Mercury
Programmable Logic
Device Family
®
January 2003, ver. 2.2
Features…
Data Sheet
■
■
Table 1. Mercury Device Features
Feature
Typical gates
HSDI channels
LEs
ESBs (1)
Maximum RAM bits
Maximum user I/O pins
EP1M120
EP1M350
120,000
350,000
8
18
4,800
14,400
12
28
49,152
114,688
303
486
Note to Table 1:
(1)
Altera Corporation
DS-MERCURY-2.2
Each ESB can be used for two dual- or single-port RAM blocks.
1
13
Development
Tools
High-performance programmable logic device (PLD) family (see
Table 1)
–
Integrated high-speed transceivers with support for clock data
recovery (CDR) at up to 1.25 gigabits per second (Gbps)
–
Look-up table (LUT)-based architecture optimized for high
speed
–
Advanced interconnect structure for fast routing of critical paths
–
Enhanced I/O structure for versatile standards and interface
support
–
Up to 14,400 logic elements (LEs)
System-level features
–
Up to four general-purpose phase-locked loops (PLLs) with
programmable multiplication and delay shifting
–
Up to 12 PLL output ports
–
Dedicated multiplier circuitry for high-speed implementation of
signed or unsigned multiplication up to 16 × 16
–
Embedded system blocks (ESBs) used to implement memory
functions including quad-port RAM, true dual-port RAM, firstin first-out (FIFO) buffers, and content-addressable memory
(CAM)
–
Each ESB contains 4,096 bits and can be split and used as two
2,048-bit unidirectional dual-port RAM blocks
Mercury Programmable Logic Device Family Data Sheet
...and More
Features
■
■
■
2
Advanced high-speed I/O features
–
Robust I/O standard support, including LVTTL, PCI up to
66 MHz, 3.3-V AGP in 1× and 2× modes, 3.3-V SSTL-3 and 2.5-V
SSTL-2, GTL+, HSTL, CTT, LVDS, LVPECL, and 3.3-V PCML
–
High-speed differential interface (HSDI) with dedicated
circuitry for CDR at up to 1.25 Gbps for LVDS, LVPECL, and
3.3-V PCML
–
Support for source-synchronous True-LVDSTM circuitry up to
840 megabits per second (Mbps) for LVDS, LVPECL, and 3.3-V
PCML
–
Up to 18 input and 18 output dedicated differential channels of
high-speed LVDS, LVPECL, or 3.3-V PCML
–
Built-in 100-Ω termination resistor on HSDI data and clock
differential pairs
–
Flexible-LVDSTM circuitry provides 624-Mbps support on up to
100 channels with the EP1M350 device
–
Versatile three-register I/O element (IOE) supporting double
data rate I/O (DDRIO), double data-rate (DDR) SDRAM, zero
bus turnaround (ZBT) SRAM, and quad data rate (QDR) SRAM
Designed for low-power operation
–
1.8-V internal supply voltage (VCCINT)
–
MultiVoltTM I/O interface voltage levels (VCCIO) compatible
with 1.5-V, 1.8-V, 2.5-V, and 3.3-V devices
–
5.0-V tolerant with external resistor
Advanced interconnect structure
–
Multi-level FastTrack® Interconnect structure providing fast,
predictable interconnect delays
–
Optimized high-speed Priority FastTrack Interconnect for
routing critical paths in a design
–
Dedicated carry chain that implements arithmetic functions such
as fast adders, counters, and comparators (automatically used by
software tools and megafunctions)
–
FastLUTTM connection allowing high speed direct connection
between LEs in the same logic array block (LAB)
–
Leap lines allowing a single LAB to directly drive LEs in adjacent
rows
–
The RapidLAB interconnect providing a high-speed connection
to a 10-LAB-wide region
–
Dedicated clock and control signal resources, including four
dedicated clocks, six dedicated fast global signals, and additional
row-global signals
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Tables 2 and 3 show the MercuryTM FineLine BGATM device package sizes,
options, and I/O pin counts.
Table 2. Mercury Package Sizes
Feature
484-Pin
FineLine BGA
780-Pin
FineLine BGA
Pitch (mm)
1.00
1.00
Area (mm2)
529
841
23 × 23
29 × 29
Length × width (mm × mm)
Table 3. Mercury Package Options & I/O Count
Device
484-Pin
FineLine BGA
EP1M120
303
EP1M350
486
13
Mercury devices integrate high-speed differential transceivers and
support for CDR with a speed-optimized PLD architecture. These
transceivers are implemented through the dedicated serializer,
deserializer, and clock recovery circuitry in the HSDI and incorporate
support for the LVDS, LVPECL, and 3.3-V PCML I/O standards. This
circuitry, together with enhanced I/O elements (IOEs) and support for
numerous I/O standards, allows Mercury devices to meet high-speed
interface requirements.
Development
Tools
General
Description
780-Pin
FineLine BGA
Mercury devices are the first PLDs optimized for core performance. These
LUT-based, enhanced memory devices use a network of fast routing
resources to achieve optimal performance. These resources are ideal for
data-path, register-intensive, mathematical, digital signal processing
(DSP), or communications designs.
Altera Corporation
3
Mercury Programmable Logic Device Family Data Sheet
Mercury devices include other features for performance such as quadport RAM, CAM, general purpose PLLs, and dedicated circuitry for
implementing multiplier circuits. Table 4 shows Mercury performance.
Table 4. Mercury Performance
Application
Resources Used
LEs
Performance
ESBs
-5 Speed
Grade
-6 Speed
Grade
-7 Speed
Grade
Units
16-bit loadable counter (1)
16
0
400
400
400
MHz
32-bit loadable counter (1)
32
0
400
400
400
MHz
32-bit accumulator (1)
32
0
400
400
400
MHz
32-to-1 multiplexer
27
0
1.864
2.466
2.723
ns
32 × 64 asynchronous FIFO
103
2
290
258
242
MHz
8-bit, 37-tap FIR filter
251
1
290
240
205
MSPS
Note to Table 4:
(1)
The clock tree supports up to 400 MHz. Although the registered performance for these designs exceed 400 MHz,
they are limited by the clock tree limit.
Configuration
The logic, circuitry, and interconnects in the Mercury architecture are
configured with CMOS SRAM elements. Mercury devices are
reconfigurable and are 100% tested prior to shipment. As a result, test
vectors do not have to be generated for fault coverage purposes. Instead,
the designer can focus on simulation and design verification. In addition,
the designer does not need to manage inventories of different ASIC
designs; Mercury devices can be configured on the board for the specific
functionality required.
Mercury devices are configured at system power-up with data stored in
an Altera® serial configuration device or provided by a system controller.
Altera offers in-system programmability (ISP)-capable configuration
devices, which configure Mercury devices via a serial data stream.
Mercury devices can be configured in under 70 ms. Moreover, Mercury
devices contain an optimized interface that permits microprocessors to
configure Mercury devices serially or in parallel, synchronously or
asynchronously. This interface also enables microprocessors to treat
Mercury devices as memory and to configure the device by writing to a
virtual memory location, simplifying reconfiguration.
4
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
After a Mercury device has been configured, it can be reconfigured
in-circuit by resetting the device and loading new data. Real-time changes
can be made during system operation, enabling innovative reconfigurable
computing applications.
Software
Mercury devices are supported by the Altera QuartusTM II development
system, a single, integrated package that offers HDL and schematic design
entry, compilation and logic synthesis, full simulation and worst-case
timing analysis, SignalTapTM logic analysis, and device configuration. The
Quartus II software also ships with Altera-specific HDL synthesis tools
from Exemplar Logic and Synopsys, and Altera-specific Register Transfer
Level (RTL) and timing simulation tools from Model Technology. The
Quartus II software supports PCs running Windows 98, Windows NT 4.0,
and Windows 2000; UNIX workstations running Solaris 2.6, 7, or 8, or
HP-UX 10.2 or 11.0; and PCs running Red Hat Linux 7.1.
For more information on the Quartus II development system, see the
Quartus II Programmable Logic Development System & Software Data Sheet.
Functional
Description
Altera Corporation
The Mercury architecture contains a row-based logic array to implement
general logic and a row-based embedded system array to implement
memory and specialized logic functions. Signal interconnections within
Mercury devices are provided by a series of row and column
interconnects with varying lengths and speeds. The priority FastTrack
Interconnect structure is faster than other interconnects; the Quartus II
Compiler places design-critical paths on these faster lines to improve
design performance.
5
13
Development
Tools
The Quartus II software provides NativeLinkTM interfaces to other
industry-standard PC- and UNIX-workstation-based EDA tools. For
example, designers can invoke the Quartus II software from within the
Mentor Graphics LeonardoSpectrum software, Synplicity’s Synplify
software, and the Synopsys FPGA Express software. The Quartus II
software also contains built-in optimized synthesis libraries; synthesis
tools can use these libraries to optimize designs for Mercury devices. For
example, the Synopsys Design Compiler library, supplied with the
Quartus II development system, includes DesignWare functions
optimized for the Mercury architecture.
Mercury Programmable Logic Device Family Data Sheet
Mercury device I/O pins are evenly distributed across the entire device
area; other Altera device families have I/O pins placed on the device
periphery. Mercury device I/O pin placement allows for higher I/O count
at a given die size; pad size is no longer a limiting issue. Each I/O pin is
fed by an IOE. IOEs are grouped in IOE row bands from the top to the
bottom of the device. IOE row bands are separated by several LAB rows.
LABs from the associated LAB row closest to the I/O row band drive IOEs
through the local interconnect. This feature allows fast clock-to-output
times when a pin is driven by any of the 10 LEs in the adjacent associated
LAB. Each IOE contains a bidirectional buffer along with an input register,
output register, output enable (OE) register, and input latch for DDR.
When used with a global clock, these dedicated registers provide
exceptional bidirectional I/O performance.
IOEs provide a variety of features, such as 3.3-V, 64-bit, 66-MHz PCI
compliance; 3.3-V, 64-bit, 133-MHz PCI-X compliance; Joint Test Action
Group (JTAG) boundary-scan test (BST) support; output drive strength
control; slew-rate control; tri-state buffers; bus-hold circuitry;
programmable pull-up resisters; programmable input and output delays;
and open-drain outputs. Mercury devices offer enhanced I/O support,
including support for 1.8-V I/O, 2.5-V I/O, LVCMOS, LVTTL, HSTL,
LVPECL, 3.3-V PCML, 3.3-V PCI, PCI-X, LVDS, GTL+, SSTL-2, SSTL-3,
CTT, and 3.3-V AGP I/O standards. CDR (up to 1.25 Gbps) and sourcesynchronous (up to 840 Mbps) transfers are supported with HSDI
circuitry for LVDS, LVPECL, and 3.3-V PCML I/O standards.
The ESB can implement a variety of memory functions, including CAM,
quad-port RAM, true dual-port RAM, dual- and single-port RAM, ROM,
and FIFO functions. ESBs are grouped into two rows: one at the top and
one at the bottom of the device. Embedding the memory directly into the
die improves performance and reduces die area compared to distributedRAM implementations. Moreover, the abundance of cascadable ESBs, in
conjunction with the ability for one ESB to implement two separate
memory blocks, ensures that the Mercury device can implement multiple
wide memory blocks for high-density designs. The ESB’s high speed
ensures the implemention of small memory blocks without any speed
penalty. The abundance of ESBs ensures that designers can create as many
different-sized memory blocks as the system requires. Figure 1 shows an
overview of the Mercury device.
6
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Figure 1. Mercury Architecture Block Diagram
ESB
Note (1)
ESB
Local Interconnect:
Connects LEs within
the Same or Adjacent
LABs
ESB
ESB
ESB
ESB
I/O Band with HSDI
Associated LAB Row
Buried LAB Row
Row and Priority Row
Interconnect: Connects
LABs within a Row
Buried LAB Row
I/O Band
Associated LAB Row
Buried LAB Row
Column and Priority
Column Interconnect:
Connects LABs within
Different Rows (Top
to Bottom)
Buried LAB Row
I/O Band
Associated LAB Row
Buried LAB Row
Leap Lines: Connects
Adjacent LABs in
Same Column
Buried LAB Row
I/O Band
Associated LAB Row
Buried LAB Row
13
Associated LAB Row
I/O Band
ESB
ESB
ESB
ESB
ESB
ESB
Note to Figure 1:
(1)
Figure 1 shows an EP1M120 device. Mercury devices have a varying number of rows, columns, and ESBs, as shown
in Table 5.
Table 5 lists the resources available in Mercury devices.
Table 5. Mercury Device Resources
Device
LAB Rows
LAB Columns
I/O Row Bands
ESBs
EP1M120
12
40
5
12
EP1M350
18
80
4
28
Altera Corporation
7
Development
Tools
RapidLAB Interconnect:
Connects Any 10
Consecutive LABs
within a Row from
a Central LAB
Mercury Programmable Logic Device Family Data Sheet
Mercury devices provide four dedicated clock input pins and six
dedicated fast I/O pins that globally drive register control inputs,
including clocks. These signals ensure efficient distribution of high-speed,
low-skew control signals. The control signals use dedicated routing
channels to provide short delays and low skew. The dedicated fast signals
can also be driven by internal logic, providing an ideal solution for a clock
divider or internally generated asynchronous control signal with high
fan-out. The dedicated clock and fast I/O pins on Mercury devices can
also feed logic. Dedicated clocks can also be used with the Mercury
general purpose PLLs for clock management.
Each I/O row band also provides two additional I/O pins that can drive
two row-global signals. Row-global signals can drive register control
inputs for the LAB row associated with that particular I/O row band.
High-Speed
Differential
Interface
The top I/O or HSDI band in Mercury devices contains dedicated
circuitry for supporting differential standards at speeds up to 1.25 Gbps.
Mercury devices have dedicated differential buffers and circuitry to
support LVDS, LVPECL, and 3.3-V PCML I/O standards. Two dedicated
high-speed PLLs (separate from the general purpose PLLs) multiply
reference clocks and drive high-speed differential serializer/deserializer
channels. In addition, clock recovery units (CRUs) at each receiver
channel enable CDR. EP1M120 devices support eight input channels,
eight output channels, and two dedicated clock inputs for feeding the
receiver and/or transmitter PLLs. EP1M350 devices support 18 input
channels, 18 output channels, and two dedicated clock inputs.
Mercury devices have optional built-in 100-Ω termination resistors on
HSDI differential receiver data pins and the HSDI_CLK1 and HSDI_CLK2
pins.
Designers can use the HSDI circuitry for the following applications:
■
■
■
■
■
■
Gigabit Ethernet backplanes
ATM, SONET
RapidIO
POS-PHY Level 4
Fibre Channel
SDTV
The HSDI band supports one of two possible modes:
■
■
8
Source-synchronous mode
Clock data recovery (CDR) mode
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
In source-synchronous mode, source synchronous interfacing is
supported at up to 840 Mbps. Serial channels are transmitted and received
along with a low speed clock. The receiving device then multiplies the
clock by a factor of 1 to 12, 14, 16, 18, or 20. The serialization/
deserialization rate can be any number from 4, 7, 8, 9 to 12, 14, 16, 18, or 20
and does not have to equal the clock multiplication value. For example, an
840-Mbps LVDS channel can be received along with a 84-MHz clock. The
84-MHz clock is multiplied by 10 to drive the serial shift register, but the
register can be clocked out in parallel at 7-, 8-, 9- to 12-, 14-, 16-, 18-, or
20-bits wide at 42 to 120 MHz. See Figures 2 and 3.
Figure 2. Receiver Diagram for Source Synchronous Mode
Notes (1), (2)
J Bits Wide
Receiver
Channel
Data to
LEs
13
1
J
Development
Tools
HSDI
×W
PLL2
HSDI_CLK2 (3)
Deserializer
+
—
To Global
Clock
Receiver Channel 1
Receiver
Channel
+
—
Receiver
Channel
+
—
Receiver Channel 2
Receiver Channel 8
Notes to Figure 2:
(1)
(2)
(3)
EP1M350 devices have 18 individual receiver channels. EP1M120 devices have 8 individual receiver channels.
W = 1 to 12, 14, 16, 18, or 20
J = 4, 7, 8, 9 to 12, 14, 16, 18, or 20
W does not have to equal J.
This clock pin drives an HSDI PLL only. It does not drive to the core.
Altera Corporation
9
Mercury Programmable Logic Device Family Data Sheet
Figure 3. Transmitter Diagram for Source Synchronous Mode
Notes (1), (2)
J Bits Wide
Serializer
Transmitter
Channel
Global Clock
from Receiver
or System Clock
HSDI
PLL1
HSDI_CLK1 (3)
×W
×W
B
Data from
LEs
1
J
Transmitter Channel 1
TXOUTCLOCK
Transmitter
Channel
Transmitter Channel 2
Transmitter
Channel
Transmitter Channel 8
Notes to Figure 3:
(1)
(2)
(3)
EP1M350 devices have 18 individual transmitter channels. EP1M120 devices have 8 individual transmitter
channels.
