[mips][msa] Added support for matching bclr, and bclri from normal IR (i.e. not intrinsics)

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194471 91177308-0d34-0410-b5e6-96231b3b80d8

1 files changed, 137 insertions, 0 deletions

diff --git a/test/CodeGen/Mips/msa/bitwise.ll b/test/CodeGen/Mips/msa/bitwise.ll
index 2104921..5831a08 100644
--- a/test/CodeGen/Mips/msa/bitwise.ll
+++ b/test/CodeGen/Mips/msa/bitwise.ll
@@ -1243,6 +1243,78 @@ define void @binsr_v2i64_i(<2 x i64>* %c, <2 x i64>* %a, <2 x i64>* %b) nounwind
   ; CHECK: .size binsr_v2i64_i
 }
 
+define void @bclr_v16i8(<16 x i8>* %c, <16 x i8>* %a, <16 x i8>* %b) nounwind {
+  ; CHECK: bclr_v16i8:
+
+  %1 = load <16 x i8>* %a
+  ; CHECK-DAG: ld.b [[R1:\$w[0-9]+]], 0($5)
+  %2 = load <16 x i8>* %b
+  ; CHECK-DAG: ld.b [[R2:\$w[0-9]+]], 0($6)
+  %3 = shl <16 x i8> <i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1>, %2
+  %4 = xor <16 x i8> %3, <i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1>
+  %5 = and <16 x i8> %1, %4
+  ; CHECK-DAG: bclr.b [[R3:\$w[0-9]+]], [[R1]], [[R2]]
+  store <16 x i8> %5, <16 x i8>* %c
+  ; CHECK-DAG: st.b [[R3]], 0($4)
+
+  ret void
+  ; CHECK: .size bclr_v16i8
+}
+
+define void @bclr_v8i16(<8 x i16>* %c, <8 x i16>* %a, <8 x i16>* %b) nounwind {
+  ; CHECK: bclr_v8i16:
+
+  %1 = load <8 x i16>* %a
+  ; CHECK-DAG: ld.h [[R1:\$w[0-9]+]], 0($5)
+  %2 = load <8 x i16>* %b
+  ; CHECK-DAG: ld.h [[R2:\$w[0-9]+]], 0($6)
+  %3 = shl <8 x i16> <i16 1, i16 1, i16 1, i16 1, i16 1, i16 1, i16 1, i16 1>, %2
+  %4 = xor <8 x i16> %3, <i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1>
+  %5 = and <8 x i16> %1, %4
+  ; CHECK-DAG: bclr.h [[R3:\$w[0-9]+]], [[R1]], [[R2]]
+  store <8 x i16> %5, <8 x i16>* %c
+  ; CHECK-DAG: st.h [[R3]], 0($4)
+
+  ret void
+  ; CHECK: .size bclr_v8i16
+}
+
+define void @bclr_v4i32(<4 x i32>* %c, <4 x i32>* %a, <4 x i32>* %b) nounwind {
+  ; CHECK: bclr_v4i32:
+
+  %1 = load <4 x i32>* %a
+  ; CHECK-DAG: ld.w [[R1:\$w[0-9]+]], 0($5)
+  %2 = load <4 x i32>* %b
+  ; CHECK-DAG: ld.w [[R2:\$w[0-9]+]], 0($6)
+  %3 = shl <4 x i32> <i32 1, i32 1, i32 1, i32 1>, %2
+  %4 = xor <4 x i32> %3, <i32 -1, i32 -1, i32 -1, i32 -1>
+  %5 = and <4 x i32> %1, %4
+  ; CHECK-DAG: bclr.w [[R3:\$w[0-9]+]], [[R1]], [[R2]]
+  store <4 x i32> %5, <4 x i32>* %c
+  ; CHECK-DAG: st.w [[R3]], 0($4)
+
+  ret void
+  ; CHECK: .size bclr_v4i32
+}
+
+define void @bclr_v2i64(<2 x i64>* %c, <2 x i64>* %a, <2 x i64>* %b) nounwind {
+  ; CHECK: bclr_v2i64:
+
+  %1 = load <2 x i64>* %a
+  ; CHECK-DAG: ld.d [[R1:\$w[0-9]+]], 0($5)
+  %2 = load <2 x i64>* %b
+  ; CHECK-DAG: ld.d [[R2:\$w[0-9]+]], 0($6)
+  %3 = shl <2 x i64> <i64 1, i64 1>, %2
+  %4 = xor <2 x i64> %3, <i64 -1, i64 -1>
+  %5 = and <2 x i64> %1, %4
+  ; CHECK-DAG: bclr.d [[R3:\$w[0-9]+]], [[R1]], [[R2]]
+  store <2 x i64> %5, <2 x i64>* %c
+  ; CHECK-DAG: st.d [[R3]], 0($4)
+
+  ret void
+  ; CHECK: .size bclr_v2i64
+}
+
 define void @bset_v16i8(<16 x i8>* %c, <16 x i8>* %a, <16 x i8>* %b) nounwind {
   ; CHECK: bset_v16i8:
 
