[llvm] d4d6e71 - [RISCV] Add empirical costs for bswap/bitreverse/ctpop/ctlz/cttz
Philip Reames via llvm-commits
llvm-commits at lists.llvm.org
Wed Aug 24 15:11:43 PDT 2022
Author: Philip Reames
Date: 2022-08-24T15:09:21-07:00
New Revision: d4d6e71ea20bfdc22f57522bb118f04ecf08ef38
URL: https://github.com/llvm/llvm-project/commit/d4d6e71ea20bfdc22f57522bb118f04ecf08ef38
DIFF: https://github.com/llvm/llvm-project/commit/d4d6e71ea20bfdc22f57522bb118f04ecf08ef38.diff
LOG: [RISCV] Add empirical costs for bswap/bitreverse/ctpop/ctlz/cttz
If anyone is looking for a source of ideas on vector codegen improvements, the lowerings for several of these seem to include pretty obvious fixits.
Added:
Modified:
llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
llvm/test/Analysis/CostModel/RISCV/integer-bit-manip.ll
Removed:
################################################################################
diff --git a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
index 446a08124957..7a26959f1116 100644
--- a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
+++ b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
@@ -353,6 +353,153 @@ static const CostTblEntry VectorIntrinsicCostTable[]{
{Intrinsic::sqrt, MVT::nxv2f64, 1},
{Intrinsic::sqrt, MVT::nxv4f64, 1},
{Intrinsic::sqrt, MVT::nxv8f64, 1},
+ {Intrinsic::bswap, MVT::v2i16, 3},
+ {Intrinsic::bswap, MVT::v4i16, 3},
+ {Intrinsic::bswap, MVT::v8i16, 3},
+ {Intrinsic::bswap, MVT::v16i16, 3},
+ {Intrinsic::bswap, MVT::nxv2i16, 3},
+ {Intrinsic::bswap, MVT::nxv4i16, 3},
+ {Intrinsic::bswap, MVT::nxv8i16, 3},
+ {Intrinsic::bswap, MVT::nxv16i16, 3},
+ {Intrinsic::bswap, MVT::v2i32, 12},
+ {Intrinsic::bswap, MVT::v4i32, 12},
+ {Intrinsic::bswap, MVT::v8i32, 12},
+ {Intrinsic::bswap, MVT::v16i32, 12},
+ {Intrinsic::bswap, MVT::nxv2i32, 12},
+ {Intrinsic::bswap, MVT::nxv4i32, 12},
+ {Intrinsic::bswap, MVT::nxv8i32, 12},
+ {Intrinsic::bswap, MVT::nxv16i32, 12},
+ {Intrinsic::bswap, MVT::v2i64, 31},
+ {Intrinsic::bswap, MVT::v4i64, 31},
+ {Intrinsic::bswap, MVT::v8i64, 31},
+ {Intrinsic::bswap, MVT::v16i64, 31},
+ {Intrinsic::bswap, MVT::nxv2i64, 31},
+ {Intrinsic::bswap, MVT::nxv4i64, 31},
+ {Intrinsic::bswap, MVT::nxv8i64, 31},
+ {Intrinsic::bitreverse, MVT::v2i8, 17},
+ {Intrinsic::bitreverse, MVT::v4i8, 17},
+ {Intrinsic::bitreverse, MVT::v8i8, 17},
+ {Intrinsic::bitreverse, MVT::v16i8, 17},
+ {Intrinsic::bitreverse, MVT::nxv2i8, 17},
+ {Intrinsic::bitreverse, MVT::nxv4i8, 17},
+ {Intrinsic::bitreverse, MVT::nxv8i8, 17},
+ {Intrinsic::bitreverse, MVT::nxv16i8, 17},
+ {Intrinsic::bitreverse, MVT::v2i16, 24},
+ {Intrinsic::bitreverse, MVT::v4i16, 24},
+ {Intrinsic::bitreverse, MVT::v8i16, 24},
+ {Intrinsic::bitreverse, MVT::v16i16, 24},
+ {Intrinsic::bitreverse, MVT::nxv2i16, 24},
+ {Intrinsic::bitreverse, MVT::nxv4i16, 24},
+ {Intrinsic::bitreverse, MVT::nxv8i16, 24},
+ {Intrinsic::bitreverse, MVT::nxv16i16, 24},
+ {Intrinsic::bitreverse, MVT::v2i32, 33},
+ {Intrinsic::bitreverse, MVT::v4i32, 33},
+ {Intrinsic::bitreverse, MVT::v8i32, 33},
+ {Intrinsic::bitreverse, MVT::v16i32, 33},
+ {Intrinsic::bitreverse, MVT::nxv2i32, 33},
+ {Intrinsic::bitreverse, MVT::nxv4i32, 33},
+ {Intrinsic::bitreverse, MVT::nxv8i32, 33},
+ {Intrinsic::bitreverse, MVT::nxv16i32, 33},
+ {Intrinsic::bitreverse, MVT::v2i64, 52},
+ {Intrinsic::bitreverse, MVT::v4i64, 52},
+ {Intrinsic::bitreverse, MVT::v8i64, 52},
+ {Intrinsic::bitreverse, MVT::v16i64, 52},
+ {Intrinsic::bitreverse, MVT::nxv2i64, 52},
+ {Intrinsic::bitreverse, MVT::nxv4i64, 52},
+ {Intrinsic::bitreverse, MVT::nxv8i64, 52},
+ {Intrinsic::ctpop, MVT::v2i8, 12},
+ {Intrinsic::ctpop, MVT::v4i8, 12},
+ {Intrinsic::ctpop, MVT::v8i8, 12},
+ {Intrinsic::ctpop, MVT::v16i8, 12},
+ {Intrinsic::ctpop, MVT::nxv2i8, 12},
+ {Intrinsic::ctpop, MVT::nxv4i8, 12},
+ {Intrinsic::ctpop, MVT::nxv8i8, 12},
+ {Intrinsic::ctpop, MVT::nxv16i8, 12},
+ {Intrinsic::ctpop, MVT::v2i16, 19},
+ {Intrinsic::ctpop, MVT::v4i16, 19},
+ {Intrinsic::ctpop, MVT::v8i16, 19},
+ {Intrinsic::ctpop, MVT::v16i16, 19},
+ {Intrinsic::ctpop, MVT::nxv2i16, 19},
+ {Intrinsic::ctpop, MVT::nxv4i16, 19},
+ {Intrinsic::ctpop, MVT::nxv8i16, 19},
+ {Intrinsic::ctpop, MVT::nxv16i16, 19},
+ {Intrinsic::ctpop, MVT::v2i32, 20},
+ {Intrinsic::ctpop, MVT::v4i32, 20},
+ {Intrinsic::ctpop, MVT::v8i32, 20},
+ {Intrinsic::ctpop, MVT::v16i32, 20},
+ {Intrinsic::ctpop, MVT::nxv2i32, 20},
+ {Intrinsic::ctpop, MVT::nxv4i32, 20},
+ {Intrinsic::ctpop, MVT::nxv8i32, 20},
+ {Intrinsic::ctpop, MVT::nxv16i32, 20},
+ {Intrinsic::ctpop, MVT::v2i64, 21},
+ {Intrinsic::ctpop, MVT::v4i64, 21},
+ {Intrinsic::ctpop, MVT::v8i64, 21},
+ {Intrinsic::ctpop, MVT::v16i64, 21},
+ {Intrinsic::ctpop, MVT::nxv2i64, 21},
+ {Intrinsic::ctpop, MVT::nxv4i64, 21},
+ {Intrinsic::ctpop, MVT::nxv8i64, 21},
+ {Intrinsic::ctlz, MVT::v2i8, 10},
+ {Intrinsic::ctlz, MVT::v4i8, 10},
+ {Intrinsic::ctlz, MVT::v8i8, 10},
+ {Intrinsic::ctlz, MVT::v16i8, 10},
+ {Intrinsic::ctlz, MVT::nxv2i8, 10},
+ {Intrinsic::ctlz, MVT::nxv4i8, 10},
+ {Intrinsic::ctlz, MVT::nxv8i8, 10},
+ {Intrinsic::ctlz, MVT::nxv16i8, 10},
+ {Intrinsic::ctlz, MVT::v2i16, 7},
+ {Intrinsic::ctlz, MVT::v4i16, 7},
+ {Intrinsic::ctlz, MVT::v8i16, 7},
+ {Intrinsic::ctlz, MVT::v16i16, 7},
+ {Intrinsic::ctlz, MVT::nxv2i16, 7},
+ {Intrinsic::ctlz, MVT::nxv4i16, 7},
+ {Intrinsic::ctlz, MVT::nxv8i16, 7},
+ {Intrinsic::ctlz, MVT::nxv16i16, 7},
+ {Intrinsic::ctlz, MVT::v2i32, 8},
+ {Intrinsic::ctlz, MVT::v4i32, 8},
+ {Intrinsic::ctlz, MVT::v8i32, 8},
+ {Intrinsic::ctlz, MVT::v16i32, 8},
