.\" 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 Jan 25, 2013
.Dt PMC.IVYBRIDGEXEON 3
.Os
.Sh NAME
.Nm pmc.ivybridgexeon
.Nd measurement events for
.Tn Intel
.Tn Ivy Bridge Xeon
family CPUs
.Sh LIBRARY
.Lb libpmc
.Sh SYNOPSIS
.In pmc.h
.Sh DESCRIPTION
.Tn Intel
.Tn "Ivy Bridge Xeon"
CPUs contain PMCs conforming to version 2 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.
.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 Ivy Bridge Xeon PMCs are documented in
.Rs
.%B "Intel(R) 64 and IA-32 Architectures Software Developer's Manual"
.%N "Order Number: 325462-045US"
.%D January 2013
.%Q "Intel Corporation"
.Re
.Ss IVYBRIDGE FIXED FUNCTION PMCS
These PMCs and their supported events are documented in
.Xr pmc.iaf 3 .
.Ss IVYBRIDGE 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_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)
Ivy Bridge 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.DEMAND_LD_MISS_CAUSES_A_WALK
.Pq Event 08H , Umask 81H
Misses in all TLB levels that cause a page walk of any page size from demand loads.
.It Li DTLB_LOAD_MISSES.DEMAND_LD_WALK_COMPLETED
.Pq Event 08H , Umask 82H
Misses in all TLB levels that caused page walk completed of any size by demand loads.
.It Li DTLB_LOAD_MISSES.DEMAND_LD_WALK_DURATION
.Pq Event 08H , Umask 84H
Cycle PMH is busy with a walk due to demand loads.
.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 = 1to count stalled cycles.
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 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 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
Counts all L2 HW prefetcher requests that hit L2.
.It Li L2_RQSTS.PF_MISS
.Pq Event 24H , Umask 80H
Counts all L2 HW prefetcher requests that missed L2.
.It Li L2_RQSTS.ALL_PF
.Pq Event 24H , Umask C0H
Counts all L2 HW prefetcher requests.
.It Li L2_STORE_LOCK_RQSTS.MISS
.Pq Event 27H , Umask 01H
RFOs that miss cache lines.
.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.MISS
.Pq Event 28H , Umask 01H
Not rejected writebacks that missed LLC.
.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 L2_L1D_WB_RQSTS.ALL
.Pq Event 28H , Umask 0FH
Not rejected writebacks from L1D to L2 cache lines in any 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 Cmaks = 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
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 01H
Number of integer Move Elimination candidate uops that were not eliminated.
.It Li MOVE_ELIMINATION.SIMD_NOT_ELIMINATED
.Pq Event 58H , Umask 02H
Number of SIMD Move Elimination candidate uops that were not eliminated.
.It Li MOVE_ELIMINATION.INT_ELIMINATED
.Pq Event 58H , Umask 04H
Number of integer Move Elimination candidate uops that were eliminated.
.It Li MOVE_ELIMINATION.SIMD_ELIMINATED
.Pq Event 58H , Umask 08H
Number of SIMD Move Elimination candidate uops that were eliminated.
.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 DTLB_LOAD_MISSES.STLB_HIT
.Pq Event 5FH , Umask 04H
Counts load operations that missed 1st level DTLB but hit the 2nd level.
.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_CODE_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.
Can combine Umask 04H and 20H.
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.
Can combine Umask 08H and 10H
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.
Can combine Umask 04H, 08H.
.It Li IDQ.MS_MITE_UOPS
.Pq Event 79H , Umask 20H
Increment each cycle # of uops delivered to IDQ when MS_busy by MITE. Set
Cmask = 1 to count cycles.
Can combine Umask 04H, 08H.
.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.
.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 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.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.
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
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.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 AnyThread to count per core.
.It Li CYCLE_ACTIVITY.CYCLES_LDM_PENDING
.Pq Event A3H , Umask 02H
Cycles with pending memory loads. Set AnyThread to count per core.
.It Li CYCLE_ACTIVITY.CYCLES_NO_EXECUTE
.Pq Event A3H , Umask 04H
Cycles of dispatch stalls. Set AnyThread to count per core.
.It Li CYCLE_ACTIVITY.CYCLES_L1D_PENDING
.Pq Event A3H , Umask 08H
Cycles with pending L1 cache miss loads. Set AnyThread to count per core.
.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.EXCEED_DSB_LINES
.Pq Event ACH , Umask 08H
DSB Fill encountered > 3 DSB lines.
.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.THREAD
.Pq Event B1H , Umask 01H
Counts total number of uops to be executed per-thread each cycle. Set Cmask
= 1, INV =1 to count stall cycles.
.It Li UOPS_EXECUTED.CORE
.Pq Event B1H , Umask 02H
Counts total number of uops to be executed per-core each cycle.
Do not need to set ANY.
.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 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.
PMC1 only.
Must quiesce other PMCs.
.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 AVX- 256 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.
Supports PEBS, use Any=1 for core granular.
.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 retired.
Supports PEBS.
.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.
Supports PEBS.
.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_BRANCHES
.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 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
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_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.
Supports PEBS.
.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 whose data source was LLC hit with no snoop required.
.It Li MEM_LOAD_UOPS_RETIRED.LLC_MISS
.Pq Event D1H , Umask 20H
Retired load uops whose data source is LLC miss.
.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.
Supports PEBS.
.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.
Supports PEBS.
.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.
Supports PEBS.
.It Li MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_DRAM
.Pq Event D3H , Umask 04H
Retired load uops whose data source was remote DRAM.
.It Li MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_HITM
.Pq Event D3H , Umask 10H
Retired load uops whose data source was remote HITM.
.It Li MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_FWD
.Pq Event D3H , Umask 20H
Retired load uops whose data source was forwards from a remote cache.
.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.
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 the MLC prefetcher.
.It Li L2_LINES_OUT.PF_DIRTY
.Pq Event F2H , Umask 08H
Dirty L2 cache lines evicted by the MLC prefetcher.
.It Li L2_LINES_OUT.DIRTY_ALL
.Pq Event F2H , Umask 0AH
Dirty L2 cache lines filling the L2.
.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.ivybridge 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 Ivy Bridge Xeon
microarchitecture was written by
.An Hiren Panchasara Aq Mt hiren.panchasara@gmail.com .