Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 19 Apr 2013 19:28:49 +0000 (UTC)
From:      Hiren Panchasara <hiren@FreeBSD.org>
To:        src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-stable@freebsd.org, svn-src-stable-9@freebsd.org
Subject:   svn commit: r249655 - in stable/9: lib/libpmc sys/dev/hwpmc sys/sys
Message-ID:  <201304191928.r3JJSn2f053749@svn.freebsd.org>

next in thread | raw e-mail | index | archive | help
Author: hiren
Date: Fri Apr 19 19:28:48 2013
New Revision: 249655
URL: http://svnweb.freebsd.org/changeset/base/249655

Log:
  MFC: r248842
  
  Update hwpmc to support Haswell class processors.
  0x3C:      /* Per Intel document 325462-045US 01/2013. */
  
  Add manpage to document all the goodness that is available in this
  processor model.
  
  Approved by:	sbruno (mentor)

Added:
  stable/9/lib/libpmc/pmc.haswell.3   (contents, props changed)
  stable/9/lib/libpmc/pmc.haswelluc.3   (contents, props changed)
Modified:
  stable/9/lib/libpmc/Makefile
  stable/9/lib/libpmc/libpmc.c
  stable/9/sys/dev/hwpmc/hwpmc_core.c
  stable/9/sys/dev/hwpmc/hwpmc_intel.c
  stable/9/sys/dev/hwpmc/hwpmc_uncore.c
  stable/9/sys/dev/hwpmc/pmc_events.h
  stable/9/sys/sys/pmc.h
Directory Properties:
  stable/9/lib/libpmc/   (props changed)
  stable/9/sys/   (props changed)
  stable/9/sys/dev/   (props changed)
  stable/9/sys/sys/   (props changed)

Modified: stable/9/lib/libpmc/Makefile
==============================================================================
--- stable/9/lib/libpmc/Makefile	Fri Apr 19 18:45:03 2013	(r249654)
+++ stable/9/lib/libpmc/Makefile	Fri Apr 19 19:28:48 2013	(r249655)
@@ -27,6 +27,8 @@ MAN+=	pmc.soft.3
 MAN+=	pmc.atom.3
 MAN+=	pmc.core.3
 MAN+=	pmc.core2.3
+MAN+=	pmc.haswell.3
+MAN+=	pmc.haswelluc.3
 MAN+=	pmc.iaf.3
 MAN+=	pmc.ivybridge.3
 MAN+=	pmc.ivybridgexeon.3

Modified: stable/9/lib/libpmc/libpmc.c
==============================================================================
--- stable/9/lib/libpmc/libpmc.c	Fri Apr 19 18:45:03 2013	(r249654)
+++ stable/9/lib/libpmc/libpmc.c	Fri Apr 19 19:28:48 2013	(r249655)
@@ -182,6 +182,11 @@ static const struct pmc_event_descr core
 	__PMC_EV_ALIAS_COREI7()
 };
 
+static const struct pmc_event_descr haswell_event_table[] =
+{
+	__PMC_EV_ALIAS_HASWELL()
+};
+
 static const struct pmc_event_descr ivybridge_event_table[] =
 {
 	__PMC_EV_ALIAS_IVYBRIDGE()
@@ -212,6 +217,11 @@ static const struct pmc_event_descr core
 	__PMC_EV_ALIAS_COREI7UC()
 };
 
