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.

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

[PTX] Enable migration of ldmatrix #2692

Open
wants to merge 4 commits into
base: SYCLomatic
Choose a base branch
from

Conversation

TejaX-Alaghari
Copy link
Contributor

This PR adds support for the migration of ldmatrix PTX ASM API

@TejaX-Alaghari TejaX-Alaghari requested a review from a team as a code owner March 6, 2025 17:29
@zhiweij1 zhiweij1 changed the title [PRX] Enable migration of ldmatrix [PTX] Enable migration of ldmatrix Mar 7, 2025
@TejaX-Alaghari TejaX-Alaghari force-pushed the ptx branch 5 times, most recently from bd88679 to 4492ec9 Compare March 13, 2025 05:16
Copy link
Contributor

@tomflinda tomflinda left a comment

Choose a reason for hiding this comment

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

LGTM

@@ -557,12 +557,15 @@ bool SYCLGenBase::emitVectorType(const InlineAsmVectorType *T) {
OS() << ", ";
switch (T->getKind()) {
case InlineAsmVectorType::v2:
case InlineAsmVectorType::x1:
Copy link
Contributor

@ziranzha ziranzha Mar 14, 2025

Choose a reason for hiding this comment

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

What's the relation between ptx x1 x2 x4 and vector size? And should have a lit test for it.

Copy link
Contributor Author

@TejaX-Alaghari TejaX-Alaghari Mar 14, 2025

Choose a reason for hiding this comment

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

In "xn", n refers to the number of output args being passed to the instruction

There seems to be no lit test added for "vn", so I'm not sure about lit test for this scenario

What do you advice?

Copy link
Contributor

Choose a reason for hiding this comment

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

By the way, I don't see the usage of the vector size, is that right?

OS() << ", ";
if (const auto *VE =
dyn_cast<InlineAsmVectorExpr>(Inst->getOutputOperand())) {
for (unsigned Inst = 0; Inst != VE->getNumElements(); ++Inst) {
Copy link
Contributor

Choose a reason for hiding this comment

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

Should we check the vector size satisfy the instruction requirement?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants