[X86][AVX512BW] Fixed 512-bit PSHUFB shuffle mask decode and added combine test.

PSHUFB decoder was assuming that input was 128 or 256-bit vector only.

llvm-svn: 262661

diff --git a/llvm/lib/Target/X86/Utils/X86ShuffleDecode.cpp b/llvm/lib/Target/X86/Utils/X86ShuffleDecode.cpp
index e2e3f8c..5c99926 100644 (file)
--- a/llvm/lib/Target/X86/Utils/X86ShuffleDecode.cpp
+++ b/llvm/lib/Target/X86/Utils/X86ShuffleDecode.cpp
@@ -314,9 +314,9 @@ void DecodePSHUFBMask(ArrayRef<uint64_t> RawMask,
       ShuffleMask.push_back(M);
       continue;
     }
-    // For AVX vectors with 32 bytes the base of the shuffle is the half of
-    // the vector we're inside.
-    int Base = i < 16 ? 0 : 16;
+    // For 256/512-bit vectors the base of the shuffle is the 128-bit
+    // subvector we're inside.
+    int Base = (i / 16) * 16;
     // If the high bit (7) of the byte is set, the element is zeroed.
     if (M & (1 << 7))
       ShuffleMask.push_back(SM_SentinelZero);

diff --git a/llvm/test/CodeGen/X86/vector-shuffle-combining-avx512bw.ll b/llvm/test/CodeGen/X86/vector-shuffle-combining-avx512bw.ll
new file mode 100644 (file)
index 0000000..f0b4080
--- /dev/null
+++ b/llvm/test/CodeGen/X86/vector-shuffle-combining-avx512bw.ll
@@ -0,0 +1,15 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+avx512bw | FileCheck %s
+
+declare <64 x i8> @llvm.x86.avx512.mask.pshuf.b.512(<64 x i8>, <64 x i8>, <64 x i8>, i64)
+
+define <64 x i8> @combine_pshufb_identity(<64 x i8> %x0) {
+; CHECK-LABEL: combine_pshufb_identity:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    retq
+  %select = bitcast <8 x i64> <i64 -1, i64 -1, i64 -1, i64 -1, i64 -1, i64 -1, i64 -1, i64 -1> to <64 x i8>
+  %mask = bitcast <16 x i32> <i32 202182159, i32 134810123, i32 67438087, i32 66051, i32 202182159, i32 134810123, i32 67438087, i32 66051, i32 202182159, i32 134810123, i32 67438087, i32 66051, i32 202182159, i32 134810123, i32 67438087, i32 66051> to <64 x i8>
+  %res0 = call <64 x i8> @llvm.x86.avx512.mask.pshuf.b.512(<64 x i8> %x0, <64 x i8> %mask, <64 x i8> %select, i64 -1)
+  %res1 = call <64 x i8> @llvm.x86.avx512.mask.pshuf.b.512(<64 x i8> %res0, <64 x i8> %mask, <64 x i8> %select, i64 -1)
+  ret <64 x i8> %res1
+}