959 lines
34 KiB
Groff
959 lines
34 KiB
Groff
.\" Copyright (c) 2012 Davide Italiano <davide@FreeBSD.org>
|
|
.\" 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 AUTHORS 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 AUTHORS 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 October 19, 2012
|
|
.Dt PMC.SANDYBRIDGE 3
|
|
.Os
|
|
.Sh NAME
|
|
.Nm pmc.sandybridge
|
|
.Nd measurement events for
|
|
.Tn Intel
|
|
.Tn Sandy Bridge
|
|
family CPUs
|
|
.Sh LIBRARY
|
|
.Lb libpmc
|
|
.Sh SYNOPSIS
|
|
.In pmc.h
|
|
.Sh DESCRIPTION
|
|
.Tn Intel
|
|
.Tn "Sandy Bridge"
|
|
CPUs contain PMCs conforming to the version 3 of the
|
|
.Tn Intel
|
|
performance measurement architecture.
|
|
These CPUs may contain up to three classes of PMCs:
|
|
.Bl -tag -width "Li PMC_CLASS_IAP"
|
|
.It Li PMC_CLASS_IAF
|
|
Fixed-function counters that count only one hardware event per counter.
|
|
.It Li PMC_CLASS_IAP
|
|
Programmable counters that may be configured to count one of a defined
|
|
set of hardware events.
|
|
.It Li PMC_CLASS_TSC
|
|
These PMCs are documented in
|
|
.Xr pmc.tsc 3 .
|
|
.El
|
|
.Pp
|
|
The number of PMCs available in each class and their widths need to be
|
|
determined at run time by calling
|
|
.Xr pmc_cpuinfo 3 .
|
|
.Pp
|
|
Intel Sandy Bridge PMCs are documented in
|
|
.Rs
|
|
.%B "Intel(R) 64 and IA-32 Architectures Software Developers Manual"
|
|
.%T "Volume 3B: System Programming Guide, Part 2"
|
|
.%N "Order Number: 253669-039US"
|
|
.%D May 2011
|
|
.%Q "Intel Corporation"
|
|
.Re
|
|
.Ss SANDY BRIDGE FIXED FUNCTION PMCS
|
|
These PMCs and their supported events are documented in
|
|
.Xr pmc.iaf 3 .
|
|
.Ss SANDY BRIDGE PROGRAMMABLE PMCS
|
|
The programmable PMCs support the following capabilities:
|
|
.Bl -column "PMC_CAP_INTERRUPT" "Support"
|
|
.It Em Capability Ta Em Support
|
|
.It PMC_CAP_CASCADE Ta \&No
|
|
.It PMC_CAP_EDGE Ta Yes
|
|
.It PMC_CAP_INTERRUPT Ta Yes
|
|
.It PMC_CAP_INVERT Ta Yes
|
|
.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 Yes
|
|
.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 rsp= Ns Ar value
|
|
Configure the Off-core Response bits.
|
|
.Bl -tag -width indent
|
|
.It Li REQ_DMND_DATA_RD
|
|
Counts the number of demand and DCU prefetch data reads of full and partial
|
|
cachelines as well as demand data page table entry cacheline reads. Does not
|
|
count L2 data read prefetches or instruction fetches.
|
|
.It Li REQ_DMND_RFO
|
|
Counts the number of demand and DCU prefetch reads for ownership (RFO)
|
|
requests generated by a write to data cacheline. Does not count L2 RFO
|
|
prefetches.
|
|
.It Li REQ_DMND_IFETCH
|
|
Counts the number of demand and DCU prefetch instruction cacheline reads.
|
|
Does not count L2 code read prefetches.
|
|
.It Li REQ_WB
|
|
Counts the number of writeback (modified to exclusive) transactions.
|
|
.It Li REQ_PF_DATA_RD
|
|
Counts the number of data cacheline reads generated by L2 prefetchers.
|
|
.It Li REQ_PF_RFO
|
|
Counts the number of RFO requests generated by L2 prefetchers.
|
|
.It Li REQ_PF_IFETCH
|
|
Counts the number of code reads generated by L2 prefetchers.
|
|
.It Li REQ_PF_LLC_DATA_RD
|
|
L2 prefetcher to L3 for loads.
|
|
.It Li REQ_PF_LLC_RFO
|
|
RFO requests generated by L2 prefetcher
|
|
.It Li REQ_PF_LLC_IFETCH
|
|
L2 prefetcher to L3 for instruction fetches.
|
|
.It Li REQ_BUS_LOCKS
|
|
Bus lock and split lock requests.
|
|
.It Li REQ_STRM_ST
|
|
Streaming store requests.
|
|
.It Li REQ_OTHER
|
|
Any other request that crosses IDI, including I/O.
|
|
.It Li RES_ANY
|
|
Catch all value for any response types.
|
|
.It Li RES_SUPPLIER_NO_SUPP
|
|
No Supplier Information available.
|
|
.It Li RES_SUPPLIER_LLC_HITM
|
|
M-state initial lookup stat in L3.
|
|
.It Li RES_SUPPLIER_LLC_HITE
|
|
E-state.
|
|
.It Li RES_SUPPLIER_LLC_HITS
|
|
S-state.
|
|
.It Li RES_SUPPLIER_LLC_HITF
|
|
F-state.
|
|
.It Li RES_SUPPLIER_LOCAL
|
|
Local DRAM Controller.
|
|
.It Li RES_SNOOP_SNPI_NONE
|
|
No details on snoop-related information.
|
|
.It Li RES_SNOOP_SNP_NO_NEEDED
|
|
No snoop was needed to satisfy the request.
|
|
.It Li RES_SNOOP_SNP_MISS
|
|
A snoop was needed and it missed all snooped caches:
|
|
-For LLC Hit, ReslHitl was returned by all cores
|
|
-For LLC Miss, Rspl was returned by all sockets and data was returned from
|
|
DRAM.
|
|
.It Li RES_SNOOP_HIT_NO_FWD
|
|
A snoop was needed and it hits in at least one snooped cache. Hit denotes a
|
|
cache-line was valid before snoop effect. This includes:
|
|
-Snoop Hit w/ Invalidation (LLC Hit, RFO)
|
|
-Snoop Hit, Left Shared (LLC Hit/Miss, IFetch/Data_RD)
|
|
-Snoop Hit w/ Invalidation and No Forward (LLC Miss, RFO Hit S)
|
|
In the LLC Miss case, data is returned from DRAM.
|
|
.It Li RES_SNOOP_HIT_FWD
|
|
A snoop was needed and data was forwarded from a remote socket.
