.\" Copyright (c) 2013 Hiren Panchasara <hiren.panchasara@gmail.com>
.\" 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 March 22, 2013
.Dt PMC.HASWELL 3
.Os
.Sh NAME
.Nm pmc.haswell
.Nd measurement events for
.Tn Intel
.Tn Haswell
family CPUs
.Sh LIBRARY
.Lb libpmc
.Sh SYNOPSIS
.In pmc.h
.Sh DESCRIPTION
.Tn Intel
.Tn "Haswell"
CPUs contain PMCs conforming to version 2 of the
.Tn Intel
performance measurement architecture.
These CPUs may contain up to two 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.
.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 Haswell PMCs are documented in
.Rs
.%B "Intel(R) 64 and IA-32 Architectures Software Developer's Manual"
.%T "Combined Volumes: 1, 2A, 2B, 2C, 3A, 3B and 3C"
.%N "Order Number: 325462-045US"
.%D January 2013
.%Q "Intel Corporation"
.Re
.Ss HASWELL FIXED FUNCTION PMCS
These PMCs and their supported events are documented in
.Xr pmc.iaf 3 .
.Ss HASWELL 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 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_SNP_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)
Haswell programmable PMCs support the following events:
.Bl -tag -width indent
.It Li LD_BLOCKS.STORE_FORWARD
.Pq Event 03H , Umask 02H
Loads blocked by overlapping with store buffer that
cannot be forwarded.
.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 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_4K
.Pq Event 08H , Umask 02H
Completed page walks due to demand load misses
that caused 4K page walks in any TLB levels.
.It Li DTLB_LOAD_MISSES.WALK_COMPLETED_2M_4K
.Pq Event 08H , Umask 02H
Completed page walks due to demand load misses
that caused 2M/4M page walks in any TLB levels.
.It Li DTLB_LOAD_MISSES.WALK_COMPLETED
.Pq Event 08H , Umask 0EH
Completed page walks in any TLB of any page size
due to demand load misses
.It Li DTLB_LOAD_MISSES.WALK_DURATION
.Pq Event 08H , Umask 10H
Cycle PMH is busy with a walk.
.It Li DTLB_LOAD_MISSES.STLB_HIT_4K
.Pq Event 08H , Umask 20H
Load misses that missed DTLB but hit STLB (4K).
.It Li DTLB_LOAD_MISSES.STLB_HIT_2M
.Pq Event 08H , Umask 40H
Load misses that missed DTLB but hit STLB (2M).
.It Li DTLB_LOAD_MISSES.STLB_HIT
.Pq Event 08H , Umask 60H
Number of cache load STLB hits. No page walk.
.It Li DTLB_LOAD_MISSES.PDE_CACHE_MISS
.Pq Event 08H , Umask 80H
DTLB demand load misses with low part of linear-to-
physical address translation missed
.It Li INT_MISC.RECOVERY_CYCLES
.Pq Event 0DH , Umask 03H
Cycles waiting to recover after Machine Clears
except JEClear. Set Cmask= 1.
.It Li UOPS_ISSUED.ANY
.Pq Event 0EH , Umask 01H
ncrements each cycle the # of Uops issued by the
RAT to RS.
Set Cmask = 1, Inv = 1, Any= 1to count stalled cycles
of this core.
.It Li UOPS_ISSUED.FLAGS_MERGE
.Pq Event 0EH , Umask 10H
Number of flags-merge uops allocated. Such uops
adds delay.
.It Li UOPS_ISSUED.SLOW_LEA
.Pq Event 0EH , Umask 20H
Number of slow LEA or similar uops allocated. Such
uop has 3 sources (e.g. 2 sources + immediate)
regardless if as a result of LEA instruction or not.
.It Li UOPS_ISSUED.SiNGLE_MUL
.Pq Event 0EH , Umask 40H
Number of multiply packed/scalar single precision
uops allocated.