+static const struct pmc_event_descr haswelluc_event_table[] =
+{
+	__PMC_EV_ALIAS_HASWELLUC()
+};
+
 static const struct pmc_event_descr sandybridgeuc_event_table[] =
 {
 	__PMC_EV_ALIAS_SANDYBRIDGEUC()
@@ -236,6 +246,7 @@ PMC_MDEP_TABLE(atom, IAP, PMC_CLASS_SOFT
 PMC_MDEP_TABLE(core, IAP, PMC_CLASS_SOFT, PMC_CLASS_TSC);
 PMC_MDEP_TABLE(core2, IAP, PMC_CLASS_SOFT, PMC_CLASS_IAF, PMC_CLASS_TSC);
 PMC_MDEP_TABLE(corei7, IAP, PMC_CLASS_SOFT, PMC_CLASS_IAF, PMC_CLASS_TSC, PMC_CLASS_UCF, PMC_CLASS_UCP);
+PMC_MDEP_TABLE(haswell, IAP, PMC_CLASS_SOFT, PMC_CLASS_IAF, PMC_CLASS_TSC, PMC_CLASS_UCF, PMC_CLASS_UCP);
 PMC_MDEP_TABLE(ivybridge, IAP, PMC_CLASS_SOFT, PMC_CLASS_IAF, PMC_CLASS_TSC);
 PMC_MDEP_TABLE(ivybridge_xeon, IAP, PMC_CLASS_SOFT, PMC_CLASS_IAF, PMC_CLASS_TSC);
 PMC_MDEP_TABLE(sandybridge, IAP, PMC_CLASS_SOFT, PMC_CLASS_IAF, PMC_CLASS_TSC, PMC_CLASS_UCF, PMC_CLASS_UCP);
@@ -275,6 +286,7 @@ PMC_CLASS_TABLE_DESC(atom, IAP, atom, ia
 PMC_CLASS_TABLE_DESC(core, IAP, core, iap);
 PMC_CLASS_TABLE_DESC(core2, IAP, core2, iap);
 PMC_CLASS_TABLE_DESC(corei7, IAP, corei7, iap);
+PMC_CLASS_TABLE_DESC(haswell, IAP, haswell, iap);
 PMC_CLASS_TABLE_DESC(ivybridge, IAP, ivybridge, iap);
 PMC_CLASS_TABLE_DESC(ivybridge_xeon, IAP, ivybridge_xeon, iap);
 PMC_CLASS_TABLE_DESC(sandybridge, IAP, sandybridge, iap);
@@ -282,6 +294,7 @@ PMC_CLASS_TABLE_DESC(sandybridge_xeon, I
 PMC_CLASS_TABLE_DESC(westmere, IAP, westmere, iap);
 PMC_CLASS_TABLE_DESC(ucf, UCF, ucf, ucf);
 PMC_CLASS_TABLE_DESC(corei7uc, UCP, corei7uc, ucp);
+PMC_CLASS_TABLE_DESC(haswelluc, UCP, haswelluc, ucp);
 PMC_CLASS_TABLE_DESC(sandybridgeuc, UCP, sandybridgeuc, ucp);
 PMC_CLASS_TABLE_DESC(westmereuc, UCP, westmereuc, ucp);
 #endif
@@ -579,6 +592,8 @@ static struct pmc_event_alias core2_alia
 #define	atom_aliases_without_iaf	core2_aliases_without_iaf
 #define corei7_aliases			core2_aliases
 #define corei7_aliases_without_iaf	core2_aliases_without_iaf
+#define haswell_aliases			core2_aliases
+#define haswell_aliases_without_iaf	core2_aliases_without_iaf
 #define ivybridge_aliases		core2_aliases
 #define ivybridge_aliases_without_iaf	core2_aliases_without_iaf
 #define ivybridge_xeon_aliases		core2_aliases
@@ -737,6 +752,31 @@ static struct pmc_masks iap_rsp_mask_sb_
 	NULLMASK
 };
 
+static struct pmc_masks iap_rsp_mask_haswell[] = {
+	PMCMASK(REQ_DMND_DATA_RD,	(1ULL <<  0)),
+	PMCMASK(REQ_DMND_RFO,		(1ULL <<  1)),
+	PMCMASK(REQ_DMND_IFETCH,	(1ULL <<  2)),
+	PMCMASK(REQ_PF_DATA_RD,		(1ULL <<  4)),
+	PMCMASK(REQ_PF_RFO,		(1ULL <<  5)),
+	PMCMASK(REQ_PF_IFETCH,		(1ULL <<  6)),
+	PMCMASK(REQ_OTHER,		(1ULL << 15)),
+	PMCMASK(RES_ANY,		(1ULL << 16)),
+	PMCMASK(RES_SUPPLIER_SUPP,	(1ULL << 17)),
+	PMCMASK(RES_SUPPLIER_LLC_HITM,	(1ULL << 18)),
+	PMCMASK(RES_SUPPLIER_LLC_HITE,	(1ULL << 19)),
+	PMCMASK(RES_SUPPLIER_LLC_HITS,	(1ULL << 20)),
+	PMCMASK(RES_SUPPLIER_LLC_HITF,	(1ULL << 21)),
+	PMCMASK(RES_SUPPLIER_LOCAL,	(1ULL << 22)),
+	PMCMASK(RES_SNOOP_SNP_NONE,	(1ULL << 31)),
+	PMCMASK(RES_SNOOP_SNP_NO_NEEDED,(1ULL << 32)),
+	PMCMASK(RES_SNOOP_SNP_MISS,	(1ULL << 33)),
+	PMCMASK(RES_SNOOP_HIT_NO_FWD,	(1ULL << 34)),
+	PMCMASK(RES_SNOOP_HIT_FWD,	(1ULL << 35)),
+	PMCMASK(RES_SNOOP_HITM,		(1ULL << 36)),
+	PMCMASK(RES_NON_DRAM,		(1ULL << 37)),
+	NULLMASK
+};
+
 static int
 iap_allocate_pmc(enum pmc_event pe, char *ctrspec,
     struct pmc_op_pmcallocate *pmc_config)
@@ -819,6 +859,11 @@ iap_allocate_pmc(enum pmc_event pe, char
 				n = pmc_parse_mask(iap_rsp_mask_sb_sbx_ib, p, &rsp);
 			} else
 				return (-1);
+		} else if (cpu_info.pm_cputype == PMC_CPU_INTEL_HASWELL) {
+			if (KWPREFIXMATCH(p, IAP_KW_RSP "=")) {
+				n = pmc_parse_mask(iap_rsp_mask_haswell, p, &rsp);
+			} else
+				return (-1);
 		} else
 			return (-1);
 