|
|
This includes:
|
|
-Snoop Forward Clean, Left Shared (LLC Hit/Miss, IFetch/Data_RD/RFT).
|
|
.It Li RES_SNOOP_HITM
|
|
A snoop was needed and it HitM-ed in local or remote cache. HitM denotes a
|
|
cache-line was in modified state before effect as a results of snoop. This
|
|
includes:
|
|
-Snoop HitM w/ WB (LLC miss, IFetch/Data_RD)
|
|
-Snoop Forward Modified w/ Invalidation (LLC Hit/Miss, RFO)
|
|
-Snoop MtoS (LLC Hit, IFetch/Data_RD).
|
|
.It Li RES_NON_DRAM
|
|
Target was non-DRAM system address. This includes MMIO transactions.
|
|
.El
|
|
.It Li cmask= Ns Ar value
|
|
Configure the PMC to increment only if the number of configured
|
|
events measured in a cycle is greater than or equal to
|
|
.Ar value .
|
|
.It Li edge
|
|
Configure the PMC to count the number of de-asserted to asserted
|
|
transitions of the conditions expressed by the other qualifiers.
|
|
If specified, the counter will increment only once whenever a
|
|
condition becomes true, irrespective of the number of clocks during
|
|
which the condition remains true.
|
|
.It Li inv
|
|
Invert the sense of comparison when the
|
|
.Dq Li cmask
|
|
qualifier is present, making the counter increment when the number of
|
|
events per cycle is less than the value specified by the
|
|
.Dq Li cmask
|
|
qualifier.
|
|
.It Li os
|
|
Configure the PMC to count events happening at processor privilege
|
|
level 0.
|
|
.It Li usr
|
|
Configure the PMC to count events occurring at privilege levels 1, 2
|
|
or 3.
|
|
.El
|
|
.Pp
|
|
If neither of the
|
|
.Dq Li os
|
|
or
|
|
.Dq Li usr
|
|
qualifiers are specified, the default is to enable both.
|
|
.Ss Event Specifiers (Programmable PMCs)
|
|
Sandy Bridge programmable PMCs support the following events:
|
|
.Bl -tag -width indent
|
|
.It Li LD_BLOCKS.DATA_UNKNOWN
|
|
.Pq EVENT_03H, Umask 01H
|
|
Blocked loads due to store buffer blocks with unknown data.
|
|
.It Li LD_BLOCKS.STORE_FORWARD
|
|
.Pq Event 03H, Umask 02H
|
|
Loads blocked by overlapping with store buffer that cannot be forwarded.
|
|
.It Li LD_BLOCKS.NO_SR
|
|
.Pq Event 03H, Umask 08H
|
|
# of Split loads blocked due to resource not available.
|
|
.It Li LD_BLOCKS.ALL_BLOCK
|
|
.Pq EVENT_03H, Umask 10H
|
|
Number of cases where any load is blocked but has no DCU miss.
|
|
.It Li MISALIGN_MEM_REF.LOADS
|
|
.Pq Event 05H, Umask 01H
|
|
Speculative cache-line split load uops dispatched to L1D.
|
|
.It Li MISALIGN_MEM_REF.STORES
|
|
.Pq Event 05H, Umask 02H
|
|
Speculative cache-line split Store-address uops dispatched to L1D.
|
|
.It Li LD_BLOCKS_PARTIAL.ADDRESS_ALIAS
|
|
.Pq Event 07H, Umask 01H
|
|
False dependencies in MOB due to partial compare on address.
|
|
.It Li LD_BLOCKS_PARTIAL.ALL_STA_BLOCK
|
|
.Pq Event 07H, Umask 08H
|
|
The number of times that load operations are temporarily blocked because of
|
|
older stores, with addresses that are not yet known.
|
|
A load operation may incur more than one block of this type.
|
|
.It LI DTLB_LOAD_MISSES.MISS_CAUSES_A_WALK
|
|
.Pq Event 08H, Umask 01H
|
|
Misses in all TLB levels that cause a page walk of any page size.
|
|
.It Li DTLB_LOAD_MISSES.WALK_COMPLETED
|
|
.Pq Event 08H, Umask 02H
|
|
Misses in all TLB levels that caused page walk completed of any size.
|
|
.It Li DTLB_LOAD_MISSES.WALK_DURATION
|
|
.Pq Event 08H, Umask 04H
|
|
Cycle PMH is busy with a walk.
|
|
.It Li DTLB_LOAD_MISSES.STLB_HIT
|
|
.Pq Event 08H, Umask 10H
|
|
Number of cache load STLB hits.
|
|
No page walk.
|
|
.It Li INT_MISC.RECOVERY_CYCLES
|
|
.Pq Event 0DH, Umask 03H
|
|
Cycles waiting to recover after Machine Clears or JEClear.
|
|
Set Cmask = 1.
|
|
Set Edge to count occurrences
|
|
.It Li INT_MISC.RAT_STALL_CYCLES
|
|
.Pq Event 0DH, Umask 40H
|
|
Cycles RAT external stall is sent to IDQ for this thread.
|
|
.It Li UOPS_ISSUED.ANY
|
|
.Pq Event 0EH, Umask 01H
|
|
Increments each cycle the # of Uops issued by the RAT to RS.
|
|
Set Cmask = 1, Inv = 1, Any= 1 to count stalled cycles of this core.
|
|
Set Cmask = 1, Inv = 1 to count stalled cycles
|
|
.It Li FP_COMP_OPS_EXE.X87
|
|
.Pq Event 10H, Umask 01H
|
|
Counts number of X87 uops executed.
|
|
.It Li FP_COMP_OPS_EXE.SSE_FP_PACKED_DOUBLE
|
|
.Pq Event 10H, Umask 10H
|
|
Counts number of SSE* double precision FP packed uops executed.
|
|
.It Li FP_COMP_OPS_EXE.SSE_FP_SCALAR_SINGLE
|
|
.Pq Event 10H, Umask 20H
|
|
Counts number of SSE* single precision FP scalar uops executed.
|
|
.It Li FP_COMP_OPS_EXE.SSE_PACKED_SINGLE
|
|
.Pq Event 10H, Umask 40H
|
|
Counts number of SSE* single precision FP packed uops executed.
|
|
.It LiFP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE
|
|
.Pq Event 10H, Umask 80H
|
|
Counts number of SSE* double precision FP scalar uops executed.
|
|
.It Li SIMD_FP_256.PACKED_SINGLE
|
|
.Pq Event 11H, Umask 01H
|
|
Counts 256-bit packed single-precision floating-point instructions.
