.y
Output Driver Enable in Power Down Mode
P1xPHEN.y = 0: output driver is disabled in power down mode
P1xPHEN.y = 1: output driver is enabled in power down mode
Data Sheet
148
rw
rw
rw
2001-02-23
C165UTAH
Parallel Ports
8.2.1
Alternate Functions of PORT1
When a demultiplexed external bus is enabled, PORT1 is used as address bus.
Note that demultiplexed bus modes use PORT1 as a 16-bit port. Otherwise all 16 port
lines can be used for general purpose I/O.
During external accesses in demultiplexed bus modes PORT1 outputs the 16-bit intrasegment address as an alternate output function.
During external accesses in multiplexed bus modes, when no BUSCON register selects
a demultiplexed bus mode, PORT1 is not used and is available for general purpose I/O.
Alternate Function
P1H
PORT1
P1L
P1H.7
P1H.6
P1H.5
P1H.4
P1H.3
P1H.2
P1H.1
P1H.0
P1L.7
P1L.6
P1L.5
P1L.4
P1L.3
P1L.2
P1L.1
P1L.0
General Purpose
Input/Output
Figure 31
a)
A15
A14
A13
A12
A11
A10
A9
A8
A7
A6
A5
A4
A3
A2
A1
A0
8/16-bit
Demux Bus
PORT1 I/O and Alternate Functions
When an external bus mode is enabled, the direction of the port pin and the loading of
data into the port output latch are controlled by the bus controller hardware. The input of
the port output latch is disconnected from the internal bus and is switched to the line
labeled “Alternate Data Output” via a multiplexer. The alternate data is the 16-bit
intrasegment address. While an external bus mode is enabled, the user software should
not write to the port output latch, otherwise unpredictable results may occur. When the
external bus modes are disabled, the contents of the direction register last written by the
user becomes active.
Figure 32 shows the structure of a PORT1 pin.
Data Sheet
149
2001-02-23
C165UTAH
Parallel Ports
Write DP1H.y/DP1L.y
’1’
1
MUX
0
Direction
Latch
Read DP1H.y/DP1L.y
Ι
n
t
e
r
n
a
l
Write P1H.y/P1L.y
Alternate
Function
Enable
Alternate
Data
Output
1
MUX
0
Port Output
Latch
B
u
s
Output
Buffer
P1H.y
P1L.y
Read P1H.y/P1L.y
Clock
1
MUX
0
Input
Latch
MCB02232
y = 7...0
Figure 32
8.3
Block Diagram of a PORT1 Pin
PORT2
In the C165UTAH Port 2 is an 8 -bit port. If Port 2 is used for general purpose I/O, the
direction of each line can be configured via the corresponding direction register DP2.
Each port line can be switched into push/pull or open drain mode via the open drain
control register ODP2.
For port pins configured as input (via DP2 or alternate function), an internal pull transistor
is connected to the pad if register P2PUDEN = ’1’, no matter wheter the C165UTAH is
in normal operation mode or in power down mode. Either pulldown transistor or pullup
transistor will be selected via P2PUDSEL.
For port pins configured as output, the internal pull transistors are always disabled. The
output driver is disabled in power down mode unless P2PHEN = ’1’.
Data Sheet
150
2001-02-23
C165UTAH
Parallel Ports
After reset, P2PUDEN and P2PUDSEL are set to LOW signal.
P2 (FFC0H / E0H)
15
14
13
SFR
12
11
10
9
8
Reset Value: - - 00H
7
6
5
4
3
2
1
0
P2.7 P2.6 P2.5 P2.4 P2.3 P2.2 P2.1 P2.0
-
-
-
-
-
-
-
Bit
Function
P2.y
Port data register P2 bit y
-
DP2 (FFC2H / E1H)
15
14
13
rw
rw
rw
rw
SFR
12
11
10
9
8
rw
rw
rw
rw
Reset Value: - - 00H
7
6
5
4
3
2
1
0
DP2.7 DP2.6 DP2.5 DP2.4 DP2.3 DP2.2 DP2.1 DP2.0
-
-
-
-
-
-
-
-
rw
rw
rw
rw
Bit
Function
DP2.y
Port direction register DP2 bit y
DP2.y = 0: Port line P2.y is an input (high-impedance)
DP2.y = 1: Port line P2.y is an output
ODP2 (F1C2H / E1H)
15
14
13
ESFR
12
11
10
9
8
7
rw
rw
rw
rw
Reset Value: - - 00H
6
5
4
3
2
1
0
ODP2 ODP2 ODP2 ODP2 ODP2 ODP2 ODP2 ODP2
.7
.6
.5
.4
.3
.2
.1
.0
-
-
-
-
-
-
-
-
rw
rw
rw
rw
rw
Bit
Function
ODP2.y
Port 2 Open Drain control register bit y
ODP2.y = 0: Port line P2.y output driver in push/pull mode
ODP2.y = 1: Port line P2.y output driver in open drain mode
Data Sheet
151
rw
rw
rw
2001-02-23
C165UTAH
Parallel Ports
P2PUDSEL (FE78H / 3CH)
SFR
15
14
13
12
11
10
9
8
-
-
-
-
-
-
-
-
Reset Value: - - 00H
7
6
5
4
3
2
1
0
P2
P2
P2
P2
P2
P2
P2
P2
PUD PUD PUD PUD PUD PUD PUD PUD
SEL.7 SEL.6 SEL.5 SEL.4 SEL.3 SEL.2 SEL.1 SEL.0
rw
rw
rw
rw
rw
rw
rw
Bit
Function
P2PUDSEL.y
Pulldown/Pullup Selection
P2PUDSEL.y = 0: internal programmable pulldown transistor is selected
P2PUDSEL.y = 1: internal programmable pullup transistor is selected
P2PUDEN (FE7AH / 3DH)
SFR
15
14
13
12
11
10
9
8
-
-
-
-
-
-
-
-
Reset Value: - - 00H
7
6
5
4
3
2
1
0
P2
P2
P2
P2
P2
P2
P2
P2
PUD PUD PUD PUD PUD PUD PUD PUD
EN.7 EN.6 EN.5 EN.4 EN.3 EN.2 EN.1 EN.0
rw
rw
rw
rw
rw
rw
Bit
Function
P2PUDEN.y
Pulldown/Pullup Enable
P2PUDEN.y = 0: internal programmable pull transistor is disabled
P2PUDEN.y = 1: internal programmable pull transistor is enabled
P2PHEN (FE7CH / 3EH)
SFR
15
14
13
12
11
10
9
8
-
-
-
-
-
-
-
-
rw
rw
Reset Value: - - 00H
7
6
5
4
3
2
1
0
P2
P2
P2
P2
P2
P2
P2
P2
PHEN PHEN PHEN PHEN PHEN PHEN PHEN PHEN
.7
.6
.5
.4
.3
.2
.1
.0
rw
rw
rw
rw
rw
Bit
Function
P2PHEN.y
Output Driver Enable in Power Down Mode
P2PHEN.y = 0: output driver is disabled in power down mode
P2PHEN.y = 1: output driver is enabled in power down mode
Data Sheet
rw
152
rw
rw
rw
2001-02-23
C165UTAH
Parallel Ports
8.3.1
Alternate Functions of PORT2
All Port 2 lines (P2.7..P2.0) can serve as Fast External Interrupt inputs (EX7IN...EX0IN).
Table 29 summarizes the alternate functions of Port 2.
Table 29
Port 2 Alternate Functions: Fast External Interrupts
Port 2 Pin
Alternate Function
P2.0
P2.1
P2.2
P2.3
P2.4
P2.5
P2.6
P2.7
EX0IN
EX1IN
EX2IN
EX3IN
EX4IN
EX5IN
EX6IN
EX7IN
Alternate Function
Port 2
a)
P2.7
P2.6
P2.5
P2.4
P2.3
P2.2
P2.1
P2.0
General Purpose
Input/Output
Figure 33
Fast External Interrupt 0 Input
Fast External Interrupt 1 Input
Fast External Interrupt 2 Input
Fast External Interrupt 3 Input
Fast External Interrupt 4 Input
Fast External Interrupt 5 Input
Fast External Interrupt 6 Input
Fast External Interrupt 7 Input
EX7IN
EX6IN
EX5IN
EX4IN
EX3IN
EX2IN
EX1IN
EX0IN
Fast External
Interrupt Input
Port 2 I/O and Alternate Functions
The pins of Port 2 combine internal bus data and alternate data output before the port
latch input.
Note: As opposed to the C165UTAH, in other existing Infineon C16x devices EX0IN is
assigned to P2.8, EX1IN is assigned to P2.9 ... and EX7IN is assigned to P2.15
using the higher byte of Port 2 instead of using the lower byte of Port 2.
Data Sheet
153
2001-02-23
C165UTAH
Parallel Ports
Write ODP2.y
Open Drain
Latch
Read ODP2.y
Ι
n
t
e
r
n
a
l
B
u
s
Write DP2.y
Direction
Latch
Read DP2.y
Write P2.y
Port Output
Latch
Output
Buffer
Read P2.y
Clock
1
MUX
0
Input
Latch
Alternate
Data Input
Figure 34
Data Sheet
P2.y
XzIn
MCB02230
Block Diagram of a Port 2 Pin (y = 7...0)
154
2001-02-23
C165UTAH
Parallel Ports
8.4
PORT3
If this 11-bit port is used for general purpose I/O, the direction of each line can be
configured via the corresponding direction register DP3. Each port lines can be switched
into push/pull or open drain mode via the open drain control register ODP3.
Note: Due to pin limitations register bit P3.0..P3.2, P3.4 and P3.14 is not connected to
an output pin. The Port 3 bit-assignment is not consecutive to for compatibility with
other C16x devices.
For port pins configured as input (via DP3 or alternate function), an internal pull transistor
is connected to the pad if register P3PUDEN = ’1’, no matter wheter the C165UTAH is
in normal operation mode or in power down mode. Either pulldown transistor or pullup
transistor will be selected via P3PUDSEL.
For port pins configured as output, the internal pull transistors are always disabled. The
output driver is disabled in power down mode unless P3PHEN = ’1’.
After reset, P3PUDEN and P3PUDSEL are set to LOW signal.
P3 (FFC4H / E2H) SFRReset Value: 0000H
15
14
P3.15
-
rw
-
13
12
11
10
9
8
6
5
P3.13 P3.12 P3.11 P3.10 P3.9 P3.8 P3.7 P3.6 P3.5
rw
rw
rw
rw
rw
rw
Bit
Function
P3.y
Port data register P3 bit y
Data Sheet
7
rw
155
rw
rw
4
3
2
1
0
-
P3.3
-
-
-
-
rw
-
-
-
2001-02-23
C165UTAH
Parallel Ports
DP3 (FFC6H / E3H)
SFR
Reset Value: 0000H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
DP3
.15
-
DP3
.13
DP3
.12
DP3
.11
DP3
.10
DP3
.9
DP3
.8
DP3
.7
DP3
.6
DP3
.5
-
DP3
.3
-
-
-
rw
-
rw
rw
rw
rw
rw
rw
rw
rw
rw
-
rw
-
-
-
Bit
Function
DP3.y
Port direction register DP3 bit y
DP3.y = 0: Port line P3.y is an input (high-impedance)
DP3.y = 1: Port line P3.y is an output
ODP3 (F1C6H / E3H)
15
14
ODP3
.15
-
rw
-
13
12
ESFR
11
10
9
8
7
Reset Value: 0000H
6
5
ODP3 ODP3 ODP3 ODP3 ODP3 ODP3 ODP3 ODP3 ODP3
.12
.13
.11
.10
.9
.8
.7
.6
.5
rw
rw
rw
rw
rw
rw
rw
rw
rw
4
3
2
1
0
-
ODP3
.3
-
-
-
-
rw
-
-
-
Bit
Function
ODP3.y
Port 3 Open Drain control register bit y
ODP3.y = 0: Port line P3.y output driver in push/pull mode
ODP3.y = 1: Port line P3.y output driver in open drain mode
P3PUDSEL (FE7EH / 3FH)
15
14
P3PU
DSEL. 15
rw
SFR
Reset Value: - - 00H
13
12
11
10
9
8
7
6
5
4
P3PU P3PU P3PU P3PU P3PU P3PU P3PU P3PU P3PU
DSEL. DSEL. DSEL. DSEL.DSEL. DSEL. DSEL. DSEL. DSEL. 13
11
10
9
8
7
6
5
12
-
rw
rw
rw
rw
rw
rw
rw
rw
rw
-
3
P3PU
DSEL.
3
2
1
0
-
-
-
rw
-
-
-
Bit
Function
P3PUDSEL.y
Pulldown/Pullup Selection
P3PUDSEL.y = 0: internal programmable pulldown transistor is selected
P3PUDSEL.y = 1: internal programmable pullup transistor is selected
Data Sheet
156
2001-02-23
C165UTAH
Parallel Ports
P3PUDEN (FE80H / 40H)
15
P3PU
DEN.
15
14
rw
-
-
SFR
Reset Value: - - 00H
13
12
11
10
9
8
7
6
5
P3PU P3PU P3PU P3PU P3PU P3PU P3PU P3PU P3PU
DEN. DEN. DEN. DEN. DEN. DEN. DEN. DEN. DEN.
13
11
10
9
8
7
6
5
12
rw
rw
rw
rw
rw
rw
rw
rw
rw
4
-
3
P3PU
DEN.
3
2
1
0
-
-
-
rw
-
-
-
Bit
Function
P3PUDEN.y
Pulldown/Pullup Enable
P3PUDEN.y = 0: internal programmable pull transistor is disabled
P3PUDEN.y = 1: internal programmable pull transistor is enabled
P3PHEN (FE82H / 41H)
15
14
P3
PHEN .15
rw
-
SFR
Reset Value: - - 00H
13
12
11
10
9
8
7
6
5
P3
P3
P3
P3
P3
P3
P3
P3
P3
PHEN PHEN PHEN PHEN PHEN PHEN PHEN PHEN PHEN
.13
.9
.8
.7
.6
.5
.11
.10
.12
rw
rw
rw
rw
rw
rw
rw
rw
rw
4
-
3
P3
PHEN
.3
2
1
0
-
-
-
rw
-
-
-
Bit
Function
P3PHEN.y
Output Driver Enable in Power Down Mode
P3PHEN.y = 0: output driver is disabled in power down mode
P3PHEN.y = 1: output driver is enabled in power down mode
8.4.1
Alternate Functions of PORT3
The pins of Port 3 serve for various functions which include external timer control lines,
the two serial interfaces and the control lines BHE and CLKOUT.
Table 30 summarizes the alternate functions of Port 3.
Data Sheet
157
2001-02-23
C165UTAH
Parallel Ports
Table 30
Alternate Functions of Port 3
Port 3 Pin
Alternate Function
P3.0
P3.1
P3.2
P3.3
P3.4
P3.5
P3.6
P3.7
P3.8
P3.9
P3.10
P3.11
P3.12
P3.13
P3.14
P3.15
------T3OUT
--T4IN
T3IN
T2IN
MRST
MTSR
TxD0
RxD0
BHE/WRH
SCLK
--CLKOUT
Alternate Function
No pin assigned
No pin assigned
No pin assigned
Timer 3 Toggle Output
No pin assigned!
Timer 4 Count Input (T3EUD Input, T2EUD Input)
Timer 3 Count Input
Timer 2 Count Input
SSC Master Receive / Slave Transmit
SSC Master Transmit / Slave Receive
ASC Transmit Data Output
ASC Receive Data Input
Byte High Enable / Write High Output
SSC Shift Clock Input/Output
No pin assigned
System Clock Output
a)
b)
P3.15
CLKOUT
P3.13
P3.12
P3.11
P3.10
P3.9
P3.8
P3.7
P3.6
P3.5
SCLK
BHE
RxD0Tx
D0
MTSR
MRST
T2IN
T3IN
T4IN
P3.3
T3OUT
No Pin
Port 3
No Pin
No Pin
No Pin
No Pin
WRH
General Purpose
Input/Output
Figure 35
Port 3 I/O and Alternate Functions
The port structure of the Port 3 pins depends on their alternate function (see figures
below).
When the on-chip peripheral associated with a Port 3 pin is configured to use the
alternate input function, it reads the input latch, which represents the state of the pin, via
Data Sheet
158
2001-02-23
C165UTAH
Parallel Ports
the line labeled “Alternate Data Input”. Port 3 pins with alternate input functions are:
T2IN, T3IN and T4IN/T3EUD/T2EUD.
When the on-chip peripheral associated with a Port 3 pin is configured to use the
alternate output function, its “Alternate Data Output” line is ANDed with the port output
latch line. When using these alternate functions, the user must set the direction of the
port line to output (DP3.y=1) and must set the port output latch (P3.y=1). Otherwise the
pin is in its high-impedance state (when configured as input) or the pin is stuck at '0'
(when the port output latch is cleared). When the alternate output functions are not used,
the “Alternate Data Output” line is in its inactive state, which is a high level ('1'). Port 3
pins with alternate output functions are:
T6OUT, T3OUT, TxD0 and CLKOUT.
When the on-chip peripheral associated with a Port 3 pin is configured to use both the
alternate input and output function, the descriptions above apply to the respective
current operating mode. The direction must be set accordingly. Port 3 pins with alternate
input/output functions are:
MTSR, MRST, RxD0 and SCLK.
Note: Enabling the CLKOUT function automatically enables the P3.15 output driver.
Setting bit DP3.15=’1’ is not required.
Data Sheet
159
2001-02-23
C165UTAH
Parallel Ports
Write ODP3.y
Open Drain
Latch
Read ODP3.y
Write DP3.y
Ι
n
t
e
r
n
a
l
Direction
Latch
Read DP3.y
Alternate
Data
Output
B
u
s
Write P3.y
&
Port Output
Latch
Output
Buffer
Read P3.y
Clock
1
MUX
0
y = 13, 11...5, 3
Figure 36
P3.y
Input
Latch
Alternate
Data
Input
MCB02229
Block Diagram of a Port 3 Pin with Alternate Input or Alternate
Output Function (y = 13, 11...5, 3)
Pin P3.12 (BHE/WRH) is one more pin with an alternate output function. However, its
structure is slightly different (see figure below), because after reset the BHE or WRH
function must be used depending on the system startup configuration. In these cases
there is no possibility to program any port latches before. Thus the appropriate alternate
function is selected automatically. If BHE/WRH is not used in the system, this pin can be
used for general purpose I/O by disabling the alternate function (BYTDIS = ‘1’ /
WRCFG=’0’).
Data Sheet
160
2001-02-23
C165UTAH
Parallel Ports
Write DP3.x
1
MUX
0
’1’
Direction
Latch
Read DP3.x
Ι
n
t
e
r
n
a
l
Write P3.x
Alternate
Function
Enable
Alternate
Data
Output
Port Output
Latch
B
u
s
P3.12/BHE
P3.15/CLKOUT
1
MUX
0
Output
Buffer
Read P3.x
Clock
1
MUX
0
Input
Latch
MCB02073
x = 15, 12
Figure 37
Block Diagram of Pins P3.15 (CLKOUT) and P3.12 (BHE/WRH)
Note: Enabling the BHE or WRH function automatically enables the P3.12 output driver.
Setting bit DP3.12=’1’ is not required.
During bus hold pin P3.12 is switched back to its standard function and is then
controlled by DP3.12 and P3.12. Keep DP3.12 = ’0’ in this case to ensure floating
in hold mode.
8.5
PORT4
If this 7-bit port is used for general purpose I/O, the direction of each line can be
configured via the corresponding direction register DP4.
Each port line can be switched into push/pull or open drain mode via the open drain
control register ODP4.
For port pins configured as input (via DP4 or alternate function), an internal pull transistor
is connected to the pad if register P4PUDEN = ’1’, no matter wheter the C165UTAH is
in normal operation mode or in power down mode. Either pulldown transistor or pullup
transistor will be selected via P4PUDSEL.
Data Sheet
161
2001-02-23
C165UTAH
Parallel Ports
For port pins configured as output, the internal pull transistors are always disabled. The
output driver is disabled in power down mode unless P4PHEN = ’1’.
After reset, P4PUDEN and P4PUDSEL are set to LOW signal.
P4 (FFC8H / E4H)
15
14
13
SFR
12
11
10
9
8
Reset Value: - - 00H
7
6
5
4
3
2
1
0
P4.6 P4.5 P4.4 P4.3 P4.2 P4.1 P4.0
-
-
-
-
-
-
-
Bit
Function
P4.y
Port data register P4 bit y
-
DP4 (FFCAH / E5H)
15
14
13
-
rw
rw
rw
SFR
12
11
10
9
8
rw
rw
rw
rw
Reset Value: - - 00H
7
6
5
4
3
2
1
0
DP4.6 DP4.5 DP4.4 DP4.3 DP4.2 DP4.1 DP4.0
-
-
-
-
-
-
-
-
-
rw
rw
rw
Bit
Function
DP4.y
Port direction register DP4 bit y
DP4.y = 0: Port line P4.y is an input (high-impedance)
DP4.y = 1: Port line P4.y is an output
ODP4 (F1CAH / E5H)
15
14
13
12
ESFR
11
10
9
8
7
rw
rw
rw
rw
Reset Value: - - 00H
6
5
4
3
2
1
0
ODP2 ODP2 ODP2 ODP2 ODP2 ODP2 ODP2
.6
.5
.4
.3
.2
.1
.0
-
-
-
-
-
-
-
-
-
rw
rw
rw
rw
Bit
Function
ODP4.y
Port 4 Open Drain control register bit y
ODP4.y = 0: Port line P4.y output driver in push/pull mode
ODP4.y = 1: Port line P4.y output driver in open drain mode
Data Sheet
162
rw
rw
rw
2001-02-23
C165UTAH
Parallel Ports
P4PUDSEL (FE84H / 42H)
SFR
Reset Value: - - 00H
15
14
13
12
11
10
9
8
7
-
-
-
-
-
-
-
-
-
6
5
4
3
2
1
0
P4
P4
P4
P4
P4
P4
P4
PUD PUD PUD PUD PUD PUD PUD
SEL.6 SEL.5 SEL.4 SEL.3 SEL.2 SEL.1 SEL.0
rw
rw
rw
rw
rw
rw
Bit
Function
P4PUDSEL.y
Pulldown/Pullup Selection
P4PUDSEL.y = 0: internal programmable pulldown transistor is selected
P4PUDSEL.y = 1: internal programmable pullup transistor is selected
P4PUDEN (FE86H / 43H)
SFR
Reset Value: - - 00H
15
14
13
12
11
10
9
8
7
-
-
-
-
-
-
-
-
-
6
5
4
3
2
1
0
P4
P4
P4
P4
P4
P4
P4
PUD PUD PUD PUD PUD PUD PUD
EN.6 EN.5 EN.4 EN.3 EN.2 EN.1 EN.0
rw
rw
rw
rw
rw
Bit
Function
P4PUDEN.y
Pulldown/Pullup Enable
P4PUDEN.y = 0: internal programmable pull transistor is disabled
P4PUDEN.y = 1: internal programmable pull transistor is enabled
P4PHEN (FE88H / 44H)
SFR
rw
rw
Reset Value: - - 00H
15
14
13
12
11
10
9
8
7
-
-
-
-
-
-
-
-
-
6
5
4
3
2
1
0
P4
P4
P4
P4
P4
P4
P4
PHEN PHEN PHEN PHEN PHEN PHEN PHEN
.6
.5
.4
.3
.2
.1
.0
rw
rw
rw
rw
Bit
Function
P4PHEN.y
Output Driver Enable in Power Down Mode
P4PHEN.y = 0: output driver is disabled in power down mode
P4PHEN.y = 1: output driver is enabled in power down mode
Data Sheet
rw
163
rw
rw
rw
2001-02-23
C165UTAH
Parallel Ports
8.5.1
Alternate Functions of PORT4
During external bus cycles that use segmentation (ie. an address space above
64 KByte) a number of Port 4 pins may output the segment address lines. The number
of pins that is used for segment address output determines the external address space
which is directly accessible. The other pins of Port 4 (if any) may be used for general
purpose I/O. If segment address lines are selected, the alternate function of Port 4 may
be necessary to access eg. external memory directly after reset. For this reason Port 4
will be switched to its alternate function automatically.
The number of segment address lines is selected via PORT0 during reset. The selected
value can be read from bitfield SALSEL in register RP0H (read only) eg. in order to check
the configuration during run time.
Table 31 summarizes the alternate functions of Port 4 depending on the number of
selected segment address lines (coded via bitfield SALSEL).
Table 31
Alternate Functions of Port 4
Port 4 Pin
Std. Function
Altern. Function
Altern. Function
Altern. Function
SALSEL=01 64 KB SALSEL=11 256 KB SALSEL=00 1 MB SALSEL=10 8 MB
P4.0
P4.1
P4.2
P4.3
P4.4
P4.5
P4.6
Gen. purpose I/O
Gen. purpose I/O
Gen. purpose I/O
Gen. purpose I/O
Gen. purpose I/O
Gen. purpose I/O
Gen. purpose I/O
Seg. Address A16
Seg. Address A17
Gen. purpose I/O
Gen. purpose I/O
Gen. purpose I/O
Gen. purpose I/O
Gen. purpose I/O
Seg. Address A16
Seg. Address A17
Seg. Address A18
Seg. Address A19
Gen. purpose I/O
Gen. purpose I/O
Gen. purpose I/O
Seg. Address A16
Seg. Address A17
Seg. Address A18
Seg. Address A19
Seg. Address A20
Seg. Address A21
Seg. Address A22
Alternate Function
Port 4
P4.6
P4.5
P4.4
P4.3
P4.2
P4.1
P4.0
A22
A21
A20
A19
A18
A17
A16
General Purpose
Input/Output
Figure 38
Data Sheet
Port 4 I/O and Alternate Functions
164
2001-02-23
C165UTAH
Parallel Ports
Write DP4.y
1
MUX
0
’1’
Direction
Latch
Ι
n
t
e
r
n
a
l
B
u
s
Read DP4.y
Write P4.y
Alternate
Function
Enable
Alternate
Data
Output
1
MUX
0
Port Output
Latch
Output
Buffer
P4.y
Read P4.y
Clock
1
MUX
0
Input
Latch
MCB02075
Figure 39
8.6
Block Diagram of a Port 4 Pin (y = 6...0)
PORT6
If this 8-bit port is used for general purpose I/O, the direction of each line can be
configured via the corresponding direction register DP6. Each port line can be switched
into push/pull or open drain mode via the open drain control register ODP6.
For port pins configured as input (via DP6 or alternate function), an internal pull transistor
is connected to the pad if register P6PUDEN = ’1’, no matter wheter the C165UTAH is
in normal operation mode or in power down mode. Either pulldown transistor or pullup
transistor will be selected via P6PUDSEL.
For port pins configured as output, the internal pull transistors are always disabled. The
output driver is disabled in power down mode unless P6PHEN = ’1’.
After reset, P6PUDEN and P6PUDSEL are set to LOW signal.
Data Sheet
165
2001-02-23
C165UTAH
Parallel Ports
P6 (FFCCH / E6H)
15
14
13
SFR
12
11
10
9
8
Reset Value: - - 00H
7
6
5
4
3
2
1
0
P6.7 P6.6 P6.5 P6.4 P6.3 P6.2 P6.1 P6.0
-
-
-
-
-
-
-
Bit
Function
P6.y
Port data register P6 bit y
-
DP6 (FFCEH / E7H)
15
14
13
rw
rw
rw
rw
SFR
12
11
10
9
8
rw
rw
rw
rw
Reset Value: - - 00H
7
6
5
4
3
2
1
0
DP6.7 DP6.6 DP6.5 DP6.4 DP6.3 DP6.2 DP6.1 DP6.0
-
-
-
-
-
-
-
-
rw
rw
rw
rw
Bit
Function
DP6.y
Port direction register DP6 bit y
DP6.y = 0: Port line P6.y is an input (high-impedance)
DP6.y = 1: Port line P6.y is an output
Data Sheet
166
rw
rw
rw
rw
2001-02-23
C165UTAH
Parallel Ports
ODP6 (F1CEH / E7H)
15
14
13
12
ESFR
11
10
9
8
7
Reset Value: - - 00H
6
5
4
3
2
1
0
ODP6 ODP6 ODP6 ODP6 ODP6 ODP6 ODP6 ODP6
.7
.6
.5
.4
.3
.2
.1
.0
-
-
-
-
-
-
-
-
rw
rw
rw
rw
rw
Bit
Function
ODP6.y
Port 6 Open Drain control register bit y
ODP6.y = 0: Port line P6.y output driver in push/pull mode
ODP6.y = 1: Port line P6.y output driver in open drain mode
P6PUDSEL (FE90H / 48H)
SFR
15
14
13
12
11
10
9
8
-
-
-
-
-
-
-
-
rw
rw
Reset Value: - - 00H
7
6
5
4
3
2
1
0
P6
P6
P6
P6
P6
P6
P6
P6
PUD PUD PUD PUD PUD PUD PUD PUD
SEL.7 SEL.6 SEL.5 SEL.4 SEL.3 SEL.2 SEL.1 SEL.0
rw
rw
rw
rw
rw
rw
rw
Bit
Function
P6PUDSEL.y
Pulldown/Pullup Selection
P6PUDSEL.y = 0: internal programmable pulldown transistor is selected
P6PUDSEL.y = 1: internal programmable pullup transistor is selected
P6PUDEN (FE92H / 49H)
SFR
15
14
13
12
11
10
9
8
-
-
-
-
-
-
-
-
rw
Reset Value: - - 00H
7
6
5
4
3
2
1
0
P6
P6
P6
P6
P6
P6
P6
P6
PUD PUD PUD PUD PUD PUD PUD PUD
EN.7 EN.6 EN.5 EN.4 EN.3 EN.2 EN.1 EN.0
rw
rw
rw
rw
rw
rw
Bit
Function
P6PUDEN.y
Pulldown/Pullup Enable
P6PUDEN.y = 0: internal programmable pull transistor is disabled
P6PUDEN.y = 1: internal programmable pull transistor is enabled
Data Sheet
rw
167
rw
rw
2001-02-23
C165UTAH
Parallel Ports
P6PHEN (FE94H / 4AH)
SFR
15
14
13
12
11
10
9
8
-
-
-
-
-
-
-
-
Reset Value: - - 00H
7
6
5
4
3
2
1
0
P6
P6
P6
P6
P6
P6
P6
P6
PHEN PHEN PHEN PHEN PHEN PHEN PHEN PHEN
.7
.6
.5
.4
.3
.2
.1
.0
rw
rw
rw
rw
rw
rw
Bit
Function
P6PHEN.y
Output Driver Enable in Power Down Mode
P6PHEN.y = 0: output driver is disabled in power down mode
P6PHEN.y = 1: output driver is enabled in power down mode
8.6.1
rw
rw
Alternate Functions of PORT6
A programmable number of chip select signals (CS4...CS0) derived from the bus control
registers (BUSCON4...BUSCON0) can be output on 5 pins of Port 6. The other 3 pins
may be used for bus arbitration to accomodate additional masters in a C165UTAH
system.
The number of chip select signals is selected via PORT0 during reset. The selected
value can be read from bitfield CSSEL in register RP0H (read only) eg. in order to check
the configuration during run time.
Table 32 summarizes the alternate functions of Port 6 depending on the number of
selected chip select lines (coded via bitfield CSSEL).
Table 32
Alternate Functions of Port 6
Port 6 Pin
Altern. Function
CSSEL = 10
Altern. Function
CSSEL = 01
Altern. Function
CSSEL = 00
Altern. Function
CSSEL = 11
P6.0
P6.1
P6.2
P6.3
P6.4
Gen. purpose I/O
Gen. purpose I/O
Gen. purpose I/O
Gen. purpose I/O
Gen. purpose I/O
Chip select CS0
Chip select CS1
Gen. purpose I/O
Gen. purpose I/O
Gen. purpose I/O
Chip select CS0
Chip select CS1
Chip select CS2
Gen. purpose I/O
Gen. purpose I/O
Chip select
Chip select
Chip select
Chip select
Chip select
P6.5
P6.6
P6.7
HOLD
HLDA
BREQ
Data Sheet
CS0
CS1
CS2
CS3
CS4
External hold request input
Hold acknowledge output
Bus request output
168
2001-02-23
C165UTAH
Parallel Ports
Alternate Function
Port 6
a)
P6.7
P6.6
P6.5
P6.4
P6.3
P6.2
P6.1
P6.0
BREQ
HLDA
HOLD
CS4
CS3
CS2
CS1
CS0
General Purpose
Input/Output
Figure 40
Port 6 I/O and Alternate Functions
The chip select lines of Port 6 additionally have an internal weak pullup device. This
device is switched on under the following conditions:
• always during reset
• if the Port 6 line is used as a chip select output, and the C165UTAH is in Hold mode
(invoked through HOLD), and the respective pin driver is in push/pull mode (ODP6.x
= ‘0’).
This feature is implemented to drive the chip select lines high during reset in order to
avoid multiple chip selection, and to allow another master to access the external memory
via the same chip select lines (Wired-AND), while the C165UTAH is in Hold mode.
With ODP6.x = ‘1’ (open drain output selected), the internal pullup device will not be
active during Hold mode; external pullup devices must be used in this case.
When entering Hold mode the CS lines are actively driven high for one clock phase, then
the output level is controlled by the pullup devices (if activated).
After reset the CS function must be used, if selected so. In this case there is no possibility
to program any port latches before. Thus the alternate function (CS) is selected
automatically in this case.
Note: The open drain output option can only be selected via software earliest during the
initialization routine; at least signal CS0 will be in push/pull output driver mode
directly after reset.
Data Sheet
169
2001-02-23
C165UTAH
Parallel Ports
Write ODP6.y
Open Drain
Latch
Read ODP6.y
Ι
n
t
e
r
n
a
l
B
u
s
Write DP6.y
1
MUX
0
’1’
Direction
Latch
Read DP6.y
Alternate
Function
Enable
Alternate
Data
Output
Write P6.y
Port Output
Latch
1
MUX
0
P6.y
Output
Buffer
Read P6.y
Clock
1
MUX
0
Input
Latch
MCB01982
Figure 41
Block Diagram of Port 6 Pins with an alternate output function
The bus arbitration signals HOLD, HLDA and BREQ are selected with bit HLDEN in
register PSW. When the bus arbitration signals are enabled via HLDEN, also these pins
are switched automatically to the appropriate direction. Note that the pin drivers for
HLDA and BREQ are automatically enabled, while the pin driver for HOLD is
automatically disabled.
Data Sheet
170
2001-02-23
C165UTAH
Parallel Ports
Write ODP6.y
Open Drain
Latch
Read ODP6.y
Ι
n
t
e
r
n
a
l
B
u
s
Write DP6.y
Direction
Latch
Read DP6.y
Write P6.y
Port Output
Latch
Output
Buffer
Read P6.y
Clock
1
MUX
0
Input
Latch
Alternate
Data Input
Figure 42
8.7
P6.5/HOLD
MCB01983
Block Diagram of Pin P6.5 (HOLD)
PORT7
In the C165UTAH Port 7 is an 6-bit general purpose I/O port. The direction of each line
can be configured via the corresponding direction register DP7. Each port line can be
switched into push/pull or open drain mode via the open drain control register ODP7.
Note: There are no alternate functions for Port 7.
For port pins configured as input via DP7, an internal pull transistor is connected to the
pad if register P7PUDEN = ’1’, no matter wheter the C165UTAH is in normal operation
mode or in power down mode. Either pulldown transistor or pullup transistor will be
selected via P7PUDSEL.
Data Sheet
171
2001-02-23
C165UTAH
Parallel Ports
For port pins configured as output, the internal pull transistors are always disabled. The
output driver is disabled in power down mode unless P7PHEN = ’1’.
After reset, P7PUDEN and P7PUDSEL are set to LOW signal.
P7 (FFD0H / E8H)
15
14
13
SFR
12
11
10
9
8
Reset Value: - - 00H
7
6
5
4
3
2
1
0
P7.5 P7.4 P7.3 P7.2 P7.1 P7.0
-
-
-
-
-
-
-
Bit
Function
P7.y
Port data register P7 bit y
-
DP7 (FFD2H / E9H)
15
14
13
-
-
rw
rw
SFR
12
11
10
9
8
rw
rw
rw
rw
Reset Value: - - 00H
7
6
5
4
3
2
1
0
DP7.5 DP7.4 DP7.3 DP7.2 DP7.1 DP7.0
-
-
-
-
-
-
-
-
-
-
rw
rw
Bit
Function
DP7.y
Port direction register DP7 bit y
DP7.y = 0: Port line P7.y is an input (high-impedance)
DP7.y = 1: Port line P7.y is an output
ODP7 (F1D2H / E9H)
15
14
13
ESFR
12
11
10
9
8
7
rw
rw
rw
rw
Reset Value: - - 00
6
5
4
3
2
1
0
ODP7 ODP7 ODP7 ODP7 ODP7 ODP7
.5
.4
.3
.2
.1
.0
-
-
-
-
-
-
-
-
-
-
rw
rw
rw
Bit
Function
ODP7.y
Port 7 Open Drain control register bit y
ODP7.y = 0: Port line P7.y output driver in push/pull mode
ODP7.y = 1: Port line P7.y output driver in open drain mode
Data Sheet
172
rw
rw
rw
2001-02-23
C165UTAH
Parallel Ports
P7PUDSEL (FE96H / 4BH)
SFR
Reset Value: - - 00H
15
14
13
12
11
10
9
8
7
6
-
-
-
-
-
-
-
-
-
-
5
4
3
2
1
0
P7
P7
P7
P7
P7
P7
PUD PUD PUD PUD PUD PUD
SEL.5 SEL.4 SEL.3 SEL.2 SEL.1 SEL.0
rw
rw
rw
rw
rw
Bit
Function
P7PUDSEL.y
Pulldown/Pullup Selection
P7PUDSEL.y = 0: internal programmable pulldown transistor is selected
P7PUDSEL.y = 1: internal programmable pullup transistor is selected
P7PUDEN (FE98H / 4CH)
SFR
Reset Value: - - 00H
15
14
13
12
11
10
9
8
7
6
-
-
-
-
-
-
-
-
-
-
5
4
3
2
1
0
P7
P7
P7
P7
P7
P7
PUD PUD PUD PUD PUD PUD
EN.5 EN.4 EN.3 EN.2 EN.1 EN.0
rw
rw
rw
rw
Bit
Function
P7PUDEN.y
Pulldown/Pullup Enable
P7PUDEN.y = 0: internal programmable pull transistor is disabled
P7PUDEN.y = 1: internal programmable pull transistor is enabled
P7PHEN (FE9AH / 4DH)
SFR
rw
rw
Reset Value: - - 00H
15
14
13
12
11
10
9
8
7
6
-
-
-
-
-
-
-
-
rw
rw
5
4
3
2
1
0
P6
P6
P6
P6
P6
P6
PHEN PHEN PHEN PHEN PHEN PHEN
.5
.4
.3
.2
.1
.0
rw
rw
rw
Bit
Function
P7PHEN.y
Output Driver Enable in Power Down Mode
P7PHEN.y = 0: output driver is disabled in power down mode
P7PHEN.y = 1: output driver is enabled in power down mode
Data Sheet
rw
173
rw
rw
rw
2001-02-23
C165UTAH
Dedicated Pins
9
Dedicated Pins
Most of the input/output or control signals of the functional the C165UTAH are realized
as alternate functions of pins of the parallel ports. There is, however, a number of signals
that use separate pins, including the USB interface, the IOM-2 interface, the oscillator,
special control signals and the power supply. Table 33 summarizes all dedicated pins of
the C165UTAH.
Table 33
Dedicated Pins
Pin(s)
Function
ALE
Address Latch Enable
RD
External Read Strobe
WR/WRL
External Write/Write Low Strobe
READY
Ready Input
EA
External Access Enable
NMI
Non-Maskable Interrupt Input
RSTIN
Reset Input
RSTOUT
Reset Output
XTAL1, XTAL2
Oscillator Input/Output
CLKMODE
Oscillator Clock Input Mode Select
DMNS, DPLS
USB
DU, DD, DCL, FSC
IOM-2
BRKIN, BRKOUT
OCDS
TDI, TDO, TCK, TMS, TRST
JTAG Interface
TEST
Test Mode Enable
VDD, GND
Power Supply and Ground (19 pins VDD, 19 pins GND)
The Address Latch Enable signal ALE controls external address latches that provide
a stable address in multiplexed bus modes.
ALE is activated for every external bus cycle independent of the selected bus mode, ie.
it is also activated for bus cycles with a demultiplexed address bus. When an external
bus is enabled (one or more of the BUSACT bits set) also X-Peripheral accesses will
generate an active ALE signal.
ALE is not activated for internal accesses, ie. the internal RAM and the special function
registers. In single chip mode, ie. when no external bus is enabled (no BUSACT bit set),
ALE will also remain inactive for X-Peripheral accesses.
Data Sheet
174
2001-02-23
C165UTAH
Dedicated Pins
External Read Strobe RD controls the output drivers of external memory or peripherals
when the C165UTAH reads data from these external devices. During reset and during
Hold mode an internal pullup ensures an inactive (high) level on the RD output.
External Write Strobe WR/WRL controls the data transfer from the C165UTAH to an
external memory or peripheral device. This pin may either provide an general WR signal
activated for both byte and word write accesses, or specifically control the low byte of an
external 16-bit device (WRL) together with the signal WRH (alternate function of P3.12/
BHE). During reset and during Hold mode an internal pullup ensures an inactive (high)
level on the WR/WRL output.
Note: Whether RD and WR/WRL remain idle during X-peripheral accesses depends on
the value of bit VISIBLE of register SYSCON.
Ready Input READY receives a control signal from an external memory or peripheral
device that is used to terminate an external bus cycle, provided that this function is
enabled for the current bus cycle. READY may be used as synchronous READY or may
be evaluated asynchronously. When waitstates are defined for a READY controlled
address window the READY input is not evaluated during these waitstates.
External Access Enable Pin EA is dedicated for on-chip ROM derivates. In this case it
determines, if the chip after reset starts fetching code from the internal ROM area
(EA=’1’) or via the external bus interface (EA=’0’). For the ROM-less C165UTAH be
sure to hold this input low.
Non-Maskable Interrupt Input NMI allows to trigger a high priority trap via an external
signal (eg. a power-fail signal). It also serves to validate the PWRDN instruction that
switches the C165UTAH into Power-Down mode. The NMI pin is sampled with every
CPU clock cycle to detect transitions.
Oscillator Input XTAL1 and Output XTAL2 connect the internal Pierce oscillator to the
external crystal. The oscillator provides an inverter and a feedback element. The
standard external oscillator circuitry (see figure below) comprises the crystal, two low
end capacitors and series resistor to limit the current through the crystal. The additional
LC combination is only required for 3rd overtone crystals to suppress oscillation in the
fundamental mode. A test resistor (RQ) may be temporarily inserted to measure the
oscillation allowance of the oscillator circuitry.
An external clock signal may be fed to the input XTAL1, leaving XTAL2 open.
Note: It is strongly recommended to measure the oscillation allowance (or margin) in the
final target system (layout) to determine the optimum parameters for the oscillator
operation.
Data Sheet
175
2001-02-23
C165UTAH
Dedicated Pins
The following starting configuration is recommended to be used for the C165UTAH:
Quarz: CL = 30 pF (max.), RS = 70 Ohm (max.), Accuracy: 96 ppm or better
External: Circuitry: CA = CB = 47 pF (max.), no serial resistor (Rx2 = 0)
Note: Please check the Infineon Application Notes in addition to this recommendation.
XTAL1
XTAL2
RQ
CA
Figure 43
Rx2
CB
External Oscillator Circuitry
The Clock Mode Select CLKMODE CLKMODE must be LOW if an external crystal is
used. HIGH signal enables the direct clock input path and switches the internal oscillator
in power down mode..
The Reset Input RSTIN allows to put the C165UTAH into the well defined reset
condition either at power-up or external events like a hardware failure or manual reset.
The input voltage threshold of the RSTIN pin is raised compared to the standard pins in
order to minimize the noise sensitivity of the reset input.
The Reset Output RSTOUT provides a special reset signal for external circuitry.
RSTOUT is activated at the beginning of the reset sequence, triggered via RSTIN, a
watchdog timer overflow or by the SRST instruction. RSTOUT remains active (low) until
the EINIT instruction is executed. This allows to initialize the controller before the
external circuitry is activated.
The Power Supply pins VDD and GND provide the power supply for the digital logic of
the C165UTAH. The respective VCC/VSS pairs should be decoupled as close to the pins
as possible. For best results it is recommended to implement two-level decoupling, eg.
(the widely used) 100 nF in parallel with 30...40 pF capacitors which deliver the peak
currents.
Note: All VDD pins and all GND pins must be connected to the power supply and ground,
respectively.
Data Sheet
176
2001-02-23
C165UTAH
External Bus Interface
10
External Bus Interface
Although the C165UTAH provides a powerful set of on-chip peripherals and on-chip
RAM areas, these internal units only cover a small fraction of its address space of up to
8 MByte. The external bus interface allows to access external peripherals and additional
volatile and non-volatile memory. The external bus interface provides a number of
configurations, so it can be taylored to fit perfectly into a given application system.
Ports & Direction Control
Alternate Functions
Address Registers
Mode Registers
P0L / P0H
BUSCON0
SYSCON
RP0H
P1L / P1H
ADDRSEL1
BUSCON1
DP3
ADDRSEL2
BUSCON2
P3
ADDRSEL3
BUSCON3
P4
ADDRSEL4
BUSCON4
ODP6
DP6
P6
P0L/P0H
P1L/P1H
DP3
P3
P4
ODP6
DP6
P6
Figure 44
PORT0
PORT1
ALE
RD
WR/WRL
BHE/WRH
PORT0 Data Registers
PORT1 Data Registers
Port 3 Direction Control Register
Port 3 Data Register
Port 4 Data Register
Port 6 Open Drain Control Register
Port 6 Direction Control Register
Port 6 Data Register
Control Registers
EA
RSTIN
READY
ADDRSELx
BUSCONx
SYSCON
RP0H
Address Range Select Register 1...4
Bus Mode Control Register 0...4
System Control Register
Port P0H Reset Configuration Register
Control Registers
SFRs and Port Pins Associated with the External Bus Interface
Accesses to external memory or peripherals are executed by the integrated External Bus
Controller (EBC). The function of the EBC is controlled via the SYSCON register and the
BUSCONx and ADDRSELx registers. The BUSCONx registers specify the external bus
cycles in terms of address (mux/demux), data (16-bit/8-bit), chip selects and length
(waitstates / READY control / ALE / RW delay). These parameters are used for accesses
within a specific address area which is defined via the corresponding register
ADDRSELx.
The four pairs BUSCON1/ADDRSEL1...BUSCON4/ADDRSEL4 allow to define four
independent “address windows”, while all external accesses outside these windows are
controlled via register BUSCON0.
Data Sheet
177
2001-02-23
C165UTAH
External Bus Interface
Single Chip Mode
Single chip mode is entered, when pin EA is high during reset. In this case register
BUSCON0 is initialized with 0000H, which also resets bit BUSACT0, so no external bus
is enabled.
In single chip mode the C165UTAH operates only with and out of internal resources. No
external bus is configured and no external peripherals and/or memory can be accessed.
Also no port lines are occupied for the bus interface. When running in single chip mode,
however, external access may be enabled by configuring an external bus under software
control.
Note: Any attempt to access a location in the external memory space in single chip mode
results in the hardware trap ILLBUS.
10.1
External Bus Modes
When the external bus interface is enabled (bit BUSACTx=’1’) and configured (bitfield
BTYP), the C165UTAH uses a subset of its port lines together with some control lines to
build the external bus.
BTYP Encoding External Data Bus Width
External Address Bus Mode
00
8-bit Data
Demultiplexed Addresses
01
8-bit Data
Multiplexed Addresses
10
16-bit Data
Demultiplexed Addresses
11
16-bit Data
Multiplexed Addresses
The bus configuration (BTYP) for the address windows (BUSCON4...BUSCON1) is
selected via software typically during the initialization of the system.
The bus configuration (BTYP) for the default address range (BUSCON0) is selected via
PORT0 during reset, provided that pin EA is low during reset. Otherwise BUSCON0 may
be programmed via software just like the other BUSCON registers.
The 16 MByte address space of the C165UTAH is divided into 256 segments of 64 KByte
each. The 16-bit intra-segment address is output on PORT0 for multiplexed bus modes
or on PORT1 for demultiplexed bus modes. When segmentation is disabled, only one 64
KByte segment can be used and accessed. Otherwise additional address lines may be
output on Port 4, and/or several chip select lines may be used to select different memory
banks or peripherals. These functions are selected during reset via bitfields SALSEL and
CSSEL of register RP0H, respectively.
Note: Bit SGTDIS of register SYSCON defines, if the CSP register is saved during
interrupt entry (segmentation active) or not (segmentation disabled).
Data Sheet
178
2001-02-23
C165UTAH
External Bus Interface
Multiplexed Bus Modes
In the multiplexed bus modes the 16-bit intra-segment address as well as the data use
PORT0. The address is time-multiplexed with the data and has to be latched externally.
The width of the required latch depends on the selected data bus width, ie. an 8-bit data
bus requires a byte latch (the address bits A15...A8 on P0H do not change, while P0L
multiplexes address and data), a 16-bit data bus requires a word latch (the least
significant address line A0 is not relevant for word accesses).
The upper address lines (An...A16) are permanently output on Port 4 (if segmentation is
enabled) and do not require latches.
The EBC initiates an external access by generating the Address Latch Enable signal
(ALE) and then placing an address on the bus. The falling edge of ALE triggers an
external latch to capture the address. After a period of time during which the address
must have been latched externally, the address is removed from the bus. The EBC now
activates the respective command signal (RD, WR, WRL, WRH). Data is driven onto the
bus either by the EBC (for write cycles) or by the external memory/peripheral (for read
cycles). After a period of time, which is determined by the access time of the memory/
peripheral, data become valid.
Read cycles: Input data is latched and the command signal is now deactivated. This
causes the accessed device to remove its data from the bus which is then tri-stated
again.
Write cycles: The command signal is now deactivated. The data remain valid on the bus
until the next external bus cycle is started.
Data Sheet
179
2001-02-23
C165UTAH
External Bus Interface
Bus Cycle
Segment (P4)
Address
ALE
BUS (P0)
Address
Data/Instr.
Address
Data
RD
BUS (P0)
WR
MCT02060
Figure 45
Multiplexed Bus Cycle
Demultiplexed Bus Modes
In the demultiplexed bus modes the 16-bit intra-segment address is permanently output
on PORT1, while the data uses PORT0 (16-bit data) or P0L (8-bit data).
The upper address lines are permanently output on Port 4 (if selected via SALSEL during
reset). No address latches are required.
The EBC initiates an external access by placing an address on the address bus. After a
programmable period of time the EBC activates the respective command signal (RD,
WR, WRL, WRH). Data is driven onto the data bus either by the EBC (for write cycles)
or by the external memory/peripheral (for read cycles). After a period of time, which is
determined by the access time of the memory/peripheral, data become valid.
Read cycles: Input data is latched and the command signal is now deactivated. This
causes the accessed device to remove its data from the data bus which is then tri-stated
again.
Write cycles: The command signal is now deactivated. If a subsequent external bus
cycle is required, the EBC places the respective address on the address bus. The data
remain valid on the bus until the next external bus cycle is started.
Data Sheet
180
2001-02-23
C165UTAH
External Bus Interface
Bus Cycle
Address (P1)
Segment (P4)
Address
ALE
BUS (P0)
Data/Instr.
RD
BUS (P0)
Data
WR
MCD02061
Figure 46
Demultiplexed Bus Cycle
Switching between the Bus Modes
The EBC allows to switch between different bus modes dynamically, ie. subsequent
external bus cycles may be executed in different ways. Certain address areas may use
multiplexed or demultiplexed buses or use READY control or predefined waitstates.
A change of the external bus characteristics can be initiated in two different ways:
Reprogramming the BUSCON and/or ADDRSEL registers allows to either change
the bus mode for a given address window, or change the size of an address window that
uses a certain bus mode. Reprogramming allows to use a great number of different
address windows (more than BUSCONs are available) on the expense of the overhead
for changing the registers and keeping appropriate tables.
Switching between predefined address windows automatically selects the bus mode
that is associated with the respective window. Predefined address windows allow to use
different bus modes without any overhead, but restrict their number to the number of
BUSCONs. However, as BUSCON0 controls all address areas, which are not covered
by the other BUSCONs, this allows to have gaps between these windows, which use the
bus mode of BUSCON0.
PORT1 will output the intra-segment address, when any of the BUSCON registers
selects a demultiplexed bus mode, even if the current bus cycle uses a multiplexed bus
Data Sheet
181
2001-02-23
C165UTAH
External Bus Interface
mode. This allows to have an external address decoder connected to PORT1 only, while
using it for all kinds of bus cycles.
Note: Never change the configuration for an address area that currently supplies the
instruction stream. Due to the internal pipelining it is very difficult to determine the
first instruction fetch that will use the new configuration. Only change the
configuration for address areas that are not currently accessed. This applies to
BUSCON registers as well as to ADDRSEL registers.
The usage of the BUSCON/ADDRSEL registers is controlled via the issued addresses.
When an access (code fetch or data) is initiated, the respective generated physical
address defines, if the access is made internally, uses one of the address windows
defined by ADDRSEL4...1, or uses the default configuration in BUSCON0. After
initializing the active registers, they are selected and evaluated automatically by
interpreting the physical address. No additional switching or selecting is necessary
during run time, except when more than the four address windows plus the default is to
be used.
Switching from demultiplexed to multiplexed bus mode represents a special case.
The bus cycle is started by activating ALE and driving the address to Port 4 and PORT1
as usual, if another BUSCON register selects a demultiplexed bus. However, in the
multiplexed bus modes the address is also required on PORT0. In this special case the
address on PORT0 is delayed by one CPU clock cycle, which delays the complete
(multiplexed) bus cycle and extends the corresponding ALE signal (see figure below).
This extra time is required to allow the previously selected device (via demultiplexed bus)
to release the data bus, which would be available in a demultiplexed bus cycle.
Data Sheet
182
2001-02-23
C165UTAH
External Bus Interface
Demultiplexed
Bus Cycle
Address (P1)
Segment (P4)
Multiplexed
Bus Cycle
Idle State
Address
Address
ALE
BUS (P0)
Address
Data/Instr.
Data/Instr.
RD
BUS (P0)
Data
Address
Data
WR
MCD02234
Figure 47
Switching from Demultiplexed to Multiplexed Bus Mode
External Data Bus Width
The EBC can operate on 8-bit or 16-bit wide external memory/peripherals. A 16-bit data
bus uses PORT0, while an 8-bit data bus only uses P0L, the lower byte of PORT0. This
saves on address latches, bus transceivers, bus routing and memory cost on the
expense of transfer time. The EBC can control word accesses on an 8-bit data bus as
well as byte accesses on a 16-bit data bus.
Word accesses on an 8-bit data bus are automatically split into two subsequent byte
accesses, where the low byte is accessed first, then the high byte. The assembly of bytes
to words and the disassembly of words into bytes is handled by the EBC and is
transparent to the CPU and the programmer.
Byte accesses on a 16-bit data bus require that the upper and lower half of the memory
can be accessed individually. In this case the upper byte is selected with the BHE signal,
while the lower byte is selected with the A0 signal. So the two bytes of the memory can
be enabled independent from each other, or together when accessing words.
When writing bytes to an external 16-bit device, which has a single CS input, but two WR
enable inputs (for the two bytes), the EBC can directly generate these two write control
Data Sheet
183
2001-02-23
C165UTAH
External Bus Interface
signals. This saves the external combination of the WR signal with A0 or BHE. In this
case pin WR serves as WRL (write low byte) and pin BHE serves as WRH (write high
byte). Bit WRCFG in register SYSCON selects the operating mode for pins WR and
BHE. The respective byte will be written on both data bus halfs.
When reading bytes from an external 16-bit device, whole words may be read and the
C165UTAH automatically selects the byte to be input and discards the other. However,
care must be taken when reading devices that change state when being read, like
FIFOs, interrupt status registers, etc. In this case individual bytes should be selected
using BHE and A0.
Bus Mode
Transfer Rate (Speed factor
for byte/word/dword access)
System Requirements
Free I/O
Lines
8-bit Multiplexed
Very low
( 1.5 / 3 / 6 )
Low (8-bit latch, byte bus)
P1H, P1L
8-bit Demultipl.
Low
(1/2/4)
Very low (no latch, byte bus) P0H
16-bit Multiplexed
High
( 1.5 / 1.5 / 3 )
High (16-bit latch, word bus) P1H, P1L
16-bit Demultipl.
Very high
(1/1/2)
Low (no latch, word bus)
---
Note: PORT1 gets available for general purpose I/O, when none of the BUSCON
registers selects a demultiplexed bus mode.
Disable/Enable Control for Pin BHE (BYTDIS)
Bit BYTDIS is provided for controlling the active low Byte High Enable (BHE) pin. The
function of the BHE pin is enabled, if the BYTDIS bit contains a '0'. Otherwise, it is
disabled and the pin can be used as standard I/O pin. The BHE pin is implicitly used by
the External Bus Controller to select one of two byte-organized memory chips, which are
connected to the C165UTAH via a word-wide external data bus. After reset the BHE
function is automatically enabled (BYTDIS = '0'), if a 16-bit data bus is selected during
reset, otherwise it is disabled (BYTDIS=’1’). It may be disabled, if byte access to 16-bit
memory is not required, and the BHE signal is not used.
Segment Address Generation
During external accesses the EBC generates a (programmable) number of address lines
on Port 4, which extend the 16-bit address output on PORT0 or PORT1, and so increase
the accessible address space. The number of segment address lines is selected during
reset and coded in bit field SALSEL in register RP0H (see table below).
SALSEL
Segment Address Lines
Directly accessible Address Space
11
Two:
A17...A16
256
KByte (Default without pull-downs)
10
Seven:
A22...A16
8
MByte (Maximum)
Data Sheet
184
2001-02-23
C165UTAH
External Bus Interface
SALSEL
Segment Address Lines
Directly accessible Address Space
01
None
64
KByte (Minimum)
00
Four:
1
MByte
A19...A16
Note: The total accessible address space may be increased by accessing several banks
which are distinguished by individual chip select signals.
CS Signal Generation
During external accesses the EBC can generate a (programmable) number of CS lines
on Port 6, which allow to directly select external peripherals or memory banks without
requiring an external decoder. The number of CS lines is selected during reset and
coded in bit field CSSEL in register RP0H (see table below).
CSSEL
Chip Select Lines
Note
11
Five:
Default without pull-downs
10
None
01
Two:
CS1...CS0
00
Three:
CS2...CS0
CS4...CS0
Port 6 pins free for I/O
The CSx outputs are associated with the BUSCONx registers and are driven active (low)
for any access within the address area defined for the respective BUSCON register. For
any access outside this defined address area the respective CSx signal will go inactive
(high). At the beginning of each external bus cycle the corresponding valid CS signal is
determined and activated. All other CS lines are deactivated (driven high) at the same
time.
Note: The CSx signals will not be updated for an access to any internal address area (ie.
when no external bus cycle is started), even if this area is covered by the
respective ADDRSELx register. An access to an on-chip X-Peripheral deactivates
all external CS signals.
Upon accesses to address windows without a selected CS line all selected CS
lines are deactivated.
The chip select signals allow to be operated in four different modes, which are selected
via bits CSWENx and CSRENx in the respective BUSCONx register.
CSWENx CSRENx
Chip Select Mode
0
0
Address Chip Select (Default after Reset, mode for CS0)
0
1
Read Chip Select
1
0
Write Chip Select
1
1
Read/Write Chip Select
Data Sheet
185
2001-02-23
C165UTAH
External Bus Interface
Address Chip Select signals remain active until an access to another address window.
An address chip select becomes active with the falling edge of ALE and becomes
inactive with the falling edge of ALE of an external bus cycle that accesses a different
address area. No spikes will be generated on the chip select lines.
Read or Write Chip Select signals remain active only as long as the associated control
signal (RD or WR) is active. This also includes the programmable read/write delay. Read
chip select is only activated for read cycles, write chip select is only activated for write
cycles, read/write chip select is activated for both read and write cycles (write cycles are
assumed, if any of the signals WRH or WRL gets active). These modes save external
glue logic, when accessing external devices like latches or drivers that only provide a
single enable input.
Note: CS0 provides an address chip select directly after reset (except for single chip
mode) when the first instruction is fetched.
Internal pullup devices hold all CS lines high during reset. After the end of a reset
sequence the pullup devices are switched off and the pin drivers control the pin levels
on the selected CS lines. Not selected CS lines will enter the high-impedance state and
are available for general purpose I/O.
The pullup devices are also active during bus hold on the selected CS lines, while HLDA
is active and the respective pin is switched to push/pull mode. Open drain outputs will
float during bus hold. In this case external pullup devices are required or the new bus
master is responsible for driving appropriate levels on the CS lines.
Segment Address versus Chip Select
The external bus interface of the C165UTAH supports many configurations for the
external memory. By increasing the number of segment address lines the C165UTAH
can address a linear address space of 256 KByte, 1 MByte or 8 MByte. This allows to
implement a large sequential memory area, and also allows to access a great number of
external devices, using an external decoder. By increasing the number of CS lines the
C165UTAH can access memory banks or peripherals without external glue logic. These
two features may be combined to optimize the overall system performance. Enabling 4
segment address lines and 5 chip select lines eg. allows to access five memory banks
of 8 MByte each. So the available address space is 40 MByte (without glue logic).
Note: Bit SGTDIS of register SYSCON defines, if the CSP register is saved during
interrupt entry (segmentation active) or not (segmentation disabled).
Data Sheet
186
2001-02-23
C165UTAH
External Bus Interface
10.2
Programmable Bus Characteristics
Important timing characteristics of the external bus interface have been made user
programmable to allow to adapt it to a wide range of different external bus and memory
configurations with different types of memories and/or peripherals.
The following parameters of an external bus cycle are programmable:
• ALE Control defines the ALE signal length and the address hold time after its falling edge
• Memory Cycle Time (extendable with 1...15 waitstates) defines the allowable access time
• Memory Tri-State Time (extendable with 1 waitstate) defines the time for a data driver to float
• Read/Write Delay Time defines when a command is activated after the falling edge of ALE
• READY Control defines, if a bus cycle is terminated internally or externally
Note: Internal accesses are executed with maximum speed and therefore are not
programmable.
External acceses use the slowest possible bus cycle after reset. The bus cycle
timing may then be optimized by the initialization software.
ALE
ADDR
RD/WR
DATA
ALE
ADDR
RD/WR
DATA
ALECTL
Figure 48
Data Sheet
MCTC
MTTC
MCD02225
Programmable External Bus Cycle
187
2001-02-23
C165UTAH
External Bus Interface
ALE Length Control
The length of the ALE signal and the address hold time after its falling edge are
controlled by the ALECTLx bits in the BUSCON registers. When bit ALECTL is set to ‘1’,
external bus cycles accessing the respective address window will have their ALE signal
prolonged by half a CPU clock. Also the address hold time after the falling edge of ALE
(on a multiplexed bus) will be prolonged by half a CPU clock, so the data transfer within
a bus cycle refers to the same CLKOUT edges as usual (ie. the data transfer is delayed
by one CPU clock). This allows more time for the address to be latched.
Note: ALECTL0 is ‘1’ after reset to select the slowest possible bus cycle, the other
ALECTLx are ‘0’ after reset.
Normal Multiplexed
Bus Cycle
Lengthened Multiplexed
Bus Cycle
Address
Address
Segment
(P4)
ALE
Setup
BUS (P0)
Address
Data/Instr.
Hold
Address
Data/Instr.
RD
BUS (P0)
Address
Data
Address
Data
WR
MCD02235
Figure 49
ALE Length Control
Programmable Memory Cycle Time
The C165UTAH allows the user to adjust the controller's external bus cycles to the
access time of the respective memory or peripheral. This access time is the total time
required to move the data to the destination. It represents the period of time during which
the controller’s signals do not change.
Data Sheet
188
2001-02-23
C165UTAH
External Bus Interface
Bus Cycle
Segment
Address
ALE
BUS (P0)
Data/Instr.
Address
RD
BUS (P0)
Data
Address
WR
MCTC Wait States (1...15)
Figure 50
MCT02063
Memory Cycle Time
The external bus cycles of the C165UTAH can be extended for a memory or peripheral,
which cannot keep pace with the controller’s maximum speed, by introducing wait states
during the access (see figure above). During these memory cycle time wait states, the
CPU is idle, if this access is required for the execution of the current instruction.
The memory cycle time wait states can be programmed in increments of one CPU clock
within a range from 0 to 15 (default after reset) via the MCTC fields of the BUSCON
registers. 15- waitstates will be inserted.
Programmable Memory Tri-State Time
The C165UTAH allows the user to adjust the time between two subsequent external
accesses to account for the tri-state time of the external device. The tri-state time defines,
when the external device has released the bus after deactivation of the read command
(RD).
Data Sheet
189
2001-02-23
C165UTAH
External Bus Interface
Bus Cycle
Segment
Address
ALE
BUS (P0)
Address
Data/Instr.
RD
MTTC Wait State
MCT02065
Figure 51
Memory Tri-State Time
The output of the next address on the external bus can be delayed for a memory or
peripheral, which needs more time to switch off its bus drivers, by introducing a wait state
after the previous bus cycle (see figure above). During this memory tri-state time wait
state, the CPU is not idle, so CPU operations will only be slowed down if a subsequent
external instruction or data fetch operation is required during the next instruction cycle.
The memory tri-state time waitstate requires one CPU clock (28 ns at fCPU = 36 MHz)
and is controlled via the MTTCx bits of the BUSCON registers. A waitstate will be
inserted, if bit MTTCx is ‘0’ (default after reset).
Note: External bus cycles in multiplexed bus modes implicitly add one tri-state time
waitstate in addition to the programmable MTTC waitstate.
Read/Write Signal Delay
The C165UTAH allows the user to adjust the timing of the read and write commands to
account for timing requirements of external peripherals. The read/write delay controls the
time between the falling edge of ALE and the falling edge of the command. Without read/
write delay the falling edges of ALE and command(s) are coincident (except for
propagation delays). With the delay enabled, the command(s) become active half a CPU
clock after the falling edge of ALE.