.It Li L2_RQSTS.DEMAND_DATA_RD_MISS
.Pq Event 24H , Umask 21H
Demand Data Read requests that missed L2, no
rejects.
.It Li L2_RQSTS.DEMAND_DATA_RD_HIT
.Pq Event 24H , Umask 41H
Demand Data Read requests that hit L2 cache.
.It Li L2_RQSTS.ALL_DEMAND_DATA_RD
.Pq Event 24H , Umask E1H
Counts any demand and L1 HW prefetch data load
requests to L2.
.It Li L2_RQSTS.RFO_HIT
.Pq Event 24H , Umask 42H
Counts the number of store RFO requests that hit
the L2 cache.
.It Li L2_RQSTS.RFO_MISS
.Pq Event 24H , Umask 22H
Counts the number of store RFO requests that miss
the L2 cache.
.It Li L2_RQSTS.ALL_RFO
.Pq Event 24H , Umask E2H
Counts all L2 store RFO requests.
.It Li L2_RQSTS.CODE_RD_HIT
.Pq Event 24H , Umask 44H
Number of instruction fetches that hit the L2 cache.
.It Li L2_RQSTS.CODE_RD_MISS
.Pq Event 24H , Umask 24H
Number of instruction fetches that missed the L2
cache.
.It Li L2_RQSTS.ALL_DEMAND_MISS
.Pq Event 24H , Umask 27H
Demand requests that miss L2 cache.
.It Li L2_RQSTS.ALL_DEMAND_REFERENCES
.Pq Event 24H , Umask E7H
Demand requests to L2 cache.
.It Li L2_RQSTS.ALL_CODE_RD
.Pq Event 24H , Umask E4H
Counts all L2 code requests.
.It Li L2_RQSTS.L2_PF_HIT
.Pq Event 24H , Umask 50H
Counts all L2 HW prefetcher requests that hit L2.
.It Li L2_RQSTS.L2_PF_MISS
.Pq Event 24H , Umask 30H
Counts all L2 HW prefetcher requests that missed
L2.
.It Li L2_RQSTS.ALL_PF
.Pq Event 24H , Umask F8H
Counts all L2 HW prefetcher requests.
.It Li L2_RQSTS.MISS
.Pq Event 24H , Umask 3FH
All requests that missed L2.
.It Li L2_RQSTS.REFERENCES
.Pq Event 24H , Umask FFH
All requests to L2 cache.
.It Li L2_DEMAND_RQSTS.WB_HIT
.Pq Event 27H , Umask 50H
Not rejected writebacks that hit L2 cache
.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 Cmaks = 1 and Edge =1 to count
occurrences.
.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_4K
.Pq Event 49H , Umask 02H
Completed page walks due to store misses in one or
more TLB levels of 4K page structure.
.It Li DTLB_STORE_MISSES.WALK_COMPLETED_2M_4M
.Pq Event 49H , Umask 04H
Completed page walks due to store misses in one or
more TLB levels of 2M/4M page structure.
.It Li DTLB_STORE_MISSES.WALK_COMPLETED
.Pq Event 49H , Umask 0EH
Completed page walks due to store miss in any TLB
levels of any page size (4K/2M/4M/1G).
.It Li DTLB_STORE_MISSES.WALK_DURATION
.Pq Event 49H , Umask 10H
Cycles PMH is busy with this walk.
.It Li DTLB_STORE_MISSES.STLB_HIT_4K
.Pq Event 49H , Umask 20H
Store misses that missed DTLB but hit STLB (4K).
.It Li DTLB_STORE_MISSES.STLB_HIT_2M
.Pq Event 49H , Umask 40H
Store misses that missed DTLB but hit STLB (2M).
.It Li DTLB_STORE_MISSES.STLB_HIT
.Pq Event 49H , Umask 60H
Store operations that miss the first TLB level but hit
the second and do not cause page walks.
.It Li DTLB_STORE_MISSES.PDE_CACHE_MISS
.Pq Event 49H , Umask 80H
DTLB store misses with low part of linear-to-physical
address translation missed.
