1; NOTE: Assertions have been autogenerated by utils/update_test_checks.py 2; RUN: opt -passes='lower-matrix-intrinsics' -S < %s | FileCheck %s 3 4 5define <4 x float> @multiply_2x2(<4 x float> %a, <4 x float> %b) { 6; CHECK-LABEL: @multiply_2x2( 7; CHECK-NEXT: entry: 8; CHECK-NEXT: [[SPLIT:%.*]] = shufflevector <4 x float> [[A:%.*]], <4 x float> poison, <2 x i32> <i32 0, i32 1> 9; CHECK-NEXT: [[SPLIT1:%.*]] = shufflevector <4 x float> [[A]], <4 x float> poison, <2 x i32> <i32 2, i32 3> 10; CHECK-NEXT: [[SPLIT2:%.*]] = shufflevector <4 x float> [[B:%.*]], <4 x float> poison, <2 x i32> <i32 0, i32 1> 11; CHECK-NEXT: [[SPLIT3:%.*]] = shufflevector <4 x float> [[B]], <4 x float> poison, <2 x i32> <i32 2, i32 3> 12; CHECK-NEXT: [[BLOCK:%.*]] = shufflevector <2 x float> [[SPLIT]], <2 x float> poison, <1 x i32> zeroinitializer 13; CHECK-NEXT: [[TMP0:%.*]] = extractelement <2 x float> [[SPLIT2]], i64 0 14; CHECK-NEXT: [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x float> poison, float [[TMP0]], i64 0 15; CHECK-NEXT: [[SPLAT_SPLAT:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT]], <1 x float> poison, <1 x i32> zeroinitializer 16; CHECK-NEXT: [[TMP1:%.*]] = fmul <1 x float> [[BLOCK]], [[SPLAT_SPLAT]] 17; CHECK-NEXT: [[BLOCK4:%.*]] = shufflevector <2 x float> [[SPLIT1]], <2 x float> poison, <1 x i32> zeroinitializer 18; CHECK-NEXT: [[TMP2:%.*]] = extractelement <2 x float> [[SPLIT2]], i64 1 19; CHECK-NEXT: [[SPLAT_SPLATINSERT5:%.*]] = insertelement <1 x float> poison, float [[TMP2]], i64 0 20; CHECK-NEXT: [[SPLAT_SPLAT6:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT5]], <1 x float> poison, <1 x i32> zeroinitializer 21; CHECK-NEXT: [[TMP3:%.*]] = fmul <1 x float> [[BLOCK4]], [[SPLAT_SPLAT6]] 22; CHECK-NEXT: [[TMP4:%.*]] = fadd <1 x float> [[TMP1]], [[TMP3]] 23; CHECK-NEXT: [[TMP5:%.*]] = shufflevector <1 x float> [[TMP4]], <1 x float> poison, <2 x i32> <i32 0, i32 poison> 24; CHECK-NEXT: [[TMP6:%.*]] = shufflevector <2 x float> poison, <2 x float> [[TMP5]], <2 x i32> <i32 2, i32 1> 25; CHECK-NEXT: [[BLOCK7:%.*]] = shufflevector <2 x float> [[SPLIT]], <2 x float> poison, <1 x i32> <i32 1> 26; CHECK-NEXT: [[TMP7:%.*]] = extractelement <2 x float> [[SPLIT2]], i64 0 27; CHECK-NEXT: [[SPLAT_SPLATINSERT8:%.*]] = insertelement <1 x float> poison, float [[TMP7]], i64 0 28; CHECK-NEXT: [[SPLAT_SPLAT9:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT8]], <1 x float> poison, <1 x i32> zeroinitializer 29; CHECK-NEXT: [[TMP8:%.*]] = fmul <1 x float> [[BLOCK7]], [[SPLAT_SPLAT9]] 30; CHECK-NEXT: [[BLOCK10:%.*]] = shufflevector <2 x float> [[SPLIT1]], <2 x float> poison, <1 x i32> <i32 1> 31; CHECK-NEXT: [[TMP9:%.*]] = extractelement <2 x float> [[SPLIT2]], i64 1 32; CHECK-NEXT: [[SPLAT_SPLATINSERT11:%.*]] = insertelement <1 x float> poison, float [[TMP9]], i64 0 33; CHECK-NEXT: [[SPLAT_SPLAT12:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT11]], <1 x float> poison, <1 x i32> zeroinitializer 34; CHECK-NEXT: [[TMP10:%.