maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/test/tools/llvm-mca/X86/cpus.s

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

102 lines
5.7 KiB
ArmAsm
Raw Normal View History

[llvm-mca] Removed an empty line generated by the timeline view. NFC. Also, regenerate all tests. llvm-svn: 332853
2018-05-22 01:11:56 +08:00
# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
[tools][tests] - Use --check-prefixes instead of multiple --check-prefix. NFCI. There is no need to use `--check-prefix` multiple times. It helps to improve readability/test maintainability. This patch does it for all tools at once. Differential revision: https://reviews.llvm.org/D78217
2020-04-15 23:28:08 +08:00
# RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=x86-64 -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefixes=ALL,BARCELONA %s
# RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=bdver2 -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefixes=ALL,BDVER2 %s
# RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=btver2 -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefixes=ALL,BTVER2 %s
# RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=znver1 -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefixes=ALL,ZNVER1 %s
# RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=znver2 -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefixes=ALL,ZNVER2 %s
[X86] AMD Zen 3 Scheduler Model Introduce basic schedule model for AMD Zen 3 CPU's, a.k.a `znver3`. This is fully built from scratch, from llvm-mca measurements and documented reference materials. Nothing was copied from `znver2`/`znver1`. I believe this is in a reasonable state of completion for inclusion, probably better than D52779 `bdver2` was :) Namely: * uops are pretty spot-on (at least what llvm-mca can measure) {F16422596} * latency is also pretty spot-on (at least what llvm-mca can measure) {F16422601} * throughput is within reason {F16422607} I haven't run much benchmarks with this, however RawSpeed benchmarks says this is beneficial: {F16603978} {F16604029} I'll call out the obvious problems there: * i didn't really bother with X87 instructions * i didn't really bother with obviously-microcoded/system instructions * There are large discrepancy in throughput for `mr` and `rm` instructions. I'm not really sure if it's a modelling defect that needs to be fixed, or it's a defect of measurments. * Pipe distributions are probably bad :) I can't do much here until AMD allows that to be fixed by documenting the appropriate counters and updating libpfm That being said, as @RKSimon notes: >>! In D94395#2647381, @RKSimon wrote: > I'll mention again that all the znver* models appear to be very inaccurate wrt SIMD/FPU instructions <...> so how much worse this could possibly be?! Things that aren't there: * Various tunings: zero idioms, etc. That is follow-ups. Differential Revision: https://reviews.llvm.org/D94395
2021-05-02 03:04:48 +08:00
# RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=znver3 -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefixes=ALL,ZNVER3 %s
[tools][tests] - Use --check-prefixes instead of multiple --check-prefix. NFCI. There is no need to use `--check-prefix` multiple times. It helps to improve readability/test maintainability. This patch does it for all tools at once. Differential revision: https://reviews.llvm.org/D78217
2020-04-15 23:28:08 +08:00
# RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=sandybridge -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefixes=ALL,SANDYBRIDGE %s
# RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=ivybridge -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefixes=ALL,IVYBRIDGE %s
# RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=haswell -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefixes=ALL,HASWELL %s
# RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=broadwell -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefixes=ALL,BROADWELL %s
# RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=knl -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefixes=ALL,KNL %s
# RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=skylake -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefixes=ALL,SKX %s
# RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=skylake-avx512 -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefixes=ALL,SKX-AVX512 %s
[MCA][X86] Add basic coverage for icelake arch Copy the skylake-avx512 tests for icelake-server coverage. Add icelake/rocketlake/tigerlake test coverage to the relevent generic tests as well.