.It Li LOAD_HIT_PRE.SW_PF
.Pq Event 4CH , Umask 01H
Non-SW-prefetch load dispatches that hit fill buffer
allocated for S/W prefetch.
.It Li LOAD_HIT_PRE.HW_PF
.Pq Event 4CH , Umask 02H
Non-SW-prefetch load dispatches that hit fill buffer
allocated for H/W prefetch.
.It Li L1D.REPLACEMENT
.Pq Event 51H , Umask 01H
Counts the number of lines brought into the L1 data
cache.
.It Li MOVE_ELIMINATION.INT_NOT_ELIMINATED
.Pq Event 58H , Umask 04H
Number of integer Move Elimination candidate uops
that were not eliminated.
.It Li MOVE_ELIMINATION.SMID_NOT_ELIMINATED
.Pq Event 58H , Umask 08H
Number of SIMD Move Elimination candidate uops
that were not eliminated.
.It Li MOVE_ELIMINATION.INT_ELIMINATED
.Pq Event 58H , Umask 01H
Unhalted core cycles when the thread is in ring 0.
.It Li MOVE_ELIMINATION.SMID_ELIMINATED
.Pq Event 58H , Umask 02H
Number of SIMD Move Elimination candidate uops
that were eliminated.
.It Li CPL_CYCLES.RING0
.Pq Event 5CH , Umask 02H
Unhalted core cycles when the thread is in ring 0.
.It Li CPL_CYCLES.RING123
.Pq Event 5CH , Umask 01H
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_CORE_RD
.Pq Event 60H , Umask 02H
Offcore outstanding Demand code 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 IDQ.MITE_UOPS
.Pq Event 79H , Umask 04H
Increment each cycle # of uops delivered to IDQ from
MITE path.
Set Cmask = 1 to count cycles.
.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.
.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. Add Edge=1 to count # of delivery.
.It Li IDQ.MS_MITE_UOPS
.Pq Event 79H , Umask 20H
ncrement each cycle # of uops delivered to IDQ
when MS_busy by MITE. Set Cmask = 1 to count
cycles.
.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.
.It Li IDQ.ALL_DSB_CYCLES_ANY_UOPS
.Pq Event 79H , Umask 18H
Counts cycles DSB is delivered at least one uops. Set
Cmask = 1.
.It Li IDQ.ALL_DSB_CYCLES_4_UOPS
.Pq Event 79H , Umask 18H
Counts cycles DSB is delivered four uops. Set Cmask
=4.
.It Li IDQ.ALL_MITE_CYCLES_ANY_UOPS
.Pq Event 79H , Umask 24H
Counts cycles MITE is delivered at least one uops. Set
Cmask = 1.
.It Li IDQ.ALL_MITE_CYCLES_4_UOPS
.Pq Event 79H , Umask 24H
Counts cycles MITE is delivered four uops. Set Cmask
=4.
.It Li IDQ.MITE_ALL_UOPS
.Pq Event 79H , Umask 3CH
# of uops delivered to IDQ from any path.
.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 ITLB that causes a page walk of any page
size.
.It Li ITLB_MISSES.WALK_COMPLETED_4K
.Pq Event 85H , Umask 02H
Completed page walks due to misses in ITLB 4K page
entries.
.It Li TLB_MISSES.WALK_COMPLETED_2M_4M
.Pq Event 85H , Umask 04H
Completed page walks due to misses in ITLB 2M/4M
page entries.
.It Li ITLB_MISSES.WALK_COMPLETED
.Pq Event 85H , Umask 0EH
Completed page walks in ITLB of any page size.
.It Li ITLB_MISSES.WALK_DURATION
.Pq Event 85H , Umask 10H
Cycle PMH is busy with a walk.
.It Li ITLB_MISSES.STLB_HIT_4K
.Pq Event 85H , Umask 20H
ITLB misses that hit STLB (4K).