*]] = fmul <1 x float> [[BLOCK10]], [[SPLAT_SPLAT12]] 35; CHECK-NEXT: [[TMP11:%.*]] = fadd <1 x float> [[TMP8]], [[TMP10]] 36; CHECK-NEXT: [[TMP12:%.*]] = shufflevector <1 x float> [[TMP11]], <1 x float> poison, <2 x i32> <i32 0, i32 poison> 37; CHECK-NEXT: [[TMP13:%.*]] = shufflevector <2 x float> [[TMP6]], <2 x float> [[TMP12]], <2 x i32> <i32 0, i32 2> 38; CHECK-NEXT: [[BLOCK13:%.*]] = shufflevector <2 x float> [[SPLIT]], <2 x float> poison, <1 x i32> zeroinitializer 39; CHECK-NEXT: [[TMP14:%.*]] = extractelement <2 x float> [[SPLIT3]], i64 0 40; CHECK-NEXT: [[SPLAT_SPLATINSERT14:%.*]] = insertelement <1 x float> poison, float [[TMP14]], i64 0 41; CHECK-NEXT: [[SPLAT_SPLAT15:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT14]], <1 x float> poison, <1 x i32> zeroinitializer 42; CHECK-NEXT: [[TMP15:%.*]] = fmul <1 x float> [[BLOCK13]], [[SPLAT_SPLAT15]] 43; CHECK-NEXT: [[BLOCK16:%.*]] = shufflevector <2 x float> [[SPLIT1]], <2 x float> poison, <1 x i32> zeroinitializer 44; CHECK-NEXT: [[TMP16:%.*]] = extractelement <2 x float> [[SPLIT3]], i64 1 45; CHECK-NEXT: [[SPLAT_SPLATINSERT17:%.*]] = insertelement <1 x float> poison, float [[TMP16]], i64 0 46; CHECK-NEXT: [[SPLAT_SPLAT18:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT17]], <1 x float> poison, <1 x i32> zeroinitializer 47; CHECK-NEXT: [[TMP17:%.*]] = fmul <1 x float> [[BLOCK16]], [[SPLAT_SPLAT18]] 48; CHECK-NEXT: [[TMP18:%.*]] = fadd <1 x float> [[TMP15]], [[TMP17]] 49; CHECK-NEXT: [[TMP19:%.*]] = shufflevector <1 x float> [[TMP18]], <1 x float> poison, <2 x i32> <i32 0, i32 poison> 50; CHECK-NEXT: [[TMP20:%.*]] = shufflevector <2 x float> poison, <2 x float> [[TMP19]], <2 x i32> <i32 2, i32 1> 51; CHECK-NEXT: [[BLOCK19:%.*]] = shufflevector <2 x float> [[SPLIT]], <2 x float> poison, <1 x i32> <i32 1> 52; CHECK-NEXT: [[TMP21:%.*]] = extractelement <2 x float> [[SPLIT3]], i64 0 53; CHECK-NEXT: [[SPLAT_SPLATINSERT20:%.*]] = insertelement <1 x float> poison, float [[TMP21]], i64 0 54; CHECK-NEXT: [[SPLAT_SPLAT21:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT20]], <1 x float> poison, <1 x i32> zeroinitializer 55; CHECK-NEXT: [[TMP22:%.*]] = fmul <1 x float> [[BLOCK19]], [[SPLAT_SPLAT21]] 56; CHECK-NEXT: [[BLOCK22:%.*]] = shufflevector <2 x float> [[SPLIT1]], <2 x float> poison, <1 x i32> <i32 1> 57; CHECK-NEXT: [[TMP23:%.*]] = extractelement <2 x float> [[SPLIT3]], i64 1 58; CHECK-NEXT: [[SPLAT_SPLATINSERT23:%.*]] = insertelement <1 x float> poison, float [[TMP23]], i64 0 59; CHECK-NEXT: [[SPLAT_SPLAT24:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT23]], <1 x float> poison, <1 x i32> zeroinitializer 60; CHECK-NEXT: [[TMP24:%.*]] = fmul <1 x float> [[BLOCK22]], [[SPLAT_SPLAT24]] 61; CHECK-NEXT: [[TMP25:%.*]] = fadd <1 x float> [[TMP22]], [[TMP24]] 62; CHECK-NEXT: [[TMP26:%.*]] = shufflevector <1 x float> [[TMP25]], <1 x float> poison, <2 x i32> <i32 0, i32 poison> 63; CHECK-NEXT: [[TMP27:%.*]] = shufflevector <2 x float> [[TMP20]], <2 x float> [[TMP26]], <2 x i32> <i32 0, i32 2> 64; CHECK-NEXT: [[TMP28:%.