2021-08-31 19:19:58 +08:00
# RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=icelake-client -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefixes=ALL,ICX %s
# RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=icelake-server -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefixes=ALL,ICX %s
# RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=rocketlake -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefixes=ALL,ICX %s
# RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=tigerlake -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefixes=ALL,ICX %s
[tools][tests] - Use --check-prefixes instead of multiple --check-prefix. NFCI. There is no need to use `--check-prefix` multiple times. It helps to improve readability/test maintainability. This patch does it for all tools at once. Differential revision: https://reviews.llvm.org/D78217
2020-04-15 23:28:08 +08:00
# RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=slm -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefixes=ALL,SLM %s
[llvm-mca] LLVM Machine Code Analyzer. llvm-mca is an LLVM based performance analysis tool that can be used to statically measure the performance of code, and to help triage potential problems with target scheduling models. llvm-mca uses information which is already available in LLVM (e.g. scheduling models) to statically measure the performance of machine code in a specific cpu. Performance is measured in terms of throughput as well as processor resource consumption. The tool currently works for processors with an out-of-order backend, for which there is a scheduling model available in LLVM. The main goal of this tool is not just to predict the performance of the code when run on the target, but also help with diagnosing potential performance issues. Given an assembly code sequence, llvm-mca estimates the IPC (instructions per cycle), as well as hardware resources pressure. The analysis and reporting style were mostly inspired by the IACA tool from Intel. This patch is related to the RFC on llvm-dev visible at this link: http://lists.llvm.org/pipermail/llvm-dev/2018-March/121490.html Differential Revision: https://reviews.llvm.org/D43951 llvm-svn: 326998
2018-03-08 21:05:02 +08:00
add %edi, %eax
[UpdateTestChecks] Improved update_mca_test_checks block analysis Previously update_mca_test_checks worked entirely at "block" level where a block is some sequence of lines delimited by at least one empty line. This generally worked well, but could sometimes lead to excessive repetition of check lines for various prefixes if some block was almost identical between prefixes, but not quite (for example, due to a different dispatch width in the otherwise identical summary views). This new analyis attempts to split blocks further in the case where the following conditions are met: a) There is some prefix common to every RUN line (typically 'ALL'). b) The first line of the block is common to the output with every prefix. c) The block has the same number of lines for the output with every prefix. Also, regenerated all llvm-mca test files with the following command: update_mca_test_checks.py "../test/tools/llvm-mca/*/*.s" "../test/tools/llvm-mca/*/*/*.s" The new analysis showed a "multiple lines not disambiguated by prefixes" warning for test "AArch64/Exynos/scheduler-queue-usage.s" so I've also added some explicit prefixes to each of the RUN lines in that test. Differential Revision: https://reviews.llvm.org/D47321 llvm-svn: 333204
2018-05-25 00:36:44 +08:00
# ALL: Iterations: 100
# ALL-NEXT: Instructions: 100
# ALL-NEXT: Total Cycles: 103
[llvm-mca] Add fields "Total uOps" and "uOps Per Cycle" to the report generated by the SummaryView. This patch adds two new fields to the perf report generated by the SummaryView. Fields are now logically organized into two small groups; only the second group contains throughput indicators. Example: ``` Iterations: 100 Instructions: 300 Total Cycles: 414 Total uOps: 700 Dispatch Width: 4 uOps Per Cycle: 1.69 IPC: 0.72 Block RThroughput: 4.0 ``` This patch also updates the docs for llvm-mca. Due to the nature of this change, several tests in the tools/llvm-mca directory were affected, and had to be updated using script `update_mca_test_checks.py`. llvm-svn: 340946
2018-08-30 01:56:39 +08:00
