mirror of
https://git.FreeBSD.org/src.git
synced 2024-12-11 09:50:12 +00:00
da4e1e319c
Add the Arm CoreLink CMN-600 Coherent Mesh Network Controller PMU driver man page. Reviewed by: pauamma_gundo.com Sponsored By: ARM Differential Revision: https://reviews.freebsd.org/D32335
598 lines
20 KiB
Groff
598 lines
20 KiB
Groff
.\" Copyright (c) 2021 ARM Ltd. All rights reserved.
|
|
.\"
|
|
.\" Redistribution and use in source and binary forms, with or without
|
|
.\" modification, are permitted provided that the following conditions
|
|
.\" are met:
|
|
.\" 1. Redistributions of source code must retain the above copyright
|
|
.\" notice, this list of conditions and the following disclaimer.
|
|
.\" 2. Redistributions in binary form must reproduce the above copyright
|
|
.\" notice, this list of conditions and the following disclaimer in the
|
|
.\" documentation and/or other materials provided with the distribution.
|
|
.\"
|
|
.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
|
|
.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
|
.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
|
.\" ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
|
|
.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
|
.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
|
.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
|
.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
|
.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
|
.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
|
.\" SUCH DAMAGE.
|
|
.\"
|
|
.\" $FreeBSD$
|
|
.\"
|
|
.Dd December 19, 2021
|
|
.Dt PMC.CMN-600 3
|
|
.Os
|
|
.Sh NAME
|
|
.Nm pmc.cmn-600
|
|
.Nd Library for accessing the Arm CoreLink CMN-600 Coherent Mesh Network
|
|
Controller performance counter events
|
|
.Sh LIBRARY
|
|
.Lb libpmc
|
|
.Sh SYNOPSIS
|
|
.In pmc.h
|
|
.Sh DESCRIPTION
|
|
CMN-600 PMU counters may be configured to count any one of a defined set of
|
|
hardware events.
|
|
Unlike other performance counters, counters for the CMN-600 require the
|
|
node ID to set up.
|
|
.Pp
|
|
Node ID information currently can be obtained one of two ways.
|
|
Using bootverbose, for example set sysctl debug.bootverbose=1 and then load the
|
|
.Xr hwpmc 4
|
|
KLD module.
|
|
The cmn600 module will be loaded automatically as a dependency:
|
|
.Dl $ sysctl debug.bootverbose=1
|
|
.Dl $ kldload hwpmc
|
|
Another way is to use sysctl to trigger
|
|
dump of nodes tree to system console:
|
|
.Dl $ sysctl dev.cmn600.0.dump_nodes=1
|
|
.Pp
|
|
Some BIOS versions of dual-socket machines have no NUMA domain information in
|
|
ACPI.
|
|
In such cases, to get more accurate events statistics, set the kernel
|
|
environment variable hint.cmn600.{unit}.domain={value}.
|
|
Where {unit} is a cmn600 device unit number and {value} is the NUMA domain of
|
|
the CPU package containing that CMN-600 controller.
|
|
Example:
|
|
.Dl $ kenv hint.cmn600.0.domain=0
|
|
.Dl $ kenv hint.cmn600.1.domain=1
|
|
.Dl $ kldunload hwpmc cmn600
|
|
.Dl $ kldload hwpmc
|
|
.Pp
|
|
Arm CoreLink CMN-600 Coherent Mesh Network Controller performance counters are
|
|
documented in
|
|
.Rs
|
|
.%B "Arm CoreLink CMN-600 Coherent Mesh Network Technical Reference Manual"
|
|
.%T "Revision: r3p2"
|
|
.%D 2020
|
|
.%Q "ARM Limited"
|
|
.Re
|
|
.Ss PMC Capabilities
|
|
CMN-600 PMU counters support the following capabilities:
|
|
.Bl -column "PMC_CAP_INTERRUPT" "Support"
|
|
.It Sy Capability Ta Em Support
|
|
.It PMC_CAP_CASCADE Ta \&No
|
|
.It PMC_CAP_EDGE Ta \&No
|
|
.It PMC_CAP_INTERRUPT Ta Yes
|
|
.It PMC_CAP_INVERT Ta \&No
|
|
.It PMC_CAP_READ Ta Yes
|
|
.It PMC_CAP_PRECISE Ta \&No
|
|
.It PMC_CAP_SYSTEM Ta Yes
|
|
.It PMC_CAP_TAGGING Ta \&No
|
|
.It PMC_CAP_THRESHOLD Ta Yes
|
|
.It PMC_CAP_USER Ta \&No
|
|
.It PMC_CAP_WRITE Ta Yes
|
|
.El
|
|
.Ss Event Qualifiers
|
|
Event specifiers for these PMCs support the following common
|
|
qualifiers:
|
|
.Bl -tag -width indent
|
|
.It Li nodeid= Ns Ar nodeid
|
|
Request counting for specific event at node
|
|
.Ar nodeid .