*]] = shufflevector <2 x float> [[TMP13]], <2 x float> [[TMP27]], <4 x i32> <i32 0, i32 1, i32 2, i32 3> 65; CHECK-NEXT: ret <4 x float> [[TMP28]] 66; 67entry: 68 %c = call <4 x float> @llvm.matrix.multiply.v4f32.v4f32.v4f32(<4 x float> %a, <4 x float> %b, i32 2, i32 2, i32 2) 69 ret <4 x float> %c 70} 71 72declare <4 x float> @llvm.matrix.multiply.v4f32.v4f32.v4f32(<4 x float>, <4 x float>, i32, i32, i32) 73 74define <4 x float> @multiply_1x2(<2 x float> %a, <2 x float> %b) { 75; CHECK-LABEL: @multiply_1x2( 76; CHECK-NEXT: entry: 77; CHECK-NEXT: [[SPLIT:%.*]] = shufflevector <2 x float> [[A:%.*]], <2 x float> poison, <2 x i32> <i32 0, i32 1> 78; CHECK-NEXT: [[SPLIT1:%.*]] = shufflevector <2 x float> [[B:%.*]], <2 x float> poison, <1 x i32> zeroinitializer 79; CHECK-NEXT: [[SPLIT2:%.*]] = shufflevector <2 x float> [[B]], <2 x float> poison, <1 x i32> <i32 1> 80; CHECK-NEXT: [[BLOCK:%.*]] = shufflevector <2 x float> [[SPLIT]], <2 x float> poison, <1 x i32> zeroinitializer 81; CHECK-NEXT: [[TMP0:%.*]] = extractelement <1 x float> [[SPLIT1]], i64 0 82; CHECK-NEXT: [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x float> poison, float [[TMP0]], i64 0 83; CHECK-NEXT: [[SPLAT_SPLAT:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT]], <1 x float> poison, <1 x i32> zeroinitializer 84; CHECK-NEXT: [[TMP1:%.*]] = fmul <1 x float> [[BLOCK]], [[SPLAT_SPLAT]] 85; CHECK-NEXT: [[TMP2:%.*]] = shufflevector <1 x float> [[TMP1]], <1 x float> poison, <2 x i32> <i32 0, i32 poison> 86; CHECK-NEXT: [[TMP3:%.*]] = shufflevector <2 x float> poison, <2 x float> [[TMP2]], <2 x i32> <i32 2, i32 1> 87; CHECK-NEXT: [[BLOCK3:%.*]] = shufflevector <2 x float> [[SPLIT]], <2 x float> poison, <1 x i32> <i32 1> 88; CHECK-NEXT: [[TMP4:%.*]] = extractelement <1 x float> [[SPLIT1]], i64 0 89; CHECK-NEXT: [[SPLAT_SPLATINSERT4:%.*]] = insertelement <1 x float> poison, float [[TMP4]], i64 0 90; CHECK-NEXT: [[SPLAT_SPLAT5:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT4]], <1 x float> poison, <1 x i32> zeroinitializer 91; CHECK-NEXT: [[TMP5:%.*]] = fmul <1 x float> [[BLOCK3]], [[SPLAT_SPLAT5]] 92; CHECK-NEXT: [[TMP6:%.*]] = shufflevector <1 x float> [[TMP5]], <1 x float> poison, <2 x i32> <i32 0, i32 poison> 93; CHECK-NEXT: [[TMP7:%.*]] = shufflevector <2 x float> [[TMP3]], <2 x float> [[TMP6]], <2 x i32> <i32 0, i32 2> 94; CHECK-NEXT: [[BLOCK6:%.*]] = shufflevector <2 x float> [[SPLIT]], <2 x float> poison, <1 x i32> zeroinitializer 95; CHECK-NEXT: [[TMP8:%.*]] = extractelement <1 x float> [[SPLIT2]], i64 0 96; CHECK-NEXT: [[SPLAT_SPLATINSERT7:%.*]] = insertelement <1 x float> poison, float [[TMP8]], i64 0 97; CHECK-NEXT: [[SPLAT_SPLAT8:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT7]], <1 x float> poison, <1 x i32> zeroinitializer 98; CHECK-NEXT: [[TMP9:%.*]] = fmul <1 x float> [[BLOCK6]], [[SPLAT_SPLAT8]] 99; CHECK-NEXT: [[TMP10:%.*]] = shufflevector <1 x float> [[TMP9]], <1 x float> poison, <2 x i32> <i32 0, i32 poison> 100; CHECK-NEXT: [[TMP11:%.