W = 1 to 12, 14, 16, 18, or 20
B = 1 to 12, 14, 16, 18, or 20
J = 4, 7, 8, 9 to 12, 14, 16, 18, or 20
W, B, and J do not have to be equal.
This clock pin drives an HSDI PLL only. It does not drive to the logic array.
The Mercury device’s source-synchronous mode also supports the
RapidIO interface protocol at up to 500 Mbps using the LVDS I/O
standard.
f
10
For more information on source synchronous interfacing see
AN 159: Using HSDI in Source-Synchronous Mode in Mercury Devices.
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Table 6 defines the support for source-synchronous mode applications.
Table 6. Source-Synchronous Mode
Data Rate
≤ 840 Mbps
I/O Standard
LVDS
LVPECL
3.3-V PCML
(1)
v
v
Note to Table 6:
(1)
You can use the CDR circuit to achieve data rates for DC coupled LVDS
applications. You must AC-couple the clock to a 2.2-V common mode voltage
(VCM) using the AC-coupling schemes in AN 134: Using Programmable I/O Standards
in Mercury Devices. The data channels should be DC-coupled. The byte alignment
relative to the clock is lost when using the CDR circuit. Therefore, a byte-alignment
circuit is required. Most Mercury source-synchronous designs already include
byte-alignment logic since they usually use DDR or SDR clocks. The CDR run
length requirement is waived if the reference clock and the receiver data come from
the same source and have the same frequency.
Altera Corporation
11
13
Development
Tools
In CDR mode, serial data is supported up to 1.25 Gbps per channel. The
system provides a reference clock which is multiplied by the receiver or
transmitter PLL to the same rate as the data is provided. For the receiver,
this multiplied reference clock is used by a CRU on each receiver channel
to generate a recovered clock in-phase with the received data. That
recovered clock drives the programmable deserializer and synchronizer.
The synchronizer is a FIFO for data transfer between the recovered clock
domain and the global clock domain. The dedicated synchronizers can be
bypassed if necessary. For every receiver channel in the EP1M350 and
EP1M120 devices, the ÷J recovered clock can drive a priority column line
for use as a clock. See Figure 4.
GCLK4
GCLK3
GCLK2
GCLK1
12
÷J
TX3 (TX8)
Serializer
Synchronizer
J
Data
(3)
+—
RX3 (RX8)
DPLL
(4)
÷J
TX4 (TX9)
Serializer
Synchronizer
J
Data
(3)
RX4 (RX9)
+—
Deserializer
DPLL
RCLK
÷J
(4)
Synchronizer
J
Data
+—
HSDI_CLK1 (5)
×W
×W
+—
HSDI_CLK2 (5)
8
PLL2
PLL1
8
REFCLK
REFCLK
÷J
(4)
TX5 (TX10)
Serializer
Synchronizer
J
Data
(3)
RX5 (RX10)
+—
Deserializer
DPLL
RCLK
÷J
(4)
Synchronizer
J
Data
÷J
TX6 (TX11)
Serializer
Synchronizer
J
Data
(3)
RX6 (RX11)
+—
Deserializer
DPLL
RCLK
÷J
Synchronizer
J
Data
Figure 4. Receiver & Transmitter Diagrams for CDR Mode
RCLK
÷J
Deserializer
Synchronizer
J
Data
4 Dedicated
Clocks
Mercury Programmable Logic Device Family Data Sheet
Notes (1), (2)
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Notes to Figure 4:
(1)
(2)
(3)
(4)
(5)
EP1M350 devices have 18 individual receiver and transmitter channels. EP1M120
devices have 8 individual receiver and transmitter channels. Receiver and
transmitter channel numbers in parenthesis are for EP1M350 devices.
W = 1 to 12, 14, 16, 18, or 20
J = 3 to 12, 14, 16, 18, or 20
W does not have to equal J.
For every receiver channel in EP1M350 and EP1M120 devices, the ÷J recovered
clock can drive the priority column interconnect for use as a clock.
The two center channels adjacent to the HSDI PLLs (channels 4 and 5 for EP1M120
devices, channels 9 and 10 for EP1M350 devices) can drive the Mercury device’s
global clocks.
HSDI_CLK1 and HSDI_CLK2 pins must be differential. These clock pins drive
HSDI PLLs only. They do not drive to the logic array.
The multiplied reference clock is also used to synchronize and serialize at
the transmitter side.
Up to two different serial data rates are supported for input channels or
output channels. Received data must be non-return-to-zero (NRZ).
Table 7 defines the support for CDR-mode applications. Table 8 shows the
supported data rates for each speed grade.
Data Rate
CDR Mode
DC-Coupled DC-Coupled DC-Coupled AC-Coupled AC-Coupled AC-Coupled
LVDS
LVPECL
3.3-V PCML LVDS (1) LVPECL (1) 3.3-V PCML
(1)
1.0 to 1.25 Gbps
(2)
v
v
v
v
v
≤ 1.0 Gbps
v
v
v
v
v
v
Notes to Table 7:
(1)
(2)
The VCM operating range for AC-coupled applications is from 0 to 0.7 V and from 1.8 to 2.4 V.
Use AC-coupled LVDS or another I/O standard. The DC-coupled LVDS I/O standard provides performance up to
1.0 Gbps.
f
Altera Corporation
For more information on CDR, see AN 130: CDR in Mercury Devices.
13
Development
Tools
Table 7. CDR-Mode Applications
13
Mercury Programmable Logic Device Family Data Sheet
1
Mercury device HSDI performance is finalized for certain speed
grades. Also, the industrial-grade CDR specification is the same
as the -6 speed grade for commercial-grade CDR specification.
See Table 8.
Table 8. CDR & Source-Synchronous Data Rates
Device
EP1M120
EP1M350
Speed Grade
Number of Channels
Maximum CDR Data
Rate (Gbps)
Maximum SourceSynchronous Data
Rate (Mbps)
-5
8
1.25
840
-6 (1)
8
1.25
840
840
-7
8
1.0
-5
18
1.25
840
-6 (1)
8 (2)
1.25
840
-7
10 (2)
1.0
840
18
1.0
840
Notes to Table 8:
(1)
(2)
The -6 speed grade specifications apply for both commercial and industrial devices.
EP1M350 devices can support any 8 channels at 1.25 Gbps. The other 10 channels must run at 1.0 Gbps or less.
Logic &
Interconnect
Mercury device logic is implemented in LEs. LE resources are used
differently according to specific operating modes and the type of logic
function being implemented. LEs are grouped into LABs in a row-based
architecture. The multi-level FastTrack Interconnect structure provides
the routing connection between LEs, ESBs, and IOEs.
Logic Array Block
Each LAB consists of 10 LEs, LE carry chains, multiplier circuitry, LAB
control signals, local interconnect, and FastLUT connection lines. The
local interconnect transfers signals between LEs within the same or
adjacent LABs. FastLUT connections transfer the output of one LE to the
adjacent LE for ultra-fast sequential LE connections within the same LAB.
The Quartus II Compiler places associated logic within a LAB or adjacent
LABs, allowing the use of fast local and FastLUT connections for high
performance. Figure 5 shows the Mercury LAB structure.
14
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Figure 5. Mercury LAB Structure
to LAB in Row Above
Row and Priority
Row Interconnect (1)
(2)
(3)
(3)
RapidLAB Interconnect
Column and Priority
Column Interconnect (1)
13
Development
Tools
Local Interconnect
Leap Line
Interconnect
The 10 LEs in the LAB are driven by
two local interconnect areas. The LAB
can drive two local interconnect areas.
to LAB in Row Below
Notes to Figure 5:
(1)
(2)
(3)
Priority column lines drive priority row lines, but not other row lines.
The RapidLAB interconnect can be driven by priority column lines, but not other column lines.
In multiplier mode, the RapidLAB interconnect drives LEs directly.
Mercury devices use an interleaved LAB structure, which allows each
LAB to drive two local interconnect areas. Every other LE drives to either
the left or right local interconnect area, alternating by LE. The local
interconnect can drive LEs within the same LAB or adjacent LABs. This
feature minimizes use of the row and column interconnects, providing
higher performance and flexibility. Each LAB structure can drive 30 LEs
through fast local interconnects.
Altera Corporation
15
Mercury Programmable Logic Device Family Data Sheet
Each LAB contains dedicated logic for driving control signals to its LEs.
The control signals include clock, clock enable, asynchronous clear,
asynchronous preset, asynchronous load, synchronous clear, and
synchronous load signals. A maximum of six control signals can be used
at a time. Although synchronous load and clear signals are generally used
when implementing counters, they can also be used with other functions.
Each LAB can use two clocks and two clock enable signals. Each LAB’s
clock and clock enable signals are linked (e.g., any LE in a particular LAB
using LABCLK1 will also use LABCLKENA1). In addition to LAB-wide
control of clock enables, Mercury devices can also control clock enable
signals on individual LEs, allowing more than two clock enables in a
given LAB. The Quartus II software automatically chooses whether a
clock enable is LAB-wide for individual LEs. If both the rising and falling
edges of a clock are used in a LAB, both LAB-wide clock signals are used.
The LAB local interconnect, fast global signals, row-global signals, and
dedicated clock pins can generate the LAB-wide control signals. The
multi-level FastTrack Interconnect’s inherent low skew allows it to be
used for clock distribution. Figure 6 shows the LAB control signal
generation circuit.
16
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Figure 6. LAB-Wide Control Signals
Dedicated
Clocks
4
Fast Global
Signals
6
Row Global
Signals
2
Local
Interconnect
Local
Interconnect
Local
Interconnect
Local
Interconnect
ASYNCLOAD
or LABPRE
LABCLK1
LABCLKENA1
13
LABCLR
Logic Element
The LE, the smallest unit of logic in the Mercury architecture, is compact
and provides efficient logic usage. Each LE contains a four-input LUT,
which is a function generator that can quickly implement any function of
four variables. In addition, each LE contains a programmable register and
carry chain with carry select look ahead capability. Each LE drives all
interconnect types: local interconnect, row and priority row interconnect,
column and priority column interconnect, leap lines, and RapidLAB
interconnect. Each LE also has the ability to drive its combinatorial output
directly to the next LE in the LAB using FastLUT connections. See
Figure 7.
Altera Corporation
17
Development
Tools
SYNCCLR
or LABCLK2
SYNCLOAD
or LABCLKENA2
Mercury Programmable Logic Device Family Data Sheet
Figure 7. Mercury LE
LAB-wide
Synchronous
Load
LAB Carry-In
Packed
Register Select
LAB-wide
Synchronous
Clear
Carry-In1
Carry-In0
data1
data2
data3
data4 (1)
Register Bypass
Look-Up
Table
(LUT)
Carry
Chain
Synchronous
Load and
Clear Logic
Programmable
Register
D
PRN
Q
FastLUT
Routing to next LE
to Local, Row, and
Column Routing
ENA
CLRN
to Local, Row, and
Column Routing
labclr
labpre
Chip-Wide
Reset
Asynchronous
Clear/Preset/
Load Logic
Clock &
Clock Enable
Select
labclk1
labclk2
labclkena1
labclkena2
Carry-Out0
Carry-Out1
LE Clock
Enable
LAB Carry-Out (2)
Notes to Figure 7:
(1)
(2)
FastLUT interconnect uses the data4 input.
LAB carry-out can only be generated by LE 4 and/or LE 10.
Each LE’s programmable register can be configured for D, T, JK, or SR
operation. The register’s clock, clock enable, and clear control signals can
be driven by global signals, general-purpose I/O pins, or any internal
logic. For combinatorial functions, the register is bypassed and the output
of the LUT drives directly to the outputs of the LE.
Each LE has four data inputs that can drive the internal LUT. One of these
inputs has a shorter delay than the others, improving overall LE
performance. This input is chosen automatically by the Quartus II
software as appropriate.
18
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Each LE has two outputs that drive the local, row, and column routing
resources. Each output can be driven independently by the LUT’s or
register’s output. For example, the LUT can drive one output, while the
register drives the other output. This feature, called register packing,
improves device utilization because the register and the LUT can be used
for unrelated functions. The LE can also drive out registered and
unregistered versions of the LUT output.
LE Operating Modes
The Mercury LE can operate in one of the following modes:
■
■
■
Normal
Arithmetic
Multiplier
The Quartus II software, in conjunction with parameterized functions
such as LPM and DesignWare functions, automatically chooses the
appropriate mode for common functions, such as counters, adders, and
multipliers. If required, the designer can also create special-purpose
functions that specify which LE operating mode to use for optimal
performance.
Normal Mode
The normal mode is suitable for general logic applications and
combinatorial functions. In normal mode, four data inputs from the LAB
local interconnect and a single carry-in are inputs to a four-input LUT. The
Quartus II Compiler automatically selects the carry-in or the data3 signal
as one of the inputs to the LUT. The LUT (combinatorial) output can be
driven to the FastLUT connection to the next LE in the LAB. LEs in normal
mode support packed registers. Figure 8 shows an LE in normal mode.
Altera Corporation
19
13
Development
Tools
Each operating mode uses LE resources differently. In each operating
mode, eight available inputs to the LE—the four data inputs from the LAB
local interconnect; carry-in0, carry-in1 from the previous LE; the
LAB carry-in from the previous carry-chain generation; and the FastLUT
Connection input from the previous LE—are directed to different
destinations to implement the desired logic function. LAB-wide signals
provide clock, asynchronous clear, asynchronous preset, asynchronous
load, synchronous clear, synchronous load, and clock enable control for
the register. These LAB-wide signals are available in all normal and
arithmetic LE modes.
Mercury Programmable Logic Device Family Data Sheet
Figure 8. Normal-Mode LE
Note (1)
LE-Out
Combinatorial
Output
data1
data2
data3
Carry-In from
Previous LE (2)
data4
LE-Out
Registered
4-Input
LUT
PRN/ALDn Output
D
Q
LE-Out
ENA
CLRN
LAB-Wide Clock Enable (3)
Notes to Figure 8:
(1)
(2)
(3)
LEs in normal mode support register packing.
When using the carry-in in normal mode, the packed register feature is unavailable.
There are two LAB-wide clock enables per LAB in addition to LE-specific clock enables.
Arithmetic Mode
The arithmetic mode is ideal for implementing adders, accumulators, and
comparators. A LE in arithmetic mode contains four 2-input LUTs. The
first two 2-input LUTs compute two summations based on a possible
carry of 1 or 0; the other two LUTs generate carry outputs for the two
possible chains of the carry-select look-ahead (CSLA) circuitry. As shown
in Figure 9, the LAB carry-in signal selects the appropriate carry-in chain
(either carry-in0 or carry-in1). The logic level of the chain selected
in turn selects which parallel sum is generated as a combinatorial or
registered output. For example, when implementing an adder, this output
is the signal comprised of the sum data1 + data2 + carry, where carry is
0 or 1. The other two LUTs use the data1 and data2 signals to generate
two possible carry-out signals—one for a carry of 1 and the other for a
carry of 0. The carry-in0 signal acts as the carry select for the
carry-out0 output; carry-in1 acts as the carry select for the
carry-out1 output. LEs in arithmetic mode can drive out registered and
unregistered versions of the LUT output. Figure 9 shows a Mercury LE in
arithmetic mode.
20
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
The arithmetic mode also offers clock enable, counter enable, synchronous
up/down control, synchronous clear, and synchronous load options. The
counter enable and synchronous up/down control signals are generated
from the data inputs of the LAB local interconnect. The synchronous clear
and synchronous load options are LAB-wide signals that affect all
registers in the LAB. Consequently, if any of the LEs in a LAB use the
counter mode, other LEs in that LAB must be used as part of the same
counter or be used for a combinatorial function. The Quartus II software
automatically places any registers that are not used by the counter into
other LABs.
Figure 9. Arithmetic Mode LE
LAB-Wide
Synchronous
Clear
LAB Carry-In
Carry-In0
Carry-In1
data1
data2
LUT
LAB-Wide
Synchronous
Load
Combinatorial
Output
LE-Out
Sum
Registered
PRN/ALDn Output
D
LUT
ENA
CLRn
13
Q
LE-Out
LUT
LAB-Wide
Clock Enable
data3
Carry-Out0
Carry-Out1
Carry-Select Look-Ahead Chain
The CSLA chain provides a very fast carry-forward function between LEs
in arithmetic mode or multiplier mode. The CSLA chain uses the
redundant carry calculation to increase the speed of carry functions. The
LE can calculate sum and carry values for a possible carry-in of 1 and
carry-in of 0 in parallel. The carry-in0 and carry-in1 signals from a
lower-order bit drive forward into the higher-order bit via the parallel
carry chain and feed into both the LUT and the next portion of the CSLA
chain. CSLA chains can begin in any LE within a LAB.
