[all-commits] [llvm/llvm-project] 5f2acf: [ScalarEvolutionExpanderTest] Avoid sign warning

[Roman Lebedev via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 5f2acfb6356f012a9191e8ea931ce7be686e1ba8
      https://github.com/llvm/llvm-project/commit/5f2acfb6356f012a9191e8ea931ce7be686e1ba8
  Author: Roman Lebedev <lebedev.ri at gmail.com>
  Date:   2023-01-05 (Thu, 05 Jan 2023)

  Changed paths:
    M llvm/unittests/Transforms/Utils/ScalarEvolutionExpanderTest.cpp

  Log Message:
  -----------
  [ScalarEvolutionExpanderTest] Avoid sign warning


  Commit: e0ad2af6916a6c743e8ff19489818381e697d4ee
      https://github.com/llvm/llvm-project/commit/e0ad2af6916a6c743e8ff19489818381e697d4ee
  Author: Roman Lebedev <lebedev.ri at gmail.com>
  Date:   2023-01-05 (Thu, 05 Jan 2023)

  Changed paths:
    M llvm/docs/CommandGuide/llvm-exegesis.rst
    M llvm/test/tools/llvm-exegesis/PowerPC/latency-by-opcode-name.s
    M llvm/test/tools/llvm-exegesis/PowerPC/unsupported-opcode.s
    M llvm/test/tools/llvm-exegesis/X86/inverse_throughput/inverse_throughput-PBLENDVBrr0.s
    M llvm/test/tools/llvm-exegesis/X86/inverse_throughput/inverse_throughput-by-opcode-name.s
    M llvm/test/tools/llvm-exegesis/X86/inverse_throughput/inverse_throughput-disable-upper-sse-registers.s
    M llvm/test/tools/llvm-exegesis/X86/latency/dump-object-to-disk.s
    M llvm/test/tools/llvm-exegesis/X86/latency/latency-CMOV32rr.s
    M llvm/test/tools/llvm-exegesis/X86/latency/latency-CVTSD2SI64rr.s
    M llvm/test/tools/llvm-exegesis/X86/latency/latency-IN16rr.s
    M llvm/test/tools/llvm-exegesis/X86/latency/latency-LEA64_32r.s
    M llvm/test/tools/llvm-exegesis/X86/latency/latency-LEA64r.s
    M llvm/test/tools/llvm-exegesis/X86/latency/latency-SBB8rr.s
    M llvm/test/tools/llvm-exegesis/X86/latency/latency-SETCCr-cond-codes-sweep.s
    M llvm/test/tools/llvm-exegesis/X86/latency/latency-SQRTSSr.s
    M llvm/test/tools/llvm-exegesis/X86/latency/latency-SYSENTER.s
    M llvm/test/tools/llvm-exegesis/X86/latency/latency-WRFSBASE.s
    M llvm/test/tools/llvm-exegesis/X86/latency/latency-by-opcode-name.s
    M llvm/test/tools/llvm-exegesis/X86/latency/max-configs.test
    A llvm/test/tools/llvm-exegesis/X86/latency/skip-codegen.s
    M llvm/test/tools/llvm-exegesis/X86/uops/uops-ADD32mi8.s
    M llvm/test/tools/llvm-exegesis/X86/uops/uops-ADD32mr.s
    M llvm/test/tools/llvm-exegesis/X86/uops/uops-ADD32rm.s
    M llvm/test/tools/llvm-exegesis/X86/uops/uops-ADD_F32m.s
    M llvm/test/tools/llvm-exegesis/X86/uops/uops-BEXTR32rm.s
    M llvm/test/tools/llvm-exegesis/X86/uops/uops-BSF16rm.s
    M llvm/test/tools/llvm-exegesis/X86/uops/uops-BTR64mr.s
    M llvm/test/tools/llvm-exegesis/X86/uops/uops-CMOV16rm-noreg-serialization.s
    M llvm/test/tools/llvm-exegesis/X86/uops/uops-FLDENVm.s
    M llvm/test/tools/llvm-exegesis/X86/uops/uops-HLT.s
    M llvm/test/tools/llvm-exegesis/X86/uops/uops-LEA64r.s
    M llvm/test/tools/llvm-exegesis/X86/uops/uops-POPCNT32rr.s
    M llvm/test/tools/llvm-exegesis/X86/uops/uops-STD.s
    M llvm/test/tools/llvm-exegesis/X86/uops/uops-VFMADDSS4rm.s
    M llvm/test/tools/llvm-exegesis/X86/uops/uops-XCHG64rr.s
    M llvm/test/tools/llvm-exegesis/X86/uops/uops-by-opcode-name.s
    M llvm/test/tools/llvm-exegesis/X86/uops/uops-misspelled-div.s
    M llvm/tools/llvm-exegesis/lib/BenchmarkResult.h
    M llvm/tools/llvm-exegesis/lib/BenchmarkRunner.cpp
    M llvm/tools/llvm-exegesis/lib/BenchmarkRunner.h
    M llvm/tools/llvm-exegesis/lib/LatencyBenchmarkRunner.cpp
    M llvm/tools/llvm-exegesis/lib/LatencyBenchmarkRunner.h
    M llvm/tools/llvm-exegesis/lib/Target.cpp
    M llvm/tools/llvm-exegesis/lib/Target.h
    M llvm/tools/llvm-exegesis/lib/UopsBenchmarkRunner.h
    M llvm/tools/llvm-exegesis/llvm-exegesis.cpp

  Log Message:
  -----------
  [exegesis] "Skip codegen" dry-run mode

While "skip measurements mode" is super useful for test coverage,
i've come to discover it's trade-offs. It still calls back-end
to actually codegen the target assembly, and that is what is taking
80%+ of the time regardless of whether or not we skip the measurements.

On the other hand, just being able to see that exegesis can come up
with a snippet to measure something, is already very useful,
and takes maybe a second for a all-opcode sweep.

Reviewed By: gchatelet

Differential Revision: https://reviews.llvm.org/D140702


Compare: https://github.com/llvm/llvm-project/compare/85d049a089d4...e0ad2af6916a