|
|
.It Li SIMD_FP_256.PACKED_DOUBLE
|
|
.Pq Event 11H, Umask 02H
|
|
Counts 256-bit packed double-precision floating-point instructions.
|
|
.It Li ARITH.FPU_DIV_ACTIVE
|
|
.Pq Event 14H, Umask 01H
|
|
Cycles that the divider is active, includes INT and FP.
|
|
Set 'edge =1, cmask=1' to count the number of divides.
|
|
.It Li INSTS_WRITTEN_TO_IQ.INSTS
|
|
.Pq Event 17H, Umask 01H
|
|
Counts the number of instructions written into the IQ every cycle.
|
|
.It Li L2_RQSTS.DEMAND_DATA_RD_HIT
|
|
.Pq Event 24H, Umask 01H
|
|
Demand Data Read requests that hit L2 cache.
|
|
.It Li L2_RQSTS.ALL_DEMAND_DATA_RD
|
|
.Pq Event 24H, Umask 03H
|
|
Counts any demand and L1 HW prefetch data load requests to L2.
|
|
.It Li L2_RQSTS.RFO_HITS
|
|
.Pq Event 24H, Umask 04H
|
|
Counts the number of store RFO requests that hit the L2 cache.
|
|
.It Li L2_RQSTS.RFO_MISS
|
|
.Pq Event 24H, Umask 08H
|
|
Counts the number of store RFO requests that miss the L2 cache.
|
|
.It Li L2_RQSTS.ALL_RFO
|
|
.Pq Event 24H, Umask 0CH
|
|
Counts all L2 store RFO requests.
|
|
.It Li L2_RQSTS.CODE_RD_HIT
|
|
.Pq Event 24H, Umask 10H
|
|
Number of instruction fetches that hit the L2 cache.
|
|
.It Li L2_RQSTS.CODE_RD_MISS
|
|
.Pq Event 24H, Umask 20H
|
|
Number of instruction fetches that missed the L2 cache.
|
|
.It Li L2_RQSTS.ALL_CODE_RD
|
|
.Pq Event 24H, Umask 30H
|
|
Counts all L2 code requests.
|
|
.It Li L2_RQSTS.PF_HIT
|
|
.Pq Event 24H, Umask 40H
|
|
Requests from L2 Hardware prefetcher that hit L2.
|
|
.It Li L2_RQSTS.PF_MISS
|
|
.Pq Event 24H, Umask 80H
|
|
Requests from L2 Hardware prefetcher that missed L2.
|
|
.It Li L2_RQSTS.ALL_PF
|
|
.Pq Event 24H, Umask C0H
|
|
Any requests from L2 Hardware prefetchers.
|
|
.It Li L2_STORE_LOCK_RQSTS.MISS
|
|
.Pq Event 27H, Umask 01H
|
|
RFOs that miss cache lines.
|
|
.It Li L2_STORE_LOCK_RQSTS.HIT_E
|
|
.Pq Event 27H, Umask 04H
|
|
RFOs that hit cache lines in E state.
|
|
.It Li L2_STORE_LOCK_RQSTS.HIT_M
|
|
.Pq EVENT_27H, Umask 08H
|
|
RFOs that hit cache lines in M state.
|
|
.It Li L2_STORE_LOCK_RQSTS.ALL
|
|
.Pq EVENT_27H, Umask 0FH
|
|
RFOs that access cache lines in any state.
|
|
.It Li L2_L1D_WB_RQSTS.HIT_E
|
|
.Pq Event 28H, Umask 04H
|
|
Not rejected writebacks from L1D to L2 cache lines in E state.
|
|
.It Li L2_L1D_WB_RQSTS.HIT_M
|
|
.Pq Event 28H, Umask 08H
|
|
Not rejected writebacks from L1D to L2 cache lines in M state.
|
|
.It Li LONGEST_LAT_CACHE.REFERENCE
|
|
.Pq Event 2EH, Umask 4FH
|
|
This event counts requests originating from the core that reference a cache
|
|
line in the last level cache.
|
|
.It Li LONGEST_LAT_CACHE.MISS
|
|
.Pq Event 2EH, Umask 41H
|
|
This event counts each cache miss condition for references to the last level
|
|
cache.
|
|
.It Li CPU_CLK_UNHALTED.THREAD_P
|
|
.Pq Event 3CH, Umask 00H
|
|
Counts the number of thread cycles while the thread is not in a halt state.
|
|
The thread enters the halt state when it is running the HLT instruction.
|
|
The core frequency may change from time to time due to power or thermal
|
|
throttling.
|
|
.It Li CPU_CLK_THREAD_UNHALTED.REF_XCLK
|
|
.Pq Event 3CH, Umask 01H
|
|
Increments at the frequency of XCLK (100 MHz) when not halted.
|
|
.It Li L1D_PEND_MISS.PENDING
|
|
.Pq Event 48H, Umask 01H
|
|
Increments the number of outstanding L1D misses every cycle.
|
|
Set Cmask = 1 and Edge =1 to count occurrences.
|
|
Counter 2 only; Set Cmask = 1 to count cycles.
|
|
.It Li DTLB_STORE_MISSES.MISS_CAUSES_A_WALK
|
|
.Pq Event 49H, Umask 01H Miss in all TLB levels causes an page walk of any
|
|
page size (4K/2M/4M/1G).
|
|
.It Li DTLB_STORE_MISSES.WALK_COMPLETED
|
|
.Pq Event 49H, Umask 02H
|
|
Miss in all TLB levels causes a page walk that completes of any page size
|
|
(4K/2M/4M/1G).
|
|
.It Li DTLB_STORE_MISSES.WALK_DURATION
|
|
.Pq Event 49H, Umask 04H
|
|
Cycles PMH is busy with this walk.
|
|
.It Li DTLB_STORE_MISSES.STLB_HIT
|
|
.Pq Event 49H, Umask 10H
|
|
Store operations that miss the first TLB level but hit the second and do not
|
|
cause page walks.
|
|
.It Li LOAD_HIT_PRE.SW_PF
|
|
.Pq Event 4CH, Umask 01H
|
|
Not SW-prefetch load dispatches that hit fill buffer allocated for S/W prefetch.
|
|
.It Li LOAD_HIT_PER.HW_PF
|
|
.Pq Event 4CH, Umask 02H
|
|
Not SW-prefetch load dispatches that hit fill buffer allocated for H/W prefetch.
|
|
.It Li HW_PRE_REQ.DL1_MISS
|
|
.Pq Event 4EH, Umask 02H
|
|
Hardware Prefetch requests that miss the L1D cache.
|
|
A request is being counted each time it access the cache & miss it, including
|
|
if a block is applicable or if hit the Fill Buffer for example.
|
|
This accounts for both L1 streamer and IP-based (IPP) HW prefetchers.
|
|
.It Li L1D.REPLACEMENT
|
|
.Pq Event 51H, Umask 01H
|
|
Counts the number of lines brought into the L1 data cache.
|
|
.It Li L1D.ALLOCATED_IN_M
|
|
.Pq Event 51H, Umask 02H
|
|
Counts the number of allocations of modified L1D cache lines.
|
|
.It Li L1D.EVICTION
|
|
.Pq Event 51H, Umask 04H
|
|
Counts the number of modified lines evicted from the L1 data cache due to
|
|
replacement.
|
|
.It Li L1D.ALL_M_REPLACEMENT
|
|
.Pq Event 51H, Umask 08H
|
|
Cache lines in M state evicted out of L1D due to Snoop HitM or dirty line
|
|
replacement.
|
|
.It Li PARTIAL_RAT_STALLS.FLAGS_MERGE_UOP
|
|
.Pq Event 59H, Umask 20H
|
|
Increments the number of flags-merge uops in flight each cycle.
|
|
Set Cmask = 1 to count cycles.
|
|
.It Li PARTIAL_RAT_STALLS.SLOW_LEA_WINDOW
|
|
.Pq Event 59H, Umask 40H
|
|
Cycles with at least one slow LEA uop allocated.
|
|
.It Li PARTIAL_RAT_STALLS.MUL_SINGLE_UOP
|
|
.Pq Event 59H, Umask 80H
|
|
Number of Multiply packed/scalar single precision uops allocated.
|
|
.It Li RESOURCE_STALLS2.ALL_FL_EMPTY
|
|
.Pq Event 5BH, Umask 0CH
|
|
Cycles stalled due to free list empty.
|
|
.It Li RESOURCE_STALLS2.ALL_PRF_CONTROL
|
|
.Pq Event 5BH, Umask 0FH
|
|
Cycles stalled due to control structures full for physical registers.
|
|
.It Li RESOURCE_STALLS2.BOB_FULL
|
|
.Pq Event 5BH, Umask 40H
|
|
Cycles Allocator is stalled due to Branch Order Buffer.
|
|
.It Li RESOURCE_STALLS2.OOO_RSRC
|
|
.Pq Event 5BH, Umask 4FH
|
|
Cycles stalled due to out of order resources full.
|
|
.It Li CPL_CYCLES.RING0
|
|
.Pq Event 5CH, Umask 01H
|
|
Unhalted core cycles when the thread is in ring 0.
|
|
Use Edge to count transition
|
|
.It Li CPL_CYCLES.RING123
|
|
.Pq Event 5CH, Umask 02H
|
|
Unhalted core cycles when the thread is not in ring 0.
|
|
.It Li RS_EVENTS.EMPTY_CYCLES
|
|
.Pq Event 5EH, Umask 01H
|
|
Cycles the RS is empty for the thread.
|
|
.It Li OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD
|
|
.Pq Event 60H, Umask 01H
|
|
Offcore outstanding Demand Data Read transactions in SQ to uncore.
|
|
Set Cmask=1 to count cycles.
|
|
.It Li OFFCORE_REQUESTS_OUTSTANDING.DEMAND_RFO
|
|
.Pq Event 60H, Umask 04H
|
|
Offcore outstanding RFO store transactions in SQ to uncore.
|
|
Set Cmask=1 to count cycles.
|
|
.It Li OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD
|
|
.Pq Event 60H, Umask 08H
|
|
Offcore outstanding cacheable data read transactions in SQ to uncore.
|
|
Set Cmask=1 to count cycles.
|
|
.It Li LOCK_CYCLES.SPLIT_LOCK_UC_LOCK_DURATION
|
|
.Pq Event 63H, Umask 01H
|
|
Cycles in which the L1D and L2 are locked, due to a UC lock or split lock.
|
|
.It Li LOCK_CYCLES.CACHE_LOCK_DURATION
|
|
.Pq Event 63H, Umask 02H
|
|
Cycles in which the L1D is locked.
|
|
.It Li IDQ.EMPTY
|
|
.Pq Event 79H, Umask 02H
|
|
Counts cycles the IDQ is empty.
|
|
.It Li IQD.MITE_UOPS
|
|
.Pq Event 79H, Umask 04H
|
|
Increment each cycle # of uops delivered to IDQ from MITE path.
|
|
Set Cmask = 1 to count cycles.
|
|
Can combine Umask 04H and 20H
|
|
.It Li IDQ.DSB_UOPS
|
|
.Pq Event 79H, Umask 08H
|
|
Increment each cycle.
|
|
# of uops delivered to IDQ from DSB path.
|
|
Set Cmask = 1 to count cycles.
|
|
Can combine Umask 08H and 10H
|
|
.It Li IDQ.MS_DSB_UOPS
|
|
.Pq Event 79H, Umask 10H
|
|
Increment each cycle # of uops delivered to IDQ when MS busy by DSB.
|
|
Set Cmask = 1 to count cycles MS is busy.
|
|
Set Cmask=1 and Edge=1 to count MS activations.
|
|
Can combine Umask 08H and 10H
|
|
.It Li IDQ.MS_MITE_UOPS
|
|
.Pq Event 79H, Umask 20H
|
|
Increment each cycle # of uops delivered to IDQ when MS is busy by MITE.
|
|
Set Cmask = 1 to count cycles.
|
|
Can combine Umask 04H and 20H
|
|
.It Li IDQ.MS_UOPS
|
|
.Pq Event 79H, Umask 30H
|
|
Increment each cycle # of uops delivered to IDQ from MS by either DSB or MITE.
|
|
Set Cmask = 1 to count cycles.
|
|
Can combine Umask 04H, 08H and 30H
|
|
.It Li ICACHE.MISSES
|
|
.Pq Event 80H, Umask 02H
|
|
Number of Instruction Cache, Streaming Buffer and Victim Cache Misses.
|
|
Includes UC accesses.
|
|
.It Li ITLB_MISSES.MISS_CAUSES_A_WALK
|
|
.Pq Event 85H, Umask 01H
|
|
Misses in all ITLB levels that cause page walks.
|
|
.It Li ITLB_MISSES.WALK_COMPLETED
|
|
.Pq Event 85H, Umask 02H
|
|
Misses in all ITLB levels that cause completed page walks.
