Skip to content

Commit

Permalink
[InstCombine] Canonicalize constant GEPs to i8 source element type (#…
Browse files Browse the repository at this point in the history 
…68882)

This patch canonicalizes getelementptr instructions with constant
indices to use the `i8` source element type. This makes it easier for
optimizations to recognize that two GEPs are identical, because they
don't need to see past many different ways to express the same offset.

This is a first step towards
https://discourse.llvm.org/t/rfc-replacing-getelementptr-with-ptradd/68699.
This is limited to constant GEPs only for now, as they have a clear
canonical form, while we're not yet sure how exactly to deal with
variable indices.

The test llvm/test/Transforms/PhaseOrdering/switch_with_geps.ll gives
two representative examples of the kind of optimization improvement we
expect from this change. In the first test SimplifyCFG can now realize
that all switch branches are actually the same. In the second test it
can convert it into simple arithmetic. These are representative of
common optimization failures we see in Rust.

Fixes #69841.
  • Loading branch information
@nikic
nikic authored Jan 24, 2024
1 parent 98509c7 commit 90ba330
Show file tree
Hide file tree
Showing 176 changed files with 2,520 additions and 2,532 deletions.
10 changes: 5 additions & 5 deletions clang/test/CodeGen/PowerPC/builtins-ppc-pair-mma.c
View file
Original file line number Diff line number Diff line change
Expand Up @@ -25,13 +25,13 @@ void test1(unsigned char *vqp, unsigned char *vpp, vector unsigned char vc, unsi
// CHECK-NEXT: [[TMP2:%.*]] = extractvalue { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> } [[TMP1]], 0
// CHECK-NEXT: store <16 x i8> [[TMP2]], ptr [[RESP:%.*]], align 16
// CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> } [[TMP1]], 1
// CHECK-NEXT: [[TMP4:%.*]] = getelementptr inbounds <16 x i8>, ptr [[RESP]], i64 1
// CHECK-NEXT: [[TMP4:%.*]] = getelementptr inbounds i8, ptr [[RESP]], i64 16
// CHECK-NEXT: store <16 x i8> [[TMP3]], ptr [[TMP4]], align 16