*]] = shufflevector <2 x float> poison, <2 x float> [[TMP10]], <2 x i32> <i32 2, i32 1> 101; CHECK-NEXT: [[BLOCK9:%.*]] = shufflevector <2 x float> [[SPLIT]], <2 x float> poison, <1 x i32> <i32 1> 102; CHECK-NEXT: [[TMP12:%.*]] = extractelement <1 x float> [[SPLIT2]], i64 0 103; CHECK-NEXT: [[SPLAT_SPLATINSERT10:%.*]] = insertelement <1 x float> poison, float [[TMP12]], i64 0 104; CHECK-NEXT: [[SPLAT_SPLAT11:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT10]], <1 x float> poison, <1 x i32> zeroinitializer 105; CHECK-NEXT: [[TMP13:%.*]] = fmul <1 x float> [[BLOCK9]], [[SPLAT_SPLAT11]] 106; CHECK-NEXT: [[TMP14:%.*]] = shufflevector <1 x float> [[TMP13]], <1 x float> poison, <2 x i32> <i32 0, i32 poison> 107; CHECK-NEXT: [[TMP15:%.*]] = shufflevector <2 x float> [[TMP11]], <2 x float> [[TMP14]], <2 x i32> <i32 0, i32 2> 108; CHECK-NEXT: [[TMP16:%.*]] = shufflevector <2 x float> [[TMP7]], <2 x float> [[TMP15]], <4 x i32> <i32 0, i32 1, i32 2, i32 3> 109; CHECK-NEXT: ret <4 x float> [[TMP16]] 110; 111entry: 112 %c = call <4 x float> @llvm.matrix.multiply.v4f32.v2f32.v2f32(<2 x float> %a, <2 x float> %b, i32 2, i32 1, i32 2) 113 ret <4 x float> %c 114} 115 116declare <4 x float> @llvm.matrix.multiply.v4f32.v2f32.v2f32(<2 x float>, <2 x float>, i32, i32, i32) 117 118define <9 x float> @multiply_2x3(<6 x float> %a, <6 x float> %b) { 119; CHECK-LABEL: @multiply_2x3( 120; CHECK-NEXT: entry: 121; CHECK-NEXT: [[SPLIT:%.*]] = shufflevector <6 x float> [[A:%.*]], <6 x float> poison, <3 x i32> <i32 0, i32 1, i32 2> 122; CHECK-NEXT: [[SPLIT1:%.*]] = shufflevector <6 x float> [[A]], <6 x float> poison, <3 x i32> <i32 3, i32 4, i32 5> 123; CHECK-NEXT: [[SPLIT2:%.*]] = shufflevector <6 x float> [[B:%.*]], <6 x float> poison, <2 x i32> <i32 0, i32 1> 124; CHECK-NEXT: [[SPLIT3:%.*]] = shufflevector <6 x float> [[B]], <6 x float> poison, <2 x i32> <i32 2, i32 3> 125; CHECK-NEXT: [[SPLIT4:%.*]] = shufflevector <6 x float> [[B]], <6 x float> poison, <2 x i32> <i32 4, i32 5> 126; CHECK-NEXT: [[BLOCK:%.*]] = shufflevector <3 x float> [[SPLIT]], <3 x float> poison, <1 x i32> zeroinitializer 127; CHECK-NEXT: [[TMP0:%.*]] = extractelement <2 x float> [[SPLIT2]], i64 0 128; CHECK-NEXT: [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x float> poison, float [[TMP0]], i64 0 129; CHECK-NEXT: [[SPLAT_SPLAT:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT]], <1 x float> poison, <1 x i32> zeroinitializer 130; CHECK-NEXT: [[TMP1:%.*]] = fmul <1 x float> [[BLOCK]], [[SPLAT_SPLAT]] 131; CHECK-NEXT: [[BLOCK5:%.*]] = shufflevector <3 x float> [[SPLIT1]], <3 x float> poison, <1 x i32> zeroinitializer 132; CHECK-NEXT: [[TMP2:%.*]] = extractelement <2 x float> [[SPLIT2]], i64 1 133; CHECK-NEXT: [[SPLAT_SPLATINSERT6:%.*]] = insertelement <1 x float> poison, float [[TMP2]], i64 0 134; CHECK-NEXT: [[SPLAT_SPLAT7:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT6]], <1 x float> poison, <1 x i32> zeroinitializer 135; CHECK-NEXT: [[TMP3:%.*]] = fmul <1 x float> [[BLOCK5]], [[SPLAT_SPLAT7]] 136; CHECK-NEXT: [[TMP4:%.*]] = fadd <1 x float> [[TMP1]], [[TMP3]] 137; CHECK-NEXT: [[TMP5:%.*]] = shufflevector <1 x float> [[TMP4]], <1 x float> poison, <3 x i32> <i32 0, i32 poison, i32 poison> 138; CHECK-NEXT: [[TMP6:%.*]] = shufflevector <3 x float> poison, <3 x float> [[TMP5]], <3 x i32> <i32 3, i32 1, i32 2> 139; CHECK-NEXT: [[BLOCK8:%.*]] = shufflevector <3 x float> [[SPLIT]], <3 x float> poison, <1 x i32> <i32 1> 140; CHECK-NEXT: [[TMP7:%.*]] = extractelement <2 x float> [[SPLIT2]], i64 0 141; CHECK-NEXT: [[SPLAT_SPLATINSERT9:%.*]] = insertelement <1 x float> poison, float [[TMP7]], i64 0 142; CHECK-NEXT: [[SPLAT_SPLAT10:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT9]], <1 x float> poison, <1 x i32> zeroinitializer 143; CHECK-NEXT: [[TMP8:%.*]] = fmul <1 x float> [[BLOCK8]], [[SPLAT_SPLAT10]] 144; CHECK-NEXT: [[BLOCK11:%.*]] = shufflevector <3 x float> [[SPLIT1]], <3 x float> poison, <1 x i32> <i32 1> 145; CHECK-NEXT: [[TMP9:%.*]] = extractelement <2 x float> [[SPLIT2]], i64 1 146; CHECK-NEXT: [[SPLAT_SPLATINSERT12:%.*]] = insertelement <1 x float> poison, float [[TMP9]], i64 0 147; CHECK-NEXT: [[SPLAT_SPLAT13:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT12]], <1 x float> poison, <1 x i32> zeroinitializer 148; CHECK-NEXT: [[TMP10:%.*]] = fmul <1 x float> [[BLOCK11]], [[SPLAT_SPLAT13]] 149; CHECK-NEXT: [[TMP11:%.*]] = fadd <1 x float> [[TMP8]], [[TMP10]] 150; CHECK-NEXT: [[TMP12:%.*]] = shufflevector <1 x float> [[TMP11]], <1 x float> poison, <3 x i32> <i32 0, i32 poison, i32 poison> 151; CHECK-NEXT: [[TMP13:%.*]] = shufflevector <3 x float> [[TMP6]], <3 x float> [[TMP12]], <3 x i32> <i32 0, i32 3, i32 2> 152; CHECK-NEXT: [[BLOCK14:%.*]] = shufflevector <3 x float> [[SPLIT]], <3 x float> poison, <1 x i32> <i32 2> 153; CHECK-NEXT: [[TMP14:%.*]] = extractelement <2 x float> [[SPLIT2]], i64 0 154; CHECK-NEXT: [[SPLAT_SPLATINSERT15:%.*]] = insertelement <1 x float> poison, float [[TMP14]], i64 0 155; CHECK-NEXT: [[SPLAT_SPLAT16:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT15]], <1 x float> poison, <1 x i32> zeroinitializer 156; CHECK-NEXT: [[TMP15:%.*]] = fmul <1 x float> [[BLOCK14]], [[SPLAT_SPLAT16]] 157; CHECK-NEXT: [[BLOCK17:%.*]] = shufflevector <3 x float> [[SPLIT1]], <3 x float> poison, <1 x i32> <i32 2> 158; CHECK-NEXT: [[TMP16:%.*]] = extractelement <2 x float> [[SPLIT2]], i64 1 159; CHECK-NEXT: [[SPLAT_SPLATINSERT18:%.*]] = insertelement <1 x float> poison, float [[TMP16]], i64 0 160; CHECK-NEXT: [[SPLAT_SPLAT19:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT18]], <1 x float> poison, <1 x i32> zeroinitializer 161; CHECK-NEXT: [[TMP17:%.*]] = fmul <1 x float> [[BLOCK17]], [[SPLAT_SPLAT19]] 162; CHECK-NEXT: [[TMP18:%.*]] = fadd <1 x float> [[TMP15]], [[TMP17]] 163; CHECK-NEXT: [[TMP19:%.*]] = shufflevector <1 x float> [[TMP18]], <1 x float> poison, <3 x i32> <i32 0, i32 poison, i32 poison> 164; CHECK-NEXT: [[TMP20:%.