shylie/llvm-project
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[X86] Rename AVX512 VEXTRACT/INSERT??x? to VEXTRACT/INSERT??X? (#116826)

Browse Source 
Use uppercase in the subvector description ("32x2" -> "32X4" etc.) - matches what we already do in VBROADCAST??X?, and we try to use uppercase for all x86 instruction mnemonics anyway (and lowercase just for the arg description suffix).
This commit is contained in:
Simon Pilgrim 2024-11-20 08:25:01 +00:00 committed by GitHub
parent 0394e08bfb
commit 3a5cf6d99b
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
19 changed files with 272 additions and 272 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
llvm/lib/Target/X86/GISel/X86InstructionSelector.cpp
View File

@ -1254,16 +1254,16 @@ bool X86InstructionSelector::selectExtract(MachineInstr &I,
if (SrcTy.getSizeInBits() == 256 && DstTy.getSizeInBits() == 128) {
if (HasVLX)
I.setDesc(TII.get(X86::VEXTRACTF32x4Z256rri));
I.setDesc(TII.get(X86::VEXTRACTF32X4Z256rri));
else if (HasAVX)
I.setDesc(TII.get(X86::VEXTRACTF128rri));
else
return false;
} else if (SrcTy.getSizeInBits() == 512 && HasAVX512) {
if (DstTy.getSizeInBits() == 128)
I.setDesc(TII.get(X86::VEXTRACTF32x4Zrri));
I.setDesc(TII.get(X86::VEXTRACTF32X4Zrri));
else if (DstTy.getSizeInBits() == 256)
I.setDesc(TII.get(X86::VEXTRACTF64x4Zrri));
I.setDesc(TII.get(X86::VEXTRACTF64X4Zrri));
else
return false;
} else
@ -1387,16 +1387,16 @@ bool X86InstructionSelector::selectInsert(MachineInstr &I,
if (DstTy.getSizeInBits() == 256 && InsertRegTy.getSizeInBits() == 128) {
if (HasVLX)
I.setDesc(TII.get(X86::VINSERTF32x4Z256rri));
I.setDesc(TII.get(X86::VINSERTF32X4Z256rri));
else if (HasAVX)
I.setDesc(TII.get(X86::VINSERTF128rri));
else
return false;
} else if (DstTy.getSizeInBits() == 512 && HasAVX512) {
if (InsertRegTy.getSizeInBits() == 128)
I.setDesc(TII.get(X86::VINSERTF32x4Zrri));
I.setDesc(TII.get(X86::VINSERTF32X4Zrri));
else if (InsertRegTy.getSizeInBits() == 256)
I.setDesc(TII.get(X86::VINSERTF64x4Zrri));
I.setDesc(TII.get(X86::VINSERTF64X4Zrri));
else
return false;
} else

206
llvm/lib/Target/X86/X86InstrAVX512.td
View File

@ -427,24 +427,24 @@ multiclass vinsert_for_type<ValueType EltVT32, int Opcode128,
X86FoldableSchedWrite sched> {
let Predicates = [HasVLX] in
defm NAME # "32x4Z256" : vinsert_for_size<Opcode128,
defm NAME # "32X4Z256" : vinsert_for_size<Opcode128,
X86VectorVTInfo< 4, EltVT32, VR128X>,
X86VectorVTInfo< 8, EltVT32, VR256X>,
vinsert128_insert, sched>, EVEX_V256;
defm NAME # "32x4Z" : vinsert_for_size<Opcode128,
defm NAME # "32X4Z" : vinsert_for_size<Opcode128,
X86VectorVTInfo< 4, EltVT32, VR128X>,
X86VectorVTInfo<16, EltVT32, VR512>,
vinsert128_insert, sched>, EVEX_V512;
defm NAME # "64x4Z" : vinsert_for_size<Opcode256,
defm NAME # "64X4Z" : vinsert_for_size<Opcode256,
X86VectorVTInfo< 4, EltVT64, VR256X>,
X86VectorVTInfo< 8, EltVT64, VR512>,
vinsert256_insert, sched>, REX_W, EVEX_V512;
// Even with DQI we'd like to only use these instructions for masking.
let Predicates = [HasVLX, HasDQI] in
defm NAME # "64x2Z256" : vinsert_for_size_split<Opcode128,
defm NAME # "64X2Z256" : vinsert_for_size_split<Opcode128,
X86VectorVTInfo< 2, EltVT64, VR128X>,
X86VectorVTInfo< 4, EltVT64, VR256X>,
null_frag, vinsert128_insert, sched>,
@ -452,13 +452,13 @@ multiclass vinsert_for_type<ValueType EltVT32, int Opcode128,
// Even with DQI we'd like to only use these instructions for masking.
let Predicates = [HasDQI] in {
defm NAME # "64x2Z" : vinsert_for_size_split<Opcode128,
defm NAME # "64X2Z" : vinsert_for_size_split<Opcode128,
X86VectorVTInfo< 2, EltVT64, VR128X>,
X86VectorVTInfo< 8, EltVT64, VR512>,
null_frag, vinsert128_insert, sched>,
REX_W, EVEX_V512;
defm NAME # "32x8Z" : vinsert_for_size_split<Opcode256,
defm NAME # "32X8Z" : vinsert_for_size_split<Opcode256,
X86VectorVTInfo< 8, EltVT32, VR256X>,
X86VectorVTInfo<16, EltVT32, VR512>,
null_frag, vinsert256_insert, sched>,
@ -472,47 +472,47 @@ defm VINSERTI : vinsert_for_type<i32, 0x38, i64, 0x3a, WriteShuffle256>;
// Codegen pattern with the alternative types,
// Even with AVX512DQ we'll still use these for unmasked operations.
