[PATCH] D73915: AMDGPU: Add flag to control mem intrinsic expansion
Matt Arsenault via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Mon Feb 3 13:33:32 PST 2020
arsenm marked an inline comment as done.
arsenm added inline comments.
================
Comment at: llvm/lib/Target/AMDGPU/AMDGPULowerIntrinsics.cpp:68
ConstantInt *CI = dyn_cast<ConstantInt>(Size);
- return !CI || (CI->getZExtValue() > MaxStaticSize);
+ return !CI || (CI->getSExtValue() > MaxStaticSize);
}
----------------
rampitec wrote:
> arsenm wrote:
> > rampitec wrote:
> > > Why signed?
> > To handle using negative values with the flag
> But why do you even need it negative?
To make sure the edge case of 0 is expanded or not. It should be possible to have the 0 case expanded or not
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D73915/new/
https://reviews.llvm.org/D73915
More information about the llvm-commits
mailing list