// CHECK-NEXT: [[TMP5:%.*]] = extractvalue { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> } [[TMP1]], 2
// CHECK-NEXT: [[TMP6:%.*]] = getelementptr inbounds <16 x i8>, ptr [[RESP]], i64 2
// CHECK-NEXT: [[TMP6:%.*]] = getelementptr inbounds i8, ptr [[RESP]], i64 32
// CHECK-NEXT: store <16 x i8> [[TMP5]], ptr [[TMP6]], align 16
// CHECK-NEXT: [[TMP7:%.*]] = extractvalue { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> } [[TMP1]], 3
// CHECK-NEXT: [[TMP8:%.*]] = getelementptr inbounds <16 x i8>, ptr [[RESP]], i64 3
// CHECK-NEXT: [[TMP8:%.*]] = getelementptr inbounds i8, ptr [[RESP]], i64 48
// CHECK-NEXT: store <16 x i8> [[TMP7]], ptr [[TMP8]], align 16
// CHECK-NEXT: ret void
//
Expand Down Expand Up @@ -60,7 +60,7 @@ void test3(unsigned char *vqp, unsigned char *vpp, vector unsigned char vc, unsi
// CHECK-NEXT: [[TMP2:%.*]] = extractvalue { <16 x i8>, <16 x i8> } [[TMP1]], 0
// CHECK-NEXT: store <16 x i8> [[TMP2]], ptr [[RESP:%.*]], align 16
// CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <16 x i8>, <16 x i8> } [[TMP1]], 1
// CHECK-NEXT: [[TMP4:%.*]] = getelementptr inbounds <16 x i8>, ptr [[RESP]], i64 1
// CHECK-NEXT: [[TMP4:%.*]] = getelementptr inbounds i8, ptr [[RESP]], i64 16
// CHECK-NEXT: store <16 x i8> [[TMP3]], ptr [[TMP4]], align 16
// CHECK-NEXT: ret void
//
Expand Down Expand Up @@ -1072,7 +1072,7 @@ void test76(unsigned char *vqp, unsigned char *vpp, vector unsigned char vc, uns
// CHECK-NEXT: [[TMP2:%.*]] = extractvalue { <16 x i8>, <16 x i8> } [[TMP1]], 0
// CHECK-NEXT: store <16 x i8> [[TMP2]], ptr [[RESP:%.*]], align 16
// CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <16 x i8>, <16 x i8> } [[TMP1]], 1
// CHECK-NEXT: [[TMP4:%.*]] = getelementptr inbounds <16 x i8>, ptr [[RESP]], i64 1
// CHECK-NEXT: [[TMP4:%.*]] = getelementptr inbounds i8, ptr [[RESP]], i64 16
// CHECK-NEXT: store <16 x i8> [[TMP3]], ptr [[TMP4]], align 16
// CHECK-NEXT: ret void
//
Expand Down
18 changes: 9 additions & 9 deletions clang/test/CodeGen/aarch64-ls64-inline-asm.c
View file
Original file line number Diff line number Diff line change
Expand Up @@ -16,8 +16,8 @@ void load(struct foo *output, void *addr)

