LogiCORE PCI Interface v3.0
LogiCORE Facts (Cont)
The core meets the setup, hold, and clock to timing require-
ments as specified in the PCI-X specification. The interface
is verified through extensive simulation.
PCI-X 64 Supported Devices
PCI64/33
Only
Virtex-E V300EBG432-8C
Virtex-II 2V1000FG456-5C
3.3v only
3.3v only
Other features that enable efficient implementation of a
PCI-X system include:
PCI-X64/66
Only
Virtex-E V300EBG432-8C
Virtex-II 2V1000FG456-5C
3.3v only
3.3v only
•
Block SelectRAM™ memory. Blocks of on-chip
ultra-fast RAM with synchronous write and dual-port
RAM capabilities. Used in PCI-X designs to implement
FIFOs.
PCI-X64/100
Only
Virtex-II 2V1000FG456-5C
Virtex-II 2V1000FG456-5C
3.3v only
3.3v only
PCI-X64/66
PCI64/33
•
SelectRAM memory. Distributed on-chip ultra-fast RAM
with synchronous write option and dual-port RAM
capabilities. Used in PCI-X designs to implement
FIFOs.
Xilinx provides technical support for this LogiCORE product when used as described
in the Design Guide and the Implementation Guide. Xilinx cannot guarantee timing,
functionality, or support of product if implemented in devices not listed, or if custom-
ized beyond that allowed in the product documentation.
Note: Fully compliant designs over 66 MHz require two bitstreams.
Note: Universal card implementations not supported.
Note: Commercial devices only; 0 C < Tj < 85 C.
The interface is carefully optimized for best possible perfor-
mance and utilization in Xilinx FPGA devices.
More Features
Smart-IP Technology
•
Supported PCI and PCI-X functions:
Drawing on the architectural advantages of Xilinx FPGAs,
Xilinx Smart-IP technology ensures the highest perfor-
mance, predictability, repeatability, and flexibility in PCI-X
designs. The Smart-IP technology is incorporated in every
LogiCORE PCI-X Interface.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Memory Write
I/O Read
I/O Write
Configuration Read
Xilinx Smart-IP technology leverages the Xilinx architectural
advantages, such as look-up tables and segmented routing,
as well as floorplanning information, such as logic mapping
and location constraints. This technology provides the best
physical layout, predictability, and performance. Addition-
ally, these features allow for significantly reduced compile
times over competing architectures.
Configuration Write
Interrupt Acknowledge
Bus Parking
Type 0 Configuration Space Header
Full 64-bit Addressing Support
Up to 6 Base Address Registers
Expansion ROM Base Address Register
Instant-On Base Address Registers
Parity Generation, Parity Error Detection
Full Command/Status Registers
To guarantee the critical setup, hold, minimum clock to out,
and maximum clock to out timing, the PCI-X interface is
delivered with Smart-IP constraint files that are unique for a
device and package combination. These constraint files
guide the implementation tools so that the critical paths
always are within specification.
Applications
Xilinx provides Smart-IP constraint files for many device
and package combinations. Constraint files for unsupported
device and package combinations may be generated using
the web-based constraint file generator.
•
Embedded applications in networking, industrial,
and telecommunication systems
•
PCI-X add-in boards such as frame buffers, network
adapters, and data acquisition boards
Functional Description
The LogiCORE PCI-X Interface is partitioned into six major
blocks and a user application as shown in Figure 1.
•
•
Hot swap CompactPCI-X boards
Any applications that need a PCI-X interface
General Description
Datapath
The LogiCORE PCI-X Interface is a preimplemented and
fully tested module for Xilinx FPGAs. Critical paths are con-
trolled by constraint and guide files to ensure predictable
timing. This significantly reduces the engineering time
required to implement the PCI-X portion of your design.
Resources can instead be focused on your unique user
application logic in the FPGA and on the system level
design. As a result, LogiCORE PCI-X products minimize
your product development time.
There are four datapaths, in and out for both target and ini-
tiator. To improve timing and ease of design, the four unidi-
rectional datapaths are multiplexed inside the interface. All
data transfers are register-to-register. Since fewer registers
are on each datapath, loading is reduced and false timing
paths are eliminated.
2
www.xilinx.com
1-800-255-7778
DS 208 (v.1.2) June 28, 2002
Data Sheet, v3.0.99