+ {Intrinsic::ctlz, MVT::nxv2i32, 11},
+ {Intrinsic::ctlz, MVT::nxv4i32, 11},
+ {Intrinsic::ctlz, MVT::nxv8i32, 11},
+ {Intrinsic::ctlz, MVT::nxv16i32, 31},
+ {Intrinsic::ctlz, MVT::v2i64, 35},
+ {Intrinsic::ctlz, MVT::v4i64, 35},
+ {Intrinsic::ctlz, MVT::v8i64, 35},
+ {Intrinsic::ctlz, MVT::v16i64, 35},
+ {Intrinsic::ctlz, MVT::nxv2i64, 35},
+ {Intrinsic::ctlz, MVT::nxv4i64, 35},
+ {Intrinsic::ctlz, MVT::nxv8i64, 35},
+ {Intrinsic::cttz, MVT::v2i8, 13},
+ {Intrinsic::cttz, MVT::v4i8, 13},
+ {Intrinsic::cttz, MVT::v8i8, 13},
+ {Intrinsic::cttz, MVT::v16i8, 13},
+ {Intrinsic::cttz, MVT::nxv2i8, 13},
+ {Intrinsic::cttz, MVT::nxv4i8, 13},
+ {Intrinsic::cttz, MVT::nxv8i8, 13},
+ {Intrinsic::cttz, MVT::nxv16i8, 13},
+ {Intrinsic::cttz, MVT::v2i16, 9},
+ {Intrinsic::cttz, MVT::v4i16, 9},
+ {Intrinsic::cttz, MVT::v8i16, 9},
+ {Intrinsic::cttz, MVT::v16i16, 9},
+ {Intrinsic::cttz, MVT::nxv2i16, 9},
+ {Intrinsic::cttz, MVT::nxv4i16, 9},
+ {Intrinsic::cttz, MVT::nxv8i16, 9},
+ {Intrinsic::cttz, MVT::nxv16i16, 9},
+ {Intrinsic::cttz, MVT::v2i32, 10},
+ {Intrinsic::cttz, MVT::v4i32, 10},
+ {Intrinsic::cttz, MVT::v8i32, 10},
+ {Intrinsic::cttz, MVT::v16i32, 10},
+ {Intrinsic::cttz, MVT::nxv2i32, 13},
+ {Intrinsic::cttz, MVT::nxv4i32, 13},
+ {Intrinsic::cttz, MVT::nxv8i32, 13},
+ {Intrinsic::cttz, MVT::nxv16i32, 24},
+ {Intrinsic::cttz, MVT::v2i64, 25},
+ {Intrinsic::cttz, MVT::v4i64, 25},
+ {Intrinsic::cttz, MVT::v8i64, 25},
+ {Intrinsic::cttz, MVT::v16i64, 25},
+ {Intrinsic::cttz, MVT::nxv2i64, 25},
+ {Intrinsic::cttz, MVT::nxv4i64, 25},
+ {Intrinsic::cttz, MVT::nxv8i64, 25},
};
InstructionCost
diff --git a/llvm/test/Analysis/CostModel/RISCV/integer-bit-manip.ll b/llvm/test/Analysis/CostModel/RISCV/integer-bit-manip.ll
index decf95973e28..d74c61a41a91 100644
--- a/llvm/test/Analysis/CostModel/RISCV/integer-bit-manip.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/integer-bit-manip.ll
@@ -4,32 +4,32 @@
define void @bswap() {
; CHECK-LABEL: 'bswap'
; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %1 = call i16 @llvm.bswap.i16(i16 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %2 = call <2 x i16> @llvm.bswap.v2i16(<2 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %3 = call <4 x i16> @llvm.bswap.v4i16(<4 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %4 = call <8 x i16> @llvm.bswap.v8i16(<8 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 176 for instruction: %5 = call <16 x i16> @llvm.bswap.v16i16(<16 x i16> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %6 = call <vscale x 2 x i16> @llvm.bswap.nxv2i16(<vscale x 2 x i16> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %7 = call <vscale x 4 x i16> @llvm.bswap.nxv4i16(<vscale x 4 x i16> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %8 = call <vscale x 8 x i16> @llvm.bswap.nxv8i16(<vscale x 8 x i16> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %9 = call <vscale x 16 x i16> @llvm.bswap.nxv16i16(<vscale x 16 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %2 = call <2 x i16> @llvm.bswap.v2i16(<2 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %3 = call <4 x i16> @llvm.bswap.v4i16(<4 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %4 = call <8 x i16> @llvm.bswap.v8i16(<8 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %5 = call <16 x i16> @llvm.bswap.v16i16(<16 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %6 = call <vscale x 2 x i16> @llvm.bswap.nxv2i16(<vscale x 2 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %7 = call <vscale x 4 x i16> @llvm.bswap.nxv4i16(<vscale x 4 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %8 = call <vscale x 8 x i16> @llvm.bswap.nxv8i16(<vscale x 8 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %9 = call <vscale x 16 x i16> @llvm.bswap.nxv16i16(<vscale x 16 x i16> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %10 = call i32 @llvm.bswap.i32(i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %11 = call <2 x i32> @llvm.bswap.v2i32(<2 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %12 = call <4 x i32> @llvm.bswap.v4i32(<4 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %13 = call <8 x i32> @llvm.bswap.v8i32(<8 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 176 for instruction: %14 = call <16 x i32> @llvm.bswap.v16i32(<16 x i32> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %15 = call <vscale x 2 x i32> @llvm.bswap.nxv2i32(<vscale x 2 x i32> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %16 = call <vscale x 4 x i32> @llvm.bswap.nxv4i32(<vscale x 4 x i32> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %17 = call <vscale x 8 x i32> @llvm.bswap.nxv8i32(<vscale x 8 x i32> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %18 = call <vscale x 16 x i32> @llvm.bswap.nxv16i32(<vscale x 16 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %11 = call <2 x i32> @llvm.bswap.v2i32(<2 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %12 = call <4 x i32> @llvm.bswap.v4i32(<4 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %13 = call <8 x i32> @llvm.bswap.v8i32(<8 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %14 = call <16 x i32> @llvm.bswap.v16i32(<16 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %15 = call <vscale x 2 x i32> @llvm.bswap.nxv2i32(<vscale x 2 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %16 = call <vscale x 4 x i32> @llvm.bswap.nxv4i32(<vscale x 4 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %17 = call <vscale x 8 x i32> @llvm.bswap.nxv8i32(<vscale x 8 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %18 = call <vscale x 16 x i32> @llvm.bswap.nxv16i32(<vscale x 16 x i32> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %19 = call i64 @llvm.bswap.i64(i64 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %20 = call <2 x i64> @llvm.bswap.v2i64(<2 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %21 = call <4 x i64> @llvm.bswap.v4i64(<4 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %22 = call <8 x i64> @llvm.bswap.v8i64(<8 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 176 for instruction: %23 = call <16 x i64> @llvm.bswap.v16i64(<16 x i64> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %24 = call <vscale x 2 x i64> @llvm.bswap.nxv2i64(<vscale x 2 x i64> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %25 = call <vscale x 4 x i64> @llvm.bswap.nxv4i64(<vscale x 4 x i64> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %26 = call <vscale x 8 x i64> @llvm.bswap.nxv8i64(<vscale x 8 x i64> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %27 = call <vscale x 16 x i64> @llvm.bswap.nxv16i64(<vscale x 16 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %20 = call <2 x i64> @llvm.bswap.v2i64(<2 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %21 = call <4 x i64> @llvm.bswap.v4i64(<4 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %22 = call <8 x i64> @llvm.bswap.v8i64(<8 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %23 = call <16 x i64> @llvm.bswap.v16i64(<16 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %24 = call <vscale x 2 x i64> @llvm.bswap.nxv2i64(<vscale x 2 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %25 = call <vscale x 4 x i64> @llvm.bswap.nxv4i64(<vscale x 4 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %26 = call <vscale x 8 x i64> @llvm.bswap.nxv8i64(<vscale x 8 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 62 for instruction: %27 = call <vscale x 16 x i64> @llvm.bswap.nxv16i64(<vscale x 16 x i64> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret void
;
call i16 @llvm.bswap.i16(i16 undef)
@@ -65,41 +65,41 @@ define void @bswap() {
define void @bitreverse() {
; CHECK-LABEL: 'bitreverse'
; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %1 = call i8 @llvm.bitreverse.i8(i8 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %2 = call <2 x i8> @llvm.bitreverse.v2i8(<2 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %3 = call <4 x i8> @llvm.bitreverse.v4i8(<4 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %4 = call <8 x i8> @llvm.bitreverse.v8i8(<8 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 176 for instruction: %5 = call <16 x i8> @llvm.bitreverse.v16i8(<16 x i8> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %6 = call <vscale x 2 x i8> @llvm.bitreverse.nxv2i8(<vscale x 2 x i8> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %7 = call <vscale x 4 x i8> @llvm.bitreverse.nxv4i8(<vscale x 4 x i8> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %8 = call <vscale x 8 x i8> @llvm.bitreverse.nxv8i8(<vscale x 8 x i8> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %9 = call <vscale x 16 x i8> @llvm.bitreverse.nxv16i8(<vscale x 16 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %2 = call <2 x i8> @llvm.bitreverse.v2i8(<2 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %3 = call <4 x i8> @llvm.bitreverse.v4i8(<4 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %4 = call <8 x i8> @llvm.bitreverse.v8i8(<8 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %5 = call <16 x i8> @llvm.bitreverse.v16i8(<16 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %6 = call <vscale x 2 x i8> @llvm.bitreverse.nxv2i8(<vscale x 2 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %7 = call <vscale x 4 x i8> @llvm.bitreverse.nxv4i8(<vscale x 4 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %8 = call <vscale x 8 x i8> @llvm.bitreverse.nxv8i8(<vscale x 8 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %9 = call <vscale x 16 x i8> @llvm.bitreverse.nxv16i8(<vscale x 16 x i8> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %10 = call i16 @llvm.bitreverse.i16(i16 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %11 = call <2 x i16> @llvm.bitreverse.v2i16(<2 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %12 = call <4 x i16> @llvm.bitreverse.v4i16(<4 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %13 = call <8 x i16> @llvm.bitreverse.v8i16(<8 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 176 for instruction: %14 = call <16 x i16> @llvm.bitreverse.v16i16(<16 x i16> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %15 = call <vscale x 2 x i16> @llvm.bitreverse.nxv2i16(<vscale x 2 x i16> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %16 = call <vscale x 4 x i16> @llvm.bitreverse.nxv4i16(<vscale x 4 x i16> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %17 = call <vscale x 8 x i16> @llvm.bitreverse.nxv8i16(<vscale x 8 x i16> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %18 = call <vscale x 16 x i16> @llvm.bitreverse.nxv16i16(<vscale x 16 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %11 = call <2 x i16> @llvm.bitreverse.v2i16(<2 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %12 = call <4 x i16> @llvm.bitreverse.v4i16(<4 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %13 = call <8 x i16> @llvm.bitreverse.v8i16(<8 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %14 = call <16 x i16> @llvm.bitreverse.v16i16(<16 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %15 = call <vscale x 2 x i16> @llvm.bitreverse.nxv2i16(<vscale x 2 