// CHECK-LABEL: @store(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP1:%.*]] = load i512, ptr [[INPUT:%.*]], align 8
// CHECK-NEXT: tail call void asm sideeffect "st64b $0,[$1]", "r,r,~{memory}"(i512 [[TMP1]], ptr [[ADDR:%.*]]) #[[ATTR1]], !srcloc !3
// CHECK-NEXT: [[TMP0:%.*]] = load i512, ptr [[INPUT:%.*]], align 8
// CHECK-NEXT: tail call void asm sideeffect "st64b $0,[$1]", "r,r,~{memory}"(i512 [[TMP0]], ptr [[ADDR:%.*]]) #[[ATTR1]], !srcloc !3
// CHECK-NEXT: ret void
//
void store(const struct foo *input, void *addr)
Expand All @@ -29,25 +29,25 @@ void store(const struct foo *input, void *addr)
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP0:%.*]] = load i32, ptr [[IN:%.*]], align 4, !tbaa [[TBAA4:![0-9]+]]
// CHECK-NEXT: [[CONV:%.*]] = sext i32 [[TMP0]] to i64
// CHECK-NEXT: [[ARRAYIDX1:%.*]] = getelementptr inbounds i32, ptr [[IN]], i64 1
// CHECK-NEXT: [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[IN]], i64 4
// CHECK-NEXT: [[TMP1:%.*]] = load i32, ptr [[ARRAYIDX1]], align 4, !tbaa [[TBAA4]]
// CHECK-NEXT: [[CONV2:%.*]] = sext i32 [[TMP1]] to i64
// CHECK-NEXT: [[ARRAYIDX4:%.*]] = getelementptr inbounds i32, ptr [[IN]], i64 4
// CHECK-NEXT: [[ARRAYIDX4:%.*]] = getelementptr inbounds i8, ptr [[IN]], i64 16
// CHECK-NEXT: [[TMP2:%.*]] = load i32, ptr [[ARRAYIDX4]], align 4, !tbaa [[TBAA4]]
// CHECK-NEXT: [[CONV5:%.*]] = sext i32 [[TMP2]] to i64
// CHECK-NEXT: [[ARRAYIDX7:%.*]] = getelementptr inbounds i32, ptr [[IN]], i64 16
// CHECK-NEXT: [[ARRAYIDX7:%.*]] = getelementptr inbounds i8, ptr [[IN]], i64 64
// CHECK-NEXT: [[TMP3:%.*]] = load i32, ptr [[ARRAYIDX7]], align 4, !tbaa [[TBAA4]]
// CHECK-NEXT: [[CONV8:%.*]] = sext i32 [[TMP3]] to i64
// CHECK-NEXT: [[ARRAYIDX10:%.*]] = getelementptr inbounds i32, ptr [[IN]], i64 25
// CHECK-NEXT: [[ARRAYIDX10:%.*]] = getelementptr inbounds i8, ptr [[IN]], i64 100
// CHECK-NEXT: [[TMP4:%.*]] = load i32, ptr [[ARRAYIDX10]], align 4, !tbaa [[TBAA4]]
// CHECK-NEXT: [[CONV11:%.*]] = sext i32 [[TMP4]] to i64
// CHECK-NEXT: [[ARRAYIDX13:%.*]] = getelementptr inbounds i32, ptr [[IN]], i64 36
// CHECK-NEXT: [[ARRAYIDX13:%.*]] = getelementptr inbounds i8, ptr [[IN]], i64 144
// CHECK-NEXT: [[TMP5:%.*]] = load i32, ptr [[ARRAYIDX13]], align 4, !tbaa [[TBAA4]]
// CHECK-NEXT: [[CONV14:%.*]] = sext i32 [[TMP5]] to i64
// CHECK-NEXT: [[ARRAYIDX16:%.*]] = getelementptr inbounds i32, ptr [[IN]], i64 49
// CHECK-NEXT: [[ARRAYIDX16:%.*]] = getelementptr inbounds i8, ptr [[IN]], i64 196
// CHECK-NEXT: [[TMP6:%.*]] = load i32, ptr [[ARRAYIDX16]], align 4, !tbaa [[TBAA4]]
// CHECK-NEXT: [[CONV17:%.*]] = sext i32 [[TMP6]] to i64
// CHECK-NEXT: [[ARRAYIDX19:%.*]] = getelementptr inbounds i32, ptr [[IN]], i64 64
// CHECK-NEXT: [[ARRAYIDX19:%.*]] = getelementptr inbounds i8, ptr [[IN]], i64 256
// CHECK-NEXT: [[TMP7:%.*]] = load i32, ptr [[ARRAYIDX19]], align 4, !tbaa [[TBAA4]]
// CHECK-NEXT: [[CONV20:%.*]] = sext i32 [[TMP7]] to i64
// CHECK-NEXT: [[S_SROA_10_0_INSERT_EXT:%.*]] = zext i64 [[CONV20]] to i512
Expand Down
48 changes: 24 additions & 24 deletions clang/test/CodeGen/attr-arm-sve-vector-bits-bitcast.c
View file
Original file line number Diff line number Diff line change
Expand Up @@ -30,21 +30,21 @@ DEFINE_STRUCT(bool)