|
|
.It Li occupancy= Ns Ar value
|
|
Filtering by occupancy type.
|
|
.It Li xpport= Ns Ar port
|
|
Count only events matched by
|
|
.Ar port .
|
|
(East, West, North, South, devport0, devport1 or numeric 0 to 5)
|
|
.It Li xpchannel= Ns Ar channel
|
|
Filter events by XP node channel.
|
|
(REQ, RSP, SNP, DAT or 0, 1, 2, 3)
|
|
.El
|
|
.Ss Class Name Prefix
|
|
These PMCs are named using a class name prefix of
|
|
.Dq Li CMN600_PMU_ .
|
|
.Ss Event Specifiers
|
|
The following list of PMC events are available:
|
|
.Ss DVM node events
|
|
.Bl -tag -width indent
|
|
.It Sy dn_rxreq_dvmop
|
|
Number of DVMOP requests received.
|
|
This includes all the sub-types include TLB invalidate, Branch predictor
|
|
invalidate, instruction cache (physical and virtual) invalidate.
|
|
.It Sy dn_rxreq_dvmsync
|
|
Number of DVM Sync requests received.
|
|
.It Sy dn_rxreq_dvmop_vmid_filtered
|
|
Number of incoming DVMOP requests that are subject to VMID based filtering.
|
|
This is a measure of the effectiveness of VMID based filtering and potential
|
|
reduction in DVM snoops.
|
|
.It Sy dn_rxreq_retried
|
|
Number of incoming requests that are retried.
|
|
This is a measure of the retry rate.
|
|
.It Sy dn_rxreq_trk_occupancy
|
|
Counts the tracker occupancy in DN. "occupancy": All, dvmop, dvmsync
|
|
.It Sy dn_rxreq_tlbi_dvmop
|
|
Number of DVMOP TLB invalidate requests received.
|
|
.It Sy dn_rxreq_bpi_dvmop
|
|
Number of DVMOP Branch predictor invalidate requests received.
|
|
.It Sy dn_rxreq_pici_dvmop
|
|
Number of DVMOP physical instruction cache invalidate requests received.
|
|
.It Sy dn_rxreq_vivi_dvmop
|
|
Number of DVMOP virtual instruction cache invalidate requests received.
