Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

pre-commit: test PR79688 #157

Closed
wants to merge 2 commits into from
Closed

pre-commit: test PR79688 #157

wants to merge 2 commits into from

Conversation

dtcxzyw
Copy link
Owner

@dtcxzyw dtcxzyw commented Jan 27, 2024

@github-actions GitHub Actions
Copy link
Contributor

baseline: llvm/llvm-project@3855757
patch: llvm/llvm-project#79688
sha256: f3161914bfea3706110d0a86fecfe31fae651cb2c68f6dc3d24a8351b2ca7f84
commit: 3edff1a
59 files changed, 18071 insertions(+), 18158 deletions(-)
3edff1a pre-commit: Update
bench/cpp-httplib/optimized/httplib.cc.ll
bench/cvc5/optimized/options_handler.cpp.ll
bench/cxxopts/optimized/example.cpp.ll
bench/folly/optimized/Conv.cpp.ll
bench/imgui/optimized/imgui.cpp.ll
bench/libquic/optimized/file_util_posix.cc.ll
bench/llama.cpp/optimized/common.cpp.ll
bench/luajit/optimized/lj_asm.ll
bench/mold/optimized/arch-riscv.cc.RV32BE.cc.ll
bench/mold/optimized/arch-riscv.cc.RV32LE.cc.ll
bench/mold/optimized/arch-riscv.cc.RV64BE.cc.ll
bench/mold/optimized/arch-riscv.cc.RV64LE.cc.ll
bench/mold/optimized/cmdline.cc.ALPHA.cc.ll
bench/mold/optimized/cmdline.cc.ARM32.cc.ll
bench/mold/optimized/cmdline.cc.ARM64.cc.ll
bench/mold/optimized/cmdline.cc.I386.cc.ll
bench/mold/optimized/cmdline.cc.LOONGARCH32.cc.ll
bench/mold/optimized/cmdline.cc.LOONGARCH64.cc.ll
bench/mold/optimized/cmdline.cc.M68K.cc.ll
bench/mold/optimized/cmdline.cc.PPC32.cc.ll
bench/mold/optimized/cmdline.cc.PPC64V1.cc.ll
bench/mold/optimized/cmdline.cc.PPC64V2.cc.ll
bench/mold/optimized/cmdline.cc.RV32BE.cc.ll
bench/mold/optimized/cmdline.cc.RV32LE.cc.ll
bench/mold/optimized/cmdline.cc.RV64BE.cc.ll
bench/mold/optimized/cmdline.cc.RV64LE.cc.ll
bench/mold/optimized/cmdline.cc.S390X.cc.ll
bench/mold/optimized/cmdline.cc.SH4.cc.ll
bench/mold/optimized/cmdline.cc.SPARC64.cc.ll
bench/mold/optimized/cmdline.cc.X86_64.cc.ll
bench/mold/optimized/multi-glob.cc.ll
bench/mold/optimized/passes.cc.ALPHA.cc.ll
bench/mold/optimized/passes.cc.ARM32.cc.ll
bench/mold/optimized/passes.cc.ARM64.cc.ll
bench/mold/optimized/passes.cc.I386.cc.ll
bench/mold/optimized/passes.cc.LOONGARCH32.cc.ll
bench/mold/optimized/passes.cc.LOONGARCH64.cc.ll
bench/mold/optimized/passes.cc.M68K.cc.ll
bench/mold/optimized/passes.cc.PPC32.cc.ll
bench/mold/optimized/passes.cc.PPC64V1.cc.ll
bench/mold/optimized/passes.cc.PPC64V2.cc.ll
bench/mold/optimized/passes.cc.RV32BE.cc.ll
bench/mold/optimized/passes.cc.RV32LE.cc.ll
bench/mold/optimized/passes.cc.RV64BE.cc.ll
bench/mold/optimized/passes.cc.RV64LE.cc.ll
bench/mold/optimized/passes.cc.S390X.cc.ll
bench/mold/optimized/passes.cc.SH4.cc.ll
bench/mold/optimized/passes.cc.SPARC64.cc.ll
bench/mold/optimized/passes.cc.X86_64.cc.ll
bench/ms-gsl/optimized/span_tests.cpp.ll
bench/ocio/optimized/BuiltinConfigRegistry.cpp.ll
bench/ocio/optimized/Config.cpp.ll
bench/ocio/optimized/FileRules.cpp.ll
bench/oiio/optimized/filesystem.cpp.ll
bench/oiio/optimized/formatspec.cpp.ll
bench/oiio/optimized/imagecache.cpp.ll
bench/qemu/optimized/accel_tcg_cpu-exec.c.ll