@@ -1379,6 +1451,71 @@ define void @bneg_v2i64(<2 x i64>* %c, <2 x i64>* %a, <2 x i64>* %b) nounwind {
   ; CHECK: .size bneg_v2i64
 }
 
+define void @bclri_v16i8(<16 x i8>* %c, <16 x i8>* %a) nounwind {
+  ; CHECK: bclri_v16i8:
+
+  %1 = load <16 x i8>* %a
+  ; CHECK-DAG: ld.b [[R1:\$w[0-9]+]], 0($5)
+  %2 = xor <16 x i8> <i8  8, i8  8, i8  8, i8  8, i8  8, i8  8, i8  8, i8  8, i8  8, i8  8, i8  8, i8  8, i8  8, i8  8, i8  8, i8  8>,
+                     <i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1>
+  %3 = and <16 x i8> %1, %2
+  ; bclri.b and andi.b are exactly equivalent.
+  ; CHECK-DAG: andi.b [[R3:\$w[0-9]+]], [[R1]], 247
+  store <16 x i8> %3, <16 x i8>* %c
+  ; CHECK-DAG: st.b [[R3]], 0($4)
+
+  ret void
+  ; CHECK: .size bclri_v16i8
+}
+
+define void @bclri_v8i16(<8 x i16>* %c, <8 x i16>* %a) nounwind {
+  ; CHECK: bclri_v8i16:
+
+  %1 = load <8 x i16>* %a
+  ; CHECK-DAG: ld.h [[R1:\$w[0-9]+]], 0($5)
+  %2 = xor <8 x i16> <i16  8, i16  8, i16  8, i16  8, i16  8, i16  8, i16  8, i16  8>,
+                     <i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1>
+  %3 = and <8 x i16> %1, %2
+  ; CHECK-DAG: bclri.h [[R3:\$w[0-9]+]], [[R1]], 3
+  store <8 x i16> %3, <8 x i16>* %c
+  ; CHECK-DAG: st.h [[R3]], 0($4)
+
+  ret void
+  ; CHECK: .size bclri_v8i16
+}
+
+define void @bclri_v4i32(<4 x i32>* %c, <4 x i32>* %a) nounwind {
+  ; CHECK: bclri_v4i32:
+
+  %1 = load <4 x i32>* %a
+  ; CHECK-DAG: ld.w [[R1:\$w[0-9]+]], 0($5)
+  %2 = xor <4 x i32> <i32  8, i32  8, i32  8, i32  8>,
+                     <i32 -1, i32 -1, i32 -1, i32 -1>
+  %3 = and <4 x i32> %1, %2
+  ; CHECK-DAG: bclri.w [[R3:\$w[0-9]+]], [[R1]], 3
+  store <4 x i32> %3, <4 x i32>* %c
+  ; CHECK-DAG: st.w [[R3]], 0($4)
+
+  ret void
+  ; CHECK: .size bclri_v4i32
+}
+
+define void @bclri_v2i64(<2 x i64>* %c, <2 x i64>* %a) nounwind {
+  ; CHECK: bclri_v2i64:
+
+  %1 = load <2 x i64>* %a
+  ; CHECK-DAG: ld.d [[R1:\$w[0-9]+]], 0($5)
+  %2 = xor <2 x i64> <i64  8, i64  8>,
+                     <i64 -1, i64 -1>
+  %3 = and <2 x i64> %1, %2
+  ; CHECK-DAG: bclri.d [[R3:\$w[0-9]+]], [[R1]], 3
+  store <2 x i64> %3, <2 x i64>* %c
+  ; CHECK-DAG: st.d [[R3]], 0($4)
+
+  ret void
+  ; CHECK: .size bclri_v2i64
+}
+
 define void @bseti_v16i8(<16 x i8>* %c, <16 x i8>* %a) nounwind {
   ; CHECK: bseti_v16i8: