Skip to content

Fixes #106142. #106369

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
tannergooding merged 2 commits into from
Aug 14, 2024
Merged

Fixes #106142. #106369

Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
9d764fa
Fixes #106142.
DeepakRajendrakumaran Aug 13, 2024
99bd44a
Removing if stmt and adding test.
DeepakRajendrakumaran Aug 14, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
24 changes: 20 additions & 4 deletions src/coreclr/jit/simd.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -802,17 +802,33 @@ TBase EvaluateBinaryScalarSpecialized(genTreeOps oper, TBase arg0, TBase arg1)
case GT_ROL:
{
// Normalize the "rotate by" value
arg1 %= (sizeof(TBase) * BITS_PER_BYTE);
Copy link
Member

@tannergooding tannergooding Aug 14, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Was the behavioral difference for negative inputs, since %= is remainder of truncating division and so handles signed values differently?

Copy link
Contributor Author

@DeepakRajendrakumaran DeepakRajendrakumaran Aug 14, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yep. It was different for negative inputs. For eg. one of the test case has the rotate count show up as -2 in here.

if ((arg1 < 0) || (arg1 >= (sizeof(TBase) * BITS_PER_BYTE)))
{
// EvaluateBinaryScalarRSZ allows overshifting and treats
// it as zeroing.
// But ROL ensures the rotateAmount is masked
// to be within range, so we pre-calculates this.
unsigned rotateCountMask = (sizeof(TBase) * BITS_PER_BYTE) - 1;
arg1 &= rotateCountMask;
}
return EvaluateBinaryScalarSpecialized<TBase>(GT_LSH, arg0, arg1) |
EvaluateBinaryScalarRSZ<TBase>(arg0, (sizeof(TBase) * 8) - arg1);
EvaluateBinaryScalarRSZ<TBase>(arg0, (sizeof(TBase) * BITS_PER_BYTE) - arg1);
}

case GT_ROR:
{
// Normalize the "rotate by" value
arg1 %= (sizeof(TBase) * BITS_PER_BYTE);
if ((arg1 < 0) || (arg1 >= (sizeof(TBase) * BITS_PER_BYTE)))
{
// EvaluateBinaryScalarRSZ allows overshifting and treats
// it as zeroing.
// But ROR ensures the rotateAmount is masked
// to be within range, so we pre-calculates this.
unsigned rotateCountMask = (sizeof(TBase) * BITS_PER_BYTE) - 1;
arg1 &= rotateCountMask;
}
return EvaluateBinaryScalarRSZ<TBase>(arg0, arg1) |
EvaluateBinaryScalarSpecialized<TBase>(GT_LSH, arg0, (sizeof(TBase) * 8) - arg1);
EvaluateBinaryScalarSpecialized<TBase>(GT_LSH, arg0, (sizeof(TBase) * BITS_PER_BYTE) - arg1);
}

case GT_RSH:
Expand Down