[AArch64] Funnel Shift now uses rev32/rev64 instructions

[jyli0116]

Fixes #130469

Now uses REV32/REV64 instructions to complete operation.

New Output:

G1:
        rev64   v0.4s, v0.4s
        ret
G2:
        rev32   v0.8h, v0.8h
        ret
G3:
        rev16   v0.16b, v0.16b
        ret
G4:
        rev32   v0.4h, v0.4h
        ret
G5:
        rev16   v0.8b, v0.8b
        ret

Old Output:

G1:
        shl     v1.2d, v0.2d, #32
        usra    v1.2d, v0.2d, #32
        mov     v0.16b, v1.16b
        ret

G2:
        shl     v1.4s, v0.4s, #16
        usra    v1.4s, v0.4s, #16
        mov     v0.16b, v1.16b
        ret

G3:
        rev16   v0.16b, v0.16b
        ret

G4:
        shl     v1.2s, v0.2s, #16
        usra    v1.2s, v0.2s, #16
        fmov    d0, d1
        ret

G5:
        rev16   v0.8b, v0.8b
        ret```

[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 (jyli0116)

Changes

Fixes #130469

Now uses REV32/REV64 instructions to complete operation.

New Output:

G1:
        rev64   v0.4s, v0.4s
        ret
G2:
        rev32   v0.8h, v0.8h
        ret
G3:
        rev16   v0.16b, v0.16b
        ret
G4:
        rev32   v0.4h, v0.4h
        ret
G5:
        rev16   v0.8b, v0.8b
        ret

Old Output:

G1:
        shl     v1.2d, v0.2d, #<!-- -->32
        usra    v1.2d, v0.2d, #<!-- -->32
        mov     v0.16b, v1.16b
        ret

G2:
        shl     v1.4s, v0.4s, #<!-- -->16
        usra    v1.4s, v0.4s, #<!-- -->16
        mov     v0.16b, v1.16b
        ret

G3:
        rev16   v0.16b, v0.16b
        ret

G4:
        shl     v1.2s, v0.2s, #<!-- -->16
        usra    v1.2s, v0.2s, #<!-- -->16
        fmov    d0, d1
        ret

G5:
        rev16   v0.8b, v0.8b
        ret```

---
Full diff: https://github.com/llvm/llvm-project/pull/136707.diff


2 Files Affected:

- (modified) llvm/lib/Target/AArch64/AArch64InstrInfo.td (+8) 
- (added) llvm/test/CodeGen/AArch64/aarch64-fshl-rev.ll (+29) 


``````````diff
diff --git a/llvm/lib/Target/AArch64/AArch64InstrInfo.td b/llvm/lib/Target/AArch64/AArch64InstrInfo.td
index a060a2f597ccd..e7aa9c53a8343 100644
--- a/llvm/lib/Target/AArch64/AArch64InstrInfo.td
+++ b/llvm/lib/Target/AArch64/AArch64InstrInfo.td
@@ -5679,6 +5679,14 @@ def : Pat<(v4i32 (bswap (v4i32 V128:$Rn))),
 def : Pat<(v2i64 (bswap (v2i64 V128:$Rn))),
           (v2i64 (REV64v16i8 (v2i64 V128:$Rn)))>;
 
+// Patterns for funnel shifts to be matched to equivalent REV instructions
+def : Pat<(v2i64 (or (v2i64 (AArch64vshl (v2i64 V128:$Rn), (i32 32))), (v2i64 (AArch64vlshr (v2i64 V128:$Rn), (i32 32))))), 
+          (v2i64 (REV64v4i32 (v2i64 V128:$Rn)))>;
+def : Pat<(v4i32 (or (v4i32 (AArch64vshl (v4i32 V128:$Rn), (i32 16))), (v4i32 (AArch64vlshr (v4i32 V128:$Rn), (i32 16))))), 
+          (v4i32 (REV32v8i16 (v4i32 V128:$Rn)))>;
+def : Pat<(v2i32 (or (v2i32 (AArch64vshl (v2i32 V64:$Rn), (i32 16))), (v2i32 (AArch64vlshr (v2i32 V64:$Rn), (i32 16))))), 
+          (v2i32 (REV32v4i16 (v2i32 V64:$Rn)))>;
+
 //===----------------------------------------------------------------------===//
 // Advanced SIMD three vector instructions.
 //===----------------------------------------------------------------------===//
diff --git a/llvm/test/CodeGen/AArch64/aarch64-fshl-rev.ll b/llvm/test/CodeGen/AArch64/aarch64-fshl-rev.ll
new file mode 100644
index 0000000000000..503bc307e7777
--- /dev/null
+++ b/llvm/test/CodeGen/AArch64/aarch64-fshl-rev.ll
@@ -0,0 +1,29 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
+; RUN: llc -mtriple=aarch64 %s -o - | FileCheck %s
+
+define <2 x i64> @fhsl_to_rev2i64(<2 x i64> %r) {
+; CHECK-LABEL: fhsl_to_rev2i64:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    rev64 v0.4s, v0.4s
+; CHECK-NEXT:    ret
+    %or = tail call <2 x i64> @llvm.fshl.v2i64(<2 x i64> %r, <2 x i64> %r, <2 x i64> splat (i64 32))
+    ret <2 x i64> %or
+}
+
+define <4 x i32> @fshl_to_rev4i32(<4 x i32> %r) {
+; CHECK-LABEL: fshl_to_rev4i32:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    rev32 v0.8h, v0.8h
+; CHECK-NEXT:    ret
+    %or = tail call <4 x i32> @llvm.fshl.v4i32(<4 x i32> %r, <4 x i32> %r, <4 x i32> splat (i32 16))
+    ret <4 x i32> %or
+}
+
+define <2 x i32> @fshl_to_rev2i32(<2 x i32> %r) {
+; CHECK-LABEL: fshl_to_rev2i32:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    rev32 v0.4h, v0.4h
+; CHECK-NEXT:    ret
+    %or = tail call <2 x i32> @llvm.fshl.v2i32(<2 x i32> %r, <2 x i32> %r, <2 x i32> splat (i32 16))
+    ret <2 x i32> %or
+}

[davemgreen]

Can you split this line to be a little shorter? It doesn't need to be too short, but maybe making the two (v2i64 (AArch64vshl/AArch64vlshr .. will help keep it readable.

[davemgreen]

AARch64 can be changed to AArch64 in the title too, but otherwise LGTM.

[c-rhodes]

couple of minor comments but otherwise LGTM, cheers

[davemgreen]

This is a small nit, but can we do it like:

def : Pat<(v2i64 (or (v2i64 (AArch64vshl (v2i64 V128:$Rn), (i32 32))), 
                     (v2i64 (AArch64vlshr (v2i64 V128:$Rn), (i32 32))))),

[davemgreen]

LGTM, thanks

[github-actions]

@jyli0116 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!