Altera Corporation
21
Development
Tools
LUT
Mercury Programmable Logic Device Family Data Sheet
The CSLA chain’s speed advantage results from the parallel precomputation of carry chains. Instead of including every LUT in the critical
path, only the propagation delays between LAB carry-in generation
circuits (LE 4 and LE 10) make up the critical path. This feature allows the
Mercury architecture to implement high-speed counters, adders,
multipliers, parity functions, and comparators of arbitrary width.
Figure 10 shows the CSLA circuitry in a LAB for a 10-bit full adder. One
portion of the LUT generates the sum of two bits using the input signals
and the appropriate carry-in bit; the sum is routed to the output of the LE.
The register can be bypassed for simple adders or used for accumulator
functions. Another portion of the LUT generates carry-out bits. A labwide carry-in bit selects which chain is used for the addition of given
inputs. The actual carry-in signal for that selected chain, carry-in0 or
carry-in1, selects the carry-out to carry forward, which is routed to the
carry-in signal of the next-higher-order bit. The final carry-out signal is
routed to an LE, where it is driven to local, row, or column interconnects.
22
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Figure 10. CSLA Details
LAB Carry-In
0
1
A1
B1
LE1
A2
B2
LE2
Sum1
LAB Carry-In
Carry-In0
Carry-In1
A3
B3
LE3
A4
B4
LE4
Sum2
Sum3
Sum4
LUT
data1
data2
Sum
LUT
LUT
LUT
0
1
LE5
A6
B6
LE6
A7
B7
LE7
A8
B8
LE8
A9
B9
LE9
A10
B10
LE10
Sum5
Sum6
Carry-Out0
13
Carry-Out1
Development
Tools
A5
B5
Sum7
Sum8
Sum9
Sum10
LAB Carry-Out
The Quartus II Compiler can create CSLA logic automatically during
design processing. Alternatively, the designer can create CSLA logic
manually during design entry. Parameterized functions such as library of
parameterized modules (LPM) and DesignWare functions automatically
take advantage of carry chains for the appropriate functions.
Altera Corporation
23
Mercury Programmable Logic Device Family Data Sheet
The Quartus II Compiler creates carry chains longer than ten LEs by
linking LABs together automatically. For enhanced fitting, a long carry
chain skips intermediate LABs in a row structure. A carry chain longer
than one LAB skips either from an even-numbered LAB to the next evennumbered LAB, or from an odd-numbered LAB to the next oddnumbered LAB. For example, the last LE of the first LAB in a LAB row
carries to the first LE of the third LAB in the same LAB row.
Multiplier Mode
Multiplier mode is used for implementing high-speed multipliers up to
16 × 16 in size. The LUT implements the partial product formation and
summation in a single stage for a N × M-bit multiply operation. A single
LE can implement the summation of ANBM + 1 + AN + 1BM for the
multiplier and multiplicand inputs. To increase the speed of the
multiplication, LAB wide signals are used to control the partial product
sum generation. These multiplier LAB-wide signals use the LABCLKENA1
and PRESET/ASYNCLOAD resources. The multiplier mode takes
advantage of the CSLA circuitry for optimized sum and carry generation
in the partial product sum. There is a special CSLA circuitry mode used
for the multiplier where the carry chain runs vertically between LABs in
the same column. The Quartus II Compiler automatically uses this special
mode for dedicated multiplier implementation only. The summation of
the multiplier and multiplicand bits is driven out along with the carryout0 and carry-out1 bits. The combinatorial or registered versions of
the sum can be driven out, allowing the multiplier to be pipelined.
The RapidLAB interconnect has dedicated fast connections to the LE
inputs in multiplier mode, further increasing the speed of the multiplier.
These dedicated connections allow RapidLAB lines to avoid delay
incurred by driving onto local interconnects and then into the LE.
The Quartus II software implements parameterized functions that use the
multiplier mode automatically when multiply operators are used.
Figure 11 shows a Mercury device LE in multiplier mode.
24
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Figure 11. Multiplier Mode LE
LAB-Wide Clock
Enable Signals (1)
LAB Carry-In
Carry-In0
Carry-In1
AN
Programmable
Inverter
BM + 1
Programmable
Inverter
AN + 1
Programmable
Inverter
BM
Programmable
Inverter
Combinatorial
Sum Output
RapidLAB
Interconnect (2)
Combinatorial
Sum Output
LE Output
D
Registered
Sum Output
Full
Adder
LAB Carry-Out
Partial Product
Generation
LE Output
Q
ENA
CLRN
Carry-Out1
13
Carry-Out0
(1)
(2)
LABCLKENA1 cannot be used in multiplier mode.
When the RapidLAB output is used, local interconnect outputs are unavailable.
The basis for the high-speed 16 × 16-bit multiplier in a Mercury device is
the binary tree multiplier. In the first stage of the binary tree, the
multiplicand bits, a[15:0], and the multiplier bits, b[15:0], are
multiplied together. The results of the first stage are sixteen 16-bit partial
products, a[15:0]b[15], a[15:0]b[14], . . . a[15:0]b[0]. The
partial products are then grouped into pairs and added together in the
second stage. In a similar fashion, the results of the previous stage are
grouped in pairs and then added forming the binary tree structure seen in
Figure 12.
Altera Corporation
25
Development
Tools
Notes to Figure 11:
26
B15
A15
B14
A14
B13
A13
B12
A12
A6B9
A7B8
A5B9
A6B8
A7B7
A8B6
A4B9
A5B8
A6B7
A7B6
A3B9
A4B8
A5B7
A6B6
A7B5
A8B4
A2B9
A3B8
A4B7
A5B6
A6B5
A7B4
A8B3
A1B9
A2B8
A3B7
A4B6
A5B5
A6B4
A7B3
A8B2
B8
A8
B7
A7
B6
A6
B5
A5
B4
A4
B3
A3
B2
A2
B1
A1
B0
A0
A0B9
A1B8
A2B7
A3B6
A4B5
A5B4
A6B3
A7B2
A8B1
A0B8
A1B7
A2B6
A3B5
A4B4
A5B3
A6B2
A7B1
A8B0
A0B7
A1B6
A2B5
A3B4
A4B3
A5B2
A6B1
A7B0
A0B6
A1B5
A2B4
A3B3
A4B2
A5B1
A6B0
A0B5
A1B4
A2B3
A3B2
A4B1
A5B0
A0B4
A1B3
A2B2
A3B1
A4B0
A0B3
A1B2
A2B1
A3B0
A0B2
A1B1
A2B0
A0B1
A1B0
A0B0
== == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == ==
A15B15 A14B15 A13B15 A12B15 A11B15 A10B15 A9B15 A8B15 A7B15 A6B15 A5B15 A4B15 A3B15 A2B15 A1B15 A0B15
A15B14 A14B14 A13B14 A12B14 A11B14 A10B14 A9B14 A8B14 A7B14 A6B14 A5B14 A4B14 A3B14 A2B14 A1B14 A0B14
A15B13 A14B13 A13B13 A12B13 A11B13 A10B13 A9B13 A8B13 A7B13 A6B13 A5B13 A4B13 A3B13 A2B13 A1B13 A0B13
A15B12 A14B12 A13B12 A12B12 A11B12 A10B12 A9B12 A8B12 A7B12 A6B12 A5B12 A4B12 A3B12 A2B12 A1B12 A0B12
A15B11 A14B11 A13B11 A12B11 A11B11 A10B11 A9B11 A8B11 A7B11 A6B11 A5B11 A4B11 A3B11 A2B11 A1B11 A0B11
A15B10 A14B10 A13B10 A12B10 A11B10 A10B10 A9B10 A8B10 A7B10 A6B10 A5B10 A4B10 A3B10 A2B10 A1B10 A0B10
A7B9
A8B8
A8B7
A15B6 A14B6 A13B6 A12B6 A11B6 A10B6 A9B6
A15B7 A14B7 A13B7 A12B7 A11B7 A10B7 A9B7
A8B5
A15B4 A14B4 A13B4 A12B4 A11B4 A10B4 A9B4
A15B5 A14B5 A13B5 A12B5 A11B5 A10B5 A9B5
A8B9
B9
A9
A15B0 A14B0 A13B0 A12B0 A11B0 A10B0 A9B0
A15B2 A14B2 A13B2 A12B2 A11B2 A10B2 A9B2
A15B8 A14B8 A13B8 A12B8 A11B8 A10B8 A9B8
B10
A10
A15B1 A14B1 A13B1 A12B1 A11B1 A10B1 A9B1
A15B3 A14B3 A13B3 A12B3 A11B3 A10B3 A9B3
A15B9 A14B9 A13B9 A12B9 A11B9 A10B9 A9B9
B11
A11
== == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == == ==
×
Sixteen 16-Bit
Partial Products
Mercury Programmable Logic Device Family Data Sheet
Figure 12. Partial Product Formation
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
For a typical 16 × 16-bit binary tree multiplier, five stages are needed to
determine the final product. The Mercury LE multiplier mode allows the
partial product formation stage (Stage 1) and the first sum of stages
(Stage 2) to be combined in a single stage, shown in Figure 13. This
feature, combined with the direct connection between RapidLAB lines
and LEs in multiplier mode, allows the fast dedicated implementation of
multipliers.
13
Development
Tools
Altera Corporation
27
28
+
Stages 1 and 2 are
combined into one level
of LEs
Stage 5
A[15:0] B[14]
Stage 4
Stage 3
A[15:0] B[15]
Stage 2
Stage 1
+
A[15:0] B[13]
+
A[15:0] B[12]
+
A[15:0] B[11]
+
A[15:0] B[10]
+
A[15:0] B[9]
+
A[15:0] B[8]
+
A[15:0] B[7]
+
A[15:0] B[6]
+
A[15:0] B[5]
+
A[15:0] B[4]
+
A[15:0] B[3]
+
A[15:0] B[2]
+
A[15:0] B[1]
+
A[15:0] B[0]
Mercury Programmable Logic Device Family Data Sheet
Figure 13. Mercury Binary Tree Implementation
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Clear & Preset Logic Control
LAB-wide signals control logic for the register’s clear and preset signals.
The LE directly supports an asynchronous clear and preset function. The
direct asynchronous preset does not require a NOT-gate push-back
technique. Mercury devices support simultaneous preset, or
asynchronous load, and clear. Asynchronous clear takes precedence if
both signals are asserted simultaneously. Each LAB supports one clear
and one preset signal. Two clears are possible in a single LAB by using a
NOT-gate push-back technique on the preset port. The Quartus II
Compiler automatically performs this second clear emulation.
In addition to the clear and preset ports, Mercury devices provide a chipwide reset pin (DEV_CLRn) that resets all registers in the device. Use of
this pin is controlled through an option in the Quartus II software that is
set before compilation. The chip-wide reset overrides all other control
signals.
Multi-Level FastTrack Interconnect
This network of routing structures provides predictable performance,
even for complex designs. In contrast, the segmented routing in FPGAs
requires switch matrices to connect a variable number of routing paths,
increasing the delays between logic resources and reducing performance.
The Multi-Level FastTrack Interconnect consists of regular and priority
lines that traverse column and row interconnect channels to span sections
and the entire device length. Each row of LABs, ESBs, and I/O bands is
served by a dedicated row interconnect, which routes signals to and from
LABs, ESBs, and I/O row bands in the same row. These row resources
include:
■
■
■
Altera Corporation
Row interconnect traversing the entire device from left to right
Priority row interconnect for high speed access across the length of
the device
RapidLAB interconnect for horizontal routing that traverses a
10-LAB-wide region from a central LAB
29
13
Development
Tools
The Mercury architecture provides connections between LEs, ESBs, and
device I/O pins via an innovative Multi-Level FastTrack Interconnect
structure. The Multi-Level FastTrack Interconnect structure is a series of
routing channels that traverse the device, providing a hierarchy of
interconnect lines. Regular resources provide efficient and capable
connections while priority resources and specialized RapidLAB, leap line,
and FastLUT resources enhance performance by accelerating timing on
critical paths. The Quartus II Compiler automatically places critical design
paths on those faster lines to improve design performance.
Mercury Programmable Logic Device Family Data Sheet
The RapidLAB interconnect provides a specialized high-speed structure
to allow a central LAB to drive other LABs within a 10-LAB-wide region.
The RapidLAB lines drive alternating local LAB interconnect regions,
allowing communication to all LABs in the 10-LAB-wide region. Even
numbered LEs in a LAB directly drive a RapidLAB line that drives one set
of alternating local interconnect regions, while odd-numbered LEs drive
a RapidLAB line that drives the opposite set of alternating local
interconnect regions. Figure 14 shows RapidLAB interconnect
connections. This 10-LAB wide region of the RapidLAB interconnect is
repeated for every LAB in the row. The region covered by the RapidLAB
interconnect is smaller than 10 for source LABs that are four or five LABs
in from either edge of the LAB row. The RapidLAB row interconnect is
used for LAB-to-LAB routing; it is only used by I/O bands or ESBs
indirectly through other interconnects. The RapidLAB interconnect drives
an LE directly when that LE is in multiplier mode.
30
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Figure 14. RapidLAB Interconnect Connections
RapidLAB Interconnect
LE 1
LE 3
LE 5
LE 7
LE 9
RapidLAB interconnects driven by odd-numbered
LEs can drive out to the four LEs to the left and five
LEs to the right through local interconnects.
Local Interconnect
LAB
LE 2
LE 4
LE 6
13
LE 8
RapidLAB interconnects driven by even-numbered
LEs can drive out to the four LEs to the right and five
LEs to the left through local interconnects.
The column interconnect vertically routes signals to and from LABs, ESBs,
and I/O bands. Each column of LABs is served by a dedicated column
interconnect. These column resources include:
■
■
■
Column interconnect traversing the entire device from top to bottom
Priority column interconnect for high speed access across the device
vertically
Leap line interconnect for vertical routing between adjacent LAB
rows and between adjacent ESP rows and LAB rows.
Leap lines are driven directly by LEs for fast access to adjacent row
interconnects. LABs can drive a leap line to the row above and/or below
(including ESB rows). The even-numbered LEs in a LAB drive leap lines
down, while odd-numbered LEs drive leap lines up. This allows a single
LAB to access row and RapidLAB interconnects within a three-row
region. Figure 15 shows the leap line interconnect.
Altera Corporation
31
Development
Tools
LE 10
Mercury Programmable Logic Device Family Data Sheet
Figure 15. Leap Line Interconnect
Leap Line
LAB Row n-1
Priority Row
and Row
LE 1
LE 1
LE 1
LE 2
LE 2
LE 2
LE 3
LE 3
LE 3
LE 4
LE 4
LE 4
LE 5
LE 5
LE 5
LE 6
LE 6
LE 6
LE 7
LE 7
LE 7
LE 8
LE 8
LE 8
LE 9
LE 9
LE 9
LE 10
LE 10
LE 10
LAB Row n
LAB Row n+1
RapidLAB
Leap Line
FastLUT Interconnect
Mercury devices include an enhanced interconnect structure within LABs
for faster routing of LE output to LE input connections. The FastLUT
connection allows the combinatorial output of an LE to directly drive the
fast input of the LE directly below it, bypassing the local interconnect. This
resource can be used as a high speed connection for wide fan-in functions
from LE 1 to LE 10 in the same LAB. Figure 16 shows a FastLUT
interconnect.
32
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Figure 16. FastLUT Interconnect
Local Interconnect
Routing Among LEs
in the LAB
FastLUT
Routing to
Adjacent LE
LE 1
LE 2
LE 3
LE 4
LE 5
LE 6
LE 7
13
Development
Tools
LE 8
LE 9
LE 10
ESB rows also have their own interconnect resources to communicate
horizontally and vertically with LAB rows. The ESB rows at the top and
bottom of the device have their own set of row and priority row
interconnect resources. For vertical communication, all LAB column
interconnect lines traverse to the ESBs. This includes leap lines, which
allow the adjacent LAB rows to communicate with the ESBs.
The row interconnect resources can be driven directly by LEs or ESBs in
that row. Further, the column interconnect resources can drive a row line,
allowing LEs, IOEs, and ESBs to drive elements in a different row via the
column and row resources.
The column interconnect resources can be directly driven by LEs, IOEs, or
ESBs within that column. The priority column and leap line resources can
be driven directly by LEs. These lines enable high-speed vertical
communication in the device for timing-critical paths. The column
resources route signals between rows. A column resource can drive row
resources directly, allowing fast connections between rows.
Altera Corporation
33
Mercury Programmable Logic Device Family Data Sheet
Table 9 summarizes how various elements of the Mercury architecture
drive each other.
