projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 6f5cdc9) | patch
AMDGPU: Lower addrspacecast on gfx6
authorMatt Arsenault <Matthew.Arsenault@amd.com>
Sat, 17 Dec 2022 21:24:38 +0000 (16:24 -0500)
committerMatt Arsenault <Matthew.Arsenault@amd.com>
Sun, 18 Dec 2022 13:02:45 +0000 (08:02 -0500)
commit9d6003c764fc4754959e4d7380f7c9d2559299f4
tree21319f9bde3fc9798999505cb407fdd947f25657tree | snapshot
parent6f5cdc93cd7656a58673a125bc37d476a6ce1c20commit | diff
AMDGPU: Lower addrspacecast on gfx6

Fixes inconsistent handling of constant-32bit case. Turns out we can
lower all the casts just fine, it's just accessing the flat results
that's a problem.
llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp diff | blob | history
llvm/lib/Target/AMDGPU/SIISelLowering.cpp diff | blob | history
llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-addrspacecast.mir diff | blob | history
llvm/test/CodeGen/AMDGPU/addrspacecast.gfx6.ll [new file with mode: 0644] blob
Domain: System / Toolchain;