@@ -2679,6 +2724,10 @@ pmc_event_names_of_class(enum pmc_class 
 			ev = corei7_event_table;
 			count = PMC_EVENT_TABLE_SIZE(corei7);
 			break;
+		case PMC_CPU_INTEL_HASWELL:
+			ev = haswell_event_table;
+			count = PMC_EVENT_TABLE_SIZE(haswell);
+			break;
 		case PMC_CPU_INTEL_IVYBRIDGE:
 			ev = ivybridge_event_table;
 			count = PMC_EVENT_TABLE_SIZE(ivybridge);
@@ -2716,6 +2765,10 @@ pmc_event_names_of_class(enum pmc_class 
 			ev = corei7uc_event_table;
 			count = PMC_EVENT_TABLE_SIZE(corei7uc);
 			break;
+		case PMC_CPU_INTEL_HASWELL:
+			ev = haswelluc_event_table;
+			count = PMC_EVENT_TABLE_SIZE(haswelluc);
+			break;
 		case PMC_CPU_INTEL_SANDYBRIDGE:
 			ev = sandybridgeuc_event_table;
 			count = PMC_EVENT_TABLE_SIZE(sandybridgeuc);
@@ -2979,6 +3032,11 @@ pmc_init(void)
 		pmc_class_table[n++] = &corei7uc_class_table_descr;
 		PMC_MDEP_INIT_INTEL_V2(corei7);
 		break;
+	case PMC_CPU_INTEL_HASWELL:
+		pmc_class_table[n++] = &ucf_class_table_descr;
+		pmc_class_table[n++] = &haswelluc_class_table_descr;
+		PMC_MDEP_INIT_INTEL_V2(haswell);
+		break;
 	case PMC_CPU_INTEL_IVYBRIDGE:
 		PMC_MDEP_INIT_INTEL_V2(ivybridge);
 		break;
@@ -3119,6 +3177,10 @@ _pmc_name_of_event(enum pmc_event pe, en
 			ev = corei7_event_table;
 			evfence = corei7_event_table + PMC_EVENT_TABLE_SIZE(corei7);
 			break;
+		case PMC_CPU_INTEL_HASWELL:
+			ev = haswell_event_table;
+			evfence = haswell_event_table + PMC_EVENT_TABLE_SIZE(haswell);
+			break;
 		case PMC_CPU_INTEL_IVYBRIDGE:
 			ev = ivybridge_event_table;
 			evfence = ivybridge_event_table + PMC_EVENT_TABLE_SIZE(ivybridge);

Added: stable/9/lib/libpmc/pmc.haswell.3
==============================================================================
--- /dev/null	00:00:00 1970	(empty, because file is newly added)
+++ stable/9/lib/libpmc/pmc.haswell.3	Fri Apr 19 19:28:48 2013	(r249655)
@@ -0,0 +1,975 @@
+.\" 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 Haswsell
+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.COND
+.Pq Event 88H , Umask 01H
+Qualify conditional near branch instructions 
+executed, but not necessarily retired. 
+.It Li BR_INST_EXEC.DIRECT_JMP
+.Pq Event 88H , Umask 02H
+Qualify all unconditional near branch instructions 
+excluding calls and indirect branches. 
+.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. 
+.It Li BR_INST_EXEC.RETURN_NEAR
+.Pq Event 88H , Umask 08H
+Qualify indirect near branches that have a return 
+mnemonic. 
+.It Li BR_INST_EXEC.DIRECT_NEAR_CALL
+.Pq Event 88H , Umask 10H
+Qualify unconditional near call branch instructions, 
+excluding non call branch, executed. 
+.It Li BR_INST_EXEC.INDIRECT_NEAR_CALL
+.Pq Event 88H , Umask 20H
+Qualify indirect near calls, including both register and
+memory indirect, executed.
+.It Li BR_INST_EXEC.NONTAKEN
+.Pq Event 88H , Umask 40H
+Qualify non-taken near branches executed. 
+.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_EXEC.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. 
+.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. 
+.It Li BR_MISP_EXEC.RETURN_NEAR
+.Pq Event 89H , Umask 08H
+Qualify mispredicted indirect near branches that 
+have a return mnemonic. 
+.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. 
+.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. 
+.It Li BR_MISP_EXEC.NONTAKEN
+.Pq Event 89H , Umask 40H
+Qualify mispredicted non-taken near branches 
+executed.
+.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. 
+.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_UOP_RETIRED.LOADS
+.Pq Event D0H , Umask 01H
+Qualify retired memory uops that are loads. Combine Supports PEBS and
+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 Supports PEBS and

*** DIFF OUTPUT TRUNCATED AT 1000 LINES ***



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201304191928.r3JJSn2f053749>