Pci/pci-x/pcie iobp, Pcie mps optimization, Pci/pci-x/pcie i/o rope groups – HP INTEGRITY RX3600 User Manual

Page 28

Advertising
background image

Table 1-1 PCI/PCI-X I/O Rope Groups (continued)

Hot Swap /
OL*

Function

Speed

Bits

PCI Bus

Rope Numbers

Slot #

Y

High-Speed PCI-X (Public)

133 MHz

64

5

2, 3 (Dual
Bandwidth)

6

Y

General PCI-X (Public)

66 MHz

64

6

9

7, 8

Y

General PCI-X (Public)

66 MHz

64

7

1

9, 10

N

UCIO (Private)

33 MHz

32

0

0

-

PCI/PCI-X/PCIe IOBP

On the 10 slot PCI/PCI-X/PCIe IOBP there are a total of eight Public slots (four PCI-X mode 1
and four PCIe), two Private Fast-core slots (PCI/PCI-X mode 1 64-bit/66-MHz), and one Private
Slow-core UCIO slot (PCI 32-bit/33-MHz). The eight Public slots are further divided into three
speed/bandwidth configurations:

Two PDHP, which operate at 64-bit/66-MHz PCI-X

Two PSHP, operating at 64-bit/133-MHz PCI-X

Four PCIe 8-lane (x8) 2.5 Gbps, two of which are switched

The two 66 MHz PCI/PCI-X slots are shared. Shared slots have many speed and mode change
restrictions during hot-plug add or remove operations.

Table 1-2 PCI/PCI-X/PCIe I/O Rope Groups

Hot Swap/OLR

Function

Speed

Bits

Rope Numbers

Slot #

N

Core I/O (Private)

66 MHz

64

8

1, 2

N

PCIe x8 (Public or Private depending upon
Core I/O)

2.5 Gbps

x8

10, 11

3, 4

Y

PCIe x8 (Public)

2.5 Gbps

x8

12, 13, 14, 15

5

Y

PCIe x8 (Public)

2.5 Gbps

x8

4, 5, 6, 7

6

Y

High-Speed PCI-X (Public)

133 MHz

64

2, 3

7

Y

High-Speed PCI-X (Public)

133 MHz

64

9

8

Y

General PCI-X (Public)

66 MHz

64

1

9, 10

N

UCIO (Private)

33 MHz

32

0

-

PCIe MPS Optimization

For PCIe-based systems, each PCIe device has a configurable MPS (maximum payload size)
parameter. Larger MPS values can enable the optimization to gain higher performance.MPS
Optimization is supported on PCIe systems running HP-UX, Open VMS, and Linux. System
firmware level greater than 02.03 performs an optimization during boot time to set the MPS
value to the largest size supported by both a PCIe root port and the devices below it.

The default server state is optimization disabled. When disabled system firmware sets MPS to
the minimum value on each PCIe device.

The info io command will display the current PCIe MPS optimization setting. See

“info”

(page 328)

.

To enable PCIe MPS optimization use the ioconfig mps_optimize command. See

“ioconfig”

(page 326)

.

For non-PCIe systems, ioconfig and info io will not display the MPS optimization policy
setting. The Set PCIe MPS Optimization boot manager menu also will not be displayed. Running

28

Overview

Advertising