|
|
.It Sy dn_rxreq_dvmop_other_filtered
|
|
Number of DVM op requests to RNDs, BPI or PICI/VICI, that were filtered
|
|
.It Sy dn_rxreq_snp_sent
|
|
Number of SNPs sent to RNs
|
|
.It Sy dn_rxreq_snp_stalled
|
|
Number of SNPs stalled to RNs due to lack of Crds
|
|
.It Sy dn_rxreq_trk_full
|
|
DVM tracker full counter
|
|
.El
|
|
.Ss HN-F node events
|
|
.Bl -tag -width indent
|
|
.It Sy hnf_cache_miss
|
|
Counts total cache misses in first lookup result (high priority)
|
|
.It Sy hnf_slc_sf_cache_access
|
|
Counts number of cache accesses in first access (high priority)
|
|
.It Sy hnf_cache_fill
|
|
Counts total allocations in HN SLC (all cache line allocations to SLC)
|
|
.It Sy hnf_pocq_retry
|
|
Counts number of retried requests
|
|
.It Sy hnf_pocq_reqs_recvd
|
|
Counts number of requests received by HN
|
|
.It Sy hnf_sf_hit
|
|
Counts number of SF hits
|
|
.It Sy hnf_sf_evictions
|
|
Counts number of SF eviction cache invalidations initiated
|
|
.It Sy hnf_dir_snoops_sent
|
|
Counts number of directed snoops sent (not including SF back invalidation)
|
|
.It Sy hnf_brd_snoops_sent
|
|
Counts number of multicast snoops sent (not including SF back invalidation)
|
|
.It Sy hnf_slc_eviction
|
|
Counts number of SLC evictions (dirty only)
|
|
.It Sy hnf_slc_fill_invalid_way
|
|
Counts number of SLC fills to an invalid way
|
|
.It Sy hnf_mc_retries
|
|
Counts number of retried transactions by the MC
|
|
.It Sy hnf_mc_reqs
|
|
Counts number of requests sent to MC
|
|
.It Sy hnf_qos_hh_retry
|
|
Counts number of times a HighHigh priority request is protocol-retried at the
|
|
HN-F
|
|
.It Sy hnf_qos_pocq
|
|
Counts the POCQ occupancy in HN-F.
|
|
Support argument "occupancy". Accept: All, Read, Write, Atomic, Stash.
|
|
Default: All.
|
|
.It Sy hnf_pocq_addrhaz
|
|
Counts number of POCQ address hazards upon allocation
|
|
.It Sy hnf_pocq_atomic_addrhaz
|
|
Counts number of POCQ address hazards upon allocation for atomic operations
|
|
.It Sy hnf_ld_st_swp_adq_full
|
|
Counts number of times ADQ is full for Ld/St/SWP type atomic operations while
|
|
POCQ has pending operations
|
|
.It Sy hnf_cmp_adq_full
|
|
Counts number of times ADQ is full for CMP type atomic operations while POCQ
|
|
has pending operations
|
|
.It Sy hnf_txdat_stall
|
|
Counts number of times HN-F has a pending TXDAT flit but no credits to upload
|
|
.It Sy hnf_txrsp_stall
|
|
Counts number of times HN-F has a pending TXRSP flit but no credits to upload
|
|
.It Sy hnf_seq_full
|
|
Counts number of times requests are replayed in SLC pipe due to SEQ being full
|
|
.It Sy hnf_seq_hit
|
|
Counts number of times a request in SLC hit a pending SF eviction in SEQ
|
|
.It Sy hnf_snp_sent
|
|
Counts number of snoops sent including directed, multicast, and SF back
|
|
invalidation
|
|
.It Sy hnf_sfbi_dir_snp_sent
|
|
Counts number of times directed snoops were sent due to SF back invalidation
|
|
.It Sy hnf_sfbi_brd_snp_sent
|
|
Counts number of times multicast snoops were sent due to SF back invalidation
|
|
.It Sy hnf_snp_sent_untrk
|
|
Counts number of times snooped were sent due to untracked RNF's
|
|
.It Sy hnf_intv_dirty
|
|
Counts number of times SF back invalidation resulted in dirty line intervention
|
|
from the RN
|
|
.It Sy hnf_stash_snp_sent
|
|
Counts number of times stash snoops were sent
|
|
.It Sy hnf_stash_data_pull
|
|
Counts number of times stash snoops resulted in data pull from the RN
|
|
.It Sy hnf_snp_fwded
|
|
Counts number of times data forward snoops were sent
|
|
.El
|
|
.Ss HN-I node events
|
|
.Bl -tag -width indent
|
|
.It Sy hni_rrt_rd_occ_cnt_ovfl
|
|