Table 9. Mercury Routing Scheme
Source
LE
Destination
LE
Local
Interconnect
v
v
IOE
ESB Row
ESB
Interconnect
Row Priority RapidLAB Column Priority Leap
Row
Interconnect
Column Lines
v
v
v
v (3)
v
v
v
v
v
v
v
v
(1)
Local
v
Interconnect
v
v (2)
IOE
(3)
v
ESB Row
Interconnect
v
ESB
Row
v
Priority Row
v
RapidLAB
v
Interconnect (4)
v
Column
v
Priority
Column
v
Leap Lines
v
v
v
v
v
v
v
v
v
v
v
v
v
Notes to Table 9:
(1)
(2)
(3)
(4)
This direct connection is possible through the FastLUT connection.
IOEs can connect to the adjacent LAB’s local interconnects in the associated LAB row.
IOEs can connect to row and priority row interconnects in the associated LAB row.
This connection is used for multiplier mode.
Embedded
System Block
The ESB can implement various types of memory blocks, including quadport, true dual-port, dual- and single-port RAM, ROM, FIFO, and CAM
blocks.
The ESB includes input and output registers; the input registers
synchronize reads and/or writes, and the output registers can pipeline
designs to further increase system performance. The ESB offers a quad
port mode, which supports up to four port operations, two reads and two
writes simultaneously, with the ability for a different clock on each of the
four ports. Figure 17 shows the ESB quad-port block diagram.
34
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Figure 17. ESB Quad-Port Block Diagram
A
B
dataA[ ]
wraddressA[ ]
wrenA
inclockA
inclockenA
inaclrA
dataB[ ]
wraddressB[ ]
wrenB
inclockB
inclockenB
inaclrB
rdaddressA[ ]
rdenA
qA[ ]
outclockA
outclockenA
outaclrA
rdaddressB[ ]
rdenB
qB[ ]
outclockB
outclockenB
outaclrB
Altera Corporation
35
13
Development
Tools
In addition to quad port memory, the ESB also supports true dual-port,
dual-port, and single-port RAM. True dual-port RAM supports any
combination of two port operations: two reads, two writes, or one read
and one write. Dual-port memory supports a simultaneous read and
write. For single-port memory, independent read and write is supported.
Figure 18 shows these different RAM memory port configurations for an
ESB.
Mercury Programmable Logic Device Family Data Sheet
Figure 18. RAM Memory Port Configurations
True Dual-Port Memory
A
dataA[ ]
addressA[ ]
wrenA
clockA
clockenA
qA[ ]
aclrA
B
dataB[ ]
addressB[ ]
wrenB
clockB
clockenB
qB[ ]
aclrB
Dual-Port Memory (1)
data[ ]
wraddress[ ]
wren
inclock
inclocken
inaclr
rdaddress[ ]
rden
q[ ]
outclock
outclocken
outaclr
Single-Port Memory (1)
data[ ]
address[ ]
wren
inclock
inclocken
inaclr
q[ ]
outclock
outclocken
outaclr
Note to Figure 18:
(1)
Two dual- or single-port memory blocks can be implemented in a single ESB.
The ESB also allows variable width data ports for reading and writing to
any of the RAM ports in any RAM configuration. For example, the ESB in
quad port configuration can be written in ×1 mode at port A, read in ×16
from port A, written in ×4 mode at port B, and read in ×2 mode from
port B.
36
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
ESBs can implement synchronous RAM, which is easier to use than
asynchronous RAM. A circuit using asynchronous RAM must generate
the RAM write enable (WE) signal while ensuring that its data and address
signals meet setup and hold time specifications relative to the WE signal.
In contrast, the ESB’s synchronous RAM generates its own WE signal and
is self-timed with respect to the global clock. Circuits using the ESB’s selftimed RAM must only meet the setup and hold time specifications relative
to the global clock.
ESBs are grouped together in rows at the top and bottom of the device for
fast horizontal communication. The ESB row interconnect can be driven
by any ESB in the row. The row interconnect drives the ESB local
interconnect, which in turn drives the ESB ports. ESB outputs drive the
ESB local interconnect, which can drive row interconnect as well as all
types of column interconnect, including leap lines. The leap lines allow
fast access between ESBs and the adjacent LAB row.
The ESB can also be split in half and used for two independent 2,048-bit
single-port or dual-port RAM blocks. For example, one half of the ESB can
be used as a 128 × 16 memory single-port memory while the other half can
be used for a 1,024 × 2 dual-port memory. This effectively doubles the
number of RAMs a Mercury device can implement for its given number
of ESBs. The Quartus II software automatically merges two logical
memory functions in a design into an ESB; the designer does not need to
merge the functions manually.
By combining multiple ESBs, the Quartus II software implements larger
memory blocks automatically. For example, two 256 × 16 RAM blocks can
be combined to form a 256 × 32 RAM block, and two 512 × 8 RAM blocks
can be combined to form a 512 × 16 RAM block. Memory performance
does not degrade for memory blocks up to 4,096 words deep. Each ESB
can implement a 4,096-word-deep memory; the ESBs are used in parallel,
eliminating the need for any external control logic and its associated
delays. To create a high-speed memory block more than 4,096 words
deep, the Quartus II software will automatically combine ESBs with LE
control logic.
Altera Corporation
37
13
Development
Tools
When implementing memory, each ESB can be configured in any of the
following sizes for quad port and true dual-port memory modes: 256 × 16;
512 × 8; 1,024 × 4; 2,048 × 2; or 4,096 × 1. For dual-port and single-port
modes, the ESB can be configured for 128 × 32 in addition to the list above.
For variable port width RAMs, any port width ratio combination must be
1, 2, 4, 8, or 16. For example, a RAM with data ports of width 1 and 16 or
2 and 32 will work, but not 1 and 32.
Mercury Programmable Logic Device Family Data Sheet
The ESB implements two forms of clocking modes for quad-port and
dual-port memory—read/write clock mode and input/output clock
mode.
Read/Write Clock Mode
An ESB implementing quad-port memory in read/write clock mode can
use up to four clocks. For port A, one clock controls all registers associated
with writing: data input, WE, and write address. The other clock controls
all registers associated with reading: read enable (RE), read address, and
data output. Another set of clocks can be used for port B of the RAM, or
the same clocks can be used. Each ESB port, A or B, also supports
independent read clock enable, write clock enable, and asynchronous
clear signals. Read/write clock mode is commonly used for applications
where reads and writes occur at different system frequencies. Figure 19
shows the ESB in read/write clock mode.
38
Altera Corporation
(1)
(2)
(3)
Altera Corporation
outclockA
inclockA
inclkenA
wrenA
outclkenA
rdenA
wraddressA[]
rdaddressA[]
dataA[]
4
6
ENA
D
ENA
D
ENA
D
Q
Q
Q
Q
Q
Write
Pulse
Generator
ENA
D
ENA
D
Six Dedicated Inputs & Global Signals
ENA
D
Q
Data Out[]
qA[]
Write Enable
Read Enable
Write Address
Data In
B
qB[]
Q
D
ENA
Data Out[]
Write Enable
Read Enable
Write Address
Read Address
RAM/ROM
128 × 32 (3)
256 × 16
512 × 8
1,024 × 4
2,048 × 2
4,096 × 1
Read Address
Data In
A
D
ENA
D
ENA
Write
Pulse
Generator
Q
Q
Q
Q
Q
D
ENA
D
ENA
D
ENA
6
4
outclockB
inclockB
inclkenB
wrenB
outclkenB
rdenB
wraddressB[]
rdaddressB[]
dataB[]
Figure 19. ESB in Read/Write Clock Mode
Development
Tools
Four Dedicated Clocks
Mercury Programmable Logic Device Family Data Sheet
Notes (1), (2)
13
Notes to Figure 19:
Only half of the ESB, either A or B, is used for dual-port configuration.
All registers can be asynchronously cleared by ESB local interconnect signals, global signals, or the chip-wide reset.
This configuration is supported for dual-port configuration.
39
Mercury Programmable Logic Device Family Data Sheet
Input/Output Clock Mode
An ESB using input/output clock mode can also use up to four clocks. On
each of the two ports, A or B, one clock controls all registers for inputs into
the ESB: data input, WE, RE, read address, and write address. The other
clock controls the ESB data output registers. Each ESB port, A or B, also
supports independent read clock enable, write clock enable, and
asynchronous clear signals. Input/output clock mode is commonly used
for applications where the reads and writes occur at the same system
frequency, but require different clock enable signals for the input and
output registers. Figure 20 shows the ESB in input/output clock mode.
40
Altera Corporation
(1)
(2)
(3)
Altera Corporation
outclockA
inclockA
inclkenA
outclkenA
wrenA
rdenA
wraddressA[ ]
rdaddressA[ ]
dataA[ ]
4
6
ENA
D
ENA
D
ENA
D
Q
Q
Q
Q
Q
Write
Pulse
Generator
ENA
D
ENA
D
Six Dedicated Inputs & Global Signals
ENA
D
Q
Data Out[]
Write Enable
Read Enable
Write Address
Read Address
Data In
A
qA[]
Data In
B
Q
D
ENA
Data Out[]
Write Enable
Read Enable
Write Address
Read Address
qB[]
RAM/ROM
128 × 32 (3)
256 × 16
512 × 8
1,024 × 4
2,048 × 2
4,096 × 1
D
ENA
D
ENA
Write
Pulse
Generator
Q
Q
Q
Q
Q
D
ENA
D
D
ENA
ENA
6
4
outclockB
inclockB
inclkenB
outclkenB
wrenB
rdenB
wraddressB[ ]
rdaddressB[ ]
dataB[ ]
Figure 20. ESB in Input/Output Clock Mode
Development
Tools
Four Dedicated Clocks
Mercury Programmable Logic Device Family Data Sheet
Notes (1), (2)
13
Notes to Figure 20:
Only half of the ESB, either A or B, is used for dual-port configuration.
All registers can be asynchronously cleared by ESB local interconnect signals, global signals, or the chip-wide reset.
This configuration is supported for dual-port configuration.
41
Mercury Programmable Logic Device Family Data Sheet
Single-Port Mode
The Mercury device’s ESB also supports a single-port mode, which is used
when simultaneous reads and writes are not required. See Figure 21. A
single ESB can support up to two single-port mode RAMs.
Figure 21. ESB in Single-Port Mode
Note (1)
Dedicated Fast
Global Signals
Dedicated Clocks
4
data[ ]
RAM/ROM (2)
128 × 16
256 × 8
512 × 4
Data In
1,024 × 2
2,048 × 1
6
D
Q
ENA
Data Out
D
Q
To FastTrack
Interconnect
ENA
Address
address[ ]
D
Q
ENA
wren
outclken
inclken
Write Enable
D
ENA
inclock
Q
Write
Pulse
Generator
outclock
Notes to Figure 21:
(1)
(2)
All registers can be asynchronously cleared by ESB local interconnect signals, global signals, or chip-wide reset.
If there is only one single-port RAM block in an ESB, it can support the following configurations: 4,096 × 1; 2,048 × 2;
1,028 × 4; 512 × 8; 256 × 16; or 128 × 32.
Content-Addressable Memory
Mercury devices can implement CAM in ESBs. CAM can be thought of as
the inverse of RAM. RAM stores data in a specific location; when the
system submits an address, the RAM block provides the data. Conversely,
when the system submits data to CAM, the CAM block provides the
address where the data is found. For example, if the data FA12 is stored
in address 14, the CAM outputs 14 when FA12 is driven into it.
42
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
CAM is used for high-speed search operations. When searching for data
within a RAM block, the search is performed serially. Thus, finding a
particular data word can take many cycles. CAM searches all addresses in
parallel and outputs the address storing a particular word. When a match
is found, a match flag is set high. CAM is ideally suited for applications
such as Ethernet address lookup, data compression, pattern recognition,
cache tags, fast routing table lookup, and high-bandwidth address
filtering. Figure 22 shows the CAM block diagram.
Figure 22. CAM Block Diagram
data[ ]
wraddress[ ]
wren
inclock
inclocken
inaclr
data_address[ ]
match
outclock
outclocken
outaclr
13
When in CAM mode, the ESB implements a 32-word, 32-bit CAM. Wider
or deeper CAM, such as a 32-word, 64-bit or 128-word, 32-bit block, can
be implemented by combining multiple CAM blocks with some ancillary
logic implemented in LEs. The Quartus II software automatically
combines ESBs and LEs to create larger CAM blocks.
CAM supports writing “don’t care” bits into words of the memory. The
don’t-care bit can be used as a mask for CAM comparisons; any bit set to
don’t-care has no effect on matches.
CAM can generate outputs in three different modes: single-match mode,
multiple-match mode, and fast multiple-match mode. In each mode, the
ESB outputs the matched data’s location as an encoded or unencoded
address. When encoded, the ESB outputs an encoded address of the data’s
location. For instance, if the data is located in address 12, the ESB output
is 12. When unencoded, each ESB port uses its 16 outputs to show the
location of the data over two clock cycles. In this case, if the data is located
in address 12, the 12th output line goes high. Figures 22 and 23 show the
encoded CAM outputs and unencoded CAM outputs, respectively.
Altera Corporation
43
Development
Tools
The Mercury on-chip CAM provides faster system performance than
traditional discrete CAM. Integrating CAM and logic into the Mercury
device eliminates off-chip and on-chip delays, improving system
performance.
Mercury Programmable Logic Device Family Data Sheet
Figure 23. Encoded CAM Address Outputs
CAM
addr[15..0] = 12
data[31..0] = 45
Data
Address
15
27
45
85
10
11
12
13
Encoded Output
match = 1
Figure 24. Unencoded CAM Address Outputs
q0
CAM
data[30..0] =45 (1)
select (2)
Data
Address
15
27
45
85
10
11
12
13
q12
Unencoded outputs.
q12 goes high to
signify a match.
q13
q14
q15
Notes to Figure 24:
(1)
(2)
For an unencoded output, the ESB only supports 31 input data bits. One input bit
is used by the select line to choose one of the two banks of 16 outputs.
If the select input is a 1, then CAM outputs odd words between 1 through 15. If
the select input is a 0, CAM outputs words even words between 0 through 14.
In single-match mode, it takes two clock cycles to write into CAM, but
only one clock cycle to read from CAM. In this mode, both encoded and
unencoded outputs are available without external logic. Single-match
mode is better suited for designs without duplicate data in the memory.
44
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
If the same data is written into multiple locations in the memory, a CAM
block can be used in multiple-match or fast multiple-match modes. The
ESB outputs the matched data’s locations as an encoded or unencoded
address. In multiple-match mode, it takes two clock cycles to write into a
CAM block. For reading, there are 16 outputs from each ESB at each clock
cycle. Therefore, it takes two clock cycles to represent the 32 words from
a single ESB port. In this mode, encoded and unencoded outputs are
available. To implement the encoded version, the Quartus II software
adds a priority encoder with LEs. Fast multiple-match is identical to the
multiple-match mode, however, it only takes one clock cycle to read from
a CAM block and generate valid outputs. To do this, the entire ESB is used
to represent 16 outputs. In fast multiple-match mode, the ESB can
implement a maximum CAM block size of 16 words.
A CAM block can be pre-loaded with data during configuration, or it can
be written during system operation. In most cases, two clock cycles are
required to write each word into CAM. When don’t-care bits are used, a
third clock cycle is required.
f
For more information on CAM, see Application Note 119 (Implementing
High-Speed Search Applications with APEX CAM).
13
ESBs provide flexible options for driving control signals. Different clocks
can be used for the ESB inputs and outputs. Registers can be inserted
independently on the data input, data output, read address, write
address, WREN, and RDEN signals on each port of the ESB. The fast global
signals and ESB local interconnect can drive the WREN and RDEN signals.
The fast global signals, dedicated clock pins, and ESB local interconnect
can drive the ESB clock signals. The ESB local interconnect is driven by the
ESB row interconnects which, in turn, are driven by all types of column
interconnects, including high-speed leap lines. Because the LEs drive the
column interconnect to the ESB local interconnect, the LEs can control the
WREN and RDEN signals and the ESB clock, clock enable, and asynchronous
clear signals. Figure 25 shows the ESB control signal generation logic.
Altera Corporation
45
Development
Tools
Driving into ESBs
Mercury Programmable Logic Device Family Data Sheet
Figure 25. ESB Control Signal Generation
Dedicated
Clocks
4
Fast Global
Signals
6
Local
Interconnect
Local
Interconnect
Local
Interconnect
Local
Interconnect
Local
Interconnect
Local
Interconnect
INCLOCK
OUTCLOCK
INCLOCKEN
OUTCLOCKEN
RDEN
INCLR
WREN
OUTCLR
The ESB can drive row interconnects within its own ESB row and can
directly drive all the column interconnects: column, priority column, and
leap lines.