// CHECK-128-LABEL: @read_int64(
// CHECK-128-NEXT: entry:
// CHECK-128-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_INT64:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-128-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 16
// CHECK-128-NEXT: [[TMP0:%.*]] = load <2 x i64>, ptr [[Y]], align 16, !tbaa [[TBAA2:![0-9]+]]
// CHECK-128-NEXT: [[CAST_SCALABLE:%.*]] = tail call <vscale x 2 x i64> @llvm.vector.insert.nxv2i64.v2i64(<vscale x 2 x i64> undef, <2 x i64> [[TMP0]], i64 0)
// CHECK-128-NEXT: ret <vscale x 2 x i64> [[CAST_SCALABLE]]
//
// CHECK-256-LABEL: @read_int64(
// CHECK-256-NEXT: entry:
// CHECK-256-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_INT64:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-256-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 32
// CHECK-256-NEXT: [[TMP0:%.*]] = load <4 x i64>, ptr [[Y]], align 16, !tbaa [[TBAA2:![0-9]+]]
// CHECK-256-NEXT: [[CAST_SCALABLE:%.*]] = tail call <vscale x 2 x i64> @llvm.vector.insert.nxv2i64.v4i64(<vscale x 2 x i64> undef, <4 x i64> [[TMP0]], i64 0)
// CHECK-256-NEXT: ret <vscale x 2 x i64> [[CAST_SCALABLE]]
//
// CHECK-512-LABEL: @read_int64(
// CHECK-512-NEXT: entry:
// CHECK-512-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_INT64:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-512-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 64
// CHECK-512-NEXT: [[TMP0:%.*]] = load <8 x i64>, ptr [[Y]], align 16, !tbaa [[TBAA2:![0-9]+]]
// CHECK-512-NEXT: [[CAST_SCALABLE:%.*]] = tail call <vscale x 2 x i64> @llvm.vector.insert.nxv2i64.v8i64(<vscale x 2 x i64> undef, <8 x i64> [[TMP0]], i64 0)
// CHECK-512-NEXT: ret <vscale x 2 x i64> [[CAST_SCALABLE]]
Expand All @@ -56,21 +56,21 @@ svint64_t read_int64(struct struct_int64 *s) {
// CHECK-128-LABEL: @write_int64(
// CHECK-128-NEXT: entry:
// CHECK-128-NEXT: [[CAST_FIXED:%.*]] = tail call <2 x i64> @llvm.vector.extract.v2i64.nxv2i64(<vscale x 2 x i64> [[X:%.*]], i64 0)
// CHECK-128-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_INT64:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-128-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 16
// CHECK-128-NEXT: store <2 x i64> [[CAST_FIXED]], ptr [[Y]], align 16, !tbaa [[TBAA2]]
// CHECK-128-NEXT: ret void
//
// CHECK-256-LABEL: @write_int64(
// CHECK-256-NEXT: entry:
// CHECK-256-NEXT: [[CAST_FIXED:%.*]] = tail call <4 x i64> @llvm.vector.extract.v4i64.nxv2i64(<vscale x 2 x i64> [[X:%.*]], i64 0)
// CHECK-256-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_INT64:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-256-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 32
// CHECK-256-NEXT: store <4 x i64> [[CAST_FIXED]], ptr [[Y]], align 16, !tbaa [[TBAA2]]
// CHECK-256-NEXT: ret void
//
// CHECK-512-LABEL: @write_int64(
// CHECK-512-NEXT: entry:
// CHECK-512-NEXT: [[CAST_FIXED:%.*]] = tail call <8 x i64> @llvm.vector.extract.v8i64.nxv2i64(<vscale x 2 x i64> [[X:%.*]], i64 0)
// CHECK-512-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_INT64:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-512-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 64
// CHECK-512-NEXT: store <8 x i64> [[CAST_FIXED]], ptr [[Y]], align 16, !tbaa [[TBAA2]]
// CHECK-512-NEXT: ret void
//
Expand All @@ -84,21 +84,21 @@ void write_int64(struct struct_int64 *s, svint64_t x) {