.It Li ITLB_MISSES.STLB_HIT_2M
.Pq Event 85H , Umask 40H
ITLB misses that hit STLB (2K).
.It Li ITLB_MISSES.STLB_HIT
.Pq Event 85H , Umask 60H
TLB misses that hit STLB. 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.NONTAKEN_COND
.Pq Event 88H , Umask 41H
Count conditional near branch instructions that were executed (but not
necessarily retired) and not taken.
.It Li BR_INST_EXEC.TAKEN_COND
.Pq Event 88H , Umask 81H
Count conditional near branch instructions that were executed (but not
necessarily retired) and taken.
.It Li BR_INST_EXEC.DIRECT_JMP
.Pq Event 88H , Umask 82H
Count all unconditional near branch instructions excluding calls and
indirect branches.
.It Li BR_INST_EXEC.INDIRECT_JMP_NON_CALL_RET
.Pq Event 88H , Umask 84H
Count executed indirect near branch instructions that are not calls nor
returns.
.It Li BR_INST_EXEC.RETURN_NEAR
.Pq Event 88H , Umask 88H
Count indirect near branches that have a return mnemonic.
.It Li BR_INST_EXEC.DIRECT_NEAR_CALL
.Pq Event 88H , Umask 90H
Count unconditional near call branch instructions, excluding non call
branch, executed.
.It Li BR_INST_EXEC.INDIRECT_NEAR_CALL
.Pq Event 88H , Umask A0H
Count indirect near calls, including both register and memory indirect,
executed.
.It Li BR_INST_EXEC.ALL_BRANCHES
.Pq Event 88H , Umask FFH
Counts all near executed branches (not necessarily retired).
.It Li BR_MISP_EXEC.NONTAKEN_COND
.Pq Event 89H , Umask 41H
Count conditional near branch instructions mispredicted as nontaken.
.It Li BR_MISP_EXEC.TAKEN_COND
.Pq Event 89H , Umask 81H
Count conditional near branch instructions mispredicted as taken.
.It Li BR_MISP_EXEC.INDIRECT_JMP_NON_CALL_RET
.Pq Event 89H , Umask 84H
Count mispredicted indirect near branch instructions that are not calls
nor returns.
.It Li BR_MISP_EXEC.RETURN_NEAR
.Pq Event 89H , Umask 88H
Count mispredicted indirect near branches that have a return mnemonic.
.It Li BR_MISP_EXEC.DIRECT_NEAR_CALL
.Pq Event 89H , Umask 90H
Count mispredicted unconditional near call branch instructions, excluding
non call branch, executed.
.It Li BR_MISP_EXEC.INDIRECT_NEAR_CALL
.Pq Event 89H , Umask A0H
Count mispredicted indirect near calls, including both register and memory
indirect, executed.
.It Li BR_MISP_EXEC.ALL_BRANCHES
.Pq Event 89H , Umask FFH
Counts all mispredicted 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.
.It Li UOPS_EXECUTED_PORT.PORT_0
.Pq Event A1H , Umask 01H
Cycles which a Uop is dispatched on port 0 in this
thread.
.It Li UOPS_EXECUTED_PORT.PORT_1
.Pq Event A1H , Umask 02H
Cycles which a Uop is dispatched on port 1 in this
thread.
.It Li UOPS_EXECUTED_PORT.PORT_2
.Pq Event A1H , Umask 04H
Cycles which a Uop is dispatched on port 2 in this
thread.
.It Li UOPS_EXECUTED_PORT.PORT_3
.Pq Event A1H , Umask 08H
Cycles which a Uop is dispatched on port 3 in this
thread.
.It Li UOPS_EXECUTED_PORT.PORT_4
.Pq Event A1H , Umask 10H
Cycles which a Uop is dispatched on port 4 in this
thread.
.It Li UOPS_EXECUTED_PORT.PORT_5
.Pq Event A1H , Umask 20H
Cycles which a Uop is dispatched on port 5 in this
thread.
