[AArch64PostCoalescer] Propagate undef flag after replacing

[XiaShark]

I encountered a compilation crash issue, and after analysis, it was caused by the AArch64PostCoalescerPass, see https://godbolt.org/z/vPeqeo5Pa.
When replacing the register, if the source register has undef flag, we should propagate the flag to all uses of the destination register.

[github-actions]

Thank you for submitting a Pull Request (PR) to the LLVM Project!

This PR will be automatically labeled and the relevant teams will be notified.

If you wish to, you can add reviewers by using the "Reviewers" section on this page.

If this is not working for you, it is probably because you do not have write permissions for the repository. In which case you can instead tag reviewers by name in a comment by using @ followed by their GitHub username.

If you have received no comments on your PR for a week, you can request a review by "ping"ing the PR by adding a comment “Ping”. The common courtesy "ping" rate is once a week. Please remember that you are asking for valuable time from other developers.

If you have further questions, they may be answered by the LLVM GitHub User Guide.

You can also ask questions in a comment on this PR, on the LLVM Discord or on the forums.

[llvmbot]

@llvm/pr-subscribers-backend-aarch64

Author: None (XiaShark)

Changes

I encountered a compilation crash issue, and after analysis, it was caused by the AArch64PostCoalescerPass, see https://godbolt.org/z/vPeqeo5Pa.
If the destination register still has active uses, removing COALESCER_BARRIER is unsafe.

---

Full diff: https://github.com/llvm/llvm-project/pull/163119.diff

2 Files Affected:

• (modified) llvm/lib/Target/AArch64/AArch64PostCoalescerPass.cpp (+2)
• (added) llvm/test/CodeGen/AArch64/aarch64-post-coalescer.ll (+14)

diff --git a/llvm/lib/Target/AArch64/AArch64PostCoalescerPass.cpp b/llvm/lib/Target/AArch64/AArch64PostCoalescerPass.cpp
index cdf2822f3ed9d..a8cb1eb07003f 100644
--- a/llvm/lib/Target/AArch64/AArch64PostCoalescerPass.cpp
+++ b/llvm/lib/Target/AArch64/AArch64PostCoalescerPass.cpp
@@ -72,6 +72,8 @@ bool AArch64PostCoalescer::runOnMachineFunction(MachineFunction &MF) {
       case AArch64::COALESCER_BARRIER_FPR128: {
         Register Src = MI.getOperand(1).getReg();
         Register Dst = MI.getOperand(0).getReg();
+        if (MRI->hasOneDef(Dst) && !MRI->use_empty(Dst))
+          continue;
         if (Src != Dst)
           MRI->replaceRegWith(Dst, Src);
 
diff --git a/llvm/test/CodeGen/AArch64/aarch64-post-coalescer.ll b/llvm/test/CodeGen/AArch64/aarch64-post-coalescer.ll
new file mode 100644
index 0000000000000..adf6c72f9fb2f
--- /dev/null
+++ b/llvm/test/CodeGen/AArch64/aarch64-post-coalescer.ll
@@ -0,0 +1,14 @@
+; RUN: llc -mtriple=aarch64 -mattr=+sme -stop-ater=aarch64-post-coalescer-pass < %s | FileCheck %s
+
+; CHECK-LABEL: name: bar
+; CHECK: [[dst:%[0-9]+]]:fpr64 = COALESCER_BARRIER_FPR64 undef [[src:%[0-9]+]]
+
+target triple = "aarch64"
+
+declare void @foo(double) "aarch64_pstate_sm_enabled"
+
+define dso_local void @bar(double noundef %a) local_unnamed_addr {
+entry:
+  tail call void @foo(double poison) "aarch64_pstate_sm_enabled"
+  ret void
+}
\ No newline at end of file

[XiaShark]

@sdesmalen-arm

[github-actions]

✅ With the latest revision this PR passed the undef deprecator.

[github-actions]

✅ With the latest revision this PR passed the C/C++ code formatter.

[sdesmalen-arm]

nit: would it be sufficient to test only for:

%1:fpr64 = COALESCER_BARRIER_FPR64 undef %1
$d0 = COPY %1
FAKE_USE implicit $d0

without the other machine instructions?

[XiaShark]

It's fine, modified

[sdesmalen-arm]

LGTM, thank you!

[MacDue]

LGTM! Would you like us to merge this now?

[XiaShark]

LGTM! Would you like us to merge this now?

Sure, thanks!

[github-actions]

@XiaShark Congratulations on having your first Pull Request (PR) merged into the LLVM Project!

Your changes will be combined with recent changes from other authors, then tested by our build bots. If there is a problem with a build, you may receive a report in an email or a comment on this PR.

Please check whether problems have been caused by your change specifically, as the builds can include changes from many authors. It is not uncommon for your change to be included in a build that fails due to someone else's changes, or infrastructure issues.

How to do this, and the rest of the post-merge process, is covered in detail here.

If your change does cause a problem, it may be reverted, or you can revert it yourself. This is a normal part of LLVM development. You can fix your changes and open a new PR to merge them again.

If you don't get any reports, no action is required from you. Your changes are working as expected, well done!