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %16 = call <vscale x 4 x i16> @llvm.bitreverse.nxv4i16(<vscale x 4 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %17 = call <vscale x 8 x i16> @llvm.bitreverse.nxv8i16(<vscale x 8 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %18 = call <vscale x 16 x i16> @llvm.bitreverse.nxv16i16(<vscale x 16 x i16> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %19 = call i32 @llvm.bitreverse.i32(i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %20 = call <2 x i32> @llvm.bitreverse.v2i32(<2 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %21 = call <4 x i32> @llvm.bitreverse.v4i32(<4 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %22 = call <8 x i32> @llvm.bitreverse.v8i32(<8 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 176 for instruction: %23 = call <16 x i32> @llvm.bitreverse.v16i32(<16 x i32> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %24 = call <vscale x 2 x i32> @llvm.bitreverse.nxv2i32(<vscale x 2 x i32> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %25 = call <vscale x 4 x i32> @llvm.bitreverse.nxv4i32(<vscale x 4 x i32> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %26 = call <vscale x 8 x i32> @llvm.bitreverse.nxv8i32(<vscale x 8 x i32> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %27 = call <vscale x 16 x i32> @llvm.bitreverse.nxv16i32(<vscale x 16 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 33 for instruction: %20 = call <2 x i32> @llvm.bitreverse.v2i32(<2 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 33 for instruction: %21 = call <4 x i32> @llvm.bitreverse.v4i32(<4 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 33 for instruction: %22 = call <8 x i32> @llvm.bitreverse.v8i32(<8 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 33 for instruction: %23 = call <16 x i32> @llvm.bitreverse.v16i32(<16 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 33 for instruction: %24 = call <vscale x 2 x i32> @llvm.bitreverse.nxv2i32(<vscale x 2 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 33 for instruction: %25 = call <vscale x 4 x i32> @llvm.bitreverse.nxv4i32(<vscale x 4 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 33 for instruction: %26 = call <vscale x 8 x i32> @llvm.bitreverse.nxv8i32(<vscale x 8 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 33 for instruction: %27 = call <vscale x 16 x i32> @llvm.bitreverse.nxv16i32(<vscale x 16 x i32> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %28 = call i64 @llvm.bitreverse.i64(i64 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %29 = call <2 x i64> @llvm.bitreverse.v2i64(<2 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %30 = call <4 x i64> @llvm.bitreverse.v4i64(<4 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %31 = call <8 x i64> @llvm.bitreverse.v8i64(<8 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 176 for instruction: %32 = call <16 x i64> @llvm.bitreverse.v16i64(<16 x i64> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %33 = call <vscale x 2 x i64> @llvm.bitreverse.nxv2i64(<vscale x 2 x i64> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %34 = call <vscale x 4 x i64> @llvm.bitreverse.nxv4i64(<vscale x 4 x i64> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %35 = call <vscale x 8 x i64> @llvm.bitreverse.nxv8i64(<vscale x 8 x i64> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %36 = call <vscale x 16 x i64> @llvm.bitreverse.nxv16i64(<vscale x 16 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %29 = call <2 x i64> @llvm.bitreverse.v2i64(<2 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %30 = call <4 x i64> @llvm.bitreverse.v4i64(<4 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %31 = call <8 x i64> @llvm.bitreverse.v8i64(<8 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %32 = call <16 x i64> @llvm.bitreverse.v16i64(<16 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %33 = call <vscale x 2 x i64> @llvm.bitreverse.nxv2i64(<vscale x 2 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %34 = call <vscale x 4 x i64> @llvm.bitreverse.nxv4i64(<vscale x 4 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %35 = call <vscale x 8 x i64> @llvm.bitreverse.nxv8i64(<vscale x 8 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 104 for instruction: %36 = call <vscale x 16 x i64> @llvm.bitreverse.nxv16i64(<vscale x 16 x i64> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret void
;
call i8 @llvm.bitreverse.i8(i8 undef)
@@ -144,41 +144,41 @@ define void @bitreverse() {
define void @ctpop() {
; CHECK-LABEL: 'ctpop'
; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %1 = call i8 @llvm.ctpop.i8(i8 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %2 = call <2 x i8> @llvm.ctpop.v2i8(<2 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %3 = call <4 x i8> @llvm.ctpop.v4i8(<4 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 40 for instruction: %4 = call <8 x i8> @llvm.ctpop.v8i8(<8 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 80 for instruction: %5 = call <16 x i8> @llvm.ctpop.v16i8(<16 x i8> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %6 = call <vscale x 2 x i8> @llvm.ctpop.nxv2i8(<vscale x 2 x i8> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %7 = call <vscale x 4 x i8> @llvm.ctpop.nxv4i8(<vscale x 4 x i8> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %8 = call <vscale x 8 x i8> @llvm.ctpop.nxv8i8(<vscale x 8 x i8> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %9 = call <vscale x 16 x i8> @llvm.ctpop.nxv16i8(<vscale