Files
clang-p2996/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.struct.buffer.load.tfe.ll
Stanislav Mekhanoshin 3277c7cd28 [AMDGPU] Skip VGPR deallocation for waveslot limited kernels (#112765)
MSG_DEALLOC_VGPRS slows down very small waveslot limited kernels. It's
been identified this message is only really needed for VGPR limited
kernels. A kernel becomes VGPR limited if a total number of VGPRs per
SIMD / number of used VGPRs is more than a number of wave slots.
2024-10-21 09:39:52 -07:00

32 KiB