Skip to content
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

AMDGPU: Legalize fminimum and fmaximum f32 for gfx950 #117634

Merged
merged 1 commit into from
Nov 26, 2024
Merged

AMDGPU: Legalize fminimum and fmaximum f32 for gfx950 #117634

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
4 changes: 4 additions & 0 deletions llvm/lib/Target/AMDGPU/SIISelLowering.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -855,6 +855,10 @@ SITargetLowering::SITargetLowering(const TargetMachine &TM,
setOperationAction({ISD::FMINIMUM, ISD::FMAXIMUM},
{MVT::v4f16, MVT::v8f16, MVT::v16f16, MVT::v32f16},
Custom);
} else {
// FIXME: For nnan fmaximum, emit the fmaximum3 instead of fmaxnum
if (Subtarget->hasMinimum3Maximum3F32())
setOperationAction({ISD::FMAXIMUM, ISD::FMINIMUM}, MVT::f32, Legal);
}

setOperationAction(ISD::INTRINSIC_WO_CHAIN,
Expand Down
17 changes: 17 additions & 0 deletions llvm/lib/Target/AMDGPU/VOP3Instructions.td
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1234,6 +1234,23 @@ def : IntClampPat<V_MQSAD_PK_U16_U8_e64, int_amdgcn_mqsad_pk_u16_u8>;
def : IntClampPat<V_QSAD_PK_U16_U8_e64, int_amdgcn_qsad_pk_u16_u8>;
def : IntClampPat<V_MQSAD_U32_U8_e64, int_amdgcn_mqsad_u32_u8>;

//===----------------------------------------------------------------------===//
// Floating-point operation Patterns
//===----------------------------------------------------------------------===//

// Implement fminimum(x, y) by using minimum3(x, y, y)
class MinimumMaximumByMinimum3Maximum3<SDPatternOperator node, ValueType vt,
Instruction inst> : GCNPat<
(vt (node (VOP3Mods vt:$src0, i32:$src0_mods), (VOP3Mods vt:$src1, i32:$src1_mods))),
(inst $src0_mods, $src0, $src1_mods, $src1, $src1_mods, $src1)
>;

// Prefer the real 2 operand form if legal
let SubtargetPredicate = HasMinimum3Maximum3F32, AddedComplexity = -1000 in {
Copy link
Contributor Author

Choose a reason for hiding this comment

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

The -1000 is really leftover debugging, will fix after the merge is completed

def : MinimumMaximumByMinimum3Maximum3<fminimum, f32, V_MINIMUM3_F32_e64>;
def : MinimumMaximumByMinimum3Maximum3<fmaximum, f32, V_MAXIMUM3_F32_e64>;
}

//===----------------------------------------------------------------------===//
// Target-specific instruction encodings.
//===----------------------------------------------------------------------===//
Expand Down
Loading