|
|
.It Li ITLB_MISSES.WALK_DURATION
|
|
.Pq Event 85H, Umask 04H
|
|
Cycle PMH is busy with a walk.
|
|
.It Li ITLB_MISSES.STLB_HIT
|
|
.Pq Event 85H, Umask 10H
|
|
Number of cache load STLB hits.
|
|
No page walk.
|
|
.It Li ILD_STALL.LCP
|
|
.Pq Event 87H, Umask 01H
|
|
Stalls caused by changing prefix length of the instruction.
|
|
.It Li ILD_STALL.IQ_FULL
|
|
.Pq Event 87H, Umask 04H
|
|
Stall cycles due to IQ is full.
|
|
.It Li BR_INST_EXEC.COND
|
|
.Pq Event 88H, Umask 01H
|
|
Qualify conditional near branch instructions executed, but not necessarily
|
|
retired.
|
|
Must combine with umask 40H, 80H
|
|
.It Li BR_INST_EXEC.DIRECT_JMP
|
|
.Pq Event 88H, Umask 02H
|
|
Qualify all unconditional near branch instructions excluding calls and indirect
|
|
branches.
|
|
Must combine with umask 80H
|
|
.It Li BR_INST_EXEC.INDIRECT_JMP_NON_CALL_RET
|
|
.Pq Event 88H, Umask 04H
|
|
Qualify executed indirect near branch instructions that are not calls nor
|
|
returns.
|
|
Must combine with umask 80H
|
|
.It Li BR_INST_EXEC.RETURN_NEAR
|
|
.Pq Event 88H, Umask 08H
|
|
Qualify indirect near branches that have a return mnemonic.
|
|
Must combine with umask 80H
|
|
.It Li BR_INST_EXEC.DIRECT_NEAR_CALL
|
|
.Pq Event 88H, Umask 10H
|
|
Qualify unconditional near call branch instructions, excluding non call branch,
|
|
executed.
|
|
Must combine with umask 80H
|
|
.It Li BR_INST_EXEC.INDIRECT_NEAR_CALL
|
|
.Pq Event 88H, Umask 20H
|
|
Qualify indirect near calls, including both register and memory indirect,
|
|
executed.
|
|
Must combine with umask 80H
|
|
.It Li BR_INST_EXEC.NONTAKEN
|
|
.Pq Event 88H, Umask 40H
|
|
Qualify non-taken near branches executed.
|
|
Applicable to umask 01H only
|
|
.It Li BR_INST_EXEC.TAKEN
|
|
.Pq Event 88H, Umask 80H
|
|
Qualify taken near branches executed.
|
|
Must combine with 01H,02H, 04H, 08H, 10H, 20H
|
|
.It Li BR_INST_EXE.ALL_BRANCHES
|
|
.Pq Event 88H, Umask FFH
|
|
Counts all near executed branches (not necessarily retired).
|
|
.It Li BR_MISP_EXEC.COND
|
|
.Pq Event 89H, Umask 01H
|
|
Qualify conditional near branch instructions mispredicted.
|
|
Must combine with umask 40H, 80H
|
|
.It Li BR_MISP_EXEC.INDIRECT_JMP_NON_CALL_RET
|
|
.Pq Event 89H, Umask 04H
|
|
Qualify mispredicted indirect near branch instructions that are not calls nor
|
|
returns.
|
|
Must combine with umask 80H
|
|
.It Li BR_MISP_EXEC.RETURN_NEAR
|
|
.Pq Event 89H, Umask 08H
|
|
Qualify mispredicted indirect near branches that have a return mnemonic.
|
|
Must combine with umask 80H
|
|
.It Li BR_MISP_EXEC.DIRECT_NEAR_CALL
|
|
.Pq Event 89H, Umask 10H
|
|
Qualify mispredicted unconditional near call branch instructions, excluding non
|
|
call branch, executed.
|
|
Must combine with umask 80H
|
|
.It Li BR_MISP_EXEC.INDIRECT_NEAR_CALL
|
|
.Pq Event 89H, Umask 20H
|
|
Qualify mispredicted indirect near calls, including both register and memory
|
|
indirect, executed.
|
|
Must combine with umask 80H
|
|
.It Li BR_MISP_EXEC.NONTAKEN
|
|
.Pq Event 89H, Umask 40H
|
|
Qualify mispredicted non-taken near branches executed.
|
|
Applicable to umask 01H only
|
|
.It Li BR_MISP_EXEC.TAKEN
|
|
.Pq Event 89H, Umask 80H
|
|
Qualify mispredicted taken near branches executed.
|
|
Must combine with 01H,02H, 04H, 08H, 10H, 20H
|
|
.It Li BR_MISP_EXEC.ALL_BRANCHES
|
|
.Pq Event 89H, Umask FFH
|
|
Counts all near executed branches (not necessarily retired).
|
|
.It Li IDQ_UOPS_NOT_DELIVERED.CORE
|
|
.Pq Event 9CH, Umask 01H
|
|
Count number of non-delivered uops to RAT per thread.
|
|
Use Cmask to qualify uop b/w
|
|
.It Li UOPS_DISPATCHED_PORT.PORT_0
|
|
.Pq Event A1H, Umask 01H
|
|
Cycles which a Uop is dispatched on port 0.
|
|
.It Li UOPS_DISPATCHED_PORT.PORT_1
|
|
.Pq Event A1H, Umask 02H
|
|
Cycles which a Uop is dispatched on port 1.
|
|
.It Li UOPS_DISPATCHED_PORT.PORT_2_LD
|
|
.Pq Event A1H, Umask 04H
|
|
Cycles which a load uop is dispatched on port 2.
|
|
.It Li UOPS_DISPATCHED_PORT.PORT_2_STA
|
|
.Pq Event A1H, Umask 08H
|
|
Cycles which a store address uop is dispatched on port 2.
|
|
.It Li UOPS_DISPATCHED_PORT.PORT_2
|
|
.Pq Event A1H, Umask 0CH
|
|
Cycles which a Uop is dispatched on port 2.
|
|
.It Li UOPS_DISPATCHED_PORT.PORT_3_LD
|
|
.Pq Event A1H, Umask 10H
|
|
Cycles which a load uop is dispatched on port 3.
|
|
.It Li UOPS_DISPATCHED_PORT.PORT_3_STA
|
|
.Pq Event A1H, Umask 20H
|
|
Cycles which a store address uop is dispatched on port 3.
|
|
.It Li UOPS_DISPATCHED_PORT.PORT_3
|
|
.Pq Event A1H, Umask 30H
|
|
.Pq Cycles which a Uop is dispatched on port 3.
