[PATCH] D56454: AMDGPU: Adjust the chain for loads writing to the HI part of a register.
Stanislav Mekhanoshin via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Tue Jan 8 14:37:53 PST 2019
rampitec added a comment.
Can you add a test where low half does not produce a chain? An arithmetic operation and an undef.
================
Comment at: lib/Target/AMDGPU/SIISelLowering.cpp:9027
+
+ SmallVector<SDValue, 4> Ops;
+ for (unsigned I = 0; I < NumOps-1; ++I)
----------------
4 is not enough, it will not be a small vector. I think 16 is ok.
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D56454/new/
https://reviews.llvm.org/D56454
More information about the llvm-commits
mailing list