x 16 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %2 = call <2 x i8> @llvm.ctpop.v2i8(<2 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %3 = call <4 x i8> @llvm.ctpop.v4i8(<4 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %4 = call <8 x i8> @llvm.ctpop.v8i8(<8 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %5 = call <16 x i8> @llvm.ctpop.v16i8(<16 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %6 = call <vscale x 2 x i8> @llvm.ctpop.nxv2i8(<vscale x 2 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %7 = call <vscale x 4 x i8> @llvm.ctpop.nxv4i8(<vscale x 4 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %8 = call <vscale x 8 x i8> @llvm.ctpop.nxv8i8(<vscale x 8 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %9 = call <vscale x 16 x i8> @llvm.ctpop.nxv16i8(<vscale x 16 x i8> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %10 = call i16 @llvm.ctpop.i16(i16 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %11 = call <2 x i16> @llvm.ctpop.v2i16(<2 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %12 = call <4 x i16> @llvm.ctpop.v4i16(<4 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 40 for instruction: %13 = call <8 x i16> @llvm.ctpop.v8i16(<8 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 80 for instruction: %14 = call <16 x i16> @llvm.ctpop.v16i16(<16 x i16> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %15 = call <vscale x 2 x i16> @llvm.ctpop.nxv2i16(<vscale x 2 x i16> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %16 = call <vscale x 4 x i16> @llvm.ctpop.nxv4i16(<vscale x 4 x i16> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %17 = call <vscale x 8 x i16> @llvm.ctpop.nxv8i16(<vscale x 8 x i16> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %18 = call <vscale x 16 x i16> @llvm.ctpop.nxv16i16(<vscale x 16 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %11 = call <2 x i16> @llvm.ctpop.v2i16(<2 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %12 = call <4 x i16> @llvm.ctpop.v4i16(<4 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %13 = call <8 x i16> @llvm.ctpop.v8i16(<8 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %14 = call <16 x i16> @llvm.ctpop.v16i16(<16 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %15 = call <vscale x 2 x i16> @llvm.ctpop.nxv2i16(<vscale x 2 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %16 = call <vscale x 4 x i16> @llvm.ctpop.nxv4i16(<vscale x 4 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %17 = call <vscale x 8 x i16> @llvm.ctpop.nxv8i16(<vscale x 8 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %18 = call <vscale x 16 x i16> @llvm.ctpop.nxv16i16(<vscale x 16 x i16> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %19 = call i32 @llvm.ctpop.i32(i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %20 = call <2 x i32> @llvm.ctpop.v2i32(<2 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %20 = call <2 x i32> @llvm.ctpop.v2i32(<2 x i32> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %21 = call <4 x i32> @llvm.ctpop.v4i32(<4 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 40 for instruction: %22 = call <8 x i32> @llvm.ctpop.v8i32(<8 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 80 for instruction: %23 = call <16 x i32> @llvm.ctpop.v16i32(<16 x i32> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %24 = call <vscale x 2 x i32> @llvm.ctpop.nxv2i32(<vscale x 2 x i32> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %25 = call <vscale x 4 x i32> @llvm.ctpop.nxv4i32(<vscale x 4 x i32> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %26 = call <vscale x 8 x i32> @llvm.ctpop.nxv8i32(<vscale x 8 x i32> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %27 = call <vscale x 16 x i32> @llvm.ctpop.nxv16i32(<vscale x 16 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %22 = call <8 x i32> @llvm.ctpop.v8i32(<8 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %23 = call <16 x i32> @llvm.ctpop.v16i32(<16 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %24 = call <vscale x 2 x i32> @llvm.ctpop.nxv2i32(<vscale x 2 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %25 = call <vscale x 4 x i32> @llvm.ctpop.nxv4i32(<vscale x 4 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %26 = call <vscale x 8 x i32> @llvm.ctpop.nxv8i32(<vscale x 8 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %27 = call <vscale x 16 x i32> @llvm.ctpop.nxv16i32(<vscale x 16 x i32> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %28 = call i64 @llvm.ctpop.i64(i64 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %29 = call <2 x i64> @llvm.ctpop.v2i64(<2 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %30 = call <4 x i64> @llvm.ctpop.v4i64(<4 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 40 for instruction: %31 = call <8 x i64> @llvm.ctpop.v8i64(<8 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 80 for instruction: %32 = call <16 x i64> @llvm.ctpop.v16i64(<16 x i64> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %33 = call <vscale x 2 x i64> @llvm.ctpop.nxv2i64(<vscale x 2 x i64> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %34 = call <vscale x 4 x i64> @llvm.ctpop.nxv4i64(<vscale x 4 x i64> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %35 = call <vscale x 8 x i64> @llvm.ctpop.nxv8i64(<vscale x 8 x i64> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %36 = call <vscale x 16 x i64> @llvm.ctpop.nxv16i64(<vscale x 16 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %29 = call <2 x i64> @llvm.ctpop.v2i64(<2 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %30 = call <4 x i64> @llvm.ctpop.v4i64(<4 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %31 = call <8 x i64> @llvm.ctpop.v8i64(<8 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %32 = call <16 x i64> @llvm.ctpop.v16i64(<16 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %33 = call <vscale x 2 x i64> @llvm.ctpop.nxv2i64(<vscale x 2 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %34 = call <vscale x 4 x i64> @llvm.ctpop.nxv4i64(<vscale x 4 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %35 = call <vscale x 8 x i64> @llvm.ctpop.nxv8i64(<vscale x 8 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 42 for instruction: %36 = call <vscale x 16 x i64> @llvm.ctpop.nxv16i64(<vscale x 16 x i64> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret void
;
call i8 @llvm.ctpop.i8(i8 undef)
@@ -223,41 +223,41 @@ define void @ctpop() {
define void @ctlz() {
; CHECK-LABEL: 'ctlz'
; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %1 = call i8 @llvm.ctlz.i8(i8 undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %2 = call <2 x i8> @llvm.ctlz.v2i8(<2 x i8> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %3 = call <4 x i8> @llvm.ctlz.v4i8(<4 x i8> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %4 = call <8 x i8> @llvm.ctlz.v8i8(<8 x i8> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 176 for instruction: %5 = call <16 x i8> @llvm.ctlz.v16i8(<16 x i8> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %6 = call <vscale x 2 x i8> @llvm.ctlz.nxv2i8(<vscale x 2 x i8> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %7 = call <vscale x 4 x i8> @llvm.ctlz.nxv4i8(<vscale x 4 x i8> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %8 = call <vscale x 8 x i8> @llvm.ctlz.nxv8i8(<vscale x 8 x i8> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %9 = call <vscale x 16 x i8> @llvm.ctlz.nxv16i8(<vscale x 16 x i8> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %2 = call <2 x i8> @llvm.ctlz.v2i8(<2 x i8> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %3 = call <4 x i8> @llvm.ctlz.v4i8(<4 x i8> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %4 = call <8 x i8> @llvm.ctlz.v8i8(<8 x i8> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %5 = call <16 x i8> @llvm.ctlz.v16i8(<16 x i8> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %6 = call <vscale x 2 x i8> @llvm.ctlz.nxv2i8(<vscale x 2 x i8> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %7 = call <vscale x 4 x i8> @llvm.ctlz.nxv4i8(<vscale x 4 x i8> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %8 = call <vscale x 8 x i8> @llvm.ctlz.nxv8i8(<vscale x 8 x i8> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %9 = call <vscale x 16 x i8> @llvm.ctlz.nxv16i8(<vscale x 16 x i8> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %10 = call i16 @llvm.ctlz.i16(i16 undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %11 = call <2 x i16> @llvm.ctlz.v2i16(<2 x i16> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %12 = call <4 x i16> @llvm.ctlz.v4i16(<4 x i16> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %13 = call <8 x i16> @llvm.ctlz.v8i16(<8 x i16> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 176 for instruction: %14 = call <16 x i16> @llvm.ctlz.v16i16(<16 x i16> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %15 = call <vscale x 2 x i16> @llvm.ctlz.nxv2i16(<vscale x 2 x i16> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %16 = call <vscale x 4 x i16> @llvm.ctlz.nxv4i16(<vscale x 4 x i16> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %17 = call <vscale x 8 x i16> @llvm.ctlz.nxv8i16(<vscale x 8 x i16> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %18 = call <vscale x 16 x i16> @llvm.ctlz.nxv16i16(<vscale x 16 x i16> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %11 = call <2 x i16> @llvm.ctlz.v2i16(<2 x i16> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %12 = call <4 x i16> @llvm.ctlz.v4i16(<4 x i16> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %13 = call <8 x i16> @llvm.ctlz.v8i16(<8 x i16> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %14 = call <16 x i16> @llvm.ctlz.v16i16(<16 x i16> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %15 = call <vscale x 2 x i16> @llvm.ctlz.nxv2i16(<vscale x 2 x i16> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %16 = call <vscale x 4 x i16> @llvm.ctlz.nxv4i16(<vscale x 4 x i16> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %17 = call <vscale x 8 x i16> @llvm.ctlz.nxv8i16(<vscale x 8 x i16> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %18 = call <vscale x 16 x i16> @llvm.ctlz.nxv16i16(<vscale x 16 x i16> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %19 = call i32 @llvm.ctlz.i32(i32 undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %20 = call <2 x i32> @llvm.ctlz.v2i32(<2 x i32> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %21 = call <4 x i32> @llvm.ctlz.v4i32(<4 x i32> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %22 = call <8 x i32> @llvm.ctlz.v8i32(<8 x i32> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 176 for instruction: %23 = call <16 x i32> @llvm.ctlz.v16i32(<16 x i32> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %24 = call <vscale x 2 x i32> @llvm.ctlz.nxv2i32(<vscale x 2 x i32> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %25 = call <vscale x 4 x i32> @llvm.ctlz.nxv4i32(<vscale x 4 x i32> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %26 = call <vscale x 8 x i32> @llvm.ctlz.nxv8i32(<vscale