|
|
.It Li UOPS_DISPATCHED_PORT.PORT_4
|
|
.Pq Event A1H, Umask 40H
|
|
Cycles which a Uop is dispatched on port 4.
|
|
.It Li UOPS_DISPATCHED_PORT.PORT_5
|
|
.Pq Event A1H, Umask 80H
|
|
Cycles which a Uop is dispatched on port 5.
|
|
.It Li RESOURCE_STALLS.ANY
|
|
.Pq Event A2H, Umask 01H
|
|
Cycles Allocation is stalled due to Resource Related reason.
|
|
.It Li RESOURCE_STALLS.LB
|
|
.Pq Event A2H, Umask 02H
|
|
Counts the cycles of stall due to lack of load buffers.
|
|
.It Li RESOURCE_STALLS.LB
|
|
.Pq Event A2H, Umask 04H
|
|
Cycles stalled due to no eligible RS entry available.
|
|
.It Li RESOURCE_STALLS.SB
|
|
.Pq Event A2H, Umask 08H
|
|
Cycles stalled due to no store buffers available.
|
|
(not including draining form sync)
|
|
.It Li RESOURCE_STALLS.ROB
|
|
.Pq Event A2H, Umask 10H
|
|
Cycles stalled due to re-order buffer full.
|
|
.It Li RESOURCE_STALLS.FCSW
|
|
.Pq Event A2H, Umask 20H
|
|
Cycles stalled due to writing the FPU control word.
|
|
.It Li RESOURCE_STALLS.MXCSR
|
|
.Pq Event A2H, Umask 40H
|
|
Cycles stalled due to the MXCSR register rename occurring to close to a previous
|
|
MXCSR rename.
|
|
.It Li RESOURCE_STALLS.OTHER
|
|
.Pq Event A2H, Umask 80H
|
|
Cycles stalled while execution was stalled due to other resource issues.
|
|
.It Li DSB2MITE_SWITCHES.COUNT
|
|
.Pq Event ABH, Umask 01H
|
|
Number of DSB to MITE switches.
|
|
.It Li DSB2MITE_SWITCHES.PENALTY_CYCLES
|
|
.Pq Event ABH, Umask 02H
|
|
Cycles DSB to MITE switches caused delay.
|
|
.It Li DSB_FILL.OTHER_CANCEL
|
|
.Pq Event ACH, Umask 02H
|
|
Cases of cancelling valid DSB fill not because of exceeding way limit.
|
|
.It Li DSB_FILL.EXCEED_DSB_LINES
|
|
.Pq Event ACH, Umask 08H
|
|
DSB Fill encountered > 3 DSB lines.
|
|
.It Li DSB_FILL.ALL_CANCEL
|
|
.Pq Event ACH, Umask 0AH
|
|
Cases of cancelling valid Decode Stream Buffer (DSB) fill not because of exceeding
|
|
way limit.
|
|
.It Li ITLB.ITLB_FLUSH
|
|
.Pq Event AEH, Umask 01H
|
|
Counts the number of ITLB flushes, includes 4k/2M/4M pages.
|
|
.It Li OFFCORE_REQUESTS.DEMAND_DATA_RD
|
|
.Pq Event B0H, Umask 01H
|
|
Demand data read requests sent to uncore.
|
|
.It Li OFFCORE_REQUESTS.DEMAND_RFO
|
|
.Pq Event B0H, Umask 04H
|
|
Demand RFO read requests sent to uncore, including regular RFOs, locks, ItoM.
|
|
.It Li OFFCORE_REQUESTS.ALL_DATA_RD
|
|
.Pq Event B0H, Umask 08H
|
|
Data read requests sent to uncore (demand and prefetch).
|
|
.It Li UOPS_DISPATCHED.THREAD
|
|
.Pq Event B1H, Umask 01H
|
|
Counts total number of uops to be dispatched per-thread each cycle.
|
|
Set Cmask = 1, INV =1 to count stall cycles.
|
|
.It Li UOPS_DISPATCHED.CORE
|
|
.Pq Event B1H, Umask 02H
|
|
Counts total number of uops to be dispatched per-core each cycle.
|
|
Do not need to set ANY
|
|
.It Li OFFCORE_REQUESTS_BUFFER.SQ_FULL
|
|
.Pq Event B2H, Umask 01H
|
|
Offcore requests buffer cannot take more entries for this thread core.
|
|
.It Li AGU_BYPASS_CANCEL.COUNT
|
|
.Pq Event B6H, Umask 01H
|
|
Counts executed load operations with all the following traits: 1. addressing
|
|
of the format [base + offset], 2. the offset is between 1 and 2047, 3. the
|
|
address specified in the base register is in one page and the address
|
|
[base+offset] is in another page.
|
|
.It Li OFF_CORE_RESPONSE_0
|
|
.Pq Event B7H, Umask 01H
|
|
Off-core Response Performance Monitoring; PMC0 only.
|
|
Requires programming MSR 01A6H
|
|
.It Li OFF_CORE_RESPONSE_1
|
|
.Pq Event BBH, Umask 01H
|
|
Off-core Response Performance Monitoring. PMC3 only.
|
|
Requires programming MSR 01A7H
|
|
.It Li TLB_FLUSH.DTLB_THREAD
|
|
.Pq Event BDH, Umask 01H
|
|
DTLB flush attempts of the thread-specific entries.
|
|
.It Li TLB_FLUSH.STLB_ANY
|
|
.Pq Event BDH, Umask 20H
|
|
Count number of STLB flush attempts.
|
|
.It Li L1D_BLOCKS.BANK_CONFLICT_CYCLES
|
|
.Pq Event BFH, Umask 05H
|
|
Cycles when dispatched loads are cancelled due to L1D bank conflicts with other
|
|
load ports.
|
|
cmask=1
|
|
.It Li INST_RETIRED.ANY_P
|
|
.Pq Event C0H, Umask 00H
|
|
Number of instructions at retirement.
|
|
.It Li INST_RETIRED.PREC_DIST
|
|
.Pq Event C0H, Umask 01H
|
|
Precise instruction retired event with HW to reduce effect of PEBS shadow in IP
|
|
distribution PMC1 only; Must quiesce other PMCs.
|
|
.It Li INST_RETIRED.X87
|
|
.Pq Event C0H, Umask 02H
|
|
X87 instruction retired event.
|
|
.It Li OTHER_ASSISTS.ITLB_MISS_RETIRED
|
|
.Pq Event C1H, Umask 02H
|
|
Instructions that experienced an ITLB miss.
