|
PCI-SIG Certified
PCI Express IP Core
Overview:
The PCI Express IP Core is compliant with the PCI Express Base Specification Revision 1.1, and implements all three layers of the
PCI Express specification (Transaction, Data Link, and Physical)
General Features:
● x8 PCI Express
IP Core
● 64-bit data path at 2.5
Gbps per lane
● x1 Core (125 MHz),
x4 Core (125 MHz), x4 Core (250 MHz), or x8 Core (250 MHz)
and 125 MHz or 250 MHz frequency for 20 Gbps full-duplex bandwidth
● Suitable for Root
Complex, Endpoint, and Dual Mode / Shared Silicon
● PCI Express Base
Specification 1.1 compliant
● Up to eight Virtual
Channels (VCs)
● Receive and Retry buffer
size configurable
● Intel’s PIPE interface
(8-bit or 16-bit mode) to interface between the PHY MAC and PHY physical
coding sub-layer (PCS) /Physical Media Attachment (PMA) layers
● Receive/transmit user
Application Interface
● Input test port for
enabling/disabling specific modes (remote boot, error generation...)
● Output test port for
monitoring (errors, state machine, flow control...)
● Fully compliant PHY PCS
sub-layer (125 MHz or 250 MHz)
- 1-bit PCS/PMA for FPGA
Altera Stratix GX
Altera Stratix GXII
Xilinx Virtex-4 FX
- Wrapper for external PHY
PXPIPE for Philips PX1012A (x1)
Wrapper for Genesys GL9714
(x4 or x8)
Wrapper for PMC Sierra PM8358 (x8)
Wrapper for TI xIO1100 (x1)

Customization
The Core can be customized with a provided wizard that permits
modification of the following parameters:
● Type of Core: Root
Port, Endpoint, Switch (upstream port), Switch (downstream
port), Bridge port, inverse Bridge port, or dual mode / shared
silicon
● Maximum Payload
Size (Max_Payload_Size): up to 4 KB
● Number of lanes: 1
to 4 (125 MHz) and 1 to 8 (250 MHz)
● Number of Virtual
Channels (VCs): 1 to 8 (the unimplemented VCs do not get
synthesized with the Core.)
● Receive Buffer
configuration
●
Dedicated blocks of DPRAM per initialized VC
●
One common block of DPRAM for all initialized VCs
● Buffer sizes:
●
Receive Buffer size: per VC
●
Retry Buffer size
● BAR definition for
Type 0 Configuration Spaces
● Memory windows for
Type 1 Configuration spaces
● Plug-and-Play
parameters
Data Transfer
● Supports up to 4KB
data payload transfer
● Supports all
Memory, I/O, Configuration, and Message transactions
● Highly optimized
Application interface for maximum effective throughput
Configuration
● Implements Type 0
Configuration space for Endpoint designs
● Implements Type 1
Configuration space for Root Complex, Switch, and Bridge designs
● Up to 6 BARs plus
expansion ROM can be implemented for Endpoints
● All I/O and memory
windows implemented for Root complex, Switch, and Bridge
components
Power Management and Interrupt
● All Power State
and associated logic implemented
● Legacy PCI Power
Management support
● Native Active
State Power Management L0s state support
● Power Management
Event (PME message) and Beacon (Wake-Up) support
● MSI (up to 32,
mapped to any TC) and INT message support
● Controlled easily
by Application signals
Reference Design
● Master/target
Application interface type for Endpoint and Root Complex designs
● Target connected
to SRAM and registers
● Highly pipelined
Master interface connected to SRAM module and data generator
● Provided as source
code for free adaptation to user’s design
Design Files
● Standard VHDL and
Verilog source code for ASIC design |
Populated with Xilinx
Virtex-4 FX60/100

Populated with
Xilinx
Virtex-4 LX 60/100/160

Populated with Altera
Stratix II GX
Populated with Altera
Stratix GX25/40
|
|
PCI Express External PHY
(4 Lanes)
PCI Express External PHY
(1 Lane)
PCI
Express to PCI Bridge |
Free Evaluation:
The Evaluation package provides Core functionality similar to the
purchased product but some features may be disabled and the evaluation
will stop functioning after a certain time. Please
contact us for registration
Price:
 |











|