[AMDGPU] [DO NOT MERGE] Nonsuccessful Attempt At Using SelectionDAG Hooks for abs i8/i16 #167064
+1,524
−608
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…rom this approach.
You can test this locally with the following command:git-clang-format --diff origin/main HEAD --extensions cpp,h -- llvm/lib/Target/AMDGPU/SIISelLowering.cpp llvm/lib/Target/AMDGPU/SIISelLowering.h --diff_from_common_commit
View the diff from clang-format here.diff --git a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
index 95910b68f..de6394b76 100644
--- a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
+++ b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
@@ -181,7 +181,7 @@ SITargetLowering::SITargetLowering(const TargetMachine &TM,
// We don't want the default expansion of 16-bit ABS since we can
// sign-extend and use the 32-bit ABS operation for 16-bit ABS with SGPRs
- setOperationAction(ISD::ABS, {MVT::i8,MVT::i16}, Custom);
+ setOperationAction(ISD::ABS, {MVT::i8, MVT::i16}, Custom);
}
addRegisterClass(MVT::v32i32, &AMDGPU::VReg_1024RegClass);
@@ -7281,7 +7281,7 @@ static SDValue lowerLaneOp(const SITargetLowering &TLI, SDNode *N,
void SITargetLowering::ReplaceNodeResults(SDNode *N,
SmallVectorImpl<SDValue> &Results,
SelectionDAG &DAG) const {
- switch (N->getOpcode()) {
+ switch (N->getOpcode()) {
case ISD::INSERT_VECTOR_ELT: {
if (SDValue Res = lowerINSERT_VECTOR_ELT(SDValue(N, 0), DAG))
Results.push_back(Res);
@@ -7462,7 +7462,7 @@ void SITargetLowering::ReplaceNodeResults(SDNode *N,
case ISD::ABS:
if (N->getValueType(0) == MVT::i16 || N->getValueType(0) == MVT::i8) {
SDValue result = lowerABSi16(SDValue(N, 0), DAG);
- if(result!=SDValue()) {
+ if (result != SDValue()) {
Results.push_back(result);
return;
}
@@ -16894,7 +16894,7 @@ SDValue SITargetLowering::PerformDAGCombine(SDNode *N,
switch (N->getOpcode()) {
case ISD::ABS:
if (N->getValueType(0) == MVT::i16 || N->getValueType(0) == MVT::i8)
- return lowerABSi16(SDValue(N,0), DCI.DAG);
+ return lowerABSi16(SDValue(N, 0), DCI.DAG);
break;
case ISD::ADD:
return performAddCombine(N, DCI);
|
arsenm
reviewed
Nov 8, 2025
|
|
||
| // We don't want the default expansion of 16-bit ABS since we can | ||
| // sign-extend and use the 32-bit ABS operation for 16-bit ABS with SGPRs | ||
| setOperationAction(ISD::ABS, {MVT::i8,MVT::i16}, Custom); |
Contributor
There was a problem hiding this comment.
It will be easier to ignore i8 for now and just get i16 working. i8 adds way more problems if you're having trouble with the easy case
| @@ -7450,6 +7459,15 @@ void SITargetLowering::ReplaceNodeResults(SDNode *N, | |||
| Results.push_back(lowerFSQRTF16(SDValue(N, 0), DAG)); | |||
| break; | |||
| } | |||
| case ISD::ABS: | |||
| if (N->getValueType(0) == MVT::i16 || N->getValueType(0) == MVT::i8) { | |||
Contributor
There was a problem hiding this comment.
i16 won't reach here in the relevant case. It would happen for <= gfx7
| assert((Op.getValueType() == MVT::i16 || Op.getValueType() == MVT::i8) && | ||
| "Tried to select abs i16 lowering with non-i16 type."); | ||
|
|
||
| // divergent means will not end up using SGPRs |
Contributor
There was a problem hiding this comment.
This is stronger of an assertion than is true, it's just more likely
| if (Op->isDivergent()) | ||
| return SDValue(); | ||
|
|
||
| //(abs i16 (i16 op1)) -> (trunc i16 (abs i32 (sext i32 (i16 op1)))) |
Contributor
There was a problem hiding this comment.
Suggested change
| //(abs i16 (i16 op1)) -> (trunc i16 (abs i32 (sext i32 (i16 op1)))) | |
| // (abs i16 (i16 op1)) -> (trunc i16 (abs i32 (sext i32 (i16 op1)))) |
|
|
||
| // divergent means will not end up using SGPRs | ||
| if (Op->isDivergent()) | ||
| return SDValue(); |
Contributor
There was a problem hiding this comment.
This is the opposite of what you want:
Suggested change
| return SDValue(); | |
| return Op; |
To get the default expansion, you return the original node. Return SDValue() means treat as legal
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This is to demonstrate that setOperationAction does not work to solve SWDEV-562122. See #165626