// CHECK-128-LABEL: @read_float64(
// CHECK-128-NEXT: entry:
// CHECK-128-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_FLOAT64:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-128-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 16
// CHECK-128-NEXT: [[TMP0:%.*]] = load <2 x double>, ptr [[Y]], align 16, !tbaa [[TBAA2]]
// CHECK-128-NEXT: [[CAST_SCALABLE:%.*]] = tail call <vscale x 2 x double> @llvm.vector.insert.nxv2f64.v2f64(<vscale x 2 x double> undef, <2 x double> [[TMP0]], i64 0)
// CHECK-128-NEXT: ret <vscale x 2 x double> [[CAST_SCALABLE]]
//
// CHECK-256-LABEL: @read_float64(
// CHECK-256-NEXT: entry:
// CHECK-256-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_FLOAT64:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-256-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 32
// CHECK-256-NEXT: [[TMP0:%.*]] = load <4 x double>, ptr [[Y]], align 16, !tbaa [[TBAA2]]
// CHECK-256-NEXT: [[CAST_SCALABLE:%.*]] = tail call <vscale x 2 x double> @llvm.vector.insert.nxv2f64.v4f64(<vscale x 2 x double> undef, <4 x double> [[TMP0]], i64 0)
// CHECK-256-NEXT: ret <vscale x 2 x double> [[CAST_SCALABLE]]
//
// CHECK-512-LABEL: @read_float64(
// CHECK-512-NEXT: entry:
// CHECK-512-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_FLOAT64:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-512-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 64
// CHECK-512-NEXT: [[TMP0:%.*]] = load <8 x double>, ptr [[Y]], align 16, !tbaa [[TBAA2]]
// CHECK-512-NEXT: [[CAST_SCALABLE:%.*]] = tail call <vscale x 2 x double> @llvm.vector.insert.nxv2f64.v8f64(<vscale x 2 x double> undef, <8 x double> [[TMP0]], i64 0)
// CHECK-512-NEXT: ret <vscale x 2 x double> [[CAST_SCALABLE]]
Expand All @@ -110,21 +110,21 @@ svfloat64_t read_float64(struct struct_float64 *s) {
// CHECK-128-LABEL: @write_float64(
// CHECK-128-NEXT: entry:
// CHECK-128-NEXT: [[CAST_FIXED:%.*]] = tail call <2 x double> @llvm.vector.extract.v2f64.nxv2f64(<vscale x 2 x double> [[X:%.*]], i64 0)
// CHECK-128-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_FLOAT64:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-128-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 16
// CHECK-128-NEXT: store <2 x double> [[CAST_FIXED]], ptr [[Y]], align 16, !tbaa [[TBAA2]]
// CHECK-128-NEXT: ret void
//
// CHECK-256-LABEL: @write_float64(
// CHECK-256-NEXT: entry:
// CHECK-256-NEXT: [[CAST_FIXED:%.*]] = tail call <4 x double> @llvm.vector.extract.v4f64.nxv2f64(<vscale x 2 x double> [[X:%.*]], i64 0)
// CHECK-256-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_FLOAT64:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-256-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 32
// CHECK-256-NEXT: store <4 x double> [[CAST_FIXED]], ptr [[Y]], align 16, !tbaa [[TBAA2]]
// CHECK-256-NEXT: ret void
//
// CHECK-512-LABEL: @write_float64(
// CHECK-512-NEXT: entry:
// CHECK-512-NEXT: [[CAST_FIXED:%.*]] = tail call <8 x double> @llvm.vector.extract.v8f64.nxv2f64(<vscale x 2 x double> [[X:%.*]], i64 0)
// CHECK-512-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_FLOAT64:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-512-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 64
// CHECK-512-NEXT: store <8 x double> [[CAST_FIXED]], ptr [[Y]], align 16, !tbaa [[TBAA2]]
// CHECK-512-NEXT: ret void
//
Expand All @@ -138,21 +138,21 @@ void write_float64(struct struct_float64 *s, svfloat64_t x) {