The read/write delay does not extend the memory cycle time, and does not slow down
the controller in general. In multiplexed bus modes, however, the data drivers of an
external device may conflict with the C165UTAH’s address, when the early RD signal is
used. Therefore multiplexed bus cycles should always be programmed with read/write
delay.
Data Sheet
190
2001-02-23
C165UTAH
External Bus Interface
Bus Cycle
Segment
Address
ALE
BUS (P0)
Data/Instr.
1)
RD
BUS (P0)
Data
Address
WR
Read/Write
Delay
1)
The Data drivers from the previous bus cycle should be disabled when the RD signal becomes active.
MCT02066
Figure 52
Read/Write Delay
The read/write delay is controlled via the RWDCx bits in the BUSCON registers. The
command(s) will be delayed, if bit RWDCx is ‘0’ (default after reset).
Early WR Signal Deactivation
The duration of an external write access can be shortened by one TCL. The WR signal
is activated (driven low) in the standard way, but can be deactivated (driven high) one
TCL earlier than defined in the standard timing. In this case, also the data output drivers
will be deactivated one TCL earlier.
This is especially useful in systems which operate on higher CPU clock frequencies and
employ external modules (memories, peripherals, etc.) which switch on their own data
drivers very fast in response to e.g. a chip select signal.
Conflicts between the C165UTAH and the external peripheral’s output drivers can be
avoided then by selecting early WR for the C165UTAH.
Note: Make sure that the reduced WR low time then still matches the requirements of
the external peripheral/memory.
Data Sheet
191
2001-02-23
C165UTAH
External Bus Interface
Early WR deactivation is controlled via the EWENx bits in the BUSCON registers (see
page 195). The WR signal will be shortened if bit EWENx is set to ’1’ signal. Default after
reset is a standard WR signal (EWENx = ’0’).
10.3
READY Controlled Bus Cycles
For situations, where the programmable waitstates are not enough, or where the
response (access) time of a peripheral is not constant, the C165UTAH provides external
bus cycles that are terminated via a READY input signal (synchronous or
asynchronous). In this case the C165UTAH first inserts a programmable number of
waitstates (0...7) and then monitors the READY line to determine the actual end of the
current bus cycle. The external device drives READY low in order to indicate that data
have been latched (write cycle) or are available (read cycle).
Bus Cycle
with active READY
Bus Cycle
extended via READY
1. WS 2. WS
1. WS 2. WS
ALE
RD/WR
SREADY
AREADY
MCD02237
: Evaluation (sampling) of the READY input
Figure 53
READY Controlled Bus Cycles
The READY function is enabled via the RDYENx bits in the BUSCON registers. When
this function is selected (RDYENx = ‘1’), only the lower 3 bits of the respective MCTC bit
field define the number of inserted waitstates (0...7), while the MSB of bit field MCTC
selects the READY operation:
MCTC.3 = ‘0’: Synchronous READY, ie. the READY signal must meet setup and hold times.
MCTC.3 = ‘1’: Asynchronous READY, ie. the READY signal is synchronized internally.
Data Sheet
192
2001-02-23
C165UTAH
External Bus Interface
Synchronous READY provides the fastest bus cycles, but requires setup and hold
times to be met. The CLKOUT signal should be enabled and may be used by the
peripheral logic to control the READY timing in this case.
Asynchronous READY is less restrictive, but requires additional waitstates caused by
the internal synchronization. As the asynchronous READY is sampled earlier (see figure
above) programmed waitstates may be necessary to provide proper bus cycles (see also
notes on “normally-ready” peripherals below).
A READY signal (especially asynchronous READY) that has been activated by an
external device may be deactivated in response to the trailing (rising) edge of the
respective command (RD or WR).
Note: When the READY function is enabled for a specific address window, each bus
cycle within this window must be terminated with an active READY signal.
Otherwise the controller hangs until the next reset. A timeout function is only
provided by the watchdog timer.
Combining the READY function with predefined waitstates is advantageous in two
cases:
Memory components with a fixed access time and peripherals operating with READY
may be grouped into the same address window. The (external) waitstate control logic in
this case would activate READY either upon the memory’s chip select or with the
peripheral’s READY output. After the predefined number of waitstates the C165UTAH
will check its READY line to determine the end of the bus cycle. For a memory access it
will be low already (see example a) in the figure above), for a peripheral access it may
be delayed (see example b) in the figure above). As memories tend to be faster than
peripherals, there should be no impact on system performance.
When using the READY function with so-called “normally-ready” peripherals, it may lead
to erroneous bus cycles, if the READY line is sampled too early. These peripherals pull
their READY output low, while they are idle. When they are accessed, they deactivate
READY until the bus cycle is complete, then drive it low again. If, however, the peripheral
deactivates READY after the first sample point of the C165UTAH, the controller samples
an active READY and terminates the current bus cycle, which, of course, is too early. By
inserting predefined waitstates the first READY sample point can be shifted to a time,
where the peripheral has safely controlled the READY line (eg. after 2 waitstates in the
figure above).
10.4
Controlling the External Bus Controller
A set of registers controls the functions of the EBC. General features like the usage of
interface pins (WR, BHE), segmentation are controlled via register SYSCON.
Note: For SYSCON register description, refer to page 66.
Data Sheet
193
2001-02-23
C165UTAH
External Bus Interface
The properties of a bus cycle like chip select mode, usage of READY, length of ALE,
external bus mode, read/write delay and waitstates are controlled via registers
BUSCON4...BUSCON0. Four of these registers (BUSCON4...BUSCON1) have an
address select register (ADDRSEL4...ADDRSEL1) associated with them, which allows
to specify up to four address areas and the individual bus characteristics within these
areas. All accesses that are not covered by these four areas are then controlled via
BUSCON0. This allows to use memory components or peripherals with different
interfaces within the same system, while optimizing accesses to each of them.
The layout of the five BUSCON registers is identical. Registers BUSCON4...BUSCON1,
which control the selected address windows, are completely under software control,
while register BUSCON0, which eg. is also used for the very first code access after reset,
is partly controlled by hardware, ie. it is initialized via PORT0 during the reset sequence.
This hardware control allows to define an appropriate external bus for systems, where
no internal program memory is provided.
Data Sheet
194
2001-02-23
C165UTAH
External Bus Interface
BUSCON0 (FF0CH / 86H)
15
14
CSW CSR
EN0 EN0
rw
rw
SFR
13
12
11
-
RDY
EN0
-
rw
-
rw
-
10
9
8
14
CSW CSR
EN1 EN1
rw
rw
13
12
rw
rw
14
CSW CSR
EN2 EN2
rw
rw
14
CSW CSR
EN3 EN3
rw
rw
11
-
RDY
EN1
-
rw
-
rw
-
10
9
14
CSW CSR
EN4 EN4
rw
rw
4
8
rw
rw
MTT RWD
C0
C0
rw
12
11
-
RDY
EN2
-
rw
-
rw
-
10
9
7
6
BTYP
rw
rw
8
rw
7
5
4
11
10
9
6
rw
8
-
RDY
EN3
-
rw
-
rw
-
rw
rw
rw
10
9
8
5
4
12
11
-
RDY
EN4
-
rw
-
rw
-
7
rw
6
BTYP
rw
1
0
MCTC
rw
3
2
1
0
MCTC
rw
rw
5
4
3
MTT RWD
C3
C3
rw
rw
5
4
2
1
0
MCTC
rw
Reset Value: 0000H
7
BUS ALE EW
ACT4 CTL4 EN4
rw
2
Reset Value: 0000H
BUS ALE EW
ACT3 CTL3 EN3
rw
3
MTT RWD
C2
C2
SFR
13
rw
rw
SFR
12
0
MCTC
MTT RWD
C1
C1
rw
BTYP
rw
13
1
Reset Value: 0000H
BUS ALE EW
ACT2 CTL2 EN2
rw
2
rw
SFR
13
3
Reset Value: 0000H
BUS ALE EW
ACT1 CTL1 EN1
BUSCON4 (FF1AH / 8DH)
15
rw
5
SFR
BUSCON3 (FF18H / 8CH)
15
6
BTYP
rw
BUSCON2 (FF16H / 8BH)
15
7
BUS ALE EW
ACT0 CTL0 EN0
BUSCON1 (FF14H / 8AH)
15
Reset Value: 0000H
6
BTYP
rw
rw
MTT RWD
C4
C4
rw
rw
3
2
1
0
MCTC
rw
Note: BUSCON0 is initialized with 0000H, if pin EA is high during reset. If pin EA is low
Bit
Function
MCTCx
Memory Cycle Time Control (Number of memory cycle time wait states)
’0000’ : 15 waitstates (Number = 15 - )
...
’1111’ : No waitstates
RWDCx
Read/Write Delay Control for BUSCONx
‘0’: With read/write delay: activate command 1 TCL after falling edge of ALE
‘1’: No read/write delay: activate command with falling edge of ALE
Data Sheet
195
2001-02-23
C165UTAH
External Bus Interface
Bit
Function
MTTCx
Memory Tristate Time Control
‘0’: 1 waitstate
‘1’: No waitstate
EWENx
Early Write Enable Bit
‘0’: Normal write
‘1’: Early write is enabled. The write signal turns off one TCL earlier.
Note: In order to have no overlapping with the following ALE signal, the write
control signal is shortened by one TCL by setting bit EWEN.
BTYPx
External Bus Configuration
0 0 : 8-bit Demultiplexed Bus
0 1 : 8-bit Multiplexed Bus
1 0 : 16-bit Demultiplexed Bus
1 1 : 16-bit Multiplexed Bus
Note: For BUSCON0 BTYP is defined via PORT0 during reset.
ALECTLx
ALE Lengthening Control
‘0’: Normal ALE signal
‘1’: Lengthened ALE signal
BUSACTx
Bus Active Control
‘0’: External bus disabled
‘1’: External bus enabled (within the respective address window, see ADDRSEL)
RDYENx
READY Input Enable
‘0’: External bus cycle is controlled by bit field MCTC only
‘1’: External bus cycle is controlled by the READY input signal
CSRENx
Read Chip Select Enable
‘0’: The CS signal is independent of the read command (RD)
‘1’: The CS signal is generated for the duration of the read command
CSWENx
Write Chip Select Enable
‘0’: The CS signal is independent of the write command (WR,WRL,WRH)
‘1’: The CS signal is generated for the duration of the write command
during reset, bits BUSACT0 and ALECTL0 are set (‘1’) and bit field BTYP is loaded
with the bus configuration selected via PORT0.
Bus Access Control
CPU accesses to internal and external busses, e.g. to internal or external memories or
peripherals, are controlled with the respective address ranges. These address ranges
are supported by ’chip select’ functions for XBUS resources or for external off-chip
resources. In the C165UTAH six address ranges with according bus definitions can be
programmed for XBUS peripherals (including memories) and additionally five ranges for
external bus peripherals.
Data Sheet
196
2001-02-23
C165UTAH
External Bus Interface
Note: In contrast to previous Infineon devices the XADRS/XBCON registers are not
hardwired but fully programmable.
Address ranges and address mapping of memories or peripherals on XBUS or external
bus are controlled with the address selection registers XADRSx for XBUS and
ADDRSELx for external bus. The respective bus type definitions are controlled with
registers XBCONx and BUSCONx.
In comparison to previous devices, C165UTAH has 3 more address selection registers
for XBUS:
• The new register pair XADRS4 / XBCON4 use the same standard scheme of address
selection and XCS control as the XADRS1-3 registers; smallest possible address
range is 256 bytes.
• The new register pairs XADRS5 / XBCON5 and XADRS6 / XBCON6 control address
selections as defined for external peripherals (as contolled by ADDRSEL); thus,
mapping of XPER addresses to the total address space is provided, with smallest
possible address range of 4 K Bytes. XBCON5/6 and XCS5/6 control are identical to
the standard XBUS address ranges.
After reset, no address selection register is selected; thus the default address range is
enabled and controlled with BUSCON0 and additionally the chip select output CS0 is
activated (as in standard C16x architecture).
Data Sheet
197
2001-02-23
C165UTAH
External Bus Interface
ADDRSEL1 (FE18H / 0CH)
15
14
13
12
11
SFR
10
9
8
14
13
12
11
14
13
12
11
14
13
12
11
5
4
3
2
1
rw
rw
SFR
10
9
8
7
6
5
4
3
2
1
RGSZ
rw
rw
SFR
10
9
8
7
6
5
4
3
2
1
RGSZ
rw
rw
SFR
9
8
7
6
5
4
3
2
1
RGSZ
rw
rw
Function
RGSZ
Range Size Selection
RGSAD
Range Start Address
0
Reset Value: 0000H
RGSAD
Bit
0
Reset Value: 0000H
RGSAD
10
0
Reset Value: 0000H
RGSAD
ADDRSEL4 (FE1EH / 0FH)
15
6
RGSZ
ADDRSEL3(FE1CH / 0EH)
15
7
RGSAD
ADDRSEL2 (FE1AH / 0DH)
15
Reset Value: 0000H
0
Note: There is no register ADDRSEL0, as register BUSCON0 controls all external
accesses outside the four address windows of BUSCON4...BUSCON1 within the
complete address space.
Definition of Address Areas
The four register pairs BUSCON4/ADDRSEL4...BUSCON1/ADDRSEL1 allow to define
4 separate address areas within the address space of the C165UTAH. Within each of
these address areas external accesses can be controlled by one of the four different bus
modes, independent of each other and of the bus mode specified in register BUSCON0.
Each ADDRSELx register in a way cuts out an address window, within which the
Data Sheet
198
2001-02-23
C165UTAH
External Bus Interface
parameters in register BUSCONx are used to control external accesses. The range start
address of such a window defines the upper address bits, which are not used within the
address window of the specified size (see table below). For a given window size only
those upper address bits of the start address are used (marked “R”), which are not
implicitly used for addresses inside the window. The lower bits of the start address
(marked “x”) are disregarded.
Bit field
RGSZ
Resulting Window
Size
Relevant Bits (R) of Start Address
(A22...A12)
0000
0001
0010
0011
0100
0101
0110
0111
1000
1001
1010
1011
11xx
4
KByte
8
KByte
16
KByte
32
KByte
64
KByte
128 KByte
256 KByte
512 KByte
1
MByte
2
MByte
4
MByte
Reserved
Reserved.
R
R
R
R
R
R
R
R
R
R
R
XADRS1(2/3/4/5/6) (F014H / 0AH)
15
14
13
12
11
R
R
R
R
R
R
R
R
R
R
x
R
R
R
R
R
R
R
R
R
x
x
Bit
R
R
R
R
R
R
R
x
x
x
x
R
R
R
R
R
R
x
x
x
x
x
R
R
R
R
R
x
x
x
x
x
x
ESFRx
10
9
8
-
-
R
R
R
R
x
x
x
x
x
x
x
R
R
R
x
x
x
x
x
x
x
x
R
R
x
x
x
x
x
x
x
x
x
R
x
x
x
x
x
x
x
x
x
x
Reset Value: 0000H
7
RGSAD
-
R
R
R
R
R
R
R
R
x
x
x
6
5
4
3
2
1
0
RGSZ
-
-
rw
Function
RGSAD
Address Range Start Address Selection
RGSZ
Address Range Size Selection
The respective SFR addresses of XADRS registers can be found in list of SFRs.
Due to the different range size options, address mapping of XPERs is possible only
within the first MByte of the total address range if XADRS1 to XADRS4 is used. The
upper four address lines (A23:A20) are set to zero. Note that the range start address can
be only on boundaries specified by the selected range size.
Data Sheet
199
2001-02-23
C165UTAH
External Bus Interface
The following tables show the different definitions of range size selections and range
start addresses for the two types of address selections:
Range
Size
RGSZ
Selected
Address
Range
Relvant(R) bits of
RGSAD
Selected Range Start Address
(Relevant(R) bits of RGSAD)
0000
0001
0010
0011
0100
0101
0110
0111
1000
1001
1010
1011
11xx
256 Byte
512 Bytes
1 KB
2 KB
4 KB
8 KB
16 KB
32 KB
64 KBy
128 KB
256 KB
512 KB
- reserved
RRRR RRRR RRRR
RRRR RRRR RRR0
RRRR RRRR RR00
RRRR RRRR R000
RRRR RRRR 0000
RRRR RRR0 0000
RRRR RR00 0000
RRRR R000 0000
RRRR 0000 0000
RRR0 0000 0000
RR00 0000 0000
R000 0000 0000
0000 RRRR RRRR RRRR 0000 0000
0000 RRRR RRRR RRR0 0000 0000
0000 RRRR RRRR RR00 0000 0000
0000 RRRR RRRR R000 0000 0000
0000 RRRR RRRR 0000 0000 0000
0000 RRRR RRR0 0000 0000 0000
0000 RRRR RR00 0000 0000 0000
0000 RRRR R000 0000 0000 0000
0000 RRRR 0000 0000 0000 0000
0000 RRR0 0000 0000 0000 0000
0000 RR00 0000 0000 0000 0000
0000 R000 0000 0000 0000 0000
Table 34
Data Sheet
Address Range and Address Range Start Definition of XADRS1/2/3/
4 register
200
2001-02-23
C165UTAH
External Bus Interface
Range
Size
RGSZ
Selected
Address
Range
Relvant(R) bits of
RGSAD
Selected Range Start Address
(Relevant(R) bits of RGSAD)
0000
0001
0010
0011
0100
0101
0110
0111
1000
1001
1010
1011
11xx
4 KB
8 KB
16 KB
32 KB
64 KB
128 KB
256 KB
512 KB
1 MB
2 MB
4 MB
8 MB
- reserved
RRRR RRRR RRRR
RRRR RRRR RRR0
RRRR RRRR RR00
RRRR RRRR R000
RRRR RRRR 0000
RRRR RRR0 0000
RRRR RR00 0000
RRRR R000 0000
RRRR 0000 0000
RRR0 0000 0000
RR00 0000 0000
R000 0000 0000
RRRR RRRR RRRR 0000 0000 0000
RRRR RRRR RRR0 0000 0000 0000
RRRR RRRR RR00 0000 0000 0000
RRRR RRRR R000 0000 0000 0000
RRRR RRRR 0000 0000 0000 0000
RRRR RRR0 0000 0000 0000 0000
RRRR RR00 0000 0000 0000 0000
RRRR R000 0000 0000 0000 0000
RRRR 0000 0000 0000 0000 0000
RRR0 0000 0000 0000 0000 0000
RR00 0000 0000 0000 0000 0000
R000 0000 0000 0000 0000 0000
Table 35
Address Range and Address Range Start Definition of XADRS5/6
Register
The XBCONx registers are defined as follows:
Data Sheet
201
2001-02-23
C165UTAH
External Bus Interface
XBCON1/2/3 (F114H / 8AH)
15
14
13
-
-
-
-
-
-
Bit
12
11
ESFR-b
10
9
8
RDY BS BUS ALE EW
ENx WCx ACTx CTLx ENx
rw
rw
rw
rw
Reset Value: 0000H
7
5
4
BTYPx
MT
TCx
RW
DCx
MCTCx
rw
rw
rw
rw
rw
6
3
2
1
0
Function
MCTCx
Memory Cycle Time Control (see BUSCON)
RWDCx
READ/WRITE Delay Control (see BUSCON)
MTTCx
Memory Tri-state Time Control (see BUSCON)
BTYPx
Bus Type Selection; only demultiplexed busses are supported on XBUS;
’00’: 8 bit bus
’10’: 16 bit bus;
’x1’: reserved.
EWENx
Early Write Enable
’0’: Standard write enable signal control
’1’: Write active state is disabled one TCL earlier
ALECTLx
ALE Lengthening Control Bit (see BUSCON)
BUSACTx*
Bus Active Control
‘0’: XBUS (peripheral) disabled
‘1’: XBUS (peripheral) enabled
Enables the XBUS and the according chip select XCSx for the respective
address window (respective XBUS peripheral), selected with according XADRSx
window; after reset, all address windows on XBUS are disabled.
*not used in FC-Cores, where XBCON is hardwired.
BSWCx
BUSCON Switch Control
’0’: Standard switch of bustype (switch of XBCON)
’1’: A bus wait state (Tri-state cycle) is included after execution of last oldbustype cycle and before the first new-bustype cycle after switch of XBCON or
BUSCON; the BSWC bit is indicated in the old-bustype XBCON/BUSCON.
RDYENx
READY Enable
’0’: The bus cycle length is controlled by the bus controller using MCTC
’1’: The bus cycle length is controlled by the peripheral using READY
Note: The ’BUSCON switch control’ BSWC is a new function, which is necessary due to
the execution with higher frequencies, to avoid bus collisions on data bus in case
of peripheral change (see BUSCON).
Note: All XADRSx/ADDRSELx registers as well as XBCONx/BUSCONx registers are
user programmable SFR registers. All BUSCONx registers are mapped into the
bitaddressable SFR memory space, all XBCONx registers are located in the
bitaddressable ESFR memory space. Although they are free programmable,
Data Sheet
202
2001-02-23
C165UTAH
External Bus Interface
programming should be performed during the initialisation phase before the first
accesses are controlled with XBCONx or BUSCONx.
Note: The respective SFR addresses of XBCON registers can be found in list of SFRs.
Note: Within the C165UTAH, register XBCON1 is related to the IOM-2 module, register
XBCON2 is related to the USB module and register XBCON2 is related to the
EPEC module. For configuration, please also refer to Chapter 10.8, "Initialization
of the C165UTAH’s X-peripherals" on page 210.
Address Window Arbitration
For each access the EBC compares the current address with all address select registers
(programmable ADDRSELx and hardwired XADRSx). This comparison is done in four
levels.
Priority 1:
The XADRSx registers are evaluated first. A match with one of these registers directs the
access to the respective X-Peripheral using the corresponding XBCONx register and ignoring
all other ADDRSELx registers.
Priority 2: Registers ADDRSEL2 and ADDRSEL4 are evaluated before ADDRSEL1 and
ADDRSEL3, respectively. A match with one of these registers directs the access to
the respective external area using the corresponding BUSCONx register and ignoring
registers ADDRSEL1/3 (see figure below).
Priority 3:
A match with registers ADDRSEL1 or ADDRSEL3 directs the access to the respective
external area using the corresponding BUSCONx register.
Priority 4:
If there is no match with any XADRSx or ADDRSELx register the access to the external bus
uses register BUSCON0.
XBCON0
BUSCON2
BUSCON4
BUSCON1
BUSCON3
Active
Window
Inactive
Window
BUSCON0
Figure 54
Address Window Arbitration
Note: Only the indicated overlaps are defined. All other overlaps lead to erroneous bus
cycles. Eg. ADDRSEL4 may not overlap ADDRSEL2 or ADDRSEL1. The
hardwired XADRSx registers are defined non-overlapping.
Data Sheet
203
2001-02-23
C165UTAH
External Bus Interface
RP0H (F108H / 84H)
15
-
14
-
13
-
SFR
12
11
-
-
10
-
9
-
8
Reset Value: - - XXH
7
-
6
5
4
3
2
1
0
CLKCFG
SALSEL
CSSEL
WRC
r
r
r
r
Bit
Function
WRC
Write Configuration
0: Pins WR and BHE operate as WRL and WRH signals
1: Pins WR and BHE operate as WR and BHE signals
CSSEL
Chip Select Line Selection (Number of active CS outputs)
0 0: 3 CS lines: CS2...CS0
0 1: 2 CS lines: CS1...CS0
1 0: No CS lines at all
1 1: 5 CS lines: CS4...CS0 (Default without pulldowns)
SALSEL
Segment Address Line Selection (Number of active segment address outputs)
0 0: 4-bit segment address: A19...A16
0 1: No segment address lines at all
1 0: 7-bit segment address: A22...A16
1 1: 2-bit segment address: A17...A16 (Default without pulldowns)
CLKCFG
Clock Generation Mode Configuration
These pins define the clock generation mode, ie. the mechanism how the the
internal CPU clock is generated from the externally applied (XTAL1) input clock.
Note: RP0H cannot be changed via software, but rather allows to check the current
configuration.
Precautions and Hints
• The external bus interface is enabled as long as at least one of the BUSCON registers
has its BUSACT bit set.
• PORT1 will output the intra-segment address as long as at least one of the BUSCON
registers selects a demultiplexed external bus, even for multiplexed bus cycles.
• Not all address areas defined via registers ADDRSELx may overlap each other. The
operation of the EBC will be unpredictable in such a case. See chapter „Address Window
Arbitration“.
• The address areas defined via registers ADDRSELx may overlap internal address
areas. Internal accesses will be executed in this case.
• For any access to an internal address area the EBC will remain inactive (see EBC Idle
State).
Data Sheet
204
2001-02-23
C165UTAH
External Bus Interface
10.5
EBC Idle State
When the external bus interface is enabled, but no external access is currently executed,
the EBC is idle. As long as only internal resources (from an architecture point of view)
like IRAM, GPRs or SFRs, etc. are used the external bus interface does not change (see
table below).
Accesses to on-chip X-Peripherals are also controlled by the EBC. However, even
though an X-Peripheral appears like an external peripheral to the controller, the
respective accesses do not generate valid external bus cycles.
Due to timing constraints address and write data of an XBUS cycle are reflected on the
external bus interface (see table below). The „address“ mentioned above includes
PORT1, Port 4, BHE and ALE which also pulses for an XBUS cycle. The external CS
signals on Port 6 are driven inactive (high) because the EBC switches to an internal XCS
signal.
The external control signals (RD and WR or WRL/WRH if enabled) remain inactive
(high).
Table 36
Status of the external bus interface during EBC idle state:
Pins
Internal accesses only
XBUS accesses
PORT0
Tristated (floating)
Tristated (floating) for read accesses
XBUS write data for write accesses
PORT1
Last used external address
(if used for the bus interface)
Last used XBUS address
(if used for the bus interface)
Port 4
Last used external segment address
(on selected pins)
Last used XBUS segment address
(on selected pins)
Port 6
Active external CS signal corresponding
to last used address
Inactive (high) for selected CS signals
BHE
Level corresponding to last external
access
Level corresponding to last XBUS
access
ALE
Inactive (low)
Pulses as defined for X-Peripheral
RD
Inactive (high)
Inactive (high)
WR/WRL
Inactive (high)
Inactive (high)
WRH
Inactive (high)
Inactive (high)
Data Sheet
205
2001-02-23
C165UTAH
External Bus Interface
10.6
External Bus Arbitration
In high performance systems it may be efficient to share external resources like memory
banks or peripheral devices among more than one controller. The C165UTAH supports
this approach with the possibility to arbitrate the access to its external bus, ie. to the
external devices.
This bus arbitration allows an external master to request the C165UTAH’s bus via the
HOLD input. The C165UTAH acknowledges this request via the HLDA output and will
float its bus lines in this case. The CS outputs provide internal pullup devices. The new
master may now access the peripheral devices or memory banks via the same interface
lines as the C165UTAH. During this time the C165UTAH can keep on executing, as long
as it does not need access to the external bus. All actions that just require internal
resources like instruction or data memory and on-chip peripherals, may be executed in
parallel.
When the C165UTAH needs access to its external bus while it is occupied by another
bus master, it demands it via the BREQ output.
The external bus arbitration is enabled by setting bit HLDEN in register PSW to ‘1’. In
this case the three bus arbitration pins HOLD, HLDA and BREQ are automatically
controlled by the EBC independent of their I/O configuration. Bit HLDEN may be cleared
during the execution of program sequences, where the external resources are required
but cannot be shared with other bus masters. In this case the C165UTAH will not answer
to HOLD requests from other external masters. If HLDEN is cleared while the
C165UTAH is in Hold State (code execution from internal RAM) this Hold State is left
only after HOLD has been deactivated again. Ie. in this case the current Hold State
continues and only the next HOLD request is not answered.
Connecting eg. two C165UTAHs in this way would require additional logic to combine
the respective output signals HLDA and BREQ. This can be avoided by switching one of
the controllers into Slave Mode where pin HLDA is switched to input. This allows to
directly connect the slave controller to another master controller without glue logic. The
Slave Mode is selected by setting bit DP6.7 to ’1’. DP6.7=’0’ (default after reset) selects
the Master Mode.
Note: The pins HOLD, HLDA and BREQ keep their alternate function (bus arbitration)
even after the arbitration mechanism has been switched off by clearing HLDEN.
All three pins are used for bus arbitration after bit HLDEN was set once.
Connecting Bus Masters
When multiple C165UTAHs or a C165UTAH and another bus master shall share
external resources some glue logic is required that defines the currently active bus
master and also enables a C165UTAH which has surrendered its bus interface to regain
control of it in case it must access the shared external resources. This glue logic is
Data Sheet
206
2001-02-23
C165UTAH
External Bus Interface
required if the „other“ bus master does not automatically remove its hold request after
having used the shared resources.
When two C165UTAHs are to be connected in this way the external glue logic can be
left out. In this case one of the controllers must be operated in its Master Mode (default
after reset, DP6.7=’0’) while the other one must be operated in its Slave Mode (selected
with DP6.7=’1’).
Figure 55
HOLD
HOLD
HLDA
HLDA
BREQ
BREQ
C165UTAH in
Slave Mode
C165UTAH in
Master Mode
In Slave Mode the C165UTAH inverts the direction of its HLDA pin and uses it as an
input, while the master’s HLDA pin remains an output. This approach does not require
any additional glue logic for the bus arbitration (see figure below).
Sharing External Resources using Slave Mode
When the bus arbitration is enabled (HLDEN=’1’) the three corresponding pins are
automatically controlled by the EBC. Normally the respective port direction register bits
retain their reset value which is ’0’. This selects Master Mode where the device operates
compatible with earlier versions. Slave Mode is enabled by intentionally switching pin
BREQ to output (DP6.7=’1’) which is neither required for Master Mode nor for earlier
devices.
Entering the Hold State
Access to the C165UTAH’s external bus is requested by driving its HOLD input low. After
synchronizing this signal the C165UTAH will complete a current external bus cycle (if
any is active), release the external bus and grant access to it by driving the HLDA output
low. During hold state the C165UTAH treats the external bus interface as follows:
• Address and data bus(es) float to tri-state
• ALE is pulled low by an internal pulldown device
Data Sheet
207
2001-02-23
C165UTAH
External Bus Interface
• Command lines are pulled high by internal pullup devices (RD, WR/WRL, BHE/WRH)
• CSx outputs are pulled high (push/pull mode) or float to tri-state (open drain mode)
Should the C165UTAH require access to its external bus during hold mode, it activates
its bus request output BREQ to notify the arbitration circuitry. BREQ is activated only
during hold mode. It will be inactive during normal operation.
~
~
HOLD
~
~
HLDA
~
~
BREQ
CSx
~
~
~ ~
Other
Signals
~ ~
~
~
MCD02238
Figure 56
External Bus Arbitration, Releasing the Bus
Note: The C165UTAH will complete the currently running bus cycle before granting bus
access as indicated by the broken lines. This may delay hold acknowledge
compared to this figure.
The figure above shows the first possibility for BREQ to get active.
During bus hold pin P3.12 is switched back to its standard function and is then
controlled by DP3.12 and P3.12. Keep DP3.12 = ’0’ in this case to ensure floating
in hold mode.
Exiting the Hold State
The external bus master returns the access rights to the C165UTAH by driving the HOLD
input high. After synchronizing this signal the C165UTAH will drive the HLDA output
high, actively drive the control signals and resume executing external bus cycles if
required.
Data Sheet
208
2001-02-23
C165UTAH
External Bus Interface
Depending on the arbitration logic, the external bus can be returned to the C165UTAH
under two circumstances:
• The external master does no more require access to the shared resources and gives
up its own access rights, or
• The C165UTAH needs access to the shared resources and demands this by activating
its BREQ output. The arbitration logic may then deactivate the other master’s HLDA and
so free the external bus for the C165UTAH, depending on the priority of the different
masters.
Note: The Hold State is not terminated by clearing bit HLDEN.
HOLD
HLDA
BREQ
CSx
Other
Signals
MCD02236
Figure 57
External Bus Arbitration, (Regaining the Bus)
Note: The falling BREQ edge shows the last chance for BREQ to trigger the indicated
regain-sequence. Even if BREQ is activated earlier the regain-sequence is
initiated by HOLD going high. BREQ and HOLD are connected via an external
arbitration circuitry. Please note that HOLD may also be deactivated without the
C165UTAH requesting the bus.
10.7
XBUS Interface
The C165UTAH provides an on-chip interface (the XBUS interface), which allows to
connect integrated customer/application specific peripherals to the standard controller
Data Sheet
209
2001-02-23
C165UTAH
External Bus Interface
core. The XBUS is an internal representation of the external bus interface, ie. it is
operated in the same way.
The current XBUS interface is prepared to support up to 3 X-Peripherals.
For each peripheral on the XBUS (X-Peripheral) there is a separate address window
controlled by an XBCON and an XADRS register. As an interface to a peripheral in many
cases is represented by just a few registers, the XADRS registers select smaller address
windows than the standard ADDRSEL registers. As the register pairs control integrated
peripherals rather than externally connected ones, they are fixed by mask programming
rather than being user programmable.
X-Peripheral accesses provide the same choices as external accesses, so these
peripherals may be bytewide or wordwide, with or without a separate address bus.
Interrupt nodes and configuration pins (on PORT0) are provided for X-Peripherals to be
integrated.
10.8
Initialization of the C165UTAH’s X-peripherals
The following registers must be set for initialization of the C165UTAH X-peripherals:
XPERCON-Register (Addr. F024, default: 0000):
Bit 5:
'1': IOM-2 active
'0': IOM-2 switched-off
Bit 6:
'1': USB module active '0': USB module switched-off
Bit 7:
'1': EPEC active
'0': EPEC switched-off
SYSCON-Register (Addr. FF12, default: 0xx0):
Bit 2:
'1': X-Peripherals enable
'0': X-Peripherals disable
Bit 1:
'1': X-Per accesses visible at externalXBUS'0': X-Per accesses not visible
SYSCON3-Register (Addr. F1D4, default: 0000):
Bit 15:
'1': All peripheral clocks disabled
'0': Individual disable control by
bits 14 thru 0
Bit 12:11: '00': USB transceiver in normal operation
’01’: Suspend mode, differential USB receiver switched off
’10’: Reserved, do not use this combination
’11’: USB transceiver switched off - full power down mode
Bit 8:
'1': Disable EPEC clock
'0': Enable EPEC clock
Bit 7:
'1': Disable USB clock
'0': Enable USB clock
Bit 6:
'1': Disable IOM-2 clock
'0': Enable IOM-2 clock
Bit 3:
'1': Disable GPT12 clock
'0': Enable GPT12 clock
Bit 2:
'1': Disable SSC clock
'0': Enable SSC clock
Bit 1:
'1': Disable ASC clock
'0': Enable ASC clock
Bit 0:
'1': Disable RTC clock
'0': Enable RTC clock
Data Sheet
210
2001-02-23
C165UTAH
External Bus Interface
XBCON1-Register (Addr.F114, default: 0000H):
Definition of the IOM-2 bus protocol.
Must be set to ’048xH’. Recommended using 0 waitstates: ’048FH’.
XBCON2-Register (Addr.F116, default: 0000H):
Definition of the USB bus protocol.
Must be set to ’048xH’. Recommended using 0 waitstates: ’048FH’.
XBCON3-Register (Addr.F118, default: 0000H):
Definition of the EPEC bus protocol.
Must be set to ’048xH’. Recommended using 0 waitstates: ’048FH’.
XADRS1-Register (Addr.F014, default: UUUU):
Definition of the IOM-2 address space.
Should be initialized with ’0EF0H’ before writing to XBCON1.
XADRS2-Register (Addr.F016, default: UUUU):
Definition of the USB address space.
Should be initialized with ’0EE0H’ before writing to XBCON2.
XADRS3-Register (Addr.F018, default: UUUU):
Definition of the EPEC address space.
Should be initialized with ’0ED0H’ before writing to XBCON3.
Note: Bits (12:11) of register SYSCON3 are related to the analog USB transceiver only
and not to the digital USB module.
Data Sheet
211
2001-02-23
C165UTAH
General Purpose Timer Unit
11
General Purpose Timer Unit
The General Purpose Timer Unit (GPT) represents very flexible multifunctional timer
structures which may be used for timing, event counting, pulse width measurement,
pulse generation, frequency multiplication, and other purposes.
In the C165UTAH, there are following alternate function pins available: P3.3/T3OUT,
P3.5/T4IN/T3EUD/T2EUD, P3.6/T3IN and P3.7/T2IN, as shown in Figure 58.
P3.7
T2IN
T2EUD
P3.6
T3IN
T3EUD
P3.5
T4IN
GPT1 Core
Timer 2
GPT1 Core
Timer 3
interrupt request
T2IC.T2IR
P3.3
T3OUT
T3OTL
interrupt request
T3IC.T3IR
GPT1 Core
Timer 4
interrupt request
T4IC.T4IR
GPT2 Core
Timer 5
interrupt request
T5IC.T5IR
GPT2 CAPREL
interrupt request
CRIC.CRIR
GPT2 Core
Timer 6
interrupt request
T6IC.T6IR
T6OTL
gpt12utahimpl
Note: In the C165UTAH, there are no external pins connected to Timer 5 and
Timer 6. Additional, there is no second external connection to Timer 4
(T4EUD) and no external connection to the register GPT2 CAPREL.
Figure 58
GPT module external pins
The GPT incorporate five 16-bit timers that are grouped into the two timer blocks GPT1
and GPT2. Each timer in each block may operate independently in a number of different
Data Sheet
212
2001-02-23
C165UTAH
General Purpose Timer Unit
modes such as gated timer or counter mode, or may be concatenated with another timer
of the same block.
Block 1 contains 3 timers/counters with a maximum resolution of fTimer/4. The auxiliary
timers of GPT1 may optionally be configured as reload or capture registers for the core
timer.
Block 2 contains 2 internal timers/counters with a maximum resolution of fTimer/2. An
additional CAPREL register supports capture and reload operation with extended
functionality.
The following enumeration summarizes all features to be supported:
• Timer Block 1:
– fTimer/4 maximum resolution.
– 3 independent timers/counters.
– Timers/counters can be concatenated.
– 4 operating modes (timer, gated timer, counter, incremental).
– Separate interrupt nodes.
• Timer Block 2:
– fTimer/2 maximum resolution.
– 2 independent timers/counters.
– Timers/counters can be concatenated.
– 2 operating modes (timer, counter).
– Capture/reload functions via 16-bit Capture/Reload register CAPREL.
– Separate interrupt nodes.
11.1
Kernel Description
11.1.1
Functional Description of Timer Block 1
All three timers of block 1 (T2, T3, T4) can run in 4 basic modes, which are timer, gated
timer, counter and incremental interface mode, and all timers can either count up or
down.
The input line (TxIN) associated with it which serves as the gate control in gated timer
mode, or as the count input in counter mode. The count direction (Up / Down) may be
programmed via software or may be dynamically altered by a signal at an external
control input line. An overflow/underflow of core timer T3 is indicated by the output toggle
latch T3OTL whose state may be output on related line T3OUT.
The auxiliary timers T2 and T4 may additionally be concatenated with the core timer, or
used as capture or reload registers for the core timer.
Data Sheet
213
2001-02-23
C165UTAH
General Purpose Timer Unit
The current contents of each timer can be read or modified by the CPU by accessing the
corresponding timer registers T2, T3, or T4, which are located in the non-bitaddressable
SFR space. When any of the timer registers is written to by the CPU in the state
immediately before a timer increment, decrement, reload, or capture is to be performed,
the CPU write operation has priority in order to guarantee correct results.
T 2E U D
f h w _c lk
U /D
2n : 1
T 2 IN
T2
M o de
C o ntrol
Inte rrup t
R e qu es t
G P T 1 T im er T 2
R e lo ad
C a pture
T 3O T L
f h w _c lk
In te rru pt
R eq ues t
2n : 1
T3
M o de
C o ntrol
T 3 IN
T o gg le FF
G P T 1 T im er T 3
T3OTL
T 3O U T
U /D
T 3E U D
T 3O T L
C a pture
R e lo ad
T 4 IN
f h w _c lk
2n : 1
T4
M o de
C o ntrol
G P T 1 T im er T 4
Inte rrup t
R e qu es t
U /D
M C T 0 21 41
Figure 59
Structure of Timer Block 1
11.1.1.1 Core Timer T3
The operation of the core timer T3 is controlled by its bitaddressable control register
T3CON.
Run Control
The timer can be started or stopped by software through bit T3R (Timer T3 Run Bit).
Setting bit T3R to ‘1’ will start the timer, clearing T3R stops the timer.
Data Sheet
214
2001-02-23
C165UTAH
General Purpose Timer Unit
In gated timer mode, the timer will only run if T3R = ‘1’ and the gate is active (high or low,
as programmed).
Note: When bit T2RC/T4RC in timer control register T2CON/T4CON is set to ’1’, T3R
will also control (start and stop) auxiliary timer T2/T4.
Count Direction Control
The count direction of the core timer can be controlled either by software or by the
external input line T3EUD (Timer T3 External Up/Down Control Input). These options are
selected by bits T3UD and T3UDE in control register T3CON. When the up/down control
is done by software (bit T3UDE = ‘0’), the count direction can be altered by setting or
clearing bit T3UD. When T3UDE = ‘1’, line T3EUD is selected to be the controlling
source of the count direction. However, bit T3UD can still be used to reverse the actual
count direction, as shown in the table below. If T3UD = ‘0’ and line T3EUD shows a low
level, the timer is counting up. With a high level at T3EUD the timer is counting down. If
T3UD = ‘1’, a high level at line T3EUD specifies counting up, and a low level specifies
counting down. The count direction can be changed regardless of whether the timer is
running or not.
When line T3EUD is used as external count direction control input, its associated port
pin must be configured as input.
Table 37
GPT1 Core Timer T3 Count Direction Control
Line TxEUD
Bit TxUDE
Bit TxUD
Count Direction
X
0
0
Count Up
X
0
1
Count Down
0
1
0
Count Up
1
1
0
Count Down
0
1
1
Count Down
1
1
1
Count Up
Note: The direction control works the same for core timer T3 and for auxiliary timer T2.
For timer T4, no T4EUD input exist, therefore the count direction can be controlled
by software only.
Timer 3 Overflow/Underflow Monitoring
An overflow or underflow of timer T3 will clock the overflow toggle latch T3OTL in control
register T3CON. T3OTL can also be set or reset by software. Bit T3OE (Overflow/
Underflow Output Enable) in register T3CON enables the state of T3OTL to be
Data Sheet
215
2001-02-23
C165UTAH
General Purpose Timer Unit
monitored via an external line T3OUT. If this line is linked to an external port pin, which
has to be configured as output, T3OTL can be used to control external HW.
In addition, T3OTL can be used in conjunction with the timer over/underflows as an input
for the counter function or as a trigger source for the reload function of the auxiliary
timers T2 and T4. For this purpose, the state of T3OTL does not have to be available at
any port pin, because an internal connection is provided for this option.
Timer 3 in Timer Mode
Timer mode for the core timer T3 is selected by setting bit field T3M in register T3CON
to ‘000B’.
In this mode, T3 is clocked with the module clock fTimer divided by a programmable
prescaler, which is controlled by bit field T3I and bit FM1. The input frequency fT3 for timer
T3 and its resolution rT3 are scaled linearly with lower module clock frequencies, as can
be seen from the following formula:
fT3 =
Table 38
fTimer
rT3 [µs] =
8 * 2
8 * 2
fTimer [MHz]
Example for Timer 3 Frequencies and Resolutions
fTimer [MHz]
T3I
FM1
fT3 [KHz]
rT3 [µs]
24
’111’
0
23.44
42.67
24
’000’
1
6000.0
0.17
36
’000’
0
4500.0
0.22
36
’100’
0
281.25
3.55
36
’111’
1
70.31
14.22
This formula also applies to the Gated Timer Mode of T3 and to the auxiliary timers T2
and T4 in timer and gated timer mode, where applicable.
Data Sheet
216
2001-02-23
C165UTAH
General Purpose Timer Unit
Txl
fhw_clk
2n : 1
Core Timer Tx
TxR
Interrupt
Request
Up/
Down
TxOTL
TxUD
x=3
TxUDE
Figure 60
MCB02028
Block Diagram of Core Timer T3 in Timer Mode
Timer 3 in Gated Timer Mode
Gated timer mode for the core timer T3 is selected by setting bit field T3M in register
T3CON to ‘010B’ or ‘011B’.
Bit T3M.0 (T3CON.3) selects the active level of the gate input. In gated timer mode the
same options for the input frequency as for the timer mode are available. However, the
input clock to the timer in this mode is gated by the external input line T3IN (Timer T3
External Input), which is an alternate function of P3.6.
To enable this operation pin P3.6/T3IN must be configured as input, ie. direction control
bit DP3.6 must contain ’0’.
Data Sheet
217
2001-02-23
C165UTAH
General Purpose Timer Unit
T xI
f h w _ c lk
T xIN
2n : 1
MUX
C ore T im er T x
T xM
T xU D
T xR
Up/
Down
T xO T L
T xO U T
T xO E
0
Interrup t
R e qu es t
MUX
T xE U D
XO R
1
x=3
T xU D E
Figure 61
M C B 02 02 9
Block Diagram of Core Timer T3 in Gated Timer Mode
If T3M = ‘010B’, the timer is enabled when T3IN shows a low level. A high level at this
line stops the timer. If T3M = ‘011B’, line T3IN must have a high level in order to enable
the timer. In addition, the timer can be turned on or off by software using bit T3R. The
timer will only run, if T3R = ‘1’ and the gate is active. It will stop, if either T3R = ‘0’ or the
gate is inactive.
Note: A transition of the gate signal at line T3IN does not cause an interrupt request.
Timer 3 in Counter Mode
Counter mode for the core timer T3 is selected by setting bit field T3M in register T3CON
to ‘001B’. In counter mode timer T3 is clocked by a transition at the external input pin
T3IN, which is an alternate function of P3.6. The event causing an increment or
decrement of the timer can be a positive, a negative, or both a positive and a negative
transition at this line. Bit field T3I in control register T3CON selects the triggering
transition (see Table 39 below).
Data Sheet
218
2001-02-23
C165UTAH
General Purpose Timer Unit
Edge
Select
TxIN
Core Timer Tx
TxR
TxUD
TxOTL
Up/
Down
TxOUT
TxOE
Txl
0
Interrupt
Request
MUX
TxEUD
XOR
1
x=3
TxUDE
Figure 62
Block Diagram of Core Timer T3 in Counter Mode
Table 39
Core Timer T3 (Counter Mode) Input Edge Selection
T3I
MCB02030
Triggering Edge for Counter Increment / Decrement
000
None. Counter T3 is disabled
001
Positive transition (rising edge) on T3IN
010
Negative transition (falling edge) on T3IN
011
Any transition (rising or falling edge) on T3IN
1XX
Reserved. Do not use this combination
For counter operation, a port pin P3.6/T3IN must be configured as input. The maximum
input frequency which is allowed in counter mode is fTimer/8 (FM1 = ’1’). To ensure that a
transition of the count input signal which is applied to T3IN is correctly recognized, its
level should be held high or low for at least 4 fTimer cycles (FM1 = ’1’) before it changes.
Timer 3 in Incremental Interface Mode
Incremental Interface mode for the core timer T3 is selected by setting bit field T3M in
register T3CON to ‘110B’ or ‘111B’. In incremental interface mode pin P3.6/T3IN
(configured as timer input T3IN) and pin P3.5/T3EUD (configured as timer input) are
used to interface to an incremental encoder.
Data Sheet
219
2001-02-23
C165UTAH
General Purpose Timer Unit
Note: In the C165UTAH, the T3EUD timer input is connected to P3.5. In this case,
Timer 4 input T4IN can be used by Software only.
T3 is clocked by each transition on one or both of the external input lines which gives 2fold or 4-fold resolution of the encoder input.
T 3IN
E d ge
S elec t
T im er T 3
T3l
T 3R
T 3O T L
U p/
D ow n
T3
E dg e
T 3O E
T3
R D IR
T 3U D
T 3E U D
XO R
0
MUX
1
T3UDE
Figure 63
Interrup t
R eq ue st
E dg e
Interrup t
T 3M
C h ang e
D e tection
P h as e
D etec t
T 3O U T
T3
C H D IR
R ota tio n
Interrup t
T 3M
M C B 03 998
Block Diagram of Core Timer T3 in Incremental Interface Mode
Bit field T3I in control register T3CON selects the triggering transitions (see table below).
In this mode the sequence of the transitions of the two input signals is evaluated and
generates count pulses as well as the direction signal. Depending on the chosen
Incremental Intrerface Mode, Rotation detection ‘110B’ or Edge Detection ‘111B’, an
interrupt can be generated. This interrupt is only generated if it’s enabled by setting bit
T3IREN in register T3CON. For the Rotation detection an interrupt will be generated
each time the count direction of timer 3 changes. For the Edge detection an interrupt will
be generated each time a count action for timer 3 occurs. Count direction, changes in
the count direction and count requests are monitored through the status bits T3RDIR,
T3CHDIR and T3EDGE in register T3CON. T3 is modified automatically according to the
speed and the direction of the incremental encoder. Therefore, the contents of timer T3
always represents the encoder’s current position.
The incremental encoder can be connected directly to the microcontroller without
external interface logic. In a standard system, however, comparators will be employed
to convert the encoder’s differential outputs (e.g. A, A) to digital signals (e.g. A). This
greatly increases noise immunity.
Data Sheet
220
2001-02-23
C165UTAH
General Purpose Timer Unit
Table 40
T3I
Core Timer T3 (Incremental Interface Mode) Input Edge Selection
Triggering Edge for Counter Increment / Decrement
000
None. Counter T3 stops.
001
Any transition (rising or falling edge) on T3IN.
010
Any transition (rising or falling edge) on T3EUD.
011
Any transition (rising or falling edge) on any T3 input (T3IN or T3EUD).
1XX
Reserved. Do not use this combination
Note: The third encoder output T0, which indicates the mechanical zero position, may
be connected to an external interrupt input and trigger a reset of timer T3.
External
Encoder
A
A
+
-
A
B
B
+
-
B
T3input
T0
T0
+-
T0
Interrupt
T3input
Microcontroller
Signal Conditioning
Figure 64
Interfacing the Encoder to the Microcontroller
For incremental interface operation the following conditions must be met:
l
l
l
Bitfield T3M must be ’110B’ or ‘111B’.
Pins associated to lines T3IN and T3EUD must be configured as input.
Bit T3UDE must be ’1’ to enable automatic direction control.
The maximum input frequency which is allowed in incremental interface mode is fTimer/8
(FM = 1). To ensure that a transition of any input signal is correctly recognized, its level
should be held high or low for at least 4 fTimer cycles (FM = 1) before it changes.
In Incremental Interface Mode the count direction is automatically derived from the
sequence in which the input signals change, which corresponds to the rotation direction
of the connected sensor. The table below summarizes the possible combinations.
The figures below give examples of T3’s operation, visualizing count signal generation
and direction control. It also shows how input jitter is compensated which might occur if
the sensor rests near to one of its switching points.
Data Sheet
221
2001-02-23
C165UTAH
General Purpose Timer Unit
Table 41
Core Timer T3 (Incremental Interface Mode) Count Direction
Level on respective
other input
Rising
T3IN Input
T3EUD Input
Falling
Rising
Falling
High
Down
Up
Up
Down
Low
Up
Down
Down
Up
Forward
Jitter
Backward
Jitter
Forward
T3IN
ow
U
p
D
Contents
of T3
U
p
T3EUD
n
Note: This example shows the timer behavior assuming that T3 counts upon any
transition on any input, i.e. T3I = ’011B’.
Figure 65
Data Sheet
Evaluation of the Incremental Encoder Signals
222
2001-02-23
C165UTAH
General Purpose Timer Unit
Forward
Jitter
Backward
Jitter
Forward
T3IN
T3EUD
p
U
ow
D
U
p
Contents
of T3
n
Note: This example shows the timer behavior assuming that T3 counts upon any
transition on input T3IN, i.e. T3I = ’001B’.
Figure 66
Evaluation of the Incremental Encoder Signals
Note: Timer T3 operating in incremental interface mode automatically provides
information on the sensor’s current position. Dynamic information (speed,
acceleration, deceleration) may be obtained by measuring the incoming signal
periods.
11.1.1.2 Auxiliary Timers T2 and T4
Note: For the external pin connection of the timer T2 and timer T4, please refer to
Figure 58, page 212.
Both auxiliary timers T2 and T4 have exactly the same functionality with the only
restriction of the availibility of external pins connected to each timer. Timer T2 can be
configured for timer, gated timer, counter, or incremental interface mode. Timer T4 can
be configured for timer, gated timer and counter mode. In addition, the auxiliary timers
can be concatenated with the core timer, or they may be used as reload or capture
registers in conjunction with the core timer.
Note: Timer 2 input T2EUD is connected to P3.5. When the external input for T2EUD is
used (P3.5 configured as input), T4IN and T3EUD can be used by Software
(internal) only.
The individual configuration for timers T2 and T4 is determined by their bitaddressable
control registers T2CON and T4CON, which are both organized identically. Note that
functions which are present in all 3 timers of timer block 1 are controlled in the same bit
positions and in the same manner in each of the specific control registers.
Data Sheet
223
2001-02-23
C165UTAH
General Purpose Timer Unit
Run control for auxiliary timers T2 and T4 can be handled by the associated Run Control
Bit T2R, T4R in register T2CON/T4CON. Alternatively, a remote control option (T2RC,
T4RC = ’1’) may be enabled to start and stop T2/T4 via the run bit T3R of core timer T3.
Timers T2 and T4 in Timer Mode or Gated Timer Mode
When the auxiliary timers T2 and T4 are programmed to timer mode or gated timer
mode, their operation is the same as described for the core timer T3. The descriptions,
figures and tables apply accordingly with two exceptions:
• There is no TxOUT output line for T2 and T4.
• There is no T4EUD input. Therefore Software must be programmed accordingly.
• Overflow/Underflow Monitoring is not supported (no output toggle latch).
Timer T2 in Counter Mode
In counter mode timer T2 can be clocked either by a transition at the respective external
input line T2IN, or by a transition of timer T3’s output toggle latch T3OTL.
Edge
Select
T2IN/
T3OTL
Auxiliary Timer T2
Interrupt
Request
Up/
Down
T2R
T2l
T2UD
T2UDE
MCB02221UTAH
Figure 67
Block Diagram of the Auxiliary Timer T2 in Counter Mode
The event causing an increment or decrement of the timer T2 can be a positive, a
negative, or both a positive and a negative transition at either the input line, or at the
output toggle latch T3OTL.
Bit field T2I in the respective control register T2CON selects the triggering transition (see
table below).
Data Sheet
224
2001-02-23
C165UTAH
General Purpose Timer Unit
Table 42
T2I
Auxiliary Timer T2 (Counter Mode) Input Edge Selection
Triggering Edge for Counter Increment / Decrement
X00
None. Counter T2 is disabled
001
Positive transition (rising edge) on T2IN
010
Negative transition (falling edge) on T2IN
011
Any transition (rising or falling edge) on T2IN
101
Positive transition (rising edge) of output toggle latch T3OTL
110
Negative transition (falling edge) of output toggle latch T3OTL
111
Any transition (rising or falling edge) of output toggle latch T3OTL
Note: Only state transitions of T3OTL which are caused by the overflows/underflows of
T3 will trigger the counter function of T2. Modifications of T3OTL via software will
NOT trigger the counter function of T2.
For counter operation, an external pin associated to line T2IN must be configured as
input. The maximum input frequency which is allowed in counter mode is fTimer/8 (FM1 =
’1’). To ensure that a transition of the count input signal which is applied to T2IN is
correctly recognized, its level should be held for at least 4 fTimer cycles (FM1 = ’1’) before
it changes.
Timer T4 in Counter Mode
The operation of Timer T4 in counter mode is exactly the same as described for Timer T2
in Chapter "Timer T2 in Counter Mode" above.
11.1.1.3 Timer Concatenation
Using the output toggle latch T3OTL as a clock source for an auxiliary timer in counter
mode concatenates the core timer T3 with the respective auxiliary timer. Depending on
which transition of T3OTL is selected to clock the auxiliary timer, this concatenation
forms a 32-bit or a 33-bit timer/counter.
• 32-bit Timer/Counter: If both a positive and a negative transition of T3OTL is used to
clock the auxiliary timer, this timer is clocked on every overflow/underflow of the core
timer T3. Thus, the two timers form a 32-bit timer.
• 33-bit Timer/Counter: If either a positive or a negative transition of T3OTL is selected
to clock the auxiliary timer, this timer is clocked on every second overflow/underflow
of the core timer T3. This configuration forms a 33-bit timer (16-bit core
timer+T3OTL+16-bit auxiliary timer).
Data Sheet
225
2001-02-23
C165UTAH
General Purpose Timer Unit
The count directions of the two concatenated timers are not required to be the same.
This offers a wide variety of different configurations.
T3 can operate in timer, gated timer or counter mode in this case.
Interrupt
Request
T3l
2n : 1
fhw_clk
Core Timer T3
T3R
T3OTL
Up/Down
T3OUT
T3OE
*)
Edge
Select
Auxiliary Timer Tx
Interrupt
Request
TxIR
TxR
x = 2,4
MCB02034gpt1
Txl
Note: Line ’*’ only affected by over/underflows of T3, but NOT by software
modifications of T3OTL.
Figure 68
Concatenation of Core Timer T3 and an Auxiliary Timer
Auxiliary Timer in Reload Mode
Reload mode for the auxiliary timers T2 and T4 is selected by setting bit field TxM in the
respective register TxCON to ‘100B’. In reload mode the core timer T3 is reloaded with
the contents of an auxiliary timer register, triggered by one of two different signals. The
trigger signal is selected the same way as the clock source for counter mode (see table
above), i.e. a transition of the auxiliary timer’s input or the output toggle latch T3OTL may
trigger the reload.
Note: When programmed for reload mode, the respective auxiliary timer (T2 or T4) stops
independent of its run flag T2R or T4R.
Data Sheet
226
2001-02-23
C165UTAH
General Purpose Timer Unit
Source/Edge
Select
Reload Register Tx
Interrupt
Request
TxIN
TxI
*)
Input
Clock
Interrupt
Request
Core Timer T3
Up/Down
T3OTL
T3OUT
T3OE
x = (2, 4)
MCB02035UTAH
Note: Line
only
Figure 69
’*’
GPT1 Auxiliary Timer in Reload Mode
Upon a trigger signal T3 is loaded with the contents of the respective timer register (T2
or T4) and the interrupt request flag (T2IR or T4IR) is set.
Note: When a T3OTL transition is selected for the trigger signal, also the interrupt
request flag T3IR will be set upon a trigger, indicating T3’s overflow or underflow.
Modifications of T3OTL via software will NOT trigger the counter function of T2/T4.
The reload mode triggered by T3OTL can be used in a number of different
configurations. Depending on the selected active transition the following functions can
be performed:
• If both a positive and a negative transition of T3OTL is selected to trigger a reload, the
core timer will be reloaded with the contents of the auxiliary timer each time it
overflows or underflows. This is the standard reload mode (reload on overflow/
underflow).
• If either a positive or a negative transition of T3OTL is selected to trigger a reload, the
core timer will be reloaded with the contents of the auxiliary timer on every second
overflow or underflow.
• Using this “single-transition” mode for both auxiliary timers allows to perform very
flexible pulse width modulation (PWM). One of the auxiliary timers is programmed to
reload the core timer on a positive transition of T3OTL, the other is programmed for a
Data Sheet
227
2001-02-23
C165UTAH
General Purpose Timer Unit
reload on a negative transition of T3OTL. With this combination the core timer is
alternately reloaded from the two auxiliary timers.
The figure below shows an example for the generation of a PWM signal using the
alternate reload mechanism. T2 defines the high time of the PWM signal (reloaded on
positive transitions) and T4 defines the low time of the PWM signal (reloaded on negative
transitions). The PWM signal can be output on line T3OUT if the control bit T3OE is set
to ‘1’. With this method the high and low time of the PWM signal can be varied in a wide
range.
Note: The output toggle latch T3OTL is accessible via software and may be changed, if
required, to modify the PWM signal. However, this will NOT trigger the reloading
of T3.
Note: An associated port pin linked to line T3OUT should be configured as output.
R eload R egister T 2
Interrupt
R equest
*)
T 2I
Input
C lock
C ore T im er T 3
T 3O T L
T 3O U T
T 3O E
U p/D ow n
Interrupt
R equest
*)
Interrupt
R equest
R eload R egister T 4
T 4I
M C B 02037
Note: Lines ’*’ only affected by over/underflows of T3, but NOT by software
modifications of T3OTL.
Figure 70
GPT1 Timer Reload Configuration for PWM Generation
Data Sheet
228
2001-02-23
C165UTAH
General Purpose Timer Unit
Note: Although it is possible, it should be avoided to select the same reload trigger event
for both auxiliary timers. In this case both reload registers would try to load the
core timer at the same time. If this combination is selected, T2 is disregarded and
the contents of T4 is reloaded.
Auxiliary Timer in Capture Mode
Capture mode for the auxiliary timers T2 and T4 is selected by setting bit field TxM in the
respective register TxCON to ‘101B’. In capture mode the contents of the core timer are
latched into an auxiliary timer register in response to a signal transition at the respective
auxiliary timer's external input line TxIN. The capture trigger signal can be a positive, a
negative, or both a positive and a negative transition.
The two least significant bits of bit field TxI are used to select the active transition (see
table in the counter mode section), while the most significant bit TxI.2 is irrelevant for
capture mode. It is recommended to keep this bit cleared (TxI.2 = ‘0’).
Note: When programmed for capture mode, the respective auxiliary timer (T2 or T4)
stops independent of its run flag T2R or T4R.
E dg e
S elec t
C a pture R e gister T x
In terru pt
R eq ue st
T xIN
In put
C lo ck
T xI
U p/D ow n
T3OTL
x = (2, 4)
Figure 71
In terru pt
R eq ue st
C ore T im er T 3
T3OUT
T3OE
M C B 0 203 8
Auxiliary Timer of Timer Block 1 in Capture Mode
Upon a trigger (selected transition) at the corresponding input line TxIN the contents of
the core timer are loaded into the auxiliary timer register and the associated interrupt
request flag TxIR will be set.
Data Sheet
229
2001-02-23
C165UTAH
General Purpose Timer Unit
Note: The direction control for T2IN and for T4IN must be set to 'Input', and the level of
the capture trigger signal should be held high or low for at least 4 fTimer (FM1 = ’1’)
cycles before it changes to ensure correct edge detection.
11.1.2
Functional Description of Timer Block 2
Timer block 2 includes the two timers T5 (referred to as the auxiliary timer) and T6
(referred to as the core timer), and the 16-bit capture/reload register CAPREL.
Note: The block 2 Timer T5 and Timer T6 can be used by Software only. There exist no
external pin for timer T5 and T6. Additional, there is no external pin connected to
register CAPREL, see also Figure 58, page 212.
The count direction (Up / Down) must be programmed via software. An overflow/
underflow of core timer T6 is indicated by the output toggle latch T6OTL whose state may
be output on line T6OFL. The auxiliary timer T6 may be reloaded with the contents of
CAPREL.
The toggle bit also supports the concatenation of T6 with auxiliary timer T5, while
concatenation of T6 with other timers is provided through line T6OFL. Triggered by an
external signal, T3IN or T3EUD, the contents of timer T5 can be captured into register
CAPREL, and T5 may optionally be cleared. Both timer T6 and T5 can count up or down,
and the current timer value can be read or modified by the CPU in the nonbitaddressable SFRs T5 and T6.
Data Sheet
230
2001-02-23
C165UTAH
General Purpose Timer Unit
2n : 1
f h w _ c lk
T5
M ode
C o n tro l
U /D
In te rru p t
Request
G P T 2 T im e r T 5
C le a r
C a p tu re
In te rru p t
Request
T 3 IN /
T3EUD
CT3
GPT2 CAPREL
In te rru p t
Request
C le a r
G P T 2 T im e r T 6
f h w _ c lk
2n : 1
T6OTL
U /D
T6
M ode
C o n tro l
T6OFL
M CB 03999
Figure 72
Structure of Timer Block 2
11.1.2.1 Core Timer T6
The operation of the core timer T6 is controlled by its bitaddressable control register
T6CON.
Timer 6 Run Bit
The timer can be started or stopped by software through bit T6R (Timer T6 Run Bit).
Setting bit T6R to ‘1’ will start the timer, clearing T6R stops the timer.
Count Direction Control
The count direction of the core timer can be controlled by software only. The count
direction can be altered by setting or clearing bit T6UD.
Note: Bit T6UDE of register T6CON must be always set to ’0’.
The count direction can be changed regardless of whether the timer is running or not.
Data Sheet
231
2001-02-23
C165UTAH
General Purpose Timer Unit
Table 43
Core Timer T6 Count Direction Control
Bit TxUDE
Bit TxUD
Count Direction
0
0
Count Up
0
1
Count Down
Note: The direction control works the same for core timer T6 and for auxiliary timer T5.
Therefore the lines and bits are named Tx...
Timer 6 Overflow/Underflow Monitoring
An overflow or underflow of timer T6 will clock the toggle latch T6OTL in control register
T6CON. T6OTL can also be set or reset by software.
In addition, T6OTL can be used in conjunction with the timer over/underflows as an input
for the counter function of the auxiliary timer T5. For this purpose, an internal connection
is provided for this option.
An overflow or underflow of timer T6 can also be used to clock other timers. For this
purpose, there is the special output line T6OFL.
Timer 6 in Timer Mode
Timer mode for the core timer T6 is selected by setting bit field T6M in register T6CON
to ‘000B’. In this mode, T6 is clocked with the module clock divided by a programmable
prescaler, which is selected by bit field T6I. The input frequency fT6 for timer T6 and its
resolution rT6 are scaled linearly with lower clock frequencies fTimer, as can be seen from
the following formula:
fT6 =
Data Sheet
fTimer
rT6 [µs] =
4 * 2
232
4 * 2
fTimer [MHz]
2001-02-23
C165UTAH
General Purpose Timer Unit
Txl
fhw_clk
2n : 1
Interrupt
Request
Core Timer Tx
TxR
Up/
Down
TxOTL
TxUD
x=6
TxUDE
Figure 73
MCB02028
Block Diagram of Core Timer T6 in Timer Mode
11.1.2.2 Auxiliary Timer T5
The auxiliary timer T5 can be configured for timer mode with the same options for the
timer frequencies and the count signal as the core timer T6. In addition, the auxiliary
timer can be concatenated with the core timer.