*]] = shufflevector <3 x float> [[TMP13]], <3 x float> [[TMP19]], <3 x i32> <i32 0, i32 1, i32 3> 165; CHECK-NEXT: [[BLOCK20:%.*]] = shufflevector <3 x float> [[SPLIT]], <3 x float> poison, <1 x i32> zeroinitializer 166; CHECK-NEXT: [[TMP21:%.*]] = extractelement <2 x float> [[SPLIT3]], i64 0 167; CHECK-NEXT: [[SPLAT_SPLATINSERT21:%.*]] = insertelement <1 x float> poison, float [[TMP21]], i64 0 168; CHECK-NEXT: [[SPLAT_SPLAT22:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT21]], <1 x float> poison, <1 x i32> zeroinitializer 169; CHECK-NEXT: [[TMP22:%.*]] = fmul <1 x float> [[BLOCK20]], [[SPLAT_SPLAT22]] 170; CHECK-NEXT: [[BLOCK23:%.*]] = shufflevector <3 x float> [[SPLIT1]], <3 x float> poison, <1 x i32> zeroinitializer 171; CHECK-NEXT: [[TMP23:%.*]] = extractelement <2 x float> [[SPLIT3]], i64 1 172; CHECK-NEXT: [[SPLAT_SPLATINSERT24:%.*]] = insertelement <1 x float> poison, float [[TMP23]], i64 0 173; CHECK-NEXT: [[SPLAT_SPLAT25:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT24]], <1 x float> poison, <1 x i32> zeroinitializer 174; CHECK-NEXT: [[TMP24:%.*]] = fmul <1 x float> [[BLOCK23]], [[SPLAT_SPLAT25]] 175; CHECK-NEXT: [[TMP25:%.*]] = fadd <1 x float> [[TMP22]], [[TMP24]] 176; CHECK-NEXT: [[TMP26:%.*]] = shufflevector <1 x float> [[TMP25]], <1 x float> poison, <3 x i32> <i32 0, i32 poison, i32 poison> 177; CHECK-NEXT: [[TMP27:%.*]] = shufflevector <3 x float> poison, <3 x float> [[TMP26]], <3 x i32> <i32 3, i32 1, i32 2> 178; CHECK-NEXT: [[BLOCK26:%.*]] = shufflevector <3 x float> [[SPLIT]], <3 x float> poison, <1 x i32> <i32 1> 179; CHECK-NEXT: [[TMP28:%.*]] = extractelement <2 x float> [[SPLIT3]], i64 0 180; CHECK-NEXT: [[SPLAT_SPLATINSERT27:%.*]] = insertelement <1 x float> poison, float [[TMP28]], i64 0 181; CHECK-NEXT: [[SPLAT_SPLAT28:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT27]], <1 x float> poison, <1 x i32> zeroinitializer 182; CHECK-NEXT: [[TMP29:%.*]] = fmul <1 x float> [[BLOCK26]], [[SPLAT_SPLAT28]] 183; CHECK-NEXT: [[BLOCK29:%.*]] = shufflevector <3 x float> [[SPLIT1]], <3 x float> poison, <1 x i32> <i32 1> 184; CHECK-NEXT: [[TMP30:%.*]] = extractelement <2 x float> [[SPLIT3]], i64 1 185; CHECK-NEXT: [[SPLAT_SPLATINSERT30:%.*]] = insertelement <1 x float> poison, float [[TMP30]], i64 0 186; CHECK-NEXT: [[SPLAT_SPLAT31:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT30]], <1 x float> poison, <1 x i32> zeroinitializer 187; CHECK-NEXT: [[TMP31:%.*]] = fmul <1 x float> [[BLOCK29]], [[SPLAT_SPLAT31]] 188; CHECK-NEXT: [[TMP32:%.*]] = fadd <1 x float> [[TMP29]], [[TMP31]] 189; CHECK-NEXT: [[TMP33:%.*]] = shufflevector <1 x float> [[TMP32]], <1 x float> poison, <3 x i32> <i32 0, i32 poison, i32 poison> 190; CHECK-NEXT: [[TMP34:%.*]] = shufflevector <3 x float> [[TMP27]], <3 x float> [[TMP33]], <3 x i32> <i32 0, i32 3, i32 2> 191; CHECK-NEXT: [[BLOCK32:%.*]] = shufflevector <3 x float> [[SPLIT]], <3 x float> poison, <1 x i32> <i32 2> 192; CHECK-NEXT: [[TMP35:%.*]] = extractelement <2 x float> [[SPLIT3]], i64 0 193; CHECK-NEXT: [[SPLAT_SPLATINSERT33:%.