projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 1c7037a) | patch
AMDGPU: Don't assert on f16 inv2pi immediates pre-gfx8
authorMatt Arsenault <Matthew.Arsenault@amd.com>
Wed, 22 Jul 2020 16:27:50 +0000 (12:27 -0400)
committerMatt Arsenault <Matthew.Arsenault@amd.com>
Wed, 22 Jul 2020 17:59:03 +0000 (13:59 -0400)
commit6f437117af2fa9f380b0d866cd203aad6c30c9b1
tree8bb32a1ba271503ecc2c1ee8c3f189d416ee1bc8tree | snapshot
parent1c7037a2a5576d0bb083db10ad947a8308e61f65commit | diff
AMDGPU: Don't assert on f16 inv2pi immediates pre-gfx8

v_cvt_f32_f16 can still accept this value as a literal constant. This
showed up in GlobalISel since it doesn't have constant folding for
G_FPEXT.
llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUInstPrinter.cpp diff | blob | history
llvm/test/MC/AMDGPU/inline-imm-inv2pi.s [new file with mode: 0644] blob
Domain: System / Toolchain;