defm : vinsert_for_size_lowering<"VINSERTF32x4Z256", v2f64x_info, v4f64x_info,
defm : vinsert_for_size_lowering<"VINSERTF32X4Z256", v2f64x_info, v4f64x_info,
vinsert128_insert, INSERT_get_vinsert128_imm, [HasVLX]>;
defm : vinsert_for_size_lowering<"VINSERTI32x4Z256", v2i64x_info, v4i64x_info,
defm : vinsert_for_size_lowering<"VINSERTI32X4Z256", v2i64x_info, v4i64x_info,
vinsert128_insert, INSERT_get_vinsert128_imm, [HasVLX]>;
defm : vinsert_for_size_lowering<"VINSERTF32x4Z", v2f64x_info, v8f64_info,
defm : vinsert_for_size_lowering<"VINSERTF32X4Z", v2f64x_info, v8f64_info,
vinsert128_insert, INSERT_get_vinsert128_imm, [HasAVX512]>;
defm : vinsert_for_size_lowering<"VINSERTI32x4Z", v2i64x_info, v8i64_info,
defm : vinsert_for_size_lowering<"VINSERTI32X4Z", v2i64x_info, v8i64_info,
vinsert128_insert, INSERT_get_vinsert128_imm, [HasAVX512]>;
defm : vinsert_for_size_lowering<"VINSERTF64x4Z", v8f32x_info, v16f32_info,
defm : vinsert_for_size_lowering<"VINSERTF64X4Z", v8f32x_info, v16f32_info,
vinsert256_insert, INSERT_get_vinsert256_imm, [HasAVX512]>;
defm : vinsert_for_size_lowering<"VINSERTI64x4Z", v8i32x_info, v16i32_info,
defm : vinsert_for_size_lowering<"VINSERTI64X4Z", v8i32x_info, v16i32_info,
vinsert256_insert, INSERT_get_vinsert256_imm, [HasAVX512]>;
// Codegen pattern with the alternative types insert VEC128 into VEC256
defm : vinsert_for_size_lowering<"VINSERTI32x4Z256", v8i16x_info, v16i16x_info,
defm : vinsert_for_size_lowering<"VINSERTI32X4Z256", v8i16x_info, v16i16x_info,
vinsert128_insert, INSERT_get_vinsert128_imm, [HasVLX]>;
defm : vinsert_for_size_lowering<"VINSERTI32x4Z256", v16i8x_info, v32i8x_info,
defm : vinsert_for_size_lowering<"VINSERTI32X4Z256", v16i8x_info, v32i8x_info,
vinsert128_insert, INSERT_get_vinsert128_imm, [HasVLX]>;
defm : vinsert_for_size_lowering<"VINSERTF32x4Z256", v8f16x_info, v16f16x_info,
defm : vinsert_for_size_lowering<"VINSERTF32X4Z256", v8f16x_info, v16f16x_info,
vinsert128_insert, INSERT_get_vinsert128_imm, [HasVLX]>;
defm : vinsert_for_size_lowering<"VINSERTF32x4Z256", v8bf16x_info, v16bf16x_info,
defm : vinsert_for_size_lowering<"VINSERTF32X4Z256", v8bf16x_info, v16bf16x_info,
vinsert128_insert, INSERT_get_vinsert128_imm, [HasVLX]>;
// Codegen pattern with the alternative types insert VEC128 into VEC512
defm : vinsert_for_size_lowering<"VINSERTI32x4Z", v8i16x_info, v32i16_info,
defm : vinsert_for_size_lowering<"VINSERTI32X4Z", v8i16x_info, v32i16_info,
vinsert128_insert, INSERT_get_vinsert128_imm, [HasAVX512]>;
defm : vinsert_for_size_lowering<"VINSERTI32x4Z", v16i8x_info, v64i8_info,
defm : vinsert_for_size_lowering<"VINSERTI32X4Z", v16i8x_info, v64i8_info,
vinsert128_insert, INSERT_get_vinsert128_imm, [HasAVX512]>;
defm : vinsert_for_size_lowering<"VINSERTF32x4Z", v8f16x_info, v32f16_info,
defm : vinsert_for_size_lowering<"VINSERTF32X4Z", v8f16x_info, v32f16_info,
vinsert128_insert, INSERT_get_vinsert128_imm, [HasAVX512]>;
defm : vinsert_for_size_lowering<"VINSERTF32x4Z", v8bf16x_info, v32bf16_info,
defm : vinsert_for_size_lowering<"VINSERTF32X4Z", v8bf16x_info, v32bf16_info,
vinsert128_insert, INSERT_get_vinsert128_imm, [HasAVX512]>;
// Codegen pattern with the alternative types insert VEC256 into VEC512
defm : vinsert_for_size_lowering<"VINSERTI64x4Z", v16i16x_info, v32i16_info,
defm : vinsert_for_size_lowering<"VINSERTI64X4Z", v16i16x_info, v32i16_info,
vinsert256_insert, INSERT_get_vinsert256_imm, [HasAVX512]>;
defm : vinsert_for_size_lowering<"VINSERTI64x4Z", v32i8x_info, v64i8_info,
defm : vinsert_for_size_lowering<"VINSERTI64X4Z", v32i8x_info, v64i8_info,
vinsert256_insert, INSERT_get_vinsert256_imm, [HasAVX512]>;
defm : vinsert_for_size_lowering<"VINSERTF64x4Z", v16f16x_info, v32f16_info,
defm : vinsert_for_size_lowering<"VINSERTF64X4Z", v16f16x_info, v32f16_info,
vinsert256_insert, INSERT_get_vinsert256_imm, [HasAVX512]>;
defm : vinsert_for_size_lowering<"VINSERTF64x4Z", v16bf16x_info, v32bf16_info,
defm : vinsert_for_size_lowering<"VINSERTF64X4Z", v16bf16x_info, v32bf16_info,
vinsert256_insert, INSERT_get_vinsert256_imm, [HasAVX512]>;