*]] = insertelement <1 x float> poison, float [[TMP35]], i64 0 194; CHECK-NEXT: [[SPLAT_SPLAT34:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT33]], <1 x float> poison, <1 x i32> zeroinitializer 195; CHECK-NEXT: [[TMP36:%.*]] = fmul <1 x float> [[BLOCK32]], [[SPLAT_SPLAT34]] 196; CHECK-NEXT: [[BLOCK35:%.*]] = shufflevector <3 x float> [[SPLIT1]], <3 x float> poison, <1 x i32> <i32 2> 197; CHECK-NEXT: [[TMP37:%.*]] = extractelement <2 x float> [[SPLIT3]], i64 1 198; CHECK-NEXT: [[SPLAT_SPLATINSERT36:%.*]] = insertelement <1 x float> poison, float [[TMP37]], i64 0 199; CHECK-NEXT: [[SPLAT_SPLAT37:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT36]], <1 x float> poison, <1 x i32> zeroinitializer 200; CHECK-NEXT: [[TMP38:%.*]] = fmul <1 x float> [[BLOCK35]], [[SPLAT_SPLAT37]] 201; CHECK-NEXT: [[TMP39:%.*]] = fadd <1 x float> [[TMP36]], [[TMP38]] 202; CHECK-NEXT: [[TMP40:%.*]] = shufflevector <1 x float> [[TMP39]], <1 x float> poison, <3 x i32> <i32 0, i32 poison, i32 poison> 203; CHECK-NEXT: [[TMP41:%.*]] = shufflevector <3 x float> [[TMP34]], <3 x float> [[TMP40]], <3 x i32> <i32 0, i32 1, i32 3> 204; CHECK-NEXT: [[BLOCK38:%.*]] = shufflevector <3 x float> [[SPLIT]], <3 x float> poison, <1 x i32> zeroinitializer 205; CHECK-NEXT: [[TMP42:%.*]] = extractelement <2 x float> [[SPLIT4]], i64 0 206; CHECK-NEXT: [[SPLAT_SPLATINSERT39:%.*]] = insertelement <1 x float> poison, float [[TMP42]], i64 0 207; CHECK-NEXT: [[SPLAT_SPLAT40:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT39]], <1 x float> poison, <1 x i32> zeroinitializer 208; CHECK-NEXT: [[TMP43:%.*]] = fmul <1 x float> [[BLOCK38]], [[SPLAT_SPLAT40]] 209; CHECK-NEXT: [[BLOCK41:%.*]] = shufflevector <3 x float> [[SPLIT1]], <3 x float> poison, <1 x i32> zeroinitializer 210; CHECK-NEXT: [[TMP44:%.*]] = extractelement <2 x float> [[SPLIT4]], i64 1 211; CHECK-NEXT: [[SPLAT_SPLATINSERT42:%.*]] = insertelement <1 x float> poison, float [[TMP44]], i64 0 212; CHECK-NEXT: [[SPLAT_SPLAT43:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT42]], <1 x float> poison, <1 x i32> zeroinitializer 213; CHECK-NEXT: [[TMP45:%.*]] = fmul <1 x float> [[BLOCK41]], [[SPLAT_SPLAT43]] 214; CHECK-NEXT: [[TMP46:%.*]] = fadd <1 x float> [[TMP43]], [[TMP45]] 215; CHECK-NEXT: [[TMP47:%.*]] = shufflevector <1 x float> [[TMP46]], <1 x float> poison, <3 x i32> <i32 0, i32 poison, i32 poison> 216; CHECK-NEXT: [[TMP48:%.*]] = shufflevector <3 x float> poison, <3 x float> [[TMP47]], <3 x i32> <i32 3, i32 1, i32 2> 217; CHECK-NEXT: [[BLOCK44:%.*]] = shufflevector <3 x float> [[SPLIT]], <3 x float> poison, <1 x i32> <i32 1> 218; CHECK-NEXT: [[TMP49:%.*]] = extractelement <2 x float> [[SPLIT4]], i64 0 219; CHECK-NEXT: [[SPLAT_SPLATINSERT45:%.*]] = insertelement <1 x float> poison, float [[TMP49]], i64 0 220; CHECK-NEXT: [[SPLAT_SPLAT46:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT45]], <1 x float> poison, <1 x i32> zeroinitializer 221; CHECK-NEXT: [[TMP50:%.