The individual configuration for timer T5 is determined by its bitaddressable control
register T5CON. Note that functions which are present in both timers of timer block 2 are
controlled in the same bit positions and in the same manner in each of the specific control
registers.
Run control for auxiliary timer T5 can be handled by the associated Run Control Bit T5R
in register T5CON. Alternatively, a remote control option (T5RC = ’1’) may be enabled
to start and stop T5 via the run bit T6R of core timer T6.
Note: The auxiliary timer has no overflow/underflow toggle latch. Therefore, an output
line for Overflow/Underflow Monitoring is not provided.
Count Direction Control for Auxiliary Timer
The count direction of the auxiliary timer can be controlled in the same way as for the
core timer T6. The description and the table apply accordingly.
Timer T5 in Counter Mode
Counter mode for the auxiliary timer T5 is selected by setting bit field T5M in register
T5CON to ‘001B’. In counter mode, timer T5 can be clocked by a transition of timer T6’s
output signal T6OFL only.
Data Sheet
233
2001-02-23
C165UTAH
General Purpose Timer Unit
Edge
Select
T6OFL
Auxiliary Timer Tx
Interrupt
Request
Up/
Down
TxR
Txl
TxUD
x=5
TxUDE
MCB02221
Figure 74
Block Diagram of Auxiliary Timer T5 in Counter Mode
The event causing an increment or decrement of the timer can be a positive, a negative,
or both a positive and a negative transition at signal T6OFL (toggle latch T6OTL).
Bit field T5P in control register T5CON selects the triggering transition (see table below).
Table 44
Auxiliary Timer (Counter Mode) Input Edge Selection
T5P
Triggering Edge for Counter Increment / Decrement
X00
None. Counter T5 is disabled
001
reserved, do not use this combination
010
reserved, do not use this combination
011
reserved, do not use this combination
101
Positive transition (rising edge) of output toggle latch T6OTL
110
Negative transition (falling edge) of output toggle latch T6OTL
111
Any transition (rising or falling edge) of output toggle latch T6OTL
Note: Only state transitions of T6OTL which are caused by the overflows/underflows of
T6 will trigger the counter function of T5. Modifications of T6OTL via software will
NOT trigger the counter function of T5.
Data Sheet
234
2001-02-23
C165UTAH
General Purpose Timer Unit
11.1.2.3 Timer Concatenation
Using the toggle bit T6OTL as a clock source for the auxiliary timer in counter mode
concatenates the core timer T6 with the auxiliary timer. Depending on which transition of
T6OTL is selected to clock the auxiliary timer, this concatenation forms a 32-bit or a 33bit timer / counter.
l
l
32-bit Timer/Counter: If both a positive and a negative transition of T6OTL is used to clock the
auxiliary timer, this timer is clocked on every overflow/underflow of the core timer T6. Thus, the
two timers form a 32-bit timer.
33-bit Timer/Counter: If either a positive or a negative transition of T6OTL is selected to clock the
auxiliary timer, this timer is clocked on every second overflow/underflow of the core timer T6.
This configuration forms a 33-bit timer (16-bit core timer+T6OTL+16-bit auxiliary timer).
The count directions of the two concatenated timers are not required to be the same.
This offers a wide variety of different configurations.
T6l
Interrupt
Request
2n : 1
fhw_clk
Core Timer T6
T6R
T6OTL
Up/Down
*)
Edge
Select
Auxiliary Timer T5
T5IR
Interrupt
Request
T5R
MCB02034gpt2
T5l
Note: Line ’*’ only affected by over/underflows of T6, but NOT by software
modifications of T6OTL.
Figure 75
Concatenation of Core Timer T6 and Auxiliary Timer T5
Data Sheet
235
2001-02-23
C165UTAH
General Purpose Timer Unit
Timer Block 2 Capture/Reload Register CAPREL in Reload Mode
The 16-bit capture/reload register CAPREL can be used as a reload register for the core
timer T6. This mode is selected by setting bit T6SR = ‘1’ in register T6CON. The event
causing a reload in this mode is an overflow or underflow of the core timer T6.
When timer T6 overflows from FFFFH to 0000H (when counting up) or when it underflows
from 0000H to FFFFH (when counting down), the value stored in register CAPREL is
loaded into timer T6. This will not set the interrupt request flag CRIR associated with the
CAPREL register. However, interrupt request flag T6IR will be set indicating the
overflow/underflow of T6.
C A P R E L R egis te r
T 6O T L
T6SR
In put
C lo ck
Inte rru pt
R e qu es t
C o re T im e r T 6
T6OFL
U p/D ow n
M CB 020 45
Figure 76
11.1.3
Timer Block 2 Register CAPREL in Reload Mode
GPT Register Set
All GPT12 related registers are summarized in the overview table below.
Table 45
GPT Register Overview
Name
Description
Address
Reset Value
GPTCLC
GPT Clock Control Register
FE4CH
0000H
T2CON
Timer 2 Function Control Register
FF40H
0000H
T3CON
Timer 3 Function Control Register
FF42H
0000H
T4CON
Timer 4 Function Control Register
FF44H
0000H
T5CON
Timer 5 Function Control Register
FF46H
0000H
T6CON
Timer 6 Function Control Register
FF48H
0000H
T2
GPT1 Timer 2 Register
FE40H
0000H
Data Sheet
236
2001-02-23
C165UTAH
General Purpose Timer Unit
Table 45
GPT Register Overview
Name
Description
Address
Reset Value
T3
GPT1 Timer 3 Register
FE42H
0000H
T4
GPT1 Timer 4 Register
FE44H
0000H
T5
GPT2 Timer 5 Register
FE46H
0000H
T6
GPT2 Timer 6 Register
FE48H
0000H
CAPREL
Capture Reload Register
FE4AH
0000H
1)
GPT1 Timer 2 Interrupt Control Register
FF60H
0000H
1)
GPT1 Timer 3 Interrupt Control Register
FF62H
0000H
1)
GPT1 Timer 4 Interrupt Control Register
FF64H
0000H
1)
GPT2 Timer 5 Interrupt Control Register
FF66H
0000H
T6IC1)
GPT2 Timer 6 Interrupt Control Register
FF68H
0000H
GPT2 CAPREL Interrupt Control Register
FF6AH
0000H
T2IC
T3IC
T4IC
T5IC
1)
CRIC
1)
For the Interrupt Control Register description, please refer to Chapter 7.2, page 108.
GPT Clock Control Register
GPTCLC (FE4CH)
Reset Value: 0000H
15
14
13
12
11
10
9
8
7
6
5
4
0
0
0
0
0
0
0
0
0
0
0
0
3
Bit
Function
GPTDISR
GPT Disable Request Bit
GPTDISR = ‘0’: GPT clock disable not requested
GPTDISR = ‘1’: GPT clock disable requested
GPTDISS
GPT Disable Status Bit
GPTDISS = ‘0’: GPT clock enabled
GPTDISS = ‘1’: GPT clock disabled
SUSPEN
Peripheral Suspend Enable Bit for OCDS
SUSPEN = ‘0’: Peripheral suspend disabled
SUSPEN = ‘1’: Peripheral suspend enabled
EXDISR
External Disable Request
EXRDIS = ‘0’: External clock disable Request is enabled
EXRDIS = ‘1’: External clock disable Request is disabled
237
1
0
EX SUS GPT GPT
DISR PEN DISS DISR
rw
Data Sheet
2
rw
r
rw
2001-02-23
C165UTAH
General Purpose Timer Unit
Function Control Registers
The operating mode of the core timer T3 is configured and controlled via its
bitaddressable control register T3CON.
T3CON
Timer 3 Control Register
15
14
13
12
T3
T3
T3 T3CH
EDG
IREN RDIR DIR
E
Reset Value: 0000H
11
FM1
10
9
8
7
6
T3
T3
T3OE
T3UD T3R
OTL
UDE
Field
Bits
Type Value Description
T3I
[2:0]
rw
T3M
[5:3]
rw
T3R
T3UD
T3UDE
6
7
8
4
3
T3M
2
1
0
T3I
Timer 3 Input Parameter Selection
Timer mode see Table 46 for encoding
Gated Timer see Table 46 for encoding
Counter mode see Table 47 for encoding
Incremental Interface mode see Table 48 for
encoding
000
001
010
011
100
101
110
111
Timer 3 Mode Control
Timer Mode
Counter Mode
Gated Timer with Gate active low
Gated Timer with Gate active high
Reserved. Do not use this combination!
Reserved. Do not use this combination!
Incremental Interface Mode ( Rotation detection )
Incremental Interface Mode ( Edge detection )
0
1
Timer 3 Run Bit
Timer / Counter 3 stops
Timer / Counter 3 runs
0
1
Timer 3 Up / Down Control
(when T3UDE = ’0)
Counting ’Up’
Counting ’Down’
rw
rw
rw
0
1
Data Sheet
5
Timer 3 External Up/Down Enable
Counting direction is internally controlled by SW
Counting direction is externally controlled by line
T3EUD
238
2001-02-23
C165UTAH
General Purpose Timer Unit
Field
Bits
Type Value Description
T3OE
9
rw
0
1
T3OTL
10
rw
0/1
FM1
11
rw
0
1
T3EDGE
12
T3CHDIR
13
T3RDIR
14
T3IREN
15
rw
Timer 3 Output Toggle Latch
Toggles on each overflow / underflow of T3. Can
be set or reset by software.
Fast Mode for Timer Block 1
The maximum input frequency
for Timer 2/3/4 is fTimer / 8.
The maximum input frequency
for Timer 2/3/4 is fTimer / 4.
0
1
Timer 3 Edge Detection
The bit is set on each successful edge detection.
The bit has to be reset by SW.
No count edge was detected
A count edge was detected
0
1
Timer 3 Count Direction Change
The bit is set on a change of the countdirection of
timer 3. The bit has to be reset by SW.
No change in count direction was detected
A change in count direction was detected
0
1
Timer 3 Rotation Direction
Timer 3 counts up.
Timer 3 counts down.
rw
r
rw
0
1
Table 46
Overflow/Underflow Output Enable
T3 overflow/underflow can not be externally
monitored
T3 overflow/underflow may be externally
monitored via T3OUT
Timer 3 Interrupt Enable
Interrupt generation for T3CHDIR and T3EDGE
is disabled.
Interrupt generation for T3CHDIR and T3EDGE
is enabled.
Timer 3 Input Parameter Selection for Timer mode and Gated mode
T3I
Prescaler for fTimer ( FM1 = 0 )
Prescaler for fTimer ( FM1 = 1 )
000
8
4
001
16
8
010
32
16
011
64
32
Data Sheet
239
2001-02-23
C165UTAH
General Purpose Timer Unit
Table 46
Timer 3 Input Parameter Selection for Timer mode and Gated mode
T3I
Prescaler for fTimer ( FM1 = 0 )
Prescaler for fTimer ( FM1 = 1 )
100
128
64
101
256
128
110
512
256
111
1014
512
Table 47
Timer 3 Input Parameter Selection for Counter mode
T3I
Triggering Edge for Counter Update
000
None. Counter T3 is disabled
001
Positive transition ( raising edge ) on T3IN
010
Negative transition ( falling edge ) on T3IN
011
Any transition ( raising or falling edge ) on T3IN
1XX
Reserved. Do not use this combination!
Table 48
Timer 3 Input Parameter Selection for Incremental Interface mode
T3I
Triggering Edge for Counter Update
000
None. Counter T3 stops
001
Any transition ( raising or falling edge ) on T3IN
010
Any transition ( raising or falling edge ) on T3EUD
011
Any transition ( raising or falling edge ) on T3IN or T3EUD
1XX
Reserved. Do not use this combination!
T2CON
Timer 2 Control Register
15
14
13
12
T2
T2
T2 T2CH
EDG
IREN RDIR DIR
E
Data Sheet
Reset Value: 0000H
11
10
0
9
T2RC
8
7
6
T2
T2UD T2R
UDE
240
5
4
T2M
3
2
1
0
T2I
2001-02-23
C165UTAH
General Purpose Timer Unit
Field
Bits
Type Value
Description
T2I
[2:0]
rw
Timer 2 Input Parameter Selection
Timer mode see Table 49 for encoding
Gated Timer see Table 49 for encoding
Counter mode see Table 50 for encoding
Incremental Interface mode see Table 51 for
encoding
T2M
[5:3]
rw
T2R
T2UD
T2UDE
6
7
8
000
001
010
011
100
101
110
111
Timer 2 Mode Control (Basic Operating Mode)
Timer Mode
Counter Mode
Gated Timer with Gate active low
Gated Timer with Gate active high
Reload Mode
Capture Mode
Incremental Interface Mode ( Rotation detection)
Incermental Interface Mode ( Edge detection )
0
1
Timer 2 Run Bit
Timer / Counter 2 stops
Timer / Counter 2 runs
0
1
Timer 2 Up / Down Control
(when T2UDE = ’0)
Counting ’Up’
Counting ’Down’
rw
rw
rw
0
1
Timer 2 External Up/Down Enable
Counting direction is internally controlled by SW
Counting direction is externally controlled by line
T2EUD.
Note: Pin P3.5 connected to T2EUD is also
connected to T4IN and to T3EUD.
T2RC
9
rw
0
1
Timer 2 Remote Control
Timer / Counter 2 is controlled by
its own run bit T2R
Timer / Counter 2 is controlled by
the run bit of core timer 3
0
[11:10] r
reserved for future use; reading returns 0;
writing to these bit positions has no effect.
T2EDGE
12
Timer 2 Edge Detection
The bit is set on each successful edge detection.
The bit has to be reset by SW.
No count edge was detected
A count edge was detected
rw
0
1
Data Sheet
241
2001-02-23
C165UTAH
General Purpose Timer Unit
Field
Bits
Type Value
Description
T2CHDIR
13
rw
0
1
Timer 2 Count Direction Change
The bit is set on a change of the countdirection of
timer 2. The bit has to be reset by SW.
No change in count direction was detected
A change in count direction was detected
0
1
Timer 2 Rotation Direction
Timer 2 counts up.
Timer 2 counts down.
T2RDIR
14
T2IREN
15
r
rw
0
1
Table 49
Timer 2 Interrupt Enable
Interrupt generation for T2CHDIR and T2EDGE
is disabled.
Interrupt generation for T2CHDIR and T2EDGE
is enabled.
Timer 2 Input Parameter Selection for Timer mode and Gated mode
T2I
Prescaler for fTimer ( FM1 = 0 )
Prescaler for fTimer ( FM1 = 1 )
000
8
4
001
16
8
010
32
16
011
64
32
100
128
64
101
256
128
110
512
256
111
1014
512
Table 50
Timer 2 Input Parameter Selection for Counter mode
T2I
Triggering Edge for Counter Update
X00
None. Counter T2 is disabled
001
Positive transition (rising edge) on T2IN
010
Negative transition (falling edge) on T2IN
011
Any transition (rising or falling edge) on T2IN
101
Positive transition (rising edge) of output toggle latch T3OTL
110
Negative transition (falling edge) of output toggle latch T3OTL
111
Any transition (rising or falling edge) of output toggle latch T3OTL
Data Sheet
242
2001-02-23
C165UTAH
General Purpose Timer Unit
Table 51
Timer 2 Input Parameter Selection for Incremental Interface mode
T2I
Triggering Edge for Counter Update
000
None. Counter T2 stops
001
Any transition ( raising or falling edge ) on T2IN
010
Any transition ( raising or falling edge ) on T2EUD
011
Any transition ( raising or falling edge ) on T2IN or T2EUD
1XX
Reserved. Do not use this combination!
T4CON
Timer 4 Control Register
15
14
13
12
Reset Value: 0000H
11
T4
T4
T4 T4CH
EDG
IREN RDIR DIR
E
10
9
8
7
6
’0’
T4RC (T4 T4UD T4R
UDE)
’0’
5
4
3
T4M
2
1
0
T4I
Field
Bits
Type Value
Description
T4I
[2:0]
rw
Timer 4 Input Parameter Selection
Timer mode see Table 49 for encoding
Gated Timer see Table 49 for encoding
Counter mode see Table 50 for encoding
T4M
[5:3]
rw
000
001
010
011
100
101
110
111
Timer 4 Mode Control (Basic Operating Mode)
Timer Mode
Counter Mode
Gated Timer with Gate active low
Gated Timer with Gate active high
Reload Mode
Capture Mode
Reserved. Do not use this combination.
Reserved. Do not use this combination.
0
1
Timer 4 Run Bit
Timer / Counter 4 stops
Timer / Counter 4 runs
0
1
Timer 4 Up / Down Control
Counting ’Up’
Counting ’Down’
T4R
T4UD
Data Sheet
6
7
rw
rw
243
2001-02-23
C165UTAH
General Purpose Timer Unit
Field
Bits
Type Value
Description
’0’ (T4UDE-bit)
8
rw
Timer 4 External Up/Down Enable
This bit must be set to ’0’ signal.
0
T4RC
9
rw
0
1
Timer 4 Remote Control
Timer / Counter 4 is controlled by
its own run bit T4R
Timer / Counter 4 is controlled by
the run bit of core timer 3
’0’
[11:10] r
reserved for future use; reading returns 0;
writing to these bit positions has no effect.
T4EDGE
12
0
1
Timer 4 Edge Detection
The bit is set on each successful edge detection.
The bit has to be reset by SW.
No count edge was detected
A count edge was detected
0
1
Timer 4 Count Direction Change
The bit is set on a change of the countdirection of
timer 4. The bit has to be reset by SW.
No change in count direction was detected
A change in count direction was detected
0
1
Timer 4 Rotation Direction
Timer 4 counts up.
Timer 4 counts down.
T4CHDIR
13
T4RDIR
14
T4IREN
15
rw
rw
r
rw
0
1
Table 52
Timer 4 Interrupt Enable
Interrupt generation for T4CHDIR and T4EDGE
is disabled.
Interrupt generation for T4CHDIR and T4EDGE
is enabled.
Timer 4 Input Parameter Selection for Timer mode and Gated mode
T4I
Prescaler for fTimer ( FM1 = 0 )
Prescaler for fTimer ( FM1 = 1 )
000
8
4
001
16
8
010
32
16
011
64
32
Data Sheet
244
2001-02-23
C165UTAH
General Purpose Timer Unit
Table 52
Timer 4 Input Parameter Selection for Timer mode and Gated mode
T4I
Prescaler for fTimer ( FM1 = 0 )
Prescaler for fTimer ( FM1 = 1 )
100
128
64
101
256
128
110
512
256
111
1014
512
Table 53
Timer 4 Input Parameter Selection for Counter mode
T4I
Triggering Edge for Counter Update
X00
None. Counter T4 is disabled
001
Positive transition (rising edge) on T4IN
010
Negative transition (falling edge) on T4IN
011
Any transition (rising or falling edge) on T4IN
101
Positive transition (rising edge) of output toggle latch T3OTL
110
Negative transition (falling edge) of output toggle latch T3OTL
111
Any transition (rising or falling edge) of output toggle latch T3OTL
T6CON
Timer 6 Control Register
15
14
T6SR
T6
CLR
13
12
’0’
Reset Value: 0000H
11
10
FM2
T6
OTL
9
8
7
6
’0’
’0’
(T6 T6UD T6R
(T6
OE) UDE)
5
4
T6M
3
2
1
T6I
Field
Bits
Type Value
Description
T6I
[2:0]
rw
Timer 6 Input Parameter Selection
Timer mode see Table 54 for encoding
T6M
[5:3]
rw
Timer 6 Mode Control (Basic Operating
Mode)
Timer Mode
Reserved. Do not use this combination!
Reserved. Do not use this combination!
Reserved. Do not use this combination!
Reserved. Do not use this combination!
000
001
010
011
1xx
Data Sheet
245
0
2001-02-23
C165UTAH
General Purpose Timer Unit
Field
Bits
Type Value
Description
T6R
6
rw
0
1
Timer 6 Run Bit
Timer / Counter 6 stops
Timer / Counter 6 runs
0
1
Timer 6 Up / Down Control
Counting ’Up’
Counting ’Down’
0
Timer 6 External Up/Down Enable
This bit must be set to ’0’ signal
0
Overflow/Underflow Output Enable
This bit must be set to ’0’ signal
T6UD
7
’0’ (T6UDE)
8
’0’ (T6OE)
9
T6OTL
10
rw
rw
rw
rw
Timer 6 Output Toggle Latch
Toggles on each overflow / underflow of T6.
Can be set or reset by software.
0/1
FM2
11
rw
Fast Mode for Timer Block 2
The maximum input frequency
for Timer 5/6 is fTimer / 4.
The maximum input frequency
for Timer 5/6 is fTimer / 2.
0
1
’0’
[13:12] r
reserved for future use; reading returns 0;
writing to these bit positions has no effect.
T6CLR
14
0
1
Timer 6 Clear Bit
Timer 6 is not cleared on a capture event
Timer 6 is cleared on a capture event
0
1
Timer 6 Reload Mode Enable
Reload from register CAPREL Disabled
Reload from register CAPREL Enabled
T6SR
Table 54
15
rw
rw
Timer 6 Input Parameter Selection for Timer mode and Gated mode
T6I
Prescaler for fTimer ( FM2 = 0 )
Prescaler for fTimer ( FM2 = 1 )
000
4
2
001
8
4
010
16
8
011
32
16
100
64
32
101
128
64
110
256
128
111
512
256
Data Sheet
246
2001-02-23
C165UTAH
General Purpose Timer Unit
T5CON
Timer 5 Control Register
15
T5SC
14
T5
CLR
13
12
CI
Reset Value: 0000H
11
CC
10
9
8
7
6
’0’
’1’
T5RC (T5 T5UD T5R
(CT3)
UDE)
5
’0’
4
3
2
1
T5M
T5I
Field
Bits
Type Value
Description
T5I
[2:0]
rw
Timer 5 Input Parameter Selection
Timer mode see Table 55 for encoding
Counter mode see Table 56 for encoding
T5M
[4:3]
rw
00
01
10
11
Timer 5 Mode Control (Basic Operating Mode)
Timer Mode
Counter Mode
Reserved. Do not use this configuration
Reserved. Do not use this configuration
’0’
5
r
reserved for future use; reading returns 0;
writing to these bit positions has no effect.
T5R
6
rw
0
1
Timer 5 Run Bit
Timer / Counter 5 stops
Timer / Counter 5 runs
0
1
Timer 5 Up / Down Control
Counting ’Up’
Counting ’Down’
0
Timer 5 External Up/Down Enable
This bit must be set to ’0’ signal
T5UD
’0’ (T5UDE)
T5RC
7
8
9
rw
rw
rw
0
1
’1’ (CT3)
CC
Data Sheet
10
11
rw
0
Timer 5 Remote Control
Timer / Counter x is controlled by
its own run bit T5R
Timer / Counter 5 is controlled by
the run bit of core timer 6 (T6R)
0
Timer 3 Capture Trigger Enable
This bit must be set to ’1’ signal
0
1
Capture Correction
T5 is just captured
T5 is decremented by 1 before being captured
rw
247
2001-02-23
C165UTAH
General Purpose Timer Unit
Field
Bits
CI
[13:12] rw
T5CLR
14
T5SC
15
Table 55
Type Value
Description
00
01
10
11
Register CAPREL Capture Trigger Selection
Capture disabled
Any transition on T3IN
Any transition on T3EUD
Any transition on T3IN or T3EUD
0
1
Timer 5 Clear Bit
Timer 5 not cleared on a capture
Timer 5 is cleared on a capture
0
1
Timer 5 Capture Mode Enable
Capture into register CAPREL Disabled
Capture into register CAPREL Enabled
rw
rw
Timer 5 Input Parameter Selection for Timer mode
T5I
Prescaler for fTimer ( FM2 = 0 )
Prescaler for fTimer ( FM2 = 1 )
000
4
2
001
8
4
010
16
8
011
32
16
100
64
32
101
128
64
110
256
128
111
512
256
Table 56
Timer 5 Input Parameter Selection for Counter mode
T5I
Triggering Edge for Counter Update
X00
None. Counter T5 is disabled
001
Reserved, do not use this combination
010
Reserved, do not use this combination
011
Reserved, do not use this combination
101
Positive transition (rising edge) of output toggle latch T6OTL
110
Negative transition (falling edge) of output toggle latch T6OTL
111
Any transition (rising or falling edge) of output toggle latch T6OTL
Data Sheet
248
2001-02-23
C165UTAH
General Purpose Timer Unit
T2/T3/T4/T5/T6
Timer Tx Register
15
14
13
Reset Value: 0000H
12
11
10
9
8
7
6
5
4
3
2
1
0
value
Field
Bits
Type Value
Description
value
[15:0]
rw
Timer Tx Register
16 bit register contains the actual timer value of
the respective Timer Tx.
CAPREL
GPT2 CAPREL Register
15
14
13
12
Reset Value: 0000H
11
10
9
8
7
6
5
4
3
2
1
0
value
Field
Bits
Type Value
Description
value
[15:0]
rw
GPT2 CAPREL Register
16 bit register contains the actual value of the
CAPREL register.
Data Sheet
249
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
12
Asynchronous/Synchr. Serial Interface
The Asynchronous/Synchronous Serial Interface (ASC) provides serial communication
between the C165UTAH and other microcontrollers, microprocessors or external
peripherals. The ASC supports a certain protocol to transfer data via a serial
interconnection.
12.1
Functional Description
The ASC supports full-duplex asynchronous communication up to 2.25 MBaud and halfduplex synchronous communication up to 4.5 MBaud (@ 36 MHz CPU clock which is
equal to the ASC module clock). In synchronous mode, data are transmitted or received
synchronous to a shift clock which is generated by the microcontroller. In asynchronous
mode, 8- or 9-bit data transfer, parity generation, and the number of stop bits can be
selected.
12.1.1
Features
• Full duplex asynchronous operating modes
– 8- or 9-bit data frames, LSB first
– Parity bit generation/checking
– One or two stop bits
– Baudrate from 2.25 MBaud to 0.5364 Baud (@ 36 MHz module clock = CPU clock)
– Multiprocessor mode for automatic address/data byte detection
– Loop-back capability
– Support for IrDA data transmission/reception up to max. 115.2 kBaud
• Autobaud detection unit for asynchronous operating modes
– Detection of standard baudrates
1200, 2400, 4800, 9600, 19200, 38400, 57600, 115200, 230400 Baud
– Detection of non-standard baudrates
– Detection of asynchronous modes
7 bit, even parity; 7 bit, odd parity;
8 bit, even parity; 8 bit, odd parity; 8 bit, no parity
– Automatic initialization of control bits and baudrate generator after detection
– Detection of a serial two-byte ASCII character frame
• Recently introduced fractional divider
– The fractional divider drastically improves the accuracy of the adjustment for
baudrates
– Standard Baud Rates generation with very small deviation (230.4 kBaud < 0.01%,
460.8 kBaud < 0.15 %, 691.2 kBaud < 0.04 %, 921.6 kBaud < 0.15 % ) @ 36 MHz
• Half-duplex 8-bit synchronous operating mode
Data Sheet
250
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
– Baudrate from 4.5 MBaud to 366.21 Baud (@ 36 MHz module clock = CPU clock)
• Double buffered transmitter/receiver
• Interrupt generation
– on a transmitter buffer empty condition
– on a transmit last bit of a frame condition
– on a receiver buffer full condition
– on an error condition (frame, parity, overrun error)
– on the start and the end of a autobaud detection
Data Sheet
251
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
12.1.2
Overview
Figure 77 shows a block diagram of the ASC with its operating modes
(asynchronous and synchronous modes.).
Prescaler /
Fractional
Divider
fMOD
fDIV
Autobaud
Detection
fMOD
Asynchronous
Mode
Serial Port
Control
RXD
Mux
IrDA
Decoding
Baudrate
Timer
÷2
or
÷3
Receive / Transmit
Buffers and
Shift Registers
Baudrate
Timer
IrDA
Coding
Mux
TXD
Synchronous
Mode
Serial Port
Control
TXD
Shift Clock
Receive / Transmit
Buffers and
Shift Registers
RXDO
RXDI
Note: RXDI and RXDO are concatenated
in the port logic to pin RXD.
Figure 77
Data Sheet
Block Diagram of the ASC
252
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
12.1.3
Register Description
The ASC registers can be basically divided into four types of registers as shown in
Figure 78.
System Registers
S0CLC
Control Register
Data Registers
Interrupt Control
S0CON
S0TBUF
S0TIC
ABS0CON
S0RBUF
S0RIC
ABSTAT
S0EIC
S0TBIC
S0BG
S0FDV
S0PMW
S0CLC
Clock Control Register
S0ID
Identification Register
S0CON Control Register
ABS0CON Autobaud Control Register
ABSTAT Autobaud Status Register
S0TIC
ASC Transmit Interrupt Control Register
S0RIC
ASC Receive Interrupt Control Register
S0BG
S0FDV
S0PMW
S0TBUF
S0RBUF
S0EIC
S0TBIC
Figure 78
SFRs associated with ASC
Table 57
ASC Register Summary
Baudrate Timer Reload Register
Fractional Divider Register
IrDA Pulse Mode and Width Register
Transmit Buffer Register
Receive Buffer Register (read only)
ASC Error Interrupt Control Register
ASC Transmit Buffer Interrupt Control Register
Name
Address
Reset Value
Type 1)
Description
S0CLC
FFBAH
0000H
rw / r
ASC Clock Control Register
S0CON
FFB0H
0000H
rwh
Control Register
ABS0CON
FEF8H
0000H
rwh
Autobaud Control Register
ABSTAT
FEFEH
0000H
rwh
Autobaud Status Register
S0BG
FEB4H
0000H
rw
Baudrate Timer Reload Register
S0FDV
FEB6H
0000H
rw
Fractional Divider Register
S0PMW
FEAAH
0000H
rw
IrDA Pulse Mode and Width Register
S0TBUF
FEB0H
0000H
rw
Transmit Buffer Register
S0RBUF
FEB2H
0000H
r
Receive Buffer Register
Data Sheet
253
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
Table 57
ASC Register Summary
1)
Name
Address
Reset Value
Type
S0TIC
FF6CH
0000H
rw
Serial Channel 0 Transmit Interrupt
Control Register
S0RIC
FF6EH
0000H
rw
Serial Channel 0 Receive Interrupt
Control Register
S0EIC
FF70H
0000H
rw
Serial Channel 0 Error Interrupt
Control Register
S0TBIC
F19CH
0000H
rw
Serial Channel 0 Transmit Buffer IC
Register
1)
Description
r: read only; w: write only; rw: read- and writeable; rwh: like rw, but SFR/bit is also affected by hardware.
ASC Clock Control Register
S0CLC (FFBAH)
Reset Value: 0000H
15
14
13
12
11
10
9
8
7
6
5
4
0
0
0
0
0
0
0
0
0
0
0
0
3
2
1
0
EX SUS S0
S0
DISR PEN DISS DISR
rw
Bit
Function
S0DISR
ASC Disable Request Bit
S0DISR = ‘0’: ASC clock disable not requested
S0DISR = ‘1’: ASC clock disable requested
S0DISS
ASC Disable Status Bit
S0DISS = ‘0’:
ASC clock enabled
S0DISS = ‘1’:
ASC clock disabled
SUSPEN
Peripheral Suspend Enable Bit for OCDS
SUSPEN = ‘0’: Peripheral suspend disabled
SUSPEN = ‘1’: Peripheral suspend enabled
EXDISR
External Disable Request
EXRDIS = ‘0’: External clock disable Request is enabled
EXRDIS = ‘1’: External clock disable Request is disabled
rw
r
rw
The serial operating modes of the ASC module are controlled by its control register
S0CON. This register contains control bits for mode and error check selection, and
status flags for error identification.
Data Sheet
254
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
S0CON
Control Register
15
14
R
LB
13
12
11
BRS ODD FDE
10
9
8
7
6
OE
FE
PE
OEN
FEN
Field
Bits
Type Value Description
M
2-0
rwh
000
001
010
011
100
101
110
111
5
4
PEN/
REN
RXDI
3
STP
2
1
0
M
Mode Selection
8-bit data
synchronous operation
8-bit data
async. operation
IrDA mode, 8-bit data async. operation
7-bit data + parity
async. operation
9-bit data
async. operation
8-bit data + wake up bit async. operation
Reserved. Do not use this combination!
8-bit data + parity
async. operation
Bits are set/cleared by hardware after a
successfull autobaud detection operation.
Note: In synchronous operation (M=’000’), the
Fractional Divider is always disabled.
STP
3
rw
0
1
REN
4
rwh
0
1
PEN
RXDI
5
rw
Data Sheet
6
Receiver Enable Control
Receiver diabled
Receiver enabled
Bit can be affected during autobaud detection
operation when bit ABEN_AUREN is set.
Bit is reset by hardware after reception of byte in
synchronous mode.
0
1
Parity Check Enable /
IrDA Input Inverter Enable
All asynchronous modes without IrDA mode:
Ignore parity
Check parity
Only in IrDA mode (M=010):
RXD input is not inverted
RXD input is inverted
0
1
Framing Check Enable (async. modes only)
Ignore framing errors
Check framing errors
0
1
FEN
Number of Stop Bit Selection
One stop bit
Two stop bits
rw
255
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
Field
Bits
Type Value Description
OEN
7
rw
0
1
Overrun Check Enable
Ignore overrun errors
Check overrun errors
PE
8
rwh
Parity Error Flag
Set by hardware on a parity error (PEN=’1’).
Must be reset by software.
FE
9
rwh
Framing Error Flag
Set by hardware on a framing error (FEN=’1’).
Must be reset by software.
OE
10
rwh
Overrun Error Flag
Set by hardware on an overrun error (OEN=’1’).
Must be reset by software.
FDE
11
rw
0
1
ODD
12
rwh
0
1
BRS
13
rw
0
1
LB
14
rw
0
1
R
15
rw
0
1
Fractional Divider Enable
Fractional divider disabled
Fractional divider is enabled and used as
prescaler for baudrate timer (bit BRS is don’t
care)
Parity Selection
Even parity selected (parity bit set on odd
number of ‘1’s in data)
Odd parity selected (parity bit set on even
number of ‘1’s in data)
Bit is be set/cleared by hardware after a
successfull autobaud detection operation.
Baudrate Selection
Baudrate timer prescaler divide-by-2 selected
Baudrate timer prescaler divide-by-3 selected
BRS is don’t care if FDE=1 (fractional divider
enabled)
Loopback Mode Enable
Loopback mode disabled
Loopback mode enabled
Baudrate Generator Run Control
Baudrate generator disabled (ASC_P inactive)
Baudrate generator enabled
BG should only be written if R=’0’.
Note: Serial data transmission or reception is only possible when the run bit S0CON.R
is set to ‘1’. Otherwise the serial interface is idle.
Do not program the mode control field S0CON.M to one of the reserved
combinations to avoid unpredictable behaviour of the serial interface.
Data Sheet
256
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
The autobaud control register ABS0CON of the ASC module is used to control the
autobaud detection operation. It contains its general enable bit, the interrupt enable
control bits, and data path control bits.
ABS0CON
Autobaud Control Register
15
14
13
12
11
10
0
0
0
0
RX
INV
TX
INV
9
8
ABEM
7
6
5
4
0
0
0
FC
DET
EN
Field
Bits
Type Value Description
ABEN
0
rwh
0
1
AUREN
1
rw
0
1
ABSTEN
ABDETEN
FCDETEN
2
3
4
2
1
AB
ABST AUR
DET
EN
EN
EN
0
AB
EN
Autobaud Detection Enable
Autobaud detection is disabled
Autobaud detection is enabled
ABEN is reset by hardware after a successful
autobaud detection; (with the stop bit detection of
the second character). Resetting ABEN by
software if it was set aborts the autobaud
detection.
Automatic Autobaud Control of CON_REN
CON_REN is not affected during autobaud
detection
CON_REN is cleared (receiver disabled) when
ABEN and AUREN are set together. CON_REN
is set (receiver enabled) after a successful
autobaud detection (with the stop bit detection of
the second character).
0
1
Start of Autobaud Detection Interrupt Enable
Start of autobaud detection interrupt disabled
Start of autobaud detection interrupt enabled
0
1
Autobaud Detection Interrupt Enable
Autobaud detection interrupt disabled
Autobaud detection interrupt enabled
rw
rw
rw
0
1
Data Sheet
3
First Character of Two-Byte Frame Detected
Enable
Autobaud detection interrupt ABDETIR becomes
active after the two-byte frame recognition
Autobaud detection interrupt ABDETIR becomes
active after detection of the first and second byte
of the two-byte frame.
257
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
Field
Bits
Type Value Description
ABEM
8-9
rw
TXINV
10
RXINV
11
–
7-5,
15-12
00
01
10
11
Autobaud Echo Mode Enable
In echo mode the serial data at RXD is switched
to TXD output.
Echo mode disabled
Echo mode is enabled during autobaud detection
Echo mode is always enabled
reserved;
0
1
Transmit Inverter Enable
Transmit inverter disabled
Transmit inverter enabled
0
1
Receive Inverter Enable
Receive inverter disabled
Receive inverter enabled
all
reserved
rw
rw
0
The autobaud status register ABSTAT of the ASC module indicates the status of the
autobaud detection operation.
ABSTAT
Autobaud Status Register
15
14
13
12
11
10
9
8
7
6
5
0
0
0
0
0
0
0
0
0
0
0
Field
Bits
Type Value Description
FCSDET
0
rwh
0
1
FCCDET
1
rwh
0
1
Data Sheet
4
3
DET SCC
WAIT DET
2
1
0
SCS
DET
FCC
DET
FCS
DET
First Character with Small Letter Detected
no small ’a’ character detected
small ’a’ character detected
Bit is cleared by hardware when ABCON_ABEN
is set or if FCCDET or SCSDET or SCCDET is
set. Bit can be also cleared by software.
First Character with Capital Letter Detected
no capital ’A’ character detected
capital ’A’ character detected
Bit is cleared by hardware when ABCON_ABEN
is set or if FCSDET or SCSDET or SCCDET is
set. Bit can be also cleared by software.
258
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
Field
Bits
Type Value Description
SCSDET
2
rwh
0
1
SCCDET
3
rwh
0
1
DETWAIT
4
rwh
0
1
–
15-5
0
all
Second Character with Small Letter Detected
no small ’t’ character detected
small ’t’ character detected
Bit is cleared by hardware when ABCON_ABEN
is set or if FCSDET or FCCDET or SCCDET is
set. Bit can be also cleared by software.
Second Character with Capital Letter
Detected
no capital ’T’ character detected
capital ’T’ character detected
Bit is cleared by hardware when ABCON_ABEN
is set or if FCSDET or FCCDET or SCSDET is
set. Bit can be also cleared by software.
Autobaud Detection is Waiting
Either character ’a’, ’A’, ’t’, or ’T’ has been
detected.
The autobaud detection unit waits for the first ’a’
or ’A’
Bit is cleared when either FCSDET or FCCDET
is set (’a’ or ’A’ detected). Bit can be also cleared
by software. DETWAIT is set by hardware when
ABCON_ABEN is set.
reserved
Note: SCSDET or SCCDET are set when the second character has been recognized.
CON_ABEN is reset and ABDETIR set after SCSDET or SCCDET have seen set.
Data Sheet
259
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
The baudrate timer reload register S0BG of the ASC module contains the 13-bit reload
value for the baudrate timer in asynchronous and sychronous mode.
S0BG
Baudrate Timer/Reload Register
15
14
13
0
0
0
12
11
10
9
8
7
6
5
4
3
2
1
0
BR_VALUE
Field
Bits
Type Value Description
BR_VALUE
12-0
rw
all
Baudrate Timer/Reload Register Value
Reading BG returns the 13-bit content of the
baudrate timer (bits 15....13 return 0); writing BG
loads the baudrate timer reload register (bits
15....13 are don’t care). BG should only be
written if CON_R=’0’.
–
15-13
0
all
reserved
The fractional divider register S0FDV of the ASC module contains the 9-bit divider value
for the fractional divider (asynchronous mode only). It is also used for reference clock
generation of the autobaud detection unit.
S0FDV
Fractional Divider Register
15
14
13
12
11
10
9
0
0
0
0
0
0
0
8
7
6
5
4
3
2
1
0
FD_VALUE
Field
Bits
Type Value Description
FD_VALUE
8-0
rw
all
Fractional Divider Register Value
FDV contains the 9-bit value n of the fractional
divider which defines the fractional divider ratio:
n/512 n=0-511). With n=0, the fractional divider
is switched off (input=output frequency,
fDIV = fMOD, see Figure 87).
–
15-9
0
all
reserved
Data Sheet
260
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
The IrDA pulse mode and width register S0PMW of the ASC module contains the 8-bit
IrDA pulse width value and the IrDA pulse width mode select bit. This register is only
required in the IrDA operating mode.
S0PMW
IrDA Pulse Mode/Width Register
15
14
13
12
11
10
9
8
7
0
0
0
0
0
0
0
IRPW
6
Bits
Type Value Description
PW_VALUE
7-0
rw
IRPW
8
rw
–
Data Sheet
15-9
0
4
3
2
1
0
PW_VALUE
Field
all
5
IrDA Pulse Width Value
PW_VALUE is the 8-bit value n, which defines
the variable pulse width of an IrDA pulse.
Depending on the ASC_P input frequency fMOD,
this value can be used to adjust the IrDA pulse
width to value which is not equal 3/16 bit time
(e.g. 1.6 µs).
0
1
IrDA Pulse Width Mode Control
IrDA pulse width is 3/16 of the bit time
IrDA pulse width is defined by PW_VALUE
all
reserved
261
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
The transmitter buffer register S0TBUF of the ASC module contains the transmit data
value in asynchronous and synchronous modes.
S0TBUF
Transmitter Buffer Register
15
14
13
12
11
10
9
0
0
0
0
0
0
0
8
7
6
5
4
3
2
1
0
TD_VALUE
Field
Bits
Type Value Description
TD_VALUE
8-0
rw
all
Transmit Data Register Value
TBUF contains the data to be transmitted in
asynchronous and synchronous operating mode
of the ASC. Data transmission is double
buffered, Therefore, a new value can be written
to TBUF before the transmission of the previous
value is complete.
–
15-9
0
all
reserved
Data Sheet
262
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
The receiver buffer register S0RBUF of the ASC module contains the receive data value
in asynchronous and synchronous modes.
S0RBUF
Receive Buffer Register
15
14
13
12
11
10
9
0
0
0
0
0
0
0
8
7
6
5
4
3
2
1
0
RD_VALUE
Field
Bits
Type Value Description
RD_VALUE
8-0
rw
all
Receive Data Register Value
S0RBUF contains the reveived data bits and,
depending on the selected mode, the parity bit in
asynchronous and synchronous operating mode
of the ASC.
In asynchronous operating mode with M=011 (7bit data + parity) the received parity bit is written
into RD7.
In asynchronous operating mode with M=111 (8bit data + parity) the received parity bit is written
into RD8.
–
15-9
0
all
reserved
S0TIC
15
-
SFR
14
-
13
-
12
-
11
-
10
-
9
-
8
-
S0RIC
15
-
-
7
6
5
4
3
S0
TIR
S0
TIE
ILVL
GLVL
rw
rw
rw
rw
SFR
14
-
13
-
12
-
11
-
10
-
9
-
8
-
S0EIC
15
Reset Value: - - 00H
-
Data Sheet
13
-
12
-
11
-
10
-
9
-
8
-
1
0
Reset Value: - - 00H
7
6
5
4
3
S0
RIR
S0
RIE
ILVL
GLVL
rw
rw
rw
rw
SFR
14
2
2
1
0
Reset Value: - - 00H
7
6
S0
EIR
S0
EIE
ILVL
GLVL
rw
rw
rw
rw
263
5
4
3
2
1
0
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
S0TBIC
15
-
SFR
14
-
13
-
12
-
11
-
10
-
9
-
8
-
Reset Value: - - 00H
7
6
5
4
3
2
1
0
S0
EIR
S0
EIE
ILVL
GLVL
rw
rw
rw
rw
Note: Please refer to the general Interrupt Control Register description on page 109 for
an explanation of the control fields.
12.1.4
General Operation
The ASC supports full-duplex asynchronous communication up to 2.25 MBaud and halfduplex synchronous communication up to 4.5 MBaud (@ 36 MHz CPU clock which is
equal to the ASC module clock). In synchronous mode, data are transmitted or received
synchronous to a shift clock which is generated by the microcontroller. In asynchronous
mode, 8- or 9-bit data transfer, parity generation, and the number of stop bits can be
selected. Parity, framing, and overrun error detection is provided to increase the
reliability of data transfers. Transmission and reception of data is double-buffered. For
multiprocessor communication, a mechanism to distinguish address from data bytes is
included. Testing is supported by a loop-back option. A 13-bit baudrate timer with a
versatile input clock divider circuitry provides the ASC with the serial clock signal. In a
special asynchronous mode, the ASC supports IrDA data transmission up to 115.2
kBaud with fixed or programmable IrDA pulse width. A autobaud detection unit allows to
detect asynchronous data frames with its baudrate and mode with automatic initialization
of the baudrate generator and the mode controll bits.
A transmission is started by writing to the Transmit Buffer register S0TBUF. Only the
number of data bits which is determined by the selected operating mode will actually be
transmitted, ie. bits written to positions 9 through 15 of register S0TBUF are always
insignificant.
Data transmission is double-buffered, so a new character may be written to the transmit
buffer register, before the transmission of the previous character is complete. This allows
the transmission of characters back-to-back without gaps.
Data reception is enabled by the Receiver Enable Bit CON_REN. After reception of a
character has been completed, the received data and, if provided by the selected
operating mode, the received parity bit can be read from the (read-only) Receive Buffer
register S0RBUF. Bits in the upper half of S0RBUF which are not valid in the selected
operating mode will be read as zeros.
Data reception is double-buffered, so that reception of a second character may already
begin before the previously received character has been read out of the receive buffer
register. In all modes, receive buffer overrun error detection can be selected through bit
Data Sheet
264
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
CON_OEN. When enabled, the overrun error status flag CON_OE and the error interrupt
request line EIR will be acitvated when the receive buffer register has not been read by
the time reception of a second character is complete. The previously received character
in the receive buffer is overwritten.
The Loop-Back option (selected by bit CON_LB) allows the data currently being
transmitted to be received simultaneously in the receive buffer. This may be used to test
serial communication routines at an early stage without having to provide an external
network. In loop-back mode the alternate input/output function of port pins is not
required.
Note: Serial data transmission or reception is only possible when the Baudrate
Generator Run Bit CON_R is set to ‘1’. Otherwise the serial interface is idle.
Do not program the mode control field COM_M to one of the reserved
combinations to avoid unpredictable behaviour of the serial interface
12.1.5
Asynchronous Operation
Asynchronous mode supports full-duplex communication, where both transmitter and
receiver use the same data frame format and the same baudrate. Data is transmitted on
pin P3.10/TXD and received on pin P3.11/RXD. IrDA data transmission/reception is
supported up to 115.2 KBit/s. Figure 79 shows the block diagram of the ASC when
operating in asynchronous mode.
Data Sheet
265
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
FDE BRS
13-Bit Reload Register
Fractional
Divider
fMOD
MUX
÷2
fDIV
÷3
R
13-Bit Baudrate Timer
fBRT
ABSTIRI
Autobaud
Detection
M
REN
FEN
PEN
OEN
LB
Sampling
Mux
IrDA
Decoding
fBR
÷16
ABDETIR
STP
ODD
Autobaud
Start Int.
Autobaud
Detect Int.
PE
FE
OE
RIR
Shift Clock
TIR
Serial Port Control
Shift Clock
TBIR
EIR
Receive Int.
Request
Transmit Int.
Request
Transmit Buffer
Int. Request
Error Int.
Request
Receive Shift
Register
Transmit Shift
Register
IrDA
Coding
Receive Buffer Reg.
RBUF
Transmit Buffer Reg.
TBUF
Mux
Mux
Internal Bus
RXD
Mux
1
Figure 79
1
TXD
Asynchronous Mode of Serial Channel ASC
12.1.5.1 Asynchronous Data Frames
8-Bit Data Frames
8-bit data frames either consist of 8 data bits D7...D0 (CON_M=’001B’), or of 7 data bits
D6...D0 plus an automatically generated parity bit (CON_M=’011B’). Parity may be odd
Data Sheet
266
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
or even, depending on bit CON_ODD. An even parity bit will be set, if the modulo-2-sum
of the 7 data bits is ‘1’. An odd parity bit will be cleared in this case. Parity checking is
enabled via bit CON_PEN (always OFF in 8-bit data mode). The parity error flag
CON_PE will be set along with the error interrupt request flag, if a wrong parity bit is
received. The parity bit itself will be stored in bit RBUF.7.
10-/11-Bit UART Frame
8 Data Bits
CON_M=001B
Start
Bit
0
D0 D1
LSB
D2
D3
D4
1
D5 D6
D7
1
(1st) (2nd)
Stop Stop
Bit
Bit
10-/11-Bit UART Frame
7 Data Bits
CON_M=011B
Figure 80
Start
Bit
0
D0 D1
LSB
D2
D3
D4
1
D5
D6 Parit
1
(1st) (2nd)
Stop Stop
Bit
Bit
Asynchronous 8-Bit Frames
9-Bit Data Frames
9-bit data frames either consist of 9 data bits D8...D0 (CON_M=’100B’), of 8 data bits
D7...D0 plus an automatically generated parity bit (CON_M=’111B’) or of 8 data bits
D7...D0 plus wake-up bit (CON_M=’101B’). Parity may be odd or even, depending on bit
CON_ODD. An even parity bit will be set, if the modulo-2-sum of the 8 data bits is ‘1’. An
odd parity bit will be cleared in this case. Parity checking is enabled via bit CON_PEN
(always OFF in 9-bit data and wake-up mode). The parity error flag CON_PE will be set
along with the error interrupt request flag, if a wrong parity bit is received. The parity bit
itself will be stored in bit RBUF.8.
Data Sheet
267
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
11-/12-Bit UART Frame
9 Data Bits
Start
Bit
0
D0 D1
LSB
D2
D3
D4
1
D5 D6
D7
Bit 9
1
(1st) (2nd)
Stop Stop
Bit
Bit
CON_M=100B : Bit 9 = Data Bit D8
CON_M=101B : Bit 9 = Wake-up Bit
CON_M=111B : Bit 9 = Parity Bit
Figure 81
Asynchronous 9-Bit Frames
In wake-up mode received frames are only transferred to the receive buffer register, if
the 9th bit (the wake-up bit) is ‘1’. If this bit is ‘0’, no receive interrupt request will be
activated and no data will be transferred.
This feature may be used to control communication in multi-processor system:
When the master processor wants to transmit a block of data to one of several slaves, it
first sends out an address byte which identifies the target slave. An address byte differs
from a data byte in that the additional 9th bit is a '1' for an address byte and a '0' for a
data byte, so no slave will be interrupted by a data 'byte'. An address 'byte' will interrupt
all slaves (operating in 8-bit data + wake-up bit mode), so each slave can examine the 8
LSBs of the received character (the address). The addressed slave will switch to 9-bit
data mode (eg. by clearing bit CON_M.0), which enables it to also receive the data bytes
that will be coming (having the wake-up bit cleared). The slaves that were not being
addressed remain in 8-bit data + wake-up bit mode, ignoring the following data bytes
IrDA Frames
The modulation schemes of IrDA is based on standard asynchronous data transmission
frames. The asynchronous data format in IrDA mode (CON_M=010B) is defined as
follows :
1 start bit / 8 data bits / 1 stop bit
The coding/decoding of/to the asynchronous data frames is shown in Figure 82. In
general, during the IrDA transmissions UART frames are encoded into IR frames and
vice cersa. A low level on the IR frame indicates a “LED off“ state. A high level on the IR
frame indicates a “LED on“ state.
Data Sheet
268
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
For a “0“ bit in the UART frame a high pulse is generated. For a “1“ bit in the UART frame
no pulse is generated. The high pulse starts in the middle of a bit cell and has a fixed
width of 3/16 of the bit time. The ASC also allows to program the length of the IrDA high
pulse. Further, the polarity of the received IrDA pulse cane be inverted in IrAD mode.
Figure 82 shows the non-inverted IrDA pulse scheme.
UART Frame
Start
Bit
Stop
Bit
8 Data Bits
0
1
0
1
0
0
1
1
0
1
IR Frame
Start
Bit
0
Bit
Time
Figure 82
Stop
Bit
8 Data Bits
1
0
1
0
0
1
1/2 Bit Time
1
0
1
Pulse Width =
3/16 Bit Time
(or variable length)
IrDA Frame Encoding/Decoding
12.1.5.2 Asynchronous Transmission
Asynchronous transmission begins at the next overflow of the divide-by-16 baudrate
timer (transition of the baudrate clock fBR), if bit S0CON.R is set and data has been
loaded into S0TBUF. The transmitted data frame consists of three basic elements:
– the start bit
– the data field (8 or 9 bits, LSB first, including a parity bit, if selected)
– the delimiter (1 or 2 stop bits)
Data transmission is double buffered. When the transmitter is idle, the transmit data
loaded into register S0TBUF is immediately moved to the transmit shift register thus
freeing S0TBUF for the next data to be sent. This is indicated by the transmit buffer
interrupt request line TBIR being activated. S0TBUF may now be loaded with the next
data, while transmission of the previous one is still going on.
The transmit interrupt request line TIR will be activated before the last bit of a frame is
transmitted, ie. before the first or the second stop bit is shifted out of the transmit shift
register.
The transmitter output pin P3.10/TXD must be configured for alternate data output’.
Data Sheet
269
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
12.1.5.3 Asynchronous Reception
Asynchronous reception is initiated by a falling edge (1-to-0 transition) on pin P3.11/
RXD, provided that bits CON_R and CON_REN are set. The receive data input pin
P3.11/RXD is sampled at 16 times the rate of the selected baudrate. A majority decision
of the 7th, 8th and 9th sample determines the effective bit value. This avoids erroneous
results that may be caused by noise.
If the detected value is not a '0' when the start bit is sampled, the receive circuit is reset
and waits for the next 1-to-0 transition at pin P3.11/RXD. If the start bit proves valid, the
receive circuit continues sampling and shifts the incoming data frame into the receive
shift register.
When the last stop bit has been received, the content of the receive shift register is
transferred to the receive data buffer register S0RBUF. Simultaneously, the receive
interrupt request line RIR is activated after the 9th sample in the last stop bit time slot (as
programmed), regardless whether valid stop bits have been received or not. The receive
circuit then waits for the next start bit (1-to-0 transition) at the receive data input pin.
The receiver input pin P3.11/RXD must be configured for input.
Asynchronous reception is stopped by clearing bit CON_REN. A currently received
frame is completed including the generation of the receive interrupt request and an error
interrupt request, if appropriate. Start bits that follow this frame will not be recognized.
Note: In wake-up mode received frames are only transferred to the receive buffer
register, if the 9th bit (the wake-up bit) is ‘1’. If this bit is ‘0’, no receive interrupt
request will be activated and no data will be transferred.
12.1.5.4 IrDA Mode
The duration of the IrDA pulse is normally 3/16 of a bit period. The IrDA standard also
allows the pulse duration being independent of the baudrate or bit period. In this case
the transmitted pulse has always the width corresponding to the 3/16 pulse width at
115.2 kBaud which is 1.627 µs. Both, bit period dependend or fixed IrDA pulse width
generation can be selected. The IrDA pulse width mode is selected by bit PMW_IRPW.
In case of fixed IrDA pulse width generation, the lower 8 bits in register PMW are used
to adapt the IrDA pulse width to a fixed value of e.g. 1.627 µs. The fixed IrDA pulse width
is generated by a programmable timer as shown in Figure 83.
Data Sheet
270
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
PMW
Start Timer
tIPW
fMOD
Figure 83
IrDA Pulse
8-Bit Timer
Fixed IrDA Pulse Generation
The IrDA pulse width can be calculated according the formulas given in Table 58.
Table 58
Formulas for the IrDA Pulse Width Calculation
PMW
PMW_IRPW
1 ... 255
0
1
Formulas
t IPW =
t IPW =
3
16 x Baudrate
PMW
t IPW min =
(PMW >> 1)
fMOD
fMOD
The name PMW in the formulas of Table 58 represents the content of the reload register
PMW (PW_VALUE), taken as unsigned 8-bit integer.
The content of PMW further defines the minimum IrDA pulse width (tIPW min) which is still
recognized during a receive operation as a valid IrDA pulse. This function is independent
of the selected IrDA pulse width mode (fixed or variable) which is defined by bit
PMW_IRPW. The minimum IrDA pulse width is calculated by a shift right operation of
PMW bit 7-0 by one bit divided by the module clock fMOD.
Note: If PMW_IRPW=0 (fixed IrDA pulse width), PW_VALUE must be a value which
assures that t IPW > t IPW min.
Table 59 gives two examples for typical frequencies of the C165UTAH: 36 MHz and 24
MHz..
Table 59
IrDA Pulse Width Adaption to 1.627 µs
fMOD
PMW
tIPW
Error
tIPW min
24 MHz
39
1.625 µs
- 0.12 %
0.79 µs
36 MHz
59
1.639 µs
+ 0.74 %
0.81 µs
Data Sheet
271
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
12.1.5.5 RXD/TXD Data Path Selection in Asynchronous Modes
The data paths for the serial input and output data of the ASC in asynchronous modes
are affected by several control bits in the registers CON and ABCON as shown in
Figure 84. The synchronous mode operation is not affected by these data path selection
capabilities.
The input signal from RXD passes an inverter which is controlled by bit ABCON_RXINV.
The output signal of this inverter is used for the autobaud detection and may bypass the
ASC logic in the echo mode (controlled by bit ABCON_ABEM). Further, two multiplexers
are in the RXD input signal path for providing the loopback mode capability (controlled
by bit CON_LB) and the IrDA receive pulse inversion capability (controlled by bit
CON_RXDI).
Depending on the asynchronous operating mode (controlled by bitfield CON_M), the
ASC output signal or the RXD input signal in echo mode (controlled by bit
ABCON_ABEM) is switched to the TXD output via an inverter (controlled by bit
ABCON_TXINV).
Autobaud
Detection
ABCON
TXINV
ABEM
IrDA
Coding
CON
LB
Figure 84
RXDI
Mux
ASC
Asynch. Mode Logic
Mux
IrDA
Decode
Mux
Mux
Mux
RXD
Mux
RXINV
TXD
M
RXD/TXD Data Path in Asynchronous Modes (ASC)
Note: In echo mode the transmit output signal of the ASC logic is blocked by the echo
mode output multiplexer. Figure 84 also shows that it is not possible to use an
IrDA coded receiver input signal for autobaud detection.
Data Sheet
272
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
12.1.6
Synchronous Operation
Synchronous mode supports half-duplex communication, basically for simple I/O
expansion via shift registers. Data is transmitted and received via pin RXD while pin TXD
outputs the shift clock. These signals are alternate functions of port pins P3.11 and
P3.10. Synchronous mode is selected with CON_M=’000B’.
Eight data bits are transmitted or received synchronous to a shift clock generated by the
internal baudrate generator. The shift clock is only active as long as data bits are
transmitted or received.
Note: The lines RXDI and RXDO are concatenated in the port logic to pin RXD.
13-Bit Reload Register
fMOD
÷2
Mux
fDIV
fBR
fBRT
÷3
R
÷4
13-Bit Baudrate Timer
BRS
M=000B
OE
RIR
Shift Clock
REN
OEN
LB
TIR
Serial Port Control
Shift Clock
TXD
RXDI
0
MUX
1
TBIR
EIR
Receive Shift
Register
Transmit Shift
Register
Receive Buffer Reg.
RBUF
Transmit Buffer Reg.
TBUF
Receive Int.
Request
Transmit Int.
Request
Transmit Buffer
Int. Request
Error Int.
Request
RXDO
Note: RXDI and RXDO are
concatenated in the
port logic to pin RXD.
Internal Bus
Figure 85
Data Sheet
Synchronous Mode of Serial Channel ASC_P
273
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
12.1.6.1 Synchronous Transmission
Synchronous transmission begins within 4 state times after data has been loaded into
S0TBUF provided that CON_R is set and CON_REN=’0’ (half-duplex, no reception).
Exception : in loopback mode (bit CON_LB set), CON_REN must be set for reception of
the transmitted byte. Data transmission is double buffered. When the transmitter is idle,
the transmit data loaded into S0TBUF is immediately moved to the transmit shift register
thus freeing S0TBUF for the next data to be sent. This is indicated by the transmit buffer
interrupt request line TBIR being activated. S0TBUF may now be loaded with the next
data, while transmission of the previous one is still going on. The data bits are
transmitted synchronous with the shift clock. After the bit time for the 8th data bit, both
TXD and RXD will go high, the transmit interrupt request line TIR is activated, and serial
data transmission stops.
Pin P3.10/TXD must be configured for alternate data output in order to provide the shift
clock. Pin P3.11/RXD must also be configured for output during transmission.
12.1.6.2 Synchronous Reception
Synchronous reception is initiated by setting bit CON_REN=’1’. If bit CON_R=1, the data
applied at RXD is clocked into the receive shift register synchronous to the clock which
is output at pin TXD. After the 8th bit has been shifted in, the content of the receive shift
register is transferred to the receive data buffer RBUF, the receive interrupt request line
RIR is activated, the receiver enable bit CON_REN is reset, and serial data reception
stops.
Pin P3.10/TXD must be configured for alternate data output in order to provide the shift
clock. Pin P3.11/RXD must be configured as alternate data input.
Synchronous reception is stopped by clearing bit CON_REN. A currently received byte
is completed including the generation of the receive interrupt request and an error
interrupt request, if appropriate. Writing to the transmit buffer register while a reception
is in progress has no effect on reception and will not start a transmission.
If a previously received byte has not been read out of the receive buffer register at the
time the reception of the next byte is complete, both the error interrupt request line EIR
and the overrun error status flag CON_OE will be activated/set, provided the overrun
check has been enabled by bit CON_OEN.
12.1.6.3 Synchronous Timing
Figure 86 shows timing diagrams of the ASC synchronous mode data reception and
data transmission. In idle state the shift clock is at high level. With the beginning of a
synchronous transmission of a data byte the data is shifted out at RXD with the falling
edge of the shift clock. If a data byte is received through RXD data is latched with the
rising edge of the shift clock.
Data Sheet
274
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
Between two consecutive receive or transmit data bytes one shift clock cycle (fBR) delay
is inserted.
Shift
Receive/Transmit Timing
Shift
Shift
Latch
Latch
Shift Clock
Transmit Data
Data
Bit n
Data
Bit n+1
Data
Bit n+2
Receive Data
Valid
Data n
Valid
Data n+1
Valid
Data n+2
Continuous Transmit Timing
Shift Clock
Transmit Data
D0
D1
D2
D3
D4
D5
D6
D7
D0
D1
1. Byte
Receive Data
D0
D1
D2
D3
D4
12.1.7
D3
2. Byte
D5
D6
1. Byte
Figure 86
D2
D7
D0
D1
D2
D3
2. Byte
ASC_P3 Synchronous Mode Waveforms
Baudrate Generation
The serial channel ASC has its own dedicated 13-bit baudrate generator with 13-bit
reload capability, allowing baudrate generation independent of the GPT timers.
The baudrate generator is clocked with a clock (fDIV) which is derived via a prescaler from
the ASC input clock fMOD, e.g. 36 MHz. The baudrate timer is counting downwards and
can be started or stopped through the baudrate generator run bit CON_R. Each
underflow of the timer provides one clock pulse to the serial channel. The timer is
reloaded with the value stored in its 13-bit reload register each time it underflows. The
resulting clock fBRT is again divided by a factor for the baudrate clock (± 16 in
asynchronous modes and ± 4 in synchronous mode). The prescaler is selected by the
bits CON_BRS and CON_FDE. In the asynchronous operating modes, additionally to
Data Sheet
275
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
the two fixed dividers a fractional divider prescaler unit is available which allows to select
prescaler divider ratios of n/512 with n=0-511. Therefore, the baudrate of ASC is
determined by the module clock, the content of S0FDV, the reload value of S0BG and
the operating mode (asynchronous or synchronous).
Register S0BG is the dual-function Baudrate Generator/Reload register. Reading BG
returns the content of the timer BR_VALUE (bits 15...13 return zero), while writing to
S0BG always updates the reload register (bits 15...13 are insiginificant).
An auto-reload of the timer with the content of the reload register is performed each time
CON_BG is written to. However, if CON_R=’0’ at the time the write operation to BG is
performed, the timer will not be reloaded until the first instruction cycle after CON_R=’1’.
For a clean baudrate initialization S0BG should only be written if CON_R=’0’. If S0BG is
written with CON_R=’1’, an unpredicted behaviour of the ASC may occur during running
transmit or receive operations.
12.1.7.1 Baudrates in Asynchronous Mode
For asynchronous operation, the baudrate generator provides a clock fBRT with 16 times
the rate of the established baudrate. Every received bit is sampled at the 7th, 8th and 9th
cycle of this clock. The clock divider circuitry, which generates the input clock for the 13bit baudrate timer, is extended by a fracxtional divider circuitry, which allows the
adjustment of more accurate baudrates and the extension of the baudrate range.
The baudrate of the baudrate generator depends on the following input clock, bits and
register values :
–
–
–
–
Input clock fMOD
Selection of the baudrate timer input clock fDIV by bits CON_FDE and CON_BRS
If bit CON_FDE=1 (fractional divider) : value of register CON_FDV
value of the 13-bit reload register S0BG
The output clock of the baudrate timer with the reload register is the sample clock in the
asynchronous modes of the ASC. For baudrate calculations, this baudrate clock fBR is
derived from the sample clock fDIV by a division by 16.
Data Sheet
276
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
13-Bit Reload Register
FDE
Fractional
Divider
fMOD
÷16
Mux
÷2
R
13-Bit Baudrate Timer
Sample
Clock
fBRT
÷3
BRS
Figure 87
fDIV
fBR Baud
Rate
Clock
FDE
BRS
Selected Divider
0
0
÷2
0
1
÷3
1
X
Fractional Divider
ASC Baudrate Generator Circuitry in Asynchronous Modes
Using the fixed Input Clock Divider
The baudrate for asynchronous operation of serial channel ASC when using the fixed
input clock divider ratios (CON_FDE=0) and the required reload value for a given
baudrate can be determined by the following formulas :
Table 60
Asynchronous Baudrate Formulas using the Fixed Input Clock
Dividers
FDE
BRS
BG
0
0
0 ... 8191
Formula
Baudrate =
BG =
1
Baudrate =
BG =
Data Sheet
277
fMOD
32 x (BG+1)
fMOD
32 x Baudrate
-1
fMOD
48 x (BG+1)
fMOD
-1
48 x Baudrate
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
BG represents the contents of the reload register S0BG (BR_VALUE), taken as
unsigned 13-bit integer.