@ -568,81 +568,81 @@ let Predicates = p in {
}
}
defm : vinsert_for_mask_cast<"VINSERTF32x4Z256", v2f64x_info, v4f64x_info,
defm : vinsert_for_mask_cast<"VINSERTF32X4Z256", v2f64x_info, v4f64x_info,
v8f32x_info, vinsert128_insert,
INSERT_get_vinsert128_imm, [HasVLX]>;
defm : vinsert_for_mask_cast<"VINSERTF64x2Z256", v4f32x_info, v8f32x_info,
defm : vinsert_for_mask_cast<"VINSERTF64X2Z256", v4f32x_info, v8f32x_info,
v4f64x_info, vinsert128_insert,
INSERT_get_vinsert128_imm, [HasDQI, HasVLX]>;
defm : vinsert_for_mask_cast<"VINSERTI32x4Z256", v2i64x_info, v4i64x_info,
defm : vinsert_for_mask_cast<"VINSERTI32X4Z256", v2i64x_info, v4i64x_info,
v8i32x_info, vinsert128_insert,
INSERT_get_vinsert128_imm, [HasVLX]>;
defm : vinsert_for_mask_cast<"VINSERTI32x4Z256", v8i16x_info, v16i16x_info,
defm : vinsert_for_mask_cast<"VINSERTI32X4Z256", v8i16x_info, v16i16x_info,
v8i32x_info, vinsert128_insert,
INSERT_get_vinsert128_imm, [HasVLX]>;
defm : vinsert_for_mask_cast<"VINSERTI32x4Z256", v16i8x_info, v32i8x_info,
defm : vinsert_for_mask_cast<"VINSERTI32X4Z256", v16i8x_info, v32i8x_info,
v8i32x_info, vinsert128_insert,
INSERT_get_vinsert128_imm, [HasVLX]>;
defm : vinsert_for_mask_cast<"VINSERTF64x2Z256", v4i32x_info, v8i32x_info,
defm : vinsert_for_mask_cast<"VINSERTF64X2Z256", v4i32x_info, v8i32x_info,
v4i64x_info, vinsert128_insert,
INSERT_get_vinsert128_imm, [HasDQI, HasVLX]>;
defm : vinsert_for_mask_cast<"VINSERTF64x2Z256", v8i16x_info, v16i16x_info,
defm : vinsert_for_mask_cast<"VINSERTF64X2Z256", v8i16x_info, v16i16x_info,
v4i64x_info, vinsert128_insert,
INSERT_get_vinsert128_imm, [HasDQI, HasVLX]>;
defm : vinsert_for_mask_cast<"VINSERTF64x2Z256", v16i8x_info, v32i8x_info,
defm : vinsert_for_mask_cast<"VINSERTF64X2Z256", v16i8x_info, v32i8x_info,
v4i64x_info, vinsert128_insert,
INSERT_get_vinsert128_imm, [HasDQI, HasVLX]>;
defm : vinsert_for_mask_cast<"VINSERTF32x4Z", v2f64x_info, v8f64_info,
defm : vinsert_for_mask_cast<"VINSERTF32X4Z", v2f64x_info, v8f64_info,
v16f32_info, vinsert128_insert,
INSERT_get_vinsert128_imm, [HasAVX512]>;
defm : vinsert_for_mask_cast<"VINSERTF64x2Z", v4f32x_info, v16f32_info,
defm : vinsert_for_mask_cast<"VINSERTF64X2Z", v4f32x_info, v16f32_info,
v8f64_info, vinsert128_insert,
INSERT_get_vinsert128_imm, [HasDQI]>;
defm : vinsert_for_mask_cast<"VINSERTI32x4Z", v2i64x_info, v8i64_info,
defm : vinsert_for_mask_cast<"VINSERTI32X4Z", v2i64x_info, v8i64_info,
v16i32_info, vinsert128_insert,
INSERT_get_vinsert128_imm, [HasAVX512]>;
defm : vinsert_for_mask_cast<"VINSERTI32x4Z", v8i16x_info, v32i16_info,
defm : vinsert_for_mask_cast<"VINSERTI32X4Z", v8i16x_info, v32i16_info,
v16i32_info, vinsert128_insert,
INSERT_get_vinsert128_imm, [HasAVX512]>;
defm : vinsert_for_mask_cast<"VINSERTI32x4Z", v16i8x_info, v64i8_info,
defm : vinsert_for_mask_cast<"VINSERTI32X4Z", v16i8x_info, v64i8_info,
v16i32_info, vinsert128_insert,
INSERT_get_vinsert128_imm, [HasAVX512]>;
defm : vinsert_for_mask_cast<"VINSERTI64x2Z", v4i32x_info, v16i32_info,
defm : vinsert_for_mask_cast<"VINSERTI64X2Z", v4i32x_info, v16i32_info,
v8i64_info, vinsert128_insert,
INSERT_get_vinsert128_imm, [HasDQI]>;
defm : vinsert_for_mask_cast<"VINSERTI64x2Z", v8i16x_info, v32i16_info,
defm : vinsert_for_mask_cast<"VINSERTI64X2Z", v8i16x_info, v32i16_info,
v8i64_info, vinsert128_insert,
INSERT_get_vinsert128_imm, [HasDQI]>;
defm : vinsert_for_mask_cast<"VINSERTI64x2Z", v16i8x_info, v64i8_info,
defm : vinsert_for_mask_cast<"VINSERTI64X2Z", v16i8x_info, v64i8_info,
v8i64_info, vinsert128_insert,
INSERT_get_vinsert128_imm, [HasDQI]>;
defm : vinsert_for_mask_cast<"VINSERTF32x8Z", v4f64x_info, v8f64_info,
defm : vinsert_for_mask_cast<"VINSERTF32X8Z", v4f64x_info, v8f64_info,
v16f32_info, vinsert256_insert,
INSERT_get_vinsert256_imm, [HasDQI]>;
defm : vinsert_for_mask_cast<"VINSERTF64x4Z", v8f32x_info, v16f32_info,
defm : vinsert_for_mask_cast<"VINSERTF64X4Z", v8f32x_info, v16f32_info,
v8f64_info, vinsert256_insert,
INSERT_get_vinsert256_imm, [HasAVX512]>;
defm : vinsert_for_mask_cast<"VINSERTI32x8Z", v4i64x_info, v8i64_info,
