projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: c0c3a6e) | patch
AMDGPU: Don't handle FP16_TO_FP in isCanonicalized
authorMatt Arsenault <Matthew.Arsenault@amd.com>
Tue, 31 Jul 2018 14:15:16 +0000 (14:15 +0000)
committerMatt Arsenault <Matthew.Arsenault@amd.com>
Tue, 31 Jul 2018 14:15:16 +0000 (14:15 +0000)
commit638a202760e222f7a8ac80ec2e4fb0412b46a3c7
treeb48dbfc42a52f496f1d68bd9ba762b6ed23867a8tree | snapshot
parentc0c3a6ed5e3b8a8ae031ee0beb89ee488a57d080commit | diff
AMDGPU: Don't handle FP16_TO_FP in isCanonicalized

This needs more special handling to do correctly.
Fixes test in subsequent commit.

llvm-svn: 338381
llvm/lib/Target/AMDGPU/SIISelLowering.cpp diff | blob | history
Domain: System / Toolchain;