The maximum baudrate that can be achieved for the asynchronous modes when using
the two fixed clock dividers and a module clock of 36 MHz is 1.125 MBaud. Table 61
below lists various commonly used baudrates together with the required reload values
and the deviation errors compared to the intended baudrate.
Table 61
Typical Asynchronous Baudrates using the Fixed Input Clock
Dividers
Baudrate
BRS = ‘0’, fMOD = 36 MHz
BRS = ‘1’, fMOD = 36 MHz
Deviation Error
Reload Value
Deviation Error
Reload Value
1.125
MBaud
---
0000H
---
---
750.0
kBaud
---
---
---
0000H
19.2
kBaud
- 0.69 %
003AH
+ 0.16 %
0026H
9600
Baud
+ 0.16 %
0074H
+ 0.16 %
004DH
4800
Baud
+ 0.16 %
00E9H
+ 0.16 %
009BH
2400
Baud
+ 0.16 %
01D3H
+ 0.16 %
0137H
1200
Baud
+ 0.05 %
03A8H
+/- 0.0 %
0270H
Note: CON_FDE must be 0 to achieve the baudrates in the table above. The deviation
errors given in the table above are rounded. Using a baudrate crystal will provide
correct baudrates without deviation errors.
Using the Fractional Divider
When the fractional divider is selected, the input clock fDIV for the baudrate timer is
derived from the module clock fMOD by a programmable divider. If CON_FDE=1, the
fractional divider is activated, It divides fMOD by a fraction of n/512 for any value of n from
0 to 511. If n=0, the divider ratio is 1 which means that fDIV=fMOD. In general, the fractional
divider allows to program the baudrrate with a much better accuracy than with the two
fixed prescaler divider stages.
Table 62
Asynchronous Baudrate Formulas using the Fractional Input Clock
Divider
FDE
BRS
BG
FDV
Formula
1
X
1 ... 8191
1 ... 511
0
Data Sheet
Baudrate =
Baudrate =
278
FDV
512
x
fMOD
16 x (BG+1)
fMOD
16 x (BG+1)
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
BG represents the content of the reload register S0BG (BR_VALUE), taken as unsigned
13-bit integer. FDV represents the content of the fractional divider register S0FDV
(FD_VALUE) taken as unsigned 9-bit integer. For example, typical asynchronous
baudrates are shown in Table 63.
Using the fractional divider and a module clock of 36 MHz (equal to the C165UTAH CPU
clock) the available baudrate range is 2.25 MBaud down to 0.5364 Baud.
.
Table 63
Typical Asynchronous Baudrates using the Fractional Input Clock
Divider
fMOD
Desired
Baudrate
BG
FDV
Resulting
Baudrate
Deviation
36 MHz
max. Baudrate
0
0
2.25
0%
230.4
kBaud
6
367
230.399 kBaud
< 0.01 %
115.2
kBaud
13
367
115.199 kBaud
< 0.01 %
57.6
kBaud
27
367
57.5997 kBaud
< 0.01 %
38.4
kBaud
41
367
38.3998 kBaud
< 0.01 %
19.2
kBaud
83
367
19.1999 kBaud
< 0.01 %
8191
1
0.53644 Baud
0%
min. Baudrate
MBaud
Note: The ApNote AP2423 provides a program ’ASC.EXE’ which allows to calculate
values for the S0FDV and S0BG registers depending on fMOD, the requested
baudrate, and the maximum deviation. Please contact your Infineon Technologies
representative.
12.1.7.2 Baudrates in Synchronous Mode
For synchronous operation, the baudrate generator provides a clock with 4 times the rate
of the established baudrate.(see Figure 88).
Data Sheet
279
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
13-Bit Reload Register
÷2
fMOD
fDIV
Mux
13-Bit Baudrate Timer
÷3
fBRT
Shift /
Sample
Clock
÷4
R
BRS
Figure 88
BRS
Selected Divider
0
÷2
1
÷3
ASC Baudrate Generator Circuitry in Synchronous Mode
The baudrate for synchronous operation of serial channel ASC can be determined by the
formulas as shown in Table 64.
Table 64
Synchronous Baudrate Formulas
BRS
BG
0
0 ... 8191
Formula
Baudrate =
1
Baudrate =
fMOD
8 x (BG+1)
BG =
fMOD
8 x Baudrate
-1
fMOD
12 x (BG+1)
BG =
fMOD
12 x Baudrate
-1
BG represents the content of the reload register S0BR (BR_VALUE), taken as unsigned
13-bit integers.
The maximum baudrate that can be achieved in synchronous mode when using a
module clock of 36 MHz is 4.5 MBaud.
Data Sheet
280
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
12.1.8
Autobaud Detection
12.1.8.1 General Operation
The autobaud detection unit in the ASC provides a capability to recognize the mode and
the baudrate of an asynchronous input signal at RXD. Generally, the baudrates to be
recognized must be known by the application. With this knowledge always a set of nine
baudrates can be detected. The autobaud detection unit is not designed to calculate a
baudrate of an unknown asynchronous frame.
Figure 89 shows how the autobaud detection unit of the ASC is integrated into its
asynchronous mode configuration. The RXD data line is an input to the autobaud
detection unit. The clock fDIV which is generated by the fractional divider is used by the
autobaud detection unit as time base. After successful recognition of baudrate and
asynchronous operating mode of the RXD data input signal, bits in the S0CON register
and the value of the S0BG register in the baudrate timer are set to the appropriate
values, and the ASC can start immediately with the reception of serial input data.
fMOD
Prescaler /
Fractional
Divider
fDIV
Autobaud
Detection
RXD
Figure 89
IrDA
Decoding
Baudrate
Timer
Asynchronous
Mode
Serial Port
Control
Mux
Receive / Transmit
Buffers and
Shift Registers
IrDA
Coding
Mux
TXD
ASC Asynchronous Mode Block Diagram
The following sequence must be generally executed to start the autobaud detection unit
for operation :
–
–
–
–
–
Definition of the baudrates to be detected : standard or non-standard baudrates
Programming of the Prescaler/Fractional Divider to select a specific value of fDIV
Starting the Prescaler/Fractional Divider (setting CON_R)
Preparing the interrupt system of the CPU
Enabling the autobaud detection (setting ABCON_ EN and the interrupt enable bits in ABCON
for interrupt generation, if required)
– Polling interrupt request flag or waiting for the autobaud detection interrupt
Data Sheet
281
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
12.1.8.2 Serial Frames for Autobaud Detection
The autobaud detection of the ASC is based on the serial reception of a specific two-byte
serial frame. This serial frame is build up by the two ASCII bytes "at" or "AT" ("aT" or "At"
are not allowed). Both byte combinations can be detected in five types of asynchronous
frames. Figure 90 and Figure 91 show the serial frames which are detected at least.
7 bit, even parity
’a’ = 61H
1
0
0
0
0
’t’ = 74H
1
1
start
1
1
0
parity stop
0
1
0
1
1
1
0
1
parity stop
start
7 bit, odd parity
’a’ = 61H
1
0
0
0
0
’t’ = 74H
1
1
start
0
1
0
parity stop
0
1
0
1
1
1
1
1
parity stop
start
8 bit, no parity
’a’ = 61H
1
0
0
0
0
’t’ = 74H
1
1
0
start
1
0
stop
0
1
0
1
1
1
0
start
1
stop
8 bit, even parity
’a’ = 61H
1
0
0
0
0
’t’ = 74H
1
1
0
start
1
1
parity stop
0
0
1
0
1
1
1
0
start
0
1
parity stop
8 bit, odd parity
’a’ = 61H
1
start
Figure 90
Data Sheet
0
0
0
0
’t’ = 74H
1
1
0
0
1
parity stop
0
0
start
1
0
1
1
1
0
1
1
parity stop
Two-Byte Serial Frames with ASCII ’at’
282
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
7 bit, even parity
’A’ = 41H
1
0
0
0
0
’T’ = 54H
0
1
start
0
1
0
parity stop
0
1
0
1
0
1
1
1
parity stop
start
7 bit, odd parity
’A’ = 41H
1
0
0
0
0
’T’ = 54H
0
1
start
1
1
0
parity stop
0
1
0
1
0
1
0
1
parity stop
start
8 bit, no parity
’A’ = 41H
1
0
0
0
0
’T’ = 54H
0
1
0
start
1
0
stop
0
1
0
1
0
1
0
start
1
stop
8 bit, even parity
’A’ = 41H
1
0
0
0
0
’T’ = 54H
0
1
0
start
0
1
parity stop
0
0
1
0
1
0
1
0
start
1
1
parity stop
8 bit, odd parity
’A’ = 41H
1
start
Figure 91
Data Sheet
0
0
0
0
’T’ = 54H
0
1
0
1
1
parity stop
0
0
start
1
0
1
0
1
0
0
1
parity stop
Two-Byte Serial Frames with ASCII ’AT’
283
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
12.1.8.3 Baudrate Selection and Calculation
The autobaud detection requires some calculations concerning the programming of the
baudrate generator and the baudrates to be detected. Two steps must be considered :
– Defining the baudrate(s) to be detected
– Programming of the baudrate timer prescaler - setup of the clock rate of fDIV
In general, the baudrate generator of the ASC in asynchronous mode is build up by two
parts (see also Figure 87) :
– the clock prescaler part which derives fDIV from fMOD
– the baudrate timer part which generates the sample clock fBRT and the baudrate clock fBR
Prior to an autobaud detection the prescaler part has to be setup by the CPU while the
baudrate timer (register BG) is initialized with a 13-bit value (BR_VALUE) automatically
after a successfull autobaud detection. For the following calculations, the fractional
divider is used (CON_FDE = 1).
Note: It is also possible to use the fixed divide-by-2 or divide-by-3 prescaler. But the
fractional divider allows to adapt fDIV much more precise to the required value.
Standard Baudrates
For standard baudrate detection the baudrates as shown in Table 65 can be e.g.
detected. Therefore, the output frequency fDIV of the ASC baudrate generator must be
set to a frequency derived from the module clock fMOD in a way that it is equal to 11.0592
MHz. The value to be written into register FDV is the nearest integer value which is
calculated according the following formula :
FDV =
512 x 11.0592 MHz
fMOD
Table 65 defines the nine standard baudrates (Br0 - Br8) which can be detected for
fDIV=11.0592 MHz.
Table 65
Autobaud Detection using Standard Baudrates (fDIV = 11.0592 MHz)
Baudrate
Numbering
Detectable Standard
Baudrate
Divide Factor d f
BG is loaded after
detection with value
Br0
230.400 kBaud
48
2
= 002H
Br1
115.200 kBaud
96
5
= 005H
Br2
57.600 kBaud
192
11
= 00BH
Br3
38.400 kBaud
288
17
= 011H
Br4
19.200 kBaud
576
35
= 023H
Data Sheet
284
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
Table 65
Autobaud Detection using Standard Baudrates (fDIV = 11.0592 MHz)
Baudrate
Numbering
Detectable Standard
Baudrate
Divide Factor d f
BG is loaded after
detection with value
Br5
9600 Baud
1152
71
= 047H
Br6
4800 Baud
2304
143
= 08FH
Br7
2400 Baud
4608
287
= 11FH
Br8
1200 Baud
9216
575
= 23FH
According Table 65 a baudrate of 9600 Baud is achieved when register BG is loaded
with a value of 047H, assuming that fDIV has been set to 11.0592 MHz.
Table 65 also lists a divide factor d f which is defined with the following formula :
Baudrate =
f DIV
df
This divide factor df defines a fixed relationship between the prescaler output frequency
fDIV and the baudrate to be detected during the autobaud detection operation. This
means, changing fDIV results in a totally different baudrate table in means of baudrate
values. For the baudrates to be detected, the following relations are always valid :
– Br0 = fDIV / 48D, Br1 = fDIV / 96D, ....... up to Br8 = fDIV / 9216D,
A requirement for detecting standard baudrates up to 230.400 kBaud is the fDIV minimum
value of 11.0592 MHz. With the value FD_VALUE in register FDV the fractional divider
fDIV is adapted to the module clock frequency fMOD. Table 66 defines the deviation of the
standard baudrates when using autobaud detection depending on the module clock
fMOD.
Table 66
Standard Baudrates - Deviations and Errors for Autobaud Detection
fMOD
FDV
10 MHz
not possible
12 MHz
472
+ 0.03 %
13 MHz
436
+ 0.1 %
16 MHz
354
+ 0.03 %
18 MHz
315
+ 0.14 %
18.432 MHz
307
- 0.07 %
20 MHz
283
- 0.04 %
24 MHz
236
+ 0.03 %
25 MHz
226
- 0.22 %
30 MHz
189
+ 0.14 %
33 MHz
172
+ 0.24 %
36 MHz
157
+ 0.18 %
Data Sheet
Error in fDIV
285
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
Note: If the deviation of the baudrate after autobaud detection is to high, the baudrate
generator (fractional divider FDV and reload register BG) can be reprogrammed if
required to get a more precise baudrate with less error.
Non-Standard Baudrates
Due to the relationship between Br0 to Br8 in Table 65 concerning the divide factor df
other baudrates than the standard baudrates can be also selected. E.g. if a baudrate of
50 kBaud has to be detected, Br2 is e.g. defined as baudrate for the 50 kBaud setection.
This further results in :
– fDIV = 50 kBaud x df@Br2 = 50 kBaud x 192 = 9.6 MHz
Therefore, depending on the module clock frequency fMOD, the value of the fractional
divider (register S0FDV must be set in this example according the formula :
FDV =
512 x fDIV
fMOD
with fDIV = 9.6 MHz
Using this selection (fDIV = 9.6 MHz), the detectable baudrates start at 200 kBaud (Br0)
down to 1042 Baud (Br8). Table 67 shows the baudrate table for this example.
Table 67
Autobaud Detection using Non-Standard Baudrates (fDIV = 9.6 MHz)
Baudrate
Numbering
Detectable Non-Standard Divide Factor d f
Baudrates
BG is loaded after
detection with value
Br0
200.000 kBaud
48
2
= 002H
Br1
100.000 kBaud
96
5
= 005H
Br2
50 kBaud
192
11
= 00BH
Br3
33.333 kBaud
288
17
= 011H
Br4
16.667 kBaud
576
35
= 023H
Br5
8333 Baud
1152
71
= 047H
Br6
4167 Baud
2304
143
= 08FH
Br7
2083 Baud
4608
287
= 11FH
Br8
1047 Baud
9216
575
= 23FH
12.1.8.4 Overwriting Registers on Successful Autobaud Detection
With a successfull autobaud detection some bits in register S0CON and S0BG are
automatically set to a value which corresponds to the mode and baudrate of the detected
serial frame conditions (see Table 68). In control register S0CON the mode control bits
Data Sheet
286
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
CON_M and the parity select bit CON_ODD are overwritten. Register S0BG is loaded
with the 13-bit reload value for the baudrate timer.
Table 68
Autobaud Detection Overwrite Values for the S0CON Register
Detected Parameters
CON_M
CON_ODD
BG_BR_VALUE
Operating Mode
7 bit, even parity
7 bit, odd parity
8 bit, even parity
8 bit, odd parity
8 bit, no parity
011
011
111
111
001
0
1
0
1
0
-
Baudrate
Br0
Br1
Br2
Br3
Br4
Br5
Br6
Br7
Br8
-
-
2
5
11
17
35
71
143
287
575
= 002H
= 005H
= 00BH
= 011H
= 023H
= 047H
= 08FH
= 11FH
= 23FH
Note: The autobaud detection interrupts are described in Chapter 12.1.10.
12.1.9
Hardware Error Detection Capabilities
To improve the safety of serial data exchange, the serial channel ASC provides an error
interrupt request flag, which indicates the presence of an error, and three (selectable)
error status flags in register S0CON, which indicate which error has been detected
during reception. Upon completion of a reception, the error interrupt request line EIR will
be activated simultaneously with the receive interrupt request line RIR, if one or more of
the following conditions are met :
– the framing error detection enable bit CON_FEN is set and any of the expected stop bits is
not high, the framing error flag CON_FE is set, indicating that the error interrupt request is due
to a framing error (Asynchronous mode only).
– If the parity error detection enable bit CON_PEN is set in the modes where a parity bit is
received, and the parity check on the received data bits proves false, the parity error flag
CON_PE is set, indicating that the error interrupt request is due to a parity error
(Asynchronous mode only).
– If the overrun error detection enable bit CON_OEN is set and the last character received was
not read out of the receive buffer by software or DMA transfer at the time the reception of a
new frame is complete, the overrun error flag CON_OE is set indicating that the error interrupt
request is due to an overrun error (Asynchronous and synchronous mode).
Data Sheet
287
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
12.1.10
Interrupts
Six interrupt sources are provided for serial channel ASC. Line TIR indicates a transmit
interrupt, TBIR indicates a transmit buffer interrupt, RIR indicates a receive interrupt and
EIR indicates an error interrupt of the serial channel. The autobaud detection unit
provides two additional interrupts, the ABSTIR start of autobaud operation interrupt and
the ABDETIR autobaud detected interrupt. The interrupt output lines TBIR, TIR, RIR,
EIR, ABSTIR, and ABDETIR are activated (active state) for two periods of the module
clock fMODfive.
The cause of an error interrupt request (framing, parity, overrun error) can be identified
by the error status flags FE, PE, and OE which are located in control register CON. For
the two autobaud detection interrupts register ABSTAT provides status information.
Note: In contrary to the error interrupt request line EIR, the error status flags FE/PE/OE
are not reset automatically but must be cleared by software.
For normal operation (ie. besides the error interrupt) the ASC provides three interrupt
requests to control data exchange via this serial channel:
– TBIR is activated when data is moved from TBUF to the transmit shift register.
– TIR is activated before the last bit of an asynchronous frame is transmitted, or after the last
bit of a synchronous frame has been transmitted.
– RIR is activated when the received frame is moved to RBUF.
The transmitter is serviced by two interrupt handlers. This provides advantages for the
servicing software.
For single transfers, it is sufficient to use the transmitter interrupt (TIR), which indicates
that the previously loaded data has been transmitted, except for the last bit of an
asynchronous frame.
For multiple back-to-back transfers it is necessary to load the following piece of data at
last until the time the last bit of the previous frame has been transmitted. In asynchronous
mode this leaves just one bit-time for the handler to respond to the transmitter interrupt
request, in synchronous mode it is impossible at all.
Using the transmit buffer interrupt (TBIR) to reload transmit data gives the time to
transmit a complete frame for the service routine, as TBUF may be reloaded while the
previous data is still being transmitted.
The ABSTIR start of autobaud operation interrupt is generated whenever the autobaud
detection unit is enabled (ABEN and ABDETEN and ABSTEN set), and a start bit has
been detected at RXD. In this case ABSTIR is generated during autobaud detection
whenever a start bit is detected.
The ABDETIR autobaud detected interrupt is always generated after recognition of the
second character of the two-byte frame, this means after a successful autobaud
detection. If ABCON_FCDETEN is set the ABDETIR autobaud detected interrupt is also
generated after the recognition of the first character of the two-byte frame.
Data Sheet
288
2001-02-23
C165UTAH
Asynchronous/Synchr. Serial Interface
Asynchronous Mode
TIR
TIR
RIR
Stop
Stop
Start
TBIR
Start
Stop
Idle
TBIR
Start
TBIR
TIR
RIR
Idle
RIR
Synchronous Mode
TIR
TBIR
TIR
TIR
TBIR
TBIR
Idle
Idle
RIR
RIR
RIR
Asynchronous Modes
Autobaud Detection
1)
Figure 92
2. character
Stop
ABDETIR
Start
1. character
ABDETIR 1)
Stop
Idle
Start
ABSTIR
only if FCDETEN=1
ASC Interrupt Generation
As shown in Figure 92, TBIR is an early trigger for the reload routine, while TIR indicates
the completed transmission. Software using handshake therefore should rely on TIR at
the end of a data block to make sure that all data has really been transmitted.
Data Sheet
289
2001-02-23
C165UTAH
Real Time Clock (RTC)
13
Real Time Clock (RTC)
13.1
Introduction
The Real Time Clock (RTC) module of the C165UTAH basically is an independent timer
chain and counts time ticks. The base frequency of the RTC can be programmed via a
reload counter. The RTC can work fully asynchronous to the system frequency and is
optimized on low power consumption.
13.1.1
Features
The RTC serves for different purposes:
–
–
–
–
System clock to determine the current time and date
Cyclic time based interrupt
Alarm interrupt for wake up on a defined time
48-bit timer for long term measurements
13.1.2
Overview
The real time clock module provides three different types of registers: two control
registers for controlling the RTC´s functionality, three data registers for setting the clock
divider for RTC base frequency programming and for flexible interrupt generation, and
three counter registers they contain the actual time and date. The interrupts are
programmed via one interrupt sub node register.
13.2
Function Description
The RTC module consists of a chain of 2 divider blocks, the reloadable 16-bit timer T14
and the 32-bit RTC timer (accessible via registers RTCH and RTCL). Both timers count
up. Timer T14 is reloaded with the value of register T14REL on every timer T14
overflow.
The count input of the RTC module (RTC_REF_CLK) can be optional divided by a
prescaler with factor 8, see Figure 94.
The RTC module operates in two different modes, an asynchronous and a synchronous
mode. In synchronous mode the RTC module is clocked with a synchronous clock
referring to the CPU clock (RTC clock). In asynchronous mode the RTC module is
clocked with the asynchronous counting input clock (RTC_REF_CLK). The
asynchronous mode is necessary in case of a very low or disabled CPU clock (eg. sleep
mode).
The mode control (asynchronous / synchronous) of the RTC is described as follows:
• The RTC is switched to asynchronous mode if SYSCON2.RCS = '1'
• The RTC is switched to asynchronous mode if device is in sleep mode.
Data Sheet
290
2001-02-23
C165UTAH
Real Time Clock (RTC)
• RTC is switched to asynchronous mode if the system is not in slowdown mode and
the system clock is not locked (observe selected CLK_CFG as set during Reset).
This means that the mode is controlled by bit RCS of register SYSCON2 (see page 469)
unless the system's clock setting indicates that no or an unreliable CPU clock is
available. In the latter case, synchronous mode is not working and thus the RTC is forced
into asynchronous mode.
As long as the CPU clock is four times faster than the RTC_REF_CLK, the RTC module
can be operated in synchronous mode. Otherwise the asynchronous mode has to be
selected by software.
In asynchronous mode no writing but only asynchronous reading to the registers via the
internal bus is possible.
13.2.1
RTC Block Diagram
Figure 93 shows the RTC block diagram:
RTC_T14INT xb(3)
RTC_INT irq(15) and
altern. source fast ext. interrupt 7
RTC_REF_CLK
Interrupt Sub Node RTCISNC
RTCR
RTC0INT RTC1INT
RTCRELL
RTC2INT
RTC3INT
RTCRELH
10 bit
6 bit
6 bit
10 bit
10 bit
6 bit
6 bit
10 bit
T14REL (16 bit)
8:1
MUX
RTCPRE
T14 (16 bit)
T14_IN
Figure 93
Data Sheet
RTCL
RTCH
RTC Block Diagram
291
2001-02-23
C165UTAH
Real Time Clock (RTC)
13.2.2
RTC Control
The operating behaviour of the RTC module is controlled by the RTCCON register. The
RTC starts counting by setting the run bit RTCR. After reset the run bit is set and the RTC
automatically starts operation. The bit RTCPRE selects a prescaler which divides the
counting clock by factor 8. Activating the prescaler reduces the resolution of the reload
counter T14. If the prescaler is not activated, the RTC may be lose counting clocks on
switching from asynchronous to synchronous mode and back. This effect can be avoided
by activating the prescaler.
Setting the bits T14DEC or T14INC decrements or increments the T14 timer with the
next count event. If at the next count event a reload has to be executed, then an
increment operation is delayed until the next count event occurs. The in/decrement
function can only be used if register T14REL is not equal to FFFFH. These bits are
cleared by hardware after the decrement/increment operation.
13.2.3
System Clock Operation
A real time system clock can be maintained that represents the current time and date. If
the RTC module is not effected by a system reset, it keeps running also during idle mode
and power down mode.
The maximum resolution (minimum stepwidth) for this clock information is determined by
timer T14’s input clock. The maximum usable timespan is achieved when T14REL is
loaded with 0000H and so T14 divides by 216.
13.2.4
Cyclic Interrupt Generation
The RTC module can generate an interrupt request RTC_T14INT whenever timer T14
overflows and is reloaded. This interrupt request may eg. be used to provide a system
time tick independent of the CPU clock frequency without loading the general purpose
timers, or to wake up regularly from idle mode. The T14 overflow interrupt
(RTC_T14INT) cycle time can be adjusted via the timer T14 reload register T14REL.
This interrupt request is also ored with all other interrupts of the RTC via the RTC
interrupt sub node RTCISN.
13.2.5
Alarm Interrupt Generation
The RTC module can also provide an alarm interrupt. For an easier programming of this
interrupt, the RTCL and RTCH timer can be divided into smaller reloadable timers. Each
sub-timer can be programmed for an overflow on different time bases (e.g. second, hour,
minute, day). With each timer overflow a RTC interrupt is generated. All these RTC
interrupts are ored via the interrupt sub node RTCISNC to one interrupt request
RTC_INT. Additionally the RTC_T14INT is connected to this interrupt sub node.
Data Sheet
292
2001-02-23
C165UTAH
Real Time Clock (RTC)
13.2.6
48-bit Timer Operation
The concatenation of the 16-bit reload timer T14 and the 32-bit RTC timer can be
regarded as a 48-bit timer which counts with the RTC count input frequency
(RTC_REF_CLK) divided by the fixed prescaler, if the prescaler is selected. The reload
registers T14REL, RTCRELL and RTCRELH should be cleared to get a 48-bit binary
timer. However, any other reload values may be used.
The maximum usable timespan is 248 (≈1014) T14 input clocks, which equals more than
100 years (referring to a RTC count input frequency RTC_REF_CLK of 625 KHz, which
is equal to a 20 MHz Oscillator divided by 32, and an activated prescaler).
13.2.7
Defining the RTC Time Base
The count input of the RTC module (RTC_REF_CLK) is connected as shown in
Figure 94. The RTC timer base is equal to timer T14 overflow and depends on the
selectable prescaler and on the reload counter T14.
RTC module (simple drawing)
XTAL1
XTAL2
RTC_T14INT
OSC
RTC_REF_CLK
32:1
8:1
M
U
X
T14
16 bit
RTC Timer
32 bit
RTC_INT
M
U
X
CPU_CLOCK
RTC_CLK_SEL
RTCmoduleclk001
Figure 94
RTC module clocking scheme
The table below lists the RTC_T14INT interrupt period range for several RTC count input
frequencies:
Table 69
RTC Interrupt Periods
Oscillator
Frequency
Divider
Factor
RTC Input
Frequency
32 KHz
1
32 KHz
1 MHz
32
312.5 KHz
4 MHz
32
125 KHz
Data Sheet
Prescaler
Factor
RTC_T14INT Period
Minimum
Maximum
31.25 µs
2.048 s
8
256.0 µs
16.77 s
8
64.0 µs
4.194 s
293
2001-02-23
C165UTAH
Real Time Clock (RTC)
Table 69
RTC Interrupt Periods
Oscillator
Frequency
Divider
Factor
RTC Input
Frequency
Prescaler
Factor
RTC_T14INT Period
Minimum
Maximum
5 MHz
32
19.531 KHz
8
51.2 µs
3.355 s
8 MHz
32
156.25 KHz
8
32.0 µs
2.097 s
10 MHz
32
312.5 KHz
8
25.6 µs
1.678 s
12 MHz
32
375 KHz
8
21.3 µs
1.398 s
16 MHz
32
500 KHz
8
16.0 µs
1.049 s
20 MHz
32
625 KHz
8
12.8 µs
0.839 s
24 MHz
32
750 KHz
8
10.67 µs
0.699 s
25 MHz
32
781.25 KHz
8
10.24 µs
0.671 s
32 MHz
32
1 MHz
8
8.0 µs
0.524 s
50 MHz
32
1.56 MHz
8
5.12 µs
0.336 s
Table 70 lists the T14 reload values for a time base of 1 s (A), 100 ms (B) and 1 ms (C)
and several RTC input frequencies:
Table 70
RTC Input
Frequency
RTC Reload Values
Reload Value A
Reload Value B
Reload Value C
T14REL
Base
T14REL
Base
T14REL
Base
32 KHz
8300H
1.000 s
F380H
100.0 ms
FFE0H
1.000 ms
312.5 KHz
F0BEH
0.999 s
FE79H
100.1 ms
FFFCH
1.024 ms
125 KHz
C2F7H
1.000 s
F9E5H
100.0 ms
FFF0H
1.024 ms
19.531 KHz
B3B5H
0.999 s
F85FH
99.9 ms
FFECH
1.024 ms
156.25 KHz
85EEH
1.000 s
F3CBH
100.0 ms
FFE1H
0.992 ms
312.5 KHz
6769H
1.000 s
F0BEH
99.9 ms
FFD9H
0.998 ms
375 KHz
48E5H
1.000 s
EDB0H
100.0 ms
FFD1H
1.003 ms
500 KHz
0BDCH
1.000 s
E796H
100.0 ms
FFC1H
1.008 ms
625 KHz
E17BH
100.0 ms
FFB2H
0.998 ms
750 KHz
DB61H
100.0 ms
FFA2H
1.003 ms
781.25 KHz
D9DAH
100.0 ms
FF9EH
1.004 ms
1 MHz
CF2CH
100.0 ms
FF83H
1.000 ms
1.56 MHz
B3B5H
99.9 ms
FF3DH
0.998 ms
Data Sheet
294
2001-02-23
C165UTAH
Real Time Clock (RTC)
13.2.8
Increased RTC Accuracy through Software Correction
The accuracy of the C165UTAH’s RTC is determined by the oscillator frequency and by
the respective prescaling factor (excluding or including T14 and the selectable
Prescaler). The accuracy limit generated by the prescaler is due to the quantization of a
binary counter (where the average is zero), while the accuracy limit generated by the
oscillator frequency is due to the difference between ideal and real frequency (and
therefore accumulates over time). The total accuracy of the RTC can be further
increased via software for specific applications that demand a high time accuracy.
The key to the improved accuracy is the knowledge of the exact oscillator frequency. The
relation of this frequency to the expected ideal frequency is a measure for the RTC’s
deviation. The number N of cycles after which this deviation causes an error of ±1 cycle
can be easily computed. So the only action is to correct the count by ±1 after each series
of N cycles.
This correction may be applied to the RTC register as well as to T14. Also the correction
may be done cyclic, eg. within T14’s interrupt service routine, or by evaluating a formula
when the RTC registers are read (for this the respective „last“ RTC value must be
available somewhere). T14 can be adjusted by a write access or better by using the in/
decrement function provided by the RTCCON register.
Note: For the majority of applications, however, the standard accuracy provided by the
RTC’s structure will be more than sufficient.
13.2.9
Hardware dependend RTC Accuracy
The RTC has different counting accuracies, depending on the operating mode (with or
without prescaler). There is only an impact on the counting accuracy when switching the
RTC from synchronous mode to asynchronous mode and back.
.
Table 71
Impact on counting accuracy
Operating mode
Inaccuracy in T14
counting ticks
without prescaler
+0 / -0.5
with prescaler
+0 / -0
13.2.10
Interrupt Sub Node RTCISNC
All RTC interrupts are connected to one interrupt node via an interrupt sub node. For this
interrupt sharing each interrupt source has additionally to the node enable and request
flag its own enable and request flag located in register RTCISNC. After a RTC interrupt
Data Sheet
295
2001-02-23
C165UTAH
Real Time Clock (RTC)
(RTC_INT) is arbitrated, the interrupt service routine has to check the request flags of all
enabled sources and run the respective software routine. The request flags have to be
deleted by software before leaving the interrupt service routine.
The following figure shows the additional necessary differentiating circuits for the
interrupt logic:
RTCISNC
T14
IE
RTC_T14INT
RTC0INT
RTC1INT
RTC2INT
RTC3INT
Diff.
Circuit
Diff.
Circuit
Diff.
Circuit
Diff.
Circuit
Diff.
Circuit
&
T14
IR1)
RTC0
IE
&
RTC0
IR1)
RTC1
IE
1
&
RTC1
IR1)
RTC2
IE
Pulse
Generation
&
RTC2
IR1)
RTC3
IE
&
RTC3
IR1)
Note: 1)The ISR (Interrupt Service Routine) must delete the IR flag in register
RTCISNC manually. Otherwise no further interrupts can be detected.
Figure 95
13.2.11
Differentiating Circuits for RTC interrupt
RTC Disable Functionality
The Peripheral Kernel of the RTC can be disabled, if the RTC functionality is not used.
In this case only the bus interface is enabled. Disabling the RTC module reduces the
power consumption and the generated noise of the complete system. The disable
request can be set in two different ways. The bit RTCDIS (if implemented in C165UTAH)
of the central peripheral control register SYSCON3 controls the disable request of the
RTC module. Additional the request can be set with bit RTCDISR inside the RTC Clock
Control register (RTCCLC). The central and the local disable requests are ored. Clearing
the respective disable request flag enables the RTC module again. Note that both
Data Sheet
296
2001-02-23
C165UTAH
Real Time Clock (RTC)
request flags have to be cleared for enabling the RTC module. An activated request flag
sets directly the disabling status flag RTCDISS inside the RTCCLC register and disables
the clock within the Clock Gating Module. Since the RTCCLC register is clocked with the
bus clock, the disabling status of the RTC can be read and the RTC can be enabled
again.
The following figure shows the disabling mechanism of the RTC module:
RTC module
OCDS_P_SUSPEND
RTC_EX_DISR
RTC_DIS_N
15
3
2
EXR
DIS
SUSP
EN
1
RTC
DISS
0
RTC
DISR
RTCCLC
Clock
Gating
Figure 96
Disabling Mechanism of the RTC
In disable state no write access to RTC registers is possible. The only exeption is the
RTCCLC register.
13.2.12
Register Definition of RTC module
The following table shows the register addresses map:
Table 72
SFR Address
Address Map Overview
b/p
F0C8H
F1CCH
Register Name
RTCCLC
b
RTCCON
F0D0H
T14REL
F0D2H
T14
F0D4H
RTCL
F0D6H
RTCH
Data Sheet
297
2001-02-23
C165UTAH
Real Time Clock (RTC)
SFR Address
b/p
Register Name
F0CCH
RTCRELL
F0CEH
RTCRELH
F1C8H
b/p
b: bitaddressable
RTCISNC
p: bit protected
RTC Clock Control Register
RTCCLC (F0C8H)
Reset Value: 0000H
15
14
13
12
11
10
9
8
7
6
5
4
0
0
0
0
0
0
0
0
0
0
0
0
3
Bit
Function
RTCDISR
RTC Disable Request Bit
RTCDISR = ‘0’: RTC clock disable not requested
RTCDISR = ‘1’: RTC clock disable requested
RTCDISS
RTC Disable Status Bit
RTCDISS = ‘0’: RTC clock enabled
RTCDISS = ‘1’: RTC clock disabled
SUSPEN
Peripheral Suspend Enable Bit for OCDS
SUSPEN = ‘0’: Peripheral suspend disabled
SUSPEN = ‘1’: Peripheral suspend enabled
EXDISR
External Disable Request
EXRDIS = ‘0’: External clock disable Request is enabled
EXRDIS = ‘1’: External clock disable Request is disabled
298
1
0
EX SUS RTC RTC
DISR PEN DISS DISR
rw
Data Sheet
2
rw
r
rw
2001-02-23
C165UTAH
Real Time Clock (RTC)
RTC Control Register
RTCCON (F1CCH)
bitaddressableReset Value:0003H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
ACC
POS
0
0
0
0
0
0
0
0
0
0
0
T14
INC
T14
DEC
rw
rw
r
1
0
RTC RTC
PRE
R
rw
rw
Bit
Function
RTCR
RTC Run Bit
RTCR = ‘0’: RTC stops
RTCR = ‘1’: RTC runs
RTCPRE
RTC Input Source Prescaler enable
RTCPRE = ‘0’: Input Prescaler disabled
RTCPRE = ‘1’: Input Prescaler enabled
T14DEC
Decrement T14 Timer Value
Setting this bit to 1 effects a decrement of the T14 timer value. The bit is cleared
by hardware after decrementation.
T14INC
Increment T14 Timer Value
Setting this bit to 1 effects an increment of the T14 timer value. The bit is cleared
by hardware after incrementation.
ACCPOS
RTC register access possible
This bit indicates that a synchronous read / write access to RTC registers is
possible. The Clock Control register RTCCLC can allways be accessed.
ACCPOS = ‘0’: No write access is possible, only asynchronous reads.
ACCPOS = ‘1’: Read / Write access is possible
Note : For compatibility reasons the bits RTCR and RTCPRE are set on asynchronous HW reset
(power on reset).
Prescaler Timer T14
Timer T14 generates the input clock for the RTC register and the periodic interrupt
T14 (F0D2H)
15
14
Data Sheet
Reset Value:0000H
13
12
11
10
9
8
7
299
6
5
4
3
2
1
0
2001-02-23
C165UTAH
Real Time Clock (RTC)
Timer T14 Reload Register
T14REL (F0D0H)
15
14
13
Reset Value:0000H
12
11
10
9
8
7
6
5
4
3
2
1
0
5
4
3
2
1
0
5
4
3
2
1
0
5
4
3
2
1
0
5
4
3
2
1
0
RTC Count Register Low Word
RTCL (F0D4H)
15
14
13
Reset Value:0000H
12
11
10
9
8
7
6
RTC Count Register High Word
RTCH (F0D6H)
15
14
13
Reset Value:0000H
12
11
10
9
8
7
6
RTC Reload Register Low Word
RTCRELL (F0CCH)
15
14
13
Reset Value:0000H
12
11
10
9
8
7
6
RTC Reload Register High Word
RTCRELH (F0CEH)
15
14
Data Sheet
13
Reset Value:0000H
12
11
10
9
8
7
300
6
2001-02-23
C165UTAH
Real Time Clock (RTC)
RTC Interrupt Sub Node Control
RTCISNC (F1C8H)
bitaddressableReset Value:UUUUH
15
14
13
12
11
10
0
0
0
0
0
0
9
8
7
6
5
4
3
2
RTC RTC RTC RTC RTC RTC RTC RTC
3 IR 3IE 2 IR 2IE 1 IR 1IE 0 IR 0IE
rw
rw
rw
rw
rw
rw
Bit
Function
T14IE
T14 Overflow Interrupt Enable Control Bit
‘0’: Interrupt request is disabled
‘1’: Interrupt request is enabled
T14IR
T14 Overflow Interrupt Request Flag (bit protected)
‘0’: No request pending
‘1’: This source has raised an interrupt request
RTCxIE
RTCx Interrupt Enable Control Bit
‘0’: Interrupt request is disabled
‘1’: Interrupt request is enabled
RTCxIR
RTCx Interrupt Request Flag (bit protected)
‘0’: No request pending
‘1’: This source has raised an interrupt request
rw
rw
1
0
T14
IR
T14
IE
rw
rw
Note : The interrupt request flags of the RTC interrupt sub node have to be cleared by software
inside the interrupt service routine.
Data Sheet
301
2001-02-23
C165UTAH
High-Speed Synchronous Serial Interface
14
High-Speed Synchronous Serial Interface
The High-Speed Synchronous Serial Interface SSC provides flexible high-speed serial
communication between the C165UTAH and other microcontrollers, microprocessors or
external peripherals.
The SSC supports full-duplex and half-duplex synchronous communication up to 18
MBaud in SSC Master Mode and 9 MBaud in SSC Slave Mode (@ 36 MHz CPU clock).
The serial clock signal can be generated by the SSC itself (master mode) or be received
from an external master (slave mode). Data width, shift direction, clock polarity and
phase are programmable. This allows communication with SPI-compatible devices.
Transmission and reception of data is double-buffered. A 16-bit baud rate generator
provides the SSC with a separate serial clock signal.
The high-speed synchronous serial interface can be configured in a very flexible way, so
it can be used with other synchronous serial interfaces (eg. the ASC in synchronous
mode), serve for master/slave or multimaster interconnections or operate compatible
with the popular SPI interface. So it can be used to communicate with shift registers (IO
expansion), peripherals (eg. EEPROMs etc.) or other controllers (networking). The SSC
supports half-duplex and full-duplex communication. Data is transmitted or received on
pins MTSR/P3.9 (Master Transmit / Slave Receive) and MRST/P3.8 (Master Receive /
Slave Transmit). The clock signal is output or input on pin SCLK/P3.13. These pins are
alternate functions of Port 3 pins.
Ports & Direction Control
Alternate Functions
System
Data Registers
Control Registers
Interrupt Control
SSCCLC
ODP3
SSCBR
DP3
SSCTB
SSCRIC
P3
SSCRB
SSCEIC
SSCCON
SSCTIC
SCLK / P3.13
MTSR / P3.9
MRST / P3.8
SSCCLC
ODP3
DP3
SSCBR
SSCTB
SSCTIC
Figure 97
Data Sheet
SSC Clock Control Register
Port 3 Open Drain Control Register
Port 3 Direction Control Register
SSC Baud Rate Generator/Reload Register
SSC Transmit Buffer Register
SSC Transmit Interrupt Control Register
P3
SSCCON
SSCRB
SSCRIC
SSCEIC
Port 3 Data Register
SSC Control Register
SSC Receive Buffer Register
SSC Receive Interrupt Control Register
SSC Error Interrupt Control Register
SFRs and Port Pins associated with the SSC
302
2001-02-23
C165UTAH
High-Speed Synchronous Serial Interface
CPU
Clock
Baud Rate
Generator
Slave Clock
Master Clock
Clock
Control
SCLK
Shift
Clock
Receive Int. Request
Transmit Int. Request
Error Int.Request
SSC Control Block
Status
Control
MTSR
Pin
Control
16-Bit Shift Register
Transmit Buffer
Register SSCTB
MRST
Receive Buffer
Register SSCRB
Internal Bus
Figure 98
MCB01957
Synchronous Serial Channel SSC Block Diagram
The operating mode of the serial channel SSC is controlled by its bit-addressable control
register SSCCON. This register serves for two purposes:
• during programming (SSC disabled by SSCEN=’0’) it provides access to a set of
control bits
• during operation (SSC enabled by SSCEN=’1’) it provides access to a set of status
flags
Register SSCCON is shown below in each of the two modes.
Data Sheet
303
2001-02-23
C165UTAH
High-Speed Synchronous Serial Interface
SSCCON (FFB2H / D9H)
15
14
SSC SSC
EN=0 MS
rw
rw
13
-
12
SFR
11
SSC SSC
AREN BEN
rw
rw
10
9
8
7
SSC SSC SSC
PEN REN TEN
rw
rw
Reset Value: 0000H
rw
-
6
5
4
3
SSC SSC SSC
PO
PH
HB
rw
rw
rw
2
1
0
SSCBM
rw
Bit
Function (Programming Mode, SSCEN = ‘0’)
SSCBM
SSC Data Width Selection
0:
Reserved. Do not use this combination.
1...15 :
Transfer Data Width is 2...16 bit (+1)
SSCHB
SSC Heading Control Bit
0:
Transmit/Receive LSB First
1:
Transmit/Receive MSB First
SSCPH
SSC Clock Phase Control Bit
0:
Shift transmit data on the leading clock edge, latch on trailing edge
1:
Latch receive data on leading clock edge, shift on trailing edge
SSCPO
SSC Clock Polarity Control Bit
0:
Idle clock line is low, leading clock edge is low-to-high transition
1:
Idle clock line is high, leading clock edge is high-to-low transition
SSCTEN
SSC Transmit Error Enable Bit
0:
Ignore transmit errors
1:
Check transmit errors
SSCREN
SSC Receive Error Enable Bit
0:
Ignore receive errors
1:
Check receive errors
SSCPEN
SSC Phase Error Enable Bit
0:
Ignore phase errors
1:
Check phase errors
SSCBEN
SSC Baudrate Error Enable Bit
0:
Ignore baudrate errors
1:
Check baudrate errors
SSCAREN
SSC Automatic Reset Enable Bit
0:
No additional action upon a baudrate error
1:
The SSC is automatically reset upon a baudrate error
SSCMS
SSC Master Select Bit
0:
Slave Mode. Operate on shift clock received via SCLK.
1:
Master Mode. Generate shift clock and output it via SCLK.
SSCEN
SSC Enable Bit = ‘0’
Transmission and reception disabled. Access to control bits.
Data Sheet
304
2001-02-23
C165UTAH
High-Speed Synchronous Serial Interface
SSCCON (FFB2H / D9H)
15
14
SSC SSC
EN=1 MS
rw
rw
13
-
12
SFR
11
SSC SSC
BSY BE
rw
rw
10
9
8
SSC SSC SSC
PE
RE
TE
rw
rw
rw
Reset Value: 0000H
7
6
5
4
3
2
1
-
-
-
-
SSCBC
-
-
-
-
r
0
Bit
Function (Operating Mode, SSCEN = ‘1’)
SSCBC
SSC Bit Count Field
Shift counter is updated with every shifted bit. Do not write to!!!
SSCTE
SSC Transmit Error Flag
1:
Transfer starts with the slave’s transmit buffer not being updated
SSCRE
SSC Receive Error Flag
1:
Reception completed before the receive buffer was read
SSCPE
SSC Phase Error Flag
1:
Received data changes around sampling clock edge
SSCBE
SSC Baudrate Error Flag
1:
More than factor 2 or 0.5 between Slave’s actual and expected
baudrate
SSCBSY
SSC Busy Flag
Set while a transfer is in progress. Do not write to!!!
SSCMS
SSC Master Select Bit
0:
Slave Mode. Operate on shift clock received via SCLK.
1:
Master Mode. Generate shift clock and output it via SCLK.
SSCEN
SSC Enable Bit = ‘1’
Transmission and reception enabled. Access to status flags and M/S control.
Note: • The target of an access to SSCCON (control bits or flags) is determined by the
state of SSCEN prior to the access, ie. writing C057H to SSCCON in programming
mode (SSCEN=’0’) will initialize the SSC (SSCEN was ‘0’) and then turn it on
(SSCEN=’1’).
• When writing to SSCCON, make sure that reserved locations receive zeros.
The shift register of the SSC is connected to both the transmit pin and the receive pin via
the pin control logic (see block diagram). Transmission and reception of serial data is
synchronized and takes place at the same time, ie. the same number of transmitted bits
is also received. Transmit data is written into the Transmit Buffer SSCTB. It is moved to
the shift register as soon as this is empty. An SSC-master (SSCMS=’1’) immediately
begins transmitting, while an SSC-slave (SSCMS=’0’) will wait for an active shift clock.
When the transfer starts, the busy flag SSCBSY is set and a transmit interrupt request
(SSCTIR) will be generated to indicate that SSCTB may be reloaded again. When the
programmed number of bits (2...16) has been transferred, the contents of the shift
register are moved to the Receive Buffer SSCRB and a receive interrupt request
Data Sheet
305
2001-02-23
C165UTAH
High-Speed Synchronous Serial Interface
(SSCRIR) will be generated. If no further transfer is to take place (SSCTB is empty),
SSCBSY will be cleared at the same time. Software should not modify SSCBSY, as this
flag is hardware controlled.
Note: Only one SSC (etc.) can be master at a given time.
The transfer of serial data bits can be programmed in many respects:
• the data width can be chosen from 2 bits to 16 bits
• transfer may start with the LSB or the MSB
• the shift clock may be idle low or idle high
• data bits may be shifted with the leading or trailing edge of the clock signal
• the baudrate may be set from 274.7 Baud up to 18 MBaud (@ 36 MHz CPU clock)
• the shift clock can be generated (master) or received (slave)
This allows the adaptation of the SSC to a wide range of applications, where serial data
transfer is required.
The Data Width Selection supports the transfer of frames of any length, from 2-bit
“characters” up to 16-bit “characters”. Starting with the LSB (SSCHB=’0’) allows
communication eg. with ASC devices in synchronous mode (C166 family) or 8051 like
serial interfaces. Starting with the MSB (SSCHB=’1’) allows operation compatible with
the SPI interface.
Regardless which data width is selected and whether the MSB or the LSB is transmitted
first, the transfer data is always right aligned in registers SSCTB and SSCRB, with the
LSB of the transfer data in bit 0 of these registers. The data bits are rearranged for
transfer by the internal shift register logic. The unselected bits of SSCTB are ignored, the
unselected bits of SSCRB will be not valid and should be ignored by the receiver service
routine.
The Clock Control allows the adaptation of transmit and receive behaviour of the SSC
to a variety of serial interfaces. A specific clock edge (rising or falling) is used to shift out
transmit data, while the other clock edge is used to latch in receive data. Bit SSCPH
selects the leading edge or the trailing edge for each function. Bit SSCPO selects the
level of the clock line in the idle state. So for an idle-high clock the leading edge is a
falling one, a 1-to-0 transition. The figure below is a summary.
Data Sheet
306
2001-02-23
C165UTAH
High-Speed Synchronous Serial Interface
Serial Clock
SCLK
SSCPO SSCPH
0
0
0
1
1
0
1
1
Pins
MTSR/MRST
First
Bit
Transmit Data
Last
Bit
MCD01960
Latch Data
Shift Data
Figure 99
14.1
Serial Clock Phase and Polarity Options
Full-Duplex Operation
The different devices are connected through three lines. The definition of these lines is
always determined by the master: The line connected to the master's data output pin
MTSR is the transmit line, the receive line is connected to its data input line MRST, and
the clock line is connected to pin SCLK. Only the device selected for master operation
generates and outputs the serial clock on pin SCLK. All slaves receive this clock, so their
pin SCLK must be switched to input mode (DP3.13=’0’). The output of the master’s shift
register is connected to the external transmit line, which in turn is connected to the
slaves’ shift register input. The output of the slaves’ shift register is connected to the
external receive line in order to enable the master to receive the data shifted out of the
slave. The external connections are hard-wired, the function and direction of these pins
is determined by the master or slave operation of the individual device.
Note: The shift direction shown in the figure applies for MSB-first operation as well as for
LSB-first operation.
When initializing the devices in this configuration, select one device for master operation
(SSCMS=’1’), all others must be programmed for slave operation (SSCMS=’0’).
Initialization includes the operating mode of the device's SSC and also the function of
the respective port lines (see “Port Control”).
Data Sheet
307
2001-02-23
C165UTAH
High-Speed Synchronous Serial Interface
Master
Device #1
Device #2
Shift Register
MTSR
MRST
Clock
Slave
Shift Register
SCLK
Transmit
Receive
Clock
MTSR
MRST
SCLK
Clock
Slave
Device #3
Shift Register
MTSR
MRST
SCLK
Clock
MCS01963
Figure 100
SSC Full Duplex Configuration
The data output pins MRST of all slave devices are connected together onto the one
receive line in this configuration. During a transfer each slave shifts out data from its shift
register. There are two ways to avoid collisions on the receive line due to different slave
data:
Only one slave drives the line, ie. enables the driver of its MRST pin. All the other
slaves have to program there MRST pins to input. So only one slave can put its data onto
the master's receive line. Only receiving of data from the master is possible. The master
selects the slave device from which it expects data either by separate select lines, or by
sending a special command to this slave. The selected slave then switches its MRST line
to output, until it gets a deselection signal or command.
The slaves use open drain output on MRST. This forms a Wired-AND connection. The
receive line needs an external pullup in this case. Corruption of the data on the receive
line sent by the selected slave is avoided, when all slaves which are not selected for
transmission to the master only send ones (‘1’). Since this high level is not actively driven
onto the line, but only held through the pullup device, the selected slave can pull this line
actively to a low level when transmitting a zero bit. The master selects the slave device
Data Sheet
308
2001-02-23
C165UTAH
High-Speed Synchronous Serial Interface
from which it expects data either by separate select lines, or by sending a special
command to this slave.
After performing all necessary initializations of the SSC, the serial interfaces can be
enabled. For a master device, the alternate clock line will now go to its programmed
polarity. The alternate data line will go to either '0' or '1', until the first transfer will start.
After a transfer the alternate data line will always remain at the logic level of the last
transmitted data bit.
When the serial interfaces are enabled, the master device can initiate the first data
transfer by writing the transmit data into register SSCTB. This value is copied into the
shift register (which is assumed to be empty at this time), and the selected first bit of the
transmit data will be placed onto the MTSR line on the next clock from the baudrate
generator (transmission only starts, if SSCEN=’1’). Depending on the selected clock
phase, also a clock pulse will be generated on the SCLK line. With the opposite clock
edge the master at the same time latches and shifts in the data detected at its input line
MRST. This “exchanges” the transmit data with the receive data. Since the clock line is
connected to all slaves, their shift registers will be shifted synchronously with the
master's shift register, shifting out the data contained in the registers, and shifting in the
data detected at the input line. After the preprogrammed number of clock pulses (via the
data width selection) the data transmitted by the master is contained in all slaves’ shift
registers, while the master's shift register holds the data of the selected slave. In the
master and all slaves the content of the shift register is copied into the receive buffer
SSCRB and the receive interrupt flag SSCRIR is set.
A slave device will immediately output the selected first bit (MSB or LSB of the transfer
data) at pin MRST, when the content of the transmit buffer is copied into the slave's shift
register. It will not wait for the next clock from the baudrate generator, as the master
does. The reason for this is that, depending on the selected clock phase, the first clock
edge generated by the master may be already used to clock in the first data bit. So the
slave's first data bit must already be valid at this time.
Note: On the SSC always a transmission and a reception takes place at the same time,
regardless whether valid data has been transmitted or received. This is different
eg. from asynchronous reception on ASC.
The initialization of the SCLK pin on the master requires some attention in order to
avoid undesired clock transitions, which may disturb the other receivers. The state of the
internal alternate output lines is '1' as long as the SSC is disabled. This alternate output
signal is ANDed with the respective port line output latch. Enabling the SSC with an idlelow clock (SSCPO=’0’) will drive the alternate data output and (via the AND) the port pin
SCLK immediately low. To avoid this, use the following sequence:
• select the clock idle level (SSCPO=’x’)
• load the port output latch with the desired clock idle level (P3.13=’x’)
• switch the pin to output (DP3.13=’1’)
Data Sheet
309
2001-02-23
C165UTAH
High-Speed Synchronous Serial Interface
• enable the SSC (SSCEN=’1’)
• if SSCPO=’0’: enable alternate data output (P3.13=’1’)
The same mechanism as for selecting a slave for transmission (separate select lines or
special commands) may also be used to move the role of the master to another device
in the network. In this case the previous master and the future master (previous slave)
will have to toggle their operating mode (SSCMS) and the direction of their port pins (see
description above).
14.2
Half Duplex Operation
In a half duplex configuration only one data line is necessary for both receiving and
transmitting of data. The data exchange line is connected to both pins MTSR and MRST
of each device, the clock line is connected to the SCLK pin.
The master device controls the data transfer by generating the shift clock, while the slave
devices receive it. Due to the fact that all transmit and receive pins are connected to the
one data exchange line, serial data may be moved between arbitrary stations.
Similar to full duplex mode there are two ways to avoid collisions on the data
exchange line:
• only the transmitting device may enable its transmit pin driver
• the non-transmitting devices use open drain output and only send ones.
Since the data inputs and outputs are connected together, a transmitting device will clock
in its own data at the input pin (MRST for a master device, MTSR for a slave). By these
means any corruptions on the common data exchange line are detected, where the
received data is not equal to the transmitted data.
Data Sheet
310
2001-02-23
C165UTAH
High-Speed Synchronous Serial Interface
Master
Device #2
Device #1
MTSR
MTSR
MRST
MRST
Clock
Slave
Shift Register
Shift Register
SCLK
Clock
SCLK
Common
Transmit/
Receive
Line
Clock
Device #3
Slave
Shift Register
MTSR
MRST
SCLK
Clock
MCS01965
Figure 101
SSC Half Duplex Configuration
Continuous Transfers
When the transmit interrupt request flag is set, it indicates that the transmit buffer SSCTB
is empty and ready to be loaded with the next transmit data. If SSCTB has been reloaded
by the time the current transmission is finished, the data is immediately transferred to the
shift register and the next transmission will start without any additional delay. On the data
line there is no gap between the two successive frames. Eg. two byte transfers would
look the same as one word transfer. This feature can be used to interface with devices
which can operate with or require more than 16 data bits per transfer. It is just a matter
of software, how long a total data frame length can be. This option can also be used eg.
to interface to byte-wide and word-wide devices on the same serial bus.
Note: Of course, this can only happen in multiples of the selected basic data width, since
it would require disabling/enabling of the SSC to reprogram the basic data width
on-the-fly.
Data Sheet
311
2001-02-23
C165UTAH
High-Speed Synchronous Serial Interface
Port Control
The SSC uses three pins of Port 3 to communicate with the external world. Pin P3.13/
SCLK serves as the clock line, while pins P3.8/MRST (Master Receive / Slave Transmit)
and P3.9/MTSR (Master Transmit / Slave Receive) serve as the serial data input/output
lines.
The operation of these pins depends on the selected operating mode (master or slave).
In order to enable the alternate output functions of these pins instead of the general
purpose I/O operation, the respective port latches have to be set to '1', since the port
latch outputs and the alternate output lines are ANDed. When an alternate data output
line is not used (function disabled), it is held at a high level, allowing I/O operations via
the port latch. The direction of the port lines depends on the operating mode. The SSC
will automatically use the correct alternate input or output line of the ports when switching
modes. The direction of the pins, however, must be programmed by the user, as shown
in the tables. Using the open drain output feature helps to avoid bus contention problems
and reduces the need for hardwired hand-shaking or slave select lines. In this case it is
not always necessary to switch the direction of a port pin. The table below summarizes
the required values for the different modes and pins.
SSC Port Control
Pin
Master Mode
Slave Mode
Function
Port Latch Direction
Function
Port Latch Direction
SCLK
Serial Clock
Output
P3.13 = ’1’
DP3.13=’1’ Serial Clock
Input
P3.13 = ’x’
DP3.13=’0’
MTSR
Serial Data
Output
P3.9 = ’1’
DP3.9 = ’1’ Serial Data
Input
P3.9 = ’x’
DP3.9 = ’0’
MRST
Serial Data
Input
P3.8 = ’x’
DP3.8 = ’0’ Serial Data
Output
P3.8 = ’1’
DP3.8 = ’1’
Note: In the table above, an 'x' means that the actual value is irrelevant in the respective
mode, however, it is recommended to set these bits to '1', so they are already in
the correct state when switching between master and slave mode.
14.3
Baud Rate Generation
The serial channel SSC has its own dedicated 16-bit baud rate generator with 16-bit
reload capability, allowing baud rate generation independent from the timers.
The baud rate generator is clocked with the CPU clock divided by 2 (fCPU/2). The timer
is counting downwards and can be started or stopped through the global enable bit
SSCEN in register SSCCON. Register SSCBR is the dual-function Baud Rate
Generator/Reload register. Reading SSCBR, while the SSC is enabled, returns the
content of the timer. Reading SSCBR, while the SSC is disabled, returns the
Data Sheet
312
2001-02-23
C165UTAH
High-Speed Synchronous Serial Interface
programmed reload value. In this mode the desired reload value can be written to
SSCBR.
Note: Never write to SSCBR, while the SSC is enabled.
The formulas below calculate either the resulting baud rate for a given reload value, or
the required reload value for a given baudrate:
fCPU
BSSC =
fCPU
SSCBR = (
2 * ( + 1)
2 * BaudrateSSC
)-1
represents the content of the reload register, taken as unsigned 16-bit
integer.
The maximum baud rate that can be achieved when using a CPU clock of 36 MHz is 18
MBaud in SSC Master Mode (= ’0d’), while in SSC Slave Mode the maximum
baud rate is 9 MBaud (= ’1d’ since =’0d’ is not allowed in Slave
Mode). The minimum baud rate is 274.66 Baud ( = ’FFFFH’ = ’65535D’).
14.4
Error Detection Mechanisms
The SSC is able to detect four different error conditions. Receive Error and Phase Error
are detected in all modes, while Transmit Error and Baudrate Error only apply to slave
mode. When an error is detected, the respective error flag is set. When the
corresponding Error Enable Bit is set, also an error interrupt request will be generated
by setting SSCEIR (see figure below). The error interrupt handler may then check the
error flags to determine the cause of the error interrupt. The error flags are not reset
automatically (like SSCEIR), but rather must be cleared by software after servicing. This
allows servicing of some error conditions via interrupt, while the others may be polled by
software.
Note: The error interrupt handler must clear the associated (enabled) errorflag(s) to
prevent repeated interrupt requests.
A Receive Error (Master or Slave mode) is detected, when a new data frame is
completely received, but the previous data was not read out of the receive buffer register
SSCRB. This condition sets the error flag SSCRE and, when enabled via SSCREN, the
error interrupt request flag SSCEIR. The old data in the receive buffer SSCRB will be
overwritten with the new value and is unretrievably lost.
A Phase Error (Master or Slave mode) is detected, when the incoming data at pin MRST
(master mode) or MTSR (slave mode), sampled with the same frequency as the CPU
clock, changes between one sample before and two samples after the latching edge of
the clock signal (see “Clock Control”). This condition sets the error flag SSCPE and,
when enabled via SSCPEN, the error interrupt request flag SSCEIR.
Data Sheet
313
2001-02-23
C165UTAH
High-Speed Synchronous Serial Interface
A Baud Rate Error (Slave mode) is detected, when the incoming clock signal deviates
from the programmed baud rate by more than 100%, ie. it either is more than double or
less than half the expected baud rate. This condition sets the error flag SSCBE and,
when enabled via SSCBEN, the error interrupt request flag SSCEIR. Using this error
detection capability requires that the slave's baud rate generator is programmed to the
same baud rate as the master device. This feature detects false additional, or missing
pulses on the clock line (within a certain frame).
Note: If this error condition occurs and bit SSCAREN=’1’, an automatic reset of the SSC
will be performed in case of this error. This is done to reinitialize the SSC, if too
few or too many clock pulses have been detected.
A Transmit Error (Slave mode) is detected, when a transfer was initiated by the master
(shift clock gets active), but the transmit buffer SSCTB of the slave was not updated
since the last transfer. This condition sets the error flag SSCTE and, when enabled via
SSCTEN, the error interrupt request flag SSCEIR. If a transfer starts while the transmit
buffer is not updated, the slave will shift out the 'old' contents of the shift register, which
normally is the data received during the last transfer.
This may lead to the corruption of the data on the transmit/receive line in half-duplex
mode (open drain configuration), if this slave is not selected for transmission. This mode
requires that slaves not selected for transmission only shift out ones, ie. their transmit
buffers must be loaded with 'FFFFH' prior to any transfer.
Note: A slave with push/pull output drivers, which is not selected for transmission, will
normally have its output drivers switched. However, in order to avoid possible
conflicts or misinterpretations, it is recommended to always load the slave's
transmit buffer prior to any transfer.
Data Sheet
314
2001-02-23
C165UTAH
High-Speed Synchronous Serial Interface
Register SSCCON
SSCTEN
Transmit
Error
&
SSCEIE
Receive
Error
SSCRE
1
SSCPEN
Phase
Error
SSCEIR
&
Error
Interrupt
SSCEINT
&
SSCPE
SSCBEN
Baudrate
Error
14.5
&
SSCTE
SSCREN
Figure 102
Register SSCEIC
&
SSCBE
MCA01968
SSC Error Interrupt Control
SSC Interrupt Control
Three bit addressable interrupt control registers are provided for serial channel SSC.
Register SSCTIC controls the transmit interrupt, SSCRIC controls the receive interrupt
and SSCEIC controls the error interrupt of serial channel SSC. Each interrupt source
also has its own dedicated interrupt vector. SCTINT is the transmit interrupt vector,
SCRINT is the receive interrupt vector, and SCEINT is the error interrupt vector.
The cause of an error interrupt request (receive, phase, baudrate,transmit error) can be
identified by the error status flags in control register SSCCON.
Note: In contrary to the error interrupt request flag SSCEIR, the error status flags SSCxE
are not reset automatically upon entry into the error interrupt service routine, but
must be cleared by software.
Data Sheet
315
2001-02-23
C165UTAH
High-Speed Synchronous Serial Interface
SSCTIC (FF72H / B9H)
15
-
14
-
13
-
12
-
SFR
11
-
10
-
9
-
8
-
SSCRIC (FF74H / BAH)
15
-
14
-
13
-
12
-
-
14
-
13
-
12
-
7
6
5
4
3
SSC
TIR
SSC
TIE
ILVL
GLVL
rw
rw
rw
rw
SFR
11
-
10
-
9
-
8
-
SSCEIC (FF76H / BBH)
15
Reset Value: - - 00H
-
10
-
9
-
8
-
1
0
Reset Value: - - 00H
7
6
5
4
3
SSC
RIR
SSC
RIE
ILVL
GLVL
rw
rw
rw
rw
SFR
11
2
2
1
0
Reset Value: - - 00H
7
6
5
4
3
2
1
0
SSC
EIR
SSC
EIE
ILVL
GLVL
rw
rw
rw
rw
Note: Please refer to the general Interrupt Control Register description on page 109 for
an explanation of the control fields.
Data Sheet
316
2001-02-23
C165UTAH
High-Speed Synchronous Serial Interface
SSC Clock Control Register
SSCCLC (F0B6H)
Reset Value: 0000H
15
14
13
12
11
10
9
8
7
6
5
4
0
0
0
0
0
0
0
0
0
0
0
0
3
Bit
Function
SSCDISR
SSC Disable Request Bit
SSCDISR = ‘0’: SSC clock disable not requested
SSCDISR = ‘1’: SSC clock disable requested
SSCDISS
SSC Disable Status Bit
SSCDISS = ‘0’: SSC clock enabled
SSCDISS = ‘1’: SSC clock disabled
SUSPEN
Peripheral Suspend Enable Bit for OCDS
SUSPEN = ‘0’: Peripheral suspend disabled
SUSPEN = ‘1’: Peripheral suspend enabled
EXDISR
External Disable Request
EXRDIS = ‘0’: External clock disable Request is enabled
EXRDIS = ‘1’: External clock disable Request is disabled
317
1
0
EX SUS SSC SSC
DISR PEN DISS DISR
rw
Data Sheet
2
rw
r
rw
2001-02-23
C165UTAH
USB Interface Controller
15
USB Interface Controller
15.1
USB Features
All USB data transfers will be initiated by the USB host. The host is also suspending the
device in order to save power and controls the remote wakeup. The device itself may
resume from suspend mode. The electrical interface and the protocol is compliant with
USB specification 1.1.