|
|
.It Li OTHER_ASSISTS.AVX_STORE
|
|
.Pq Event C1H, Umask 08H
|
|
Number of assists associated with 256-bit AVX store operations.
|
|
.It Li OTHER_ASSISTS.AVX_TO_SSE
|
|
.Pq Event C1H, Umask 10H
|
|
Number of transitions from AVX256 to legacy SSE when penalty applicable.
|
|
.It Li OTHER_ASSISTS.SSE_TO_AVX
|
|
.Pq Event C1H, Umask 20H
|
|
Number of transitions from SSE to AVX-256 when penalty applicable.
|
|
.It Li UOPS_RETIRED.ALL
|
|
.Pq Event C2H, Umask 01H
|
|
Counts the number of micro-ops retired.
|
|
Use cmask=1 and invert to count active cycles or stalled cycles.
|
|
.It Li UOPS_RETIRED.RETIRE_SLOTS
|
|
.Pq Event C2H, Umask 02H
|
|
Counts the number of retirement slots used each cycle.
|
|
.It Li MACHINE_CLEARS.MEMORY_ORDERING
|
|
.Pq Event C3H, Umask 02H
|
|
Counts the number of machine clears due to memory order conflicts.
|
|
.It Li MACHINE_CLEARS.SMC
|
|
.Pq Event C3H, Umask 04H
|
|
Counts the number of times that a program writes to a code section.
|
|
.It Li MACHINE_CLEARS.MASKMOV
|
|
.Pq Event C3H, Umask 20H
|
|
Counts the number of executed AVX masked load operations that refer to an
|
|
illegal address range with the mask bits set to 0.
|
|
.It Li BR_INST_RETIRED.ALL_BRANCH
|
|
.Pq Event C4H, Umask 00H
|
|
Branch instructions at retirement.
|
|
.It Li BR_INST_RETIRED.CONDITIONAL
|
|
.Pq Event C4H, Umask 01H
|
|
Counts the number of conditional branch instructions retired.
|
|
.It Li BR_INST_RETIRED.NEAR_CALL
|
|
.Pq Event C4H, Umask 02H
|
|
Direct and indirect near call instructions retired.
|
|
.It Li BR_INST_RETIRED.ALL_BRANCHES
|
|
.Pq Event C4H, Umask 04H
|
|
Counts the number of branch instructions retired.
|
|
.It Li BR_INST_RETIRED.NEAR_RETURN
|
|
.Pq Event C4H, Umask 08H
|
|
Counts the number of near return instructions retired.
|
|
.It Li BR_INST_RETIRED.NOT_TAKEN
|
|
.Pq Event C4H, Umask 10H
|
|
Counts the number of not taken branch instructions retired.
|
|
.It Li BR_INST_RETIRED.NEAR_TAKEN
|
|
.Pq Event C4H, Umask 20H
|
|
Number of near taken branches retired.
|
|
.It Li BR_INST_RETIRED.FAR_BRANCH
|
|
.Pq Event C4H, Umask 40H
|
|
Number of far branches retired.
|
|
.It Li BR_MISP_RETIRED.ALL_BRANCHES
|
|
.Pq Event C5H, Umask 00H
|
|
Mispredicted branch instructions at retirement.
|
|
.It Li BR_MISP_RETIRED.CONDITIONAL
|
|
.Pq Event C5H, Umask 01H
|
|
Mispredicted conditional branch instructions retired.
|
|
.It Li BR_MISP_RETIRED.NEAR_CALL
|
|
.Pq Event C5H, Umask 02H
|
|
Direct and indirect mispredicted near call instructions retired.
|
|
.It Li BR_MISP_RETIRED.ALL_BRANCH
|
|
.Pq Event C5H, Umask 04H
|
|
Mispredicted macro branch instructions retired.
|
|
.It Li BR_MISP_RETIRED.NOT_TAKEN
|
|
.Pq Event C5H, Umask 10H
|
|
Mispredicted not taken branch instructions retired.
|
|
.It Li BR_MISP_RETIRED.TAKEN
|
|
.Pq Event C5H, Umask 20H
|
|
Mispredicted taken branch instructions retired.
|
|
.It Li FP_ASSIST.X87_OUTPUT
|
|
.Pq Event CAH, Umask 02H
|
|
Number of X87 assists due to output value.
|
|
.It Li FP_ASSIST.X87_INPUT
|
|
.Pq Event CAH, Umask 04H
|
|
Number of X87 assists due to input value.
|
|
.It Li FP_ASSIST.SIMD_OUTPUT
|
|
.Pq Event CAH, Umask 08H
|
|
Number of SIMD FP assists due to Output values.
|
|
.It Li FP_ASSIST.SIMD_INPUT
|
|
.Pq Event CAH, Umask 10H
|
|
Number of SIMD FP assists due to input values.
|
|
.It Li FP_ASSIST.ANY
|
|
.Pq Event CAH, Umask 1EH
|
|
Cycles with any input/output SSE* or FP assists.
|
|
.It Li ROB_MISC_EVENTS.LBR_INSERTS
|
|
.Pq Event CCH, Umask 20H
|
|
Count cases of saving new LBR records by hardware.
|
|
.It Li MEM_TRANS_RETIRED.LOAD_LATENCY
|
|
.Pq Event CDH, Umask 01H
|
|
Sample loads with specified latency threshold.
|
|
PMC3 only.
|
|
Specify threshold in MSR 0x3F6.
|
|
.It Li MEM_TRANS_RETIRED.PRECISE_STORE
|
|
.Pq Event CDH, Umask 02H
|
|
Sample stores and collect precise store operation via PEBS record.
|
|
PMC3 only.
|
|
.It Li MEM_UOP_RETIRED.LOADS
|
|
.Pq Event D0H, Umask 01H
|
|
Qualify retired memory uops that are loads.
|
|
Combine with umask 10H, 20H, 40H, 80H.
|
|
.It Li MEM_UOP_RETIRED.STORES
|
|
.Pq Event D0H, Umask 02H
|
|
Qualify retired memory uops that are stores.
|
|
Combine with umask 10H, 20H, 40H, 80H.
|
|
.It Li MEM_UOP_RETIRED.STLB_MISS
|
|
.Pq Event D0H, Umask 10H
|
|
Qualify retired memory uops with STLB miss.
|
|
Must combine with umask 01H, 02H, to produce counts.
|
|
.It Li MEM_UOP_RETIRED.LOCK
|
|
.Pq Event D0H, Umask 20H
|
|
Qualify retired memory uops with lock.
