diff options
| author | David Candler <david.candler@arm.com> | 2022-06-20 16:03:59 +0100 |
|---|---|---|
| committer | David Candler <david.candler@arm.com> | 2022-06-20 16:41:46 +0100 |
| commit | d3919a8cc5031fd3d8c1ce3188abc0d7551306fb (patch) | |
| tree | 7de3f9cb712c2031b1752ca692b57ff8aff966da /llvm/lib/CodeGen/LiveDebugVariables.cpp | |
| parent | 398834f45b615f394496fadba5e72cfd4b8be2e5 (diff) | |
| download | llvm-d3919a8cc5031fd3d8c1ce3188abc0d7551306fb.zip llvm-d3919a8cc5031fd3d8c1ce3188abc0d7551306fb.tar.gz llvm-d3919a8cc5031fd3d8c1ce3188abc0d7551306fb.tar.bz2 | |
[ConstantFolding] Respect denormal handling mode attributes when folding instructions
Depending on the environment, a floating point instruction should
treat denormal inputs as zero, and/or flush a denormal output to zero.
Denormals are not currently accounted for when an instruction gets
folded to a constant, which can lead to differences in output between
a folded and a unfolded instruction when running on the target. The
denormal handling mode can be set by the function level attribute
denormal-fp-math, which this patch uses to determine whether any
denormal inputs to or outputs from folding should be zero, and that
the sign is set appropriately.
Reviewed By: spatel
Differential Revision: https://reviews.llvm.org/D116952
Diffstat (limited to 'llvm/lib/CodeGen/LiveDebugVariables.cpp')
0 files changed, 0 insertions, 0 deletions