RRT read occupancy count overflow
|
|
.It Sy hni_rrt_wr_occ_cnt_ovfl
|
|
RRT write occupancy count overflow
|
|
.It Sy hni_rdt_rd_occ_cnt_ovfl
|
|
RDT read occupancy count overflow
|
|
.It Sy hni_rdt_wr_occ_cnt_ovfl
|
|
RDT write occupancy count overflow
|
|
.It Sy hni_wdb_occ_cnt_ovfl
|
|
WDB occupancy count overflow
|
|
.It Sy hni_rrt_rd_alloc
|
|
RRT read allocation
|
|
.It Sy hni_rrt_wr_alloc
|
|
RRT write allocation
|
|
.It Sy hni_rdt_rd_alloc
|
|
RDT read allocation
|
|
.It Sy hni_rdt_wr_alloc
|
|
RDT write allocation
|
|
.It Sy hni_wdb_alloc
|
|
WDB allocation
|
|
.It Sy hni_txrsp_retryack
|
|
RETRYACK TXRSP flit sent
|
|
.It Sy hni_arvalid_no_arready
|
|
ARVALID set without ARREADY event
|
|
.It Sy hni_arready_no_arvalid
|
|
ARREADY set without ARVALID event
|
|
.It Sy hni_awvalid_no_awready
|
|
AWVALID set without AWREADY event
|
|
.It Sy hni_awready_no_awvalid
|
|
AWREADY set without AWVALID event
|
|
.It Sy hni_wvalid_no_wready
|
|
WVALID set without WREADY event
|
|
.It Sy hni_txdat_stall
|
|
TXDAT stall (TXDAT valid but no link credit available)
|
|
.It Sy hni_nonpcie_serialization
|
|
Non-PCIe serialization event
|
|
.It Sy hni_pcie_serialization
|
|
PCIe serialization event
|
|
.El
|
|
.Ss XP node events
|
|
.Bl -tag -width indent
|
|
.It Sy xp_txflit_valid
|
|
Number of flits transmitted on a specified port and CHI channel.
|
|
This is a measure of the flit transfer bandwidth from an XP.
|
|
Note: On device ports, this event also includes link flit transfers.
|
|
.It Sy xp_txflit_stall
|
|
Number of cycles when a flit is stalled at an XP waiting for link credits at a
|
|
specified port and CHI channel.
|
|
This is a measure of the flit traffic congestion on the mesh and at the flit
|
|
download ports.
|
|
.It Sy xp_partial_dat_flit
|
|
Number of times when a partial DAT flit is uploaded onto the mesh from a
|
|
RN-F_CHIA port.
|
|
Partial DAT flit transmission occurs when XP is not able to combine two 128b
|
|
DAT flits and send them over the 256b DAT channel.
|
|
This can happen under 2 circumstances:
|
|
1. Only one 128b DAT flit is received within a transmission time window.
|
|
2. Two 128b DAT flits are received but they are not two halves of a
|
|
single 256b word.
|
|
.El
|
|
.Ss SBSX node events
|
|
.Bl -tag -width indent
|
|
.It Sy sbsx_rd_req
|
|
Read request
|
|
.It Sy sbsx_wr_req
|
|
Write request
|
|
.It Sy sbsx_cmo_req
|
|
CMO request
|
|
.It Sy sbsx_txrsp_retryack
|
|
RETRYACK TXRSP flit sent
|
|
.It Sy sbsx_txdat_flitv
|
|
TXDAT flit seen
|
|
.It Sy sbsx_txrsp_flitv
|
|
TXRSP flit seen
|
|
.It Sy sbsx_rd_req_trkr_occ_cnt_ovfl
|
|
Read request tracker occupancy count overflow
|
|
.It Sy sbsx_wr_req_trkr_occ_cnt_ovfl
|
|
Write request tracker occupancy count overflow
|
|
.It Sy sbsx_cmo_req_trkr_occ_cnt_ovfl
|
|
CMO request tracker occupancy count overflow
|
|
.It Sy sbsx_wdb_occ_cnt_ovfl
|
|
WDB occupancy count overflow
|
|
.It Sy sbsx_rd_axi_trkr_occ_cnt_ovfl
|
|
Read AXI pending tracker occupancy count overflow
|
|
.It Sy sbsx_cmo_axi_trkr_occ_cnt_ovfl
|
|
CMO AXI pending tracker occupancy count overflow
|
|
.It Sy sbsx_arvalid_no_arready
|
|
ARVALID set without ARREADY
|
|
.It Sy sbsx_awvalid_no_awready
|
|
AWVALID set without AWREADY
|
|
.It Sy sbsx_wvalid_no_wready
|
|
WVALID set without WREADY
|
|
.It Sy sbsx_txdat_stall
|
|
TXDAT stall (TXDAT valid but no link credit available)
|
|
.It Sy sbsx_txrsp_stall
|
|
TXRSP stall (TXRSP valid but no link credit available)
|
|
.El
|
|
.Ss RN-D node events
|
|
.Bl -tag -width indent
|
|
.It Sy rnd_s0_rdata_beats
|
|
Number of RData beats, RVALID and RREADY, dispatched on port 0.
|
|
This is a measure of the read bandwidth, including CMO responses.
|
|
.It Sy rnd_s1_rdata_beats
|
|
Number of RData beats, RVALID and RREADY, dispatched on port 1.
|
|
This is a measure of the read bandwidth, including CMO responses.
|
|
.It Sy rnd_s2_rdata_beats
|
|
Number of RData beats, RVALID and RREADY, dispatched on port 2.
|
|
This is a measure of the read bandwidth, including CMO responses.
|
|
.It Sy rnd_rxdat_flits
|
|
Number of RXDAT flits received.
|
|
This is a measure of the true read data bandwidth, excluding CMOs.
|
|
.It Sy rnd_txdat_flits
|
|
Number of TXDAT flits dispatched.
|
|
This is a measure of the write bandwidth.
|
|
.It Sy rnd_txreq_flits_total
|
|
Number of TXREQ flits dispatched.
|
|
This is a measure of the total request
|
|
bandwidth.
|
|
.It Sy rnd_txreq_flits_retried
|
|
Number of retried TXREQ flits dispatched.
|
|
This is a measure of the retry rate.
|
|
.It Sy rnd_rrt_occ_ovfl
|
|
All entries in the read request tracker are occupied.
|
|
This is a measure of oversubscription in the read request tracker.
|
|
.It Sy rnd_wrt_occ_ovfl
|
|
All entries in the write request tracker are occupied.
|
|
This is a measure of oversubscription in the write request tracker.
|
|
.It Sy rnd_txreq_flits_replayed
|
|
Number of replayed TXREQ flits.
|
|
This is the measure of replay rate.
|
|
.It Sy rnd_wrcancel_sent
|
|
Number of write data cancels sent.
|
|
This is the measure of write cancel rate.
|
|
.It Sy rnd_s0_wdata_beats
|
|
Number of WData beats, WVALID and WREADY, dispatched on port 0.
|
|
This is a measure of write bandwidth on AXI port 0.
|
|
.It Sy rnd_s1_wdata_beats
|
|
Number of WData beats, WVALID and WREADY, dispatched on port 1.
|
|
This is a measure of write bandwidth on AXI port 1.
|
|
.It Sy rnd_s2_wdata_beats
|
|
Number of WData beats, WVALID and WREADY, dispatched on port 2.
|
|
This is a measure of write bandwidth on AXI port 2.
|
|
.It Sy rnd_rrt_alloc
|
|
Number of allocations in the read request tracker.
|
|
This is a measure of read transaction count.
|
|
.It Sy rnd_wrt_alloc
|
|
Number of allocations in the write request tracker.
|
|
This is a measure of write transaction count.
|
|
.It Sy rnd_rdb_unord
|
|
Number of cycles for which Read Data Buffer state machine is in Unordered Mode.
|
|
.It Sy rnd_rdb_replay
|
|
Number of cycles for which Read Data Buffer state machine is in Replay mode
|
|
.It Sy rnd_rdb_hybrid
|
|
Number of cycles for which Read Data Buffer state machine is in hybrid mode.