@XChy
Copy link

XChy commented Jan 28, 2024

Could you rerun it again? I think it covers the original real-world ir now. I'm curious about the result 😄

@dtcxzyw dtcxzyw closed this Jan 28, 2024
@dtcxzyw dtcxzyw reopened this Jan 28, 2024
@github-actions GitHub Actions
Copy link
Contributor

baseline: llvm/llvm-project@7089c01
patch: llvm/llvm-project#79688
sha256: 81acde680fe370af12e371ba7286740b9b0f5d06999cbc3ed7f32f732f4aae06
commit: 38946da
257 files changed, 63400 insertions(+), 64584 deletions(-)
38946da pre-commit: Update
bench/abseil-cpp/optimized/bind.cc.ll
bench/abseil-cpp/optimized/civil_time_test.cc.ll
bench/abseil-cpp/optimized/exponential_biased.cc.ll
bench/abseil-cpp/optimized/float_conversion.cc.ll
bench/abseil-cpp/optimized/generators_test.cc.ll
bench/abseil-cpp/optimized/marshalling_test.cc.ll
bench/abseil-cpp/optimized/str_cat_test.cc.ll
bench/arrow/optimized/bitmap_ops.cc.ll
bench/assimp/optimized/FBXTokenizer.cpp.ll
bench/assimp/optimized/X3DImporter_Geometry3D.cpp.ll
bench/assimp/optimized/clipper.cpp.ll
bench/cpp-httplib/optimized/httplib.cc.ll
bench/cpython/optimized/cfield.ll
bench/curl/optimized/libcurl_la-http.ll
bench/curl/optimized/libcurl_la-multi.ll
bench/cvc5/optimized/bag_solver.cpp.ll
bench/cvc5/optimized/ite_utilities.cpp.ll
bench/cvc5/optimized/linear_equality.cpp.ll
bench/cvc5/optimized/options_handler.cpp.ll
bench/cxxopts/optimized/example.cpp.ll
bench/duckdb/optimized/ub_duckdb_func_string_main.cpp.ll
bench/eastl/optimized/BenchmarkAlgorithm.cpp.ll
bench/eastl/optimized/EASprintfCore.cpp.ll
bench/entt/optimized/version.cpp.ll
bench/folly/optimized/Conv.cpp.ll
bench/folly/optimized/EpollBackend.cpp.ll
bench/git/optimized/apply.ll
bench/git/optimized/commit-graph.ll
bench/git/optimized/fsck.ll
bench/git/optimized/trailer.ll
bench/grpc/optimized/hpack_parser.cc.ll
bench/grpc/optimized/legacy_inproc_transport.cc.ll
bench/hermes/optimized/Array.cpp.ll
bench/hermes/optimized/BytecodeFormConverter.cpp.ll
bench/hermes/optimized/CSE.cpp.ll
bench/hermes/optimized/Callable.cpp.ll
bench/hermes/optimized/CodeMotion.cpp.ll
bench/hermes/optimized/Error.cpp.ll
bench/hermes/optimized/Host.cpp.ll
bench/hermes/optimized/JSCallableProxy.cpp.ll
bench/hermes/optimized/JSError.cpp.ll
bench/hermes/optimized/JSLexer.cpp.ll
bench/hermes/optimized/JSObject.cpp.ll
bench/hermes/optimized/Lowering.cpp.ll
bench/hermes/optimized/Operations.cpp.ll
bench/hermes/optimized/OrderedHashMap.cpp.ll
bench/hermes/optimized/SimpleStackPromotion.cpp.ll
bench/hermes/optimized/String.cpp.ll
bench/hermes/optimized/TDZDedup.cpp.ll
bench/hermes/optimized/TypeInference.cpp.ll
bench/hyperscan/optimized/Utf8ComponentClass.cpp.ll
bench/hyperscan/optimized/ng_literal_component.cpp.ll
bench/hyperscan/optimized/ng_misc_opt.cpp.ll
bench/hyperscan/optimized/smallwrite_build.cpp.ll
bench/icu/optimized/collationbuilder.ll