*]] = fmul <1 x float> [[BLOCK44]], [[SPLAT_SPLAT46]] 222; CHECK-NEXT: [[BLOCK47:%.*]] = shufflevector <3 x float> [[SPLIT1]], <3 x float> poison, <1 x i32> <i32 1> 223; CHECK-NEXT: [[TMP51:%.*]] = extractelement <2 x float> [[SPLIT4]], i64 1 224; CHECK-NEXT: [[SPLAT_SPLATINSERT48:%.*]] = insertelement <1 x float> poison, float [[TMP51]], i64 0 225; CHECK-NEXT: [[SPLAT_SPLAT49:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT48]], <1 x float> poison, <1 x i32> zeroinitializer 226; CHECK-NEXT: [[TMP52:%.*]] = fmul <1 x float> [[BLOCK47]], [[SPLAT_SPLAT49]] 227; CHECK-NEXT: [[TMP53:%.*]] = fadd <1 x float> [[TMP50]], [[TMP52]] 228; CHECK-NEXT: [[TMP54:%.*]] = shufflevector <1 x float> [[TMP53]], <1 x float> poison, <3 x i32> <i32 0, i32 poison, i32 poison> 229; CHECK-NEXT: [[TMP55:%.*]] = shufflevector <3 x float> [[TMP48]], <3 x float> [[TMP54]], <3 x i32> <i32 0, i32 3, i32 2> 230; CHECK-NEXT: [[BLOCK50:%.*]] = shufflevector <3 x float> [[SPLIT]], <3 x float> poison, <1 x i32> <i32 2> 231; CHECK-NEXT: [[TMP56:%.*]] = extractelement <2 x float> [[SPLIT4]], i64 0 232; CHECK-NEXT: [[SPLAT_SPLATINSERT51:%.*]] = insertelement <1 x float> poison, float [[TMP56]], i64 0 233; CHECK-NEXT: [[SPLAT_SPLAT52:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT51]], <1 x float> poison, <1 x i32> zeroinitializer 234; CHECK-NEXT: [[TMP57:%.*]] = fmul <1 x float> [[BLOCK50]], [[SPLAT_SPLAT52]] 235; CHECK-NEXT: [[BLOCK53:%.*]] = shufflevector <3 x float> [[SPLIT1]], <3 x float> poison, <1 x i32> <i32 2> 236; CHECK-NEXT: [[TMP58:%.*]] = extractelement <2 x float> [[SPLIT4]], i64 1 237; CHECK-NEXT: [[SPLAT_SPLATINSERT54:%.*]] = insertelement <1 x float> poison, float [[TMP58]], i64 0 238; CHECK-NEXT: [[SPLAT_SPLAT55:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT54]], <1 x float> poison, <1 x i32> zeroinitializer 239; CHECK-NEXT: [[TMP59:%.*]] = fmul <1 x float> [[BLOCK53]], [[SPLAT_SPLAT55]] 240; CHECK-NEXT: [[TMP60:%.*]] = fadd <1 x float> [[TMP57]], [[TMP59]] 241; CHECK-NEXT: [[TMP61:%.*]] = shufflevector <1 x float> [[TMP60]], <1 x float> poison, <3 x i32> <i32 0, i32 poison, i32 poison> 242; CHECK-NEXT: [[TMP62:%.*]] = shufflevector <3 x float> [[TMP55]], <3 x float> [[TMP61]], <3 x i32> <i32 0, i32 1, i32 3> 243; CHECK-NEXT: [[TMP63:%.*]] = shufflevector <3 x float> [[TMP20]], <3 x float> [[TMP41]], <6 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5> 244; CHECK-NEXT: [[TMP64:%.*]] = shufflevector <3 x float> [[TMP62]], <3 x float> poison, <6 x i32> <i32 0, i32 1, i32 2, i32 poison, i32 poison, i32 poison> 245; CHECK-NEXT: [[TMP65:%.*]] = shufflevector <6 x float> [[TMP63]], <6 x float> [[TMP64]], <9 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8> 246; CHECK-NEXT: ret <9 x float> [[TMP65]] 247; 248entry: 249 %c = call <9 x float> @llvm.matrix.multiply.v6f32.v6f32.v6f32(<6 x float> %a, <6 x float> %b, i32 3, i32 2, i32 3) 250 ret <9 x float> %c 251} 252 253declare <9 x float> @llvm.matrix.multiply.v6f32.v6f32.v6f32(<6 x float>, <6 x float>, i32, i32, i32) 254