|
|
Must combine with umask 01H, 02H, to produce counts.
|
|
.It Li MEM_UOP_RETIRED.SPLIT
|
|
.Pq Event D0H, Umask 40H
|
|
Qualify retired memory uops with line split.
|
|
Must combine with umask 01H, 02H, to produce counts.
|
|
.It Li MEM_UOP_RETIRED_ALL
|
|
.Pq Event D0H, Umask 80H
|
|
Qualify any retired memory uops.
|
|
Must combine with umask 01H, 02H, to produce counts.
|
|
.It Li MEM_LOAD_UOPS_RETIRED.L1_HIT
|
|
.Pq Event D1H, Umask 01H
|
|
Retired load uops with L1 cache hits as data sources.
|
|
Must combine with umask 01H, 02H, to produce counts.
|
|
.It Li MEM_LOAD_UOPS_RETIRED.L2_HIT
|
|
.Pq Event D1H, Umask 02H
|
|
Retired load uops with L2 cache hits as data sources.
|
|
.It Li MEM_LOAD_UOPS_RETIRED.LLC_HIT
|
|
.Pq Event D1H, Umask 04H
|
|
Retired load uops which data sources were data hits in LLC without snoops
|
|
required.
|
|
.It Li MEM_LOAD_UOPS_RETIRED.HIT_LFB
|
|
.Pq Event D1H, Umask 40H
|
|
Retired load uops which data sources were load uops missed L1 but hit FB due
|
|
to preceding miss to the same cache line with data not ready.
|
|
.It Li MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS
|
|
.Pq Event D2H, Umask 01H
|
|
Retired load uops which data sources were LLC hit and cross-core snoop missed in
|
|
on-pkg core cache.
|
|
.It Li MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT
|
|
.Pq Event D2H, Umask 02H
|
|
Retired load uops which data sources were LLC and cross-core snoop hits in
|
|
on-pkg core cache.
|
|
.It Li MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM
|
|
.Pq Event D2H, Umask 04H
|
|
Retired load uops which data sources were HitM responses from shared LLC.
|
|
.It Li MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_NONE
|
|
.Pq Event D2H, Umask 08H
|
|
Retired load uops which data sources were hits in LLC without snoops required.
|
|
.It Li MEM_LOAD_UOPS_LLC_HIT_RETIRED.LLC_MISS
|
|
.Pq Event D4H, Umask 02H
|
|
Retired load uops with unknown information as data source in cache serviced the load.
|
|
.It Li L2_TRANS.DEMAND_DATA_RD
|
|
.Pq Event F0H, Umask 01H
|
|
Demand Data Read requests that access L2 cache.
|
|
.It Li L2_TRANS.RF0
|
|
.Pq Event F0H, Umask 02H
|
|
RFO requests that access L2 cache.
|
|
.It Li L2_TRANS.CODE_RD
|
|
.Pq Event F0H, Umask 04H
|
|
L2 cache accesses when fetching instructions.
|
|
.It Li L2_TRANS.ALL_PF
|
|
.Pq Event F0H, Umask 08H
|
|
L2 or LLC HW prefetches that access L2 cache.
|
|
.It Li L2_TRANS.L1D_WB
|
|
.Pq Event F0H, Umask 10H
|
|
L1D writebacks that access L2 cache.
|
|
.It Li L2_TRANS.L2_FILL
|
|
.Pq Event F0H, Umask 20H
|
|
L2 fill requests that access L2 cache.
|
|
.It Li L2_TRANS.L2_WB
|
|
.Pq Event F0H, Umask 40H
|
|
L2 writebacks that access L2 cache.
|
|
.It Li L2_TRANS.ALL_REQUESTS
|
|
.Pq Event F0H, Umask 80H
|
|
Transactions accessing L2 pipe.
|
|
.It Li L2_LINES_IN.I
|
|
.Pq Event F1H, Umask 01H
|
|
L2 cache lines in I state filling L2.
|
|
Counting does not cover rejects.
|
|
.It Li L2_LINES_IN.S
|
|
.Pq Event F1H, Umask 02H
|
|
L2 cache lines in S state filling L2.
|
|
Counting does not cover rejects.
|
|
.It Li L2_LINES_IN.E
|
|
.Pq Event F1H, Umask 04H
|
|
L2 cache lines in E state filling L2.
|
|
Counting does not cover rejects.
|
|
.It Li L2_LINES-IN.ALL
|
|
.Pq Event F1H, Umask 07H
|
|
L2 cache lines filling L2.
|
|
Counting does not cover rejects.
|
|
.It Li L2_LINES_OUT.DEMAND_CLEAN
|
|
.Pq Event F2H, Umask 01H
|
|
Clean L2 cache lines evicted by demand.
|
|
.It Li L2_LINES_OUT.DEMAND_DIRTY
|
|
.Pq Event F2H, Umask 02H
|
|
Dirty L2 cache lines evicted by demand.
|
|
.It Li L2_LINES_OUT.PF_CLEAN
|
|
.Pq Event F2H, Umask 04H
|
|
Clean L2 cache lines evicted by L2 prefetch.
|
|
.It Li L2_LINES_OUT.PF_DIRTY
|
|
.Pq Event F2H, Umask 08H
|
|
Dirty L2 cache lines evicted by L2 prefetch.
|
|
.It Li L2_LINES_OUT.DIRTY_ALL
|
|
.Pq Event F2H, Umask 0AH
|
|
Dirty L2 cache lines filling the L2.
|
|
Counting does not cover rejects.
|
|
.It Li SQ_MISC.SPLIT_LOCK
|
|
.Pq Event F4H, Umask 10H
|
|
Split locks in SQ.
|
|
.El
|
|
.Sh SEE ALSO
|
|
.Xr pmc 3 ,
|
|
.Xr pmc.atom 3 ,
|
|
.Xr pmc.core 3 ,
|
|
.Xr pmc.corei7 3 ,
|
|
.Xr pmc.corei7uc 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.sandybridgeuc 3 ,
|
|
.Xr pmc.sandybridgexeon 3 ,
|
|
.Xr pmc.soft 3 ,
|
|
.Xr pmc.tsc 3 ,
|
|
.Xr pmc.ucf 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 .
|
|
.Sh AUTHORS
|
|
The
|
|
.Lb libpmc
|
|
library was written by
|
|
.An "Joseph Koshy"
|
|
.Aq jkoshy@FreeBSD.org.
|
|
The support for the Sandy Bridge
|
|
microarchitecture was written by
|
|
.An "Davide Italiano"
|
|
.Aq davide@FreeBSD.org .
|