Implementing Logic in ROM
In addition to implementing RAM functions, the ESB can implement logic
functions when it is programmed with a read-only pattern during
configuration, creating a large LUT. With LUTs, combinatorial functions
are implemented by looking up the results, rather than by computing
them. This implementation of combinatorial functions can be faster than
using algorithms implemented in general logic, a performance advantage
further enhanced by the fast access times of ESBs. The large capacity of
ESBs enables designers to implement complex functions in one logic level
without the routing delays associated with linked LEs or distributed RAM
blocks. Parameterized functions such as LPM functions can take
advantage of the ESB automatically. Further, the Quartus II software can
implement portions of a design with ESBs where appropriate.
46
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Programmable Speed/Power Control
Mercury device ESBs offer the Turbo BitTM option, a high-speed mode that
supports fast operation on an ESB-by-ESB basis. When high speed is not
required, the Turbo Bit option can be turned off to reduce power
dissipation by up to 50%. ESBs that run at low power incur a nominal
timing delay adder. An ESB that is not used will be powered down so it
does not consume DC current.
Designers can program each ESB in the Mercury device for either highspeed or low-power operation. As a result, speed-critical paths in the
design can run at high speed, while the remaining paths operate at
reduced power.
I/O Structure
For normal bidirectional operation, the input register can have its own
clock input separate from the OE and output registers. The OE and output
register share the same clock source. Each register can have its own clock
enable signal from local interconnect in the associated LAB, fast global
signals, or row global signals.
Altera Corporation
47
13
Development
Tools
The IOE in Mercury devices contains a bidirectional I/O buffer and three
registers for a complete embedded bidirectional IOE. The IOE contains
individual input, output, and output enable registers. The input register
can be used for external data requiring fast setup times. The output
register can be used for data requiring fast clock-to-output performance.
The output enable (OE) register can be used for fast clock-to-output enable
timing. The Quartus II software automatically duplicates a single OE
register that controls multiple output or bidirectional pins.
Mercury Programmable Logic Device Family Data Sheet
The Mercury IOE includes programmable delays that can be activated to
ensure zero hold times, minimum clock-to-output times, input IOE
register-to-core register transfers, or core-to-output IOE register transfers.
A path in which a pin directly drives a register may require the delay to
ensure zero hold time, whereas a path in which a pin drives a register
through combinatorial logic may not require the delay. Programmable
delays exist for decreasing input pin to core and IOE input register delays.
The Quartus II Compiler can program these delays automatically to
minimize setup time while providing a zero hold time. Delays are also
programmable for increasing the register to pin delays for output and/or
output enable registers. A programmable delay exists for increasing the
tZX delay to the output pin, which is required for ZBT interfaces. Table 10
shows the programmable delays for Mercury devices.
Table 10. Mercury Programmable Delay Chain
Programmable Delays
Quartus II Logic Option
Input pin to core delay (1)
Decrease input delay to internal cells
Input pin to input register delay
Decrease input delay to input register
Output propagation delay
Increase delay to output pin
Output enable register tCO delay
Increase delay to OE pin
Output tZX delay
Increase tZX delay to output pin
Note to Table 10:
(1)
This delay has four settings: off and three levels of delay.
The IOE registers in Mercury devices share the same source for clear or
preset. Use of the preset/clear is programmable for each individual IOE.
The register(s) can be programmed to power up high or low after
configuration is complete. If programmed to power up low, an
asynchronous clear can control the register(s). If programmed to power
up high, an asynchronous preset can control the register(s). This feature
prevents the inadvertent activation of another device’s active-low input
upon power-up. Figure 26 shows the IOE for Mercury devices.
48
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Figure 26. Mercury IOE
Associated LAB
Local Interconnect
Six Fast
Global
Signals
Two Row
Local Fast
Signals
Four
Dedicated
Clocks
Column and
Priority
Column
Interconnect
OE Register
D
Q
Output tZX
Delay
VCCIO
Optional
PCI Clamp
ENA
CLRN/PRN
OE Register
tCO Delay
VCCIO
Programmable
Pull-Up
Chip-Wide Reset
Output Register
D
Q
ENA
CLRN/PRN
Output
Propagation Delay
Drive Strength Control
Open-Drain Output
Slew Control
13
Development
Tools
Input Pin to
Core Delay (1)
Bus-Hold
Circuit
Input Pin to Input
Register Delay
Input Register
D
Q
ENA
CLRN/PRN
Priority Row Interconnect (for Associated LAB Row)
Row Interconnect (for Associated LAB Row)
Note to Figure 26:
(1)
This programmable delay has four settings: off and three levels of delay.
Double Data Rate I/O
Mercury device’s have three register IOEs to support the DDRIO feature,
which makes double data rate interfaces possible by clocking data on both
positive and negative clock edges. The IOE in Mercury devices supports
double data rate input and double data rate output modes.
Altera Corporation
49
Mercury Programmable Logic Device Family Data Sheet
In Mercury device IOEs, the OE register is a multi-purpose register
available as a second input or output register. When using the IOE for
double data rate inputs, the input register and OE register are
automatically configured as input registers to clock input double rate data
on alternating edges. An input latch is also used within the IOE for DDR
input acquisition. The latch holds the data that is present during the clock
high times, driving it to the OE register. This allows the OE register and
input register to clock both bits of data into LEs, synchronous to the same
clock edge (either rising or falling). Figure 27 shows an IOE configured for
DDR input.
Figure 27. IOE Configured for DDR Input
Associated LAB
Local Interconnect
Two Row
Local Fast
Signals
Six Fast
Global
Signals
Four
Dedicated
Clocks
VCCIO
Optional
PCI Clamp
OE Register
Column and
Priority
Column
D
VCCIO
Programmable
Pull-Up
Q
ENA
CLRN/PRN
Chip-Wide Reset
Input Pin to Input
Register Delay
Bus-Hold
Circuit
Input Register
D
Q
ENA
CLRN/PRN
PRN
D
Q
ENA
Latch
Priority Row (for Associated LAB Row)
Row (for Associated LAB Row)
50
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
When using the IOE for double data rate outputs, the output register and
OE register are automatically configured to clock two data paths from LEs
on rising clock edges. These register outputs are multiplexed by the clock
to drive the output pin at a ×2 rate. The output register clocks the first bit
out on the clock high time, while the OE register clocks the second bit out
on the clock low time. Figure 28 shows the IOE configured for DDR
output.
Figure 28. IOE Configured for DDR Output
Associated LAB
Local Interconnect
Two Row
Local Fast
Signals
Six Fast
Global
Signals
Four
Dedicated
Clocks
13
OE Register
Q
VCCIO
Optional
PCI Clamp
ENA
CLRN/PRN
VCCIO
Programmable
Pull-Up
Chip-Wide Reset
Output Register
D
Q
ENA
CLRN/PRN
0
1
Output
Propagation Delay
Drive Strength Control
Open-Drain Output
Slew Control
Bus-Hold
Circuit
Bidirectional DDR on an I/O pin is possible by using the IOE for DDR
output and using LEs to acquire the double data rate input. Bidirectional
DDR I/O pins support double data rate synchronous DRAM (DDR
SDRAM) at 166 MHz (334 Mbps), which transfer data on a double data
rate bidirectional bus. QDR SRAMs are also supported with DDR I/O
pins on separate read and write ports.
Altera Corporation
51
Development
Tools
D
Mercury Programmable Logic Device Family Data Sheet
Zero Bus Turnaround SRAM Interface Support
In addition to DDR SDRAM support, Mercury device I/O pins also
support interfacing with ZBT SRAM blocks at up to 200 MHz. ZBT SRAM
blocks are designed to eliminate dead bus cycles when turning a
bidirectional bus around between reads and writes, or writes and reads.
ZBT allows for 100% bus utilization because ZBT SRAM can read or write
on every clock cycle.
To avoid bus contention, the output tZX delay ensures that the clock-tolow-impedance time (tZX) is greater than the clock-to-high-impedance
time (tXZ). Time delay control of clocks to the OE/output and input
register, using a single general purpose PLL, enable the Mercury device to
meet ZBT tCO and tSU times.
Programmable Drive Strength
The output buffer for each Mercury device I/O pin has a programmable
drive strength control for certain I/O standards. The LVTTL standard has
several levels of drive strength that can be controlled by the user. SSTL-3
class I and II, SSTL-2 class I and II, HSTL class I and II, and 3.3-V GTL+
support a minimum or maximum setting. The minimum setting is the
lowest drive strength that guarantees the IOH/IOL of the standard. The
maximum setting provides higher drive strength that allows for faster
switching and is the default setting. Using settings below the maximum
provides signal slew-rate control to reduce system noise and signal
overshoot. Table 11 shows the possible settings for the I/O standards with
drive strength control.
52
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Table 11. Programmable Drive Strength
I/O Standard
LVTTL (3.3 V)
IOH/IOL Current Strength
Setting
4 mA
8 mA
12 mA
16 mA
24 mA (default)
LVTTL (2.5 V)
4 mA
8 mA
12 mA
16 mA (default)
LVTTL (1.8 V)
2 mA
4 mA (default)
Minimum
13
Maximum (default)
Open-Drain Output
Mercury devices provide an optional open-drain (equivalent to an opencollector) output for each I/O pin. This open-drain output enables the
device to provide system-level control signals (e.g., interrupt and write
enable signals) that can be asserted by any of several devices.
Slew-Rate Control
The output buffer for each Mercury device I/O pin has a programmable
output slew rate control that can be configured for low-noise or highspeed performance. A faster slew rate provides high-speed transitions for
high-performance systems. However, these fast transitions may introduce
noise transients into the system. A slow slew rate reduces system noise,
but adds a nominal delay to rising and falling edges. Each I/O pin has an
individual slew rate control, allowing the designer to specify the slew rate
on a pin-by-pin basis. The slew rate control affects both the rising and
falling edges.
Altera Corporation
53
Development
Tools
SSTL-3 class I and II
SSTL-2 class I and II
HSTL class I and II
GTL+ (3.3 V)
Mercury Programmable Logic Device Family Data Sheet
Bus Hold
Each Mercury device I/O pin provides an optional bus-hold feature.
When this feature is enabled for an I/O pin, the bus-hold circuitry weakly
holds the signal at its last driven state. By holding the last driven state of
the pin until the next input signal is present, the bus-hold feature
eliminates the need to add external pull-up or pull-down resistors to hold
a signal level when the bus is tri-stated. The bus-hold circuitry also pulls
undriven pins away from the input threshold voltage where noise can
cause unintended high-frequency switching. This feature can be selected
individually for each I/O pin. The bus-hold output will drive no higher
than VCCIO to prevent overdriving signals. If the bus-hold feature is
enabled, the programmable pull-up option cannot be used. The bus-hold
feature should also be disabled if open-drain outputs are used with the
GTL+ I/O standard.
The bus-hold circuitry weakly pulls the signal level to the last driven state
through a resistor with a nominal resistance (RBH) of approximately 8 kΩ.
Table 42 gives specific sustaining current that will be driven through this
resistor and overdrive current that will identify the next driven input
level. This information is provided for each VCCIO voltage level.
The bus-hold circuitry is active only after configuration. When going into
user mode, the bus-hold circuit captures the value on the pin present at
the end of configuration.
Programmable Pull-Up Resistor
Each Mercury device I/O pin provides an optional programmable pullup resistor during user mode. When this feature is enabled for an I/O pin,
the pull-up resistor (50 kΩ) weakly holds the output to the VCCIO level of
the bank that the output pin resides in.
I/O Row Bands
The I/O row bands are one of the advanced features of the Mercury
architecture. All IOEs are grouped in I/O row bands across the device.
The number of I/O row bands depends on the Mercury device size. The
I/O row bands are designed for flip-chip technology, allowing I/O pins
to be distributed across the entire chip, not only in the periphery. This
array driver technology allows higher I/O pin density (I/O pins per
device area) than peripheral I/O pins.
54
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Each row of I/O pins has an associated LAB row for driving to and from
the core of the Mercury device. For a given I/O band row, its associated
LAB row is located below it with the exception of the bottom I/O band
row. The bottom I/O band is located at the bottom periphery of the
device, hence its associated LAB row is located above it. Figure 29 shows
an example of an I/O band to associated LAB row interconnect in a
Mercury device.
There is a maximum of two IOEs associated with each LAB in the
associated LAB row. The local interconnect of the associated LAB drives
the IOEs. Since local interconnect is shared with the LAB neighbor, any
given LAB can directly drive up to four IOEs. The local interconnect
drives the data and OE signals when the IOE is used as an output or
bidirectional pin.
Figure 29. IOE Connection to Interconnects and Adjacent LAB
IOE Pair
IOE
INC
IOE
IOE Pair
INC
INB OUT INA
IOE
INC
INB OUT INA
All Column Interconnects
IOE Pair
IOE
INC
INB OUT INA
IOE
INC
INB OUT INA
IOE
INC
INB OUT INA
13
I/O Band
Row
Row Interconnect
Priority Row
Interconnect
LAB
LAB
LAB
To Next
LAB
To Local
Interconnect
IOE pairs drive unregistered
inputs to the associated
LAB's local interconnect.
The associated LAB and its
neighbor can drive a given IOE
pair through the local interconnect.
Associated LAB
to IOE Pair
Associated
LAB Row
Local Interconnect
Note to Figure 29:
(1)
INA: unregistered input; INB: registered/unregistered input; INC: registered/unregistered input or OE register
output in DDR mode.
The IOEs drive registered or combinatorial versions of input data into the
device. The unregistered input data can be driven to the local interconnect
(for fast input setup), row and priority row interconnect, and column and
priority column interconnects. The registered data can also be driven to
the same row and column resources. The OE register output can be fed
back through column and row interconnects to implement DDR I/O pins.
Altera Corporation
55
Development
Tools
INB OUT INA
Note (1)
Mercury Programmable Logic Device Family Data Sheet
Dedicated Fast Lines & I/O Pins
Mercury devices incorporate dedicated bidirectional pins for signals with
high internal fanout, such as PCI control signals. These pins are called
dedicated fast I/O pins (FAST1, FAST2, FAST3, FAST4, FAST5, and
FAST6) and can drive the six global fast lines throughout the device, ideal
for fast clock, clock enable, clear, preset, or high fanout logic signal
distribution. The dedicated fast I/O pins have the same IOE as a regular
I/O pin. The dedicated fast lines can also be driven by a LE local
interconnect to generate internal global signals.
In addition to the device global fast lines, each LAB row has two dedicated
fast lines local to the row. This is ideal for high fanout control signals for
a section of a design that may fit into a single LAB row. Each I/O band
(with the exception of the top I/O band) has two dedicated row-global
fast I/O pins to drive the row-global fast resources for the associated LAB.
The dedicated local fast I/O pins have the same IOE as a regular I/O pin.
The LE local interconnect can drive dedicated row-global fast lines to
generate internal global signals specific to a row. There are no pin
connections for buried LAB rows; LE local interconnects drive the rowglobal signals in those rows.
I/O Standard Support
Mercury device IOEs support the following I/O standards:
■
■
■
■
■
■
■
■
■
■
■
■
■
■
■
56
LVTTL
LVCMOS
1.8-V
2.5-V
3.3-V PCI
3.3-V PCI-X
3.3-V AGP (1×, 2×)
LVDS
LVPECL
3.3-V PCML
GTL+
HSTL class I and II
SSTL-3 class I and II
SSTL-2 class I and II
CTT
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Table 12 describes the I/O standards supported by Mercury devices.
Table 12. Mercury Supported I/O Standards
I/O Standard
Type
Input
Reference
Voltage (VREF)
(V)
Output
Supply
Voltage
(VCCIO) (V)
Board
Termination
Voltage
(VTT) (V)
Single-ended
N/A
3.3
N/A
LVCMOS
Single-ended
N/A
3.3
N/A
2.5 V
Single-ended
N/A
2.5
N/A
1.8 V
Single-ended
N/A
1.8
N/A
3.3-V PCI
Single-ended
N/A
3.3
N/A
3.3-V PCI-X
Single-ended
N/A
3.3
N/A
LVDS
Differential
N/A
3.3
N/A
LVPECL
Differential
N/A
3.3
N/A
3.3-V PCML
Differential
N/A
3.3
3.3
GTL+
Voltage referenced
1.0
N/A
1.5
HSTL class I and II
Voltage referenced
0.75
1.5
0.75
SSTL-2 class I and II
Voltage referenced
1.25
2.5
1.25
SSTL-3 class I and II
Voltage referenced
1.5
3.3
1.5
AGP
Voltage referenced
1.32
3.3
N/A
CTT
Voltage referenced
1.5
3.3
1.5
13
Development
Tools
LVTTL
Each regular I/O band row contains two I/O banks. The number of I/O
banks in a Mercury device depends on the number of I/O band rows. The
top I/O band contains four regular I/O banks specifically designed for
HSDI. The top I/O band banks and dedicated clock inputs support LVDS,
LVPECL, and 3.3-V PCML. 3.3-V PCML is an open-drain standard and
therefore requires external termination to 3.3 V. All other standards are
supported by all I/O banks. The top I/O banks 1, 2, 3, and 4 only support
non-HSDI I/O pins if the design does not use HSDI circuitry. If the design
uses any HSDI channel, banks 1, 2, 3, and 4 all do not support regular I/O
pins.