defm : vinsert_for_mask_cast<"VINSERTI32X8Z", v4i64x_info, v8i64_info,
v16i32_info, vinsert256_insert,
INSERT_get_vinsert256_imm, [HasDQI]>;
defm : vinsert_for_mask_cast<"VINSERTI32x8Z", v16i16x_info, v32i16_info,
defm : vinsert_for_mask_cast<"VINSERTI32X8Z", v16i16x_info, v32i16_info,
v16i32_info, vinsert256_insert,
INSERT_get_vinsert256_imm, [HasDQI]>;
defm : vinsert_for_mask_cast<"VINSERTI32x8Z", v32i8x_info, v64i8_info,
defm : vinsert_for_mask_cast<"VINSERTI32X8Z", v32i8x_info, v64i8_info,
v16i32_info, vinsert256_insert,
INSERT_get_vinsert256_imm, [HasDQI]>;
defm : vinsert_for_mask_cast<"VINSERTI64x4Z", v8i32x_info, v16i32_info,
defm : vinsert_for_mask_cast<"VINSERTI64X4Z", v8i32x_info, v16i32_info,
v8i64_info, vinsert256_insert,
INSERT_get_vinsert256_imm, [HasAVX512]>;
defm : vinsert_for_mask_cast<"VINSERTI64x4Z", v16i16x_info, v32i16_info,
defm : vinsert_for_mask_cast<"VINSERTI64X4Z", v16i16x_info, v32i16_info,
v8i64_info, vinsert256_insert,
INSERT_get_vinsert256_imm, [HasAVX512]>;
defm : vinsert_for_mask_cast<"VINSERTI64x4Z", v32i8x_info, v64i8_info,
defm : vinsert_for_mask_cast<"VINSERTI64X4Z", v32i8x_info, v64i8_info,
v8i64_info, vinsert256_insert,
INSERT_get_vinsert256_imm, [HasAVX512]>;
@ -732,19 +732,19 @@ multiclass vextract_for_type<ValueType EltVT32, int Opcode128,
ValueType EltVT64, int Opcode256,
SchedWrite SchedRR, SchedWrite SchedMR> {
let Predicates = [HasAVX512] in {
defm NAME # "32x4Z" : vextract_for_size<Opcode128,
defm NAME # "32X4Z" : vextract_for_size<Opcode128,
X86VectorVTInfo<16, EltVT32, VR512>,
X86VectorVTInfo< 4, EltVT32, VR128X>,
vextract128_extract, SchedRR, SchedMR>,
EVEX_V512, EVEX_CD8<32, CD8VT4>;
defm NAME # "64x4Z" : vextract_for_size<Opcode256,
defm NAME # "64X4Z" : vextract_for_size<Opcode256,
X86VectorVTInfo< 8, EltVT64, VR512>,
X86VectorVTInfo< 4, EltVT64, VR256X>,
vextract256_extract, SchedRR, SchedMR>,
REX_W, EVEX_V512, EVEX_CD8<64, CD8VT4>;
}
let Predicates = [HasVLX] in
defm NAME # "32x4Z256" : vextract_for_size<Opcode128,
defm NAME # "32X4Z256" : vextract_for_size<Opcode128,
X86VectorVTInfo< 8, EltVT32, VR256X>,
X86VectorVTInfo< 4, EltVT32, VR128X>,
vextract128_extract, SchedRR, SchedMR>,
@ -752,7 +752,7 @@ multiclass vextract_for_type<ValueType EltVT32, int Opcode128,
// Even with DQI we'd like to only use these instructions for masking.
let Predicates = [HasVLX, HasDQI] in
defm NAME # "64x2Z256" : vextract_for_size_split<Opcode128,
defm NAME # "64X2Z256" : vextract_for_size_split<Opcode128,
X86VectorVTInfo< 4, EltVT64, VR256X>,
X86VectorVTInfo< 2, EltVT64, VR128X>,
null_frag, vextract128_extract, SchedRR, SchedMR>,
@ -760,12 +760,12 @@ multiclass vextract_for_type<ValueType EltVT32, int Opcode128,
// Even with DQI we'd like to only use these instructions for masking.
let Predicates = [HasDQI] in {
defm NAME # "64x2Z" : vextract_for_size_split<Opcode128,
defm NAME # "64X2Z" : vextract_for_size_split<Opcode128,
X86VectorVTInfo< 8, EltVT64, VR512>,
X86VectorVTInfo< 2, EltVT64, VR128X>,
null_frag, vextract128_extract, SchedRR, SchedMR>,
REX_W, EVEX_V512, EVEX_CD8<64, CD8VT2>;
defm NAME # "32x8Z" : vextract_for_size_split<Opcode256,
defm NAME # "32X8Z" : vextract_for_size_split<Opcode256,
X86VectorVTInfo<16, EltVT32, VR512>,
X86VectorVTInfo< 8, EltVT32, VR256X>,
null_frag, vextract256_extract, SchedRR, SchedMR>,
@ -779,48 +779,48 @@ defm VEXTRACTI : vextract_for_type<i32, 0x39, i64, 0x3b, WriteShuffle256, WriteV
// extract_subvector codegen patterns with the alternative types.
// Even with AVX512DQ we'll still use these for unmasked operations.
defm : vextract_for_size_lowering<"VEXTRACTF32x4Z", v8f64_info, v2f64x_info,
defm : vextract_for_size_lowering<"VEXTRACTF32X4Z", v8f64_info, v2f64x_info,
vextract128_extract, EXTRACT_get_vextract128_imm, [HasAVX512]>;
defm : vextract_for_size_lowering<"VEXTRACTI32x4Z", v8i64_info, v2i64x_info,
defm : vextract_for_size_lowering<"VEXTRACTI32X4Z", v8i64_info, v2i64x_info,
vextract128_extract, EXTRACT_get_vextract128_imm, [HasAVX512]>;
defm : vextract_for_size_lowering<"VEXTRACTF64x4Z", v16f32_info, v8f32x_info,