// CHECK-128-LABEL: @read_bfloat16(
// CHECK-128-NEXT: entry:
// CHECK-128-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_BFLOAT16:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-128-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 16
// CHECK-128-NEXT: [[TMP0:%.*]] = load <8 x bfloat>, ptr [[Y]], align 16, !tbaa [[TBAA2]]
// CHECK-128-NEXT: [[CAST_SCALABLE:%.*]] = tail call <vscale x 8 x bfloat> @llvm.vector.insert.nxv8bf16.v8bf16(<vscale x 8 x bfloat> undef, <8 x bfloat> [[TMP0]], i64 0)
// CHECK-128-NEXT: ret <vscale x 8 x bfloat> [[CAST_SCALABLE]]
//
// CHECK-256-LABEL: @read_bfloat16(
// CHECK-256-NEXT: entry:
// CHECK-256-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_BFLOAT16:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-256-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 32
// CHECK-256-NEXT: [[TMP0:%.*]] = load <16 x bfloat>, ptr [[Y]], align 16, !tbaa [[TBAA2]]
// CHECK-256-NEXT: [[CAST_SCALABLE:%.*]] = tail call <vscale x 8 x bfloat> @llvm.vector.insert.nxv8bf16.v16bf16(<vscale x 8 x bfloat> undef, <16 x bfloat> [[TMP0]], i64 0)
// CHECK-256-NEXT: ret <vscale x 8 x bfloat> [[CAST_SCALABLE]]
//
// CHECK-512-LABEL: @read_bfloat16(
// CHECK-512-NEXT: entry:
// CHECK-512-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_BFLOAT16:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-512-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 64
// CHECK-512-NEXT: [[TMP0:%.*]] = load <32 x bfloat>, ptr [[Y]], align 16, !tbaa [[TBAA2]]
// CHECK-512-NEXT: [[CAST_SCALABLE:%.*]] = tail call <vscale x 8 x bfloat> @llvm.vector.insert.nxv8bf16.v32bf16(<vscale x 8 x bfloat> undef, <32 x bfloat> [[TMP0]], i64 0)
// CHECK-512-NEXT: ret <vscale x 8 x bfloat> [[CAST_SCALABLE]]
Expand All @@ -164,21 +164,21 @@ svbfloat16_t read_bfloat16(struct struct_bfloat16 *s) {
// CHECK-128-LABEL: @write_bfloat16(
// CHECK-128-NEXT: entry:
// CHECK-128-NEXT: [[CAST_FIXED:%.*]] = tail call <8 x bfloat> @llvm.vector.extract.v8bf16.nxv8bf16(<vscale x 8 x bfloat> [[X:%.*]], i64 0)
// CHECK-128-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_BFLOAT16:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-128-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 16
// CHECK-128-NEXT: store <8 x bfloat> [[CAST_FIXED]], ptr [[Y]], align 16, !tbaa [[TBAA2]]
// CHECK-128-NEXT: ret void
//
// CHECK-256-LABEL: @write_bfloat16(
// CHECK-256-NEXT: entry:
// CHECK-256-NEXT: [[CAST_FIXED:%.*]] = tail call <16 x bfloat> @llvm.vector.extract.v16bf16.nxv8bf16(<vscale x 8 x bfloat> [[X:%.*]], i64 0)
// CHECK-256-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_BFLOAT16:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-256-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 32
// CHECK-256-NEXT: store <16 x bfloat> [[CAST_FIXED]], ptr [[Y]], align 16, !tbaa [[TBAA2]]
// CHECK-256-NEXT: ret void
//
// CHECK-512-LABEL: @write_bfloat16(
// CHECK-512-NEXT: entry:
// CHECK-512-NEXT: [[CAST_FIXED:%.*]] = tail call <32 x bfloat> @llvm.vector.extract.v32bf16.nxv8bf16(<vscale x 8 x bfloat> [[X:%.*]], i64 0)
// CHECK-512-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_BFLOAT16:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-512-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 64
// CHECK-512-NEXT: store <32 x bfloat> [[CAST_FIXED]], ptr [[Y]], align 16, !tbaa [[TBAA2]]
// CHECK-512-NEXT: ret void
//
Expand All @@ -192,23 +192,23 @@ void write_bfloat16(struct struct_bfloat16 *s, svbfloat16_t x) {

