[AArch64] Expand v8i8 cttz (PR39729)

[nikic]

Summary

Fix for https://bugs.llvm.org/show_bug.cgi?id=39729.

Rather than adding just a case for v8i8 I'm setting cttz to expand for all vector types.

[efriedma]

LGTM.

(I think we can improve the generated sequences for some of these, but you obviously don't need to do that here.)

llvm/trunk/lib/Target/AArch64/AArch64ISelLowering.cpp

if (Subtarget->hasNEON()) {
     // Or, direct i32 -> f16 vector conversion.  Set it so custom, so the
     // conversion happens in two steps: v4i32 -> v4f32 -> v4f16
     setOperationAction(ISD::SINT_TO_FP, MVT::v4i32, Custom);
     setOperationAction(ISD::UINT_TO_FP, MVT::v4i32, Custom);
 
     setOperationAction(ISD::CTLZ,       MVT::v1i64, Expand);
     setOperationAction(ISD::CTLZ,       MVT::v2i64, Expand);
 
-    setOperationAction(ISD::CTTZ,       MVT::v2i8,  Expand);
-    setOperationAction(ISD::CTTZ,       MVT::v4i16, Expand);
-    setOperationAction(ISD::CTTZ,       MVT::v2i32, Expand);
-    setOperationAction(ISD::CTTZ,       MVT::v1i64, Expand);
-    setOperationAction(ISD::CTTZ,       MVT::v16i8, Expand);
-    setOperationAction(ISD::CTTZ,       MVT::v8i16, Expand);
-    setOperationAction(ISD::CTTZ,       MVT::v4i32, Expand);
-    setOperationAction(ISD::CTTZ,       MVT::v2i64, Expand);
-
     // AArch64 doesn't have MUL.2d:
     setOperationAction(ISD::MUL, MVT::v2i64, Expand);
     // Custom handling for some quad-vector types to detect MULL.
     setOperationAction(ISD::MUL, MVT::v8i16, Custom);
     setOperationAction(ISD::MUL, MVT::v4i32, Custom);
     setOperationAction(ISD::MUL, MVT::v2i64, Custom);
 
     // Vector reductions
for (MVT VT : MVT::vector_valuetypes()) {
       } else {
         setOperationAction(ISD::MULHS, VT, Expand);
         setOperationAction(ISD::MULHU, VT, Expand);
       }
       setOperationAction(ISD::SMUL_LOHI, VT, Expand);
       setOperationAction(ISD::UMUL_LOHI, VT, Expand);
 
       setOperationAction(ISD::BSWAP, VT, Expand);
+      setOperationAction(ISD::CTTZ, VT, Expand);
 
       for (MVT InnerVT : MVT::vector_valuetypes()) {
         setTruncStoreAction(VT, InnerVT, Expand);
         setLoadExtAction(ISD::SEXTLOAD, VT, InnerVT, Expand);
         setLoadExtAction(ISD::ZEXTLOAD, VT, InnerVT, Expand);
         setLoadExtAction(ISD::EXTLOAD, VT, InnerVT, Expand);
       }
     }