defm : vextract_for_size_lowering<"VEXTRACTF64X4Z", v16f32_info, v8f32x_info,
vextract256_extract, EXTRACT_get_vextract256_imm, [HasAVX512]>;
defm : vextract_for_size_lowering<"VEXTRACTI64x4Z", v16i32_info, v8i32x_info,
defm : vextract_for_size_lowering<"VEXTRACTI64X4Z", v16i32_info, v8i32x_info,
vextract256_extract, EXTRACT_get_vextract256_imm, [HasAVX512]>;
defm : vextract_for_size_lowering<"VEXTRACTF32x4Z256", v4f64x_info, v2f64x_info,
defm : vextract_for_size_lowering<"VEXTRACTF32X4Z256", v4f64x_info, v2f64x_info,
vextract128_extract, EXTRACT_get_vextract128_imm, [HasVLX]>;
defm : vextract_for_size_lowering<"VEXTRACTI32x4Z256", v4i64x_info, v2i64x_info,
defm : vextract_for_size_lowering<"VEXTRACTI32X4Z256", v4i64x_info, v2i64x_info,
vextract128_extract, EXTRACT_get_vextract128_imm, [HasVLX]>;
// Codegen pattern with the alternative types extract VEC128 from VEC256
defm : vextract_for_size_lowering<"VEXTRACTI32x4Z256", v16i16x_info, v8i16x_info,
defm : vextract_for_size_lowering<"VEXTRACTI32X4Z256", v16i16x_info, v8i16x_info,
vextract128_extract, EXTRACT_get_vextract128_imm, [HasVLX]>;
defm : vextract_for_size_lowering<"VEXTRACTI32x4Z256", v32i8x_info, v16i8x_info,
defm : vextract_for_size_lowering<"VEXTRACTI32X4Z256", v32i8x_info, v16i8x_info,
vextract128_extract, EXTRACT_get_vextract128_imm, [HasVLX]>;
defm : vextract_for_size_lowering<"VEXTRACTF32x4Z256", v16f16x_info, v8f16x_info,
defm : vextract_for_size_lowering<"VEXTRACTF32X4Z256", v16f16x_info, v8f16x_info,
vextract128_extract, EXTRACT_get_vextract128_imm, [HasVLX]>;
defm : vextract_for_size_lowering<"VEXTRACTF32x4Z256", v16bf16x_info, v8bf16x_info,
defm : vextract_for_size_lowering<"VEXTRACTF32X4Z256", v16bf16x_info, v8bf16x_info,
vextract128_extract, EXTRACT_get_vextract128_imm, [HasVLX]>;
// Codegen pattern with the alternative types extract VEC128 from VEC512
defm : vextract_for_size_lowering<"VEXTRACTI32x4Z", v32i16_info, v8i16x_info,
defm : vextract_for_size_lowering<"VEXTRACTI32X4Z", v32i16_info, v8i16x_info,
vextract128_extract, EXTRACT_get_vextract128_imm, [HasAVX512]>;
defm : vextract_for_size_lowering<"VEXTRACTI32x4Z", v64i8_info, v16i8x_info,
defm : vextract_for_size_lowering<"VEXTRACTI32X4Z", v64i8_info, v16i8x_info,
vextract128_extract, EXTRACT_get_vextract128_imm, [HasAVX512]>;
defm : vextract_for_size_lowering<"VEXTRACTF32x4Z", v32f16_info, v8f16x_info,
defm : vextract_for_size_lowering<"VEXTRACTF32X4Z", v32f16_info, v8f16x_info,
vextract128_extract, EXTRACT_get_vextract128_imm, [HasAVX512]>;
defm : vextract_for_size_lowering<"VEXTRACTF32x4Z", v32bf16_info, v8bf16x_info,
defm : vextract_for_size_lowering<"VEXTRACTF32X4Z", v32bf16_info, v8bf16x_info,
vextract128_extract, EXTRACT_get_vextract128_imm, [HasAVX512]>;
// Codegen pattern with the alternative types extract VEC256 from VEC512
defm : vextract_for_size_lowering<"VEXTRACTI64x4Z", v32i16_info, v16i16x_info,
defm : vextract_for_size_lowering<"VEXTRACTI64X4Z", v32i16_info, v16i16x_info,
vextract256_extract, EXTRACT_get_vextract256_imm, [HasAVX512]>;
defm : vextract_for_size_lowering<"VEXTRACTI64x4Z", v64i8_info, v32i8x_info,
defm : vextract_for_size_lowering<"VEXTRACTI64X4Z", v64i8_info, v32i8x_info,
vextract256_extract, EXTRACT_get_vextract256_imm, [HasAVX512]>;
defm : vextract_for_size_lowering<"VEXTRACTF64x4Z", v32f16_info, v16f16x_info,
defm : vextract_for_size_lowering<"VEXTRACTF64X4Z", v32f16_info, v16f16x_info,
vextract256_extract, EXTRACT_get_vextract256_imm, [HasAVX512]>;
defm : vextract_for_size_lowering<"VEXTRACTF64x4Z", v32bf16_info, v16bf16x_info,
defm : vextract_for_size_lowering<"VEXTRACTF64X4Z", v32bf16_info, v16bf16x_info,
vextract256_extract, EXTRACT_get_vextract256_imm, [HasAVX512]>;
@ -861,31 +861,31 @@ def : Pat<(v16i8 (extract_subvector (v64i8 VR512:$src), (iPTR 16))),
// smaller extract to enable EVEX->VEX.
let Predicates = [HasVLX] in {
def : Pat<(v2i64 (extract_subvector (v8i64 VR512:$src), (iPTR 2))),
(v2i64 (VEXTRACTI32x4Z256rri
(v2i64 (VEXTRACTI32X4Z256rri
(v4i64 (EXTRACT_SUBREG (v8i64 VR512:$src), sub_ymm)),
(iPTR 1)))>;
def : Pat<(v2f64 (extract_subvector (v8f64 VR512:$src), (iPTR 2))),
(v2f64 (VEXTRACTF32x4Z256rri
(v2f64 (VEXTRACTF32X4Z256rri
(v4f64 (EXTRACT_SUBREG (v8f64 VR512:$src), sub_ymm)),
(iPTR 1)))>;
def : Pat<(v4i32 (extract_subvector (v16i32 VR512:$src), (iPTR 4))),
(v4i32 (VEXTRACTI32x4Z256rri
(v4i32 (VEXTRACTI32X4Z256rri
(v8i32 (EXTRACT_SUBREG (v16i32 VR512:$src), sub_ymm)),
(iPTR 1)))>;
def : Pat<(v4f32 (extract_subvector (v16f32 VR512:$src), (iPTR 4))),
(v4f32 (VEXTRACTF32x4Z256rri
(v4f32 (VEXTRACTF32X4Z256rri
(v8f32 (EXTRACT_SUBREG (v16f32 VR512:$src), sub_ymm)),
(iPTR 1)))>;
def : Pat<(v8i16 (extract_subvector (v32i16 VR512:$src), (iPTR 8))),
(v8i16 (VEXTRACTI32x4Z256rri
(v8i16 (VEXTRACTI32X4Z256rri
(v16i16 (EXTRACT_SUBREG (v32i16 VR512:$src), sub_ymm)),
(iPTR 1)))>;
def : Pat<(v8f16 (extract_subvector (v32f16 VR512:$src), (iPTR 8))),
(v8f16 (VEXTRACTF32x4Z256rri
(v8f16 (VEXTRACTF32X4Z256rri
(v16f16 (EXTRACT_SUBREG (v32f16 VR512:$src), sub_ymm)),
(iPTR 1)))>;
def : Pat<(v16i8 (extract_subvector (v64i8 VR512:$src), (iPTR 16))),
(v16i8 (VEXTRACTI32x4Z256rri
(v16i8 (VEXTRACTI32X4Z256rri
(v32i8 (EXTRACT_SUBREG (v64i8 VR512:$src), sub_ymm)),
(iPTR 1)))>;
}
@ -919,81 +919,81 @@ let Predicates = p in {
}
}
defm : vextract_for_mask_cast<"VEXTRACTF32x4Z256", v4f64x_info, v2f64x_info,
defm : vextract_for_mask_cast<"VEXTRACTF32X4Z256", v4f64x_info, v2f64x_info,
v4f32x_info, vextract128_extract,
EXTRACT_get_vextract128_imm, [HasVLX]>;
defm : vextract_for_mask_cast<"VEXTRACTF64x2Z256", v8f32x_info, v4f32x_info,
defm : vextract_for_mask_cast<"VEXTRACTF64X2Z256", v8f32x_info, v4f32x_info,
v2f64x_info, vextract128_extract,
EXTRACT_get_vextract128_imm, [HasDQI, HasVLX]>;
defm : vextract_for_mask_cast<"VEXTRACTI32x4Z256", v4i64x_info, v2i64x_info,
defm : vextract_for_mask_cast<"VEXTRACTI32X4Z256", v4i64x_info, v2i64x_info,
v4i32x_info, vextract128_extract,
EXTRACT_get_vextract128_imm, [HasVLX]>;
defm : vextract_for_mask_cast<"VEXTRACTI32x4Z256", v16i16x_info, v8i16x_info,
defm : vextract_for_mask_cast<"VEXTRACTI32X4Z256", v16i16x_info, v8i16x_info,
v4i32x_info, vextract128_extract,
EXTRACT_get_vextract128_imm, [HasVLX]>;
defm : vextract_for_mask_cast<"VEXTRACTI32x4Z256", v32i8x_info, v16i8x_info,
defm : vextract_for_mask_cast<"VEXTRACTI32X4Z256", v32i8x_info, v16i8x_info,
v4i32x_info, vextract128_extract,
EXTRACT_get_vextract128_imm, [HasVLX]>;
defm : vextract_for_mask_cast<"VEXTRACTI64x2Z256", v8i32x_info, v4i32x_info,
defm : vextract_for_mask_cast<"VEXTRACTI64X2Z256", v8i32x_info, v4i32x_info,
v2i64x_info, vextract128_extract,
EXTRACT_get_vextract128_imm, [HasDQI, HasVLX]>;
defm : vextract_for_mask_cast<"VEXTRACTI64x2Z256", v16i16x_info, v8i16x_info,
defm : vextract_for_mask_cast<"VEXTRACTI64X2Z256", v16i16x_info, v8i16x_info,
v2i64x_info, vextract128_extract,
EXTRACT_get_vextract128_imm, [HasDQI, HasVLX]>;
defm : vextract_for_mask_cast<"VEXTRACTI64x2Z256", v32i8x_info, v16i8x_info,
defm : vextract_for_mask_cast<"VEXTRACTI64X2Z256", v32i8x_info, v16i8x_info,
v2i64x_info, vextract128_extract,
EXTRACT_get_vextract128_imm, [HasDQI, HasVLX]>;
defm : vextract_for_mask_cast<"VEXTRACTF32x4Z", v8f64_info, v2f64x_info,
defm : vextract_for_mask_cast<"VEXTRACTF32X4Z", v8f64_info, v2f64x_info,
v4f32x_info, vextract128_extract,
EXTRACT_get_vextract128_imm, [HasAVX512]>;
defm : vextract_for_mask_cast<"VEXTRACTF64x2Z", v16f32_info, v4f32x_info,
defm : vextract_for_mask_cast<"VEXTRACTF64X2Z", v16f32_info, v4f32x_info,
v2f64x_info, vextract128_extract,
EXTRACT_get_vextract128_imm, [HasDQI]>;
defm : vextract_for_mask_cast<"VEXTRACTI32x4Z", v8i64_info, v2i64x_info,
defm : vextract_for_mask_cast<"VEXTRACTI32X4Z", v8i64_info, v2i64x_info,
v4i32x_info, vextract128_extract,
EXTRACT_get_vextract128_imm, [HasAVX512]>;
defm : vextract_for_mask_cast<"VEXTRACTI32x4Z", v32i16_info, v8i16x_info,
defm : vextract_for_mask_cast<"VEXTRACTI32X4Z", v32i16_info, v8i16x_info,
v4i32x_info, vextract128_extract,
EXTRACT_get_vextract128_imm, [HasAVX512]>;
defm : vextract_for_mask_cast<"VEXTRACTI32x4Z", v64i8_info, v16i8x_info,
defm : vextract_for_mask_cast<"VEXTRACTI32X4Z", v64i8_info, v16i8x_info,
v4i32x_info, vextract128_extract,
EXTRACT_get_vextract128_imm, [HasAVX512]>;
defm : vextract_for_mask_cast<"VEXTRACTI64x2Z", v16i32_info, v4i32x_info,
defm : vextract_for_mask_cast<"VEXTRACTI64X2Z", v16i32_info, v4i32x_info,
v2i64x_info, vextract128_extract,
EXTRACT_get_vextract128_imm, [HasDQI]>;
defm : vextract_for_mask_cast<"VEXTRACTI64x2Z", v32i16_info, v8i16x_info,
defm : vextract_for_mask_cast<"VEXTRACTI64X2Z", v32i16_info, v8i16x_info,
v2i64x_info, vextract128_extract,
EXTRACT_get_vextract128_imm, [HasDQI]>;
defm : vextract_for_mask_cast<"VEXTRACTI64x2Z", v64i8_info, v16i8x_info,
defm : vextract_for_mask_cast<"VEXTRACTI64X2Z", v64i8_info, v16i8x_info,
v2i64x_info, vextract128_extract,
EXTRACT_get_vextract128_imm, [HasDQI]>;
defm : vextract_for_mask_cast<"VEXTRACTF32x8Z", v8f64_info, v4f64x_info,
defm : vextract_for_mask_cast<"VEXTRACTF32X8Z", v8f64_info, v4f64x_info,
v8f32x_info, vextract256_extract,
EXTRACT_get_vextract256_imm, [HasDQI]>;
defm : vextract_for_mask_cast<"VEXTRACTF64x4Z", v16f32_info, v8f32x_info,
defm : vextract_for_mask_cast<"VEXTRACTF64X4Z", v16f32_info, v8f32x_info,
v4f64x_info, vextract256_extract,
EXTRACT_get_vextract256_imm, [HasAVX512]>;
defm : vextract_for_mask_cast<"VEXTRACTI32x8Z", v8i64_info, v4i64x_info,
defm : vextract_for_mask_cast<"VEXTRACTI32X8Z", v8i64_info, v4i64x_info,
v8i32x_info, vextract256_extract,
EXTRACT_get_vextract256_imm, [HasDQI]>;
defm : vextract_for_mask_cast<"VEXTRACTI32x8Z", v32i16_info, v16i16x_info,
defm : vextract_for_mask_cast<"VEXTRACTI32X8Z", v32i16_info, v16i16x_info,
v8i32x_info, vextract256_extract,
EXTRACT_get_vextract256_imm, [HasDQI]>;
defm : vextract_for_mask_cast<"VEXTRACTI32x8Z", v64i8_info, v32i8x_info,
defm : vextract_for_mask_cast<"VEXTRACTI32X8Z", v64i8_info, v32i8x_info,
v8i32x_info, vextract256_extract,
EXTRACT_get_vextract256_imm, [HasDQI]>;
defm : vextract_for_mask_cast<"VEXTRACTI64x4Z", v16i32_info, v8i32x_info,
defm : vextract_for_mask_cast<"VEXTRACTI64X4Z", v16i32_info, v8i32x_info,
v4i64x_info, vextract256_extract,
EXTRACT_get_vextract256_imm, [HasAVX512]>;
defm : vextract_for_mask_cast<"VEXTRACTI64x4Z", v32i16_info, v16i16x_info,
defm : vextract_for_mask_cast<"VEXTRACTI64X4Z", v32i16_info, v16i16x_info,
v4i64x_info, vextract256_extract,
EXTRACT_get_vextract256_imm, [HasAVX512]>;
defm : vextract_for_mask_cast<"VEXTRACTI64x4Z", v64i8_info, v32i8x_info,
defm : vextract_for_mask_cast<"VEXTRACTI64X4Z", v64i8_info, v32i8x_info,
v4i64x_info, vextract256_extract,
EXTRACT_get_vextract256_imm, [HasAVX512]>;

8
llvm/lib/Target/X86/X86InstrInfo.cpp
View File

@ -6293,16 +6293,16 @@ bool X86InstrInfo::expandPostRAPseudo(MachineInstr &MI) const {
get(X86::VBROADCASTF64X4Zrm), X86::sub_ymm);
case X86::VMOVAPSZ128mr_NOVLX:
return expandNOVLXStore(MIB, &getRegisterInfo(), get(X86::VMOVAPSmr),
get(X86::VEXTRACTF32x4Zmri), X86::sub_xmm);
get(X86::VEXTRACTF32X4Zmri), X86::sub_xmm);
case X86::VMOVUPSZ128mr_NOVLX:
return expandNOVLXStore(MIB, &getRegisterInfo(), get(X86::VMOVUPSmr),
get(X86::VEXTRACTF32x4Zmri), X86::sub_xmm);
get(X86::VEXTRACTF32X4Zmri), X86::sub_xmm);
case X86::VMOVAPSZ256mr_NOVLX:
return expandNOVLXStore(MIB, &getRegisterInfo(), get(X86::VMOVAPSYmr),
get(X86::VEXTRACTF64x4Zmri), X86::sub_ymm);
get(X86::VEXTRACTF64X4Zmri), X86::sub_ymm);
case X86::VMOVUPSZ256mr_NOVLX:
return expandNOVLXStore(MIB, &getRegisterInfo(), get(X86::VMOVUPSYmr),
get(X86::VEXTRACTF64x4Zmri), X86::sub_ymm);
get(X86::VEXTRACTF64X4Zmri), X86::sub_ymm);
case X86::MOV32ri64: {
Register Reg = MIB.getReg(0);
Register Reg32 = RI.getSubReg(Reg, X86::sub_32bit);

48
llvm/lib/Target/X86/X86ReplaceableInstrs.def
View File

@ -110,30 +110,30 @@ ENTRY(VBROADCASTSDZ256rr, VBROADCASTSDZ256rr, VPBROADCASTQZ256rr)
ENTRY(VBROADCASTSDZ256rm, VBROADCASTSDZ256rm, VPBROADCASTQZ256rm)
ENTRY(VBROADCASTSDZrr, VBROADCASTSDZrr, VPBROADCASTQZrr)
ENTRY(VBROADCASTSDZrm, VBROADCASTSDZrm, VPBROADCASTQZrm)
ENTRY(VINSERTF32x4Zrri, VINSERTF32x4Zrri, VINSERTI32x4Zrri)