The C165UTAH provides eight endpoints: in addition to Endpoint Zero as the Default
Control Pipe, seven Endpoints can be configured. Each of these endpoints can be of
type isochronous, bulk or interrupt. Two configurations with alternate settings for multiple
modes of device operation will be supported.
C165UTAH is designed to provide support for all USB device classes, including
Communication Device, Audio and HID Class.
The maximum packet length supported will be 1023 bytes.
Acknowledged (ACK) and Not Acknowledged (NACK) handshake will be generated by
Hardware (HW) within the UDC, whereas each individual endpoint can be STALLED by
Software (SW).
15.2
USB Protocol
The USB protocol is packet based and consists of the following key elements: tokens,
packets, transactions and transfers.
Tokens
Tokens will identify the type of packet in the Packet Identifier field (PID) of the token
packet. Four PIDs are defined: SOF (start of frame), SETUP (for device control), IN and
OUT (for control and other data).
Packets
Four types of packets are defined: SOF, token, data and handshake. Each packet begins
with a SYNC byte followed by the Packet Identifier (PID).
A SOF packet consists of the following fields:
–
–
–
–
SYNC byte (8 bits)
Packet Identifier (8 bits)
Frame Number (11 bits)
CRC-5 (5 bits)
A token packet consists of the following fields:
– SYNC byte (8 bits)
– Packet Identifier (8 bits)
Data Sheet
318
2001-02-23
C165UTAH
USB Interface Controller
– Address (7 bits)
– Endpoint (4 bits)
– CRC-5 (5 bits)
A data packet consists of the following fields:
–
–
–
–
SYNC byte (8 bits)
Packet Identifier (8 bits)
Data (n bytes)
CRC-5 (5 bits)
A handshake packet consists of the following fields:
– SYNC byte (8 bits)
– Packet Identifier (8 bits)
Transaction
A transaction consists of a token packet, optional data packets and a handshake packet.
Transfer
A transfer consists of one or more transactions. Control transfers consists of a setup
transaction, optional data transactions and a handshake/status transaction.
15.3
USB Endpoints
Data transactions will be handled by the UDC core as a bus master via the application
bus interface. The 8-byte SETUP control data will be written into the USBD_SETUP
registers only, whereas subsequent data transactions will use endpoint 0 or another
endpoint specified by the SETUP packet.
In addition, the USBD FIFO/Control block provides eight (i.e. per endpoint) 8-byte
Transmit FIFOs for IN transfers and eight 8-byte Receive FIFOs for OUT transfers. The
Transmit FIFO is accessed by the USBD_TXWRn register and the Receive FIFO by the
USBD_RXRDn register from the CPU via XBUS. Each Receive and Transmit FIFO
provides a Packet-Complete indication register and interrupt USBD_TXDONEn/
USBD_RXDONEn, indicating a complete packet transfer from/to the CPU.
The basic FIFO structure is shown in Figure 103 below. Each endpoint FIFO pair
generates two transfer request interrupt to the EPEC. The USBD_TXREQn interrupt
indicates that the Transmit FIFO is able to accept words from the XBUS. The
USBD_RXREQn interrupt indicates an valid word in the Receive FIFO which can be
read.
The USBD control logic multiplexes the active FIFO input/output on the unidirectional
DEV_DATA or UDC_DATA bus.
Data Sheet
319
2001-02-23
C165UTAH
USB Interface Controller
A 10-bit receive byte counter counts the length of the currently receiving packet. This
length information together with an endpoint/packet status will be provided in the length
register after the packet is received completely.
For SW syncronization purposes, an 1 ms Start-of-Frame (SOF) interrupt will be
generated periodically by the USB host. The frame number will be captured for each
received frame.
No data available for an Interrupt, Bulk or Control endpoint for the dedicated Transmit
FIFO will result in a NACK, generated by the UDC. No Receive FIFO space available will
result also in a NACK. Additionally, for receive function the RX byte count register has to
be read. Otherwise the next packet will be NACKed.
In transmit direction, the software has to read the STATUS bit from the
USBD_STATUS_REG1 register after transferring the packet in order to determine
whether the transfer was ACKed or NACKed by the host.
The SW can STALL or resume the device from suspend mode by using the command
register USBD_CMD_REG.
Data Sheet
320
2001-02-23
C165UTAH
USB Interface Controller
XBUS
TX_DIR_CTRL
3
8
0
DEV_DATA
7
USBD_TXWRn
SFR
high
byte
8-byte Transmit FIFO (IN)
16 low
byte
high
8
0
UDC_DATA
7
16 byte
8-byte Receive FIFO (OUT)
10-bit packet byte counter
Endpoint#0
Endpoint#1..6
Endpoint#7
3
RX_DIR_CTRL
low
byte
SFR
USBD_RXRDn
Setup Register #0
Application Bus
Setup Register #1
Setup Register #2
Setup Register #3
8
USBD
Control
8
8
8
8
Figure 103
USBD_INT_RXREQn (to EPEC)
USBD_INT_TXREQn (to EPEC)
EPEC_DONEn (from EPEC)
USBD_INT_RXDONEn (to CPU)
USBD_INT_TXDONEn (to CPU)
USBD basic FIFO/Control structure
Data requested by Transmit FIFOs (including control IN packets following SETUP
packets) will be transferred from an internal/external memory location to the endpoint
FIFO by the EPEC injecting MOV-instructions into the decode stage of the instruction
pipeline.
Data available by Receive FIFOs (including control OUT packets following SETUP
packets) will be transferred from the endpoint FIFO to an internal/external memory
Data Sheet
321
2001-02-23
C165UTAH
USB Interface Controller
location by the EPEC injecting MOV-instructions into the decode stage of the instruction
pipeline.
In both directions the SW will control the EPEC providing DMA-like source and
destination pointers on a per-packet basis. This provides a maximum degee of flexibility
for the application SW, e.g. a linked list structure or circular FIFOs implementation with
buffer pool elements can be used.
IN-Transactions (Device to Host: TX)
The host requests device data by using the IN transaction. The IN transaction
mechanism handled by the device HW/SW is shown in Figure 104 below.
Transmission
Timeout Error
IDLE
Polling active
Timer expired
Poll
USBD_INT_TX_DONE
Enable EPEC
Set EPEC_START
= ’1’
ISR active
USBD_INT_TX_DONE
from USBD
EPEC or FIFO
busy with
previous packet
USBD_INT_TX_DONE
= ’1’
program EPEC:
ByteCount, SrcPtr,
DestPtr
Figure 104
USBD_INT_TX_DONE
= ’0’
Transmit FIFO
Ready
Reset
USB IN-Transfer controlled by SW
Since the UDC requests data handshake from the application within 1UDC clock cycle
(@12MHz) the SW has to provide the next data packet to the Transmit FIFO in time, i.e.
before the host is requesting data, in order to prevent the FIFO from starvation resulting
in a NACK to the host. This can be achieved by either polling for the USBD_TXDONEn
for bit being asserted, indicating a free FIFO or by providing the next packet as soon as
the previous packet has been transmitted completely using the interrupt
USBD_INT_TXDONE directly.
Zero-data-length packets can be sent from SW to the host by writing into the TxEOD
register.
Data Sheet
322
2001-02-23
C165UTAH
USB Interface Controller
The device SW provides a new packet by setting up the new source and destination
pointer within the EPEC. Since the packet length of the last packet is either 0 or less than
the maximum defined packet length, the EPEC provides a 10-bit byte counter. After
reaching its terminal count value, the EPEC stops transferring data, sets the
EPEC_DONEn pulse signalizing to the USB core the end of packet and sets the EPEC
interrupt. The counter will be loaded by SW for each packet to be transmitted.
The new packet transfer is started by setting the EPEC_STARTn bit in the EPEC register
EPEC_CMD. This bit will enable the word transfer request USBD_INT_TXREQ
generated by the Transmit FIFO.
Whether the last EPEC transfer was a word or byte transfer will be handled by the EPEC
for transmit data.
By this mechanism only a minimum of SW transaction will be required which guarantees
minimum latency times for the UDC handshake procedure.
OUT Transactions (Host to Device: RX)
The host transmits device data by using the OUT transaction. The OUT transaction
mechanism handled by the device HW/SW is shown in Figure 105.
Since the UDC requests data handshake from the application within 4 UDC clock cycles
(@12MHz) the SW has to provide the next empty receive buffer to the Receive FIFO in
time, in order to prevent the FIFO from overflow resulting in a NACK to the host. This can
be achieved by either polling for the USBD_RXDONEn bit for being asserted, indicating
an empty Receive FIFO or by providing the next receive buffer as soon as the previous
packet has been received completely using the interrupt USBD_INT_RXDONE directly.
The USBD provides a 10-bit byte counter USBD_RXBYTECNTn for each endpoint
Receive FIFO counting the data strobe pulses asserted by the UDC. A completed packet
transfer over the application bus (either XferAck or XferNack) will stop the byte counter
and the counter value will be copied into the USBD_RXBYTECNTn register along with
a packet status information (e.g. packet valid). As soon as the complete packet has been
transferred by the EPEC, an endpoint receive interrupt USBD_INT_RXDONE will be
generated.
A new packet can only be received if the RX FIFO is empty, i.e. the EPEC has transfered
the last byte/word. The next packet cannot be accepted by the very same Receive FIFO
until the counter value has been cleared by a CPU read access. Therefore, once the
counter is read and cleared the EPEC is enabled again, the SW has to load the EPEC
with a new source and destination pointers before reading the counter register. This will
enable the word transfer request USBD_INT_RXREQ generated by the Receive FIFO
and EPEC starts transferring words.
Data Sheet
323
2001-02-23
C165UTAH
USB Interface Controller
Receive
Timeout Error
IDLE
Polling active
Timer expired
Poll
USBD_INT_RX_DONE
Read RXBYTECNT
Write to head of
packet
ISR active
USBD_INT_RX_DONE
from USBD
EPEC or FIFO
busy with
previous packet
USBD_INT_RX_DONE
= ’1’
program EPEC:
ByteCount, SrcPtr,
DestPtr
Figure 105
USBD_INT_RX_DONE
= ’0’
Packet transfer
to buffer complete
Reset
USB OUT-Transfer controlled by SW
Whether the last EPEC transfer was a word or byte transfer will be indicated by a
USBD_RXBYTECNT value, which will be read by SW.
By this mechanism only a minimum of SW transaction will be required which guarantees
minimum latency times for the UDC handshake procedure.
Control Endpoint 0
The default control pipe is reserved to Endpoint 0. The host communicates over the
control endpoint in order to retrieve device information, device configurations, interfaces,
alternate settings and other device endpoint characteristics. Control endpoints are
bidirectional and the data delivery will be guaranteed.
The hosts starts with an 8-bytes SETUP packet, which will be written to the SETUP
register bank USBD_Setup(3:0). A SETUP interrupt request will inform the CPU in order
to process the SETUP packet if the packet was received without any errors. A pending/
processing SETUP request, which will be overwritten by the host, has to be retired by
the device SW.
In the bidirectional data stage, optional data packets can be read/written by the host
through the 8-bytes bidirectional Endpoint 0 FIFO.
Data Sheet
324
2001-02-23
C165UTAH
USB Interface Controller
The following handshake/status stage will be generated by the UDC core itself. The
application interface has to check for the zero length data IN/OUT token in reverse
direction of the actual transfer sent by the host at the end of the data stage. I.e. the
control read (IN token) will be terminated by an OUT token with zero length data transfer
and the control write by an IN token.
The status stage will be NACK’d by the device until the SW has processed the request.
A STALL will indicate to the host that the request cannot be completed successfully by
the device.
Isochronous, Bulk and Interrupt Endpoints
All these types of transfers are handled by the FIFO/EPEC mechanism described for IN
and OUT transactions above.
Standard Device Requests
Most of the standard device requests will be handled by the UDC internally, including
SETUP stage, optional DATA stage and STATUS stage. Only GET_DESCRIPTOR,
SET_DESCRIPTOR, SYNC_FRAME will be forwarded to the application bus and
handled by SW.
The 8-byte SETUP request will be captured in the SETUP registers. IN/OUT transactions
from the host will be NACK’d until the SW is ready. The SETUP registers cannot be
overwritten by the host until all 4 registers have been read.
The USB specification allows an early termination of control transfers. For control
requests with data transmission from device to host, this means that host can send a
status out packet even thought it had requested more bytes than actually were sent
during this transfer. If Software has already set up the next packet in the FIFO’s, this data
must not be transfered with the next control in request. The data can be flushed by
Software after receiving the empty status out packet or automatically this can be done
by setting the Auto Flash Enable of the CMD register.
Requests to device, interface or endpoints with no DATA stage have to be successfully
completed (incl. STATUS stage) within 2 ms. Requests with DATA stage require the first
data packet within 10 ms and all subsequent packets within 5 ms. The STATUS stage
must then be completed within 2 ms.
The UDC performs a zero data transfer write/read transaction on the application bus.
The application itself, i.e. USBD control has to identify this STATUS stage of the Control
Read/Write by a change of read/write direction and has to respond to this transaction by
proper handshake.
Vendor/Class Requests
All Vendor/Class requests will be forwarded to the application bus and handled by SW.
Data Sheet
325
2001-02-23
C165UTAH
USB Interface Controller
The 8-byte SETUP request will be captured in the SETUP registers. IN/OUT transactions
from the host will be NACK’d until the SW is ready. The SETUP registers cannot be
overwritten by the host until all 4 registers have been read.
Requests to device, interface or endpoints with no DATA stage have to be successfully
completed (incl. STATUS stage) within 2 ms. Requests with DATA stage require the first
data packet within 10 ms and all subsequent packets within 5 ms. The STATUS stage
must then be completed within 2 ms.
The UDC performs a zero data transfer write/read transaction on the application bus.
The application itself, i.e. USBD control has to identify this STATUS stage of the Control
Read/Write by a change of read/write direction and has to respond to this transaction by
proper handshake.
Load Configuration Data
After power-up the UDC has to be loaded by the SW via control endpoint#0 Transmit
FIFO. The only buffers the UDC maintains will be the 17 EndPtBufs one for each physical
endpoint. The number of bytes written by the SW then equals 5x17, i.e. 85 bytes.
The first strobe will load the first byte into the MSB byte of EndPtBuf0(39:32) of
endpoint#0 and so on.
Latency Considerations
The latency time is considered to be the minimum/maximum accumulated time between
two EPEC transfers. Since the EPEC transfer is injected into the decode pipeline, the
latency time is determined by the previous commands which have already entered the
FETCH and DECODE stage of the pipeline.
The USB data rate for a packet to be processed by the application is 12 Mbit/s. Since the
UDC samples 8 bits before the byte is transferred over the application bus, the FIFOs
are either read or written every 1.333 µs to provide a full word access to/from the XBUS
interface. The first EPEC transfer at the start of a packet transfer over the XBUS, i.e. its
initial latency time, may be delayed by up to 4 * 1.333 µs (i.e. 5.33 µs) since each FIFO
is 8-byte deep.
Worst case: The worst case EPEC interrupt response time including external accesses
will occur, when instructions N and N+1 are executed out of external memory,
instructions N-1 and N require external operand read accesses and instructions N-3, N2 and N-1 write back external operands. In this case the EPEC response time is the time
to perform 7 word bus accesses.
with fCPU @24MHz (TCL=21 ns) and 1 external waitstate:
= 7 * ext. bus accesses with 1 waitstate + 2 * states + 1 * ext. access for src/dest. pointer
= 7 * (4 * TCL + 2 * TCL) + 2 * (2* TCL) + (4 * TCL + 2 * TCL)
= 52 * TCL = 1.092 µs
Data Sheet
326
2001-02-23
C165UTAH
USB Interface Controller
with fCPU @36MHz (TCL=13.89 ns) and 1 external waitstate:
= 7 * ext. bus accesses with 1 waitstate + 2 * states + 1 * ext. access for src/dest. pointer
= 7 * (4 * TCL + 2 * TCL) + 2 * (2 * TCL) + (4 * TCL + 2 * TCL)
= 52 * TCL = 722.3 ns
Best case: When instructions N and N-1 are executed out of external memory, but all
operands for instructions N-3 through N-1 are in internal memory, then the EPEC
response time is the time to execute 1 word bus access plus 2 state times (2*TCL).
with fCPU @24MHz (TCL=21ns) and 1 external waitstate:
= 1* ext. bus accesses with 1 waitstate + 2 * states
= 1* (4 * TCL + 2 * TCL) + 2 * (2* TCL)
= 10 * TCL = 210 ns
with fCPU @36MHz (TCL=13.89 ns) and 1 external waitstate:
= 1* ext. bus accesses with 1 waitstate + 2 * states
= 1* (4 * TCL + 2 * TCL) + 2 * (2 * TCL)
= 10* TCL = 138.9 ns
Once a request for EPEC has been acknowledged by the CPU, the execution of the next
instruction is delayed by 2 state times plus the additional time it might take to fetch the
source operand from internal RAM or external memory and to write the destination over
the external bus in an external program environment.
A bus access in this context also includes delays by an external READY signal or by bus
arbitration (HOLD mode).
Suspend and Host Resume Support
The UDC has built in counters, to count 6 ms idle time on the USB, which detects a
Suspend and 3 ms counter to send the RemoteWakeup sequence to the host. The
UDC_Suspend_Set interrupt will gate the indicate the Suspend mode to the application.
In response to this signal Software is supposed to turn off the clock of all the modules
which are not used at that moment in order to support a low power consumption. The
USB interface module clock can be switched off by SW using the USBCLC register.
The RemoteWakeup detection will deassert the UDC_Suspend for at least 20 ms, which
will generate a second interrupt UDC_Suspend_Off. The application will restart the UDC
and XBus clock supply as soon as possible.
Note: The normal suspend_off interrupt can not be generated if the USB clock is
switched off. In this case, the fast external interrupt EX5INT (firq(5)) must be used
instead.
Data Sheet
327
2001-02-23
C165UTAH
USB Interface Controller
Suspend and Remote Resume Support
After entering the Suspend state, the UDC may start a RemoteWakeup sequence on the
USB, after the device has enabled the UDC clock supply. After the PLL VCO frequency
has been stabilized the DEV_Resume signal can be asserted which causes the UDC to
drive a resume (K-state) to the USB for 3 ms.
The UDC provides a DEVICE_REMOTE_WAKEUP feature bit in the EndpointInfo block
of the UDC core. At power-on reset this bit is disabled and on a SetFeature command to
the device, this bit can be set by the host. The state of this bit is reflected on the
STATUS2 register.
15.4
USB Interface Controller (USBD) Architecture
D-
PLL
Protocol
Layer
SIE
UDC
End
Point 0
Control,
Arbiter &
MUX
EndPoint Info
Application
Bus Interface
Figure 106
15.5
BPI
UBL
SMIF
D+
Full-Speed USB I/Os
The USB Interface Controller architecture is shown in Figure 106 below. The UDC
provides the 48MHz PLL for full-speed clock & data recovery, Serial Interface Engine
(SIE) and the USB Bridge Layer Block. The UDC is connected with its Application Bus
Interface to the Standard Module Interface (SMIF), which itself is hooked onto the BUS
Peripheral interface (BPI).
XBUS
USB Interface Controller Architecture
Endpoint Info Block
The EPINFO Block is a configurable block which will be programmed at compile time .
Since the GetDescriptor command is forwarded to the Application, the EPINFO provides
only the registers used by the Application to download application specific configuration
data. The EPINFO block features in summary are:
– DATA0/DATA1 Synchronization
– EndPtStalled Bit support
Data Sheet
328
2001-02-23
C165UTAH
USB Interface Controller
– Two different Configurations, with
- Configuration #1 with 4 Interfaces (three interfaces with 3 alternate settings, one
interface with 2 alternate settings) and each interface with maximal seven unique
endpoints where each endpoint can be assigned to one interface only.
- Configuration #2 with 2 Interfaces, each Interface with 2 alternate settings and 2
Endpoints for each setting
– Address pointer and length support for 2 strings
– EndPtBuf information for each of the 6 currently active non-Control Endpoints
All these registers are loaded under SW control as part of the Configuration Process at
Power-On time.
Examples for Configurations, Alternate Settings and Endpoints supported are shown
below.
Config 0
EP 0
Figure 107
USB Configuration0 with Control Endpoint 0 supported.
Data Sheet
329
2001-02-23
C165UTAH
USB Interface Controller
Config 1
IF 0
AS 0
EP 0
IF 1
AS 1
AS 2
EP 1 EP 2
EP 1
AS 1
AS 2
EP 3 EP 4
EP 3
AS 0
IF 2
AS 0
IF 3
EP 5
Figure 108
AS 0
AS 1
AS 1
EP 6
EP 7
Example for USB Configuration1 with Alternate Settings and
Endpoints
Config 2
IF 0
AS 0
EP 0
Figure 109
Data Sheet
IF 1
AS 0
AS 1
AS 1
EP 3
EP 1 EP 2
EP 4
Example for USB Configuration2 with Alternate Settings and
Endpoints
330
2001-02-23
C165UTAH
USB Interface Controller
EndPtBuf’s
The endpoint configuration data file structure shown in the figures above, which will be
downloaded into the USB endpoint info block is summarized in Table 73.
Table 73
Endpoint, Configuration and Interface Structure
PhyEP EpNum Ep_Config Ep_
Interface
Ep_AltSetting
Ep_Type Ep_Dir
MaxPktSize
1
0
0
0
0
Control
I/O
64 bytes
2
1
1
0
1
**
I/O
1023 bytes
3
2
1
0
1
**
I/O
1023 bytes
4
1
1
0
2
**
I/O
1023 bytes
5
3
1
1
1
**
I/O
1023 bytes
6
4
1
1
1
**
I/O
1023 bytes
7
3
1
1
2
**
I/O
1023 bytes
8
5
1
2
1
**
I/O
1023 bytes
9
6
1
2
1
**
I/O
1023 bytes
10
5
1
2
2
**
I/O
1023 bytes
11
7
1
3
1
**
I/O
1023 bytes
12
0
2
0
1
**
I/O
1023 bytes
13
1
2
0
1
**
I/O
1023 bytes
14
2
2
0
2
**
I/O
1023 bytes
15
1
2
1
1
**
I/O
1023 bytes
16
3
2
1
1
**
I/O
1023 bytes
17
4
2
1
2
**
I/O
1023 bytes
18
3
2
1
2
**
I/O
1023 bytes
The EndPtBuf’s format is shown in Table 74..
Table 74
EndPtBuf Format
Bit Field
Type
Description
(39:36)
EpNum
Logical Endpoint number (Endpoint number on host side)
(35:34)
Ep_Config
Configuration number related to this Endpoint
(33:32)
Ep_Interface
Interface number related to this Endpoint
(31:30)
Ep_AltSetting
Alternate Setting related to this Endpoint
Data Sheet
331
2001-02-23
C165UTAH
USB Interface Controller
Table 74
EndPtBuf Format
Bit Field
Type
Description
(29:28)
Ep_Type
Type of EndPoint
’00’ - Control
’01’ - Isochronous
’10’ - Bulk
’11’ - Interrupt
(27)
Ep_Dir
Direction of Endpoint (O: out, I: in)
(26:17)
Ep_MaxPktSize
Maximum packet size for this endpoint in bytes
(16)
reserved
Reserved.
(15:13)
Ep_BufAdrPtr
Reflects as UDC_BufAdrPtr to the Application Bus and
indicates the endpoint number.
(12:0)
reserved
Reserved. Must be set to ’0’.
Configuration Buffer
Since the UDC is forwarding the GetDescriptor Command to the Application, the UDC
does not maintain Configuration Buffer space within the UDC core.
String Buffers
Since the UDC is forwarding the GetDescriptor Command to the Application, the UDC
does not maintain String Buffer space within the UDC core.
15.6
USB Microprocessor Registers
The USB register set provides Special Function Registers (SFR’s) for eight receiving
Endpoints and eight transmitting Endpoints. In addition, Interrupt and Status SFRs will
be provided for fast data processing. Each individual endpoint’s Transmit and Receive
FIFO can be accessed via XBUS. Transmit and Receive FIFO interrupts will be
generated for each endpoint.
Note: Please note, that the SETUPxx and RXRR registers are one time read-only
registers. Multiple read operations without the approriate handshake request will
provide unconsistent data.
Data Sheet
332
2001-02-23
C165UTAH
USB Interface Controller
The USBD register set is shown in Table 75. The Base Address is 00EE00H.
Table 75
USBD Register Set
00EE00H + ... Register
Function
00H
USBCLC
USB clock control register.
08H
USBD_ID
USB peripheral identification register, set to
ZERO in the current version.
10H
USBD_CMD_REG
Command register
12H
USBD_STATUS_REG0
USB endpoint FIFO status
14H
USBD_STATUS_REG1
USB endpoint FIFO handshake control
16H
USBD_STATUS_REG2
USB Device Remote Wake-Up Feature Status
18H
reserved
1AH
reserved
1CH
reserved
1EH
reserved
20H
reserved
22H
reserved
24H
USBD_TIME_REG
USB timestamp info: Frame number of the
transmitted frame
26H
USBD_SETUP_REG01
USB setup bytes (1:0)
28H
USBD_SETUP_REG23
USB setup bytes (3:2)
2AH
USBD_SETUP_REG45
USB setup bytes (5:4)
2CH
USBD_SETUP_REG67
USB setup bytes (7:6)
2EH
USBD_TXWR0
USB Transmit FIFO data register
30H
USBD_TXEOD0
EPEC/SW End-of-packet indication for USBD.
32H
RR0
USB Receive FIFO data register
34H
USBD_RX_BYTECNT0
USB receive packet length in bytes
36H
USBD_TXWR1
USB Transmit FIFO data register
38H
USBD_TXEOD1
EPEC/SW End-of-packet indication for USBD.
3AH
USBD_RXRR1
USB Receive FIFO data register
3CH
USBD_RX_BYTECNT1
USB receive packet length in bytes
3EH
USBD_TXWR2
USB Transmit FIFO data register
40H
USBD_TXEOD2
EPEC/SW End-of-packet indication for USBD.
42H
USBD_RXRR2
USB Receive FIFO data register
44H
USBD_RX_BYTECNT2
USB receive packet length in bytes
46H
USBD_TXWR3
USB Transmit FIFO data register
Data Sheet
333
2001-02-23
C165UTAH
USB Interface Controller
Table 75
USBD Register Set
00EE00H + ... Register
Function
48H
USBD_TXEOD3
EPEC/SW End-of-packet indication for USBD.
4AH
USBD_RXRR3
USB Receive FIFO data register
4CH
USBD_RX_BYTECNT3
USB receive packet length in bytes
4EH
USBD_TXWR4
USB Transmit FIFO data register
50H
USBD_TXEOD4
EPEC/SW End-of-packet indication for USBD.
52H
USBD_RXRR4
USB Receive FIFO data register
54H
USBD_RX_BYTECNT4
USB receive packet length in bytes
56H
USBD_TXWR5
USB Transmit FIFO data register
58H
USBD_TXEOD5
EPEC/SW End-of-packet indication for USBD.
5AH
USBD_RXRR5
USB Receive FIFO data register
5CH
USBD_RX_BYTECNT5
USB receive packet length in bytes
5EH
USBD_TXWR6
USB Transmit FIFO data register
60H
USBD_TXEOD6
EPEC/SW End-of-packet indication for USBD.
62H
USBD_RXRR6
USB Receive FIFO data register
64H
USBD_RX_BYTECNT6
USB receive packet length in bytes
66H
USBD_TXWR7
USB Transmit FIFO data register
68H
USBD_TXEOD7
EPEC/SW End-of-packet indication for USBD.
6AH
USBD_RXRR7
USB Receive FIFO data register
6CH
USBD_RX_BYTECNT7
USB receive packet length in bytes
6EH
USBD_CFGVAL
Current Configuration & Alternate Setting
selected by Host
70H
USBC_CMD_RESET
USB Block Reset
The detailed register description is shown below.
Data Sheet
334
2001-02-23
C165UTAH
USB Interface Controller
USBCLC
Reset Value: 0000H
Table 76
USBCLC Register
Bit No. Bit
Function
(15:4)
reserved. Always ’0’
3
USBEX_DIS
USB Controller Clock Disable
0: The kernel clock of the USB interface
controller is enabled, normal operation.
1: The kernel clock of the USB interface controller is
disabled.
2
USBGPSEN
USB Controller Clock OCDS Disable
0: The kernel clock of the USB interface
controller is enabled, normal operation.
1: The kernel clock of the USB interface controller is
disabled during debugging mode (OCDS)
1
USBDIS
USB Controller Clock Status
0: The status of the USB interface
controller kernel clock is ’enabled’.
1: The status of the USB interface controller clock
is ’disabled’.
0
USBDISR
USB Controller Clock Disable
0: The kernel clock of the USB interface
controller is enabled, normal operation.
1: The kernel clock of the USB interface controller is
disabled.
The state of the USB interface kernel clock is controlled by the register bit USBDISR.
The actual USB kernel clock state will be shown by the state bit USBDIS.
The register USBCLC is clocked with the bus clock to be able to switch the USB interface
controller clock on again, if it was off. If required, switching off the clock can be prevented
by the USB controller.
For on chip debugging support (OCDS) an additional bit USBGPSEN is introduced to
stop the peripheral clock for arbitrary lengths of time during debugging if this function is
enabled. If debugging mode is active, the peripheral core rejects write access to
registers connected to the peripheral clock.
To be compatible with previous C16x products an USBEX_DISR signal is provided to
disable the peripheral clock.
Data Sheet
335
2001-02-23
C165UTAH
USB Interface Controller
USBD_CMD_REG
Table 77
Reset Value: 0000H
USBD_CMD_REG Command Register
Bit No. Bit
Function
15
Enable HW controlled flushing of TX channel, if IN
transfer is not completed by host, i.e. if a SETUP
transfer is immediately followed by a Status-OUT
transfer. This also clears EPEC channel 0. (For
Control endpoint 0 only. Will not be reset by HW).
Autoflush_Enable
Note: The Autoflush_Enable feature can be used
@ 36 MHz CPU clock only and is NOT
supported @ 24 MHz CPU clock.
14:12
Flush_TX_Channel_Select
Select the channel to be flushed
’xxx’: binary coded 7 downto 0
11
Flush_TX _Channel
Flush the TX channel, i.e. clear FIFO and control state
machine
1: Flush Command active, will be reset by HW
0: no action
10
UDC_Suspend
1: Device is in Suspend Mode
0: Normal Mode: device has initiated Device Resume
proc.
(read-only)
9
DEV_Resume
1: Start RemoteWakeup procedure
0: normal operation
8
USB_TXProtect
1: Enables protection feature of the USB transceiver
in case of short circuit between DPLS/DMNS and
VDDU/VSSU
0: short circuit protection disabled
(7:0)
STALL_EP(7:0)
1: Set Endpoint N (N=7..0) to STALLED
0: normal operation
The USBD command register provides set STALL capability per Endpoint,
RemoteWakeup, Resume and Suspend control for the device.
Data Sheet
336
2001-02-23
C165UTAH
USB Interface Controller
USBD_STATUS_REG0
Table 78
Reset Value: 0000H
USBD_STATUS_REG0 Status Register
Bit No. Bit
Function
15:8
1: Indicates RX Fifo Endpoint N (N=7..0) is empty
0: not empty
RX_EMPTY(7:0)
Note: The reset value is ’0’, but after reset the value
changes immediately.
7:0
TX_FULL(7:0)
1: Indicates TX Fifo Endpoint N (N=7..0) is full
0: normal operation
The USBD status register 0 provides the SW with status indication for RX and TX FIFOs
and provides handshake control for both FIFOs. This register is read-only.
USBD_STATUS_REG1
Table 79
Reset Value: 0000H
USBD_STATUS_REG1 Status Register
Bit No. Bit
Function
15:8
RX_XFR_ACK(7:0)
1: Indicates RX Transfer Acknowledge for USBD
FIFO Endpoint N (N=7..0)
0: idle/busy
7:0
TX_XFR_ACK(7:0)
1: Indicates TX Transfer Acknowledge for USBD FIFO
Endpoint N (N=7..0)
0: idle/busy
The USBD status register 1 provides the SW with internal handshake indication for RX
and TX FIFOs. This register is read/write (writing ’1’ resets the bit).
Note: RX_XFR_ACK are reflected in the corresponding RX_BYTECNT registers.
USBD_STATUS_REG2
Table 80
Reset Value: 0000H
USBD_STATUS_REG2 Status Register
Bit No. Bit
Function
0
DEV_REM_WAKEUP_FEAT
1: Indicates that the DeviceRemoteWakeup Feature
has been set by the host
0: DeviceRemoteWakeup Feature not allowed
15:1
Reserved
reserved
Data Sheet
337
2001-02-23
C165UTAH
USB Interface Controller
This register is read-only.
USBD_TIME_REG
Table 81
Reset Value: 0000H
USBD_TIME_REG Register
Bit No. Bit
Function
(15:11)
reserved. Always ’0’
(10:0)
FRAME_NUMBER
Actual frame number
The USB Time register provides the actual 11-bit frame number information of the actual
received start of frame packet.
USBD_Setup_REG01
Table 82
Reset Value: 0000H
USBD_Setup_REG01 Register
Bit No. Bit
Function
(15:8)
Setup_Byte1
Byte 1 of SETUP command
(7:0)
Setup_Byte0
Byte 0 of SETUP command
The USB Setup_REG01 register provides setup byte(1:0) of the SETUP command from
the host. The register can not be overwritten by the host until the last setup register (67)
is read by the device.
USBD_Setup_REG23
Table 83
Reset Value: 0000H
USBD_Setup_REG23 Register
Bit No. Bit
Function
(15:8)
Setup_Byte3
Byte 3 of SETUP command
(7:0)
Setup_Byte2
Byte 2 of SETUP command
The USB Setup_REG23 register provides setup byte(3:2) of the SETUP command from
the host. The register can not be overwritten by the host until the last setup register (67)
is read by the device.
Data Sheet
338
2001-02-23
C165UTAH
USB Interface Controller
USBD_Setup_REG45
Table 84
Reset Value: 0000H
USBD_Setup_REG45 Register
Bit No. Bit
Function
(15:8)
Setup_Byte5
Byte 5 of SETUP command
(7:0)
Setup_Byte4
Byte 4 of SETUP command
The USB Setup_REG45 register provides setup byte(5:4) of the SETUP command from
the host. The register can not be overwritten by the host until the last setup register (67)
is read by the device.
USBD_Setup_REG67
Table 85
Reset Value: 0000H
USBD_Setup_REG67 Register
Bit No. Bit
Function
(15:8)
Setup_Byte7
Byte 7 of SETUP command
(7:0)
Setup_Byte6
Byte 6 of SETUP command
The USB Setup_REG67 register provides setup byte(7:6) of the SETUP command from
the host. The register can not be overwritten by the host until the last setup register (67)
is read by the device.
Note: The Transmit/Receive handling by SW and EPEC is controlled by register groups
per endpoint in address range 24H to 6EH. The register group description is shown
in detail for one endpoint and will be repetitive for the others.
USBD_TXWRn (n=7..0)
Table 86
Reset Value: 0000H
USBD_TXWRn Transmit Data FIFO EP#n Register
Bit No. Bit
Function
(15:0)
16-bit word for Transmit FIFO endpoint#n (n=7..0)
TXWRn
The USB TXWR0 register provides 16-bit write access to the Endpoint#n Transmit FIFO.
Data Sheet
339
2001-02-23
C165UTAH
USB Interface Controller
USBD_TXEODn (n=7..0)
Table 87
Reset Value: 0000H
USBD_TXEODn Transmit End-of-packet Register
Bit No. Bit
Function
(15:1)
reserved
0
TXEOD
’1’: indicates last byte of packet transferred to Transmit
FIFO endpoint#n (n=7..0)
’0’: EPEC/SW is idle/busy
The USB TXEODn register provides 16-bit read/write access. Indicates that the EPEC/
SW has sent the last valid data byte to the Endpoint’s#n Transmit FIFO. Writing ’1’ to the
register makes the next request for a data packet from Host answered by C165UTAH
with an empty data packet.
As long as a ’1’ is stored in the register, the packet has not been transfered to the host.
If transferred, the bit is cleared.
Note: If bit TXEOD is written at the time an IN request from the host - due to missing
transmit data - is answered by a NACK packet, bit TXEOD is cleared and the
tx_done interrupt is asserted although the NACK packet has been sent instead of
the empty packet.
Therefore, prior to setting the TXEOD bit, Software must reset the endpoint’s
acknowledge-bit TX_XFR_ACKn in register USBD_STATUS_REG1. In reaction
to endpoint’s tx_done interrupt, Software always should check the value of
TX_XFR_ACKn to verify that the previous data transfer succeeded. If
TX_XFR_ACKn indicates an unsuccessful data transmission, Software must
repeat the transfer.
USBD_RXRRn (n=7..0)
Table 88
Reset Value: 0000H
USBD_RXRRn Receive Data FIFO EP#n Register
Bit No. Bit
Function
(15:0)
16-bit word for Receive FIFO endpoint#n (n=7..0)
RXRRn
The USB RXRR0 register provides 16-bit read access to the Endpoint#n Receive FIFO.
Data Sheet
340
2001-02-23
C165UTAH
USB Interface Controller
USBD_RX_BYTECNTn (n=7..0)
Table 89
Reset Value: 0000H
USBD_RX_BYTECNTn Byte Counter Receive FIFO EP#n Register
Bit No. Bit
Function
15
RX_STATUS
packet status indication
’1’: status ok
’0’: packet error
(14:10)
reserved
always ’0’
(9:0)
RX_BYTECNT
10-bit byte counter for received packet in
endpoint#n (n=7..0)
The USB BYTE_CNTn register provides 16-bit read access to the Endpoint#n Receive
FIFO’s 10-bit byte counter after a complete package is available. The status of the
packet is indicated by bit 10 (see USB_STATUS_REG1).
USBD_CFGVAL_REG
Table 90
Reset Value: 0000H
USBD_CFGVAL_REG Command Register
Bit No. Bit
Function
15:10
Reserved (always ’0’)
9:8
AS_IF3
Alternate Setting selected for Interface 3:
’xx’: Alternate Setting 3-0 binary coded.
(read-only)
7:6
AS_IF2
Alternate Setting selected for Interface 2:
’xx’: Alternate Setting 3-0 binary coded.
(read-only)
5:4
AS_IF1
Alternate Setting selected for Interface 1:
’xx’: Alternate Setting 3-0 binary coded.
(read-only)
3:2
AS_IF0
Alternate Setting selected for Interface 0:
’xx’: Alternate Setting 3-0 binary coded.
(read-only)
1:0
CFG
Configuration selected by host
’xx’:Configuration 2-0 binary coded.
(read-only)
The USBD CFGVAL register provides the current selection of the configuration and
alternate setting done by the host. The SET_CONFIGURATION resets all settings to
Alternate Setting 0, i.e. the Control Endpoint 0.
Data Sheet
341
2001-02-23
C165UTAH
USB Interface Controller
USBC_CMD_RESET
Table 91
Reset Value: 0000H
USBC_CMD_RESET USB RESET REGISTER
Bit No. Bit
Function
(15:1)
Reserved
reserved
(0)
USBC_RST
Resets the USB block including the transmit and receive logic.
Note:
1. The USB RESET does not reset the USBCLC register.
2. The USB reset must be active for at least 10 clock cycles.
3. The USB reset must always be activated after the USB
device is connected to the USB
15.7
Programmers Guidlines: Using USB and EPEC
For normal functionality, the following interrupts must be enabled:
all udc_tx_done- and udc_rx_done-interrupts of those endpoints and in the direction in
which they are used
udc_setup, udc_suspend, udc_suspendoff, udc_start of frame, udc_configval
optional interrupts are the EPEC-interrupt (this interrupt might only be used if SW needs
to have complete control over the contents of the fifos and always wants to keep track of
the status of the transmission) and the UDC_txwr- and UDC_rxrr-interrupts (which are
generated every time, a fifo in the USB-block is ready to accept data or receive-data
might be read from the fifo; this interrupt is connected to the EPEC which in nomal
functionality does this transfer).
By default, the bits 8 and 15 of the CMD register (auto flush enable and tx_protect)
should be set.
15.7.1
Writing the configuration-value
After a system-reset, the USB-block expects the configurator of the UDC to be
transferred via the tx-fifo(0) like a normal in-transfer. Best is using the EPEC for this
endpoint. The source-pointer must be programmed to a memory-block containing the 85
bytes of the configurator, destination to usbd_txwr-register(0) and the length-register to
85 bytes. After starting the EPEC, it will transfer the configurator to the USB-block and
after having transferred all the data, the EPEC-interrupt is generated and if the
configurator went through the fifo, the udc_tx_done-interrupt(0) is also generated.
In order to allow generation of a next interrupt-pulse of the EPEC, SW must read the
EPEC-interrupt-register and clear the interrupt-bit of endpoint 0 by writing a ‘1’ to it.
Data Sheet
342
2001-02-23
C165UTAH
USB Interface Controller
15.7.2
In-Transfer (Transmit)
An In-transfer means that the device will send data to the host by using endpointX. This
transfer is started by a bulk_in, interrupt, control_in or iso_in request from the host.
• SW sets up source-pointer (to a memory-block containing the data to be sent), a
destination-pointer (usbd_txwr-registerX) and the packetlenght (usually
maxpacketlength) of the EPEC for the endpoint on which the transfer is to be done;
the packetlength must always be the maxpacketlength for this endpoint for interrupt-,
bulk_in or control_in-transfers except for the last packet to be transferred, for
isochronous_in-endpoints this endpoint must be according to the sequence of
packetlengths and the data which is available to be sent.
If SW wants to send a packet of zero bytes, it must not use the EPEC, but must write
‘1’ to the corresponding usbd_tx_eod-register of the USB
• SW writes EPEC-start, EPEC transferres the data from the memory to the USB-fifoX;
when host requests for the data, it is transferred through the fifo to the host; each time,
the USB-block has space in the fifo and may accept a write into the usbd_txwrregister, the udc_txwr-interrupt is generated (for normal functionality this can be
ignored as the EPEC uses this as a handshake for the next transfer)
• when EPEC has transferred all the data, it generates the EPEC-interrupt; SW must
read the EPEC-interrupt-register and clear the (to the endpoint) corresponding bit by
writing a ‘1’ to it
• when the transfer over the USB is finished, the udc_tx_done-interruptX is generated
and SW can check the corresponding bit in the Status-Register, wheather the transfer
was successful or not; this is for bulk, interrupt and control-Endpoints only, not for
Isochonous Endpoints (where no ACK will be sent as Handshake)
• for non-Iso-transfers: if fhe transfer was ACK’d, the next packet can be set up for
transmission, otherwise, host expects the same data to be resent
• SW must set up again the EPEC source-, destination-pointer and packetlength and
start the transfer
If SW has already set up data in a tx-fifo and now, e.g. host changes the configuration
or interfaces, SW can use a write into the command-register to flush the fifo of the
corresponding endpoint. Before doing this, the EPEC-channel must be disabled or
reprorgammed, otherwise the next pending bytes will be transferred into the tx-fifo.
15.7.3
Out-Transfer (Receive)
During an Out-transfer, host is transferring data to the device. SW must provide an free
memory-block for each endpoint and set up the EPEC for moving arriving data from the
USB-block to a free memory-location.
• SW provides an free memory-block and sets up source- (usbd_rxrr-registerX),
destination-pointer (free memory-block) and packet lenght of the EPEC and sets the
TXR_ENAx bit (refer to Table 20, “EPEC_CTRL_REGx Source Pointer Register,” on
Data Sheet
343
2001-02-23
C165UTAH
USB Interface Controller
•
•
•
•
page 99); the packetlength supported by the EPEC must always be an even number
of bytes (in receive-direction the EPEC only does word-transfers) and have at least
space for the maxpacketlength of the endpoint.
when host sends data, it is forwarded through the fifo’s and with every transferred
word that can be read from usbd_rxrr-registerX the usbd_rxrr-interruptX is set (for
normal functionality this can be ignored); EPEC transferes this data into the memory
EPEC generates an EPEC-interrupt when it has transferred all the data into the
memory (this interrupt is generated shortly after the USB-block has generated the
udc_rx_done-interrupt); SW must read the EPEC-interrupt-register and clear the (to
the endpoint) corresponding bit in this register by writing a ‘1’ into it
when USB-block has finished the whole transfer, it generates the udc_rx_doneinterrupt; SW must then read the usbd_rx_bytecnt-register, in order to determine the
number of bytes of the received packet and to release the interlocking of the fifo for
the next transfer; the most significant bit in this register contains also the status-bit of
the status-register and shows whether this packet had transmission-errors or not
if host has sent a packet of zero lenght, no rxrr-interrupt is generated but only a
rx_done-interrupt; here also SW must read the rxbytecountX-register
in order to prepare a receive on this endpoint again SW must provide a new free
memory-block, set up the source-, destination-pointer and packetlength of the EPEC
and write the start-bit
15.7.4
Reading out Setup-Packets
Setup-packets are treated without the EPEC. If host sends a setup-packet which is
forwarded to the CPU (there are only three commands: get_descriptor, set_descriptor
and synch_frame all the other ones are treated internally), and the packet is valid, the
udc_setup-interrupt is generated and SW must read all four setup-registers. By reading
the last one (setup 61, an overwrite-protection for those registers will be released and
the device will accept the next setup-packet. As long as this register is not read, each
setup-packet on the USB will be NAK’d.
15.7.5
Special case: Setup-Transfer
The endpoint zero has an additional feature in transmit-direction in order to handle an
early end of a setup-transaction. If, as an example, host requests for a device-desciptor
(by sending a get_descriptor(device_descriptor)) with a length of 12 bytes and the
maxpacketlength of endpoint zeo is eight, host must request for two control_in-packets
and then accnwowledge them with an status_out-packet. According to the USB-spec,
host can also request only one packet or even none before it sends out the statuspacket. This is called an early end of the setup-transaction.
If SW has already set up the data in the tx-fifo after the get_descriptor-command and
host will immediately after this request for different data with another setup-packet, the
old data, already in the fifo would be sent. By enabling the AutoFlushEnable-feature, with
Data Sheet
344
2001-02-23
C165UTAH
USB Interface Controller
every setup-packet which is visible for the CPU will flush the tx-fifo for endpoint zero. This
will avoid wrong, old data to be sent over control-enpoint zero.
This flush-mechanism could also be done by flushing the usb with a write into the
command-register of the USB-block but the flush initiated by SW might happen too late.
Note: The AutoFlushEnable-feature described above is only available if the C165UTAH
CPU is running with 36 MHz. This feature can not be used if the device is running
with 24 MHz. In this case, pending data has to be flushed explicitly by Software
via register USBD_CMD_REG on reception of the OUT packet that ends setup
transfer, indicated by the rx_done interrupt of endpoint 0.
15.7.6
Setting of configuration and alternate settings of interfaces
Each time the host send a valid set_configuration- or set_interface-command, this will
show up for the CPU as a configval-interrupt. In order to determine the actual
confitguration and the alternate setting of an interface, SW must read the configvalregister and set up the endpoints which are actually enabled.
There is no overwirte-protection on this register, the value is always updated if a valid
set_configuration or set_interface-packet is received.
15.7.7
Stalling Endpoints
All transmit-endpoints can be stalled by writing into the command-register, where it
makes no sense to stall an isochronous endpoint (for a isochronous packet there is no
stall-handshake, so the host will never notice that the endpoint is stalled and thus will
never try to abolish the stall-condition). The stall will be kept as long the bit in the
command-register is set and the host did not send a clear_feature-command.
If an endpoint was stalled during a in-transfer, this transfer will be finished and the next
request by the host will return a stall-handshake. If there is data in the fifo for transission,
the data will be kept and sent, if the stall-condition is abolished and host requests for this
data.
A stall-condition during an out-transfer will finish first and the next request will return a
stall-handshake.
Endpoint zero has only one stall-bit for both directions, for in and out. A stall on one
direction will also lead to a stall in the other transfer-direction.
15.7.8
Start of Frame
Host sends a Start of Frame-(SOF)-signal every ms. A SOF-interrupt is generated and
the value of the actual frame-number is stored in the SOF-register. There is no overwirteprotection on this register, the value is always updated if a valid SOF-packid is received.
Data Sheet
345
2001-02-23
C165UTAH
USB Interface Controller
15.7.9
Suspend and Suspendoff
In normal operation, there is the UDC-clock of 48 MHz enabled and the normal CPUclock which may vary according to the divider in the clock-generation-unit.
If the host is sending a suspend-request (by driving an idle-state for more than 6 ms),
after 6 ms the suspend-interrupt will be generated. This must cause SW to go in lowpower mode. There are different modes in which the chip can be set. According to the
mode the wakeup initiated by the host, must be detected differently:
• Using the bit 0 in clc register of every peripheral to turn off the clock. The suspendoff
interrupt is generated even though the rest of the usbblk is turned off.
• Using the SYSCON3 register to turn off the clock of xbus and pdbus peripherals
(peripheral disable only). The SYSCON3 register is a write protected register and SW
first must go into low protected mode to be able to do this (see page 470 for
SYSCON3 register description). In this mode also, the suspendoff interrupt is
generated.
• Using the SYSCON3 register group disable (msb of the register) to turn off all the xbus
and pdbus peripherals. In this mode, the normal suspendoff interrupt is not generated,
wakeup must be done with the falling edge of the fast external interrupt alternate
function firq_alt(5).
• Going into sleep mode which stops program execution and turns off the clock for most
of the entire chip. In this mode the fast external interrupt alternate function firq_alt(5)
is also generated and will wakeup the cpu. Program execution will start with the
interrupt procedure of the interrupt, or, if SW was in an interrupt routine with a higher
priority before, program execution will continue at the point, it was stopped.
If SW wants to send a device-wakeup this feature must have been enabled by the host.
Whether this feature was enabled or not, is reflected in the STATUS3 register. If this
feature is enabled, and SW wants to wake up the USB, it must turn on the clocks and
write the resume-bit of the command-register. This will drive the non-idle-state on the
USB for 3 ms, host will start a wakeup-procedure.
15.7.10
Device disconnecting
Either our device is bus- or self-powered. In the case of being bus-powered, every time
the device is disconnected from the bus, the power supply will break down and a replugging will restart with a reset of the entire chip.
In case of a self-powered device, if there is a disconnection from the USB, the logic of
the usbblk must be reset. There must be external logic added to provide the detection of
a disconnection. If SW detects a dis- and reconnection, it must disable all the epec
channels and reset the whole usbblk by writing into the usbd_cmd_reset register. After
de-asserting the reset, the whole configuration process (with writing of the configurator)
must be redone.
Data Sheet
346
2001-02-23
C165UTAH
IOM-2 Interface Controller
16
IOM-2 Interface Controller
The C165UTAH supports the IOM-2 interface with single/double bitrate clock in Terminal
(TE) mode, as well as the Linecard (LT) and PCM mode. The IOM-2 interface consists
of four lines: the inputs FSC and DCL as well as the bidirectional pins DD and DU.
Note: In the C165UTAH the bitrate clock (BCL) is driven on the DCL (double bitrate
clock) line. In this document the term ’BCL’ is used to indicate the single bitrate
clock.
The rising edge of FSC indicates the start of an IOM-2 frame. Typically, the FSC signal
is generated by the upstream transceiver, which synchronizes to the received framing
indication. The DCL input clock signal synchronizes the data transfer on both data lines.
For double bitrate clock the bits are shifted out with the rising edge of the first DCL clock
cycle and sampled at the falling edge of the second clock cycle. For single bitrate clock
the bits are shifted with the rising edge of the BCL and sampled with the falling edge of
the BCL.
16.1
IOM-2 and PCM Frame Structure
16.1.1
Definitions
The following definitions are used throughout the documentaion of the IOM-2 Interface
Controller.
16.1.1.1 Frame Structure
The IOM-2 frame structure is defined as:
1. A start of a frame is detected by the rising edge of the FSC signal. The length of the
FCS signal being high is irrelevant for the IOM-2 unit, as long as it comprises with the
signal specification.
2. The IOM-2 module synchronizes automatically to the frame length (i.e. number of bits/
octets) generated by the DCL signal.
3. The CLKM bit of the IOM_CR register specifies whether the DCL is a single or double
bitrate clock.
4. A frame is always 125 µs on IOM.
5. A frame always consists of multiple of 8 bit time slots.
6. The IOM-2 module does not distinguish between the different frame modes (PCM,TE,
LT). Each frame structure can be configured by specifying the corresponding time slot.
Table 92 shows a summary of design parameters for different frame structures
(e.g.Terminal Mode, the Lincard Mode and the PCM Mode).
Data Sheet
347
2001-02-23
C165UTAH
IOM-2 Interface Controller
Table 92
IOM-2 parameters for different modes (TE, LT, PCM)
min
max
TE mode
LT mode
PCM mode
octets per FSC
1
32
12
n·4,
n ∈{1,2,3,..,8}
m ∈{1,2,3,..,32}
number of bits
per FSC
8
256
96
n·4·8,
n ∈{1,2,3,..,8}
m·8,
m ∈{1,2,3,..,32}
FSC clock rate
8 kHz 8 kHz
8 kHz
8 kHz
8 kHz
BCL clock rate
(FSC · bits)
64
kHz
2048
kHz
768 kHz
n·256 kHz,
n ∈{1,2,3,..,8}
m·64 kHz,
m ∈{1,2,3,..,32}
DCL clock rate
(2·BCL)
128
kHz
4096
kHz
1536 kHz
n·512 kHz,
n ∈{1,2,3,..,8}
m·128 kHz,
m ∈{1,2,3,..,32}
16.1.1.2 HDLC Channels on the IOM-2: B1, B2, D1, D2
The IOM-2 Interface Controller module contains two 8-bit HDLC channels and two 2-bit
HDLC channels (bit 0 and 1 of an octet). These channels can be configured to any of the
32 IOM-2 time slots. The main use for these HDLC Channels is: B1 and B2 for the 8-bit
channels and D1 and D2 for the 2-bit channels. Therefore the terms B1, B2, D1, and D2
are used throughout this document.
16.1.2
PCM Mode
The frame structure in PCM mode consists of 1..32 individual octets. This is a bit rate of
64..2048 kbit/s in steps of 64 kbit/s.
All the IOM-2 channels (e.g. B1, B2, D1, D2, CDA) can be individually mapped to any of
the 32 timeslots. Therefor any other frame mode type can be configured.
Note: There is no protection or checking for exclusive use of one timeslot by different
IOM-2 channels. This must be insured by the SW.
16.1.3
Terminal Mode
The frame structure on the IOM-2 data ports (DU,DD) in IOM-2 terminal mode is shown
in figure 110.
Data Sheet
348
2001-02-23
C165UTAH
IOM-2 Interface Controller
Figure 110
IOM“-2 Frame Structure in Terminal Mode
The frame is composed of three channels
• Channel 0 contains 144-kbit/s of user and signaling data (2B + D1), an optional
MONITOR programming channel (MON0) and an optional command/indication
channel (CI0) for control and programming of a layer-1 transceiver, if any.
• Channel 1 contains two 64-kbit/s intercommunication channels (IC) plus a MONITOR
and command/indicate channel (MON1, CI1) to program or transfer data to other IOM2 devices.
• Channel 1 provides acces to an optional second signaling channel D2 in CI1(7:6)
• Channel 2 can be used for the TlC-bus access. Additionally channel 2 supports further
IC and MON channels.
Note: Each octet related to any integrated functional block can be programmed to any
of the 12 timeslots. Exceptions are the C/I0-channel, which is always related to
channel 0 (i.e. timeslot 3), C/1-channel, which is always related to channel 1 (i.e.
timeslot 7). The Monitor channel can be configured to one of 3 IOM channels in
TM mode (timeslot 2, 6 or 10) or one of 8 IOM channels in LT mode (timeslot 2, 6,
10, 14, 18, 22, 26 or 30).
Data Sheet
349
2001-02-23
C165UTAH
IOM-2 Interface Controller
16.1.4
Linecard Mode
The frame structure on the IOM-2 data ports (DU,DD) in IOM-2 linecard mode is shown
in Figure 111.
125 µ s
FSC
DCL
DD
IOM R CH0
CH1
CH2
CH3
CH4
CH5
CH6
CH7
CH0
DU
IOM CH0
R
CH1
CH2
CH3
CH4
CH5
CH6
CH7
CH0
B1
Figure 111
B2
MONITOR
D
C/I
MM
RX
ITD09635
IOM“-2 Frame Structure in Linecard Mode
The frame is composed of up to eight IOM channels (see also TM mode).
16.2
IOM-2 Handler
The IOM-2 handler offers a great flexibility for handling the data transfer between the
different functional units of the C165UTAH and voice/data devices connected to the
IOM-2 interface. Additionally it provides direct CPU access to all time slots of the IOM-2
interface via the four controller data access registers (CDA). Figure 112 shows the
architecture of the IOM-2 handler. For illustrating the functional description it contains all
configuration and control registers of the IOM-2 handler.
The Controller data access (CDA) can be configured by programming the time slot and
data port selection registers (TSDP). With the TSS bits (Time Slot Selection) the data of
the functional units can be assigned to the available time slots of the IOM-2 frame. With
the DPS bit (Data Port Selection) the output of each functional unit is assigned to either
DU or DD line respectively. The input is assigned vice versa. With the control registers
(CR) the access to the data of the functional units can be controlled by setting the
Data Sheet
350
2001-02-23
C165UTAH
IOM-2 Interface Controller
corresponding control bits (EN, SWAP), see Chapter 16.5, "Controller Data Access
Handler".
The IOM-2 handler provides also access to the
•
•
•
•
•
MONITOR channel (MON)
Command Indication (C/I) channels (CI0,CI1)
TIC bus (TIC)
two 8-bit HDLC data channels (B1, B2)
two 2-bit HDLC data channels (D1, D2)
The access to these channels is controlled by the registers MON_CR, CIC_CR,
IOMSEL_x. The IOM-2 interface is controlled by the control register IOM_CR.
DU DD
IOM-2 Handler
IOM_CR
FSC DCL
IOM-2 Interface
(EN, OD)
DU
DU
DD
DD
TIC Bus
Data
CDA Data
Monitor Data
D1/D2/B1/B2
D1/D2/B1/B2
Data
Data
D1/D2/B1/B2
D1/D2/B1/B2
Data
Data
CI1 Data
CI0 Data
Controller Data
Access (CDA)
CDA
Register
CDA10
CDA11
CDA20
CDA21
.
Control
Data Access
(TSDP, DPS,
EN, SWAP,
MCDA, STI)
Control
Monitor
Data
(DPS,EN
MCS)
CDA_TSDPxy
CDA_CRx
MCDA
STI
MSTI
ASTI
MON_CR
Control
TIC Bus
Disable
Control
CI0
CI1
Data
Data
(DPS, EN) (DPS,EN)
CIC_CR
HDLC-0 HDLC-1 HDLC-2 HDLC-3
FIFO
FIFO
FIFO
FIFO
DPS, EN, TSS
x,y = 1 or 2
MON
TIC
Handler
IOMHAND.DRW
Figure 112
Data Sheet
CI0
CI1
Data
Microcontroller Interface
Architecture of the IOM-2 Handler
351
2001-02-23
C165UTAH
IOM-2 Interface Controller
16.3
IOM-2 Monitor Handler
The IOM-2 MONITOR channel is utilized for information exchange between the
C165UTAH and other devices connected to the MONITOR channel.
The MONITOR channel data can be controlled by the bits in the MONITOR control
register (MON_CR). For the MONITOR data one of the up to 3 (TE) or up to 8 (LT/PCM)
IOM channels can be selected by setting the MONITOR channel selection bits (MCS).
The DPS bit in the same register selects between an output on DU or DD respectively
and with EN_MON the MONITOR functionlity can be enabled/disabled. The default
value is MONITOR channel 0 (MON0) enabled and transmisson on DD. The following
applications may apply.
• The C165UTAH can program and control other devices attached to the IOM-2 which
do not need a microcontroller interface.
• For data exchange between two microcontroller systems attached to two different
devices on one IOM-2 backplane. Use of the MONITOR channel avoids the necessity
of a dedicated serial communication path between the two systems. This simplifies the
system design.
16.3.1
Handshake Procedure
The MONITOR channel operates on an asynchronous basis. While data transfers on the
bus take place synchronized to frame sync, the flow of data is controlled by a handshake
procedure using the MONITOR Channel Receive (MR) and MONITOR Channel
Transmit (MX) bits. Data is placed onto the MONITOR channel and the MX bit is
activated (MX =’0’). This data will be transmitted once per 8-kHz frame until the transfer
is acknowledged via the MR bit (MR=’0’).
The MONITOR channel protocol between a Transmitter µC (µCT) and a Receiver µC
(µCR) is described In the following section and illustrated in Figure 113. The relevant
control and status bits for transmission and reception are listed in Table 93 and Table
94.
Table 93
Transmission of MONITOR Data
Control/
Status Bit
Register
Bit
Function
Control
MOCR
MXC
MX Bit Control
MIE
MDA and MAB Interrupt Enable
MDA
Data Acknowledged Interrupt
MAB
Data Abort Interrupt
MAC
Transmission Active
Status
MOSR
MSTA
Data Sheet
352
2001-02-23
C165UTAH
IOM-2 Interface Controller
Table 94
Reception of MONITOR Data
Control/
Status Bit
Register
Bit
Function
Control
MOCR
MRC
MR Bit Control
MIE
MER Interrupt Enable
MRE
MDR Interrupt Enable
MDR
Data Received Interrupt
MER
End of Reception Interrupt
Status
Data Sheet
MOSR
353
2001-02-23
C165UTAH
IOM-2 Interface Controller
Transmission
Reception
µC (µCT)
MIE=1
MOX=ADR
MXC=1
MAC=1
MDA Int.
MOX=DATA1
MDA Int.
MOX=DATA2
MDA Int.
MXC=0
MAC=0
Figure 113
Data Sheet
µC (µCR)
MON
MX
MR
FF
FF
ADR
ADR
1
1
0
0
1
1
1
1
ADR
ADR
DATA1
DATA1
0
0
1
0
0
0
0
0
DATA1
DATA1
0
0
1
0
DATA2
DATA2
1
0
0
0
DATA2
DATA2
0
0
1
0
FF
FF
1
1
0
0
1
1
1
1
FF
FF
MRE=1
125µs
MDR Int.
RD MOR (=ADR)
MRC=1
MIE=1
MDR Int.
RD MOR
(=DATA1)
MDR Int.
RD MOR
(=DATA2)
MER Int.
MRC=0
MONITOR Channel Protocol (IOM-2)
354
2001-02-23
C165UTAH
IOM-2 Interface Controller
Before starting a transmission, the µCT should verify that the transmitter is inactive, i.e.
that a possible previous transmission has been terminated. This is indicated by a ’0’ in
the MONITOR Channel Active MAC status bit.
After having written the MONITOR Data Transmit (MOX) register, the µCT sets the
MONITOR Transmit Control bit MXC to ’1’. This enables the MX bit to go active (0),
indicating the presence of valid MONITOR data (contents of MOX) in the corresponding
frame. As a result, the receiving device stores the MONITOR byte in its MONITOR
Receive MOR register and generates an MDR interrupt status (MRE must be ’1’).
Alerted by the MDR interrupt, the µCR reads the MONITOR Receive (MOR) register.
When it is ready to accept data (e.g. based on the value in MOR, which in a point-tomultipoint application might be the address of the destination device), it sets the MR
control bit MRC to ’1’ to enable the receiver to store succeeding MONITOR channel
bytes and acknowledge them according to the MONITOR channel protocol. In addition,
it enables other MONITOR channel interrupts by setting MONITOR Interrupt Enable
(MIE) to ’1’.
As a result, the first MONITOR byte is acknowledged by the receiving device setting the
MR bit to ’0’. This causes a MONITOR Data Acknowledge MDA interrupt status at the
transmitter.
A new MONITOR data byte can now be written by the µCT in MOX. The MX bit is still in
the active (0) state. The transmitter indicates a new byte in the MONITOR channel by
returning the MX bit active after sending one frame in the inactive state. As a result, the
receiver stores the MONITOR byte in MOR and generates a new MDR interrupt status.
When the µCR has read the MOR register, the receiver acknowledges the data by
returning the MR bit active after sending one frame in the inactive state. This in turn
causes the transmitter to generate an MDA interrupt status.
This "MDA interrupt – write data – MDR interrupt – read data – MDA interrupt"
handshake is repeated as long as the transmitter has data to send.
When the last byte has been acknowledged by the receiver (MDA interrupt status), the
microcontroller sets the MONITOR Transmit Control bit MXC to ’0’. This enforces an
inactive (’1’) state in the MX bit. Two frames of MX inactive signifies the end of a
message. Thus, a MONITOR Channel End of Reception MER interrupt status is
generated by the receiver when the MX bit is received in the inactive state in two
consecutive frames. As a result, the microcontroller sets the MR control bit MRC to 0,
which in turn enforces an inactive state in the MR bit. This marks the end of the
transmission, making the MONITOR Channel Active MAC bit return to ’0’.
The MONITOR transfer protocol rules are summarized in the following section
• A pair of MX and MR in the inactive state for two or more consecutive frames indicates
an idle state or an end of transmission.
• A start of a transmission is initiated by the transmitter by setting the MXC bit to ’1’
enabling the internal MX control. The receiver acknowledges the received first byte by
setting the MR control bit to ’1’ enabling the internal MR control.
Data Sheet
355
2001-02-23
C165UTAH
IOM-2 Interface Controller
• The internal MX,MR control indicates or acknowledges a new byte in the MON slot by
toggling MX,MR from the active to the inactive state for one frame.
• Two frames with the MX-bit in the inactive state indicate the end of transmission.
• Two frames with the MR-bit set to inactive indicate a receiver request for abort (see
Chapter 16.3.2).
• The transmitter can delay a transmission sequence by sending the same byte
continuously. In that case the MX-bit remains active in the IOM-2 frame following the
first byte occurrence.