.It Li UOPS_EXECUTED_PORT.PORT_6
.Pq Event A1H , Umask 40H
Cycles which a Uop is dispatched on port 6 in this
thread.
.It Li UOPS_EXECUTED_PORT.PORT_7
.Pq Event A1H , Umask 80H
Cycles which a Uop is dispatched on port 7 in this
thread.
.It Li RESOURCE_STALLS.ANY
.Pq Event A2H , Umask 01H
Cycles Allocation is stalled due to Resource Related
reason.
.It Li RESOURCE_STALLS.RS
.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 CYCLE_ACTIVITY.CYCLES_L2_PENDING
.Pq Event A3H , Umask 01H
Cycles with pending L2 miss loads. Set Cmask=2 to
count cycle.
.It Li CYCLE_ACTIVITY.CYCLES_LDM_PENDING
.Pq Event A3H , Umask 02H
Cycles with pending memory loads. Set Cmask=2 to
count cycle.
.It Li CYCLE_ACTIVITY.STALLS_L2_PENDING
.Pq Event A3H , Umask 05H
Number of loads missed L2.
.It Li CYCLE_ACTIVITY.CYCLES_L1D_PENDING
.Pq Event A3H , Umask 08H
Cycles with pending L1 cache miss loads. Set
Cmask=8 to count cycle.
.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_CODE_RD
.Pq Event B0H , Umask 02H
Demand code 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_EXECUTED.CORE
.Pq Event B1H , Umask 02H
Counts total number of uops to be executed per-core
each cycle.
.It Li OFF_CORE_RESPONSE_0
.Pq Event B7H , Umask 01H
Requires MSR 01A6H
.It Li OFF_CORE_RESPONSE_1
.Pq Event BBH , Umask 01H
Requires MSR 01A7H
.It Li PAGE_WALKER_LOADS.DTLB_L1
.Pq Event BCH , Umask 11H
Number of DTLB page walker loads that hit in the
L1+FB.
.It Li PAGE_WALKER_LOADS.ITLB_L1
.Pq Event BCH , Umask 21H
Number of ITLB page walker loads that hit in the
L1+FB.
.It Li PAGE_WALKER_LOADS.DTLB_L2
.Pq Event BCH , Umask 12H
Number of DTLB page walker loads that hit in the L2.
.It Li PAGE_WALKER_LOADS.ITLB_L2
.Pq Event BCH , Umask 22H
Number of ITLB page walker loads that hit in the L2.
.It Li PAGE_WALKER_LOADS.DTLB_L3
.Pq Event BCH , Umask 14H
Number of DTLB page walker loads that hit in the L3.
.It Li PAGE_WALKER_LOADS.ITLB_L3
.Pq Event BCH , Umask 24H
Number of ITLB page walker loads that hit in the L3.
.It Li PAGE_WALKER_LOADS.DTLB_MEMORY
.Pq Event BCH , Umask 18H
Number of DTLB page walker loads from memory.
.It Li PAGE_WALKER_LOADS.ITLB_MEMORY
.Pq Event BCH , Umask 28H
Number of ITLB page walker loads from memory.
.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 INST_RETIRED.ANY_P
.Pq Event C0H , Umask 00H
Number of instructions at retirement.
.It Li INST_RETIRED.ALL
.Pq Event C0H , Umask 01H
Precise instruction retired event with HW to reduce
effect of PEBS shadow in IP distribution.
.It Li OTHER_ASSISTS.AVX_TO_SSE
.Pq Event C1H , Umask 08H
Number of transitions from AVX-256 to legacy SSE
when penalty applicable.
.It Li OTHER_ASSISTS.SSE_TO_AVX
.Pq Event C1H , Umask 10H
Number of transitions from SSE to AVX-256 when
penalty applicable.
.It Li OTHER_ASSISTS.ANY_WB_ASSIST
.Pq Event C1H , Umask 40H
Number of microcode assists invoked by HW upon
uop writeback.