# ALL-NEXT: Total uOps: 100
[UpdateTestChecks] Improved update_mca_test_checks block analysis Previously update_mca_test_checks worked entirely at "block" level where a block is some sequence of lines delimited by at least one empty line. This generally worked well, but could sometimes lead to excessive repetition of check lines for various prefixes if some block was almost identical between prefixes, but not quite (for example, due to a different dispatch width in the otherwise identical summary views). This new analyis attempts to split blocks further in the case where the following conditions are met: a) There is some prefix common to every RUN line (typically 'ALL'). b) The first line of the block is common to the output with every prefix. c) The block has the same number of lines for the output with every prefix. Also, regenerated all llvm-mca test files with the following command: update_mca_test_checks.py "../test/tools/llvm-mca/*/*.s" "../test/tools/llvm-mca/*/*/*.s" The new analysis showed a "multiple lines not disambiguated by prefixes" warning for test "AArch64/Exynos/scheduler-queue-usage.s" so I've also added some explicit prefixes to each of the RUN lines in that test. Differential Revision: https://reviews.llvm.org/D47321 llvm-svn: 333204
2018-05-25 00:36:44 +08:00
[NFC][MCA][X86] Add baseline test coverage for AMD Barcelona (aka K10, fam10h) Looking into sched model for that CPU ... llvm-svn: 363497
2019-06-16 00:12:05 +08:00
# BARCELONA: Dispatch Width: 4
# BARCELONA-NEXT: uOps Per Cycle: 0.97
# BARCELONA-NEXT: IPC: 0.97
# BARCELONA-NEXT: Block RThroughput: 0.3
[NFC][X86] Baseline tests for AMD BdVer2 (Piledriver) Scheduler model Adding the baseline tests in a preparatory NFC commit, so that the actual commit shows the *diff*. Yes, i'm aware that a few of these codegen-based sched tests are testing wrong instructions, i will fix that afterwards. For https://reviews.llvm.org/D52779 llvm-svn: 345462
2018-10-28 04:36:11 +08:00
# BDVER2: Dispatch Width: 4
# BDVER2-NEXT: uOps Per Cycle: 0.97
# BDVER2-NEXT: IPC: 0.97
[X86] AMD Piledriver (BdVer2): major cleanup (mainly inverse throughput) I've started this cleanup more several times now, but got sidetracked elsewhere, e.g. by llvm-exegesis problems. Not this time, finally! This is mainly cleaning up the inverse throughput values, and a few latencies/uops, based on the llvm-exegesis measured values. Though this is not complete by any means, there's certainly more cleanup to be done. The performance numbers (i've only checked by RawSpeed benchmark) aren't really surprising - overall this *slightly* (< -1%) improves perf. llvm-svn: 360341
2019-05-09 21:54:51 +08:00
# BDVER2-NEXT: Block RThroughput: 1.0
[NFC][X86] Baseline tests for AMD BdVer2 (Piledriver) Scheduler model Adding the baseline tests in a preparatory NFC commit, so that the actual commit shows the *diff*. Yes, i'm aware that a few of these codegen-based sched tests are testing wrong instructions, i will fix that afterwards. For https://reviews.llvm.org/D52779 llvm-svn: 345462
2018-10-28 04:36:11 +08:00
[llvm-mca] Add fields "Total uOps" and "uOps Per Cycle" to the report generated by the SummaryView. This patch adds two new fields to the perf report generated by the SummaryView. Fields are now logically organized into two small groups; only the second group contains throughput indicators. Example: ``` Iterations: 100 Instructions: 300 Total Cycles: 414 Total uOps: 700 Dispatch Width: 4 uOps Per Cycle: 1.69 IPC: 0.72 Block RThroughput: 4.0 ``` This patch also updates the docs for llvm-mca. Due to the nature of this change, several tests in the tools/llvm-mca directory were affected, and had to be updated using script `update_mca_test_checks.py`. llvm-svn: 340946
2018-08-30 01:56:39 +08:00
# BROADWELL: Dispatch Width: 4
# BROADWELL-NEXT: uOps Per Cycle: 0.97
# BROADWELL-NEXT: IPC: 0.97
[UpdateTestChecks] Improved update_mca_test_checks block analysis Previously update_mca_test_checks worked entirely at "block" level where a block is some sequence of lines delimited by at least one empty line. This generally worked well, but could sometimes lead to excessive repetition of check lines for various prefixes if some block was almost identical between prefixes, but not quite (for example, due to a different dispatch width in the otherwise identical summary views). This new analyis attempts to split blocks further in the case where the following conditions are met: a) There is some prefix common to every RUN line (typically 'ALL'). b) The first line of the block is common to the output with every prefix. c) The block has the same number of lines for the output with every prefix. Also, regenerated all llvm-mca test files with the following command: update_mca_test_checks.py "../test/tools/llvm-mca/*/*.s" "../test/tools/llvm-mca/*/*/*.s" The new analysis showed a "multiple lines not disambiguated by prefixes" warning for test "AArch64/Exynos/scheduler-queue-usage.s" so I've also added some explicit prefixes to each of the RUN lines in that test. Differential Revision: https://reviews.llvm.org/D47321 llvm-svn: 333204
