[PATCH] D69878: Consoldiate internal denormal flushing controls
Anastasia Stulova via Phabricator via cfe-commits
cfe-commits at lists.llvm.org
Thu Nov 7 04:37:50 PST 2019
Anastasia added a comment.
> Stop emitting the denorms-are-zero attribute for the OpenCL flag. It
> has no in-tree users. The meaning would also be target dependent, such
> as the AMDGPU choice to treat this as only meaning allow flushing of
> f32 and not f16 or f64. The naming is also potentially confusing,
> since DAZ in other contexts refers to instructions implicitly treating
> input denormals as zero, not necessarily flushing output denormals to
> zero.
Would the targets supporting OpenCL need to define their own behavior in `getDefaultDenormalModeForType`?
================
Comment at: clang/include/clang/Driver/ToolChain.h:619
+ const llvm::fltSemantics *FPType = nullptr) const {
+ // FIXME: This should be IEEE when default handling is fixed.
+ return llvm::DenormalMode::Invalid;
----------------
Can you elaborate what has to be done in order to fix this?
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D69878/new/
https://reviews.llvm.org/D69878
More information about the cfe-commits
mailing list