• Since a double last-look criterion is implemented the receiver is able to receive the
MON slot data at least twice (in two consecutive frames). The receiver acknowledge
the data after the reception of two identical bytes in two successive frames.
• To control this handshake procedure a collision detection mechanism is implemented
in the transmitter. This is done by making a collision check per bit on the transmitted
MONITOR data and the MX bit.
• Monitor data will be transmitted repeatedly until its reception is acknowledged or the
transmission time-out timer expires.
• Two frames with the MX bit in the inactive state indicates the end of a message
(EOM).
• MONITOR control commands are processed sequential that means e.g. during a read
on a register no further command is executed.
16.3.2
Abort Procedure
During a transmission process, it is possible for the receiver to ask a transmission to be
aborted by sending an inactive MR bit value in two consecutive frames. This is effected
by the microcontroller writing the MR control bit MRC to ’0’. An aborted transmission is
indicated by a MONITOR Channel Data Abort MAB interrupt status at the transmitter.
The MX/MR bits are under control of the microcontroller through MXC or MRC
respectively. An abort is indicated by an MAB interrupt or MER interrupt respectively.
A transmission is aborted by the C165UTAH if
• abort request from receiver is detected (see Figure 114)
• a collision on the IOM bus of the MONITOR data or MX bit occurs
• a recommended software timer (GPT) expires
A reception is aborted by the C165UTAH if
• an abort request from the transmitting device occurs, see Figure 115. This abort
request has basically the same waveform as an end of transmission, see Figure 116.
Note: In case the C165UTAH does not detect identical monitor messages in two
successive frames, transmission is not aborted. Instead the C165UTAH will wait
until two identical bytes are received in succession.
Data Sheet
356
2001-02-23
C165UTAH
IOM-2 Interface Controller
IOM -2 Frame No.
MX (DU)
1
2
3
4
5
7
1
EOM
0
MR (DD)
6
1
0
Abort Request from Receiver
mon_rec-abort.vsd
Figure 114
Monitor Channel, Transmission Abort requested by the Receiver
IOM -2 Frame No.
MR (DU)
1
2
3
5
6
7
1
EOM
0
MX (DD)
4
1
0
Abort Request from Transmitter
mon_tx-abort.vsd
Figure 115
Data Sheet
Monitor Channel, Transmission Abort requested by the Transmitter
357
2001-02-23
C165UTAH
IOM-2 Interface Controller
IOM -2 Frame No.
MR (DU)
1
2
3
4
5
6
8
1
EOM
0
MX (DD)
7
1
0
mon_norm.vsd
Figure 116
Monitor Channel, normal End of Transmission
Note: The Monitor Time-Out register bit, known from other Infineon devices, is not
implemented within the C165UTAH. To prevent lock-up situations in a MONITOR
transmission, a GPT Timer can be programmed accordingly.
16.3.3
MONITOR Interrupt Logic
Figure 117 shows the MONITOR interrupt structure of the C165UTAH. The MONITOR
Data Receive interrupt status MDR has two enable bits, MONITOR Receive interrupt
Enable (MRE) and MR bit Control (MRC). The MONITOR channel End of Reception
MER, MONITOR channel Data Acknowledged MDA and MONITOR channel Data Abort
MAB interrupt status bits have a common enable bit MONITOR Interrupt Enable MIE.
MRE inactive (0) prevents the occurrence of MDR status, including when the first byte of
a packet is received. When MRE is active (1) but MRC is inactive, the MDR interrupt
status is generated only for the first byte of a receive packet. When both MRE and MRC
are active, MDR is always generated and all received MONITOR bytes - marked by a 1to-0 transition in MX bit - are stored. (Additionally, an active MRC enables the control of
the MR handshake bit according to the MONITOR channel protocol.)
Data Sheet
358
2001-02-23
C165UTAH
IOM-2 Interface Controller
MASK
ISTA
ST
CIC
MOS
4x
HDLC
ST
CIC
MOS
4x
HDLC
MRE
MIE
MOCR
MDR
MER
MDA
MAB
MOSR
IOMINT
Figure 117
16.4
MONITOR Interrupt Structure
C/I Channel Handler
The Command/Indication channel carries real-time status information between the
C165UTAH and another device connected to the IOM.
16.4.1
C/I0 - Command/Indication 0
One C/I channel (called C/I0) conveys the commands and indications between the
C165UTAH and an external layer-1 device. C/I0 channel access may be arbitrated via
the TIC bus access protocol. In this case the arbitration is done in C/I channel 2
(timeslot 1), see Figure 110.
The C/I0 channel is accessed via register bit CIC0_D.CODR0 (in receive direction, layer1 to layer-2) and register CIC0_D.CODX0 (in transmit direction, layer-2 to layer-1). The
C/I0 code is four bits long. In the receive direction, the code from layer-1 is continuously
monitored, with an interrupt being generated anytime a change occurs (ISTA.CIC). A
new code must be found in two consecutive IOM frames to be considered valid and to
trigger a C/I code change interrupt status (double last look criterion).
In the transmit direction, the code written in CIX0 is continuously transmitted in C/I0.
16.4.2
C/I1 - Command/Indication 1
A second C/I channel (called C/I1) can be used to convey real time status information
between the C165UTAH and various non-layer-1 peripheral devices. The C/I1 channel
consists of four or six bits in each direction.The width can be changed from 4bit to 6bit
by setting bit CIC_CMD.CICW.
The C/I1 channel is accessed via registers CIC1_D.CODR1 and CIC1_D.CODX1. A
change in the received C/I1 code is indicated by an interrupt status without double last
look criterion.
Data Sheet
359
2001-02-23
C165UTAH
IOM-2 Interface Controller
Note: The function of the C/I1 channel change detection (CIC_ST.CI1) gets disabled
when disabling the according interrupt (CIC_CMD.CI1E set to ’0’).
16.4.3
CIC Interrupt Logic
Figure 118 shows the CIC interrupt structure.
A CIC interrupt may originate
• from a change in received C/I channel 0 code (CIC0_D.CODR_0) or
• from a change in received C/I channel 1 code (CIC1_D.CODR_1).
The two corresponding status bits CIC0 and CIC1 are in CIC_ST register. CIC1 can be
individually disabled by clearing the enable bit CI1E in the CIX1 register. In this case the
occurrence of a code change in CIR1 will not be displayed by CIC1 until the
corresponding enable bit has been set to one.
Bits CIC0 and CIC1 are cleared by writing ’1’ into the specific bit.
An interrupt status is issued every time a valid new code is loaded into CIR0 or CIR1.
The CIR0 is buffered with a FIFO size of two. If a second code change occurs in the
received C/I channel 0 before the first one has been read, immediately after reading of
CIR0 a new interrupt will be generated and the new code will be stored in CIR0. If several
consecutive codes are detected, only the first and the last code is obtained at the first
and second register read, respectively.
For CIR1 no buffering is available. The actual code of the received C/I channel 1 is
always stored in CIR1.
MASK
ISTA
ST
CIC
MOS
4x
HDLC
ST
CIC
MOS
4x
HDLC
CI1E
CIX1
CIC0
CIC1
CIR0
IOMINT
Figure 118
16.4.4
CIC Interrupt Structure
D-Channel Access Control
D-channel access control is defined to guarantee all connected HDLC controllers a fair
chance to transmit data in the D-channel. Collisions are possible on the IOM-2 interface,
Data Sheet
360
2001-02-23
C165UTAH
IOM-2 Interface Controller
if there is more than one HDLC controller connected. This arbitration mechanism is
implemented in the C165UTAH and will be described in the following chapter.
Note: D-channel access control is only necessary if several D-channel controllers try to
access the same space on the IOM-2 bus.
Note: The TIC Bus D-Channel Access Control mechanism, as described in
Chapter 16.4.4.1, is implemented for HDLC-2 controller only.
16.4.4.1 TIC Bus D-Channel Access Control for HDLC-2 Controller
The TIC bus is implemented to organize the access to the layer-1 functions of the
transceiver device via IOM-2 interface (C/I-channel) and to the D-channel from up to 7
external communication controllers.
The arbitration mechanism is implemented in the last octet in IOM channel 2 of the IOM-2
interface (IOM-2 TE Mode only, TIC_DIS in IOM_CR is disabled). An access request to
the TIC bus may either be generated by software (MP access to the C/I channel) or by
the C165UTAH itself (transmission of an HDLC frame in the D-channel). A software
access request to the bus is effected by setting the BAC bit (CIC_CMD register) to ’1’.
In the case of an access request, the C165UTAH checks the Bus Accessed-bit BAC (bit
5 of DU last octet of channel 2) for the status "bus free“, which is indicated by a logical ’1’.
If the bus is free, the C165UTAH transmits its individual TIC bus address TAD
programmed in the CIC_CMD register and compares it bit by bit with the value on DU. If
a sent bit set to ’1’ is read back as ’0’ because of the access of another D-channel source
with a lower TAD, the C165UTAH withdraws immediately from the TIC bus. The TIC bus
is occupied by the device which sends its address error-free. If more than one device
attempt to seize the bus simultaneously, the one with the lowest address wins and starts
D-channel transmission.
MR
MX
DU
B1
B2
MON0 D CI0
IC1
MR
MX
IC2
MON1
TAD
BAC
CI1
BAC
TAD
2
1
0
TIC-Bus Address (TAD 2-0)
Bus Accessed ('1' no TIC-Bus Access)
tic_octet-du.vsd
Figure 119
Data Sheet
Structure of Last Octet of Ch2 on DU
361
2001-02-23
C165UTAH
IOM-2 Interface Controller
When the TIC bus is seized by the C165UTAH, the bus is identified to other devices as
occupied via the DU channel 2 Bus Accessed-bit state ’0’ until the access request is
withdrawn. After a successful bus access, the C165UTAH is automatically set into a
lower priority class, that is, a new bus access cannot be performed until the status "bus
free" is indicated in two successive frames.
If none of the devices connected to the IOM interface requests access to the D channel,
the TIC bus address 7 will be present. The device with this address will therefore have
access, by default, to the D channels.
Note: Bit BAC (CIC_CMD register) should be reset by the µP when access to the C/I
channels is no more requested, to grant other devices access to the D and C/I
channels.
The availability of the line interface D channel is indicated in bit 5 "S/G" (S/G) of the DD
last octet of channel 2.
S/G = 1: stop
S/G = 0: go
After the BAC bit has been set to active and the UTAH has achieved the lowest TAD, the
S/G bit is checked in order to avoid collision with external layer 1 devices sending on the
D-channel.
MR
MX
DD
B1
B2
MON0 D CI0
IC1
MR
MX
IC2
E
MON1
E
S/G
CI1
S/G
Stop/Go
tic_octet-du.vsd
Figure 120
16.5
Structure of Last Octet of Ch2 on DD
Controller Data Access Handler
The C165UTAH IOM-2 handler provides with his four controller data access registers
(CDA10, CDA11, CDA20, CDA21) a very flexible solution for the access to any possible
timeslot.
The functional unit CDA (controller data access) allows with its control and configuration
registers
Data Sheet
362
2001-02-23
C165UTAH
IOM-2 Interface Controller
• looping of up to four independent IOM-2 interface channels from DU to DD or vice
versa over the four CDA registers
• shifting or switching of two independent IOM-2 interface channels to another two
independent IOM-2 interface channels on both data ports (DU, DD)
• monitoring of up to four time slots on the IOM-2 interface simultaneously
• microcontroller read and write access to each IOM-2 interface channel
16.5.1
Description
The access principle which is identical for the two channel register pairs CDA10/11 and
CDA20/21 is illustrated in Figure 121. The index variables x,y used in the following
description can be 1 or 2 for x, and 0 or 1 for y. The prefix ’CDA_’ from the register names
has been omitted for simplification.
To each of the four CDAxy data registers a TSDPxy register is assigned which specifies
the time slot and the data port. With the TSS (Time Slot Selection) bits a time slot from
0...31 can be selected. With the DPS (Data Port Selection) bit the output of the CDAxy
register can be assigned to DU or DD respectively. The time slot and data port for the
output of CDAxy is always defined by its own TSDPxy register. The input of CDAxy
depends on the SWAP bit in the control registers CDAx_CR.
If the SWAP bit = ’0’ the time slot and data port for the input and output of the CDAxy
register is defined by its own TSDPxy register. The data port for the CDAxy input is vice
versa to the output setting for CDAxy.
If the SWAP bit = ’1’, the input port and time slot of the CDAx0 is defined by the TSDP
register of CDAx1 and the input port and time slot of CDAx1 is defined by the TSDP
register of CDAx0.
The input and output of every CDAxy register can be enabled or disabled by setting the
corresponding EN (-able) bit in the control register CDAx_CR. If the input of a register is
disabled the output value in the register is retained.
Data Sheet
363
2001-02-23
C165UTAH
IOM-2 Interface Controller
.
TSa
TSb
DU
Control
Register
CDAx0
0
1
1
Time Slot
Selection (TSS)
Enable
input
output
(EN_I1)
(EN_O1)
Input
Swap
(SWAP)
1
1
1
1
CDAx1
1
1
0
CDA_TSDPx1
1
0
Data Port
Selection (DPS)
Time Slot
Selection (TSS)
Enable
output
input
(EN_O0) (EN_I0)
Data Port
Selection (DPS)
CDA_TSDPx0
CDA_CRx
0
1
DD
TSa
TSb
IOM_HAND.FM4
x = 1 or 2; a,b = 0...11
Figure 121
16.5.2
Data Access via CDAx0 and CDAx1 register pairs
Looping and Shifting Data
Figure 122 gives examples for typical configurations with the above explained control
and configuration possibilities with the bits TSS, DPS, EN and SWAP in the registers
TSDPxy or CDAx_CR:
a) looping IOM-2 time slot data from DU to DD or vice versa (SWAP = ’0’)
b) shifting data from TSa to TSb on DU and DD (SWAP = ’1’)
c) switching data from TSa (DU) to TSb(DD) and TSb (DU) to TSa (DD)
Data Sheet
364
2001-02-23
C165UTAH
IOM-2 Interface Controller
a) Looping Data
TSa
TSb
CDAx0
CDAx0
.TSS: TSa
TSb
.DPS ’0’
’1’
.SWAP
’0’
DU
DD
b) Shifting Data
TSa
TSb
CDAx0
DU
CDAx0
DD
.TSS: TSa
.DPS ’0’
.SWAP
TSb
’1’
’1’
c) Switching Data
TSa
TSb
CDAx0
DU
CDAx0
DD
.TSS: TSa
.DPS ’0’
.SWAP
.x = 1 or 2
Figure 122
TSb
’0’
’1’
Examples for Data Access via CDAxy Registers
a) Looping Data
b) Shifting Data
c) Switching Data
Data Sheet
365
2001-02-23
C165UTAH
IOM-2 Interface Controller
16.5.3
Monitoring Data
Figure 123 gives an example for monitoring of two IOM-2 time slots each on DU or DD
simultaneously.
.
a) Monitoring Data
EN_O: ’0’
CDA_CR1. EN_I: ’1’
DPS: ’0’
TSS: TS(2n)
’0’
’1’
’0’
TS(2m+1)
DU
CDA10
CDA11
CDA20
CDA21
TSS: TS(2n)
’1’
DPS:
CDA_CR2.
EN_I: ’1’
EN_O: ’0’
Figure 123
TS(2m+1)
’1’
’1’
’0’
DD
n,m = 0...5
Example for Monitoring Data
Note: Due to the hardware design, there are some restrictions for the CDA Monitoring
Data Function. These restrictions apply if the CDA Monitoring Function is used to
monitor other active functions (e.g. B1, B2, D1, D2, TIC, CIC, Monitor).
These restrictions do not apply if the DPS setting of the monitoring CDA channel
and the active channel are different.
If the DPS setting of the monitoring CDA channel is equal to the DPS setting of the
monitored active channel, following rules apply:
• In case of monitoring B1, Monitor, TIC the data is stored in the specified CDA_xy
register (normal function).
• In case of monitoring B2, D1, D2, CIC the data is stored in the reverse CDA_xy
register. For example, when activating CDA_10, it is stored in CDA_11 and if CDA_11
activated, it is stored in CDA_10.
Data Sheet
366
2001-02-23
C165UTAH
IOM-2 Interface Controller
16.5.4
Synchronous Transfer
While looping, shifting and switching the data can be accessed by the controller between
the synchronous transfer interrupt (STI) and the status overflow interrupt (STOV).
The microcontroller access to the CDAxy registers can be synchronized by means of
four programmable synchronous transfer interrupts (STIxy) and synchronous transfer
overflow interrupts (STOVxy) in the STI register.
Depending on the DPS bit in the corresponding CDA_TSDPxy register the STIxy is
generated two (for DPS=’0’) or one (for DPS=’1’) BCL clock after the selected time slot
(CDA_TSDPxy.TSS). One BCL clock is equivalent to two DCL clocks.
A non masked synchronous transfer overflow (STOVx0y0) interrupt is generated if the
appropriate STIx1y1 is not acknowledged in time. The STIx1y1 is acknowledged in time if
bit ACKx1y1 in the ASTI register is set to ’1’ one BCL clock (for DPS=’0’) or zero BCL
clocks (for DPS=’1’) before the time slot which is selected for the appropriate STOVx0y0.
If STIx1y1 and STOVx1y1 are not masked STOVx1y1 is only related to STIx1y1 (see
example a), c) and d) of Figure 125).
If STIx1y1 is masked but STOVx1y1 is not masked, STOVx0y0 is related to each enabled
STIxy (see example b) and d) of Figure 125).
Setting the corresponding bits in the MSTI (Mask Synchronous Transfer Interrupts)
register masks the STIxy and the STOVxy interrupt. The interrupt structure of the
synchronous transfer is shown in Figure 124. Examples of the described synchronous
transfer interrupt controlling are illustrated in Figure 125. A read to the STI register
clears the STIxy and STOVxy interrupts.
.
IOMINT
ST
CIC
MOS
4x
HDLC
ST
CIC
MOS
4x
HDLC
MASK
ISTA
STOV21
STOV20
STOV11
STOV10
STI21
STI20
STI11
STI10
MSTI
Figure 124
Data Sheet
STOV21
STOV20
STOV11
STOV10
STI21
STI20
STI11
STI10
STI
ACK21
ACK20
ACK11
ACK10
ASTI
Interrupt Structure of the Synchronous Data Transfer
367
2001-02-23
C165UTAH
IOM-2 Interface Controller
.
: STI interrupt generated
: STOV interrupt generated for a not acknowledged STI interrupt
a) Interrupts for data access to time slot 0 (B1 after reset), MSTI.STI10 and MSTI.STOV10 enabled
xy:
CDA_TDSPxy.TSS:
MSTI.STIxy:
MSTI.STOVxy:
10
TS0
'0'
'0'
11
TS1
'1'
'1'
21
TS5
'1'
'1'
TS11 TS0 TS1 TS2 TS3
20
TS11
'1'
'1'
TS4 TS5 TS6 TS7 TS8 TS9 TS10 TS11 TS0
b) Interrupts for data access to time slot 0 (B1 after reset), STOV interrupt used as flag for "last possible CDA
access"; MSTI.STI10 and MSTI.STOV20 enabled
xy:
CDA_TDSPxy.TSS:
MSTI.STIxy:
MSTI.STOVxy:
10
TS0
'0'
'1'
11
TS1
'1'
'1'
21
TS5
'1'
'1'
TS11 TS0 TS1 TS2 TS3
20
TS11
'1'
'0'
TS4 TS5 TS6 TS7 TS8 TS9 TS10 TS11 TS0
c) Interrupts for data access to time slot 0 and 1 (B1 and B2 after reset), MSTI.STI10, MSTI.STOV10,
MSTI.STI11 and MSTI.STOV11 enabled
xy:
CDA_TDSPxy.TSS:
MSTI.STIxy:
MSTI.STOVxy:
10
TS0
'0'
'0'
11
TS1
'0'
'0'
21
TS5
'1'
'1'
TS11 TS0 TS1 TS2 TS3
20
TS11
'1'
'1'
TS4 TS5 TS6 TS7 TS8 TS9 TS10 TS11 TS0
d) Interrupts for data access to time slot 0 (B1 after reset), STOV20 interrupt used as flag for "last possible CDA
access", STOV10 interrupt used as flag for "CDA access failed"; MSTI.STI10, MSTI.STOV10 and
MSTI.STOV20 enabled
xy:
CDA_TDSPxy.TSS:
MSTI.STIxy:
MSTI.STOVxy:
10
TS0
'0'
'0'
11
TS1
'1'
'1'
21
TS5
'1'
'1'
TS11 TS0 TS1 TS2 TS3
20
TS11
'1'
'0'
TS4 TS5 TS6 TS7 TS8 TS9 TS10 TS11 TS0
sti_stov.vsd
Figure 125
Examples for the Synchronous Transfer Interrupt Control with one
enabled STIxy
Figure 126 shows the timing of looping TSa on DU to TSa on DD (a = 0...11) via CDAxy
register. TSa is read in the CDAxy register from DU and is written one frame later on DD.
Data Sheet
368
2001-02-23
C165UTAH
IOM-2 Interface Controller
.
a = 0...11
FSC
DU
TSa
TSa
WR
RD
DD
TSa
STOV
µC *)
STI
ACK
STI
CDAxy
TSa
*) if access by the µC is required
Figure 126
Data Access when Looping TSa from DU to DD
Figure 127 shows the timing of shifting data from TSa to TSb on DU(DD). In
Figure 127a) shifting is done in one frame because TSa and TSb didn’t succeed direct
one another (a,b = 0...9 and b = a+2). In Figure 127b) shifting is done from one frame
to the following frame. This is the case when the time slots succeed one other (b = a+1)
or b is smaller than a (b < a).
Data Sheet
369
2001-02-23
C165UTAH
IOM-2 Interface Controller
a) Shifting TSa → TSb within one frame
(a,b: 0...11 and b ≥ a+2)
FSC
DU
(DD)
TSa
TSb
TSa
µC
ACK
*)
STI
STOV
WR
STI
RD
CDAxy
b) Shifting TSa → TSb in the next frame
(a,b: 0...11 and (b = a+1 or b 1
mode 1
SAP1,SAP2,SAPG
–
High-byte address
compare.
1
0
1 Transparent > 1
mode 2
–
TEI1,TEI2,TEIG Low-byte address
compare.
Note: SAP1, SAP2: two programmable address values for the first received address
byte (in the case of an address field longer than 1 byte);
SAPG = fixed value FC / FEH.
TEI1, TEI2: two programmable address values for the second (or the only, in the
case of a one-byte address) received address byte; TEIG = fixed value FFH
Two different methods of the high byte and/or low byte address comparision can
be selected by setting SAP1.MHA and/or SAP2.ML
SAPI1 Register
Address:
Name:
15
14
s_adr_x + 10H
SAP1_x
13
12
11
10
9
8
7
RESERVED
6
5
SAPI1
Field
Bits
Type Value Description
SAPI1
7:2
W
Data Sheet
3FH
4
3
2
1
0
RES
ERV MHA
ED
SAPI1 value
Value of the first programmable Service
Access Point Identifier (SAPI) according
to the ISDN LAPD protocol.
423
2001-02-23
C165UTAH
IOM-2 Interface Controller
Field
Bits
Type Value Description
MHA
0
W
RESERVED
14:13 R
9,4
0
Mask High Address
0: The SAPI address of an incomming
frame is compared with SAP1, SAP2,
SAPG
1: The SAPI address of an incomming
frame is compared with SAP1 and
SAPG.
SAP1 can be masked with SAP2 thereby
bitpositions of SAP1 are not compared
if they are set to ’1’ in SAP2.
These bits are reserved.
SAPI2 Register
Address:
Name:
15
14
s_adr_x + 12H
SAP2_x
13
12
11
10
9
8
7
RESERVED
6
5
SAPI2
Field
Bits
Type Value Description
SAPI2
7:2
W
Data Sheet
3FH
4
3
2
1
0
RES
ERV MLA
ED
SAPI2 value
Value of the second programmable
Service Access Point Identifier (SAPI)
according to the ISDN LAPD-protocol.
424
2001-02-23
C165UTAH
IOM-2 Interface Controller
Field
Bits
Type Value Description
MLA
0
W
RESERVED
14:13 R
9,4
0
Mask Low Address
0: The TEI address of an incomming
frame is compared with TEI1, TEI2,
TEIG
1: The TEI address of an incomming
frame is compared with TEI1
andTEIG.
TEI1 can be masked with TEI2 thereby
bitpositions of TEI1 are not compared
if they are set to ’1’ in TEI2
These bits are reserved.
Receive Frame Byte Count
Address:
Name:
15
14
s_adr_x + 14H
RBC_x
13
RESERVED
Data Sheet
12
11
10
9
8
7
OV
6
5
4
3
2
1
0
RBC
425
2001-02-23
C165UTAH
IOM-2 Interface Controller
Field
Bits
Type Value Description
RBC
11:0
R/W
000H
12 bits of the total number of bytes in a
received message. Bit 11 is the most
significant bit, bit 0 the least significant
bit, respective.
Note: Normally RBC should be read by
the microcontroller after an RMEinterrupt in order to determine the
number of bytes to be read from
the RFIFO, and the total message
length. The contents of the
registers are valid only after an
RME or RPF interrupt, and remain
so until the frame is acknowledged
via resetting RBC (I.e. writing a
value to RBC) or via resetting the
RFIFO (i.e. setting RRES).
OV
12
RESERVED
15:13 R
Data Sheet
R/W
0
Overflow
A ’1’ in this bit position indicates a
message longer than (212 - 1) = 4095
bytes .
These bits are reserved.
426
2001-02-23
C165UTAH
IOM-2 Interface Controller
TEI1 Register
Address:
Name:
15
s_adr_x + 16H
TEI1_x
14
13
12
11
10
9
8
7
6
RESERVED
5
4
3
2
TEI1
Field
Bits
Type Value Description
TEI1
7:1
W
7FH
1
0
EA
Terminal Endpoint Identifier
In all message transfer modes except in
transparent modes 0, 1 and extended
transparent mode, TEI1 is used for
address recognition. In the case of a twobyte address field, it contains the value of
the
first
programmable
Terminal
Endpoint Identifier according to the ISDN
LAPD-protocol.
In non-auto-modes with one-byte
address field, TEI1 is a command
address, according to X.25 LAPB.
EA
0
W
RESERVED
15:8
R
Data Sheet
1
Address field Extension bit
This bit is set to ’1’ according to HDLC/
LAPD.
These bits are reserved.
427
2001-02-23
C165UTAH
IOM-2 Interface Controller
TEI2 Register
Address:
Name:
15
s_adr_x + 18H
TEI2_x
14
13
12
11
10
9
8
7
6
RESERVED
5
4
3
TEI1
2
1
0
EA
Field
Bits
Type Value Description
TEI1
7:1
W
7FH
Terminal Endpoint Identifier
In all message transfer modes except in
transparent modes 0, 1 and extended
transparent mode, TEI2 is used for
address recognition. In the case of a twobyte address field, it contains the value of
the second programmable Terminal
Endpoint Identifier according of the ISDN
LAPD-protocol.
In non-auto-modes with one-byte
address field, TEI2 is a response
address, according to X.25 LAPD.
EA
0
W
1
Address field Extension bit
This bit is to be set to ’1’ according to
HDLC/LAPD.
RESERVED
15:8
R
Data Sheet
These bits are reserved.
428
2001-02-23
C165UTAH
IOM-2 Interface Controller
Looping Register
Address:
Name:
15
14
s_adr_x + 1AH
LOOP_x
13
12
11
10
9
8
7
6
5
RESERVED
4
3
2
1
0
ELP FAST TLP
Field
Bits
Type Value Description
TLP
0
R/W
0
Transmit Loop
Setting TLP to ’1’ causes outgoing data
being looped back into the receiver. The
received data contains all HDLC
procession, but does not contain the start
and ending Flag.
FAST
1
R/W
0
Fast Looping
Setting FAST = ’1’ affects the TLP timing.
’0’: The TLP is processed with the IOM-2
lines condition (8 kHz signal).
’1’: The TLP looping runs with the BCL
clock.
ELP
Data Sheet
2
R/W
0
External Loop
Setting ELP to ’1’ causes incomming
HDLC data being looped back to the line.
429
2001-02-23
C165UTAH
Watchdog Timer (WDT)
17
Watchdog Timer (WDT)
To allow recovery from software or hardware failure, the C165UTAH provides a
Watchdog Timer. If the software fails to service this timer before an overflow occurs, an
internal reset sequence will be initiated. This internal reset will also pull the RSTOUT pin
low, which also resets the peripheral hardware, which might be the cause for the
malfunction. When the watchdog timer is enabled and the software has been designed
to service it regularly before it overflows, the watchdog timer will supervise the program
execution, as it only will overflow if the program does not progress properly. The
watchdog timer will also time out, if a software error was due to hardware related failures.
This prevents the controller from malfunctioning for longer than a user-specified time.
The watchdog timer provides two registers: a read-only timer register that contains the
current count, and a control register for initialization.
Reset Indication Pin
Data Registers
WDT
RSTOUT
Figure 133
Control Registers
WDTCON
SFRs and Port Pins associated with the Watchdog Timer
The watchdog timer is a 16-bit up counter which can be clocked with the CPU clock (fCPU)
either divided by 2 or divided by 128. This 16-bit timer is realized as two concatenated
8-bit timers (see figure below). The upper 8 bits of the watchdog timer can be preset to
a user-programmable value via a watchdog service access in order to vary the watchdog
expire time. The lower 8 bits are reset on each service access.
÷2
f CPU
MUX
÷ 128
WDT Low Byte
WDT High Byte
WDTR
Clear
RSTOUT
WDTIN
Reset
WDT
Control
WDTREL
MCB02052
Figure 134
Data Sheet
Watchdog Timer Block Diagram
430
2001-02-23
C165UTAH
Watchdog Timer (WDT)
17.1
Operation of the Watchdog Timer
The current count value of the Watchdog Timer is contained in the Watchdog Timer
Register WDT, which is a non-bitaddressable read-only register. The operation of the
Watchdog Timer is controlled by its bitaddressable Watchdog Timer Control Register
WDTCON. This register specifies the reload value for the high byte of the timer, selects
the input clock prescaling factor and provides a flag that indicates a watchdog timer
overflow.
WDTCON (FFAEH / D7H)
15
14
13
12
SFR
11
10
9
8
Reset Value: 00XXH
7
5
-
-
Bit
Function
WDTIN
Watchdog Timer Input Frequency
Selection
‘0’: Input frequency is fCPU / 2
‘1’: Input frequency is fCPU / 128
WDTR
Watchdog Timer Reset Indication Flag
Set by the watchdog timer on an overflow.
Cleared by the SRVWDT instruction.
SWR
Software Reset
Set by the command SRST
SHWR
Short Hardware Reset
Set by the input RSTIN
LHWR
Long Hardware Reset
Set by the input RSTIN
reserved
Reserved
These bits are reserved
WDTREL
Watchdog Timer Reload Value (for the high
byte)
4
3
2
1
0
LHW SHW
WDT WDT
R
R
SWR
R
IN
RESERVED
WDTREL
rw
6
-
r
r
r
r
rw
Note: The C165UTAH does not
distinguish between short and
long hardware reset.
The reset sources supported by the C165UTAH are summarized in Table 101.
Note: Differentiation between long and short hardware reset, known from other Infineon
C16x devices, is not supported.
Note: An internal power-on detection circuitry, also known from other C16x devices, is
not implemented. Therefore, bit WDTCON.5 (in other devices called PONR power-on reset) is reserved.
Data Sheet
431
2001-02-23
C165UTAH
Watchdog Timer (WDT)
Table 101
WDTCON Register: Reset Source Identification
Type of Reset
WDTCON
Reset Value
WDTCON Flags being set
Hardware reset via pin RSTIN
001CH
LHWR, SHWR, SWR
Software reset via command SRST
0004H
SWR
Watchdog Timer reset
0006H
SWR, WDTR
Note: The WDTCON register bits [7, 6, 5] 4, 3, 2 and 1 are cleared by the EINIT
command.
After any software reset, external hardware reset, or watchdog timer reset, the watchdog
timer is enabled and starts counting up from 0000H with the frequency fCPU/2. The input
frequency may be switched to fCPU/128 by setting bit WDTIN. The watchdog timer can
be disabled via the instruction DISWDT (Disable Watchdog Timer). Instruction DISWDT
is a protected 32-bit instruction which will ONLY be executed during the time between a
reset and execution of either the EINIT (End of Initialization) or the SRVWDT (Service
Watchdog Timer) instruction. Either one of these instructions disables the execution of
DISWDT.
When the watchdog timer is not disabled via instruction DISWDT, it will continue
counting up, even during Idle Mode. If it is not serviced via the instruction SRVWDT by
the time the count reaches FFFFH the watchdog timer will overflow and cause an internal
reset. This reset will pull the external reset indication pin RSTOUT low. It differs from a
software or external hardware reset in that bit WDTR (Watchdog Timer Reset Indication
Flag) of register WDTCON will be set. A hardware reset or the SRVWDT instruction will
clear this bit. Bit WDTR can be examined by software in order to determine the cause of
the reset.
A watchdog reset will also complete a running external bus cycle before starting the
internal reset sequence if this bus cycle does not use READY or samples READY active
(low) after the programmed waitstates. Otherwise the external bus cycle will be aborted.
Note: After a hardware reset that activates the Bootstrap Loader the watchdog timer will
be disabled.
To prevent the watchdog timer from overflowing, it must be serviced periodically by the
user software. The watchdog timer is serviced with the instruction SRVWDT, which is a
protected 32-bit instruction. Servicing the watchdog timer clears the low byte and reloads
the high byte of the watchdog timer register WDT with the preset value from bitfield
WDTREL which is the high byte of register WDTCON. Servicing the watchdog timer will
also reset bit WDTR. After being serviced the watchdog timer continues counting up from
the value ( * 28). Instruction SRVWDT has been encoded in such a way that
the chance of unintentionally servicing the watchdog timer (eg. by fetching and executing
Data Sheet
432
2001-02-23
C165UTAH
Watchdog Timer (WDT)
a bit pattern from a wrong location) is minimized. When instruction SRVWDT does not
match the format for protected instructions the Protection Fault Trap will be entered,
rather than the instruction be executed.
The time period for an overflow of the watchdog timer is programmable in two ways:
• the input frequency to the watchdog timer can be selected via bit WDTIN in register
WDTCON to be either fCPU/2 or fCPU/128.
• the reload value WDTREL for the high byte of WDT can be programmed in register
WDTCON.
The period PWDT between servicing the watchdog timer and the next overflow can
therefore be determined by the following formula:
PWDT =
2(1 + *6) * (216 - * 28)
fCPU
Note: For safety reasons, the user is advised to rewrite WDTCON each time before the
watchdog timer is serviced.
Data Sheet
433
2001-02-23
C165UTAH
Bootstrap Loader
18
Bootstrap Loader
The built-in bootstrap loader of the C165UTAH provides a mechanism to load the startup
program, which is executed after reset, via the serial interface.
The bootstrap loader moves code/data into the internal RAM, but it is also possible to
transfer data via the serial interface into an external RAM using a second level loader
routine. It may be used to provide lookup tables or may provide “core-code”, ie. a set of
general purpose subroutines, eg. for I/O operations, number crunching, system
initialization, etc.
RSTIN
P0L.4
1)
2)
4)
RxD0
3)
TxD0
5)
CSP:IP
6)
Int. Boot ROM BSL-routine
32 bytes
user software
1)
BSL initialization time, > 2µs @ fCPU = 20 MHz.
2)
Zero byte (1 start bit, eight ‘0’ data bits, 1 stop bit), sent by host.
3)
Identification byte, sent by C165UTAH.
4)
32 bytes of code / data, sent by host.
Figure 135
Bootstrap Loader Sequence
The Bootstrap Loader may be used to load the complete application software into
ROMless systems, it may load temporary software into complete systems for testing or
calibration.
The BSL mechanism may be used for standard system startup as well as only for special
occasions like system maintenance (firmware update) or end-of-line programming or
testing.
Data Sheet
434
2001-02-23
C165UTAH
Bootstrap Loader
Entering the Bootstrap Loader
The C165UTAH enters BSL mode if pin P0L.4 is sampled low at the end of a hardware
reset. In this case the built-in bootstrap loader is activated independent of the selected
bus mode.
After entering BSL mode and the respective initialization the C165UTAH scans the
RXD0 line to receive a zero byte, ie. one start bit, eight ‘0’ data bits and one stop bit. From
the duration of this zero byte it calculates the corresponding baudrate factor with respect
to the current CPU clock, initializes the serial interface ASC accordingly and switches pin
TxD0 to output. Using this baudrate, an identification byte is returned to the host that
provides the loaded data.
This identification byte identifies the device to be booted. The following codes are
defined for Infineon Technologies microcontrollers:
55H:
A5H:
B5H:
C5H:
D5H:
8xC166.
Previous versions of the C167 (obsolete).
C165.
C167 derivatives.
C165UTAH (and all other devices equipped with identification registers).
Note: The identification byte D5H does not directly identify a specific derivative. This
information can in this case be obtained from the identification registers.
When the C165UTAH has entered BSL mode, the following configuration is
automatically set (values that deviate from the normal reset values, are marked):
Watchdog Timer:
Context Pointer CP:
Stack Pointer SP:
Register S0CON:
Register S0BG:
Disabled
FA00H
FA40H
8011H
acc. to ‘00’ byte
Register SYSCON:
Register STKUN:
Register STKOV:
Register BUSCON0:
P3.10 / TXD0:
DP3.10:
0E00H
FA40H
FA0CH 0C
acc. to startup config.
‘1’
‘1’
Other than after a normal reset the watchdog timer is disabled, so the bootstrap loading
sequence is not time limited. Pin TXD0 is configured as output, so the C165UTAH can
return the identification byte.
The hardware that activates the BSL during reset may be a simple pull-down resistor on
P0L.4 for systems that use this feature upon every hardware reset. You may want to use
a switchable solution (via jumper or an external signal) for systems that only temporarily
use the bootstrap loader.
Data Sheet
435
2001-02-23
C165UTAH
Bootstrap Loader
External Signal
Circuit_1
Circuit_2
P0L.4
P0L.4
R P0L.4
Normal Boot
BSL
R P0L.4
8 kΩ
8 kΩ
MCA02261
Figure 136
Hardware Provisions to Activate the BSL
After sending the identification byte the ASC receiver is enabled and is ready to receive
the initial 32 bytes from the host. A half duplex connection is therefore sufficient to feed
the BSL.
Note: In order to properly enter BSL mode it is not only required to pull P0L.4 low,
but also pins P0L.2, P0L.3, P0L.5 must receive defined levels.
This is described in chapter "System Reset“.
Loading the Startup Code
After sending the identification byte the BSL enters a loop to receive 32 bytes via ASC.
These bytes are stored sequentially into locations 00’FA40H through 00’FA5FH of the
internal RAM. So up to 16 instructions may be placed into the RAM area. To execute the
loaded code the BSL then jumps to location 00’FA40H, ie. the first loaded instruction. The
bootstrap loading sequence is now terminated, the C165UTAH remains in BSL mode,
however. Most probably the initially loaded routine will load additional code or data, as
an average application is likely to require substantially more than 16 instructions. This
second receive loop may directly use the pre-initialized interface ASC to receive data
and store it to arbitrary user-defined locations.
This second level of loaded code may be the final application code. It may also be
another, more sophisticated, loader routine that adds a transmission protocol to enhance
the integrity of the loaded code or data. It may also contain a code sequence to change
the system configuration and enable the bus interface to store the received data into
external memory.
This process may go through several iterations or may directly execute the final
application. In all cases the C165UTAH will still run in BSL mode, ie. with the watchdog
timer disabled and limited access to the internal code memory.
Data Sheet
436
2001-02-23
C165UTAH
Bootstrap Loader
Exiting Bootstrap Loader Mode
In order to execute a program in normal mode, the BSL mode must be terminated first.
The C165UTAH exits BSL mode upon a software reset (ignores the level on P0L.4) or a
hardware reset (P0L.4 must be high then!). After a reset the C165UTAH will start
executing from location 00’0000H of the external memory (make sure, pin EA is tied to
’0’ signal).
Choosing the Baudrate for the BSL
The calculation of the serial baudrate for ASC from the length of the first zero byte that
is received, allows the operation of the bootstrap loader of the C165UTAH with a wide
range of baudrates. However, the upper and lower limits have to be kept, in order to
insure proper data transfer.
f
BC165UT
CPU
-----------------------------------------32 ⋅ ( S0BRL + 1 )
The C165UTAH uses timer T6 to measure the length of the initial zero byte. The
quantization uncertainty of this measurement implies the first deviation from the real
baudrate, the next deviation is implied by the computation of the S0BRL reload value
from the timer contents. The formula below shows the association:
T6 – 36
S0BRL = ------------------72
,
9
T6 = -- ⋅
4
f
CPU
---------------
B Host
For a correct data transfer from the host to the C165UTAH the maximum deviation
between the internal initialized baudrate for ASC and the real baudrate of the host should
be below 2.5%. The deviation (FB, in percent) between host baudrate and C165UTAH
baudrate can be calculated via the formula below:
F
B Contr – B Host
=
---------------------------------------- ⋅ 100 % ,
B
B
Contr
F
B
≤ 2,5 %
Note: Function (FB) does not consider the tolerances of oscillators and other devices
supporting the serial communication.
This baudrate deviation is a nonlinear function depending on the CPU clock and the
baudrate of the host. The maxima of the function (FB) increase with the host baudrate
due to the smaller baudrate prescaler factors and the implied higher quantization error
(see figure below).
Data Sheet
437
2001-02-23
C165UTAH
Bootstrap Loader
Ι
FB
2.5%
B Low
B High
B Host
ΙΙ
Figure 137
MCA02260
Baudrate deviation between host and C165UTAH
Minimum baudrate (BLow in the figure above) is determined by the maximum count
capacity of timer T6, when measuring the zero byte, ie. it depends on the CPU clock.
Using the maximum T6 count 216 in the formula the minimum baudrate for fCPU=20 MHz
is 687 Baud. The lowest standard baudrate in this case would be 1200 Baud. Baudrates
below BLow would cause T6 to overflow. In this case ASC cannot be initialized properly.
Maximum baudrate (BHigh in the figure above) is the highest baudrate where the
deviation still does not exceed the limit, ie. all baudrates between BLow and BHigh are
below the deviation limit. The maximum standard baudrate that fulfills this requirement
is 19200 Baud.
Higher baudrates, however, may be used as long as the actual deviation does not
exceed the limit. A certain baudrate (marked I) in the figure) may eg. violate the deviation
limit, while an even higher baudrate (marked II) in the figure) stays very well below it. This
depends on the host interface.
Data Sheet
438
2001-02-23
C165UTAH
System Reset
19
System Reset
The internal system reset function provides initialization of the C165UTAH into a defined
default state and is invoked either by asserting a hardware reset signal on pin RSTIN
(Hardware Reset Input), upon the execution of the SRST instruction (Software Reset) or
by an overflow of the watchdog timer (WDT).
Whenever one of these conditions occurs, the microcontroller is reset into its predefined
default state through an internal reset procedure. When a reset is initiated, pending
internal hold states are cancelled and the current internal access cycle (if any) is
completed. An external bus cycle is aborted, except for a watchdog reset (see
description). After that the bus pin drivers and the I/O pin drivers are switched off
(tristate). RSTOUT is activated depending on the reset source.
The internal reset procedure requires 516 CPU clock cycles in order to perform a
complete reset sequence. This 516 cycle reset sequence is started upon a watchdog
timer overflow, a SRST instruction or when the reset input signal RSTIN is latched low
(hardware reset). The internal reset condition is active at least for the duration of the
reset sequence and then until the RSTIN input is inactive. When this internal reset
condition is removed (reset sequence complete and RSTIN inactive), the reset
configuration is latched from PORT0, and pins ALE, RD and WR are driven to their
inactive levels.
Note: Bit ADP, which selects the Adapt mode during low RSTIN signal, is latched with
the rising edge of RSTIN.
After the internal reset condition is removed, the microcontroller will start program
execution from memory location 00’0000H in code segment zero. This start location will
typically hold a branch instruction to the start of a software initialization routine for the
application specific configuration of peripherals and CPU Special Function Registers.
Data Sheet
439
2001-02-23
C165UTAH
System Reset
RSTOUT
C165UTAH
External
Hardware
RSTIN
VCC
a)
Reset
+ b)
&
External
Reset
Sources
a) Generated Warm Reset
b) Automatic Power-ON Reset
MCA02259
Figure 138
External Reset Circuitry
Hardware Reset
A hardware reset is triggered when the reset input signal RSTIN is latched low. To
ensure the recognition of the RSTIN signal (latching), it must be held low for at least
8 CPU clock cycles.
Note: During reset, the CPU is clocked with the free-running PLL clock which may run
as slow as < 1 MHz.
Also shorter RSTIN pulses may trigger a hardware reset, if they coincide with the latch’s
sample point. However, it is recommended to keep RSTIN low for ca. 1 ms. After the
reset sequence has been completed, the RSTIN input is sampled. When the reset input
signal is active at that time the internal reset condition is prolonged until RSTIN gets
inactive.
During a hardware reset the PORT0 inputs for the reset configuration need some time
to settle on the required levels, especially if the hardware reset aborts a read operation
form an external peripheral. During this settling time the configuration may intermittently
be wrong. In such a case also the PLL clock selection may be wrong.
Note: To ensure a glitch free start-up of the C165UTAH, it is strongly recommended to
provide an external reset pulse of ca. 1 ms in order to allow the PLL to settle on
the desired CPU clock frequency.
The input RSTIN provides an internal pullup device equalling a resistor of 100 KΩ to
660 KΩ (the minimum reset time must be determined by the lowest value). Simply
connecting an external capacitor is sufficient for an automatic power-on reset, see b) in
Data Sheet
440
2001-02-23
C165UTAH
System Reset
Figure 138. RSTIN may also be connected to the output of other logic gates, see a)
same figure.
Note: A power-on reset requires an active time of two reset sequences (1036 CPU clock
cycles) after a stable clock signal is available (about 10...50 ms to allow the onchip oscillator to stabilize).
Software Reset
The reset sequence can be triggered at any time via the protected instruction SRST
(Software Reset). This instruction can be executed deliberately within a program, eg. to
leave bootstrap loader mode, or upon a hardware trap that reveals a system failure.
C165UTAH’s latched in reset configuration on software reset is shown in Figure 140,
page 447.
Watchdog Timer Reset
When the watchdog timer is not disabled during the initialization or serviced regularly
during program execution it will overflow and trigger the reset sequence. Other than
hardware and software reset the watchdog reset completes a running external bus cycle
if this bus cycle either does not use READY at all, or if READY is sampled active (low)
after the programmed waitstates. When READY is sampled inactive (high) after the
programmed waitstates the running external bus cycle is aborted. Then the internal reset
sequence is started.
Note: For latched in watchdog reset configuration, refer to Figure 140, page 447.
The watchdog reset cannot occur while the C165UTAH is in bootstrap loader
mode!
The C165UTAH’s Pins after Reset
After the reset sequence the different groups of pins of the C165UTAH are activated in
different ways depending on their function. Bus and control signals are activated
immediately after the reset sequence according to the configuration latched from
PORT0, so either external accesses can take place or the external control signals are
inactive. The general purpose I/O pins remain in input mode (high impedance) until
reprogrammed via software (see figure below). The RSTOUT pin remains active (low)
until the end of the initialization routine (see description).
Data Sheet
441
2001-02-23
C165UTAH
System Reset
8)
6)
RSTIN
7)
RD, WR
ALE
Bus
1)
IO
2)
RSTOUT
2)
4)
3)
Internal Reset Condition
5)
Initialization
3)
6)
RSTIN
Internal Reset Condition
Initialization
When the internal reset condition is extended by RSTIN, the activation of the output signals is
delayed until the end of the internal reset condition.
1)
2)
3)
4)
5)
6)
7)
8)
Current bus cycle is completed or aborted.
Switches asinchronously with RSTIN, sinchronously upon software or watchdog reset.
The reset condition ends here. The C 167CR starts program execution.
Activation of the IO pins is controlled by software.
Execution of the EINIT instruction.
The shaded area designates the internal reset sequence, which starts after synchronization of RSTIN.
A short hardware reset is extended until the end of the reset sequence in Bidirectional reset mode.
A software or WDT reset activates the RSTIN line in Bidirectional reset mode.
MCS02258
Figure 139
Reset Input and Output Signals
Reset Output Pin
The RSTOUT pin is dedicated to generate a reset signal for the system components
besides the controller itself. RSTOUT will be driven active (low) at the begin of any reset
sequence (triggered by hardware, the SRST instruction or a watchdog timer overflow).
RSTOUT stays active (low) beyond the end of the internal reset sequence until the
protected EINIT (End of Initialization) instruction is executed (see figure above). This
Data Sheet
442
2001-02-23
C165UTAH
System Reset
allows the complete configuration of the controller including its on-chip peripheral units
before releasing the reset signal for the external peripherals of the system.
Watchdog Timer Operation after Reset
The watchdog timer starts running after the internal reset has completed. It will be
clocked with the internal system clock divided by 2 (18 MHz @ fCPU=36 MHz), and its
default reload value is 00H, so a watchdog timer overflow will occur 131072 CPU clock
cycles (3.64 ms @ fCPU=36 MHz) after completion of the internal reset, unless it is
disabled, serviced or reprogrammed meanwhile. When the system reset was caused by
a watchdog timer overflow, the WDTR (Watchdog Timer Reset Indication) flag in register
WDTCON will be set to '1'. This indicates the cause of the internal reset to the software
initialization routine. WDTR is reset to '0' by an external hardware reset or by servicing
the watchdog timer. After the internal reset has completed, the operation of the watchdog
timer can be disabled by the DISWDT (Disable Watchdog Timer) instruction. This
instruction has been implemented as a protected instruction. For further security, its
execution is only enabled in the time period after a reset until either the SRVWDT
(Service Watchdog Timer) or the EINIT instruction has been executed. Thereafter the
DISWDT instruction will have no effect.
Note: For a complete description of register WDTCON, refer to Chapter 17.1, page 431.
Reset Values for the C165UTAH Registers
During the reset sequence the registers of the C165UTAH are preset with a default
value. Most SFRs, including system registers and peripheral control and data registers,
are cleared to zero, so all peripherals and the interrupt system are off or idle after reset.
A few exceptions to this rule provide a first pre-initialization, which is either fixed or
controlled by input pins.
DPP1:
DPP2:
DPP3:
CP:
STKUN:
STKOV:
SP:
WDTCON:
S0RBUF:
SSCRB:
SYSCON:
BUSCON0:
RP0H:
ONES:
Data Sheet
0001H (points to data page 1)
0002H (points to data page 2)
0003H (points to data page 3)
FC00H
FC00H
FA00H
FC00H
00XXH, (value depends on the reset configuration)
XXH (undefined)
XXXXH (undefined)
0XX0H (set according to reset configuration)
0XX0H (set according to reset configuration)
XXH (reset levels of P0H)
FFFFH (fixed value)
443
2001-02-23
C165UTAH
System Reset
The Internal RAM after Reset
The contents of the internal RAM are not affected by a system reset. However, after
power-on the contents of the internal RAM are undefined. This implies that the GPRs
(R15...R0) and the PEC source and destination pointers (SRCP7...SRCP0,
DSTP7...DSTP0) which are mapped into the internal RAM are also unchanged after a
hardware reset, software reset or watchdog reset, but are undefined after power-on.
Ports and External Bus Configuration during Reset
During the internal reset sequence all of the C165UTAH's port pins are configured as
inputs by clearing the associated direction registers, and their pin drivers are switched to
the high impedance state. This ensures that the C165UTAH and external devices will not
try to drive the same pin to different levels. Pin ALE is held low through an internal
pulldown, and pins RD and WR are held high through internal pullups. Also the pins
selected for CS output will be pulled high.
The registers SYSCON and BUSCON0 are initialized according to the configuration
selected via PORT0:
• the Bus Type field (BTYP) in register BUSCON0 is initialized according to P0L.7 and
P0L.6
• bit BUSACT0 in register BUSCON0 is set to ‘1’
• bit ALECTL0 in register BUSCON0 is set to ‘1’
• bit ROMEN in register SYSCON will be cleared to ‘0’
• bit BYTDIS in register SYSCON is set according to the data bus width
Note: In the C165UTAH, pin EA must always be set to ’0’. The "internal start" (EA=’1’),
known from other Infineon C16x devices is not supported.
The other bits of register BUSCON0, and the other BUSCON registers are cleared. This
default initialization selects the slowest possible external accesses using the configured
bus type. The Ready function is disabled at the end of the internal system reset.
When the internal reset has completed, the configuration of PORT0, PORT1, Port 4,
Port 6 and of the BHE signal (High Byte Enable, alternate function of P3.12) depends on
the bus type which was selected during reset. When any of the external bus modes was
selected during reset, PORT0 (and PORT1) will operate in the selected bus mode. Port 4
will output the selected number of segment address lines (all zero after reset) and Port 6
will drive the selected number of CS lines (CS0 will be ‘0’, while the other active CS lines
will be ‘1’). When no memory accesses above 64 K are to be performed, segmentation
may be disabled.
When the on-chip bootstrap loader was activated during reset, pin TxD0 (alternate
function of P3.10) will be switched to output mode after the reception of the zero byte.
All other pins remain in the high-impedance state until they are changed by software or
peripheral operation.
Data Sheet
444
2001-02-23
C165UTAH
System Reset
Application-Specific Initialization Routine
After the internal reset condition is removed the C165UTAH fetches the first instruction
from location 00’0000H, which is the first vector in the trap/interrupt vector table, the reset
vector. 4 words (locations 00’0000H through 00’0007H) are provided in this table to start
the initialization after reset. As a rule, this location holds a branch instruction to the actual
initialization routine that may be located anywhere in the address space.
Note: When the Bootstrap Loader Mode was activated during a hardware reset the
C165UTAH does not fetch instructions from location 00’0000H but rather expects
data via serial interface ASC.
The first instruction is fetched from external memory. To decrease the number of
instructions required to initialize the C165UTAH, each peripheral is programmed to a
default configuration upon reset, but is disabled from operation. These default
configurations can be found in the descriptions of the individual peripherals.
During the software design phase, portions of the internal memory space must be
assigned to register banks and system stack. When initializating the stack pointer (SP)
and the context pointer (CP), it must be ensured that these registers are initialized before
any GPR or stack operation is performed. This includes interrupt processing, which is
disabled upon completion of the internal reset, and should remain disabled until the SP
is initialized.
Note: Traps (incl. NMI) may occur, even though the interrupt system is still disabled.
In addition, the stack overflow (STKOV) and the stack underflow (STKUN) registers
should be initialized. After reset, the CP, SP, and STKUN registers all contain the same
reset value 00’FC00H, while the STKOV register contains 00’FA00H. With the default
reset initialization, 256 words of system stack are available, where the system stack
selected by the SP grows downwards from 00’FBFEH, while the register bank selected
by the CP grows upwards from 00’FC00H.
Based on the application, the user may wish to initialize portions of the internal memory
before normal program operation. Once the register bank has been selected by
programming the CP register, the desired portions of the internal memory can easily be
initialized via indirect addressing.
At the end of the initialization, the interrupt system may be globally enabled by setting bit
IEN in register PSW. Care must be taken not to enable the interrupt system before the
initialization is complete.
The software initialization routine should be terminated with the EINIT instruction. This
instruction has been implemented as a protected instruction. Execution of the EINIT
instruction...
• disables the action of the DISWDT instruction,
• disables write accesses to register SYSCON,
Data Sheet
445
2001-02-23
C165UTAH
System Reset
Note: All configurations regarding register SYSCON (enable CLKOUT, stacksize,
etc.) must be selected before the execution of EINIT.
• disables write access to registers SYSCON2 and SYSCON3 (further write accesses
to SYSCON2 and SYSCON3 can be executed only using a special unlock
mechanism),
• clears the reset source detection bits in register WDTCON,
• causes the RSTOUT pin to go HIGH. This signal can be used to indicate the end of
the initialization routine and the proper operation of the microcontroller to external
hardware.
19.1
System Startup Configuration
Although most of the programmable features of the C165UTAH are either selected
during the initialization phase or repeatedly during program execution, there are some
features that must be selected earlier, because they are used for the first access of the
program execution.
These selections are made during reset via the pins of PORT0, which are read at the
end of the internal reset sequence. During reset internal pullup devices are active on the
PORT0 lines, so their input level is high, if the respective pin is left open, or is low, if the
respective pin is connected to an external pulldown device. With the coding of the
selections, as shown below, in many cases the default option, ie. high level, can be used.
The value on the upper byte of PORT0 (P0H) is latched into register RP0H upon reset,
the value on the lower byte (P0L) directly influences the BUSCON0 register (bus mode)
or the internal control logic of the C165UTAH.
Data Sheet
446
2001-02-23
C165UTAH
System Reset
H.7 H.6 H.5 H.4 H.3 H.2 H.1 H.0 L.7 L.6 L.5 L.4 L.3 L.2 L.1 L.0
SALSEL
CSSEL
WRC
Clock
Generator
Port 4
Logic
SMOD
BUSTYP
RP0H
CLKCFG
ADP
’1’
Internal Control Logic
P0L.0
must
always
be ’1’
Port 6
Logic
SYSCON
BUSCON0
Configuration latched in on hardware reset
H.7 H.6 H.5 H.4 H.3 H.2 H.1 H.0 L.7 L.6 L.5 L.4 L.3 L.2 L.1 L.0
Configuration latched in on software and/or WDT reset
H.4 H.3 H.2 H.1 H.0 L.7 L.6
Figure 140
L.1 L.0
PORT0 Configuration during Reset
Note: The configuration on pins P0H.7:P0H.5 (CLKCFG) and P0L5:P0L.2 (SMOD) is
latched in on a hardware triggered reset only and will not be evaluated by the
C165UTAH on a software and/or WDT reset.
The configuration via P0H is latched in register RP0H for subsequent evaluation by
software. Register RP0H is described in chapter “The External Bus Interface”.
Note: The reset configuration needs to be held on port P0 throughout the start-up phase
until the C165UTAH takes over control of the external XBUS. This is first indicated
by driving the XBUS output lines ALE, RD, WR/WRL, CS, P4, P1H and P1L. Since
it might prove infeasible to detect the change from tristate to a strongly driven
value, the first rising edge of ALE can be used for indication of the end of the reset
configuration hold time. The first rising edge of ALE occurs 4 CPU cycles after
taking control of the external bus.
The following describes the different selections that are offered for reset configuration.
The default modes refer to pins at high level, ie. without external pulldown devices
connected. Table 102 shows a summary of all modes, supported by the C165UTAH.
Data Sheet
447
2001-02-23
C165UTAH
System Reset
Note: The Emulation Mode, known from other C16x Infineon devices, is not supported
by the C165UTAH. Make sure, on pin P0L.0 a HIGH signal is always latched in.
HIGH on P0L.0 is the default configuration and is supported by the internal pull-up
device.
Table 102
C165UTAH’s Supported Modes and Related Reset Configurations
P0L.5 : P0L.2
(SMOD)
P0L.1 (ADP)
Selected Mode
xxxx
0
Adapt Mode
1111
1
Normal Mode
0001
1
Internal Boot-ROM Read-Out
1011
1
Bootstrap-Loader Mode
1101
1
Selftest
Adapt Mode
Pin P0L.1 (ADP) selects the Adapt Mode when low during reset. It is latched with the
rising edge of RSTIN. In this mode, the C165UTAH goes into a passive state, which is
similar to its state during reset. The pins of the C165UTAH float to tristate or are
deactivated via internal pullup/pulldown devices, as described for the reset state. In
addition also the RSTOUT pin floats to tristate rather than be driven low, and the on-chip
oscillator is switched off.
This mode allows switching a C165UTAH that is mounted to a board virtually off, so an
emulator may control the board’s circuitry, even though the original C165UTAH remains
in its place. The original C165UTAH also may resume to control the board after a reset
sequence with P0L.1 high.
Default: Adapt Mode is off.
Bootstrap Loader Mode
Pin P0L.4 (BSL) activates the on-chip bootstrap loader, when low during reset. The
bootstrap loader allows moving the start code into the internal RAM of the C165UTAH
via the serial interface ASC. The C165UTAH will remain in bootstrap loader mode until
a hardware reset with P0L.4 high or a software reset.
Default: The C165UTAH starts fetching code from location 00’0000H, the bootstrap
loader is off.
External Bus Type
Pins P0L.7 and P0L.6 (BUSTYP) select the external bus type during reset, if an external
start is selected via pin EA. This allows the configuration of the external bus interface of
Data Sheet
448
2001-02-23
C165UTAH
System Reset
the C165UTAH even for the first code fetch after reset. The two bits are copied into bit
field BTYP of register BUSCON0. P0L.7 controls the data bus width, while P0L.6
controls the address output (multiplexed or demultiplexed). This bit field may be changed
via software after reset, if required.
BTYP
Encoding
External Data Bus Width
External Address Bus Mode
00
8-bit Data
Demultiplexed Addresses
01
8-bit Data
Multiplexed Addresses
10
16-bit Data
Demultiplexed Addresses
11
16-bit Data
Multiplexed Addresses
PORT0 and PORT1 are automatically switched to the selected bus mode. In multiplexed
bus modes PORT0 drives both the 16-bit intra-segment address and the output data,
while PORT1 remains in high impedance state as long as no demultiplexed bus is
selected via one of the BUSCON registers. In demultiplexed bus modes PORT1 drives
the 16-bit intra-segment address, while PORT0 or P0L (according to the selected data
bus width) drives the output data.
For a 16-bit data bus BHE is automatically enabled, for an 8-bit data bus BHE is disabled
via bit BYTDIS in register SYSCON.
Default: 16-bit data bus with multiplexed addresses.
Note: If an internal start is selected via pin EA, these two pins are disregarded and bit
field BTYP of register BUSCON0 is cleared.
Write Configuration
Pin P0H.0 (WRC) selects the initial operation of the control pins WR and BHE during
reset. When high, this pin selects the standard function, ie. WR control and BHE. When
low, it selects the alternate configuration, ie. WRH and WRL. Thus even the first access
after a reset can go to a memory controlled via WRH and WRL. This bit is latched in
register RP0H and its inverted value is copied into bit WRCFG in register SYSCON.
Default: Standard function (WR control and BHE).
Chip Select Lines
Pins P0H.2 and P0H.1 (CSSEL) define the number of active chip select signals during
reset. This allows the selection which pins of Port 6 drive external CS signals and which
are used for general purpose IO. The two bits are latched in register RP0H.
Default: All 5 chip select lines active (CS4...CS0).
Note: The selected number of CS signals cannot be changed via software after reset.
Data Sheet
449
2001-02-23
C165UTAH
System Reset
CSSEL
Chip Select Lines
Note
11
Five:
Default without pull-downs
10
None
01
Two:
CS1...CS0
P6.4..P6.2 free for GPI/O
00
Three:
CS2...CS0
P6.4..P6.3 free for GPI/O
CS4...CS0
Port 6 pins free for I/O
Segment Address Lines
Pins P0H.4 and P0H.3 (SALSEL) define the number of active segment address lines
during reset. This allows the selection which pins of Port 4 drive address lines and which
are used for general purpose IO. The two bits are latched in register RP0H. Depending
on the system architecture the required address space is chosen and accessible right
from the start, so the initialization routine can directly access all locations without prior
programming. The required pins of Port 4 are automatically switched to address output
mode.
SALSEL
Segment Address Lines
Directly accessible Address Space
11
Two:
A17...A16
256
KByte (Default without pull-downs)
10
Eight:
A22...A16
8
MByte (Maximum)
01
None
64
KByte (Minimum)
00
Four:
1
MByte
A19...A16
Even if not all segment address lines are enabled on Port 4, the C165UTAH internally
uses its complete 24-bit addressing mechanism. This allows the restriction of the width
of the effective address bus, while still deriving CS signals from the complete addresses.
Default: 2-bit segment address (A17...A16) allowing access to 256 KByte.
Note: The selected number of segment address lines cannot be changed via software
after reset.
Clock Generation Control
Pins P0H.7, P0H.6 and P0H.5 (CLKCFG) select the clock generation mode (on-chip
PLL) during reset. Please refer to Chapter 3.3, "Clock Generation Concept".
Data Sheet
450
2001-02-23
C165UTAH
Power Reduction Modes
20
Power Reduction Modes
Two different power reduction modes with different levels of power reduction have been
implemented in the C165UTAH, which may be entered under software control.
In Idle mode the CPU is stopped, while the peripherals continue their operation. Idle
mode can be terminated by any reset or interrupt request.
In Power Down mode both the CPU and the peripherals are stopped. Power Down
mode can only be terminated by a hardware reset.
Note: All external bus actions are completed before Idle or Power Down mode is
entered. However, Idle or Power Down mode is not entered if READY is enabled,
but has not been activated (driven low) during the last bus access.
20.1
Idle Mode
The power consumption of the C165UTAH microcontroller can be decreased by entering
Idle mode. In this mode all peripherals, including the watchdog timer, continue to
operate normally, only the CPU operation is halted.
Idle mode is entered after the IDLE instruction has been executed and the instruction
before the IDLE instruction has been completed. To prevent unintentional entry into Idle
mode, the IDLE instruction has been implemented as a protected 32-bit instruction.
Idle mode is terminated by interrupt requests from any enabled interrupt source whose
individual Interrupt Enable flag was set before the Idle mode was entered, regardless of
bit IEN.
For a request selected for CPU interrupt service the associated interrupt service routine
is entered if the priority level of the requesting source is higher than the current CPU
priority and the interrupt system is globally enabled. After the RETI (Return from
Interrupt) instruction of the interrupt service routine is executed the CPU continues
executing the program with the instruction following the IDLE instruction. Otherwise, if
the interrupt request cannot be serviced because of a too low priority or a globally
disabled interrupt system the CPU immediately resumes normal program execution with
the instruction following the IDLE instruction.
For a request which was programmed for PEC service a PEC data transfer is performed
if the priority level of this request is higher than the current CPU priority and the interrupt
system is globally enabled. After the PEC data transfer has been completed the CPU
remains in Idle mode. Otherwise, if the PEC request cannot be serviced because of a
too low priority or a globally disabled interrupt system the CPU does not remain in Idle
mode but continues program execution with the instruction following the IDLE
instruction.
