Reapply "[msan] Add avx512-intrinsics.ll and avx512-intrinsics-upgrade.ll test case (#123980)" #124500
Conversation
This is forked from llvm/test/CodeGen/X86/avx512-intrinsics-upgrade.ll. It tests intrinsics that LLVM "auto-upgrades"; for example, @llvm.x86.avx512.mask.store is converted into @llvm.masked.store (which has the interesting side effect that MemorySanitizer can already handle it via its existing handleMaskedStore).
thurstond
requested review from
vitalybuka,
alexander-shaposhnikov and
kstoimenov
|
✅ With the latest revision this PR passed the undef deprecator. |
It can be more convenient to review reland, it consist of 2 commits, "revert of revert", and fixes Is possible to fix "undef deprecator" ? |
thurstond
force-pushed
the
reland_msan_avx512_tests
branch
from
7736f61 to
8555d06
Compare
I've split the commit into "revert of revert" + the fix.
Since these are MSan tests, and we want to know how MSan deals with uninitialized memory, undef seems appropriate? |
Undef handling is simpler and have dedicated tests. |
thurstond
force-pushed
the
reland_msan_avx512_tests
branch
from
8555d06 to
3e0deee
Compare
|
Failing tests in CI (https://buildkite.com/llvm-project/github-pull-requests/builds/141497) are unrelated: LLVM.CodeGen/AArch64/aarch64-build-attributes-all.ll |
This reverts commit b2647ff i.e., relands 980e86f.
I had reverted the original patch because of buildbot failures (e.g., https://lab.llvm.org/buildbot/#/builders/13/builds/4938/steps/5/logs/FAIL__LLVM__avx512-intrinsics-upgrade_ll). This reland addresses the likely root cause; namely, that the target datalayout had not been specified.