x 8 x i32> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %27 = call <vscale x 16 x i32> @llvm.ctlz.nxv16i32(<vscale x 16 x i32> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %20 = call <2 x i32> @llvm.ctlz.v2i32(<2 x i32> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %21 = call <4 x i32> @llvm.ctlz.v4i32(<4 x i32> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %22 = call <8 x i32> @llvm.ctlz.v8i32(<8 x i32> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %23 = call <16 x i32> @llvm.ctlz.v16i32(<16 x i32> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %24 = call <vscale x 2 x i32> @llvm.ctlz.nxv2i32(<vscale x 2 x i32> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %25 = call <vscale x 4 x i32> @llvm.ctlz.nxv4i32(<vscale x 4 x i32> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %26 = call <vscale x 8 x i32> @llvm.ctlz.nxv8i32(<vscale x 8 x i32> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %27 = call <vscale x 16 x i32> @llvm.ctlz.nxv16i32(<vscale x 16 x i32> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %28 = call i64 @llvm.ctlz.i64(i64 undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %29 = call <2 x i64> @llvm.ctlz.v2i64(<2 x i64> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %30 = call <4 x i64> @llvm.ctlz.v4i64(<4 x i64> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %31 = call <8 x i64> @llvm.ctlz.v8i64(<8 x i64> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 176 for instruction: %32 = call <16 x i64> @llvm.ctlz.v16i64(<16 x i64> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %33 = call <vscale x 2 x i64> @llvm.ctlz.nxv2i64(<vscale x 2 x i64> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %34 = call <vscale x 4 x i64> @llvm.ctlz.nxv4i64(<vscale x 4 x i64> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %35 = call <vscale x 8 x i64> @llvm.ctlz.nxv8i64(<vscale x 8 x i64> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %36 = call <vscale x 16 x i64> @llvm.ctlz.nxv16i64(<vscale x 16 x i64> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 35 for instruction: %29 = call <2 x i64> @llvm.ctlz.v2i64(<2 x i64> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 35 for instruction: %30 = call <4 x i64> @llvm.ctlz.v4i64(<4 x i64> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 35 for instruction: %31 = call <8 x i64> @llvm.ctlz.v8i64(<8 x i64> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 35 for instruction: %32 = call <16 x i64> @llvm.ctlz.v16i64(<16 x i64> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 35 for instruction: %33 = call <vscale x 2 x i64> @llvm.ctlz.nxv2i64(<vscale x 2 x i64> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 35 for instruction: %34 = call <vscale x 4 x i64> @llvm.ctlz.nxv4i64(<vscale x 4 x i64> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 35 for instruction: %35 = call <vscale x 8 x i64> @llvm.ctlz.nxv8i64(<vscale x 8 x i64> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 70 for instruction: %36 = call <vscale x 16 x i64> @llvm.ctlz.nxv16i64(<vscale x 16 x i64> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret void
;
call i8 @llvm.ctlz.i8(i8 undef)
@@ -302,41 +302,41 @@ define void @ctlz() {
define void @cttz() {
; CHECK-LABEL: 'cttz'
; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %1 = call i8 @llvm.cttz.i8(i8 undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %2 = call <2 x i8> @llvm.cttz.v2i8(<2 x i8> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %3 = call <4 x i8> @llvm.cttz.v4i8(<4 x i8> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %4 = call <8 x i8> @llvm.cttz.v8i8(<8 x i8> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 176 for instruction: %5 = call <16 x i8> @llvm.cttz.v16i8(<16 x i8> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %6 = call <vscale x 2 x i8> @llvm.cttz.nxv2i8(<vscale x 2 x i8> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %7 = call <vscale x 4 x i8> @llvm.cttz.nxv4i8(<vscale x 4 x i8> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %8 = call <vscale x 8 x i8> @llvm.cttz.nxv8i8(<vscale x 8 x i8> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %9 = call <vscale x 16 x i8> @llvm.cttz.nxv16i8(<vscale x 16 x i8> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %2 = call <2 x i8> @llvm.cttz.v2i8(<2 x i8> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %3 = call <4 x i8> @llvm.cttz.v4i8(<4 x i8> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %4 = call <8 x i8> @llvm.cttz.v8i8(<8 x i8> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %5 = call <16 x i8> @llvm.cttz.v16i8(<16 x i8> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %6 = call <vscale x 2 x i8> @llvm.cttz.nxv2i8(<vscale x 2 x i8> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %7 = call <vscale x 4 x i8> @llvm.cttz.nxv4i8(<vscale x 4 x i8> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %8 = call <vscale x 8 x i8> @llvm.cttz.nxv8i8(<vscale x 8 x i8> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %9 = call <vscale x 16 x i8> @llvm.cttz.nxv16i8(<vscale x 16 x i8> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %10 = call i16 @llvm.cttz.i16(i16 undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %11 = call <2 x i16> @llvm.cttz.v2i16(<2 x i16> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %12 = call <4 x i16> @llvm.cttz.v4i16(<4 x i16> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %13 = call <8 x i16> @llvm.cttz.v8i16(<8 x i16> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 176 for