Additionally, the EP1M350 device includes the Flexible-LVDS feature,
providing support for up to 100 LVDS channels on all regular I/O banks.
Regular I/O banks in EP1M350 devices include dedicated LVDS input
and output buffers that do not require any external components except for
100-Ω termination resistors on receiver channels.
Altera Corporation
57
Mercury Programmable Logic Device Family Data Sheet
For the HSDI I/O band, half of the dedicated banks support LVDS, 3.3-V
PCML or LVPECL, and receiver inputs, while the other half support
LVDS, PCML or LVPECL, and transmitter outputs. A single device can
support 1.5-V, 1.8-V, 2.5-V, and 3.3-V interfaces; each bank can support a
VCCIO standard independently. Each bank can also use a separate VREF
level so that each bank can support any of the terminated standards (such
as SSTL-3) independently. A bank can support a single VREF level. Each
bank contains a fixed VREF pin for voltage referenced standards. This pin
can be used as a regular I/O if a VREF standard is not used. Table 13 shows
the number of I/O banks in each Mercury device.
Table 13. Number of I/O Banks per Device
Device
Regular I/O Banks
HSDI Band I/O Banks
EP1M120
8
4
EP1M350
12
4
Each bank can support multiple standards with the same VCCIO for output
pins. For EP1M120 devices, each bank can support one voltage-referenced
I/O standard, but can support multiple I/O standards with the same
VCCIO and VREF voltage levels. For example, when VCCIO is 3.3 V, a bank
can support LVTTL, LVCMOS, 3.3-V PCI, and SSTL-3 for inputs and
outputs. Figure 30 shows the I/O bank layout for an EP1M120 device. For
EP1M350 devices, each bank can support two voltage-reverenced I/O
standards; each I/O bank is split into two voltage-referenced sub-banks.
When using the two HSDI transmitter banks as regular I/O banks in a
non-HSDI mode, those two banks require the same VCCIO level. However,
each HSDI transmitter bank supports its own VREF level.
58
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Figure 30. I/O Bank Layout
ESB
I/O or HSDI
Banks
ESB
ESB
ESB
ESB
ESB
Input, Output, or HSDI Receiver (1)
Input, Output, or HSDI Receiver (1)
Input, Output, or HSDI Transmitter (1), (2)
Input, Output, or HSDI Transmitter (1), (2)
Regular I/O
Banks
I/O Bank
I/O Bank
I/O Bank
I/O Bank
I/O Bank
I/O Bank
13
ESB
ESB
Development
Tools
I/O Bank
I/O Bank
ESB
ESB
ESB
ESB
Notes to Figure 30:
(1)
(2)
If HSDI I/O channels are not used, the HSDI banks can be used as regular I/O banks.
When used as regular I/O banks, these banks must be set to the same VCCIO level, but can have separate VREF bank
settings.
f
For more information on I/O standards, see Application Note 117 (Using
Selectable I/O Standards in Altera Devices).
MultiVolt I/O Interface
The Mercury architecture supports the MultiVolt I/O interface feature,
which allows Mercury devices in all packages to interface with devices
with different supply voltages. The devices have one set of VCC pins for
internal operation and input buffers (VCCINT), and another set for I/O
output drivers (VCCIO).
Altera Corporation
59
Mercury Programmable Logic Device Family Data Sheet
The Mercury VCCINT pins must always be connected to a 1.8-V power
supply. With a 1.8-V VCCINT level, input pins are 1.8-V, 2.5-V and 3.3-V
tolerant. The VCCIO pins can be connected to either a 1.5-V, 1.8-V, 2.5-V or
3.3-V power supply, depending on the output requirements. When VCCIO
pins are connected to a 1.5-V power supply, the output levels are
compatible with HSTL systems. When VCCIO pins are connected to a
1.8-V power supply, the output levels are compatible with 1.8-V systems.
When VCCIO pins are connected to a 2.5-V power supply, the output
levels are compatible with 2.5-V systems. When the VCCIO pins are
connected to a 3.3-V power supply, the output high is 3.3 V and is
compatible with 3.3-V or 5.0-V systems.
Table 14 summarizes Mercury MultiVolt I/O support.
Table 14. Mercury MultiVolt I/O Support
VCCIO
(V)
Note (1)
Input Signal
Output Signal
1.5 V
1.8 V
2.5 V
3.3 V
1.5
v
v
v
v
1.8
v (2)
v
v
v
2.5
v (2)
v (2)
v
v
3.3
v (2)
v (2)
v (2)
v
5.0 V
1.5 V
1.8 V
2.5 V
3.3 V
5.0 V
v
v
v
v (3)
v (4)
v (6)
v
v (5)
Notes to Table 14:
(1)
(2)
(3)
(4)
(5)
(6)
The PCI clamping diode must be disabled to drive an input with voltages higher than V CCIO unless an external
resistor is used.
These input levels are only available if the input standard is set to any VREF-based input standard (SSTL-2, SSTL-3,
HSTL, GTL+, AGP 2×). The input buffers are powered from VCCINT when using V REF-based input standards.
LVTTL, PCI, PCI-X, AGP 1× input buffers are powered by VCCIO. Therefore, these standards cannot be driven with
input levels below the V CCIO setting except for when VCCIO = 3.3 V and the input voltage (VI) = 2.5 V.
When VCCIO = 1.8 V, the Mercury device can drive a 1.5-V device with 1.8-V tolerant inputs.
When VCCIO = 2.5 V, the Mercury device can drive a 1.8-V device with 2.5-V tolerant inputs.
When VCCIO = 3.3 V, the Mercury device can drive a 2.5-V device with 3.3-V tolerant inputs.
Designers can set Mercury devices to be 5.0-V tolerant by adding an external resistor and enabling the PCI clamping
diode.
Power Sequencing & Hot-Socketing
Because Mercury devices can be used in a mixed-voltage environment,
the devices are designed specifically to tolerate any possible power-up
sequence. Therefore, the VCCIO and VCCINT power supplies may be
powered in any order.
Signals can be driven into Mercury devices before and during power-up
without damaging the device. In addition, Mercury devices do not drive
out during power-up. Once operating conditions are reached and the
device is configured, Mercury devices operate as specified by the user.
60
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
General
Purpose PLL
Mercury devices have ClockLockTM, ClockBoostTM, and advanced
ClockShiftTM features, which use up to four general-purpose PLLs
(separate from the two HSDI PLLs) to province clock management and
clock-frequency synthesis. EP1M120 devices contain two general purpose
PLLs; EP1M350 devices contain four general purpose PLLs. These PLLs
allow designers to increase performance and provide clock-frequency
synthesis. The PLL reduces the clock delay within a device. This reduction
minimizes clock-to-output and setup times while maintaining zero hold
times. The PLLs, which provide programmable multiplication, allow the
designer to distribute a low-speed clock and multiply that clock ondevice. Mercury devices include a high-speed clock tree: unlike ASICs, the
user does not have to design and optimize the clock tree. The PLLs work
in conjunction with the Mercury device’s high-speed clock to provide
significant improvements in system performance and bandwidth.
Table 15 shows the general purpose PLL features for Mercury devices.
Figure 31 shows a Mercury PLL.
Table 15. Mercury General Purpose PLL Features
Device
ClockBoost
Feature (1)
Number of External
Clock Outputs
Number of
Feedback Inputs
Advanced
ClockShift
EP1M120
2
m/(n × k, p, q, v)
2
2
v
EP1M350
4
m/(n × k, p, q, v)
4
4
v
Note to Table 15:
(1)
n represents the prescale divider for the PLL input. k, p, q, and v represent the different post scale dividers for the
four possible PLL outputs. m, k, p, and q are integers that range from 1 to 160. n and v are integers that can range
from 1 to 16.
Figure 31. Mercury General-Purpose PLL
Phase
Comparator
inclock
Voltage-Controlled
Oscillator
n
Phase Shift
Circuitry
k
Time
Delay/Shift
clock0
p
Time
Delay/Shift
clock1
q
Time
Delay/Shift
clock2
Time
Delay/Shift
clock_ext
m
v
fbin
ClockLock
Altera Corporation
61
13
Development
Tools
Number of PLLs
Mercury Programmable Logic Device Family Data Sheet
The PLLs in Mercury devices are enabled through the Quartus II software.
External devices are not required to use these features.
Advanced ClockBoost Multiplication & Division
Each Mercury PLL includes circuitry that provides clock synthesis for up
to four outputs (three internal outputs and one external output) using
m/(n × output divider) scaling. When a PLL is locked, the locked output
clock aligns to the rising edge of the input clock. The closed loop equation
for Figure 31 gives an output frequency fclock0 = (m/(n × k))fIN,
fclock1 = (m/(n × p))fIN, fclock2 = (m/(n × q))fIN, and
fclock_ext = (m/(n × v))fIN or fclock1. These equations allow the
multiplication or division of clocks by a programmable number. The
Quartus II software automatically chooses the appropriate scaling factors
according to the frequency, multiplication, and division values entered.
A single PLL in a Mercury device allows for multiple user-defined
multiplication and division ratios that are not possible even with multiple
delay-locked loops (DLLs). For example, if a frequency scaling factor of
3.75 is needed for a given input clock, a multiplication factor of 15 and a
division factor of 4 can be entered. This advanced multiplication scaling
can be performed with a single PLL, making it unnecessary to cascade
PLL outputs.
External Clock Outputs
Mercury devices have four low-jitter external clocks available for external
clock sources. Other devices on the board can use these outputs as clock
sources.
There are three modes for external clock outputs. Multiplication is
allowed in all external clock output modes.
■
■
62
Zero Delay Buffer: The external clock output pin is phase aligned
with the clock input pin for zero delay. Programmable phase shift
and time delay shift are not allowed in this configuration.
Multiplication is allowed with the zero delay buffer mode. The
MegaWizard interface for altclklock should be used to verify
possible clock settings.
External Feedback: The external feedback input pin is phase aligned
with clock input pin. By aligning these clocks, you can actively
remove clock delay and skew between devices. Multiplication is
allowed with the external feedback mode. This mode has the same
restrictions as zero delay buffer mode.
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
■
Normal Mode: The external clock output pin will have phase delay
relative to the clock input pin. If an internal clock is used in this mode,
the IOE register clock will be phase aligned to the input clock pin.
Multiplication is allowed with the normal mode.
Advanced ClockShift Circuitry
General purpose PLLs in Mercury devices have advanced ClockShiftTM
circuitry that provides programmable phase shift and fine tune time delay
shift. For phase shifting, users can enter a phase shift (in degrees or time
units) that affects all PLL outputs. Phase shifts of 90, 180, and 270 can be
implemented exactly. Other values of phase shifting, or delay shifting in
time units, are allowed with a resolution range of 0.3 ns to 1.0 ns. This
resolution varies with frequency input and the user-entered
multiplication and division factors. The phase shift ability is only possible
on a multiplied or divided clock if the input and output frequency have
an integer multiple relationship (i.e., fIN/fOUT or fOUT/fIN must be an
integer).
Clock Enable Signal
Mercury PLLs have a CLKLK_ENA pin for enabling/disabling all of the
device PLLs. When the CLKLK_ENA pin is high, the PLL drives a clock to
all its output ports. When the CLKLK_ENA pin is low, the clock0,
clock1, clock2 and extclock ports are driven by GND and all of the
PLLs go out of lock. When the CLKLK_ENA pin goes high again, the PLL
must relock.
The individual enable port for each general purpose PLL is
programmable. If more than one general-purpose PLL is instantiated,
each one does not have to use the clock enable. To enable/disable the
device PLLs with the CLKLK_ENA pin, the inclocken port on the
altclklock instance must be connected to the CLKLK_ENA input pin.
Altera Corporation
63
13
Development
Tools
In addition to the phase shift feature that affects all outputs, there is an
advanced fine time delay shift control on each of the four PLL outputs.
Each PLL output can be shifted in 250-ps increments for a range of –2.0 ns
to +2.0 ns. This ability can be used in conjunction with the phase shifting
ability that affects all outputs. fIN/fOUT does not need to have an integer
relationship for the advanced fine time delay shift control.
Mercury Programmable Logic Device Family Data Sheet
Lock Signals
The Mercury device general purpose PLL circuits support individual
LOCK signals. The LOCK signal drives high when the PLL has locked onto
the input clock. Lock remains high as long as the input remains within
specification. It will go low if the input is out of specification. A LOCK pin
is optional for each PLL used in the Mercury devices; when not used, they
are I/O pins. This signal is not available internally; if it is used in the core,
it must be fed back in with an input pin.
SignalTap
Embedded
Logic Analyzer
Mercury devices include device enhancements to support the SignalTap
embedded logic analyzer. By including this circuitry, the Mercury device
provides the ability to monitor design operation over a period of time
through the IEEE Std. 1149.1 JTAG circuitry; a designer can analyze
internal logic at speed without bringing internal signals to the I/O pins.
This feature is particularly important for advanced packages such as
FineLine BGA packages, because it can be difficult to add a connection to
a pin during the debugging process after a board is designed and
manufactured.
IEEE Std.
1149.1 (JTAG)
Boundary-Scan
Support
All Mercury devices provide JTAG BST circuitry that complies with the
IEEE Std. 1149.1-1990 specification. JTAG boundary-scan testing can be
performed before or after configuration, but not during configuration.
Mercury devices can also use the JTAG port for configuration with the
Quartus II software or with hardware using either Jam Standard Test and
Programming Language (STAPL) Files (.jam) or Jam STAPL Byte-Code
Files (.jbc). Mercury devices also use the JTAG port to monitor the logic
operation of the device with the SignalTap embedded logic analyzer.
Mercury devices support the JTAG instructions shown in Table 16.
64
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Table 16. Mercury JTAG Instructions
JTAG Instruction
Description
SAMPLE/PRELOAD Allows a snapshot of signals at the device pins to be captured and examined during
normal device operation and permits an initial data pattern to be output at the device pins.
Also used by the SignalTap embedded logic analyzer.
Allows the external circuitry and board-level interconnections to be tested by forcing a test
pattern at the output pins and capturing test results at the input pins.
BYPASS
Places the 1-bit bypass register between the TDI and TDO pins, which allows the BST data
to pass synchronously through selected devices to adjacent devices during normal device
operation.
USERCODE
Selects the 32-bit USERCODE register and places it between the TDI and TDO pins,
allowing the USERCODE to be serially shifted out of TDO.
IDCODE
Selects the IDCODE register and places it between TDI and TDO, allowing the IDCODE
to be serially shifted out of TDO.
ICR Instructions
These instructions are used when configuring a Mercury device via the JTAG port with a
ByteBlasterMVTM download cable, or using a Jam STAPL or Jam Byte-Code file via an
embedded processor.
SignalTap
Instructions
These instructions monitor internal device operation with the SignalTap embedded logic
analyzer.
The Mercury device instruction register length is 10 bits. The Mercury
device USERCODE register length is 32 bits. Tables 17 and 18 show the
boundary-scan register length and device IDCODE information for
Mercury devices.
Table 17. Mercury Boundary-Scan Register Length
Device
Boundary-Scan Register Length (Bits)
EP1M120
1,125
EP1M350
1,695
Table 18. 32-Bit Mercury Device IDCODE
Device
IDCODE (32 Bits) (1)
Version
(4 Bits)
Part Number (16 Bits)
Manufacturer Identity
(11 Bits)
1 (1 Bit) (2)
EP1M120
0000
0011 0000 0000 0000
000 0110 1110
1
EP1M350
0000
0011 0000 0000 0001
000 0110 1110
1
Notes to Table 18:
(1)
(2)
The most significant bit (MSB) is on the left.
The IDCODE’s least significant bit (LSB) is always 1.
Altera Corporation
65
13
Development
Tools
EXTEST
Mercury Programmable Logic Device Family Data Sheet
Figure 32 shows the timing requirements for the JTAG signals.
Figure 32. Mercury JTAG Waveforms
TMS
TDI
t JCP
t JCH
t JCL
t JPSU
t JPH
TCK
tJPZX
t JPXZ
t JPCO
TDO
tJSH
tJSSU
Signal
to Be
Captured
Signal
to Be
Driven
tJSCO
tJSZX
tJSXZ
Table 19 shows the JTAG timing parameters and values for Mercury
devices.