|
|
Hybrid mode is where there is mix of ordered/unordered traffic.
|
|
.It Sy rnd_rdb_ord
|
|
Number of cycles for which Read Data Buffer state machine is in ordered Mode.
|
|
.El
|
|
.Ss RN-I node events
|
|
.Bl -tag -width indent
|
|
.It Sy rni_s0_rdata_beats
|
|
Number of RData beats, RVALID and RREADY, dispatched on port 0.
|
|
This is a measure of the read bandwidth, including CMO responses.
|
|
.It Sy rni_s1_rdata_beats
|
|
Number of RData beats, RVALID and RREADY, dispatched on port 1.
|
|
This is a measure of the read bandwidth, including CMO responses.
|
|
.It Sy rni_s2_rdata_beats
|
|
Number of RData beats, RVALID and RREADY, dispatched on port 2.
|
|
This is a measure of the read bandwidth, including CMO responses.
|
|
.It Sy rni_rxdat_flits
|
|
Number of RXDAT flits received.
|
|
This is a measure of the true read data bandwidth, excluding CMOs.
|
|
.It Sy rni_txdat_flits
|
|
Number of TXDAT flits dispatched.
|
|
This is a measure of the write bandwidth.
|
|
.It Sy rni_txreq_flits_total
|
|
Number of TXREQ flits dispatched.
|
|
This is a measure of the total request bandwidth.
|
|
.It Sy rni_txreq_flits_retried
|
|
Number of retried TXREQ flits dispatched.
|
|
This is a measure of the retry rate.
|
|
.It Sy rni_rrt_occ_ovfl
|
|
All entries in the read request tracker are occupied.
|
|
This is a measure of oversubscription in the read request tracker.
|
|
.It Sy rni_wrt_occ_ovfl
|
|
All entries in the write request tracker are occupied.
|
|
This is a measure of oversubscription in the write request tracker.
|
|
.It Sy rni_txreq_flits_replayed
|
|
Number of replayed TXREQ flits.
|
|
This is the measure of replay rate.
|
|
.It Sy rni_wrcancel_sent
|
|
Number of write data cancels sent.
|
|
This is the measure of write cancel rate
|
|
.It Sy rni_s0_wdata_beats
|
|
Number of WData beats, WVALID and WREADY, dispatched on port 0.
|
|
This is a measure of write bandwidth on AXI port 0.
|
|
.It Sy rni_s1_wdata_beats
|
|
Number of WData beats, WVALID and WREADY, dispatched on port 1.
|
|
This is a measure of write bandwidth on AXI port 1.
|
|
.It Sy rni_s2_wdata_beats
|
|
Number of WData beats, WVALID and WREADY, dispatched on port 2.
|
|
This is a measure of write bandwidth on AXI port 2.
|
|
.It Sy rni_rrt_alloc
|
|
Number of allocations in the read request tracker.
|
|
This is a measure of read transaction count.
|
|
.It Sy rni_wrt_alloc
|
|
Number of allocations in the write request tracker.
|
|
This is a measure of write transaction count
|
|
.It Sy rni_rdb_unord
|
|
Number of cycles for which Read Data Buffer state machine is in Unordered Mode.
|
|
.It Sy rni_rdb_replay
|
|
Number of cycles for which Read Data Buffer state machine is in Replay mode
|
|
.It Sy rni_rdb_hybrid
|
|
Number of cycles for which Read Data Buffer state machine is in hybrid mode.
|
|
Hybrid mode is where there is mix of ordered/unordered traffic.
|
|
.It Sy rni_rdb_ord
|
|
Number of cycles for which Read Data Buffer state machine is in ordered Mode.