instruction: %14 = call <16 x i16> @llvm.cttz.v16i16(<16 x i16> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %15 = call <vscale x 2 x i16> @llvm.cttz.nxv2i16(<vscale x 2 x i16> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %16 = call <vscale x 4 x i16> @llvm.cttz.nxv4i16(<vscale x 4 x i16> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %17 = call <vscale x 8 x i16> @llvm.cttz.nxv8i16(<vscale x 8 x i16> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %18 = call <vscale x 16 x i16> @llvm.cttz.nxv16i16(<vscale x 16 x i16> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %11 = call <2 x i16> @llvm.cttz.v2i16(<2 x i16> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %12 = call <4 x i16> @llvm.cttz.v4i16(<4 x i16> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %13 = call <8 x i16> @llvm.cttz.v8i16(<8 x i16> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %14 = call <16 x i16> @llvm.cttz.v16i16(<16 x i16> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %15 = call <vscale x 2 x i16> @llvm.cttz.nxv2i16(<vscale x 2 x i16> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %16 = call <vscale x 4 x i16> @llvm.cttz.nxv4i16(<vscale x 4 x i16> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %17 = call <vscale x 8 x i16> @llvm.cttz.nxv8i16(<vscale x 8 x i16> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %18 = call <vscale x 16 x i16> @llvm.cttz.nxv16i16(<vscale x 16 x i16> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %19 = call i32 @llvm.cttz.i32(i32 undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %20 = call <2 x i32> @llvm.cttz.v2i32(<2 x i32> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %21 = call <4 x i32> @llvm.cttz.v4i32(<4 x i32> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %22 = call <8 x i32> @llvm.cttz.v8i32(<8 x i32> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 176 for instruction: %23 = call <16 x i32> @llvm.cttz.v16i32(<16 x i32> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %24 = call <vscale x 2 x i32> @llvm.cttz.nxv2i32(<vscale x 2 x i32> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %25 = call <vscale x 4 x i32> @llvm.cttz.nxv4i32(<vscale x 4 x i32> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %26 = call <vscale x 8 x i32> @llvm.cttz.nxv8i32(<vscale x 8 x i32> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %27 = call <vscale x 16 x i32> @llvm.cttz.nxv16i32(<vscale x 16 x i32> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %20 = call <2 x i32> @llvm.cttz.v2i32(<2 x i32> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %21 = call <4 x i32> @llvm.cttz.v4i32(<4 x i32> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %22 = call <8 x i32> @llvm.cttz.v8i32(<8 x i32> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %23 = call <16 x i32> @llvm.cttz.v16i32(<16 x i32> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %24 = call <vscale x 2 x i32> @llvm.cttz.nxv2i32(<vscale x 2 x i32> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %25 = call <vscale x 4 x i32> @llvm.cttz.nxv4i32(<vscale x 4 x i32> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %26 = call <vscale x 8 x i32> @llvm.cttz.nxv8i32(<vscale x 8 x i32> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %27 = call <vscale x 16 x i32> @llvm.cttz.nxv16i32(<vscale x 16 x i32> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %28 = call i64 @llvm.cttz.i64(i64 undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %29 = call <2 x i64> @llvm.cttz.v2i64(<2 x i64> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %30 = call <4 x i64> @llvm.cttz.v4i64(<4 x i64> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %31 = call <8 x i64> @llvm.cttz.v8i64(<8 x i64> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 176 for instruction: %32 = call <16 x i64> @llvm.cttz.v16i64(<16 x i64> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %33 = call <vscale x 2 x i64> @llvm.cttz.nxv2i64(<vscale x 2 x i64> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %34 = call <vscale x 4 x i64> @llvm.cttz.nxv4i64(<vscale x 4 x i64> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %35 = call <vscale x 8 x i64> @llvm.cttz.nxv8i64(<vscale x 8 x i64> undef, i1 false)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %36 = call <vscale x 16 x i64> @llvm.cttz.nxv16i64(<vscale x 16 x i64> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 25 for instruction: %29 = call <2 x i64> @llvm.cttz.v2i64(<2 x i64> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 25 for instruction: %30 = call <4 x i64> @llvm.cttz.v4i64(<4 x i64> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 25 for instruction: %31 = call <8 x i64> @llvm.cttz.v8i64(<8 x i64> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 25 for instruction: %32 = call <16 x i64> @llvm.cttz.v16i64(<16 x i64> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 25 for instruction: %33 = call <vscale x 2 x i64> @llvm.cttz.nxv2i64(<vscale x 2 x i64> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 25 for instruction: %34 = call <vscale x 4 x i64> @llvm.cttz.nxv4i64(<vscale x 4 x i64> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 25 for instruction: %35 = call <vscale x 8 x i64> @llvm.cttz.nxv8i64(<vscale x 8 x i64> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 50 for instruction: %36 = call <vscale x 16 x i64> @llvm.cttz.nxv16i64(<vscale x 16 x i64> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret void
;
call i8 @llvm.cttz.i8(i8 undef)
More information about the llvm-commits
mailing list