llvm/trunk/test/CodeGen/AArch64/vec_cttz.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=aarch64-none-linux-gnu -mattr=+neon | FileCheck %s --check-prefix=CHECK
+
+declare <8 x i8> @llvm.cttz.v8i8(<8 x i8>, i1)
+declare <4 x i16> @llvm.cttz.v4i16(<4 x i16>, i1)
+declare <2 x i32> @llvm.cttz.v2i32(<2 x i32>, i1)
+declare <1 x i64> @llvm.cttz.v1i64(<1 x i64>, i1)
+
+declare <16 x i8> @llvm.cttz.v16i8(<16 x i8>, i1)
+declare <8 x i16> @llvm.cttz.v8i16(<8 x i16>, i1)
+declare <4 x i32> @llvm.cttz.v4i32(<4 x i32>, i1)
+declare <2 x i64> @llvm.cttz.v2i64(<2 x i64>, i1)
+
+define <8 x i8> @cttz_v8i8(<8 x i8> %a) nounwind {
+; CHECK-LABEL: cttz_v8i8:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    movi v1.8b, #1
+; CHECK-NEXT:    sub v1.8b, v0.8b, v1.8b
+; CHECK-NEXT:    bic v0.8b, v1.8b, v0.8b
+; CHECK-NEXT:    cnt v0.8b, v0.8b
+; CHECK-NEXT:    ret
+    %b = call <8 x i8> @llvm.cttz.v8i8(<8 x i8> %a, i1 true)
+    ret <8 x i8> %b
+}
+
+define <4 x i16> @cttz_v4i16(<4 x i16> %a) nounwind {
+; CHECK-LABEL: cttz_v4i16:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    movi v1.4h, #1
+; CHECK-NEXT:    sub v1.4h, v0.4h, v1.4h
+; CHECK-NEXT:    bic v0.8b, v1.8b, v0.8b
+; CHECK-NEXT:    clz v0.4h, v0.4h
+; CHECK-NEXT:    movi v1.4h, #16
+; CHECK-NEXT:    sub v0.4h, v1.4h, v0.4h
+; CHECK-NEXT:    ret
+    %b = call <4 x i16> @llvm.cttz.v4i16(<4 x i16> %a, i1 true)
+    ret <4 x i16> %b
+}
+
+define <2 x i32> @cttz_v2i32(<2 x i32> %a) nounwind {
+; CHECK-LABEL: cttz_v2i32:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    movi v1.2s, #1
+; CHECK-NEXT:    sub v1.2s, v0.2s, v1.2s
+; CHECK-NEXT:    bic v0.8b, v1.8b, v0.8b
+; CHECK-NEXT:    clz v0.2s, v0.2s
+; CHECK-NEXT:    movi v1.2s, #32
+; CHECK-NEXT:    sub v0.2s, v1.2s, v0.2s
+; CHECK-NEXT:    ret
+    %b = call <2 x i32> @llvm.cttz.v2i32(<2 x i32> %a, i1 true)
+    ret <2 x i32> %b
+}
+
+define <1 x i64> @cttz_v1i64(<1 x i64> %a) nounwind {
+; CHECK-LABEL: cttz_v1i64:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    orr w8, wzr, #0x1
+; CHECK-NEXT:    fmov d1, x8
+; CHECK-NEXT:    sub d1, d0, d1
+; CHECK-NEXT:    bic v0.8b, v1.8b, v0.8b
+; CHECK-NEXT:    cnt v0.8b, v0.8b
+; CHECK-NEXT:    uaddlp v0.4h, v0.8b
+; CHECK-NEXT:    uaddlp v0.2s, v0.4h
+; CHECK-NEXT:    uaddlp v0.1d, v0.2s
+; CHECK-NEXT:    ret
+    %b = call <1 x i64> @llvm.cttz.v1i64(<1 x i64> %a, i1 true)
+    ret <1 x i64> %b
+}
+
+define <16 x i8> @cttz_v16i8(<16 x i8> %a) nounwind {
+; CHECK-LABEL: cttz_v16i8:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    movi v1.16b, #1
+; CHECK-NEXT:    sub v1.16b, v0.16b, v1.16b
+; CHECK-NEXT:    bic v0.16b, v1.16b, v0.16b
+; CHECK-NEXT:    cnt v0.16b, v0.16b
+; CHECK-NEXT:    ret
+    %b = call <16 x i8> @llvm.cttz.v16i8(<16 x i8> %a, i1 true)
+    ret <16 x i8> %b
+}
+
+define <8 x i16> @cttz_v8i16(<8 x i16> %a) nounwind {
+; CHECK-LABEL: cttz_v8i16:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    movi v1.8h, #1
+; CHECK-NEXT:    sub v1.8h, v0.8h, v1.8h
+; CHECK-NEXT:    bic v0.16b, v1.16b, v0.16b
+; CHECK-NEXT:    clz v0.8h, v0.8h
+; CHECK-NEXT:    movi v1.8h, #16
+; CHECK-NEXT:    sub v0.8h, v1.8h, v0.8h
+; CHECK-NEXT:    ret
+    %b = call <8 x i16> @llvm.cttz.v8i16(<8 x i16> %a, i1 true)
+    ret <8 x i16> %b
+}
+
+define <4 x i32> @cttz_v4i32(<4 x i32> %a) nounwind {
+; CHECK-LABEL: cttz_v4i32:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    movi v1.4s, #1
+; CHECK-NEXT:    sub v1.4s, v0.4s, v1.4s
+; CHECK-NEXT:    bic v0.16b, v1.16b, v0.16b
+; CHECK-NEXT:    clz v0.4s, v0.4s
+; CHECK-NEXT:    movi v1.4s, #32
+; CHECK-NEXT:    sub v0.4s, v1.4s, v0.4s
+; CHECK-NEXT:    ret
+    %b = call <4 x i32> @llvm.cttz.v4i32(<4 x i32> %a, i1 true)
+    ret <4 x i32> %b
+}
+
+define <2 x i64> @cttz_v2i64(<2 x i64> %a) nounwind {
+; CHECK-LABEL: cttz_v2i64:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    orr w8, wzr, #0x1
+; CHECK-NEXT:    dup v1.2d, x8
+; CHECK-NEXT:    sub v1.2d, v0.2d, v1.2d
+; CHECK-NEXT:    bic v0.16b, v1.16b, v0.16b
+; CHECK-NEXT:    cnt v0.16b, v0.16b
+; CHECK-NEXT:    uaddlp v0.8h, v0.16b
+; CHECK-NEXT:    uaddlp v0.4s, v0.8h
+; CHECK-NEXT:    uaddlp v0.2d, v0.4s
+; CHECK-NEXT:    ret
+    %b = call <2 x i64> @llvm.cttz.v2i64(<2 x i64> %a, i1 true)
+    ret <2 x i64> %b
+}