2018-05-25 00:36:44 +08:00
# BROADWELL-NEXT: Block RThroughput: 0.3
[llvm-mca] Add fields "Total uOps" and "uOps Per Cycle" to the report generated by the SummaryView. This patch adds two new fields to the perf report generated by the SummaryView. Fields are now logically organized into two small groups; only the second group contains throughput indicators. Example: ``` Iterations: 100 Instructions: 300 Total Cycles: 414 Total uOps: 700 Dispatch Width: 4 uOps Per Cycle: 1.69 IPC: 0.72 Block RThroughput: 4.0 ``` This patch also updates the docs for llvm-mca. Due to the nature of this change, several tests in the tools/llvm-mca directory were affected, and had to be updated using script `update_mca_test_checks.py`. llvm-svn: 340946
2018-08-30 01:56:39 +08:00
[utils] Ensure that update_mca_test_checks.py writes prefixes in alphabetical order llvm-svn: 343783
2018-10-04 22:42:19 +08:00
# BTVER2: Dispatch Width: 2
# BTVER2-NEXT: uOps Per Cycle: 0.97
# BTVER2-NEXT: IPC: 0.97
# BTVER2-NEXT: Block RThroughput: 0.5
[llvm-mca] Add fields "Total uOps" and "uOps Per Cycle" to the report generated by the SummaryView. This patch adds two new fields to the perf report generated by the SummaryView. Fields are now logically organized into two small groups; only the second group contains throughput indicators. Example: ``` Iterations: 100 Instructions: 300 Total Cycles: 414 Total uOps: 700 Dispatch Width: 4 uOps Per Cycle: 1.69 IPC: 0.72 Block RThroughput: 4.0 ``` This patch also updates the docs for llvm-mca. Due to the nature of this change, several tests in the tools/llvm-mca directory were affected, and had to be updated using script `update_mca_test_checks.py`. llvm-svn: 340946
2018-08-30 01:56:39 +08:00
# HASWELL: Dispatch Width: 4
# HASWELL-NEXT: uOps Per Cycle: 0.97
# HASWELL-NEXT: IPC: 0.97
[UpdateTestChecks] Improved update_mca_test_checks block analysis Previously update_mca_test_checks worked entirely at "block" level where a block is some sequence of lines delimited by at least one empty line. This generally worked well, but could sometimes lead to excessive repetition of check lines for various prefixes if some block was almost identical between prefixes, but not quite (for example, due to a different dispatch width in the otherwise identical summary views). This new analyis attempts to split blocks further in the case where the following conditions are met: a) There is some prefix common to every RUN line (typically 'ALL'). b) The first line of the block is common to the output with every prefix. c) The block has the same number of lines for the output with every prefix. Also, regenerated all llvm-mca test files with the following command: update_mca_test_checks.py "../test/tools/llvm-mca/*/*.s" "../test/tools/llvm-mca/*/*/*.s" The new analysis showed a "multiple lines not disambiguated by prefixes" warning for test "AArch64/Exynos/scheduler-queue-usage.s" so I've also added some explicit prefixes to each of the RUN lines in that test. Differential Revision: https://reviews.llvm.org/D47321 llvm-svn: 333204
2018-05-25 00:36:44 +08:00
# HASWELL-NEXT: Block RThroughput: 0.3
[llvm-mca] Add fields "Total uOps" and "uOps Per Cycle" to the report generated by the SummaryView. This patch adds two new fields to the perf report generated by the SummaryView. Fields are now logically organized into two small groups; only the second group contains throughput indicators. Example: ``` Iterations: 100 Instructions: 300 Total Cycles: 414 Total uOps: 700 Dispatch Width: 4 uOps Per Cycle: 1.69 IPC: 0.72 Block RThroughput: 4.0 ``` This patch also updates the docs for llvm-mca. Due to the nature of this change, several tests in the tools/llvm-mca directory were affected, and had to be updated using script `update_mca_test_checks.py`. llvm-svn: 340946
2018-08-30 01:56:39 +08:00
[MCA][X86] Add basic coverage for icelake arch Copy the skylake-avx512 tests for icelake-server coverage. Add icelake/rocketlake/tigerlake test coverage to the relevent generic tests as well.
