384 Hardware Reference Manual
Intel® IXP2800 Network Processor
Performance Monitor Unit
Chassis/Push-Pull
Command Bus Utilization
These statistics give the number of the command requests issued by
the different Masters in a particular period of time.
This measurement also indicates how long it takes to issue the grant
from the request being issued by the different Masters.
Push and Pull Bus Utilization This measurement keeps track of the number of accesses issued and
how long it takes to send the data to its destination.
Hash
Number of Accesses by Command
Type
This measurement indicates the number of hash accesses issued; this
count is maskable, based on command type.
Latency of Histogram This monitors the latency through each of the HASH queues.
Scratch
Number of Accesses by Command
Type
This measurement indicates the number of Scratch accesses issued
and this count is maskable, based on command type.
Number of Bytes Transfer This measurement indicates total number of bytes transferred to or from
Scratch.
Latency of Histogram This measurement indicates the latency of performing read or write from
the Scratch. Latency in command executions may also be measured.
PCI
Master Accesses
These statistics give the number of Master accesses that were
generated by the PCI blocks. This measurement can be counted based
on individual command type.
Slave Accesses
These statistics give the number of Slave accesses that were generated
by the PCI blocks. This measurement can be counted based on
individual command type.
Master/Slave Read Byte Count
This statistics give the total number of bytes of data that were generated
by the PCI Master/Slave reads access. This measurement can be
counted based on individual command type.
Master/Slave Write Byte Count
These statistics give the total number of bytes of data that were
generated by the PCI Master/Slave write accesses. This measurement
can be counted based on individual command type.
Burst Size Histogram These statistics give a histogram of the number of various burst sizes.
Media Interface
TBUF Occupancy Histogram This measurement shows the occupancy rate at different depths of the
FIFO. This can help in better utilization of TBUF.
RBUF Occupancy Histogram This measurement shows the occupancy rate at different depths of the
FIFO. This can help in better utilization of RBUF.
Packet/Cell/Frame Count on a Per-
Port Basis
This measurement gives the count of number of packets or cells or
frames transferred in Transmitting mode. This measurement gives the
count of number of packets or cells or frames transferred in the
receiving mode. This may be measured using a per-port basis.
Inter-arrival Time for Packets on a
Per-Port Basis
This measurement can provide information on gaps between packets,
thereby indicating effective line rate.
Burst Size Histogram This measurement gives the various burst sizes of packets being
transmitted and received.
Table 152. Hardware Blocks and Their Performance Measurement Events (Sheet 2 of 2)
Hardware
Block Performance Measurement Event Description