|
|
.El
|
|
.Ss CXHA node events
|
|
.Pq Note: CXHA events descriptions are guessed
|
|
.Bl -tag -width indent
|
|
.It Sy cxha_rddatbyp
|
|
Number of Read DAT Bypass
|
|
.It Sy cxha_chirsp_up_stall
|
|
Number of CHI RSP up Stall
|
|
.It Sy cxha_chidat_up_stall
|
|
Number of CHI DAT up Stall
|
|
.It Sy cxha_snppcrd_lnk0_stall
|
|
Number of Snoop Pcrd Stall on Link 0
|
|
.It Sy cxha_snppcrd_lnk1_stall
|
|
Number of Snoop Pcrd Stall on Link 1
|
|
.It Sy cxha_snppcrd_lnk2_stall
|
|
Number of Snoop Pcrd Stall on Link 2
|
|
.It Sy cxha_reqtrk_occ
|
|
Request Tracker Occupancy
|
|
.It Sy cxha_rdb_occ
|
|
Read Data Buffer Occupancy
|
|
.It Sy cxha_rdbbyp_occ
|
|
Read Data Buffer Bypass Occupancy
|
|
.It Sy cxha_wdb_occ
|
|
Write Data Buffer Occupancy
|
|
.It Sy cxha_snptrk_occ
|
|
Snoop Tracker Occupancy
|
|
.It Sy cxha_sdb_occ
|
|
SDB Occupancy
|
|
.It Sy cxha_snphaz_occ
|
|
Snoop Hazard Occupancy
|
|
.El
|
|
.Ss CXRA node events
|
|
.Bl -tag -width indent
|
|
.It Sy cxra_req_trk_occ
|
|
Request tracker occupancy
|
|
.It Sy cxra_snp_trk_occ
|
|
Snoop tracker occupancy
|
|
.It Sy cxra_rd_dat_buf_occ
|
|
Read data buffer occupancy
|
|
.It Sy cxra_wr_dat_buf_occ
|
|
Write data buffer occupancy
|
|
.It Sy cxra_snp_sink_buf_occ
|
|
Snoop sink buffer occupancy
|
|
.It Sy cxra_snp_bcasts
|
|
Snoop broadcasts
|
|
.It Sy cxra_req_chains
|
|
Number of request chains formed larger than one
|
|
.It Sy cxra_req_chain_avg_len
|
|
Average size of request chains, only for chain sizes larger than one
|
|
.It Sy cxra_chi_rsp_upload_stalls
|
|
Local RA upload stalls to CHI because of contention with HA
|
|
.It Sy cxra_chi_dat_upload_stalls
|
|
Local RA upload stalls to CHI because of contention with HA
|
|
.It Sy cxra_dat_pcrd_stalls_lnk0
|
|
Memory Data Request available, but no DAT Pcrd to send over CCIX per LinkEnd 0
|
|
.It Sy cxra_dat_pcrd_stalls_lnk1
|
|
Memory Data Request available, but no DAT Pcrd to send over CCIX per LinkEnd 1
|
|
.It Sy cxra_dat_pcrd_stalls_lnk2
|
|
Memory Data Request available, but no DAT Pcrd to send over CCIX per LinkEnd 2
|
|
.It Sy cxra_req_pcrd_stalls_lnk0
|
|
Memory Data Request available but no Req Pcrd to send over CCIX per LinkEnd 0
|
|
.It Sy cxra_req_pcrd_stalls_lnk1
|
|
Memory Data Request available but no Req Pcrd to send over CCIX per LinkEnd 1
|
|
.It Sy cxra_req_pcrd_stalls_lnk2
|
|
Memory Data Request available but no Req Pcrd to send over CCIX per LinkEnd 2
|
|
.It Sy cxra_ext_rsp_stall
|
|
CHI external RSP stall
|
|
