From: Arnaldo Carvalho de Melo <acme@kernel.org>
To: Ingo Molnar <mingo@kernel.org>
Cc: linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org,
Andi Kleen <ak@linux.intel.com>,
Arnaldo Carvalho de Melo <acme@redhat.com>
Subject: [PATCH 10/24] perf vendor events: Update JSON metrics for Skylake Server
Date: Mon, 23 Oct 2017 20:47:22 -0300 [thread overview]
Message-ID: <20171023234736.5335-11-acme@kernel.org> (raw)
In-Reply-To: <20171023234736.5335-1-acme@kernel.org>
From: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Link: http://lkml.kernel.org/r/20170914200748.GA13837@tassilo.jf.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
.../pmu-events/arch/x86/skylakex/skx-metrics.json | 42 +++++++---------------
1 file changed, 12 insertions(+), 30 deletions(-)
diff --git a/tools/perf/pmu-events/arch/x86/skylakex/skx-metrics.json b/tools/perf/pmu-events/arch/x86/skylakex/skx-metrics.json
index 68f12a26c816..36c903faed0b 100644
--- a/tools/perf/pmu-events/arch/x86/skylakex/skx-metrics.json
+++ b/tools/perf/pmu-events/arch/x86/skylakex/skx-metrics.json
@@ -13,19 +13,19 @@
},
{
"BriefDescription": "Rough Estimation of fraction of fetched lines bytes that were likely consumed by program instructions",
- "MetricExpr": "min( 1 , UOPS_ISSUED.ANY / (UOPS_RETIRED.RETIRE_SLOTS / INST_RETIRED.ANY * 64 * ( ICACHE_64B.IFTAG_HIT + ICACHE_64B.IFTAG_MISS ) / 4.1) )",
+ "MetricExpr": "min( 1 , UOPS_ISSUED.ANY / ((UOPS_RETIRED.RETIRE_SLOTS / INST_RETIRED.ANY) * 64 * ( ICACHE_64B.IFTAG_HIT + ICACHE_64B.IFTAG_MISS ) / 4.1) )",
"MetricGroup": "Frontend",
"MetricName": "IFetch_Line_Utilization"
},
{
- "BriefDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache)",
+ "BriefDescription": "Fraction of Uops delivered by the DSB (aka Decoded Icache; or Uop Cache)",
"MetricExpr": "IDQ.DSB_UOPS / ( IDQ.DSB_UOPS + LSD.UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS )",
"MetricGroup": "DSB; Frontend_Bandwidth",
"MetricName": "DSB_Coverage"
},
{
"BriefDescription": "Cycles Per Instruction (threaded)",
- "MetricExpr": "1 / INST_RETIRED.ANY / cycles",
+ "MetricExpr": "1 / (INST_RETIRED.ANY / cycles)",
"MetricGroup": "Pipeline;Summary",
"MetricName": "CPI"
},
@@ -36,8 +36,8 @@
"MetricName": "CLKS"
},
{
- "BriefDescription": "Total issue-pipeline slots (per-core)",
- "MetricExpr": "4*cycles if not #SMT_on else (( CPU_CLK_UNHALTED.THREAD / 2) * (CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK )) if #EBS_Mode else ( CPU_CLK_UNHALTED.THREAD_ANY / 2 )",
+ "BriefDescription": "Total issue-pipeline slots",
+ "MetricExpr": "4*(( CPU_CLK_UNHALTED.THREAD_ANY / 2 ) if #SMT_on else cycles)",
"MetricGroup": "TopDownL1",
"MetricName": "SLOTS"
},
@@ -49,25 +49,25 @@
},
{
"BriefDescription": "Instructions Per Cycle (per physical core)",
- "MetricExpr": "INST_RETIRED.ANY / cycles if not #SMT_on else (( CPU_CLK_UNHALTED.THREAD / 2) * (CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK )) if #EBS_Mode else ( CPU_CLK_UNHALTED.THREAD_ANY / 2 )",
+ "MetricExpr": "INST_RETIRED.ANY / (( CPU_CLK_UNHALTED.THREAD_ANY / 2 ) if #SMT_on else cycles)",
"MetricGroup": "SMT",
"MetricName": "CoreIPC"
},
{
"BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is at least 1 uop executed)",
- "MetricExpr": "UOPS_EXECUTED.THREAD / ( UOPS_EXECUTED.CORE_CYCLES_GE_1 / 2) if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_1",
+ "MetricExpr": "UOPS_EXECUTED.THREAD / (( UOPS_EXECUTED.CORE_CYCLES_GE_1 / 2) if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_1)",