Data Sheet
451
2001-02-23
C165UTAH
Power Reduction Modes
denied
CPU Interrupt Request
accepted
Active
Mode
IDLE instruction
Denied PEC Request
Figure 141
Idle
Mode
Executed
PEC Request
Transitions between Idle mode and active mode
Idle mode can also be terminated by a Non-Maskable Interrupt, ie. a high to low transition
on the NMI pin. After Idle mode has been terminated by an interrupt or NMI request, the
interrupt system performs a round of prioritization to determine the highest priority
request. In the case of an NMI request, the NMI trap will always be entered.
Any interrupt request whose individual Interrupt Enable flag was set before Idle mode
was entered will terminate Idle mode regardless of the current CPU priority. The CPU
will not go back into Idle mode when a CPU interrupt request is detected, even when the
interrupt was not serviced because of a higher CPU priority or a globally disabled
interrupt system (IEN=’0’). The CPU will only go back into Idle mode when the interrupt
system is globally enabled (IEN=’1’) and a PEC service on a priority level higher than
the current CPU level is requested and executed.
Note: An interrupt request which is individually enabled and assigned to priority level 0
will terminate Idle mode. The associated interrupt vector will not be accessed,
however.
The watchdog timer may be used to monitor the Idle mode: an internal reset will be
generated if no interrupt or NMI request occurs before the watchdog timer overflows. To
prevent the watchdog timer from overflowing during Idle mode it must be programmed
to a reasonable time interval before Idle mode is entered.
The standard Idle mode can be additionally configured by programming the SYSCON3
register, using the flexible peripheral management functions. This is especially
advantages, because it is thus possible to activate only these peripherals also in Idle
mode which are really required for standby operation or for wakeup, reducing power
consumption to the absolute minimum for a specific peripheral operation during Idle
mode.
Data Sheet
452
2001-02-23
C165UTAH
Power Reduction Modes
20.2
Power Down Mode
To further reduce the power consumption the microcontroller can be switched to Power
Down mode. Clocking of all internal blocks is stopped, the contents of the internal RAM,
however, are preserved through the voltage supplied via the VCC pins. The watchdog
timer is stopped in Power Down mode. This mode can only be terminated by an external
hardware reset, ie. by asserting a low level on the RSTIN pin. This reset will initialize all
SFRs and ports to their default state, but will not change the contents of the internal
RAM.
There are two levels of protection against unintentionally entering Power Down mode.
First, the PWRDN (Power Down) instruction which is used to enter this mode has been
implemented as a protected 32-bit instruction. Second, this instruction is effective only
if the NMI (Non Maskable Interrupt) pin is externally pulled low while the PWRDN
instruction is executed. The microcontroller will enter Power Down mode after the
PWRDN instruction has completed.
This feature can be used in conjunction with an external power failure signal which pulls
the NMI pin low when a power failure is imminent. The microcontroller will enter the NMI
trap routine which can save the internal state into RAM. After the internal state has been
saved, the trap routine may set a flag or write a certain bit pattern into specific RAM
locations, and then execute the PWRDN instruction. If the NMI pin is still low at this time,
Power Down mode will be entered, otherwise program execution continues. During
power down the voltage at the VCC pins can be lowered to 2.5 V while the contents of the
internal RAM will still be preserved.
The initialization routine (executed upon reset) can check the identification flag or bit
pattern within RAM to determine whether the controller was initially switched on, or
whether it was properly restarted from Power Down mode.
20.3
Status of Output Pins during Idle and Power Down Mode
During Idle mode the CPU clocks are turned off, while all peripherals continue their
operation in the normal way. Therefore all ports pins, which are configured as general
purpose output pins, output the last data value which was written to their port output
latches. If the alternate output function of a port pin is used by a peripheral, the state of
the pin is determined by the operation of the peripheral.
Port pins which are used for bus control functions go into that state which represents the
inactive state of the respective function (eg. WR), or to a defined state which is based on
the last bus access (eg. BHE). Port pins which are used as external address/data bus
hold the address/data which was output during the last external memory access before
entry into Idle mode under the following conditions:
P0H outputs the high byte of the last address if a multiplexed bus mode with 8-bit data
bus is used, otherwise P0H is floating. P0L is always floating in Idle mode.
Data Sheet
453
2001-02-23
C165UTAH
Power Reduction Modes
PORT1 outputs the lower 16 bits of the last address if a demultiplexed bus mode is used,
otherwise the output pins of PORT1 represent the port latch data.
Port 4 outputs the segment address for the last access on those pins that were selected
during reset, otherwise the output pins of Port 4 represent the port latch data.
During Power Down mode the oscillator and the clocks to the CPU and to the
peripherals are turned off. Like in Idle mode, all port pins which are configured as general
purpose output pins output the last data value which was written to their port output
latches.
When the alternate output function of a port pin is used by a peripheral the state of this
pin is determined by the last action of the peripheral before the clocks were switched off.
The table below summarizes the state of all C165UTAH output pins during Idle and
Power Down mode.
C165UTAH
Idle Mode
Output Pin(s) No
external bus
Power Down Mode
External bus
enabled
No
external bus
External bus
enabled
ALE
Low
Low
Low
Low
RD, WR
High
High
High
High
CLKOUT
Active
Active
High
High
RSTOUT
1)
1)
1)
1)
P0L
Port Latch Data
Floating
Port Latch Data
Floating
Port Latch Data
A15...A8 2) / Float
2)
P0H
Port Latch Data
A15...A8
PORT1
Port Latch Data
Last Address 3) /
Port Latch Data
Port Latch Data
Last Address 3) /
Port Latch Data
Port 4
Port Latch Data
Port Latch Data/
Last segment
Port Latch Data
Port Latch Data/
Last segment
BHE
Port Latch Data
Last value
Port Latch Data
Last value
HLDA
Port Latch Data
Last value
Port Latch Data
Last value
BREQ
Port Latch Data
High
Port Latch Data
High
CSx
Port Latch Data
Last value 4)
Port Latch Data
Last value 4)
Other Port
Output Pins
Port Latch Data /
Port Latch Data /
Port Latch Data /
Port Latch Data /
Alternate Function Alternate Function Alternate Function Alternate Function
/ Float
Note:
1)
: High if EINIT was executed before entering Idle or Power Down mode, Low otherwise.
2)
: For multiplexed buses with 8-bit data bus.
Data Sheet
454
2001-02-23
C165UTAH
Power Reduction Modes
3)
: For demultiplexed buses.
4)
: The CS signal that corresponds to the last address remains active (low), all other enabled CS signals
remain inactive (high). By accessing an on-chip X-Periperal prior to entering a power save mode all
external CS signals can be deactivated.
20.4
Extended Power Management
Infineon Technologies C16x’s well known basic power reduction modes (Idle and Power
Down) are enhanced by a number of additional power management features. These
features can be combined or selectively used to reduce the controller’s power
consumption to the respective application’s possible minimum. According to the sense
of platform modularity, the extended power management functions are controlled by
different submodules and registers, as follows::
Sub Module
Control Register
Extended Power Management /Sleep Mode Control
SYSCON1
Flexible Clock Generation Management
SYSCON2
Flexible Peripheral Management
SYSCON3
The C165UTAH’s power management functions can be supplemented by the Real Time
Clock (RTC) timer with optional periodic wakeup from Sleep or Idle mode. The periodic
wakeup combines the drastically reduced power consumption in power reduction modes
(in conjunction with the additional power management features) with a high level of
system availability. External signals and events can be scanned (at a lower rate) by
periodically activating the CPU and selected peripherals which then return to powersave
mode after a short time. This greatly reduces the system’s average power consumption.
The RTC is fully controlled by the C165UTAH’s power reduction submodules.
The Extended Power Management Module controls the Sleep mode. The Sleep mode is
a new power management function which represents and is equal to a Power Down
mode but with exit/wakeup handling as in Idle mode. Wakeup out of Sleep state is
possible with all external interrupts (including alternate sources e.g. from ASC interface),
with NMI and with RTC interrupts. As in Idle mode also PEC requests are executed in
Sleep mode, resulting in an interruption and resumption of Sleep mode. The watchdog
timer is stopped in Sleep mode. The contents of internal RAM and of CBC’s registers are
preserved through the voltage supplied via the VDD pins.
As in Power Down mode, the Sleep mode may also be combined with a running real time
clock RTC. In Sleep mode the oscillators (RTC and selected oscillator optionally), the
PLL as well as the whole clock system is stopped as in power down state. This implies
- contrary to Idle mode - , that after wakeup the exit of Sleep mode and thus the start of
any CPU operation is normally delayed by the ramp-up time of the clock system
Data Sheet
455
2001-02-23
C165UTAH
Power Reduction Modes
(oscillator, PLL). Only when the PLL clock is locked on configured frequency, the system
clock is started and the following processing is identical to wakeup from Idle mode.
For description of Idle mode and its possibilities of configuration see Chapter 20.1, "Idle
Mode".
Register in Extended Power Management Module
:
Register
Description
SYSCON1
System configuration control register for sleep management
Note: SYSCON1 is a protected register; its security level is automatically set to full write
protection after execution of EINIT instruction.
The power reduction modes Idle and power down are extended by the Infineons C16x
devices newly introduced sleep mode.
20.4.1
Sleep Mode
The Sleep mode is a new power management function which represents and is equal to
a Power Down mode but with exit/wakeup handling as in Idle mode. Wakeup from Sleep
state is possible with all external interrupts (including alternate sources e.g. from SSC
interface), with NMI and with RTC interrupts. As in Idle mode also PEC requests are
executed in Sleep mode, resulting in an interruption and resumption of Sleep mode. The
watchdog timer is stopped in Sleep mode. The contents of internal RAM and registers
are preserved through the voltage supplied via the VDD pins.
Generally, the external bus and the XBUS are released during Sleep mode if enabled by
the Hold Enable bit HLDEN in the last Program Status Word PSW. If enabled, the signal
HLDA is active as long as the Sleep mode (as well as the Idle or Power Down mode) is
active. Only when the clock is available again after wakeup, the HOLD request signal is
sampled and the HLDA state continued until HOLD is deactivated.
As in Power Down mode, the Sleep mode may also be combined with a running real time
clock RTC. In Sleep mode the oscillators (RTC and selected oscillator optionally), the
PLL as well as the whole clock system is stopped as in Power Down state. This implies,
that after wakeup the exit of Sleep mode normally is delayed by the ramp-up time of the
clock system (oscillator, PLL).
Sleep mode is entered after the standard IDLE instruction (protected 32 bit instruction)
has been executed and the instruction before the IDLE instruction has been completed.
The selection between standard Idle mode and Sleep mode is controlled with the new
register SYSCON1 (see below).
Note: Sleep mode cannot be entered in Slow Down mode - the start of sleep mode and
wakeup is only possible in the normal clocking mode (PLL or direct drive) as
defined with the startup configuration on port P0. If Sleep mode shall be entered
Data Sheet
456
2001-02-23
C165UTAH
Power Reduction Modes
during Slow Down mode, automatically the standard Idle mode is selected as
configured with SYSCON3 register.
The Sleep mode is controlled by bitfield SLEEPCON within register SYSCON1.
SYSCON1 (F1DCH / EEH)
ESFR-b
Reset Value: 0000H
15
14
13
12
11
10
9
8
7
6
5
4
3
2
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
1
0
SLEEPCON
rw
rw
Note: SYSCON1 is write protected after the execution of EINIT unless it is released via
the unlock sequence.
General description of SYSCON1 bits:
Bit
Function
SLEEPCON
SLEEP Mode Configuration
‘0 0’: normal IDLE mode
‘0 1’: SLEEP mode with running RTC
‘1 0’: reserved
‘1 1’: SLEEP mode with stopped RTC and stopped OSC
Before entering Sleep mode with the IDLE instruction, the continuation of instruction
processing after termination of Sleep mode must be prepared as known from standard
Idle mode. For wakeup with interrupt, four general possibilities of continuation can be
selected, which are controlled (prepared) as follows:
• Continuation with first instruction after the IDLE instruction will be enabled if
– interrupts are globally disabled with the Interrupt Enable bit in PSW, or
– the interrupt is enabled by global (PSW) and by individual (interrupt control register)
enable bit, but the current CPU priority level (in PSW) of IDLE instruction is higher
than the interrupt level.
• Continuation with first instruction of dedicated interrupt service routine will be
selected if
the interrupt is enabled by global (in PSW) and by individual (interrupt control register)
enable bit, and the CPU priority level of IDLE instruction is lower than the interrupt
level, thus the enabled interrupt has highest priority. Additionally, PEC Transfer for this
interrupt is not enabled. The continuation with the dedicated service routine is always
performed in case of NMI hardware traps, independently of any enable bit or CPU
priority level.
• Execution of one PEC Transfer and resumption of Sleep mode will be selected if
the interrupt is enabled by global (in PSW) and by individual (interrupt control register)
enable bit, and the CPU priority level of IDLE instruction is lower than the interrupt
Data Sheet
457
2001-02-23
C165UTAH
Power Reduction Modes
level, thus the enabled interrupt has highest priority. Additionally, PEC Transfer for this
interrupt is enabled.
• Continuation with standard Idle mode as configured with register SYSCON3 if
the interrupt is not enabled with the individual Interrupt Enable flag in its interrupt
control register. Note: In standard Idle mode the watchdog timer has to be serviced.
For description of SYSCON3 see ´description of Peripheral Management Module.
As wakeup from Idle mode, wakeup from Sleep mode is performed with any enabled
interrupt request. Sleep mode is terminated and the before selected (and above
described) continuation of processing is executed, if one of the following interrupts occur:
• Fast External Interrupts (EXxINT). All fast external interrupts can be selected for
wakeup from Sleep mode by defining the related trigger transitions (edges) in the
EXICON register. All transition types are allowed also in Sleep mode.
• Alternate sources for Fast External Interrupts (EXxINT) as defined by the EXISEL
register. If selected, transitions on receive lines of serial interface controllers (ASC,
SSC, IOM-2, USB) can be used for wakeup from Sleep mode.
• RTC Timer T14 cyclic interrupt. For waking up from Sleep mode via RTC T14
interrupt, the RTC operation during Sleep mode must be selected in bitfield
SLEEPCON within register SYSCON1. Additionally, the RTC interrupt must be
enabled in the Interrupt Subnode Control register ISNC.
• RTC interrupt(s). With new real time clock, additional RTC interrupts can be enabled
via the Interrupt Subnode Control register RTCISNC. For wakeup, RTC operation
during Sleep mode must be selected in SYSCON1. This function is not supported in
C167CS.
• Non-Maskable Interrupt NMI. A high-to-low transition on NMI pin always terminates
the Sleep mode. The NMI input is filtered for spike suppression. (Planned: Input
signals shorter than 10ns are suppressed, detection is guaranteed for minimum 150ns
NMI signal).
Setup Lengthening Control (Start Delay)
Contrary to Idle mode, after wakup from Sleep mode at first the ramp-up of clock system
(oscillator and PLL) has to be controlled before any CPU operation can be started. Only
when the clock system is locked on configured frequency, the following processing is
identical to wakeup from Idle mode.
Note: This setup lengthening function is very similar to the start delay after HW-reset
because of reset lengthening conditions (see reset section). Setup lengthening
uses the same lengthening control signals as reset lengthening, but after setup the
program start is controlled with the trailing edge of a setup-active signal (contrary
to the RST signal in case of reset lengthening) which is provided to the core to
Data Sheet
458
2001-02-23
C165UTAH
Power Reduction Modes
delay the execution of first instruction. The system hold state during setup is
controlled by start delay of clock distribution.
Data Sheet
459
2001-02-23
C165UTAH
System Control Unit (CSCU)
21
System Control Unit (CSCU)
21.1
Introduction
The System Control Unit CSCU is used to control system specific tasks such as reset
control or power management (see previous Chapter "Power Reduction Modes") within
an on-chip system build around the Infineons Cell-Based Core C166. The power
management features of the CSCU provide effective means to realize standby
conditions for the system with an optimum balance between power reduction, peripheral
operation and system functionality. Additionally, the CSCU controls the modes and
operation of Real Time Clock RTC.
Summary of Features and Functions
The CSCU is characterized by the following functions:
• Central Control of system operation
– External interrupt and frequency output control
– Protection management for system control registers
• General XBUS peripherals control
– Control of visibility of XPERs
• Power management additional to the standard Idle and Power Down modes
– Sleep mode with wakeup from Power Down state by external interrupts
• Peripheral Management with individual clock and power control of peripherals
– Control of power down state of Flash modules during Idle
• Flexible clock generation management
– Programmable system Slow Down control with or without PLL
– Control interface for Clock Generation Unit
• Identification register block for chip and CSCU identification
– Device, revision, manufacturer
– CSCU identification register
21.2
Operational Overview
21.2.1
Overview of CSCU submodules
In the following paragraphs a functional overview of the different blocks and submodules
of the System Control Unit is presented.
XBUS Peripheral Configuration Block
In the C165UTAH, XBUS peripherals can be separately switched on or off by
programming the XPERCON register. If switched off, the respective peripheral is not
visible, meaning, that its address space and its functional pins are not occupied.
Data Sheet
460
2001-02-23
C165UTAH
System Control Unit (CSCU)
Note: In parallel to the XPER control with XPERCON register, the visibility of XPER
address spaces also is controlled with the BUSACT bits in respective XBCON
registers (in the C166 core)
Note: The XPER configuration is additionally controlled by means of flexible peripheral
management control (see Peripheral Management Module below) for power
reduction.
Register in XPER Configuration Block:
Register
Description
XPERCON
XBUS peripheral control of XPER visibility
System Control Block
This block has several system management functions.
The System Control Block controls the system register write protection, introduced for
the system control registers SYSCON1-3.
Note: The new register write protection especially supports modularity of design, and is
therefore not compatible with the previously known C16x release function, using
the release bitfield in SYSCON2 for write protection.
Additional control functions of the System Control Block:
–
–
–
–
–
Control of fast external interrupt inputs
Control of external interrupt source selection
Control of interrupt subnode for PLL and realtime clock interrupts
Control of spike suppression for fast external interrupts and NMI in Sleep mode
Clock output frequency control
The System Control Block provides the following registers:
:
Register
Description
SCUSLC
SCU security level command register
SCUSLS
SCU security level status and password register
EXICON
External interrupt control register (see Chapter 7.8.1, page 125)
EXISEL
External interrupt source selection control register (see Chapter 7.8.2, page 126)
ISNC
Interrupt subnode control register (see Chapter 7.8.3, page 127)
FOCON
Frequency output control register
Data Sheet
461
2001-02-23
C165UTAH
System Control Unit (CSCU)
Identification Register Block
All new derivatives of Infineons C16x microcontroller family provide a set of min. four
identification registers (expandable to eight). These registers offer information on the
chip manufacturer, the chip type and its memory properties.
Identification registers in the ID Block: :
Register
Description
IDMANUF
Manufacturer and department
IDCHIP
Identification of device and revision code
IDMEM
Identification of on-chip program memory (type, size)
IDPROG
Identification of programming/erasing voltage of on-chip program memory
IDMEM2
Identification of additional EEPROM, OTP, DRAM or Flash memory
21.3
XBUS Peripheral Configuration Block
The XBUS peripherals can be separately selected for being visible to the user by means
of corresponding selection bits in the XPERCON register. If not selected and therefore
not enabled (not activated with XPERCON bit), the peripheral’s address space including
SFR addresses and port pins are not occupied by the peripheral, thus the peripheral is
not visible and not available. To make an XBUS peripheral visible, its related bit in
XPERCON register must be set before the XPERs are globally enabled with XPEN-bit
in SYSCON register (during system initialization before EINIT instruction).
Note: After reset, no XBUS peripheral is selected in XPERCON register.
The XPERCON register is defined as follows:
XPERCON (F024H / 12H)
15
14
13
12
ESFRReset Value : 0000H
11
10
9
8
7
6
5
4
XPER XPER XPER
7
6
5
reserved
3
2
1
reserved
Bit Field
Bits
Type Value Description
reserved
15..8
rw
0
These bits are reserved and must be set to ’0’.
XPER7
7
rw
0
1
EPEC module is not visible
EPEC is selected and visible
XPER6
6
rw
0
1
USB module is not visible
USB is selected and visible
XPER5
5
rw
0
1
IOM-2 module is not visible
IOM-2 is selected and visible
reserved
4..0
rw
0
These bits are reserved and must be set to Zero
Data Sheet
462
0
2001-02-23
C165UTAH
System Control Unit (CSCU)
Note: The CSCU provides per XPERCON bit one enable signal for XPER visibility
control. These enable signals are routed to the core to be combined with
selectable (per module pins) BUSACT functions of XBCON registers.
21.4
System Control Block
21.4.1
Register Write Protection
The System Control Unit CSCU provides two different protection types of registers:
• Unprotected Registers
• Protectable Registers
The unprotected registers allow reading and writing (if not read-only) of register values
without any restrictions. However, the write access of the protectable registers (security
registers) can be programmed for three different modes of security level, whereas the
read access is always unprotected:
• Write Protected Mode
• Low Protected Mode
• Unprotected Mode
In write protected mode the registers can not be accessed by a write command. However
in low protected mode the registers can be written with a special command sequence
(see desription below). If the registers are set to unprotected mode, all write accesses
are possible.
Some register controlled functions and modes which are critical for the C165UTAH’s
operation are locked after the execution of EINIT, so these vital system functions cannot
be changed inadvertently eg. by software errors. However, as these security registers
control also the power management they need to be accessed during operation to select
the appropriate mode.
The switching between the different security levels is controlled by a state machine. Via
a password and a command sequence the security levels can be changed. After reset
always the unprotected mode is automatically selected. The EINIT command switches
the security level automatically to protected mode.
The low protected mode is especially important for a standby state of the application.
This mode allows fast accesses within two commands to the protected registers without
removing the protection completely.
Security Level Switching
Two registers are provided for switching the security level, the security level command
register SCUSLC and the security level status register SCUSLS . The security level
command register SCUSLC is used to control the state machine for switching the
security level. The SCUSLC register is loaded with the different commands of the
Data Sheet
463
2001-02-23
C165UTAH
System Control Unit (CSCU)
command sequence necessary to control a change of the security level. It is also used
for the one unlock command, which is necessary in the low protected mode to access
one protected register. The commands of the (unlock) command sequence are
characterized by certain pattern words (as AAAAH) or by patterns combined with an 8bit password. For command definition see the following state diagram (figure below). The
new password is defined with command 3 and stored in the according 8-bit field in the
SCUSLS register.
The SCUSLC register is defined as follow
SCUSLC (F0C0H / 60H)
15
14
13
ESFRReset Value: 0000H
12
11
10
9
8
7
6
5
4
3
2
1
0
COMMAND
rw
The command definition is described in Figure 142.
The security level status register SCUSLS is a read only register which shows the
current password, the actual security level and the state of the switching statemachine.
The SCUSLS is defined as followes:
SCUSLS (F0C2H / 61H)
15
14
13
ESFRReset Value: 0000H
12
11
STATE
SL
r
r
10
9
8
7
6
5
reserved
4
3
2
1
PASSWORD
r
Bit
Function
PASSWORD
Current Password
SL
Security Level
0 0:
Unprotected write mode
0 1:
Low protected mode
1 0:
Reserved
1 1:
Write protected mode
STATE
Actual State
0 0 0: Wait for first command (command 0)
0 0 1: Wait for command 1
0 1 0: Wait for command 2
0 1 1: Wait for new security level and for new password (command 3)
1 0 0: Security registers are unlocked; access to one register is possible
( only in low protected mode)
1 0 1: Reserved
1 1 0: Reserved
1 1 1: Reserved
Data Sheet
0
464
2001-02-23
C165UTAH
System Control Unit (CSCU)
The following registers are defined as protected (security) registers:
• SYSCON1
• SYSCON2
• SYSCON3
The following state diagram, Figure 142, shows the state machine for security level
switching and for unlock command execution in low protected mode:
Command3
Command4
and low
protected
mode
Command0
State 0
any CSCU
register
write access
State 4
1)
or any other CSCU
register write access
Command1
State 3
Command2
or any other CSCU
register write
access
or any other CSCU
register write access
State 1
Command2
Reset
State 2
Command1
Command Command
Number
Note :
0
AAAAh
1
5554h
2
96h + inverse (old) password
3
000b + new level + 000b + new password
4
8Eh + inverse (new) password: Unlock
security register in low protected mode
1)
Only if the security level command register is accessed, the new security level
and the new password is valid.
Figure 142
Statemachine for Security Level Switching
Write Access in Low Protected Mode
The write access in low protected mode is also done via a command sequence. First the
specific command 4 (see figure above) with the current password has to be written to
register SCUSLC. After this command all security registers are unlocked until the next
Data Sheet
465
2001-02-23
C165UTAH
System Control Unit (CSCU)
write access to any CSCU register is done. Read access is always possible to all
registers of the CSCU and will not influence the command sequences. In register
SCUSCS the actual status of the command state machine can always be read.
It is recommendet to use an atomic sequence for all command sequences.
21.4.2
Clock Output Frequency Control
A clock output signal with programmable frequency (fOUT) can be output via pin FOUT.
This clock signal is generated via a reload counter, so the output frequency can be
selected in small steps. An optional toggle latch provides a clock signal with a 50% duty
cycle.
Signal fOUT always provides complete output periods (see Signal Waveforms below):
• When fOUT is started (FOEN-->’1’) FOCNT is loaded from FORV
• When fOUT is stopped (FOEN-->’0’) FOCNT is stopped when fOUT has reached (or is)
’0’.
FORV
FOEN
Ctrl.
MUX
fOUT
fCPU
FOCNT
FOTL
FOSS
Figure 143
Clock Output Signal Generation
Register FOCON provides control over the output signal generation.
FOCON (FFAAH / D5H)
15
14
13
12
SFR-b
11
FOEN FOSS
rw
rw
10
9
8
Reset Value: 0000H
7
6
FORV
-
FOTL
FOCNT
rw
-
rw
rw
Bit
Function
FOCNT
Frequency Output Counter
FOTL
Frequency Output Toggle Latch
Is toggled upon each underflow of FOCNT.
Data Sheet
466
5
4
3
2
1
0
2001-02-23
C165UTAH
System Control Unit (CSCU)
Bit
Function
FORV
Frequency Output Reload Value
Is copied to FOCNT upon each underflow of FOCNT.
FOSS
Frequency Output Signal Select
0: Output of the toggle latch: DC=50%.
1: Output of the reload counter: DC depends on FORV.
FOEN
Frequency Output Enable
0: Frequency output generation stops when signal fOUT is/gets low.
1: FOCNT is running, fOUT is gated to pin. 1st reload after 0-1 transition.
Note: It is not recommended to write to any part of bitfield FOCNT, especially not while
the counter is running. Writing to FOCNT prior to starting the counter is obsolete
because it will immediatley be reloaded from FORV. Writing to FOCNT during
operation may produce uninteded counter values.
Output Frequency Calculation
The output frequency can be calculated as fOUT = fCPU / ( (FORV+1) * 2(1-FOSS) ),
so fOUTmin = fCPU / 128 (FORV = 3FH, FOSS = ’0’),
and fOUTmax = fCPU / 1 (FORV = 00H, FOSS = ’1’).
fCPU
fOUT
1)
(FORV=0)
2)
fOUT
1)
(FORV=2)
2)
fOUT
1)
(FORV=5)
2)
FOEN-->’1’
1) FOSS=’1’, output of counter
2) FOSS=’0’, output of toggle latch
Figure 144
Data Sheet
FOEN-->’0’
The counter starts here
The counter stops here
Signal Waveforms
467
2001-02-23
C165UTAH
System Control Unit (CSCU)
Note: The output signal (for FOSS=’1’) is high for the duration of 1 fCPU cycle for all reload
values FORV > 0. For FORV = 0 the output signal corresponds to fCPU.
Connection to Output
Signal fOUT in the C165UTAH is an alternate function of pin P3.15/CLKOUT/FOUT.
The priority ranking is: P3.15 < FOUT < CLKOUT.
Direction
0
"1"
MUX
1
CLKEN
FOUT_active
0
PortLatch
0
MUX
1
fOUT
MUX
1
fCPU
Figure 145
Connection to Port Logic (Functional Approach)
Note: For the generation of fOUT pin FOUT must be switched to output, ie. DP3.15 =’1’.
While fOUT is disabled the pin is controlled by the port latch (see figure above). The
port latch P3.15 must be ’0’ in order to maintain the fOUT inactive level on the pin.
Clock Management Module
Flexible Clock Management
This module especially serves for power management support. Flexible clock
management includes programmable system slow down with additional control of power
down and optional real time clock. The slowdown operation is achieved by dividing the
oscillator clock by a programmable factor (1...32) resulting in a low frequency device
operation which significantly reduces the overall power consumption. The PLL may be
completely switched off in this mode.
This module also controls the oscillator selection (main or auxiliary) for Real Time Clock
and for Slow Down Divider. During Power Down mode, this block controls the operation
of RTC and ports.
The clock generation is controlled via register SYSCON2.
Data Sheet
468
2001-02-23
C165UTAH
System Control Unit (CSCU)
SYSCON2 (F1D0H / E8H) ESFR-b Reset Value: 0000.0000.UU00.0000B
15
14
CLK
LOCK
r
13
12
11
10
9
8
CLKREL
CLKCON
rw
rw
7
6
SCS RCS
rw
rw
5
4
3
2
1
PDCON
reserved
rw
rw
0
Bit
Function
PDCON
Power Down Control (during power down mode)
x0: RTC = Off,
Ports = On (default after reset).
x1: RTC = Off,
Ports = Off.
In power down mode, the RTC of the C165UTAH is always off. Bit 5 of
SYSCON2 is don’t care.
RCS
RTC Clock Source (not affected by a reset)
0:
RTC is switched to synchronous mode.
The input is derived from the CPU clock.
1:
RTC is switched to asynchronous mode. The input is derived
from the RTC_REF_CLK (oscillator clock).
SCS
SDD Clock Source (not affected by a reset)
Has to be set to ’0’.
CLKCON
Clock State Control
00: Running on configured basic frequency.
01: Running on slow down frequency, PLL ON.
10: Running on slow down frequency, PLL OFF.
11: Reserved. Do not use this combination.
CLKREL
Reload Counter Value for Slowdown Divider
CLKLOCK
Clock Signal Status Bit
0:
Main oscillator is unstable or PLL is unlocked.
1:
Main oscillator is stable and PLL is locked.
Note: SYSCON2 is a security register. The security level is automatically set to write
protectection after execution of EINIT.
Note: To be compatibile to Infineon’s C167CR / 167CS, the Power Down Control
PDCON must be programmed to ’10’ (RTC = Off, Ports = On) during the
initialisation phase before the execution of EINIT instruction. The initial state after
reset is so defined that a reset does not interrupt the real time clock.
Data Sheet
469
2001-02-23
C165UTAH
System Control Unit (CSCU)
21.5
Peripheral Management Module
This module especially serves for power management support, controlling dynamically
the operation and thus the power consumption of the different peripherals on PD Bus
and XBUS. In each situation (eg. several system operating modes, standby, etc.) only
those peripherals may be kept running which are required for the respective functionality.
All others can be switched off. It also allows the operation control of whole groups of
peripherals.
Peripheral’s operation is disabled or enabled by controlling the specific clock input. This
function also is supported in idle and/or slow down mode.
The Real Time Clock (RTC) may be fed by a separate clock driver, so it can be kept
running even in power down mode.
While a peripheral is disabled its output pins remain in the state they had at the time of
disabling.
Note: In contrast to the peripheral management of Infineon’s 16x family the registers of
a disabled module are not accessable. Only the clock control register of the
platform peripheral is accessable. Note, the register access is not compatible to
the C167CS.
The user gets access to the flexible operation control of peripherals via the SYSCON3
register. This register is defined as follows:
SYSCON3 (F1D4H / EAH)
15
14
GRP reDIS serv
ed
rw
-
13
12
11
PLL
DIS
USBTDIS
rw
rw
ESFR-b
10
9
reserved
-
-
8
7
Reset Value:0000H
6
PER PER PER
DIS8 DIS7 DIS6
rw
rw
rw
5
4
reserved
-
-
3
2
1
0
PER
PER
PER PER
DIS3 DIS2 DIS1 DIS0
rw
rw
rw
rw
.
Bit
Function
PERDISx
Peripheral Disable Flag 0 - 14
‘0’: Module is enabled; the peripheral is supplied with the clock signal
‘1’: Module is disabled; the clock input of peripheral is disabled
GRPDIS
Peripheral Group Disable Flag (PD-Bus and X-Bus Peripherals)
‘0’: Peripheral clock driver for peripheral group is enabled
‘1’: Peripheral clock driver for peripheral group is disabled
Data Sheet
470
2001-02-23
C165UTAH
System Control Unit (CSCU)
Bit
Function
USBTDIS
USB Transceiver Disable Flag ONLY IF BIT XPERCON.6 = ’1’
‘00’: Normal operation, USB transceiver enabled
‘01’: Suspend mode, differencial transceiver switched off
‘10’: Reserved, do not use this combination.
‘11’: Full power down.
Note: If bit 6 of register XPERCON set to ’0’, the USB transceiver is always
switched off (power down mode), independently of bit USBTDIS.
PLLDIS
PLL Disable Flag (additional power savings / noise reduction feature)
‘0’: The PLL of the C165UTAH is switched on. This is the default configuration.
‘1’: The PLL is completely switched off. The free running feature and the
oscillator watchdog will not work, since there is no PLL clock at all.
Note: It makes sense to switch off the PLL in direct drive clock mode only.
Note : Please refer to Chapter 10.8, "Initialization of the C165UTAH’s X-peripherals", for complete
register initialization.
Note : SYSCON3 is an security register. The security level is automatically set to write protection
after execution of EINIT
PERDISx
Module Type
Module Function (examples for associated peripheral
modules)
0
PD-Bus Unit
RTC
Real Time Clock
1
PD-Bus Unit
ASC
USART
2
PD-Bus Unit
SSC
Synchronous Serial Channel
3
PD-Bus Unit
GPT12
General Purpose Timer Block
4..5
reserved
-
Reserved, has to be set to ’0’.
6
X-Bus Unit
IOM-2
IOM-2 Interface
7
X-Bus Unit
USB
Universal Serial Bus Interface
8
X-Bus Unit
EPEC
Extended PEC
9..14
reserved
-
Reserved, has to be set to ’0’.
21.6
Identification Registers
21.6.1
Introduction
The C165UTAH provides a set of 5 identification registers that offer information on the
chip, as manufacturer, chip type and its memory properties.
The ID registers are read only registers. A device that incorporates ID registers shall
return D5H as its Bootstrap Loader identification byte. A standardized routine may then
Data Sheet
471
2001-02-23
C165UTAH
System Control Unit (CSCU)
be downloaded which sends the ID registers to the serial interface, so the host gets exact
information about its partner.
21.6.2
ID Register Description
The ID registers are placed in the extended SFR area.
IDMANUF (F07EH / 3FH)
15
14
13
12
ESFR
11
10
9
8
7
6
5
4
3
2
MANUF
DEPT
r
r
Bit
Function
MANUF
Manufacturer
0C1H: Infineon Technologies JEDEC normalized manufacturer code
DEPT
Department
04H: Infineon’s Datacom Department
IDCHIP (F07CH / 3EH)
15
14
13
12
11
10
9
8
7
6
3
2
r
r
Device Revision Code
03H: actual device revision code
CHIPID
Device Identification
05H: Infineons C16x device identification
IDMEM (F07AH / 3DH)
12
1
0
1
0
ESFR
11
10
9
8
7
6
5
Type (’0H’)
Size (’000H’)
r
r
Data Sheet
4
Chip Revision Number
Revision
13
5
CHIPID
Function
14
0
ESFR
Bit
15
1
472
4
3
2
2001-02-23
C165UTAH
System Control Unit (CSCU)
Bit
Function
Size
Size of on-chip Program Memory
The size of the implemented program memory in terms of 4 K blocks, i.e..
Memory-size = *4 KByte.
000H: No program memory on the C165UTAH.
Type
Type of on-chip Program Memory
Identifies the memory type on this silicon.
0H:
ROMless
1H:
Mask ROM
EPROM
3H:
Flash
2H:
4H:
OTP
5H:
EEPROM
6H:
DRAM/SRAM
IDPROG (F078H / 3CH)
15
14
13
ESFR
12
11
10
9
8
7
6
5
4
3
PROGVPP (’00H’)
PROGVDD (’00H’)
r
r
2
1
0
Bit
Function
PROGVDD
Programming VDD Voltage
The voltage of the standard power supply pins required when programming or
erasing (if applicable) the on-chip program memory.
Formula: VDD = 20* / 256 [V]
00H: No program memory on the C165UTAH.
PROGVPP
Programming VPP Voltage
The voltage of the special programming power supply (if existent) required to
program or erase (if applicable) the on-chip program memory.
Formula: VPP = 20* / 256 [V]
00H: No program memory on the C165UTAH.
IDMEM2 (F076H / 3BH)
15
14
13
12
ESFR
11
10
9
8
7
6
5
Type (’0H’)
Size (’000H’)
r
r
4
3
2
1
0
Note: IDMEM2 describes the second block of (program) memory. This register is
dedicated to other Infineon devices containing Flash, EEPROM or DRAM
sections. Static RAM modules are not described with ID registers. Since there is
no program memory on the C165UTAH, IDMEM2 is set to ’0000H’.
Data Sheet
473
2001-02-23
C165UTAH
System Programming
22
System Programming
To aid in software development, a number of features has been incorporated into the
instruction set of the C165UTAH, including constructs for modularity, loops, and context
switching. In many cases commonly used instruction sequences have been simplified
while providing greater flexibility. The following programming features help to fully utilize
this instruction set.
Instructions Provided as Subsets of Instructions
In many cases, instructions found in other microcontrollers are provided as subsets of
more powerful instructions in the C165UTAH. This allows the same functionality to be
provided while decreasing the hardware required and decreasing decode complexity. In
order to aid assembly programming, these instructions, familiar from other
microcontrollers, can be built in macros, thus providing the same names.
Directly Substitutable Instructions are instructions known from other microcontrollers
that can be replaced by the following instructions of the C165UTAH:
Substituted Instruction
C165UTAH Instruction
Function
CLR
Rn
AND
Rn, #0H
Clear register
CPLB
Bit
BMOVN
Bit, Bit
Complement bit
DEC
Rn
SUB
Rn, #1H
Decrement register
INC
Rn
ADD
Rn, #1H
Increment register
SWAPB
Rn
ROR
Rn, #8H
Swap bytes within word
Modification of System Flags is performed using bit set or bit clear instructions (BSET,
BCLR ). All bit and word instructions can access the PSW register, so no instructions like
CLEAR CARRY or ENABLE INTERRUPTS are required.
External Memory Data Access does not require special instructions to load data
pointers or explicitly load and store external data. The C165UTAH provides a VonNeumann memory architecture and its on-chip hardware automatically detects accesses
to internal RAM, GPRs, and SFRs.
Multiplication and Division
Multiplication and division of words and double words is provided through multiple cycle
instructions implementing a Booth algorithm. Each instruction implicitly uses the 32-bit
register MD (MDL = lower 16 bits, MDH = upper 16 bits). The MDRIU flag (Multiply or
Divide Register In Use) in register MDC is set whenever either half of this register is
written to or when a multiply/divide instruction is started. It is cleared whenever the MDL
Data Sheet
474
2001-02-23
C165UTAH
System Programming
register is read. Because an interrupt can be acknowledged before the contents of
register MD are saved, this flag is required to alert interrupt routines, which require the
use of the multiply/divide hardware, so they can preserve register MD. This register,
however, only needs to be saved when an interrupt routine requires use of the MD
register and a previous task has not saved the current result. This flag is easily tested by
the Jump-on-Bit instructions.
Multiplication or division is simply performed by specifying the correct (signed or
unsigned) version of the multiply or divide instruction. The result is then stored in register
MD. The overflow flag (V) is set if the result from a multiply or divide instruction is greater
than 16 bits. This flag can be used to determine whether both word halfs must be
transferred from register MD. The high portion of register MD (MDH) must be moved into
the register file or memory first, in order to ensure that the MDRIU flag reflects the correct
state.
The following instruction sequence performs an unsigned 16 by 16-bit multiplication:
SAVE:
JNB
SCXT
MDRIU, START
MDC, #0010H
BSET
SAVED
PUSH
MDH
PUSH
MDL
START:
MULU
R1, R2
JMPR
cc_NV, COPYL
MOV
R3, MDH
COPYL:
MOV
R4, MDL
RESTORE:
JNB
SAVED, DONE
POP
MDL
POP
MDH
POP
MDC
BCLR
SAVED
DONE:
;Test if MD was in use.
;Save and clear control register,
;leaving MDRIU set
;(only required for interrupted
;multiply/divide instructions)
;Indicate the save operation
;Save previous MD contents...
;...on system stack
;Multiply 16·16 unsigned, Sets MDRIU
;Test for only 16-bit result
;Move high portion of MD
;Move low portion of MD, Clears MDRIU
;Test if MD registers were saved
;Restore registers
;Multiplication is completed,
;program continues
...
The above save sequence and the restore sequence after COPYL are only required if
the current routine could have interrupted a previous routine which contained a MUL or
DIV instruction. Register MDC is also saved because it is possible that a previous
routine's Multiply or Divide instruction was interrupted while in progress. In this case the
information about how to restart the instruction is contained in this register. Register
MDC must be cleared to be correctly initialized for a subsequent multiplication or
Data Sheet
475
2001-02-23
C165UTAH
System Programming
division. The old MDC contents must be popped from the stack before the RETI
instruction is executed.
For a division the user must first move the dividend into the MD register. If a 16/16-bit
division is specified, only the low portion of register MD must be loaded. The result is also
stored into register MD. The low portion (MDL) contains the integer result of the division,
while the high portion (MDH) contains the remainder.
The following instruction sequence performs a 32 by 16-bit division:
MOV
MOV
DIV
JMPR
MOV
MOV
MDH, R1
MDL, R2
R3
cc_V, ERROR
R3, MDH
R4, MDL
;Move dividend to MD register. Sets MDRIU
;Move low portion to MD
;Divide 32/16 signed, R3 holds divisor
;Test for divide overflow
;Move remainder to R3
;Move integer result to R4. Clears MDRIU
Whenever a multiply or divide instruction is interrupted while in progress, the address of
the interrupted instruction is pushed onto the stack and the MULIP flag in the PSW of the
interrupting routine is set. When the interrupt routine is exited with the RETI instruction,
this bit is implicitly tested before the old PSW is popped from the stack. If MULIP=’1’ the
multiply/divide instruction is re-read from the location popped from the stack (return
address) and will be completed after the RETI instruction has been executed.
Note: The MULIP flag is part of the context of the interrupted task. When the
interrupting routine does not return to the interrupted task (eg. scheduler switches
to another task) the MULIP flag must be set or cleared according to the context of
the task that is switched to.
BCD Calculations
No direct support for BCD calculations is provided in the C165UTAH. BCD calculations
are performed by converting BCD data to binary data, performing the desired
calculations using standard data types, and converting the result back to BCD data. Due
to the enhanced performance of division instructions binary data is quickly converted to
BCD data through division by 10D. Conversion from BCD data to binary data is enhanced
by multiple bit shift instructions. This provides similar performance compared to
instructions directly supporting BCD data types, while no additional hardware is required.
22.1
Stack Operations
The C165UTAH supports two types of stacks. The system stack is used implicitly by the
controller and is located in the internal RAM. The user stack provides stack access to the
user in either the internal or external memory. Both stack types grow from high memory
addresses to low memory addresses.
Data Sheet
476
2001-02-23
C165UTAH
System Programming
Internal System Stack
A system stack is provided to store return vectors, segment pointers, and processor
status for procedures and interrupt routines. A system register, SP, points to the top of
the stack. This pointer is decremented when data is pushed onto the stack, and
incremented when data is popped.
The internal system stack can also be used to temporarily store data or pass it between
subroutines or tasks. Instructions are provided to push or pop registers on/from the
system stack. However, in most cases the register banking scheme provides the best
performance for passing data between multiple tasks.
Note: The system stack allows the storage of words only. Bytes must either be
converted to words or the respective other byte must be disregarded.
Register SP can only be loaded with even byte addresses (The LSB of SP is
always '0').
Detection of stack overflow/underflow is supported by two registers, STKOV (Stack
Overflow Pointer) and STKUN (Stack Underflow Pointer). Specific system traps (Stack
Overflow trap, Stack Underflow trap) will be entered whenever the SP reaches either
boundary specified in these registers.
The contents of the stack pointer are compared to the contents of the overflow register,
whenever the SP is DECREMENTED either by a CALL, PUSH or SUB instruction. An
overflow trap will be entered, when the SP value is less than the value in the stack
overflow register.
The contents of the stack pointer are compared to the contents of the underflow register,
whenever the SP is INCREMENTED either by a RET, POP or ADD instruction. An
underflow trap will be entered, when the SP value is greater than the value in the stack
underflow register.
Note: When a value is MOVED into the stack pointer, NO check against the overflow/
underflow registers is performed.
In many cases the user will place a software reset instruction (SRST) into the stack
underflow and overflow trap service routines. This is an easy approach, which does not
require special programming. However, this approach assumes that the defined internal
stack is sufficient for the current software and that exceeding its upper or lower boundary
represents a fatal error.
It is also possible to use the stack underflow and stack overflow traps to cache portions
of a larger external stack. Only the portion of the system stack currently being used is
placed into the internal memory, thus allowing a greater portion of the internal RAM to
be used for program, data or register banking. This approach assumes no error but
requires a set of control routines (see below).
Data Sheet
477
2001-02-23
C165UTAH
System Programming
Circular (virtual) Stack
This basic technique allows pushing until the overflow boundary of the internal stack is
reached. At this point a portion of the stacked data must be saved into external memory
to create space for further stack pushes. This is called “stack flushing”. When executing
a number of return or pop instructions, the upper boundary (since the stack empties
upward to higher memory locations) is reached. The entries that have been previously
saved in external memory must now be restored. This is called “stack filling”. Because
procedure call instructions do not continue to nest infinitely and call and return
instructions alternate, flushing and filling normally occurs very infrequently. If this is not
true for a given program environment, this technique should not be used because of the
overhead of flushing and filling.
The basic mechanism is the transformation of the addresses of a virtual stack area,
controlled via registers SP, STKOV and STKUN, to a defined physical stack area within
the internal RAM via hardware. This virtual stack area covers all possible locations that
SP can point to, ie. 00’F000H through 00’FFFEH. STKOV and STKUN accept the same
4 KByte address range.
The size of the physical stack area within the internal RAM that effectively is used for
standard stack operations is defined via bitfield STKSZ in register SYSCON (see below).
Stack Size Internal RAM Addresses (Words)
(Words)
of Physical Stack
Significant Bits of
Stack Pointer SP
000B
256
00’FBFEH...00’FA00H (Default after Reset)
SP.8...SP.0
001B
128
00’FBFEH...00’FB00H
SP.7...SP.0
010B
64
00’FBFEH...00’FB80H
SP.6...SP.0
011B
32
00’FBFEH...00’FBC0H
SP.5...SP.0
100B
512
00’FBFEH...00’F800H (not for 1 Kbyte IRAM) SP.9...SP.0
101B
---
Reserved. Do not use this combination.
---
110B
---
Reserved. Do not use this combination.
---
111B
1024
00’FDFEH...00’FX00H (Note: No circular
SP.11...SP.0
stack)
00’FX00H represents the lower IRAM limit, ie.
1 KB: 00’FA00H, 2 KB: 00’F600H, 3 KB:
00’F200H
The virtual stack addresses are transformed to physical stack addresses by
concatenating the significant bits of the stack pointer register SP (see table) with the
complementary most significant bits of the upper limit of the physical stack area
(00’FBFEH). This transformation is done via hardware (see figure below).
Data Sheet
478
2001-02-23
C165UTAH
System Programming
The reset values (STKOV=FA00H, STKUN=FC00H, SP=FC00H, STKSZ=000B) map the
virtual stack area directly to the physical stack area and allow using the internal system
stack without any changes, provided that the 256 word area is not exceeded.
FBFEH 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 0
FBFEH 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 0
FB80H 1 1 1 1 1 0 1 1 1 0 0 0 0 0 0 0
Phys.A.
FA00H 1 1 1 1 1 0 1 0 0 0 0 0 0 0 0 0
FB80H 1 1 1 1 1 0 1 1 1 0 0 0 0 0 0 0
F800H 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0
After PUSH
After PUSH
FBFEH 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 0
FBFEH 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 0
FBFEH 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 0
Phys.A.
FBFEH 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 0
FB7EH 1 1 1 1 1 0 1 1 0 1 1 1 1 1 1 0
F7FEH 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 0
64 words
Figure 146
Stack Size
256 words
Physical Stack Address Generation
The following example demonstrates the circular stack mechanism which is also an
effect of this virtual stack mapping: First, register R1 is pushed onto the lowest physical
stack location according to the selected maximum stack size. With the following
instruction, register R2 will be pushed onto the highest physical stack location although
the SP is decremented by 2 as for the previous push operation.
MOV
SP, #0F802H
;Set SP before last entry...
;...of physical stack of 256 words
...
;(SP)=F802H: Physical stack addr.=FA02H
PUSH
R1
;(SP)=F800H: Physical stack addr.=FA00H
PUSH
R2
;(SP)=F7FEH: Physical stack addr.=FBFEH
The effect of the address transformation is that the physical stack addresses wrap
around from the end of the defined area to its beginning. When flushing and filling the
internal stack, this circular stack mechanism only requires to move that portion of stack
Data Sheet
479
2001-02-23
C165UTAH
System Programming
data which is really to be re-used (ie. the upper part of the defined stack area) instead of
the whole stack area. Stack data that remain in the lower part of the internal stack need
not be moved by the distance of the space being flushed or filled, as the stack pointer
automatically wraps around to the beginning of the freed part of the stack area.
Note: This circular stack technique is applicable for stack sizes of 32 to 512 words
(STKSZ = ‘000B’ to ‘100B’), it does not work with option STKSZ = ‘111B’, which
uses the complete internal RAM for system stack.
In the latter case the address transformation mechanism is deactivated.
When a boundary is reached, the stack underflow or overflow trap is entered, where the
user moves a predetermined portion of the internal stack to or from the external stack.
The amount of data transferred is determined by the average stack space required by
routines and the frequency of calls, traps, interrupts and returns. In most cases this will
be approximately one quarter to one tenth the size of the internal stack. Once the transfer
is complete, the boundary pointers are updated to reflect the newly allocated space on
the internal stack. Thus, the user is free to write code without concern for the internal
stack limits. Only the execution time required by the trap routines affects user programs.
The following procedure initializes the controller for usage of the circular stack
mechanism:
• Specify the size of the physical system stack area within the internal RAM (bitfield
STKSZ in register SYSCON).
• Define two pointers, which specify the upper and lower boundary of the external stack.
These values are then tested in the stack underflow and overflow trap routines when
moving data.
• Set the stack overflow pointer (STKOV) to the limit of the defined internal stack area
plus six words (for the reserved space to store two interrupt entries).
The internal stack will now fill until the overflow pointer is reached. After entry into the
overflow trap procedure, the top of the stack will be copied to the external memory. The
internal pointers will then be modified to reflect the newly allocated space. After exiting
from the trap procedure, the internal stack will wrap around to the top of the internal
stack, and continue to grow until the new value of the stack overflow pointer is reached.
When the underflow pointer is reached while the stack is meptied the bottom of stack is
reloaded from the external memory and the internal pointers are adjusted accordingly.
Linear Stack
The C165UTAH also offers a linear stack option (STKSZ = ‘111B’), where the system
stack may use the complete internal RAM area. This provides a large system stack
without requiring procedures to handle data transfers for a circular stack. However, this
method also leaves less RAM space for variables or code. The RAM area that may
effectively be consumed by the system stack is defined via the STKUN and STKOV
Data Sheet
480
2001-02-23
C165UTAH
System Programming
pointers. The underflow and overflow traps in this case serve for fatal error detection
only.
For the linear stack option all modifiable bits of register SP are used to access the
physical stack. Although the stack pointer may cover addresses from 00’F000H up to
00’FFFEH the (physical) system stack must be located within the internal RAM and
therefore may only use the address range 00’F600H to 00’FDFEH. It is the user’s
responsibility to restrict the system stack to the internal RAM range.
Note: Avoid stack accesses below the IRAM area (ESFR space and reserved area) and
within address range 00’FE00H and 00’FFFEH (SFR space).
Otherwise unpredictable results will occur.
User Stacks
User stacks provide the ability to create task specific data stacks and to off-load data
from the system stack. The user may push both bytes and words onto a user stack, but
is responsible for using the appropriate instructions when popping data from the specific
user stack. No hardware detection of overflow or underflow of a user stack is provided.
The following addressing modes allow implementation of user stacks:
[– Rw], Rb or [– Rw], Rw: Pre-decrement Indirect Addressing.
Used to push one byte or word onto a user stack. This mode is only available for MOV
instructions and can specify any GPR as the user stack pointer.
Rb, [Rw+] or Rw, [Rw+]: Post-increment Index Register Indirect Addressing.
Used to pop one byte or word from a user stack. This mode is available to most
instructions, but only GPRs R0-R3 can be specified as the user stack pointer.
Rb, [Rw+] or Rw, [Rw+]: Post-increment Indirect Addressing.
Used to pop one byte or word from a user stack. This mode is only available for MOV
instructions and can specify any GPR as the user stack pointer.
22.2
Register Banking
Register banking provides the user with an extremely fast method to switch user context.
A single machine cycle instruction saves the old bank and enters a new register bank.
Each register bank may assign up to 16 registers. Each register bank should be
allocated during coding based on the needs of each task. Once the internal memory has
been partitioned into a register bank space, internal stack space and a global internal
memory area, each bank pointer is then assigned. Thus, upon entry into a new task, the
appropriate bank pointer is used as the operand for the SCXT (switch context)
instruction. Upon exit from a task a simple POP instruction to the context pointer (CP)
restores the previous task's register bank.
Data Sheet
481
2001-02-23
C165UTAH
System Programming
22.3
Procedure Call Entry and Exit
To support modular programming a procedure mechanism is provided to allow coding of
frequently used portions of code into subroutines. The CALL and RET instructions store
and restore the value of the instruction pointer (IP) on the system stack before and after
a subroutine is executed.
Procedures may be called conditionally with instructions CALLA or CALLI, or be called
unconditionally using instructions CALLR or CALLS.
Note: Any data pushed onto the system stack during execution of the subroutine must
be popped before the RET instruction is executed.
Passing Parameters on the System Stack
Parameters may be passed via the system stack through PUSH instructions before the
subroutine is called, and POP instructions during execution of the subroutine. Base plus
offset indirect addressing also permits access to parameters without popping these
parameters from the stack during execution of the subroutine. Indirect addressing
provides a mechanism of accessing data referenced by data pointers, which are passed
to the subroutine.
In addition, two instructions have been implemented to allow one parameter to be
passed on the system stack without additional software overhead.
The PCALL (push and call) instruction first pushes the 'reg' operand and the IP contents
onto the system stack and then passes control to the subroutine specified by the 'caddr'
operand.
When exiting from the subroutine, the RETP (return and pop) instruction first pops the IP
and then the 'reg' operand from the system stack and returns to the calling program.
Cross Segment Subroutine Calls
Calls to subroutines in different segments require the use of the CALLS (call intersegment subroutine) instruction. This instruction preserves both the CSP (code segment
pointer) and IP on the system stack.
Upon return from the subroutine, a RETS (return from inter-segment subroutine)
instruction must be used to restore both the CSP and IP. This ensures that the next
instruction after the CALLS instruction is fetched from the correct segment.
Note: It is possible to use CALLS within the same segment, but still two words of the
stack are used to store both the IP and CSP.
Providing Local Registers for Subroutines
For subroutines which require local storage, the following methods are provided:
Data Sheet
482
2001-02-23
C165UTAH
System Programming
Alternate Bank of Registers: Upon entry into a subroutine, it is possible to specify a
new set of local registers by executing the SCXT (switch context) instruction. This
mechanism does not provide a method to recursively call a subroutine.
Saving and Restoring of Registers: To provide local registers, the contents of the
registers which are required for use by the subroutine can be pushed onto the stack and
the previous values be popped before returning to the calling routine. This is the most
common technique used today and it does provide a mechanism to support recursive
procedures. This method, however, requires two machine cycles per register stored on
the system stack (one cycle to PUSH the register, and one to POP the register).
Use of the System Stack for Local Registers: It is possible to use the SP and CP to
set up local subroutine register frames. This enables subroutines to dynamically allocate
local variables as needed within two machine cycles. A local frame is allocated by simply
subtracting the number of required local registers from the SP, and then moving the
value of the new SP to the CP.
This operation is supported through the SCXT (switch context) instruction with the
addressing mode 'reg, mem'. Using this instruction saves the old contents of the CP on
the system stack and moves the value of the SP into CP (see example below). Each local
register is then accessed as if it was a normal register. Upon exit from the subroutine,
first the old CP must be restored by popping it from the stack and then the number of
used local registers must be added to the SP to restore the allocated local space back
to the system stack.
Note: The system stack is growing downwards, while the register bank is growing
upwards.
Old Stack
Area
Old SP
New CP
New SP
R4
R3
R2
R1
R0
Old CP Contents
Newly
Allocated
Register
Bank
New
Stack
Area
Figure 147
Data Sheet
Local Registers
483
2001-02-23
C165UTAH
System Programming
The software to provide the local register bank for the example above is very compact:
After entering the subroutine:
SUB
SP, #10D
;Free 5 words in the current system stack
SCXT
CP, SP
;Set the new register bank pointer
Before exiting the subroutine:
POP
CP
;Restore the old register bank
ADD
SP, #10D
;Release the 5 words...
;...of the current system stack
22.4
Table Searching
A number of features have been included to decrease the execution time required to
search tables. First, branch delays are eliminated by the branch target cache after the
first iteration of the loop. Second, in non-sequentially searched tables, the enhanced
performance of the ALU allows more complicated hash algorithms to be processed to
obtain better table distribution. For sequentially searched tables, the auto-increment
indirect addressing mode and the E (end of table) flag stored in the PSW decrease the
number of overhead instructions executed in the loop.
The two examples below illustrate searching ordered tables and non-ordered tables,
respectively:
MOV
R0, #BASE
;Move table base into R0
CMP
R1, [R0+]
;Compare target to table entry
JMPR
cc_SGT, LOOP
;Test whether target has not been found
LOOP:
Note: The last entry in the table must be greater than the largest possible target.
MOV
R0, #BASE
;Move table base into R0
CMP
R1, [R0+]
;Compare target to table entry
JMPR
cc_NET, LOOP
;Test whether target is not found AND..
LOOP:
;..the end of table has not been reached.
Note: The last entry in the table must be equal to the lowest signed integer (8000H).
Data Sheet
484
2001-02-23
C165UTAH
System Programming
22.5
Peripheral Control and Interface
All communication between peripherals and the CPU is performed either by PEC
transfers to and from internal memory, or by explicitly addressing the SFRs associated
with the specific peripherals. After resetting the C165UTAH all peripherals (except the
watchdog timer) are disabled and initialized to default values. A desired configuration of
a specific peripheral is programmed using MOV instructions of either constants or
memory values to specific SFRs. Specific control flags may also be altered via bit
instructions.
Once in operation, the peripheral operates autonomously until an end condition is
reached at which time it requests a PEC transfer or requests CPU servicing through an
interrupt routine. Information may also be polled from peripherals through read accesses
to SFRs or bit operations including branch tests on specific control bits in SFRs. To
ensure proper allocation of peripherals among multiple tasks, a portion of the internal
memory has been made bit addressable to allow user semaphores. Instructions have
also been provided to lock out tasks via software by setting or clearing user specific bits
and conditionally branching based on these specific bits.
It is recommended that bit fields in control SFRs are updated using the BFLDH and
BFLDL instructions or a MOV instruction to avoid undesired intermediate modes of
operation which can occur, when BCLR/BSET or AND/OR instruction sequences are
used.
22.6
Floating Point Support
All floating point operations are performed using software. Standard multiple precision
instructions are used to perform calculations on data types that exceed the size of the
ALU. Multiple bit rotate and logic instructions allow easy masking and extracting of
portions of floating point numbers.
To decrease the time required to perform floating point operations, two hardware
features have been implemented in the CPU core. First, the PRIOR instruction aids in
normalizing floating point numbers by indicating the position of the first set bit in a GPR.
This result can the be used to rotate the floating point result accordingly. The second
feature aids in properly rounding the result of normalized floating point numbers through
the overflow (V) flag in the PSW. This flag is set when a one is shifted out of the carry bit
during shift right operations. The overflow flag and the carry flag are then used to round
the floating point result based on the desired rounding algorithm.
22.7
Trap/Interrupt Entry and Exit
Interrupt routines are entered when a requesting interrupt has a priority higher than the
current CPU priority level. Traps are entered regardless of the current CPU priority.
When either a trap or interrupt routine is entered, the state of the machine is preserved
on the system stack and a branch to the appropriate trap/interrupt vector is made.
Data Sheet
485
2001-02-23
C165UTAH
System Programming
All trap and interrupt routines require the use of the RETI (return from interrupt)
instruction to exit from the called routine. This instruction restores the system state from
the system stack and then branches back to the location where the trap or interrupt
occurred.
22.8
Unseparable Instruction Sequences
The instructions of the C165UTAH are very efficient (most instructions execute in one
machine cycle) and even the multiplication and division are interruptable in order to
minimize the response latency to interrupt requests (internal and external). In many
microcontroller applications this is vital.
Some special occasions, however, require certain code sequences (eg. semaphore
handling) to be uninterruptable to function properly. This can be provided by inhibiting
interrupts during the respective code sequence by disabling and enabling them before
and after the sequence. The necessary overhead may be reduced by means of the
ATOMIC instruction which allows locking 1...4 instructions to an unseparable code
sequence, during which the interrupt system (standard interrupts and PEC requests)
and Class A Traps (NMI, stack overflow/underflow) are disabled. A Class B Trap
(illegal opcode, illegal bus access, etc.), however, will interrupt the atomic sequence,
since it indicates a severe hardware problem. The interrupt inhibit caused by an ATOMIC
instruction gets active immediately, ie. no other instruction will enter the pipeline except
the one that follows the ATOMIC instruction, and no interrupt request will be serviced in
between. All instructions requiring multiple cycles or hold states are regarded as one
instruction in this sense (eg. MUL is one instruction). Any instruction type can be used
within an unseparable code sequence.
ATOMIC
#3
;The next 3 instr. are locked (No NOP requ.)
MOV
R0, #1234H
;Instr. 1 (no other instr. enters pipeline!)
MOV
R1, #5678H
;Instr. 2
MUL
R0, R1
;Instr. 3: MUL regarded as one instruction
MOV
R2, MDL
;This instruction is out of the scope...
;...of the ATOMIC instruction sequence
22.9
Overriding the DPP Addressing Mechanism
The standard mechanism to access data locations uses one of the four data page
pointers (DPPx), which selects a 16 KByte data page, and a 14-bit offset within this data
page. The four DPPs allow immediate access to up to 64 KByte of data. In applications
with big data arrays, especially in HLL applications using large memory models, this may
require frequent reloading of the DPPs, even for single accesses.
Data Sheet
486
2001-02-23
C165UTAH
System Programming
The EXTP (extend page) instruction allows switching to an arbitrary data page for 1...4
instructions without having to change the current DPPs.
EXTP
R15, #1
;The override page number is stored in R15
MOV
R0, [R14]
;The (14-bit) page offset is stored in R14
MOV
R1, [R13]
;This instruction uses the std. DPP scheme!
The EXTS (extend segment) instruction allows switching to a 64 KByte segment
oriented data access scheme for 1...4 instructions without having to change the current
DPPs. In this case all 16 bits of the operand address are used as segment offset, with
the segment taken from the EXTS instruction. This greatly simplifies address calculation
with continuous data like huge arrays in “C”.
EXTS
#15, #1
;The override seg. is 15 (0F’0000H..0F’FFFFH)
MOV
R0, [R14]
;The (16-bit) segment offset is stored in R14
MOV
R1, [R13]
;This instruction uses the std. DPP scheme!
Note: Instructions EXTP and EXTS inhibit interrupts the same way as ATOMIC.
Short Addressing in the Extended SFR (ESFR) Space
The short addressing modes of the C165UTAH (REG or BITOFF) implicitly access the
SFR space. The additional ESFR space would have to be accessed via long addressing
modes (MEM or [Rw]). The EXTR (extend register) instruction redirects accesses in
short addressing modes to the ESFR space for 1...4 instructions, so the additional
registers can be accessed this way, too.
The EXTPR and EXTSR instructions combine the DPP override mechanism with the
redirection to the ESFR space using a single instruction.
Note: Instructions EXTR, EXTPR and EXTSR inhibit interrupts the same way as
ATOMIC.
The switching to the ESFR area and data page overriding is checked by the
development tools or handled automatically.
Nested Locked Sequences
Each of the described extension instruction and the ATOMIC instruction starts an
internal “extension counter” counting the effected instructions. When another extension
or ATOMIC instruction is contained in the current locked sequence this counter is
restarted with the value of the new instruction. This allows the construction of locked
sequences longer than 4 instructions.
Note: • Interrupt latencies may be increased when using locked code sequences.
• PEC requests are not serviced during idle mode, if the IDLE instruction is part of
a locked sequence.
Data Sheet
487
2001-02-23
C165UTAH
System Programming
Code Memory Configuration during Reset
The control input pin EA (External Access) enables the user to define the address area
from which the first instructions after reset are fetched. When EA is low (‘0’) during reset,
the internal code memory is disabled and the first instructions are fetched from external
memory. When EA is high (‘1’) during reset, the internal code memory is globally enabled
and the first instructions are fetched from the internal memory.
Enabling and Disabling the Internal Code Memory After Reset
If the internal code memory does not contain an appropriate startup code, the system
may be booted from external memory, while the internal memory is enabled afterwards
to provide access to library routines, tables, etc.
If the internal code memory only contains the startup code and/or test software, the
system may be booted from internal memory, which may then be disabled, after the
software has switched to executing from (eg.) external memory, in order to free the
address space occupied by the internal code memory, which is now unnecessary.
22.10
Pits, Traps and Mines
Although handling the internal code memory provides powerful means to enhance the
overall performance and flexibility of a system, extreme care must be taken in order to
avoid a system crash. Instruction memory is the most crucial resource for the
C165UTAH and it must be made sure that it never runs out of it. The following
precautions help to take advantage of the methods mentioned above without
jeopardizing system security.
