This can be thought as a subcase of x & ((1 << nbits) - 1) where x == -1
Details
Diff Detail
- Repository
- rG LLVM Github Monorepo
Event Timeline
llvm/test/CodeGen/X86/extract-lowbits.ll | ||
---|---|---|
3036 | Yes, just don't see any tests with a mask. vreg_width - 1 really. I expect it to work fine, just want to watch out for the fact that shifts can ignore word mask, but bzhi can't |
llvm/test/CodeGen/AArch64/extract-lowbits.ll | ||
---|---|---|
24 ↗ | (On Diff #541968) | This is a X86 patch - skip the aarch64 regeneration changes (or commit separately) |
llvm/test/CodeGen/AArch64/extract-lowbits.ll | ||
---|---|---|
24 ↗ | (On Diff #541968) | Ok, it's just this test explicitly asks to keep them in sync. |
llvm/test/CodeGen/X86/extract-lowbits.ll | ||
---|---|---|
516 | Can you split the new tests to a seperate patch? I.e Then submit both to phab and use "Edit Related Revisions" to make the tests patch the child of the impl patch. That way it easier to track the changes caused by the impl itself. |
llvm/test/CodeGen/X86/extract-lowbits.ll | ||
---|---|---|
516 | commit 1 would be empty in this case. Everything below this line is just label renumbering caused by insertion of the new test function. |
llvm/test/CodeGen/X86/extract-lowbits.ll | ||
---|---|---|
516 | Disregard previous reply. Phabricator is being unwieldy. Tests are now in https://reviews.llvm.org/D155734 |
nit: needs clang format.