"MetricGroup": "Pipeline;Ports_Utilization",
"MetricName": "ILP"
},
{
"BriefDescription": "Average Branch Address Clear Cost (fraction of cycles)",
- "MetricExpr": "( RS_EVENTS.EMPTY_CYCLES - (ICACHE_16B.IFDATA_STALL +2* ICACHE_16B.IFDATA_STALL:c1:e1) - ICACHE_64B.IFTAG_STALL ) / RS_EVENTS.EMPTY_END",
+ "MetricExpr": "2* (( RS_EVENTS.EMPTY_CYCLES - ICACHE_16B.IFDATA_STALL - ICACHE_64B.IFTAG_STALL ) / RS_EVENTS.EMPTY_END)",
"MetricGroup": "Unknown_Branches",
"MetricName": "BAClear_Cost"
},
{
"BriefDescription": "Core actual clocks when any thread is active on the physical core",
- "MetricExpr": "CPU_CLK_UNHALTED.THREAD if not #SMT_on else (( CPU_CLK_UNHALTED.THREAD / 2) * (CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK )) if 1 else ( CPU_CLK_UNHALTED.THREAD_ANY / 2 )",
+ "MetricExpr": "( CPU_CLK_UNHALTED.THREAD_ANY / 2 ) if #SMT_on else CPU_CLK_UNHALTED.THREAD",
"MetricGroup": "SMT",
"MetricName": "CORE_CLKS"
},
@@ -79,35 +79,17 @@
},
{
"BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least 1 such miss)",
- "MetricExpr": "L1D_PEND_MISS.PENDING / ( L1D_PEND_MISS.PENDING_CYCLES_ANY / 2) if #SMT_on else L1D_PEND_MISS.PENDING_CYCLES",
+ "MetricExpr": "L1D_PEND_MISS.PENDING / (( L1D_PEND_MISS.PENDING_CYCLES_ANY / 2) if #SMT_on else L1D_PEND_MISS.PENDING_CYCLES)",
"MetricGroup": "Memory_Bound;Memory_BW",
"MetricName": "MLP"
},
{
"BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses",
- "MetricExpr": "( ITLB_MISSES.WALK_PENDING + DTLB_LOAD_MISSES.WALK_PENDING + DTLB_STORE_MISSES.WALK_PENDING + EPT.WALK_PENDING ) / ( 2 * cycles if not #SMT_on else (( CPU_CLK_UNHALTED.THREAD / 2) * (CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK )) if #EBS_Mode else ( CPU_CLK_UNHALTED.THREAD_ANY / 2 ) )",
+ "MetricExpr": "( ITLB_MISSES.WALK_PENDING + DTLB_LOAD_MISSES.WALK_PENDING + DTLB_STORE_MISSES.WALK_PENDING + EPT.WALK_PENDING ) / ( 2 * (( CPU_CLK_UNHALTED.THREAD_ANY / 2 ) if #SMT_on else cycles) )",
"MetricGroup": "TLB",
"MetricName": "Page_Walks_Utilization"
},
{
- "BriefDescription": "L1 cache miss per kilo instruction for demand loads",
- "MetricExpr": "1000 * MEM_LOAD_RETIRED.L1_MISS_PS / INST_RETIRED.ANY",
- "MetricGroup": "Cache_Misses;",
- "MetricName": "L1MPKI"
- },
- {
- "BriefDescription": "L2 cache miss per kilo instruction for demand loads",
- "MetricExpr": "1000 * MEM_LOAD_RETIRED.L2_MISS_PS / INST_RETIRED.ANY",
- "MetricGroup": "Cache_Misses;",
- "MetricName": "L2MPKI"
- },
- {
- "BriefDescription": "L3 cache miss per kilo instruction for demand loads",
- "MetricExpr": "1000 * MEM_LOAD_RETIRED.L3_MISS_PS / INST_RETIRED.ANY",
- "MetricGroup": "Cache_Misses;",
- "MetricName": "L3MPKI"
- },
- {
"BriefDescription": "Average CPU Utilization",
"MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / msr@tsc@",
"MetricGroup": "Summary",
@@ -115,7 +97,7 @@
},
{
"BriefDescription": "Giga Floating Point Operations Per Second",
- "MetricExpr": "( 1*( FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE ) + 2* FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4*( FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE ) + 8* (FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE ) + 16* FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / 1000000000 / duration_time",