Table 19. Mercury JTAG Timing Parameters & Values
Symbol
66
Parameter
Min
Max
Unit
tJCP
TCK clock period
tJCH
TCK clock high time
100
ns
50
ns
tJCL
TCK clock low time
50
ns
tJPSU
JTAG port setup time
20
ns
tJPH
JTAG port hold time
45
ns
tJPCO
JTAG port clock to output
25
ns
tJPZX
JTAG port high impedance to valid output
25
ns
tJPXZ
JTAG port valid output to high impedance
25
tJSSU
Capture register setup time
20
ns
tJSH
Capture register hold time
45
ns
tJSCO
Update register clock to output
35
ns
tJSZX
Update register high impedance to valid output
35
ns
tJSXZ
Update register valid output to high impedance
35
ns
ns
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
f
For more information, see the following documents:
■
■
Generic Testing
Application Note 39 (IEEE Std. 1149.1 (JTAG) Boundary-Scan Testing in
Altera Devices)
Jam Programming & Test Language Specification
Each Mercury device is functionally tested. Complete testing of each
configurable static random access memory (SRAM) bit and all logic
functionality ensures 100% yield. AC test measurements for Mercury
devices are made under conditions equivalent to those shown in
Figure 33. Multiple test patterns can be used to configure devices during
all stages of the production flow.
13
Development
Tools
Altera Corporation
67
Mercury Programmable Logic Device Family Data Sheet
Figure 33. Mercury AC Test Conditions
Power supply transients can affect AC
measurements. Simultaneous transitions
of multiple outputs should be avoided for
accurate measurement. Threshold tests
must not be performed under AC
conditions. Large-amplitude, fast-groundcurrent transients normally occur as the
device outputs discharge the load
capacitances. When these transients flow
through the parasitic inductance between
the device ground pin and the test system
ground, significant reductions in
observable noise immunity can result.
Operating
Conditions
VCCINT
To Test
System
Device input
rise and fall
times < 3 ns
C1 (includes
jig capacitance)
Table 20 through 43 provide information on absolute maximum ratings,
recommended operating conditions, DC operating conditions, and
capacitance for 1.8-V Mercury devices.
Table 20. Mercury Device Absolute Maximum Ratings
Symbol
Device
Output
Parameter
Supply voltage
Note (1)
Conditions
With respect to ground (2)
VCCIO
Minimum
Maximum
Unit
–0.5
2.5
V
–0.5
4.6
V
VI
DC input voltage
–0.5
4.6
V
IOUT
DC output current, per pin
–34
34
mA
TSTG
Storage temperature
No bias
–65
150
°C
TAMB
Ambient temperature
Under bias
–65
135
°C
TJ
Junction temperature
BGA packages under bias
135
°C
68
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Table 21. Mercury Device Recommended Operating Conditions
Symbol
Parameter
Conditions
Minimum
Maximum
Unit
1.71
1.89
V
3.00 (3.135)
3.60 (3.465)
V
Supply voltage for output buffers, (3)
2.5-V operation
2.375
2.625
V
Supply voltage for output buffers, (3)
1.8-V operation
1.71
1.89
V
Supply voltage for output buffers, (3)
1.5-V operation
1.4
1.6
V
–0.5
4.1
V
0
VCCIO
V
0
85
°C
–40
VCCINT
Supply voltage for internal logic
and input buffers
(3)
VCCIO
Supply voltage for output buffers, (3), (4)
3.3-V operation
Input voltage
Output voltage
(2), (5)
TJ
Operating temperature
100
°C
tR
Input rise time
40
ns
tF
Input fall time
40
ns
For commercial
use
For industrial use
Table 22. Mercury Device DC Operating Conditions
Symbol
Parameter
13
Development
Tools
VI
VO
Note (6), (7)
Conditions
Minimum
Typical
Maximum
Unit
II
Input pin leakage
current
VI = VCCIOmax to 0 V (5)
–10
10
µA
IOZ
Tri-stated I/O pin
leakage current
VO = VCCIOmax to 0 V (5)
–10
10
µA
ICC0
VCC supply current
(standby) for
EP1M120 devices
For commercial use (8)
30
mA
For Industrial use (8)
40
mA
VCC supply current
(standby) for
EP1M350 devices
For commercial use (8)
50
mA
For Industrial use (8)
60
mA
RCONF
Value of I/O pin pull- VCCIO = 3.0 V (9)
up resistor before
VCCIO = 2.375 V (9)
and during
VCCIO = 1.71 V (9)
configuration
Altera Corporation
20
50
kΩ
30
80
kΩ
60
150
kΩ
69
Mercury Programmable Logic Device Family Data Sheet
Table 23. LVTTL Specifications
Symbol
Note (10)
Parameter
Conditions
Minimum
Maximum
Units
V
VCCIO
Output supply voltage
3.0
3.6
VI H
High-level input voltage
1.7
4.1
V
VIL
Low-level input voltage
–0.5
0.7
V
II
Input pin leakage current
VIN = 0 V or VCCIO
–10
10
µA
VOH
High-level output voltage
IOH = –4 mA
2.4
VOL
Low-level output voltage
IOL = 4 mA
V
0.45
V
Minimum
Maximum
Units
3.0
3.6
V
Table 24. LVCMOS Specifications
Symbol
Parameter
Conditions
VCCIO
Power supply voltage range
VIH
High-level input voltage
1.7
4.1
V
VIL
Low-level input voltage
–0.5
0.7
V
–10
10
µA
II
Input pin leakage current
VIN = 0 V or VCCIO
VOH
High-level output voltage
VCCIO = 3.0,
IOH = –0.1 mA
VOL
Low-level output voltage
VCCIO = 3.0,
IOL = 0.1 mA
Table 25. 2.5-V I/O Specifications
Symbol
Parameter
VCCIO
Output supply voltage
VIH
High-level input voltage
VIL
Low-level input voltage
II
Input pin leakage current
VOH
High-level output voltage
VOL
70
Low-level output voltage
VCCIO – 0.2
V
0.2
V
Minimum
Maximum
Units
2.375
2.625
V
1.7
4.1
V
Note (10)
Conditions
–0.5
0.7
V
VIN = 0 V or VCCIO
10
10
µA
IOH = –0.1 mA
2.1
V
IOH = –1 mA
2.0
V
IOH = –2 mA
1.7
V
IOL = 0.1 mA
0.2
V
IOH = 1 mA
0.4
V
IOH = 2 mA
0.7
V
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Table 26. 1.8-V I/O Specifications
Symbol
Note (10)
Parameter
VCCIO
Output supply voltage
VI H
High-level input voltage
Conditions
VIL
Low-level input voltage
II
Input pin leakage current
VIN = 0 V or VCCIO
VOH
High-level output voltage
IOH = –2 mA
VOL
Low-level output voltage
IOL = 2 mA
Minimum
Maximum
Units
1.71
1.89
V
0.65 × VCCIO
4.1
V
–0.5
0.35 × VCCIO
V
–10
10
µA
VCCIO – 0.45
V
0.45
V
Figures 34 and 35 show receiver input and transmitter output waveforms,
respectively, for all differential I/O standards (LVPECL, 3.3-V PCML,
LVDS, and HyperTransport technology).
Figure 34. Receiver Input Waveforms for Differential I/O Standards
Single-Ended Waveform
Positive Channel (p) = VIH
13
Negative Channel (n) = VIL
Development
Tools
±VID
VCM
Ground
Differential Waveform
+VID
p−n=0V
VID (Peak-to-Peak)
Altera Corporation
− VID
71
Mercury Programmable Logic Device Family Data Sheet
Figure 35. Transmitter Output Waveforms for Differential I/O Standards
Single-Ended Waveform
Positive Channel (p) = VOH
±VOD
Negative Channel (n) = VOL
VCM
Ground
Differential Waveform
+VOD
p−n=0V
VOD (Peak-to-Peak)
− VOD
Table 27. 3.3-V LVDS I/O Specifications
Symbol
Parameter
Conditions
VCCIO
I/O supply voltage
VOD
Differential output voltage
RL = 100 Ω
∆ VOD
Change in VOD between
high and low
RL = 100 Ω
Minimum
Typical
Maximum
3.135
3.3
3.465
V
250
510
600
mV
50
mV
VOS
Output offset voltage
RL = 100 Ω
∆ VOS
Change in VOS between
high and low
RL = 100 Ω
VTH
Differential input threshold VCM = 1.2 V
VIN
Receiver input voltage
range
RL
Receiver differential input
resistor (external to
Mercury devices)
90
72
1.125
1.25
Units
1.375
V
50
mV
–100
100
mV
0.0
2.4
V
110
Ω
100
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Table 28. 3.3-V PCML Specifications
Symbol
Parameter
VCCIO
I/O supply voltage
VIL
Low-level input voltage
VIH
High-level input voltage
VOL
Low-level output voltage
Conditions
Minimum
Typical
3.135
3.3
Maximum
Units
3.465
V
VCCIO –
0.4
V
VCCIO
V
VCCIO –
0.4
VCCIO
V
VOH
High-level output voltage
VT
Output termination voltage
V
VID
Differential input voltage
400
VOD
Differential output voltage
400
tR
Rise time (20 to 80%)
tF
Fall time (20 to 80%)
200
ps
tDSKEW
Differential skew
25
ps
R1 (11)
Output load
90
100
110
Ω
R2 (11)
Receiver differential input
resistor
45
50
55
Ω
Minimum
Typical
Maximum
Units
3.135
3.3
VCCIO
700
V
800
mV
800
mV
200
ps
13
Symbol
Parameter
VCCIO
I/O supply voltage
VIL
Low-level input voltage
VIH
VOL
Conditions
3.465
V
0
2,000
mV
High-level input voltage
400
2,470
mV
Low-level output voltage
1,400
1,650
mV
VOH
High-level output voltage
2,275
2,470
mV
VID
Differential input voltage
400
600
1,200
mV
VOD
Differential output voltage
525
1,050
1,200
mV
tR
Rise time (20 to 80%)
85
325
ps
tF
Fall time (20 to 80%)
85
325
ps
tDSKEW
Differential skew
25
ps
RL
Receiver differential input
resistor
Altera Corporation
100
Ω
73
Development
Tools
Table 29. LVPECL Specifications
Mercury Programmable Logic Device Family Data Sheet
Table 30. 3.3-V PCI Specifications
Symbol
Parameter
Conditions
Minimum
Typical
3.0
3.3
Maximum
Units
VCCIO
I/O supply voltage
3.6
V
VIH
High-level input voltage
0.5 ×
VCCIO
VCCIO +
0.5
V
VIL
Low-level input voltage
–0.5
0.3 ×
VCCIO
V
II
Input pin leakage current
0 < VIN < VCCIO
–10
10
µA
VOH
High-level output voltage
IOUT = –500 µA
0.9 ×
VCCIO
VOL
Low-level output voltage
IOUT = 1,500 µA
V
0.1 ×
VCCIO
V
Table 31. PCI-X Specifications
Symbol
Parameter
Maximum
Units
3.0
3.6
V
High-level input voltage
0.5 ×
VCCIO
VCCIO +
0.5
V
VIL
Low-level input voltage
–0.5
0.35 ×
VCCIO
V
VIPU
Input pull-up voltage
IIL
Input leakage current
0 < VIN < VCCIO
–10
VOH
High-level output voltage
IOUT = –500 µA
0.9 ×
VCCIO
VOL
Low-level output voltage
IOUT = 1,500 µA
LPIN
Pin inductance
VCCIO
I/O supply voltage
VIH
Table 32. GTL+ I/O Specifications
Symbol
Parameter
Conditions
Minimum
Typical
0.7 ×
VCCIO
V
10
µA
V
0.1 ×
VCCIO
V
15
nH
Maximum
Units
Note (10)
Conditions
Minimum
Typical
VTT
Termination voltage
1.35
1.5
1.65
V
VREF
Reference voltage
0.88
1.0
1.12
V
VREF – 0.1
V
0.65
V
VIH
High-level input voltage
VIL
Low-level input voltage
VOL
Low-level output voltage
74
VREF + 0.1
IOL = 34 mA
V
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Table 33. SSTL-2 Class I Specifications
Symbol
Parameter
VCCIO
I/O supply voltage
VTT
Termination voltage
VREF
Reference voltage
VIH
High-level input voltage
Note (10)
Conditions
VIL
Low-level input voltage
VOH
High-level output voltage
IOH = –7.6 mA
VOL
Low-level output voltage
IOL = 7.6 mA
Table 34. SSTL-2 Class II Specifications
Symbol
Parameter
VCCIO
I/O supply voltage
Conditions
Termination voltage
Reference voltage
VIH
High-level input voltage
VIL
Low-level input voltage
VOH
High-level output voltage
IOH = –15.2 mA
VOL
Low-level output voltage
IOL = 15.2 mA
Parameter
VCCIO
I/O supply voltage
VTT
Termination voltage
VREF
Reference voltage
VIH
High-level input voltage
Units
2.375
2.5
2.625
V
VREF – 0.04
VREF
VREF + 0.04
V
1.15
1.25
1.35
V
VREF + 0.18
3.0
V
–0.3
VREF – 0.18
V
VTT – 0.57
V
VTT + 0.57
V
Minimum
Typical
Maximum
Units
2.3
2.5
2.7
V
VREF – 0.04
VREF
VREF + 0.04
V
1.15
1.25
1.35
V
VREF + 0.18
VCCIO + 0.3
V
–0.3
VREF – 0.18
V
VTT + 0.76
13
V
VTT – 0.76
V
Maximum
Units
Note (10)
Conditions
VIL
Low-level input voltage
VOH
High-level output voltage
IOH = –8 mA
VOL
Low-level output voltage
IOL = 8 mA
Altera Corporation
Maximum
Development
Tools
VTT
Symbol
Typical
Note (10)
VREF
Table 35. SSTL-3 Class I Specifications
Minimum
Minimum
Typical
3.0
3.3
3.6
V
VREF – 0.05
VREF
VREF + 0.05
V
1.3
1.5
1.7
V
VREF + 0.2
VCCIO + 0.3
V
–0.3
VREF – 0.2
V
VTT – 0.6
V
VTT + 0.6
V
75
Mercury Programmable Logic Device Family Data Sheet
Table 36. SSTL-3 Class II Specifications
Symbol
Parameter
Note (10)
Conditions
VCCIO
I/O supply voltage
VTT
Termination voltage
VREF
Reference voltage
VIH
High-level input voltage
VIL
Low-level input voltage
VOH
High-level output voltage
IOH = –16 mA
VOL
Low-level output voltage
IOL = 16 mA
Minimum
Typical
Maximum
Units
3.0
3.3
3.6
V
VREF – 0.05
VREF
VREF + 0.05
V
1.3
1.5
1.7
V
VREF + 0.2
VCCIO + 0.3
V
–0.3
VREF – 0.2
V
VTT – 0.8
V
Maximum
Units
VT T + 0.8
V
Table 37. 3.3-V AGP -2X Specifications
Symbol
Parameter
Conditions
Minimum
Typical
3.3
VCCIO
I/O supply voltage
3.15
3.45
V
VREF
Reference voltage
0.39 × VCCIO
0.41 × VCCIO
V
VIH
High-level input voltage
(12)
0.5 × VCCIO
VCCIO + 0.5
V
VIL
Low-level input voltage
(12)
0.3 × VCCIO
V
VOH
High-level output voltage
IOUT = –20 µA
VOL
Low-level output voltage
IOUT = 20 µA
II
Input pin leakage current
0 < VI N < V CCIO
0.9 × VCCIO
3.6
V
0.1 × VCCIO
V
±10
µA
Maximum
Units
Table 38. 3.3-V AGP -1X Specifications
Symbol
Parameter
Conditions
VCCIO
I/O supply voltage
VIH
High-level input voltage
(12)
VIL
Low-level input voltage
(12)
VOH
High-level output voltage
IOUT = –20 µA
VOL
Low-level output voltage
IOUT = 20 µA
II
Input pin leakage current
0 < VI N < V CCIO
76
Minimum
Typical
3.15
3.3
0.5 × VCCIO
0.9 × VCCIO
3.45
V
VCCIO + 0.5
V
0.3 × VCCIO
V
3.6
V
0.1 × VCCIO
V
±10
µA
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Table 39. 1.5-V HSTL Class I Specifications
Symbol
Parameter
Note (10)
Conditions
Minimum
Typical
Maximum
Units
VCCIO
I/O supply voltage
1.4
1.5
1.6
V
VREF
Input reference voltage
0.68
0.75
0.9
V
0.7
0.75
0.8
VTT
Termination voltage
VIH (DC)
DC high-level input voltage
VREF + 0.1
VIL (DC)
DC low-level input voltage
–0.3
VIH (AC)
AC high-level input voltage
VREF + 0.2
VIL (AC)
AC low-level input voltage
VOH
High-level output voltage
IOH = 8 mA
VOL
Low-level output voltage
IOH = –8 mA
Table 40. 1.5-V HSTL Class II Specifications
Symbol
Parameter
V
V
VREF – 0.1
V
VREF – 0.2
V
V
VCCIO – 0.4
V
0.4
V
Maximum
Units
Note (10)
Conditions
Minimum
Typical
I/O supply voltage
1.4
1.5
1.6
V
Input reference voltage
0.68
0.75
0.9
V
0.7
0.75
0.8
VTT
Termination voltage
VIH (DC)
DC high-level input voltage
VREF + 0.1
VIL (DC)
DC low-level input voltage
–0.3
VIH (AC)
AC high-level input voltage
VREF + 0.2
VIL (AC)
AC low-level input voltage
VOH
High-level output voltage
IOH = 16 mA
VOL
Low-level output voltage
IOH = –16 mA
13
Development
Tools
VCCIO
VREF
V
V
VREF – 0.1
V
VREF – 0.2
V
V
VCCIO – 0.4
V
0.4
V
Maximum
Units
Table 41. CTT I/O Specifications
Symbol
Parameter
Conditions
Minimum
Typical
VCCIO
I/O supply voltage
3.0
3.3
3.6
V
VTT/VREF
Termination and input
reference voltage
1.35
1.5
1.65
V
VREF – 0.2
V
±10
µA
VREF – 0.4
V
±10
µA
VIH
High-level input voltage
VIL
Low-level input voltage
II
Input pin leakage current
0 < VIN < V CCIO
VOH
High-level output voltage
IOH = –8 mA
VOL
Low-level output voltage
IOL = 8 mA
IO
Output leakage current
(when output is high Z)
GND ð VOUT ð
VCCIO
Altera Corporation
VREF + 0.2
V
VREF + 0.4
V
77
Mercury Programmable Logic Device Family Data Sheet
Table 42. Bus Hold Parameters
Parameter
Conditions
VCCIO Level
1.8 V
Units
2.5 V
3.3 V
Minimum Maximum Minimum Maximum Minimum Maximum
VIN > VIL
(maximum)
30
50
70
µA
High sustaining VIN < VIH
(minimum)
current
–30
–50
–70
µA
Low sustaining
current
Low overdrive
current
0 V < VIN <
VCCIO
200
300
500
µA
High overdrive
current
0 V < VIN <
VCCIO
–200
–300
–500
µA
Table 43. Mercury Device Capacitance
Symbol
Note (13)
Parameter
Minimum
Typical
Maximum
Unit
CIO
I/O pin capacitance
13.5
pF
CCLK
Input capacitance on CLK[4..1] pins
16.9
pF
CRXHSDI
Input capacitance on HSDI receiver pins
8.0
pF
CTXHSDI
Input capacitance on HSDI transmitter pins
18.0
pF
CCLKHSDI
Input capacitance on HSDI clock pins
7.5
pF
CFLEXLVDSRX
Input capacitance on flexible LVDS receiver
pins
13.4
pF
CFLEXLVDSTX
Input capacitance on flexible LVDS
transmitter pins
13.4
pF
78
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Notes to Tables 20 – 43.:
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
See the Operating Requirements for Altera Devices Data Sheet.