2021-08-31 19:19:58 +08:00
# ICX: Dispatch Width: 6
# ICX-NEXT: uOps Per Cycle: 0.97
# ICX-NEXT: IPC: 0.97
# ICX-NEXT: Block RThroughput: 0.3
[llvm-mca] Add fields "Total uOps" and "uOps Per Cycle" to the report generated by the SummaryView. This patch adds two new fields to the perf report generated by the SummaryView. Fields are now logically organized into two small groups; only the second group contains throughput indicators. Example: ``` Iterations: 100 Instructions: 300 Total Cycles: 414 Total uOps: 700 Dispatch Width: 4 uOps Per Cycle: 1.69 IPC: 0.72 Block RThroughput: 4.0 ``` This patch also updates the docs for llvm-mca. Due to the nature of this change, several tests in the tools/llvm-mca directory were affected, and had to be updated using script `update_mca_test_checks.py`. llvm-svn: 340946
2018-08-30 01:56:39 +08:00
# IVYBRIDGE: Dispatch Width: 4
# IVYBRIDGE-NEXT: uOps Per Cycle: 0.97
# IVYBRIDGE-NEXT: IPC: 0.97
[UpdateTestChecks] Improved update_mca_test_checks block analysis Previously update_mca_test_checks worked entirely at "block" level where a block is some sequence of lines delimited by at least one empty line. This generally worked well, but could sometimes lead to excessive repetition of check lines for various prefixes if some block was almost identical between prefixes, but not quite (for example, due to a different dispatch width in the otherwise identical summary views). This new analyis attempts to split blocks further in the case where the following conditions are met: a) There is some prefix common to every RUN line (typically 'ALL'). b) The first line of the block is common to the output with every prefix. c) The block has the same number of lines for the output with every prefix. Also, regenerated all llvm-mca test files with the following command: update_mca_test_checks.py "../test/tools/llvm-mca/*/*.s" "../test/tools/llvm-mca/*/*/*.s" The new analysis showed a "multiple lines not disambiguated by prefixes" warning for test "AArch64/Exynos/scheduler-queue-usage.s" so I've also added some explicit prefixes to each of the RUN lines in that test. Differential Revision: https://reviews.llvm.org/D47321 llvm-svn: 333204
2018-05-25 00:36:44 +08:00
# IVYBRIDGE-NEXT: Block RThroughput: 0.3
[llvm-mca] Add fields "Total uOps" and "uOps Per Cycle" to the report generated by the SummaryView. This patch adds two new fields to the perf report generated by the SummaryView. Fields are now logically organized into two small groups; only the second group contains throughput indicators. Example: ``` Iterations: 100 Instructions: 300 Total Cycles: 414 Total uOps: 700 Dispatch Width: 4 uOps Per Cycle: 1.69 IPC: 0.72 Block RThroughput: 4.0 ``` This patch also updates the docs for llvm-mca. Due to the nature of this change, several tests in the tools/llvm-mca directory were affected, and had to be updated using script `update_mca_test_checks.py`. llvm-svn: 340946
2018-08-30 01:56:39 +08:00
# KNL: Dispatch Width: 4
# KNL-NEXT: uOps Per Cycle: 0.97
# KNL-NEXT: IPC: 0.97
[UpdateTestChecks] Improved update_mca_test_checks block analysis Previously update_mca_test_checks worked entirely at "block" level where a block is some sequence of lines delimited by at least one empty line. This generally worked well, but could sometimes lead to excessive repetition of check lines for various prefixes if some block was almost identical between prefixes, but not quite (for example, due to a different dispatch width in the otherwise identical summary views). This new analyis attempts to split blocks further in the case where the following conditions are met: a) There is some prefix common to every RUN line (typically 'ALL'). b) The first line of the block is common to the output with every prefix. c) The block has the same number of lines for the output with every prefix. Also, regenerated all llvm-mca test files with the following command: update_mca_test_checks.py "../test/tools/llvm-mca/*/*.s" "../test/tools/llvm-mca/*/*/*.s" The new analysis showed a "multiple lines not disambiguated by prefixes" warning for test "AArch64/Exynos/scheduler-queue-usage.s" so I've also added some explicit prefixes to each of the RUN lines in that test. Differential Revision: https://reviews.llvm.org/D47321 llvm-svn: 333204
2018-05-25 00:36:44 +08:00