+ "MetricExpr": "(( 1*( FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE ) + 2* FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4*( FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE ) + 8* FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE )) / 1000000000 / duration_time",
"MetricGroup": "FLOPS;Summary",
"MetricName": "GFLOPs"
},
--
2.13.6
next prev parent reply other threads:[~2017-10-23 23:48 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-10-23 23:47 [GIT PULL 00/24] perf/core improvements and fixes Arnaldo Carvalho de Melo
2017-10-23 23:47 ` [PATCH 01/24] perf vendor events: Update JSON metrics for Broadwell Arnaldo Carvalho de Melo
2017-10-23 23:47 ` [PATCH 02/24] perf vendor events: Update JSON metrics for Broadwell Server Arnaldo Carvalho de Melo
2017-10-23 23:47 ` [PATCH 03/24] perf vendor events: Update JSON metrics for Haswell Arnaldo Carvalho de Melo
2017-10-23 23:47 ` [PATCH 04/24] perf vendor events: Update JSON metrics for Haswell Server Arnaldo Carvalho de Melo
2017-10-23 23:47 ` [PATCH 05/24] perf vendor events: Update JSON metrics for IvyBridge Arnaldo Carvalho de Melo
2017-10-23 23:47 ` [PATCH 06/24] perf vendor events: Update JSON metrics for IvyTown Arnaldo Carvalho de Melo
2017-10-23 23:47 ` [PATCH 07/24] perf vendor events: Update JSON metrics for JakeTown Arnaldo Carvalho de Melo
2017-10-23 23:47 ` [PATCH 08/24] perf vendor events: Update JSON metrics for Sandy Bridge Arnaldo Carvalho de Melo
2017-10-23 23:47 ` [PATCH 09/24] perf vendor events: Update JSON metrics for Skylake Arnaldo Carvalho de Melo
2017-10-23 23:47 ` Arnaldo Carvalho de Melo [this message]
2017-10-23 23:47 ` [PATCH 11/24] perf mmap: Move perf_mmap and methods to separate mmap.[ch] files Arnaldo Carvalho de Melo
2017-10-23 23:47 ` [PATCH 12/24] perf record: Make record__mmap_read generic Arnaldo Carvalho de Melo
2017-10-23 23:47 ` [PATCH 13/24] perf mmap: Adopt push method from builtin-record.c Arnaldo Carvalho de Melo
2017-10-23 23:47 ` [PATCH 14/24] perf tests attr: Make hw events optional Arnaldo Carvalho de Melo
2017-10-23 23:47 ` [PATCH 15/24] perf list: Fix group description in the man page Arnaldo Carvalho de Melo
2017-10-23 23:47 ` [PATCH 16/24] perf annotate: Remove arch::cpuid_parse callback Arnaldo Carvalho de Melo
2017-10-23 23:47 ` [PATCH 17/24] perf tools: Do not check ABI headers in a detached tarball build Arnaldo Carvalho de Melo
2017-10-23 23:47 ` [PATCH 18/24] perf vendor events: Fix incorrect cmask syntax for some Intel metrics Arnaldo Carvalho de Melo
2017-10-23 23:47 ` [PATCH 19/24] perf tools: Introduce binary__fprintf() Arnaldo Carvalho de Melo
2017-10-23 23:47 ` [PATCH 20/24] perf script: Use fprintf like printing uniformly Arnaldo Carvalho de Melo
2017-10-23 23:47 ` [PATCH 21/24] perf script: Fix error handling path Arnaldo Carvalho de Melo
2017-10-23 23:47 ` [PATCH 22/24] perf kmem: Perform some cleanup if '--time' is given an invalid value Arnaldo Carvalho de Melo
2017-10-23 23:47 ` [PATCH 23/24] perf namespaces: Add more appropriate set of headers Arnaldo Carvalho de Melo
2017-10-23 23:47 ` [PATCH 24/24] perf vendor events: Add Goldmont Plus V1 event file Arnaldo Carvalho de Melo
2017-10-24 9:13 ` [GIT PULL 00/24] perf/core improvements and fixes Ingo Molnar
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20171023234736.5335-11-acme@kernel.org \
--to=acme@kernel.org \
--cc=acme@redhat.com \
--cc=ak@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=mingo@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).