LLVM should be smarter about *known* malloc's alignment and this knowledge may enable other optimizations.
Diff Detail
Diff Detail
- Repository
- rG LLVM Github Monorepo
Event Timeline
llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp | ||
---|---|---|
2052 | Basically reused info from Clang: | |
2068 | Open question for review |
Comment Actions
Yeah. But on the other side, LLVM already does a lot ot work with libcalls and attrs (it should be Clang job, but...).
But yes, doing this in Clang seems to be a better idea.
clang-tidy: warning: invalid case style for function 'GuaranteedAlignmentForAllocFns' [readability-identifier-naming]
not useful