// CHECK-128-LABEL: @read_bool(
// CHECK-128-NEXT: entry:
// CHECK-128-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_BOOL:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-128-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 2
// CHECK-128-NEXT: [[TMP0:%.*]] = load <2 x i8>, ptr [[Y]], align 2, !tbaa [[TBAA2]]
// CHECK-128-NEXT: [[CAST_SCALABLE:%.*]] = tail call <vscale x 2 x i8> @llvm.vector.insert.nxv2i8.v2i8(<vscale x 2 x i8> undef, <2 x i8> [[TMP0]], i64 0)
// CHECK-128-NEXT: [[TMP1:%.*]] = bitcast <vscale x 2 x i8> [[CAST_SCALABLE]] to <vscale x 16 x i1>
// CHECK-128-NEXT: ret <vscale x 16 x i1> [[TMP1]]
//
// CHECK-256-LABEL: @read_bool(
// CHECK-256-NEXT: entry:
// CHECK-256-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_BOOL:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-256-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 4
// CHECK-256-NEXT: [[TMP0:%.*]] = load <4 x i8>, ptr [[Y]], align 2, !tbaa [[TBAA2]]
// CHECK-256-NEXT: [[CAST_SCALABLE:%.*]] = tail call <vscale x 2 x i8> @llvm.vector.insert.nxv2i8.v4i8(<vscale x 2 x i8> undef, <4 x i8> [[TMP0]], i64 0)
// CHECK-256-NEXT: [[TMP1:%.*]] = bitcast <vscale x 2 x i8> [[CAST_SCALABLE]] to <vscale x 16 x i1>
// CHECK-256-NEXT: ret <vscale x 16 x i1> [[TMP1]]
//
// CHECK-512-LABEL: @read_bool(
// CHECK-512-NEXT: entry:
// CHECK-512-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_BOOL:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-512-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 8
// CHECK-512-NEXT: [[TMP0:%.*]] = load <8 x i8>, ptr [[Y]], align 2, !tbaa [[TBAA2]]
// CHECK-512-NEXT: [[CAST_SCALABLE:%.*]] = tail call <vscale x 2 x i8> @llvm.vector.insert.nxv2i8.v8i8(<vscale x 2 x i8> undef, <8 x i8> [[TMP0]], i64 0)
// CHECK-512-NEXT: [[TMP1:%.*]] = bitcast <vscale x 2 x i8> [[CAST_SCALABLE]] to <vscale x 16 x i1>
Expand All @@ -222,23 +222,23 @@ svbool_t read_bool(struct struct_bool *s) {
// CHECK-128-NEXT: entry:
// CHECK-128-NEXT: [[TMP0:%.*]] = bitcast <vscale x 16 x i1> [[X:%.*]] to <vscale x 2 x i8>
// CHECK-128-NEXT: [[CAST_FIXED:%.*]] = tail call <2 x i8> @llvm.vector.extract.v2i8.nxv2i8(<vscale x 2 x i8> [[TMP0]], i64 0)
// CHECK-128-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_BOOL:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-128-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 2
// CHECK-128-NEXT: store <2 x i8> [[CAST_FIXED]], ptr [[Y]], align 2, !tbaa [[TBAA2]]
// CHECK-128-NEXT: ret void
//
// CHECK-256-LABEL: @write_bool(
// CHECK-256-NEXT: entry:
// CHECK-256-NEXT: [[TMP0:%.*]] = bitcast <vscale x 16 x i1> [[X:%.*]] to <vscale x 2 x i8>
// CHECK-256-NEXT: [[CAST_FIXED:%.*]] = tail call <4 x i8> @llvm.vector.extract.v4i8.nxv2i8(<vscale x 2 x i8> [[TMP0]], i64 0)
// CHECK-256-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_BOOL:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-256-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 4
// CHECK-256-NEXT: store <4 x i8> [[CAST_FIXED]], ptr [[Y]], align 2, !tbaa [[TBAA2]]
// CHECK-256-NEXT: ret void
//
// CHECK-512-LABEL: @write_bool(
// CHECK-512-NEXT: entry:
// CHECK-512-NEXT: [[TMP0:%.*]] = bitcast <vscale x 16 x i1> [[X:%.*]] to <vscale x 2 x i8>
// CHECK-512-NEXT: [[CAST_FIXED:%.*]] = tail call <8 x i8> @llvm.vector.extract.v8i8.nxv2i8(<vscale x 2 x i8> [[TMP0]], i64 0)
// CHECK-512-NEXT: [[Y:%.*]] = getelementptr inbounds [[STRUCT_STRUCT_BOOL:%.*]], ptr [[S:%.*]], i64 0, i32 1
// CHECK-512-NEXT: [[Y:%.*]] = getelementptr inbounds i8, ptr [[S:%.*]], i64 8
// CHECK-512-NEXT: store <8 x i8> [[CAST_FIXED]], ptr [[Y]], align 2, !tbaa [[TBAA2]]
// CHECK-512-NEXT: ret void
//
Expand Down
Loading

0 comments on commit 90ba330

Please sign in to comment.