General Rules
When mapping the code memory no instruction or data accesses should be made to the
internal memory, otherwise unpredictable results may occur.
To avoid these problems, the instructions that configure the internal code memory
should be executed from external memory or from the on-chip RAM.
Whenever the internal code memory is disabled, enabled or remapped the DPPs must
be explicitly (re)loaded to enable correct data accesses to the internal and/or external
memory.
Data Sheet
488
2001-02-23
C165UTAH
Register Set
23
Register Set
This section summarizes all registers, which are implemented in the C165UTAH and
explains the description format which is used in the chapters describing the function and
layout of the SFRs.
For easy reference the registers are ordered according to two different keys (except for
GPRs):
• Ordered by address, to check which register a given address references,
• Ordered by register name, to find the location of a specific register.
23.1
Register Description Format
In the respective chapters the function and the layout of the SFRs is described in a
specific format which provides a number of details about the described special function
register. The example below shows how to interpret these details.
A word register looks like this:
REG_NAME (A16H / A8H)
E/SFR
Reset Value: * * * *H
15
14
13
12
11
10
9
8
7
6
res.
res.
res.
res.
res.
write
only
hw
bit
read
only
std
bit
hw
bit
bitfield
bitfield
-
-
-
-
-
w
rw
r
rw
rw
rw
rw
Bit
5
4
3
2
1
0
Function
bit(field)name Explanation of bit(field)name
Description of the functions controlled by this bit(field).
A byte register looks like this:
REG_NAME (A16H / A8H)
15
-
14
-
13
-
12
-
E/SFR
11
-
10
-
9
-
8
-
Reset Value: - - * *H
7
6
5
4
3
2
1
0
std
bit
hw
bit
bitfield
bitfield
rw
rw
rw
rw
Elements:
REG_NAME
A16 / A8
SFR/ESFR/XReg
(* *) * *
hwbit
Data Sheet
Name of this register
Long 16-bit address / Short 8-bit address
Register space (SFR, ESFR or External/XBUS Register)
Register contents after reset
0/1: defined value, ’X’: undefined, ’U’: unchanged (undefined (’X’) after power up)
Bits that are set/cleared by hardware are marked with a shaded access box
489
2001-02-23
C165UTAH
Register Set
23.2
CPU General Purpose Registers (GPRs)
The GPRs form the register bank that the CPU works with. This register bank may be
located anywhere within the internal RAM via the Context Pointer (CP). Due to the
addressing mechanism, GPR banks can only reside within the internal RAM. All GPRs
are bit-addressable.
Name
Physical 8-Bit
Address Address
Description
Reset
Value
R0
(CP) + 0
F0H
CPU General Purpose (Word) Register R0
UUUUH
R1
(CP) + 2
F1H
CPU General Purpose (Word) Register R1
UUUUH
R2
(CP) + 4
F2H
CPU General Purpose (Word) Register R2
UUUUH
R3
(CP) + 6
F3H
CPU General Purpose (Word) Register R3
UUUUH
R4
(CP) + 8
F4H
CPU General Purpose (Word) Register R4
UUUUH
R5
(CP) + 10 F5H
CPU General Purpose (Word) Register R5
UUUUH
R6
(CP) + 12 F6H
CPU General Purpose (Word) Register R6
UUUUH
R7
(CP) + 14 F7H
CPU General Purpose (Word) Register R7
UUUUH
R8
(CP) + 16 F8H
CPU General Purpose (Word) Register R8
UUUUH
R9
(CP) + 18 F9H
CPU General Purpose (Word) Register R9
UUUUH
R10
(CP) + 20 FAH
CPU General Purpose (Word) Register R10
UUUUH
R11
(CP) + 22 FBH
CPU General Purpose (Word) Register R11
UUUUH
R12
(CP) + 24 FCH
CPU General Purpose (Word) Register R12
UUUUH
R13
(CP) + 26 FDH
CPU General Purpose (Word) Register R13
UUUUH
R14
(CP) + 28 FEH
CPU General Purpose (Word) Register R14
UUUUH
R15
(CP) + 30 FFH
CPU General Purpose (Word) Register R15
UUUUH
The first 8 GPRs (R7...R0) may also be accessed bytewise. Other than with SFRs,
writing to a GPR byte does not affect the other byte of the respective GPR.
The respective halves of the byte-accessible registers receive special names:
Name
Physical 8-Bit
Address Address
Description
RL0
(CP) + 0
F0H
CPU General Purpose (Byte) Register RL0
UUH
RH0
(CP) + 1
F1H
CPU General Purpose (Byte) Register RH0
UUH
RL1
(CP) + 2
F2H
CPU General Purpose (Byte) Register RL1
UUH
Data Sheet
490
Reset
Value
2001-02-23
C165UTAH
Register Set
Name
Physical 8-Bit
Address Address
Description
RH1
(CP) + 3
F3H
CPU General Purpose (Byte) Register RH1
UUH
RL2
(CP) + 4
F4H
CPU General Purpose (Byte) Register RL2
UUH
RH2
(CP) + 5
F5H
CPU General Purpose (Byte) Register RH2
UUH
RL3
(CP) + 6
F6H
CPU General Purpose (Byte) Register RL3
UUH
RH3
(CP) + 7
F7H
CPU General Purpose (Byte) Register RH3
UUH
RL4
(CP) + 8
F8H
CPU General Purpose (Byte) Register RL4
UUH
RH4
(CP) + 9
F9H
CPU General Purpose (Byte) Register RH4
UUH
RL5
(CP) + 10 FAH
CPU General Purpose (Byte) Register RL5
UUH
RH5
(CP) + 11 FBH
CPU General Purpose (Byte) Register RH5
UUH
RL6
(CP) + 12 FCH
CPU General Purpose (Byte) Register RL6
UUH
RH6
(CP) + 13 FDH
CPU General Purpose (Byte) Register RH6
UUH
RL7
(CP) + 14 FEH
CPU General Purpose (Byte) Register RL7
UUH
RH7
(CP) + 14 FFH
CPU General Purpose (Byte) Register RH7
UUH
23.3
Reset
Value
Special Function Registers ordered by Address
The following table lists all SFRs which are implemented in the C165UTAH ordered by physical address.
Bit-addressable SFRs are marked with the letter “b” in column “Type”.
SFRs within the Extended SFR-Space (ESFRs) are marked with the letter “E” in column “Type”.
Table 103
Registers ordered by Address
Physical Register
Addr
Name
Type
8-bit
Addr
Description
Reset
Value
R0
SFR-b
F0H
General Purpose Register 0
UUUUH
R0
ESFR-b
F0H
General Purpose Register 0
UUUUH
R1
SFR-b
F1H
General Purpose Register 1
UUUUH
R1
ESFR-b
F1H
General Purpose Register 1
UUUUH
R10
ESFR-b
FAH
General Purpose Register 10
UUUUH
R10
SFR-b
FAH
General Purpose Register 10
UUUUH
R11
SFR-b
FBH
General Purpose Register 11
UUUUH
R11
ESFR-b
FBH
General Purpose Register 11
UUUUH
R12
SFR-b
FCH
General Purpose Register 12
UUUUH
R12
ESFR-b
FCH
General Purpose Register 12
UUUUH
R13
SFR-b
FDH
General Purpose Register 13
UUUUH
R13
ESFR-b
FDH
General Purpose Register 13
UUUUH
R14
SFR-b
FEH
General Purpose Register 14
UUUUH
Data Sheet
491
2001-02-23
C165UTAH
Register Set
Physical Register
Addr
Name
Type
8-bit
Addr
Description
Reset
Value
R14
ESFR-b
FEH
General Purpose Register 14
UUUUH
R15
ESFR-b
FFH
General Purpose Register 15
UUUUH
R15
SFR-b
FFH
General Purpose Register 15
UUUUH
R2
SFR-b
F2H
General Purpose Register 2
UUUUH
R2
ESFR-b
F2H
General Purpose Register 2
UUUUH
R3
SFR-b
F3H
General Purpose Register 3
UUUUH
R3
ESFR-b
F3H
General Purpose Register 3
UUUUH
R4
ESFR-b
F4H
General Purpose Register 4
UUUUH
R4
SFR-b
F4H
General Purpose Register 4
UUUUH
R5
ESFR-b
F5H
General Purpose Register 5
UUUUH
R5
SFR-b
F5H
General Purpose Register 5
UUUUH
R6
ESFR-b
F6H
General Purpose Register 6
UUUUH
R6
SFR-b
F6H
General Purpose Register 6
UUUUH
R7
ESFR-b
F7H
General Purpose Register 7
UUUUH
R7
SFR-b
F7H
General Purpose Register 7
UUUUH
R8
SFR-b
F8H
General Purpose Register 8
UUUUH
R8
ESFR-b
F8H
General Purpose Register 8
UUUUH
R9
SFR-b
F9H
General Purpose Register 9
UUUUH
R9
ESFR-b
F9H
General Purpose Register 9
UUUUH
F014H
XADRS1
ESFR
0AH
XBUS Address Select Register 1
F016H
XADRS2
ESFR
0BH
XBUS Address Select Register 2
F018H
XADRS3
ESFR
0CH
XBUS Address Select Register 3
F01AH
XADRS4
ESFR
0DH
XBUS Address Select Register 4
F01CH
XADRS5
ESFR
0EH
XBUS Address Select Register 5
F01EH
XADRS6
ESFR
0FH
XBUS Address Select Register 6
F024H
XPERCON
ESFR
12H
XBUS Peripheral Control Register
0401H
F076H
IDMEM2
ESFR
3BH
Identifier
0000H
F078H
IDPROG
ESFR
3CH
Identifier
0000H
F07AH
IDMEM
ESFR
3DH
Identifier
0000H
F07CH
IDCHIP
ESFR
3EH
Identifier
0503H
F07EH
IDMANUF
ESFR
3FH
Identifier
1824H
F0B0H
SSCTB
ESFR
58H
SSC Transmit Buffer (WO)
0000H
F0B2H
SSCRB
ESFR
59H
SSC Receive Buffer (RO)
xxxxH
F0B4H
SSCBR
ESFR
5AH
SSC Baudrate Register
0000H
F0B6H
SSCCLC
ESFR
5BH
SSC Clock Control Register
0000H
Data Sheet
492
2001-02-23
C165UTAH
Register Set
Physical Register
Addr
Name
Type
8-bit
Addr
Description
Reset
Value
F0C0H
SCUSLC
ESFR
60H
Security Level Control Register
0000H
F0C2H
SCUSLS
ESFR
61H
Security Level Status Register
0000H
F0C8H
RTCCLC
ESFR
64H
RTC Clock Control Register
0000H
F0CCH
RTCRELL
ESFR
66H
RTC Timer Reload Register Low
0000H
F0CEH
RTCRELH
ESFR
67H
RTC Timer Reload Register High
0000H
F0D0H
T14REL
ESFR
68H
Timer 14 Reload Register
nH
F0D2H
T14
ESFR
69H
Timer 14 Register
nH
F0D4H
RTCL
ESFR
6AH
RTC Timer Register Low
nH
F0D6H
RTCH
ESFR
6BH
RTC Timer Register High
nH
F0D8H
DTIDR
ESFR
6CH
Task ID register1)
0000H
F100H
DP0L
ESFR-b
80H
P0L Direction Control Register
00H
F102H
DP0H
ESFR-b
81H
P0H Direction Control Register
00H
F104H
DP1L
ESFR-b
82H
P1L Direction Control Register
00H
F106H
DP1H
ESFR-b
83H
P1H Direction Control Register
00H
F108H
RP0H
ESFR-b
84H
System Startup Configuration Register (RO)
xxH
F114H
XBCON1
ESFR-b
8AH
XBUS Control register 1: IOM-2 module
0000H
F116H
XBCON2
ESFR-b
8BH
XBUS Control register 2: USB module
0000H
F118H
XBCON3
ESFR-b
8CH
XBUS Control register 3: EPEC module
0000H
F11AH
XBCON4
ESFR-b
8DH
XBUS Control register 4: reserved
0000H
F11CH
XBCON5
ESFR-b
8EH
XBUS Control register 5: reserved
0000H
F11EH
XBCON6
ESFR-b
8FH
XBUS Control register 6: reserved
0000H
F160H
UTD3IC
ESFR-b
B0H
UDC TX Done3 Interrupt Control Register
0000H
F162H
UTD4IC
ESFR-b
B1H
UDC TX Done4 Interrupt Control Register
0000H
F164H
UTD5IC
ESFR-b
B2H
UDC TX Done5 Interrupt Control Register
0000H
F166H
UTD6IC
ESFR-b
B3H
UDC TX Done6 Interrupt Control Register
0000H
F168H
UTD7IC
ESFR-b
B4H
UDC TX Done7 Interrupt Control Register
0000H
F16AH
URXRIC
ESFR-b
B5H
UDC RXRR Interrupt Control Register
0000H
F16CH
UTXRIC
ESFR-b
B6H
UDC TXWR Interrupt Control Register
0000H
F16EH
UCFGVIC
ESFR-b
B7H
UDC Config Val Interrupt Control Register
0000H
F170H
USOFIC
ESFR-b
B8H
UDC Start of Frame Interrupt Control Register
0000H
F172H
USSOIC
ESFR-b
B9H
UDC Suspend off Interrupt Control Register
0000H
F174H
USSIC
ESFR-b
BAH
UDC Suspend Interrupt Control Register
0000H
F176H
ULCDIC
ESFR-b
BBH
UDC Load Config Done Interrupt Control
Register
0000H
F178H
USETIC
ESFR-b
BCH
UDC SETUP Interrupt Control Register
0000H
F17AH
URD0IC
ESFR-b
BDH
UDC RX Done0 Interrupt Control Register
0000H
Data Sheet
493
2001-02-23
C165UTAH
Register Set
Physical Register
Addr
Name
Type
8-bit
Addr
Description
Reset
Value
F17CH
EPECIC
ESFR-b
BEH
EPEC Interrupt
0000H
F17EH
IOMC0TIC
ESFR-b
BFH
IOM-2 Channel0 TX Interrupt Control Register
0000H
F180H
PECCLIC
ESFR-b
C0H
PEC Channel Link Interrupt Control Register
0000H
F182H
IOMC0RIC
ESFR-b
C1H
IOM-2 Channel0 RX Interrupt Control Register
0000H
F184H
RTC_INTIC
ESFR-b
C2H
RTC_INT Sub Node Interrupt Register
0000H
F186H
XP0IC
ESFR-b
C3H
X-Bus Peripheral 0 UDC TXWR Interrupt Control
Register
0000H
F18AH
IOMC1TIC
ESFR-b
C5H
IOM-2 Channel1 TX Interrupt Control Register
0000H
F18CH
ABENDIC
ESFR-b
C6H
ASC Autobaud End Interrupt Control Register
0000H
F18EH
XP1IC
ESFR-b
C7H
X-Bus Peripheral 1 EPEC Interrupt Control
Register
0000H
F192H
IOMC1RIC
ESFR-b
C9H
IOM-2 Channel1 RX Interrupt Control Register
0000H
F194H
ABSTIC
ESFR-b
CAH
ASC Autobaud Start Interrupt Control Register
0000H
F196H
XP2IC
ESFR-b
CBH
X-Bus Peripheral 2 IOM-2 IO Interrupt Control
Register
0000H
F19AH
RES6IC
ESFR-b
CDH
reserved
0000H
F19CH
S0TBIC
ESFR-b
CEH
Serial Channel 0 Transmit Buffer IC Register
0000H
F19EH
XP3IC
ESFR-b
CFH
X-Bus Peripheral 3 PLL/RTC Interrupt Control
Register
0000H
F1C0H
EXICON
ESFR-b
E0H
External Interrupt Control Register
0000H
F1C2H
ODP2
ESFR-b
E1H
Port 2 Open Drain Control Register
0000H
F1C6H
ODP3
ESFR-b
E3H
Port 3 Open Drain Control Register
0000H
F1C8H
RTCISNC
ESFR-b
E4H
RTC Interrupt Sub Node Control Register
0000H
F1CAH
ODP4
ESFR-b
E5H
Port 4 Open Drain Control Register
00H
F1CCH
RTCCON
ESFR-b
E6H
RTC Control Register
00H
F1CEH
ODP6
ESFR-b
E7H
Port 6 Open Drain Control Register
00H
F1D0H
SYSCON2
ESFR-b
E8H
System Configuration Register 2/Clock Control
0000H
F1D2H
ODP7
ESFR-b
E9H
Port 7 Open Drain Control Register
00H
F1D4H
SYSCON3
ESFR-b
EAH
System Configuration Register 3/Periph.
Managem.
0000H
F1D6H
reserved
ESFR-b
EBH
reserved - do not use
0000H
F1D8H
reserved
ESFR-b
ECH
reserved - do not use
0000H
F1DAH
EXISEL
ESFR-b
EDH
External Interrupt Select Register
0000H
F1DCH
SYSCON1
ESFR-b
EEH
System Configuration Register 1/Sleep Mode
0000H
F1DEH
ISNC
ESFR-b
EFH
Interrupt Sub Node Control Register
0000H
FE00H
DPP0
SFR
00H
CPU Data Page Pointer 0 Register (10 bits)
0000H
FE02H
DPP1
SFR
01H
CPU Data Page Pointer 1 Register (10 bits)
0001H
Data Sheet
494
2001-02-23
C165UTAH
Register Set
Physical Register
Addr
Name
Type
8-bit
Addr
Description
Reset
Value
FE04H
DPP2
SFR
02H
CPU Data Page Pointer 2 Register (10 bits)
0002H
FE06H
DPP3
SFR
03H
CPU Data Page Pointer 3 Register (10 bits)
0003H
FE08H
CSP
SFR
04H
CPU Code Segment Pointer Register (8 bits)
0000H
FE0AH
EMUCON
SFR
05H
Emulation Control Register2)
xxxxH
FE0CH
MDH
SFR
06H
CPU Multiply Divide Register - High Word
0000H
FE0EH
MDL
SFR
07H
CPU Multiply Divide Register - Low Word
0000H
FE10H
CP
SFR
08H
CPU Context Pointer Register
FC00H
FE12H
SP
SFR
09H
CPU System Stack Pointer Register
FC00H
FE14H
STKOV
SFR
0AH
CPU Stack Overflow Pointer Register
FA00H
FE16H
STKUN
SFR
0BH
CPU Stack Underflow Pointer Register
FC00H
FE18H
ADDRSEL1
SFR
0CH
Address Select Register 1
0000H
FE1AH
ADDRSEL2
SFR
0DH
Address Select Register 2
0000H
FE1CH
ADDRSEL3
SFR
0EH
Address Select Register 3
0000H
FE1EH
ADDRSEL4
SFR
0FH
Address Select Register 4
0000H
FE22H
ODP0H
SFR
11H
Port 0 Open Drain Control Register High
0000H
FE24H
ODP1L
SFR
12H
Port 1 Open Drain Control Register Low
0000H
FE26H
ODP1H
SFR
13H
Port 1 Open Drain Control Register High
0000H
FE40H
T2
SFR
20H
GPT1 Timer 2 Register
0000H
FE42H
T3
SFR
21H
GPT1 Timer 3 Register
0000H
FE44H
T4
SFR
22H
GPT1 Timer 4 Register
0000H
FE46H
T5
SFR
23H
GPT2 Timer 5 Register
0000H
FE48H
T6
SFR
24H
GPT2 Timer 6 Register
0000H
FE4AH
CAPREL
SFR
25H
GPT1/2 Capture / Reload Register
0000H
FE4CH
GPTCLC
SFR
26H
GPT1/2 Clock Control Register
0000H
FE60H
P0LPUDSEL
SFR
30H
Port 0 Low Pull-Up/Down Select Register
xxFFH
FE62H
P0HPUDSEL
SFR
31H
Port 0 High Pull-Up/Down Select Register
xxFFH
FE64H
P0LPUDEN
SFR
32H
Port 0 Low Pull Switch On/Off Register
xxFF
FE66H
P0HPUDEN
SFR
33H
Port 0 High Pull Switch On/Off Register
xxFFH
FE68H
P0LPHEN
SFR
34H
Port 0 Low Pin Hold Enable Register
0000H
FE6AH
P0HPHEN
SFR
35H
Port 0 High Pin Hold Enable Register
0000H
FE6CH
P1LPUDSEL
SFR
36H
Port 1 Low Pull-Up/Down Select Register
0000H
FE6EH
P1HPUDSEL
SFR
37H
Port 1 High Pull-Up/Down Select Register
0000H
FE70H
P1LPUDEN
SFR
38H
Port 1 Low Pull Switch On/Off Register
0000H
FE72H
P1HPUDEN
SFR
39H
Port 1 High Pull Switch On/Off Register
0000H
FE74H
P1LPHEN
SFR
3AH
Port 1 Low Pin Hold Enable Register
0000H
Data Sheet
495
2001-02-23
C165UTAH
Register Set
Physical Register
Addr
Name
Type
8-bit
Addr
Description
Reset
Value
FE76H
P1HPHEN
SFR
3BH
Port 1 High Pin Hold Enable Register
0000H
FE78H
P2PUDSEL
SFR
3CH
Port 2 Pull-Up/Down Select Register
0000H
FE7AH
P2PUDEN
SFR
3DH
Port 2 Pull Switch On/Off Register
0000H
FE7CH
P2PHEN
SFR
3EH
Port 2 Pin Hold Enable Register
0000H
FE7EH
P3PUDSEL
SFR
3FH
Port 3 Pull-Up/Down Select Register
0000H
FE80H
P3PUDEN
SFR
40H
Port 3 Pull Switch On/Off Register
0000H
FE82H
P3PHEN
SFR
41H
Port 3 Pin Hold Enable Register
0000H
FE84H
P4PUDSEL
SFR
42H
Port 4 Pull-Up/Down Select Register
0000H
FE86H
P4PUDEN
SFR
43H
Port 4 Pull Switch On/Off Register
0000H
FE88H
P4PHEN
SFR
44H
Port 4 Pin Hold Enable Register
0000H
FE90H
P6PUDSEL
SFR
48H
Port 6 Pull-Up/Down Select Register
0000H
FE92H
P6PUDEN
SFR
49H
Port 6 Pull Switch On/Off Register
0000H
FE94H
P6PHEN
SFR
4AH
Port 6 Pin Hold Enable Register
0000H
FE96H
P7PUDSEL
SFR
4BH
Port 7 Pull-Up/Down Select Register
0000H
FE98H
P7PUDEN
SFR
4CH
Port 7 Pull Switch On/Off Register
0000H
FE9AH
P7PHEN
SFR
4DH
Port 7 Pin Hold Enable Register
0000H
FEAAH
S0PMW
SFR
55H
ASC IrDA PMW Control Register
0000H
FEAEH
WDT
SFR
57H
Watchdog Timer Register (RO)
0000H
FEB0H
S0TBUF
SFR
58H
Serial Channel 0 Transmit Buffer Register (WO)
0000H
FEB2H
S0RBUF
SFR
59H
Serial Channel 0 Receive Buffer Register (RO)
xxxxH
FEB4H
S0BG
SFR
5AH
Serial Channel 0 Baud Rate Generator Reload
Register
0000H
FEB6H
S0FDV
SFR
5BH
ASC Fractional Divide Register
0000H
FEC0H
PECC0
SFR
60H
PEC Channel 0 Control Register
0000H
FEC2H
PECC1
SFR
61H
PEC Channel 1 Control Register
0000H
FEC4H
PECC2
SFR
62H
PEC Channel 2 Control Register
0000H
FEC6H
PECC3
SFR
63H
PEC Channel 3 Control Register
0000H
FEC8H
PECC4
SFR
64H
PEC Channel 4 Control Register
0000H
FECAH
PECC5
SFR
65H
PEC Channel 5 Control Register
0000H
FECCH
PECC6
SFR
66H
PEC Channel 6 Control Register
0000H
FECEH
PECC7
SFR
67H
PEC Channel 7 Control Register
0000H
FED0H
PECSN0
SFR
68H
PEC Segment No Register
FED2H
PECSN1
SFR
69H
PEC Segment No Register
FED4H
PECSN2
SFR
6AH
PEC Segment No Register
FED6H
PECSN3
SFR
6BH
PEC Segment No Register
FED8H
PECSN4
SFR
6CH
PEC Segment No Register
Data Sheet
496
2001-02-23
C165UTAH
Register Set
Physical Register
Addr
Name
Type
8-bit
Addr
Description
FEDAH
PECSN5
SFR
6DH
PEC Segment No Register
FEDCH
PECSN6
SFR
6EH
PEC Segment No Register
FEDEH
PECSN7
SFR
6FH
PEC Segment No Register
FEF0H
PECXC0
SFR
78H
PEC Channel 0 Extended Control Register
FEF2H
PECXC2
SFR
79H
PEC Channel 2 Extended Control Register
FEF8H
ABS0CON
SFR
7CH
ASC Autobaud Control Register
0000H
FEFEH
ABSTAT
SFR
7FH
ASC Autobaud Status Register
0000H
FF00H
P0L
SFR-b
80H
Port 0 Low Register (Lower half)
00H
FF02H
P0H
SFR-b
81H
Port 0 High Register (Upper half)
00H
FF04H
P1L
SFR-b
82H
Port 1 Low Register (Lower half)
00H
FF06H
P1H
SFR-b
83H
Port 1 High Register (Upper half)
00H
FF0CH
BUSCON0
SFR-b
86H
Bus Configuration Register 0
0000H
FF0EH
MDC
SFR-b
87H
CPU Multiply Divide Control Register
0000H
FF10H
PSW
SFR-b
88H
CPU Program Status Word
0000H
FF12H
SYSCON
SFR-b
89H
CPU System Configuration Register
0xx0H
FF14H
BUSCON1
SFR-b
8AH
Bus Configuration Register 1
0000H
FF16H
BUSCON2
SFR-b
8BH
Bus Configuration Register 2
0000H
FF18H
BUSCON3
SFR-b
8CH
Bus Configuration Register 3
0000H
FF1AH
BUSCON4
SFR-b
8DH
Bus Configuration Register 4
0000H
FF1CH
ZEROS
SFR-b
8EH
Constant Value 0sRegister'
0000H
FF1EH
ONES
SFR-b
8FH
Constant Value 1sRegister'
FFFFH
FF40H
T2CON
SFR-b
A0H
GPT1 Timer 2 Control Register
0000H
FF42H
T3CON
SFR-b
A1H
GPT1 Timer 3 Control Register
0000H
FF44H
T4CON
SFR-b
A2H
GPT1 Timer 4 Control Register
0000H
FF46H
T5CON
SFR-b
A3H
GPT2 Timer 5 Control Register
0000H
FF48H
T6CON
SFR-b
A4H
GPT2 Timer 6 Control Register
0000H
FF60H
T2IC
SFR-b
B0H
GPT1 Timer 2 Interrupt Control Register
0000H
FF62H
T3IC
SFR-b
B1H
GPT1 Timer 3 Interrupt Control Register
0000H
FF64H
T4IC
SFR-b
B2H
GPT1 Timer 4 Interrupt Control Register
0000H
FF66H
T5IC
SFR-b
B3H
GPT2 Timer 5 Interrupt Control Register
0000H
FF68H
T6IC
SFR-b
B4H
GPT2 Timer 6 Interrupt Control Register
0000H
FF6AH
CRIC
SFR-b
B5H
GPT2 CAPREL Interrupt Control Register
0000H
FF6CH
S0TIC
SFR-b
B6H
Serial Channel 0 Transmit Interrupt Control
Register
0000H
FF6EH
S0RIC
SFR-b
B7H
Serial Channel 0 Receive Interrupt Control
Register
0000H
Data Sheet
497
Reset
Value
2001-02-23
C165UTAH
Register Set
Physical Register
Addr
Name
Type
8-bit
Addr
Description
FF70H
S0EIC
SFR-b
B8H
Serial Channel 0 Error Interrupt Control Register 0000H
FF72H
SSCTIC
SFR-b
B9H
SSC Transmit Interrupt Control Register
0000H
FF74H
SSCRIC
SFR-b
BAH
SSC Receive Interrupt Control Register
0000H
FF76H
SSCEIC
SFR-b
BBH
SSC Error Interrupt Control Register
0000H
FF78H
URD3IC
SFR-b
BCH
UDC RX Done3 Interrupt Control Register
0000H
FF7AH
URD4IC
SFR-b
BDH
UDC RX Done4 Interrupt Control Register
0000H
FF7CH
URD5IC
SFR-b
BEH
UDC RX Done5 Interrupt Control Register
0000H
FF7EH
URD6IC
SFR-b
BFH
UDC RX Done6 Interrupt Control Register
0000H
FF80H
URD7IC
SFR-b
C0H
UDC RX Done7 Interrupt Control Register
0000H
FF82H
UTD0IC
SFR-b
C1H
UDC TX Done0 Interrupt Control Register
0000H
FF84H
UTD1IC
SFR-b
C2H
UDC TX Done1 Interrupt Control Register
0000H
FF86H
UTD2IC
SFR-b
C3H
UDC TX Done2 Interrupt Control Register
0000H
FF88H
FEI0IC
SFR-b
C4H
Fast External Interrupt 0 Control Register
0000H
FF8AH
FEI1IC
SFR-b
C5H
Fast External Interrupt 1 Control Register
0000H
FF8CH
FEI2IC
SFR-b
C6H
Fast External Interrupt 2 Control Register
0000H
FF8EH
FEI3IC
SFR-b
C7H
Fast External Interrupt 3 Control Register
0000H
FF90H
FEI4IC
SFR-b
C8H
Fast External Interrupt 4 Control Register
0000H
FF92H
FEI5IC
SFR-b
C9H
Fast External Interrupt 5 Control Register
0000H
FF94H
FEI6IC
SFR-b
CAH
Fast External Interrupt 6 Control Register
0000H
FF96H
FEI7IC
SFR-b
CBH
Fast External Interrupt 7 Control Register
0000H
FF98H
RES4IC
SFR-b
CBH
reserved
0000H
FF9AH
IOMIOIC
SFR-b
CDH
IOM-2 IO Interrupt Control Register
0000H
FF9CH
URD2IC
SFR-b
CEH
UDC RX Done2 Interrupt Control Register
0000H
FF9EH
URD1IC
SFR-b
CFH
UDC RX Done1 Interrupt Control Register
0000H
FFA8H
CLISNC
SFR-b
D4H
The channel link interrupt subnode register
0000H
FFAAH
FOCON
SFR-b
D5H
Frequency Output Control Register
0000H
FFACH
TFR
SFR-b
D6H
Trap Flag Register
0000H
FFAEH
WDTCON
SFR-b
D7H
Watchdog Timer Control Register
000xH
FFB0H
S0CON
SFR-b
D8H
Serial Channel 0 Control Register
0000H
FFB2H
SSCCON
SFR-b
D9H
SSC Control Register
0000H
FFBAH
S0CLC
SFR-b
DDH
ASC Clock Control Register
0000H
FFC0H
P2
SFR-b
E0H
Port 2 Register
0000H
FFC2H
DP2
SFR-b
E1H
Port 2 Direction Control Register
0000H
FFC4H
P3
SFR-b
E2H
Port 3 Register
0000H
FFC6H
DP3
SFR-b
E3H
Port 3 Direction Control Register
0000H
Data Sheet
498
Reset
Value
2001-02-23
C165UTAH
Register Set
Physical Register
Addr
Name
Type
8-bit
Addr
Description
Reset
Value
FFC8H
P4
SFR-b
E4H
Port 4 Register (8 bits)
00H
FFCAH
DP4
SFR-b
E5H
Port 4 Direction Control Register
00H
FFCCH
P6
SFR-b
E6H
Port 6 Register (8 bits)
00H
FFCEH
DP6
SFR-b
E7H
Port 6 Direction Control Register
00H
FFD0H
P7
SFR-b
E8H
Port 7 Register (8 bits)
00H
FFD2H
DP7
SFR-b
E9H
Port 7 Direction Control Register
00H
1)
2)
The DTIDR register is a data register which is used by advanced real time operating systems to store the task
ID of the active task. It is used for hardware trigger events in the OCDS.
The EMUCON register is a reserved test register and is not to be used by other software.
23.4
Special Function Registers ordered by Name
The following table lists all SFRs which are implemented in the C165UTAH ordered by their name. Bitaddressable SFRs are marked with the letter “b” in column “Type”.
SFRs within the Extended SFR-Space (ESFRs) are marked with the letter “E” in column “Type”.
Table 104
Register
Name
Registers ordered by Name
Phys.
Addr
Type
8-bit Description
Addr
Reset
Value
ABENDIC
F18CH
ESFR-b
C6H
ASC Autobaud End Interrupt Control Register
0000H
ABS0CON
FEF8H
SFR
7CH
ASC Autobaud Control Register
0000H
ABSTAT
FEFEH
SFR
7FH
ASC Autobaud Status Register
0000H
ABSTIC
F194H
ESFR-b
CAH
ASC Autobaud Start Interrupt Control Register
0000H
ADDRSEL1
FE18H
SFR
0CH
Address Select Register 1
0000H
ADDRSEL2
FE1AH
SFR
0DH
Address Select Register 2
0000H
ADDRSEL3
FE1CH
SFR
0EH
Address Select Register 3
0000H
ADDRSEL4
FE1EH
SFR
0FH
Address Select Register 4
0000H
BUSCON0
FF0CH
SFR-b
86H
Bus Configuration Register 0
0000H
BUSCON1
FF14H
SFR-b
8AH
Bus Configuration Register 1
0000H
BUSCON2
FF16H
SFR-b
8BH
Bus Configuration Register 2
0000H
BUSCON3
FF18H
SFR-b
8CH
Bus Configuration Register 3
0000H
BUSCON4
FF1AH
SFR-b
8DH
Bus Configuration Register 4
0000H
CAPREL
FE4AH
SFR
25H
GPT1/2 Capture / Reload Register
0000H
CLISNC
FFA8H
SFR-b
D4H
The channel link interrupt subnode register
0000H
CP
FE10H
SFR
08H
CPU Context Pointer Register
FC00H
CRIC
FF6AH
SFR-b
B5H
GPT2 CAPREL Interrupt Control Register
0000H
CSP
FE08H
SFR
04H
CPU Code Segment Pointer Register (8 bits)
0000H
Data Sheet
499
2001-02-23
C165UTAH
Register Set
Register
Name
Phys.
Addr
Type
8-bit Description
Addr
Reset
Value
DP0H
F102H
ESFR-b
81H
P0H Direction Control Register
00H
DP0L
F100H
ESFR-b
80H
P0L Direction Control Register
00H
DP1H
F106H
ESFR-b
83H
P1H Direction Control Register
00H
DP1L
F104H
ESFR-b
82H
P1L Direction Control Register
00H
DP2
FFC2H
SFR-b
E1H
Port 2 Direction Control Register
0000H
DP3
FFC6H
SFR-b
E3H
Port 3 Direction Control Register
0000H
DP4
FFCAH
SFR-b
E5H
Port 4 Direction Control Register
00H
DP6
FFCEH
SFR-b
E7H
Port 6 Direction Control Register
00H
DP7
FFD2H
SFR-b
E9H
Port 7 Direction Control Register
00H
DPP0
FE00H
SFR
00H
CPU Data Page Pointer 0 Register (10 bits)
0000H
DPP1
FE02H
SFR
01H
CPU Data Page Pointer 1 Register (10 bits)
0001H
DPP2
FE04H
SFR
02H
CPU Data Page Pointer 2 Register (10 bits)
0002H
DPP3
FE06H
SFR
03H
CPU Data Page Pointer 3 Register (10 bits)
0003H
DTIDR
F0D8H
ESFR
6CH
Task ID register
0000H
EMUCON
FE0AH
SFR
05H
Emulation Control Register
xxxxH
EPECIC
F17CH
ESFR-b
BEH
EPEC Interrupt
0000H
EXICON
F1C0H
ESFR-b
E0H
External Interrupt Control Register
0000H
EXISEL
F1DAH
ESFR-b
EDH
External Interrupt Select Register
0000H
FEI0IC
FF88H
SFR-b
C4H
Fast External Interrupt 0 Control Register
0000H
FEI1IC
FF8AH
SFR-b
C5H
Fast External Interrupt 1 Control Register
0000H
FEI2IC
FF8CH
SFR-b
C6H
Fast External Interrupt 2 Control Register
0000H
FEI3IC
FF8EH
SFR-b
C7H
Fast External Interrupt 3 Control Register
0000H
FEI4IC
FF90H
SFR-b
C8H
Fast External Interrupt 4 Control Register
0000H
FEI5IC
FF92H
SFR-b
C9H
Fast External Interrupt 5 Control Register
0000H
FEI6IC
FF94H
SFR-b
CAH
Fast External Interrupt 6 Control Register
0000H
FEI7IC
FF96H
SFR-b
CBH
Fast External Interrupt 7 Control Register
0000H
FOCON
FFAAH
SFR-b
D5H
Frequency Output Control Register
0000H
GPTCLC
FE4CH
SFR
26H
GPT1/2 Clock Control Register
0000H
IDCHIP
F07CH
ESFR
3EH
Identifier
0503H
IDMANUF
F07EH
ESFR
3FH
Identifier
1824H
IDMEM
F07AH
ESFR
3DH
Identifier
0000H
IDMEM2
F076H
ESFR
3BH
Identifier
0000H
IDPROG
F078H
ESFR
3CH
Identifier
0000H
IOMC0RIC
F182H
ESFR-b
C1H
IOM-2 Channel0 RX Interrupt Control Register
0000H
IOMC0TIC
F17EH
ESFR-b
BFH
IOM-2 Channel0 TX Interrupt Control Register
0000H
Data Sheet
500
2001-02-23
C165UTAH
Register Set
Register
Name
Phys.
Addr
Type
8-bit Description
Addr
Reset
Value
IOMC1RIC
F192H
ESFR-b
C9H
IOM-2 Channel1 RX Interrupt Control Register
0000H
IOMC1TIC
F18AH
ESFR-b
C5H
IOM-2 Channel1 TX Interrupt Control Register
0000H
IOMIOIC
FF9AH
SFR-b
CDH
IOM-2 IO Interrupt Control Register
0000H
ISNC
F1DEH
ESFR-b
EFH
Interrupt Sub Node Control Register
0000H
MDC
FF0EH
SFR-b
87H
CPU Multiply Divide Control Register
0000H
MDH
FE0CH
SFR
06H
CPU Multiply Divide Register - High Word
0000H
MDL
FE0EH
SFR
07H
CPU Multiply Divide Register - Low Word
0000H
ODP0H
FE22H
SFR
11H
Port 0 Open Drain Control Register High
0000H
ODP1H
FE26H
SFR
13H
Port 1 Open Drain Control Register High
0000H
ODP1L
FE24H
SFR
12H
Port 1 Open Drain Control Register Low
0000H
ODP2
F1C2H
ESFR-b
E1H
Port 2 Open Drain Control Register
0000H
ODP3
F1C6H
ESFR-b
E3H
Port 3 Open Drain Control Register
0000H
ODP4
F1CAH
ESFR-b
E5H
Port 4 Open Drain Control Register
00H
ODP6
F1CEH
ESFR-b
E7H
Port 6 Open Drain Control Register
00H
ODP7
F1D2H
ESFR-b
E9H
Port 7 Open Drain Control Register
00H
ONES
FF1EH
SFR-b
8FH
Constant Value 1sRegister'
FFFFH
P0H
FF02H
SFR-b
81H
Port 0 High Register (Upper half)
00H
P0HPHEN
FE6AH
SFR
35H
Port 0 High Pin Hold Enable Register
0000H
P0HPUDEN
FE66H
SFR
33H
Port 0 High Pull Switch On/Off Register
xxFFH
P0HPUDSEL
FE62H
SFR
31H
Port 0 High Pull-Up/Down Select Register
xxFFH
P0L
FF00H
SFR-b
80H
Port 0 Low Register (Lower half)
00H
P0LPHEN
FE68H
SFR
34H
Port 0 Low Pin Hold Enable Register
0000H
P0LPUDEN
FE64H
SFR
32H
Port 0 Low Pull Switch On/Off Register
xxFFH
P0LPUDSEL
FE60H
SFR
30H
Port 0 Low Pull-Up/Down Select Register
xxFFH
P1H
FF06H
SFR-b
83H
Port 1 High Register (Upper half)
00H
P1HPHEN
FE76H
SFR
3BH
Port 1 High Pin Hold Enable Register
0000H
P1HPUDEN
FE72H
SFR
39H
Port 1 High Pull Switch On/Off Register
0000H
P1HPUDSEL
FE6EH
SFR
37H
Port 1 High Pull-Up/Down Select Register
0000H
P1L
FF04H
SFR-b
82H
Port 1 Low Register (Lower half)
00H
P1LPHEN
FE74H
SFR
3AH
Port 1 Low Pin Hold Enable Register
0000H
P1LPUDEN
FE70H
SFR
38H
Port 1 Low Pull Switch On/Off Register
0000H
P1LPUDSEL
FE6CH
SFR
36H
Port 1 Low Pull-Up/Down Select Register
0000H
P2
FFC0H
SFR-b
E0H
Port 2 Register
0000H
P2PHEN
FE7CH
SFR
3EH
Port 2 Pin Hold Enable Register
0000H
P2PUDEN
FE7AH
SFR
3DH
Port 2 Pull Switch On/Off Register
0000H
Data Sheet
501
2001-02-23
C165UTAH
Register Set
Register
Name
Phys.
Addr
Type
8-bit Description
Addr
Reset
Value
P2PUDSEL
FE78H
SFR
3CH
Port 2 Pull-Up/Down Select Register
0000H
P3
FFC4H
SFR-b
E2H
Port 3 Register
0000H
P3PHEN
FE82H
SFR
41H
Port 3 Pin Hold Enable Register
0000H
P3PUDEN
FE80H
SFR
40H
Port 3 Pull Switch On/Off Register
0000H
P3PUDSEL
FE7EH
SFR
3FH
Port 3 Pull-Up/Down Select Register
0000H
P4
FFC8H
SFR-b
E4H
Port 4 Register (8 bits)
00H
P4PHEN
FE88H
SFR
44H
Port 4 Pin Hold Enable Register
0000H
P4PUDEN
FE86H
SFR
43H
Port 4 Pull Switch On/Off Register
0000H
P4PUDSEL
FE84H
SFR
42H
Port 4 Pull-Up/Down Select Register
0000H
P6
FFCCH
SFR-b
E6H
Port 6 Register (8 bits)
00H
P6PHEN
FE94H
SFR
4AH
Port 6 Pin Hold Enable Register
0000H
P6PUDEN
FE92H
SFR
49H
Port 6 Pull Switch On/Off Register
0000H
P6PUDSEL
FE90H
SFR
48H
Port 6 Pull-Up/Down Select Register
0000H
P7
FFD0H
SFR-b
E8H
Port 7 Register (8 bits)
00H
P7PHEN
FE9AH
SFR
4DH
Port 7 Pin Hold Enable Register
0000H
P7PUDEN
FE98H
SFR
4CH
Port 7 Pull Switch On/Off Register
0000H
P7PUDSEL
FE96H
SFR
4BH
Port 7 Pull-Up/Down Select Register
0000H
PECC0
FEC0H
SFR
60H
PEC Channel 0 Control Register
0000H
PECC1
FEC2H
SFR
61H
PEC Channel 1 Control Register
0000H
PECC2
FEC4H
SFR
62H
PEC Channel 2 Control Register
0000H
PECC3
FEC6H
SFR
63H
PEC Channel 3 Control Register
0000H
PECC4
FEC8H
SFR
64H
PEC Channel 4 Control Register
0000H
PECC5
FECAH
SFR
65H
PEC Channel 5 Control Register
0000H
PECC6
FECCH
SFR
66H
PEC Channel 6 Control Register
0000H
PECC7
FECEH
SFR
67H
PEC Channel 7 Control Register
0000H
PECCLIC
F180H
ESFR-b
C0H
PEC Channel Link Interrupt Control Register
0000H
PECSN0
FED0H
SFR
68H
PEC Segment No Register
PECSN1
FED2H
SFR
69H
PEC Segment No Register
PECSN2
FED4H
SFR
6AH
PEC Segment No Register
PECSN3
FED6H
SFR
6BH
PEC Segment No Register
PECSN4
FED8H
SFR
6CH
PEC Segment No Register
PECSN5
FEDAH
SFR
6DH
PEC Segment No Register
PECSN6
FEDCH
SFR
6EH
PEC Segment No Register
PECSN7
FEDEH
SFR
6FH
PEC Segment No Register
PECXC0
FEF0H
SFR
78H
PEC Channel 0 Extended Control Register
Data Sheet
502
2001-02-23
C165UTAH
Register Set
Register
Name
Phys.
Addr
Type
8-bit Description
Addr
Reset
Value
PECXC2
FEF2H
SFR
79H
PEC Channel 2 Extended Control Register
PSW
FF10H
SFR-b
88H
CPU Program Status Word
0000H
R0
SFR-b
F0H
General Purpose Register 0
UUUUH
R0
ESFR-b
F0H
General Purpose Register 0
UUUUH
R1
SFR-b
F1H
General Purpose Register 1
UUUUH
R1
ESFR-b
F1H
General Purpose Register 1
UUUUH
R10
ESFR-b
FAH
General Purpose Register 10
UUUUH
R10
SFR-b
FAH
General Purpose Register 10
UUUUH
R11
SFR-b
FBH
General Purpose Register 11
UUUUH
R11
ESFR-b
FBH
General Purpose Register 11
UUUUH
R12
SFR-b
FCH
General Purpose Register 12
UUUUH
R12
ESFR-b
FCH
General Purpose Register 12
UUUUH
R13
SFR-b
FDH
General Purpose Register 13
UUUUH
R13
ESFR-b
FDH
General Purpose Register 13
UUUUH
R14
SFR-b
FEH
General Purpose Register 14
UUUUH
R14
ESFR-b
FEH
General Purpose Register 14
UUUUH
R15
ESFR-b
FFH
General Purpose Register 15
UUUUH
R15
SFR-b
FFH
General Purpose Register 15
UUUUH
R2
SFR-b
F2H
General Purpose Register 2
UUUUH
R2
ESFR-b
F2H
General Purpose Register 2
UUUUH
R3
SFR-b
F3H
General Purpose Register 3
UUUUH
R3
ESFR-b
F3H
General Purpose Register 3
UUUUH
R4
ESFR-b
F4H
General Purpose Register 4
UUUUH
R4
SFR-b
F4H
General Purpose Register 4
UUUUH
R5
ESFR-b
F5H
General Purpose Register 5
UUUUH
R5
SFR-b
F5H
General Purpose Register 5
UUUUH
R6
ESFR-b
F6H
General Purpose Register 6
UUUUH
R6
SFR-b
F6H
General Purpose Register 6
UUUUH
R7
ESFR-b
F7H
General Purpose Register 7
UUUUH
R7
SFR-b
F7H
General Purpose Register 7
UUUUH
R8
SFR-b
F8H
General Purpose Register 8
UUUUH
R8
ESFR-b
F8H
General Purpose Register 8
UUUUH
R9
SFR-b
F9H
General Purpose Register 9
UUUUH
R9
ESFR-b
F9H
General Purpose Register 9
UUUUH
SFR-b
CBH
reserved
0000H
503
2001-02-23
RES4IC
Data Sheet
FF98H
C165UTAH
Register Set
Register
Name
Phys.
Addr
Type
8-bit Description
Addr
Reset
Value
RES6IC
F19AH
ESFR-b
CDH
reserved
0000H
reserved
F1D6H
ESFR-b
EBH
reserved - do not use
0000H
reserved
F1D8H
ESFR-b
ECH
reserved - do not use
0000H
RP0H
F108H
ESFR-b
84H
System Startup Configuration Register (RO)
xxH
RTC_INTIC
F184H
ESFR-b
C2H
RTC_INT Sub Node Interrupt Register
0000H
RTCCLC
F0C8H
ESFR
64H
RTC Clock Control Register
0000H
RTCCON
F1CCH
ESFR-b
E6H
RTC Control Register
00H
RTCH
F0D6H
ESFR
6BH
RTC Timer Register High
nH
RTCISNC
F1C8H
ESFR-b
E4H
RTC Interrupt Sub Node Control Register
0000H
RTCL
F0D4H
ESFR
6AH
RTC Timer Register Low
nH
RTCRELH
F0CEH
ESFR
67H
RTC Timer Reload Register High
0000H
RTCRELL
F0CCH
ESFR
66H
RTC Timer Reload Register Low
0000H
S0BG
FEB4H
SFR
5AH
Serial Channel 0 Baud Rate Generator Reload
Register
0000H
S0CLC
FFBAH
SFR-b
DDH
ASC Clock Control Register
0000H
S0CON
FFB0H
SFR-b
D8H
Serial Channel 0 Control Register
0000H
S0EIC
FF70H
SFR-b
B8H
Serial Channel 0 Error Interrupt Control Register 0000H
S0FDV
FEB6H
SFR
5BH
ASC Fractional Divide Register
0000H
S0PMW
FEAAH
SFR
55H
ASC IrDA PMW Control Register
0000H
S0RBUF
FEB2H
SFR
59H
Serial Channel 0 Receive Buffer Register (RO)
xxxxH
S0RIC
FF6EH
SFR-b
B7H
Serial Channel 0 Receive Interrupt Control
Register
0000H
S0TBIC
F19CH
ESFR-b
CEH
Serial Channel 0 Transmit Buffer IC Register
0000H
S0TBUF
FEB0H
SFR
58H
Serial Channel 0 Transmit Buffer Register (WO)
0000H
S0TIC
FF6CH
SFR-b
B6H
Serial Channel 0 Transmit Interrupt Control
Register
0000H
SCUSLC
F0C0H
ESFR
60H
Security Level Control Register
SCUSLS
F0C2H
ESFR
61H
Security Level Status Register
SP
FE12H
SFR
09H
CPU System Stack Pointer Register
FC00H
SSCBR
F0B4H
ESFR
5AH
SSC Baudrate Register
0000H
SSCCLC
F0B6H
ESFR
5BH
SSC Clock Control Register
0000H
SSCCON
FFB2H
SFR-b
D9H
SSC Control Register
0000H
SSCEIC
FF76H
SFR-b
BBH
SSC Error Interrupt Control Register
0000H
SSCRB
F0B2H
ESFR
59H
SSC Receive Buffer (RO)
xxxxH
SSCRIC
FF74H
SFR-b
BAH
SSC Receive Interrupt Control Register
0000H
SSCTB
F0B0H
ESFR
58H
SSC Transmit Buffer (WO)
0000H
Data Sheet
504
2001-02-23
C165UTAH
Register Set
Register
Name
Phys.
Addr
Type
8-bit Description
Addr
Reset
Value
SSCTIC
FF72H
SFR-b
B9H
SSC Transmit Interrupt Control Register
0000H
STKOV
FE14H
SFR
0AH
CPU Stack Overflow Pointer Register
FA00H
STKUN
FE16H
SFR
0BH
CPU Stack Underflow Pointer Register
FC00H
SYSCON
FF12H
SFR-b
89H
CPU System Configuration Register
0xx0H
SYSCON1
F1DCH
ESFR-b
EEH
System Configuration Register 1/Sleep Mode
0000H
SYSCON2
F1D0H
ESFR-b
E8H
System Configuration Register 2/Clock Control
0000H
SYSCON3
F1D4H
ESFR-b
EAH
System Configuration Register 3/Periph.
Managem.
0000H
T14
F0D2H
ESFR
69H
Timer 14 Register
nH
T14REL
F0D0H
ESFR
68H
Timer 14 Reload Register
nH
T2
FE40H
SFR
20H
GPT1 Timer 2 Register
0000H
T2CON
FF40H
SFR-b
A0H
GPT1 Timer 2 Control Register
0000H
T2IC
FF60H
SFR-b
B0H
GPT1 Timer 2 Interrupt Control Register
0000H
T3
FE42H
SFR
21H
GPT1 Timer 3 Register
0000H
T3CON
FF42H
SFR-b
A1H
GPT1 Timer 3 Control Register
0000H
T3IC
FF62H
SFR-b
B1H
GPT1 Timer 3 Interrupt Control Register
0000H
T4
FE44H
SFR
22H
GPT1 Timer 4 Register
0000H
T4CON
FF44H
SFR-b
A2H
GPT1 Timer 4 Control Register
0000H
T4IC
FF64H
SFR-b
B2H
GPT1 Timer 4 Interrupt Control Register
0000H
T5
FE46H
SFR
23H
GPT2 Timer 5 Register
0000H
T5CON
FF46H
SFR-b
A3H
GPT2 Timer 5 Control Register
0000H
T5IC
FF66H
SFR-b
B3H
GPT2 Timer 5 Interrupt Control Register
0000H
T6
FE48H
SFR
24H
GPT2 Timer 6 Register
0000H
T6CON
FF48H
SFR-b
A4H
GPT2 Timer 6 Control Register
0000H
T6IC
FF68H
SFR-b
B4H
GPT2 Timer 6 Interrupt Control Register
0000H
TFR
FFACH
SFR-b
D6H
Trap Flag Register
0000H
UCFGVIC
F16EH
ESFR-b
B7H
UDC Config Val Interrupt Control Register
0000H
ULCDIC
F176H
ESFR-b
BBH
UDC Load Config Done Interrupt Control
Register
0000H
URD0IC
F17AH
ESFR-b
BDH
UDC RX Done0 Interrupt Control Register
0000H
URD1IC
FF9EH
SFR-b
CFH
UDC RX Done1 Interrupt Control Register
0000H
URD2IC
FF9CH
SFR-b
CEH
UDC RX Done2 Interrupt Control Register
0000H
URD3IC
FF78H
SFR-b
BCH
UDC RX Done3 Interrupt Control Register
0000H
URD4IC
FF7AH
SFR-b
BDH
UDC RX Done4 Interrupt Control Register
0000H
URD5IC
FF7CH
SFR-b
BEH
UDC RX Done5 Interrupt Control Register
0000H
URD6IC
FF7EH
SFR-b
BFH
UDC RX Done6 Interrupt Control Register
0000H
Data Sheet
505
2001-02-23
C165UTAH
Register Set
Register
Name
Phys.
Addr
Type
8-bit Description
Addr
Reset
Value
URD7IC
FF80H
SFR-b
C0H
UDC RX Done7 Interrupt Control Register
0000H
URXRIC
F16AH
ESFR-b
B5H
UDC RXRR Interrupt Control Register
0000H
USETIC
F178H
ESFR-b
BCH
UDC SETUP Interrupt Control Register
0000H
USOFIC
F170H
ESFR-b
B8H
UDC Start of Frame Interrupt Control Register
0000H
USSIC
F174H
ESFR-b
BAH
UDC Suspend Interrupt Control Register
0000H
USSOIC
F172H
ESFR-b
B9H
UDC Suspend off Interrupt Control Register
0000H
UTD0IC
FF82H
SFR-b
C1H
UDC TX Done0 Interrupt Control Register
0000H
UTD1IC
FF84H
SFR-b
C2H
UDC TX Done1 Interrupt Control Register
0000H
UTD2IC
FF86H
SFR-b
C3H
UDC TX Done2 Interrupt Control Register
0000H
UTD3IC
F160H
ESFR-b
B0H
UDC TX Done3 Interrupt Control Register
0000H
UTD4IC
F162H
ESFR-b
B1H
UDC TX Done4 Interrupt Control Register
0000H
UTD5IC
F164H
ESFR-b
B2H
UDC TX Done5 Interrupt Control Register
0000H
UTD6IC
F166H
ESFR-b
B3H
UDC TX Done6 Interrupt Control Register
0000H
UTD7IC
F168H
ESFR-b
B4H
UDC TX Done7 Interrupt Control Register
0000H
UTXRIC
F16CH
ESFR-b
B6H
UDC TXWR Interrupt Control Register
0000H
WDT
FEAEH
SFR
57H
Watchdog Timer Register (RO)
0000H
WDTCON
FFAEH
SFR-b
D7H
Watchdog Timer Control Register
000xH
XADRS1
F014H
ESFR
0AH
XBUS Address Select Register 1
XADRS2
F016H
ESFR
0BH
XBUS Address Select Register 2
XADRS3
F018H
ESFR
0CH
XBUS Address Select Register 3
XADRS4
F01AH
ESFR
0DH
XBUS Address Select Register 4
XADRS5
F01CH
ESFR
0EH
XBUS Address Select Register 5
XADRS6
F01EH
ESFR
0FH
XBUS Address Select Register 6
XBCON1
F114H
ESFR-b
8AH
XBUS Control register 1: IOM-2 module
0000H
XBCON2
F116H
ESFR-b
8BH
XBUS Control register 2: USB module
0000H
XBCON3
F118H
ESFR-b
8CH
XBUS Control register 3: EPEC module
0000H
XBCON4
F11AH
ESFR-b
8DH
XBUS Control register 4: reserved
0000H
XBCON5
F11CH
ESFR-b
8EH
XBUS Control register 5: reserved
0000H
XBCON6
F11EH
ESFR-b
8FH
XBUS Control register 6: reserved
XP0IC
F186H
ESFR-b
C3H
X-Bus Peripheral 0 UDC TXWR Interrupt Control
Register
0000H
XP1IC
F18EH
ESFR-b
C7H
X-Bus Peripheral 1 EPEC Interrupt Control
Register
0000H
Data Sheet
506
2001-02-23
C165UTAH
Register Set
Register
Name
Phys.
Addr
Type
8-bit Description
Addr
Reset
Value
XP2IC
F196H
ESFR-b
CBH
X-Bus Peripheral 2 IOM-2 IO Interrupt Control
Register
0000H
XP3IC
F19EH
ESFR-b
CFH
X-Bus Peripheral 3 PLL/RTC Interrupt Control
Register
0000H
XPERCON
F024H
ESFR
12H
XBUS Peripheral Control Register
0401H
ZEROS
FF1CH
SFR-b
8EH
Constant Value 0sRegister'
0000H
23.5
Special Notes
PEC Pointer Registers
The source and destination pointers for the peripheral event controller are mapped to a
special area within the internal RAM. Pointers that are not occupied by the PEC may
therefore be used like normal RAM. During Power Down mode or any warm reset the
PEC pointers are preserved.
The PEC and its registers are described in chapter “Interrupt and Trap Functions”.
GPR Access in the ESFR Area
The locations 00’F000H...00’F01EH within the ESFR area are reserved and allow to
access the current register bank via short register addressing modes. The GPRs are
mirrored to the ESFR area which allows access to the current register bank even after
switching register spaces (see example below).
MOV
EXTR
MOV
R5, DP3
#1
R5, ODP3
;GPR access via SFR area
;GPR access via ESFR area
Writing Bytes to SFRs
All special function registers may be accessed wordwise or bytewise (some of them even
bitwise). Reading bytes from word SFRs is a non-critical operation. However, when
writing bytes to word SFRs the complementary byte of the respective SFR is cleared with
the write operation.
Data Sheet
507
2001-02-23
C165UTAH
Instruction Set Summary
24
Instruction Set Summary
This chapter briefly summarizes the C165UTAH's instructions ordered by instruction
classes. This provides a basic understanding of the C165UTAH’s instruction set, the
power and versatility of the instructions and their general usage.
A detailed description of each single instruction, including its operand data type,
condition flag settings, addressing modes, length (number of bytes) and object code
format is provided in the “Instruction Set Manual” for the C16x Family. This manual
also provides tables ordering the instructions according to various criteria, to allow quick
references.
Summary of Instruction Classes
Grouping the various instruction into classes aids in identifying similar instructions (eg.
SHR, ROR) and variations of certain instructions (eg. ADD, ADDB). This provides an
easy access to the possibilities and the power of the instructions of the C165UTAH.
Note: The used mnemonics refer to the detailled description.
Arithmetic Instructions
•
•
•
•
•
•
•
•
•
Addition of two words or bytes:
Addition with Carry of two words or bytes:
Subtraction of two words or bytes:
Subtraction with Carry of two words or bytes:
16*16 bit signed or unsigned multiplication:
16/16 bit signed or unsigned division:
32/16 bit signed or unsigned division:
1's complement of a word or byte:
2's complement (negation) of a word or byte:
ADD
ADDC
SUB
SUBC
MUL
DIV
DIVL
CPL
NEG
ADDB
ADDCB
SUBB
SUBCB
MULU
DIVU
DIVLU
CPLB
NEGB
AND
OR
XOR
ANDB
ORB
XORB
CMP
CMPB
CMPI1
CMPI2
CMPD1
CMPD2
Logical Instructions
• Bitwise ANDing of two words or bytes:
• Bitwise ORing of two words or bytes:
• Bitwise XORing of two words or bytes:
Compare and Loop Control Instructions
• Comparison of two words or bytes:
• Comparison of two words with post-increment
by either 1 or 2:
• Comparison of two words with post-decrement
by either 1 or 2:
Data Sheet
508
2001-02-23
C165UTAH
Instruction Set Summary
Boolean Bit Manipulation Instructions
• Manipulation of a maskable bit field
in either the high or the low byte of a word:
• Setting a single bit (to ‘1’):
• Clearing a single bit (to ‘0’):
• Movement of a single bit:
• Movement of a negated bit:
• ANDing of two bits:
• ORing of two bits:
• XORing of two bits:
• Comparison of two bits:
BFLDH
BSET
BCLR
BMOV
BMOVN
BAND
BOR
BXOR
BCMP
BFLDL
Shift and Rotate Instructions
•
•
•
•
•
Shifting right of a word:
Shifting left of a word:
Rotating right of a word:
Rotating left of a word:
Arithmetic shifting right of a word (sign bit shifting):
SHR
SHL
ROR
ROL
ASHR
Prioritize Instruction
• Determination of the number of shift cycles required
to normalize a word operand (floating point support):
PRIOR
Data Movement Instructions
• Standard data movement of a word or byte:
• Data movement of a byte to a word location
with either sign or zero byte extension:
MOV
MOVB
MOVBS
MOVBZ
Note: The data movement instructions can be used with a big number of different
addressing modes including indirect addressing and automatic pointer in-/
decrementing.
System Stack Instructions
• Pushing of a word onto the system stack:
• Popping of a word from the system stack:
• Saving of a word on the system stack,
and then updating the old word with a new value
(provided for register bank switching):
Data Sheet
509
PUSH
POP
SCXT
2001-02-23
C165UTAH
Instruction Set Summary
Jump Instructions
• Conditional jumping to an either absolutely,
indirectly, or relatively addressed target instruction
within the current code segment:
• Unconditional jumping to an absolutely addressed
target instruction within any code segment:
• Conditional jumping to a relatively addressed
target instruction within the current code segment
depending on the state of a selectable bit:
• Conditional jumping to a relatively addressed
target instruction within the current code segment
depending on the state of a selectable bit
with a post-inversion of the tested bit
in case of jump taken (semaphore support):
JMPA
JMPI
JMPR
JMPS
JB
JNB
JBC
JNBS
CALLA
CALLI
Call Instructions
• Conditional calling of an either absolutely
or indirectly addressed subroutine within
the current code segment:
• Unconditional calling of a relatively addressed
subroutine within the current code segment:
• Unconditional calling of an absolutely addressed
subroutine within any code segment:
• Unconditional calling of an absolutely addressed
subroutine within the current code segment plus
an additional pushing of a selectable register onto
the system stack:
• Unconditional branching to the interrupt or
trap vector jump table in code segment 0:
CALLR
CALLS
PCALL
TRAP
Return Instructions
• Returning from a subroutine
within the current code segment:
• Returning from a subroutine
within any code segment:
• Returning from a subroutine within the current
code segment plus an additional popping of a
selectable register from the system stack:
• Returning from an interrupt service routine:
Data Sheet
510
RET
RETS
RETP
RETI
2001-02-23
C165UTAH
Instruction Set Summary
System Control Instructions
•
•
•
•
•
•
Resetting the C165UTAH via software:
Entering the Idle mode:
Entering the Power Down mode:
Servicing the Watchdog Timer:
Disabling the Watchdog Timer:
Signifying the end of the initialization routine
(pulls pin RSTOUT high, and disables the effect of
any later execution of a DISWDT instruction):
SRST
IDLE
PWRDN
SRVWDT
DISWDT
EINIT
Miscellaneous
• Null operation which requires 2 bytes of
storage and the minimum time for execution:
• Definition of an unseparable instruction sequence:
• Switch ‘reg’, ‘bitoff’ and ‘bitaddr’ addressing modes
to the Extended SFR space:
• Override the DPP addressing scheme
using a specific data page instead of the DPPs,
and optionally switch to ESFR space:
• Override the DPP addressing scheme
using a specific segment instead of the DPPs,
and optionally switch to ESFR space:
NOP
ATOMIC
EXTR
EXTP
EXTPR
EXTS
EXTSR
Note: The ATOMIC and EXT* instructions provide support for uninterruptable code
sequences eg. for semaphore operations. They also support data addressing
beyond the limits of the current DPPs (except ATOMIC), which is advantageous
for bigger memory models in high level languages. Refer to chapter “System
Programming” for examples.
Protected Instructions
Some instructions of the C165UTAH which are critical for the functionality of the
controller are implemented as so-called Protected Instructions. These protected
instructions use the maximum instruction format of 32 bits for decoding, while the regular
instructions only use a part of it (eg. the lower 8 bits) with the other bits providing
additional information like involved registers. Decoding all 32 bits of a protected
doubleword instruction increases the security in cases of data distortion during
instruction fetching. Critical operations like a software reset are therefore only executed
if the complete instruction is decoded without an error. This enhances the safety and
reliability of a microcontroller system.
Data Sheet
511
2001-02-23
C165UTAH
AC/DC Characteristics
25
AC/DC Characteristics
25.1
Absolute Maximum Ratings
• Storage temperature (TST) -65 to +150 °C
• Voltage on VDD pins with respect to ground (VSS) -0.5 to + 4.0 V
• Voltage on any pin with respect to ground (VSS) (except VDD, VSS, XTAL and USB pins)
-0.5 to 5.5 V
• Absolute maximum total I/O current 250 mA
• Absolute maximum current on any pin, sink or source 10 mA
Stresses beyond those listed above may cause permanent damage to the device. This
is a stress rating only, and functional operation of the C165UTAH is not implied at these
or any other conditions above those indicated in the operational sections of this
specification. Exposure to absolute maximum rating conditions for extended periods
may affect device reliability.
25.2
Recommended Operating Conditions
The following conditions are to be met for correct operation of the device.
• Ambient temperature under bias (TA): -40 to +85 °C
• Load capacitance (CL):
工商网监
湘ICP备2023018690号