# KNL-NEXT: Block RThroughput: 0.3
[llvm-mca] Add fields "Total uOps" and "uOps Per Cycle" to the report generated by the SummaryView. This patch adds two new fields to the perf report generated by the SummaryView. Fields are now logically organized into two small groups; only the second group contains throughput indicators. Example: ``` Iterations: 100 Instructions: 300 Total Cycles: 414 Total uOps: 700 Dispatch Width: 4 uOps Per Cycle: 1.69 IPC: 0.72 Block RThroughput: 4.0 ``` This patch also updates the docs for llvm-mca. Due to the nature of this change, several tests in the tools/llvm-mca directory were affected, and had to be updated using script `update_mca_test_checks.py`. llvm-svn: 340946
2018-08-30 01:56:39 +08:00
# SANDYBRIDGE: Dispatch Width: 4
# SANDYBRIDGE-NEXT: uOps Per Cycle: 0.97
# SANDYBRIDGE-NEXT: IPC: 0.97
[UpdateTestChecks] Improved update_mca_test_checks block analysis Previously update_mca_test_checks worked entirely at "block" level where a block is some sequence of lines delimited by at least one empty line. This generally worked well, but could sometimes lead to excessive repetition of check lines for various prefixes if some block was almost identical between prefixes, but not quite (for example, due to a different dispatch width in the otherwise identical summary views). This new analyis attempts to split blocks further in the case where the following conditions are met: a) There is some prefix common to every RUN line (typically 'ALL'). b) The first line of the block is common to the output with every prefix. c) The block has the same number of lines for the output with every prefix. Also, regenerated all llvm-mca test files with the following command: update_mca_test_checks.py "../test/tools/llvm-mca/*/*.s" "../test/tools/llvm-mca/*/*/*.s" The new analysis showed a "multiple lines not disambiguated by prefixes" warning for test "AArch64/Exynos/scheduler-queue-usage.s" so I've also added some explicit prefixes to each of the RUN lines in that test. Differential Revision: https://reviews.llvm.org/D47321 llvm-svn: 333204
2018-05-25 00:36:44 +08:00
# SANDYBRIDGE-NEXT: Block RThroughput: 0.3
[llvm-mca] Add fields "Total uOps" and "uOps Per Cycle" to the report generated by the SummaryView. This patch adds two new fields to the perf report generated by the SummaryView. Fields are now logically organized into two small groups; only the second group contains throughput indicators. Example: ``` Iterations: 100 Instructions: 300 Total Cycles: 414 Total uOps: 700 Dispatch Width: 4 uOps Per Cycle: 1.69 IPC: 0.72 Block RThroughput: 4.0 ``` This patch also updates the docs for llvm-mca. Due to the nature of this change, several tests in the tools/llvm-mca directory were affected, and had to be updated using script `update_mca_test_checks.py`. llvm-svn: 340946
2018-08-30 01:56:39 +08:00
# SKX: Dispatch Width: 6
# SKX-NEXT: uOps Per Cycle: 0.97
# SKX-NEXT: IPC: 0.97
[UpdateTestChecks] Improved update_mca_test_checks block analysis Previously update_mca_test_checks worked entirely at "block" level where a block is some sequence of lines delimited by at least one empty line. This generally worked well, but could sometimes lead to excessive repetition of check lines for various prefixes if some block was almost identical between prefixes, but not quite (for example, due to a different dispatch width in the otherwise identical summary views). This new analyis attempts to split blocks further in the case where the following conditions are met: a) There is some prefix common to every RUN line (typically 'ALL'). b) The first line of the block is common to the output with every prefix. c) The block has the same number of lines for the output with every prefix. Also, regenerated all llvm-mca test files with the following command: update_mca_test_checks.py "../test/tools/llvm-mca/*/*.s" "../test/tools/llvm-mca/*/*/*.s" The new analysis showed a "multiple lines not disambiguated by prefixes" warning for test "AArch64/Exynos/scheduler-queue-usage.s" so I've also added some explicit prefixes to each of the RUN lines in that test. Differential Revision: https://reviews.llvm.org/D47321 llvm-svn: 333204
2018-05-25 00:36:44 +08:00
# SKX-NEXT: Block RThroughput: 0.3
[llvm-mca] Add fields "Total uOps" and "uOps Per Cycle" to the report generated by the SummaryView. This patch adds two new fields to the perf report generated by the SummaryView. Fields are now logically organized into two small groups; only the second group contains throughput indicators. Example: ``` Iterations: 100 Instructions: 300 Total Cycles: 414 Total uOps: 700 Dispatch Width: 4 uOps Per Cycle: 1.69 IPC: 0.72 Block RThroughput: 4.0 ``` This patch also updates the docs for llvm-mca. Due to the nature of this change, several tests in the tools/llvm-mca directory were affected, and had to be updated using script `update_mca_test_checks.py`. llvm-svn: 340946