.It Sy cxra_ext_dat_stall
|
|
CHI external DAT stall
|
|
.El
|
|
.Ss CXLA node events
|
|
.Bl -tag -width indent
|
|
.It Sy cxla_rx_tlp_link0
|
|
RX TLP for Link 0
|
|
.It Sy cxla_rx_tlp_link1
|
|
RX TLP for Link 1
|
|
.It Sy cxla_rx_tlp_link2
|
|
RX TLP for Link 2
|
|
.It Sy cxla_tx_tlp_link0
|
|
TX TLP for Link 0
|
|
.It Sy cxla_tx_tlp_link1
|
|
TX TLP for Link 1
|
|
.It Sy cxla_tx_tlp_link2
|
|
TX TLP for Link 2
|
|
.It Sy cxla_rx_cxs_link0
|
|
RX CXS for Link 0
|
|
.It Sy cxla_rx_cxs_link1
|
|
RX CXS for Link 1
|
|
.It Sy cxla_rx_cxs_link2
|
|
RX CXS for Link 2
|
|
.It Sy cxla_tx_cxs_link0
|
|
TX CXS for Link 0
|
|
.It Sy cxla_tx_cxs_link1
|
|
TX CXS for Link 1
|
|
.It Sy cxla_tx_cxs_link2
|
|
TX CXS for Link 2
|
|
.It Sy cxla_avg_rx_tlp_sz_dws
|
|
Average RX TLP size in DWs
|
|
.It Sy cxla_avg_tx_tlp_sz_dws
|
|
Average TX TLP size in DWs
|
|
.It Sy cxla_avg_rx_tlp_sz_ccix_msg
|
|
Average RX TLP size in CCIX messages
|
|
.It Sy cxla_avg_tx_tlp_sz_ccix_msg
|
|
Average TX TLP size in CCIX messages
|
|
.It Sy cxla_avg_sz_rx_cxs_dw_beat
|
|
Average size of RX CXS in DWs within a beat
|
|
.It Sy cxla_avg_sz_tx_cxs_dw_beat
|
|
Average size of TX CXS in DWs within a beat
|
|
.It Sy cxla_tx_cxs_link_credit_backpressure
|
|
TX CXS link credit backpressure
|
|
.It Sy cxla_rx_tlp_buffer_full
|
|
RX TLP buffer full and backpressured
|
|
.It Sy cxla_tx_tlp_buffer_full
|
|
TX TLP buffer full and backpressured
|
|
.It Sy cxla_avg_latency_process_rx_tlp
|
|
Average latency to process an RX TLP
|
|
.It Sy cxla_avg_latency_form_tx_tlp
|
|
Average latency to form a TX TLP
|
|
.El
|
|
.Sh SEE ALSO
|
|
.Xr pmc 3 ,
|
|
.Xr pmc.atom 3 ,
|
|
.Xr pmc.core 3 ,
|
|
.Xr pmc.core2 3 ,
|
|
.Xr pmc.corei7 3 ,
|
|
.Xr pmc.corei7uc 3 ,
|
|
.Xr pmc.iaf 3 ,
|
|
.Xr pmc.iaf 3 ,
|
|
.Xr pmc.k7 3 ,
|
|
.Xr pmc.k8 3 ,
|
|
.Xr pmc.p4 3 ,
|
|
.Xr pmc.p5 3 ,
|
|
.Xr pmc.p6 3 ,
|
|
.Xr pmc.soft 3 ,
|
|
.Xr pmc.tsc 3 ,
|
|
.Xr pmc.westmere 3 ,
|
|
.Xr pmc.westmereuc 3 ,
|
|
.Xr pmc_cpuinfo 3 ,
|
|
.Xr pmclog 3 ,
|
|
.Xr hwpmc 4
|
|
.Sh HISTORY
|
|
The
|
|
.Nm pmc
|
|
library first appeared in
|
|
.Fx 6.0 .
|
|
.br
|
|
The
|
|
.Nm pmc.cmn-600
|
|
driver was added in
|
|
.Fx 14.0 .
|
|
.Sh AUTHORS
|
|
.An -nosplit
|
|
The
|
|
.Lb libpmc
|
|
library was written by
|
|
.An Joseph Koshy Aq Mt jkoshy@FreeBSD.org ,
|
|
.An Oleksandr Rybalko Aq Mt ray@FreeBSD.org .
|
|
.br
|
|
The CMN-600 PMU driver was sponsored by ARM Ltd.
|
|
This manual page was written by
|
|
.An Oleksandr Rybalko Aq Mt ray@FreeBSD.org .
|