.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
Number of self-modifying-code machine clears
detected.
.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_BRANCHES
.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 Supports PEBS
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.CONDITIONAL
.Pq Event C5H , Umask 04H
Mispredicted macro branch instructions retired.
.It Li FP_ASSIST.X87_OUTPUT
.Pq Event CAH , Umask 02H
Number of X87 FP assists due to Output values.
.It Li FP_ASSIST.X87_INPUT
.Pq Event CAH , Umask 04H
Number of X87 FP assists due to input values.
.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
Randomly sampled loads whose latency is above a
user defined threshold. A small fraction of the overall
loads are sampled due to randomization.
.It Li MEM_UOPS_RETIRED.STLB_MISS_LOADS
.Pq Event D0H , Umask 11H
Count retired load uops that missed the STLB.
.It Li MEM_UOPS_RETIRED.STLB_MISS_STORES
.Pq Event D0H , Umask 12H
Count retired store uops that missed the STLB.
.It Li MEM_UOPS_RETIRED.SPLIT_LOADS
.Pq Event D0H , Umask 41H
Count retired load uops that were split across a cache line.
.It Li MEM_UOPS_RETIRED.SPLIT_STORES
.Pq Event D0H , Umask 42H
Count retired store uops that were split across a cache line.
.It Li MEM_UOPS_RETIRED.ALL_LOADS
.Pq Event D0H , Umask 81H
Count all retired load uops.
.It Li MEM_UOPS_RETIRED.ALL_STORES
.Pq Event D0H , Umask 82H
Count all retired store uops.
.It Li MEM_LOAD_UOPS_RETIRED.L1_HIT
.Pq Event D1H , Umask 01H
Retired load uops with L1 cache hits as data sources.
.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 with LLC cache hits as data
sources.
.It Li MEM_LOAD_UOPS_RETIRED.L2_MISS
.Pq Event D1H , Umask 10H
Retired load uops missed L2. Unknown data source
excluded.
.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_MISS_RETIRED.LOCAL_DRAM
.Pq Event D3H , Umask 01H
Retired load uops which data sources missed LLC but
serviced from local dram.
.It Li BACLEARS.ANY
.Pq Event E6H , Umask 1FH
Number of front end re-steers due to BPU
misprediction.
.It Li L2_TRANS.DEMAND_DATA_RD
.Pq Event F0H , Umask 01H
Demand Data Read requests that access L2 cache.
.It Li L2_TRANS.RFO
.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
Any MLC or LLC HW prefetch accessing L2, including
rejects.
.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.
.It Li L2_LINES_IN.S
.Pq Event F1H , Umask 02H
L2 cache lines in S state filling L2.
.It Li L2_LINES_IN.E
.Pq Event F1H , Umask 04H
L2 cache lines in E state filling L2.
.It Li L2_LINES_IN.ALL
.Pq Event F1H , Umask 07H
L2 cache lines filling L2.
.It Li L2_LINES_OUT.DEMAND_CLEAN
.Pq Event F2H , Umask 05H
Clean L2 cache lines evicted by demand.
.It Li L2_LINES_OUT.DEMAND_DIRTY
.Pq Event F2H , Umask 06H
Dirty L2 cache lines evicted by demand.
.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.haswelluc 3 ,
.Xr pmc.iaf 3 ,
.Xr pmc.ivybridge 3 ,
.Xr pmc.ivybridgexeon 3 ,
.Xr pmc.k7 3 ,
.Xr pmc.k8 3 ,
.Xr pmc.p4 3 ,
.Xr pmc.p5 3 ,
.Xr pmc.p6 3 ,
.Xr pmc.sandybridge 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
.An -nosplit
The
.Lb libpmc
library was written by
.An Joseph Koshy Aq Mt jkoshy@FreeBSD.org .
The support for the Haswell
microarchitecture was written by
.An Hiren Panchasara Aq Mt hiren.panchasara@gmail.com .