2018-08-30 01:56:39 +08:00
# SKX-AVX512: Dispatch Width: 6
# SKX-AVX512-NEXT: uOps Per Cycle: 0.97
# SKX-AVX512-NEXT: IPC: 0.97
[UpdateTestChecks] Improved update_mca_test_checks block analysis Previously update_mca_test_checks worked entirely at "block" level where a block is some sequence of lines delimited by at least one empty line. This generally worked well, but could sometimes lead to excessive repetition of check lines for various prefixes if some block was almost identical between prefixes, but not quite (for example, due to a different dispatch width in the otherwise identical summary views). This new analyis attempts to split blocks further in the case where the following conditions are met: a) There is some prefix common to every RUN line (typically 'ALL'). b) The first line of the block is common to the output with every prefix. c) The block has the same number of lines for the output with every prefix. Also, regenerated all llvm-mca test files with the following command: update_mca_test_checks.py "../test/tools/llvm-mca/*/*.s" "../test/tools/llvm-mca/*/*/*.s" The new analysis showed a "multiple lines not disambiguated by prefixes" warning for test "AArch64/Exynos/scheduler-queue-usage.s" so I've also added some explicit prefixes to each of the RUN lines in that test. Differential Revision: https://reviews.llvm.org/D47321 llvm-svn: 333204
2018-05-25 00:36:44 +08:00
# SKX-AVX512-NEXT: Block RThroughput: 0.3
[utils] Ensure that update_mca_test_checks.py writes prefixes in alphabetical order llvm-svn: 343783
2018-10-04 22:42:19 +08:00
# SLM: Dispatch Width: 2
# SLM-NEXT: uOps Per Cycle: 0.97
# SLM-NEXT: IPC: 0.97
# SLM-NEXT: Block RThroughput: 0.5
# ZNVER1: Dispatch Width: 4
# ZNVER1-NEXT: uOps Per Cycle: 0.97
# ZNVER1-NEXT: IPC: 0.97
# ZNVER1-NEXT: Block RThroughput: 0.3
[X86] AMD Znver2 (Rome) Scheduler enablement The patch gives out the details of the znver2 scheduler model. There are few improvements with respect to execution units, latencies and throughput when compared with znver1. The tests that were present for znver1 for llvm-mca tool were replicated. The latencies, execution units, timeline and throughput information are updated for znver2. Reviewers: craig.topper, Simon Pilgrim Differential Revision: https://reviews.llvm.org/D66088
2020-01-09 22:41:32 +08:00
# ZNVER2: Dispatch Width: 4
# ZNVER2-NEXT: uOps Per Cycle: 0.97
# ZNVER2-NEXT: IPC: 0.97
# ZNVER2-NEXT: Block RThroughput: 0.3
[X86] AMD Zen 3 Scheduler Model Introduce basic schedule model for AMD Zen 3 CPU's, a.k.a `znver3`. This is fully built from scratch, from llvm-mca measurements and documented reference materials. Nothing was copied from `znver2`/`znver1`. I believe this is in a reasonable state of completion for inclusion, probably better than D52779 `bdver2` was :) Namely: * uops are pretty spot-on (at least what llvm-mca can measure) {F16422596} * latency is also pretty spot-on (at least what llvm-mca can measure) {F16422601} * throughput is within reason {F16422607} I haven't run much benchmarks with this, however RawSpeed benchmarks says this is beneficial: {F16603978} {F16604029} I'll call out the obvious problems there: * i didn't really bother with X87 instructions * i didn't really bother with obviously-microcoded/system instructions * There are large discrepancy in throughput for `mr` and `rm` instructions. I'm not really sure if it's a modelling defect that needs to be fixed, or it's a defect of measurments. * Pipe distributions are probably bad :) I can't do much here until AMD allows that to be fixed by documenting the appropriate counters and updating libpfm That being said, as @RKSimon notes: >>! In D94395#2647381, @RKSimon wrote: > I'll mention again that all the znver* models appear to be very inaccurate wrt SIMD/FPU instructions <...> so how much worse this could possibly be?! Things that aren't there: * Various tunings: zero idioms, etc. That is follow-ups. Differential Revision: https://reviews.llvm.org/D94395
2021-05-02 03:04:48 +08:00
# ZNVER3: Dispatch Width: 6
# ZNVER3-NEXT: uOps Per Cycle: 0.97
# ZNVER3-NEXT: IPC: 0.97
# ZNVER3-NEXT: Block RThroughput: 0.3