Minimum DC input is –0.5 V. During transitions, the inputs may undershoot to –0.5 V or overshoot to 4.1 V for input
currents less than 100 mA and periods shorter than 20 ns.
Maximum VCC rise time is 100 ms, and VCC must rise monotonically.
VCCIO maximum and minimum conditions for LVPECL, LVDS, RapidIO, and 3.3-V PCML are shown in
parentheses.
All pins, including dedicated inputs, clock, I/O, and JTAG pins, may be driven before VCCINT and VCCIO are
powered.
Typical values are for TA = 25° C, VCCINT = 1.8 V, and VCCIO = 1.8 V, 2.5 V, and 3.3 V.
These values are specified under the Mercury Device Recommended Operating Conditions shown in Table 3 on
page 3.
Input pins are grounded. In the test design, internal logic does not toggle. The test design does not use PLL or HSDI
circuitry. All ESBs are in power-down mode.
Pin pull-up resistance values will lower if an external source drives the pin higher than VCCIO.
Drive strength is programmable according to values in Table 11 on page 53.
For more information on termination, see AN 134: Using Programmable I/O Standards in Mercury Devices or
AN 159: Using HSDI in Source-Synchronous Mode in Mercury Devices.
VREF specifies the center point of the switching range.
Capacitance is sample-tested only. Capacitance is measured using time-domain reflections (TDR). Measurement
accuracy is within ±5%.
Timing Model
Figure 36 shows the timing model for bidirectional IOE pin timing. All
registers are within the IOE.
Figure 36. Synchronous Bidirectional Pin External Timing Model
OE Register
Dedicated
Clock
D
PRN
Q
tXZBIDIR
tZXBIDIR
CLRN
tOUTCOBIDIR
Output Register
D
PRN
Q
CLRN
Bidirectional
Pin
tINSUBIDIR
tINHBIDIR
Input Register
PRN
D
Q
CLRN
Altera Corporation
79
13
Development
Tools
The high-performance multi-level FastTrack Interconnect routing
resources ensure predictable performance, accurate simulation, and
accurate timing analysis. The predictable performance of Mercury devices
offer an advantage over FPGAs, which use a segmented connection
scheme and therefore have unpredictable performance.
Mercury Programmable Logic Device Family Data Sheet
Tables 44 and 45 describe the Mercury device’s external timing
parameters.
Table 44. Mercury External Timing Parameters
Symbol
Notes (1), (2)
Parameter
Conditions
tINSU
Setup time with global clock at IOE register
tINH
Hold time with global clock at IOE register
tOUTCO
Clock-to-output delay with global clock at IOE register
tINSUPLL
Setup time with PLL clock at IOE input register
tINHPLL
Hold time with PLL clock at IOE input register
tOUTCOPLL
Clock-to-output delay with PLL clock at IOE output register
Table 45. Mercury External Bidirectional Timing Parameters
Symbol
C1 = 35 pF
C1 = 35 pF
Notes (1), (2)
Parameter
Conditions
tINSUBIDIR
Setup time for bidirectional pins with gobal clock at IOE input register
tINHBIDIR
Hold time for bidirectional pins with global clock at IOE input register
tOUTCOBIDIR
Clock-to-output delay for bidirectional pins with global clock at IOE
output register
tXZBIDIR
Synchronous IOE output enable register to output buffer disable delay C1 = 35 pF
tZXBIDIR
Synchronous IOE output enable register output buffer enable delay
tINSUBIDIRPLL
Setup time for bidirectional pins with PLL clock at IOE input register
tINHBIDIRPLL
Hold time for bidirectional pins with PLL clock at IOE input register
tOUTCOBIDIRPLL
Clock-to-output delay for bidirectional pins with PLL clock at IOE output C1 = 35 pF
register
tXZBIDIRPLL
Synchronous IOE output enable register to output buffer disable delay C1 = 35 pF
with PLL
tZXBIDIRPLL
Synchronous IOE output enable register output buffer enable delay
with PLL
C1 = 35 pF
C1 = 35 pF
C1 = 35 pF
Notes to Tables 44 and 45:
(1)
(2)
80
These timing parameters are sample-tested only.
All timing parameters are either to and/or from pins, including global clock pins.
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Tables 46 through 51 show external timing parameters for Mercury
devices.
Table 46. EP1M120 External Timing Parameters
Symbol
-5 Speed Grade
Min
tINSU
0.67
tINH
0.00
tOUTCO
2.00
tINSUPLL
0.59
tINHPLL
0.00
tOUTCOPLL
0.50
Max
Note (1)
-6 Speed Grade
Min
Max
2.00
2.00
0.64
0.50
0.00
2.08
ns
2.15
ns
Note (1)
-6 Speed Grade
Min
0.50
ns
ns
Max
-7 Speed Grade
Min
Unit
Max
tINSUBIDIR
0.67
0.70
0.73
ns
tINHBIDIR
0.00
0.00
0.00
ns
tOUTCOBIDIR
2.00
3.49
ns
tXZBIDIR
3.52
3.30
2.00
3.53
3.32
2.00
3.74
ns
tZXBIDIR (2)
3.52
3.53
3.74
ns
tZXBIDIR (3)
3.72
3.73
3.99
ns
tINSUBIDIRPLL
0.59
0.64
0.62
ns
tINHBIDIRPLL
0.00
0.00
0.00
ns
tOUTCOBIDIRPLL
0.50
2.15
ns
tXZBIDIRPLL
2.29
2.29
2.39
ns
tZXBIDIRPLL (2)
2.29
2.29
2.39
ns
tZXBIDIRPLL (3)
2.49
2.49
2.64
ns
Altera Corporation
2.08
0.50
2.08
0.50
13
Development
Tools
Max
ns
3.49
0.62
0.00
2.08
ns
0.00
3.32
Unit
Max
0.73
0.00
3.30
-5 Speed Grade
Min
Min
0.70
Table 47. EP1M120 External Bidirectional Timing Parameters
Symbol
-7 Speed Grade
81
Mercury Programmable Logic Device Family Data Sheet
Table 48. EP1M120 External Timing Parameters
Symbol
Note (1)
-7A Speed Grade
Min
-8A Speed Grade
Max
Min
Unit
Max
tINSU
0.74
0.79
ns
tINH
0.00
0.00
ns
tOUTCO
2.00
tINSUPLL
0.62
tINHPLL
0.00
tOUTCOPLL
0.50
3.50
2.00
Min
ns
0.50
-7A Speed Grade
2.43
Min
tINSUBIDIR
0.74
0.79
tINHBIDIR
0.00
0.00
tOUTCOBIDIR
2.00
3.50
ns
Note (1)
-8A Speed Grade
Max
ns
ns
0.00
2.15
Table 49. EP1M120 External Bidirectional Timing Parameters
Symbol
4.10
0.75
Unit
Max
ns
ns
2.00
4.10
ns
tXZBIDIR
3.75
4.30
ns
tZXBIDIR (2)
3.75
4.30
ns
tZXBIDIR (3)
4.00
4.58
ns
tINSUBIDIRPLL
0.62
0.75
tINHBIDIRPLL
0.00
0.00
tOUTCOBIDIRPLL
0.50
2.15
ns
ns
0.50
2.43
ns
ns
tXZBIDIRPLL
2.39
2.67
tZXBIDIRPLL (2)
2.39
2.67
ns
tZXBIDIRPLL (3)
2.64
2.95
ns
Table 50. EP1M350 External Timing Parameters
Symbol
-5 Speed Grade
Min
Max
Note (1)
-6 Speed Grade
Min
Max
-7 Speed Grade
Min
Unit
Max
tINSU
0.60
0.57
0.71
tINH
0.00
0.00
0.00
tOUTCO
2.00
tINSUPLL
0.69
0.70
0.82
ns
tINHPLL
0.00
0.00
0.00
ns
tOUTCOPLL
0.50
82
3.95
2.23
2.00
0.50
3.97
2.23
2.00
0.50
ns
ns
4.75
2.69
ns
ns
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Table 51. EP1M350 External Bidirectional Timing Parameters
Symbol
-5 Speed Grade
Min
Max
Note (1)
-6 Speed Grade
Min
Max
-7 Speed Grade
Min
tINSUBIDIR
0.60
0.57
0.71
tINHBIDIR
0.00
0.00
0.00
tOUTCOBIDIR
2.00
3.95
2.00
3.97
2.00
Unit
Max
ns
ns
4.75
ns
tXZBIDIR
3.90
3.93
4.70
ns
tZXBIDIR (2)
3.90
3.93
4.70
ns
tZXBIDIR (3)
4.10
4.13
4.94
ns
tINSUBIDIRPLL
0.69
0.70
0.82
ns
tINHBIDIRPLL
0.00
0.00
0.00
ns
tOUTCOBIDIRPLL
0.50
2.69
ns
tXZBIDIRPLL
2.19
2.23
0.50
2.18
2.23
0.50
2.63
ns
tZXBIDIRPLL (2)
2.19
2.18
2.63
ns
tZXBIDIRPLL (3)
2.39
2.38
2.87
ns
13
Notes to Tables 46 – 51:
(2)
(3)
Timing will vary by I/O pin placement. Therefore, use the Quartus II software to determine exact I/O timing for
each pin.
This parameter is measured with the Increase tZX Delay to Output Pin option set to Off.
This parameter is measured with the Increase tZX Delay to Output Pin option set to On.
Power
Consumption
Detailed power consumption information for Mercury devices will be
released when available.
Configuration &
Operation
The Mercury architecture supports several configuration schemes. This
section summarizes the device operating modes and available device
configuration schemes.
Operating Modes
The Mercury architecture uses SRAM configuration elements that require
configuration data to be loaded each time the circuit powers up. The
process of physically loading the SRAM data into the device is called
configuration. During initialization, which occurs immediately after
configuration, the device resets registers, enables I/O pins, and begins to
operate as a logic device. The I/O pins are tri-stated during power-up and
before and during configuration. Together, the configuration and
initialization processes are called command mode; normal device
operation is called user mode.
Altera Corporation
83
Development
Tools
(1)
Mercury Programmable Logic Device Family Data Sheet
Before and during device configuration, all I/O pins are pulled to VCCIO
by a built-in weak pull-up resistor.
SRAM configuration elements allow Mercury devices to be reconfigured
in-circuit by loading new configuration data into the device. Real-time
reconfiguration is performed by forcing the device into command mode
with a device pin, loading different configuration data, reinitializing the
device, and resuming user-mode operation. In-field upgrades can be
performed by distributing new configuration files.
Configuration Schemes
The configuration data for a Mercury device can be loaded with one of five
configuration schemes (see Table 52), chosen on the basis of the target
application. A configuration device, intelligent controller, or the JTAG
port can be used to control the configuration of a Mercury device. When a
configuration device is used, the system can configure automatically at
system power-up.
By connecting the configuration enable (nCE) and configuration enable
output (nCEO) pins on each device, multiple Mercury devices can be
configured in any of five configuration schemes.
Table 52. Data Sources for Configuration
Configuration Scheme
Data Source
Configuration device
Configuration device
Passive serial (PS)
MasterBlasterTM or ByteBlasterMVTM download cable
or serial data source
Passive parallel asynchronous (PPA)
Parallel data source
Passive parallel synchronous (PPS)
Parallel data source
JTAG
MasterBlaster or ByteBlasterMV download cable or a
microprocessor with a Jam STAPL or JBC file
f
Device PinOuts
84
For more information on configuration, see Application Note 116
(Configuring APEX 20K, FLEX 10K & FLEX 6000 Devices).
See the Altera web site (http://www.altera.com) or the Altera Digital
Library for pin-out information.
Altera Corporation
Mercury Programmable Logic Device Family Data Sheet
Revision
History
The information contained in the Mercury Programmable Logic Device
Family Data Sheet version 2.2 supersedes information published in
previous versions.
Version 2.2
The following changes were made to the Mercury Programmable Logic
Device Family Data Sheet version 2.2:
■
Updated the condition values (symbols II and IOZ) in Table 22.
Version 2.1
The following changes were made to the Mercury Programmable Logic
Device Family Data Sheet version 2.1:
■
■
■
Updated Table 8.
Updated EP1M350 regular I/O banks in Table 13.
Updated Note (6) in Table 14.
13
Development
Tools
Version 2.0
The following changes were made to the Mercury Programmable Logic
Device Family Data Sheet version 2.0:
■
■
■
■
■
■
■
■
■
Altera Corporation
Changed all references to PCML to 3.3-V PCML.
Updated Table 4.
Updated “High-Speed Differential Interface” on page 8.
Added Tables 6 through 8.
Added Figures 34 and 35.
Updated I/O specifications in Tables 28 and 29.
Updated Mercury device capacitance in Table 43.
Updated EP1M120 device timing in Tables 46 through 49.
Added EP1M350 device timing in Tables 50 and 51.
85
Mercury Programmable Logic Device Family Data Sheet
®
101 Innovation Drive
San Jose, CA 95134
(408) 544-7000
http://www.altera.com
Applications Hotline:
(800) 800-EPLD
Customer Marketing:
(408) 544-7104
Literature Services:
lit_req@altera.com
Copyright © 2003 Altera Corporation. All rights reserved. Altera, The Programmable Solutions Company, the
stylized Altera logo, specific device designations, and all other words and logos that are identified as
trademarks and/or service marks are, unless noted otherwise, the trademarks and service marks of Altera
Corporation in the U.S. and other countries. All other product or service names are the property of their
respective holders. Altera products are protected under numerous U.S. and foreign patents and pending
applications, mask work rights, and copyrights. Altera warrants performance of its semiconductor products
to current specifications in accordance with Altera’s standard warranty, but reserves the right to make changes
to any products and services at any time without notice. Altera assumes no responsibility
or liability arising out of the application or use of any information, product, or service
described herein except as expressly agreed to in writing by Altera Corporation. Altera
customers are advised to obtain the latest version of device specifications before relying on
any published information and before placing orders for products or services.
Altera Corporation
86
工商网监
湘ICP备2023018690号