bench/icu/optimized/collationrootelements.ll
bench/icu/optimized/locdistance.ll
bench/libquic/optimized/e_rc2.c.ll
bench/libquic/optimized/file_util_posix.cc.ll
bench/libquic/optimized/spdy_frame_builder.cc.ll
bench/libquic/optimized/trees.c.ll
bench/libsodium/optimized/libsodium_la-ed25519_ref10.ll
bench/libuv/optimized/idna.c.ll
bench/libzmq/optimized/proxy.cpp.ll
bench/llama.cpp/optimized/common.cpp.ll
bench/lodepng/optimized/pngdetail.cpp.ll
bench/luajit/optimized/lj_parse.ll
bench/luajit/optimized/lj_parse_dyn.ll
bench/miniaudio/optimized/unity.c.ll
bench/mold/optimized/arch-arm32.cc.ll
bench/mold/optimized/arch-riscv.cc.RV32BE.cc.ll
bench/mold/optimized/arch-riscv.cc.RV32LE.cc.ll
bench/mold/optimized/arch-riscv.cc.RV64BE.cc.ll
bench/mold/optimized/arch-riscv.cc.RV64LE.cc.ll
bench/mold/optimized/cmdline.cc.ALPHA.cc.ll
bench/mold/optimized/cmdline.cc.ARM32.cc.ll
bench/mold/optimized/cmdline.cc.ARM64.cc.ll
bench/mold/optimized/cmdline.cc.I386.cc.ll
bench/mold/optimized/cmdline.cc.LOONGARCH32.cc.ll
bench/mold/optimized/cmdline.cc.LOONGARCH64.cc.ll
bench/mold/optimized/cmdline.cc.M68K.cc.ll
bench/mold/optimized/cmdline.cc.PPC32.cc.ll
bench/mold/optimized/cmdline.cc.PPC64V1.cc.ll
bench/mold/optimized/cmdline.cc.PPC64V2.cc.ll
bench/mold/optimized/cmdline.cc.RV32BE.cc.ll
bench/mold/optimized/cmdline.cc.RV32LE.cc.ll
bench/mold/optimized/cmdline.cc.RV64BE.cc.ll
bench/mold/optimized/cmdline.cc.RV64LE.cc.ll
bench/mold/optimized/cmdline.cc.S390X.cc.ll
bench/mold/optimized/cmdline.cc.SH4.cc.ll
bench/mold/optimized/cmdline.cc.SPARC64.cc.ll
bench/mold/optimized/cmdline.cc.X86_64.cc.ll
bench/mold/optimized/multi-glob.cc.ll
bench/mold/optimized/passes.cc.ALPHA.cc.ll
bench/mold/optimized/passes.cc.ARM32.cc.ll
bench/mold/optimized/passes.cc.ARM64.cc.ll
bench/mold/optimized/passes.cc.I386.cc.ll
bench/mold/optimized/passes.cc.LOONGARCH32.cc.ll
bench/mold/optimized/passes.cc.LOONGARCH64.cc.ll

XChy
XChy reviewed Jan 28, 2024
View reviewed changes
Copy link

@XChy XChy left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It seems that the real-world ir in original issue doesn't folds util llvm/llvm-project#77231 is merged. But the effect is actually good.

bench/hermes/optimized/Host.cpp.ll
%Features.16.ph.i = select i1 %31, i32 %Features.14.ph.i, i32 %spec.select74.i
%or135.i = or i32 %Features.16.ph.i, 4194304
%spec.select75.i = select i1 %14, i32 %or135.i, i32 %Features.14.ph.i
%32 = extractelement <8 x i1> %29, i64 5
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

To my surprise, a large portion of ir is optimized away

@dtcxzyw dtcxzyw closed this Jan 31, 2024
@dtcxzyw dtcxzyw deleted the test-pr79688 branch September 16, 2024 12:06
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@XChy XChy XChy left review comments

Assignees
No one assigned
Labels
reviewed
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@dtcxzyw @XChy