ENTRY(VINSERTF32x4Zrmi, VINSERTF32x4Zrmi, VINSERTI32x4Zrmi)
ENTRY(VINSERTF32x8Zrri, VINSERTF32x8Zrri, VINSERTI32x8Zrri)
ENTRY(VINSERTF32x8Zrmi, VINSERTF32x8Zrmi, VINSERTI32x8Zrmi)
ENTRY(VINSERTF64x2Zrri, VINSERTF64x2Zrri, VINSERTI64x2Zrri)
ENTRY(VINSERTF64x2Zrmi, VINSERTF64x2Zrmi, VINSERTI64x2Zrmi)
ENTRY(VINSERTF64x4Zrri, VINSERTF64x4Zrri, VINSERTI64x4Zrri)
ENTRY(VINSERTF64x4Zrmi, VINSERTF64x4Zrmi, VINSERTI64x4Zrmi)
ENTRY(VINSERTF32x4Z256rri, VINSERTF32x4Z256rri, VINSERTI32x4Z256rri)
ENTRY(VINSERTF32x4Z256rmi, VINSERTF32x4Z256rmi, VINSERTI32x4Z256rmi)
ENTRY(VINSERTF64x2Z256rri, VINSERTF64x2Z256rri, VINSERTI64x2Z256rri)
ENTRY(VINSERTF64x2Z256rmi, VINSERTF64x2Z256rmi, VINSERTI64x2Z256rmi)
ENTRY(VEXTRACTF32x4Zrri, VEXTRACTF32x4Zrri, VEXTRACTI32x4Zrri)
ENTRY(VEXTRACTF32x4Zmri, VEXTRACTF32x4Zmri, VEXTRACTI32x4Zmri)
ENTRY(VEXTRACTF32x8Zrri, VEXTRACTF32x8Zrri, VEXTRACTI32x8Zrri)
ENTRY(VEXTRACTF32x8Zmri, VEXTRACTF32x8Zmri, VEXTRACTI32x8Zmri)
ENTRY(VEXTRACTF64x2Zrri, VEXTRACTF64x2Zrri, VEXTRACTI64x2Zrri)
ENTRY(VEXTRACTF64x2Zmri, VEXTRACTF64x2Zmri, VEXTRACTI64x2Zmri)
ENTRY(VEXTRACTF64x4Zrri, VEXTRACTF64x4Zrri, VEXTRACTI64x4Zrri)
ENTRY(VEXTRACTF64x4Zmri, VEXTRACTF64x4Zmri, VEXTRACTI64x4Zmri)
ENTRY(VEXTRACTF32x4Z256rri, VEXTRACTF32x4Z256rri, VEXTRACTI32x4Z256rri)
ENTRY(VEXTRACTF32x4Z256mri, VEXTRACTF32x4Z256mri, VEXTRACTI32x4Z256mri)
ENTRY(VEXTRACTF64x2Z256rri, VEXTRACTF64x2Z256rri, VEXTRACTI64x2Z256rri)
ENTRY(VEXTRACTF64x2Z256mri, VEXTRACTF64x2Z256mri, VEXTRACTI64x2Z256mri)
ENTRY(VINSERTF32X4Zrri, VINSERTF32X4Zrri, VINSERTI32X4Zrri)
ENTRY(VINSERTF32X4Zrmi, VINSERTF32X4Zrmi, VINSERTI32X4Zrmi)
ENTRY(VINSERTF32X8Zrri, VINSERTF32X8Zrri, VINSERTI32X8Zrri)
ENTRY(VINSERTF32X8Zrmi, VINSERTF32X8Zrmi, VINSERTI32X8Zrmi)
ENTRY(VINSERTF64X2Zrri, VINSERTF64X2Zrri, VINSERTI64X2Zrri)
ENTRY(VINSERTF64X2Zrmi, VINSERTF64X2Zrmi, VINSERTI64X2Zrmi)
ENTRY(VINSERTF64X4Zrri, VINSERTF64X4Zrri, VINSERTI64X4Zrri)
ENTRY(VINSERTF64X4Zrmi, VINSERTF64X4Zrmi, VINSERTI64X4Zrmi)
ENTRY(VINSERTF32X4Z256rri, VINSERTF32X4Z256rri, VINSERTI32X4Z256rri)
ENTRY(VINSERTF32X4Z256rmi, VINSERTF32X4Z256rmi, VINSERTI32X4Z256rmi)
ENTRY(VINSERTF64X2Z256rri, VINSERTF64X2Z256rri, VINSERTI64X2Z256rri)
ENTRY(VINSERTF64X2Z256rmi, VINSERTF64X2Z256rmi, VINSERTI64X2Z256rmi)
ENTRY(VEXTRACTF32X4Zrri, VEXTRACTF32X4Zrri, VEXTRACTI32X4Zrri)
ENTRY(VEXTRACTF32X4Zmri, VEXTRACTF32X4Zmri, VEXTRACTI32X4Zmri)
ENTRY(VEXTRACTF32X8Zrri, VEXTRACTF32X8Zrri, VEXTRACTI32X8Zrri)
ENTRY(VEXTRACTF32X8Zmri, VEXTRACTF32X8Zmri, VEXTRACTI32X8Zmri)
ENTRY(VEXTRACTF64X2Zrri, VEXTRACTF64X2Zrri, VEXTRACTI64X2Zrri)
ENTRY(VEXTRACTF64X2Zmri, VEXTRACTF64X2Zmri, VEXTRACTI64X2Zmri)
ENTRY(VEXTRACTF64X4Zrri, VEXTRACTF64X4Zrri, VEXTRACTI64X4Zrri)
ENTRY(VEXTRACTF64X4Zmri, VEXTRACTF64X4Zmri, VEXTRACTI64X4Zmri)
ENTRY(VEXTRACTF32X4Z256rri, VEXTRACTF32X4Z256rri, VEXTRACTI32X4Z256rri)
ENTRY(VEXTRACTF32X4Z256mri, VEXTRACTF32X4Z256mri, VEXTRACTI32X4Z256mri)
ENTRY(VEXTRACTF64X2Z256rri, VEXTRACTF64X2Z256rri, VEXTRACTI64X2Z256rri)
ENTRY(VEXTRACTF64X2Z256mri, VEXTRACTF64X2Z256mri, VEXTRACTI64X2Z256mri)
ENTRY(VPERMILPSmi, VPERMILPSmi, VPSHUFDmi)
ENTRY(VPERMILPSri, VPERMILPSri, VPSHUFDri)
ENTRY(VPERMILPSZ128mi, VPERMILPSZ128mi, VPSHUFDZ128mi)

16
llvm/lib/Target/X86/X86SchedIceLake.td
View File

@ -1591,14 +1591,14 @@ def: InstRW<[ICXWriteResGroup121, ReadAfterVecYLd],
"VBROADCASTI64X4Zrm(b?)",
"VBROADCASTSD(Z|Z256)rm(b?)",
"VBROADCASTSS(Z|Z256)rm(b?)",
"VINSERTF32x4(Z|Z256)rm(b?)",
"VINSERTF32x8Zrm(b?)",
"VINSERTF64x2(Z|Z256)rm(b?)",
"VINSERTF64x4Zrm(b?)",
"VINSERTI32x4(Z|Z256)rm(b?)",
"VINSERTI32x8Zrm(b?)",
"VINSERTI64x2(Z|Z256)rm(b?)",
"VINSERTI64x4Zrm(b?)",
"VINSERTF32X4(Z|Z256)rm(b?)",
"VINSERTF32X8Zrm(b?)",
"VINSERTF64X2(Z|Z256)rm(b?)",
"VINSERTF64X4Zrm(b?)",
"VINSERTI32X4(Z|Z256)rm(b?)",
"VINSERTI32X8Zrm(b?)",
"VINSERTI64X2(Z|Z256)rm(b?)",
"VINSERTI64X4Zrm(b?)",
"VMOVAPD(Z|Z256)rm(b?)",
"VMOVAPS(Z|Z256)rm(b?)",
"VMOVDDUP(Z|Z256)rm(b?)",

6
llvm/lib/Target/X86/X86SchedSapphireRapids.td
View File

@ -1666,8 +1666,8 @@ def : InstRW<[SPRWriteResGroup131], (instregex "^VBROADCAST(F|I)32X(8|2)Zrmk(z?)
"^VMOVDQ(A|U)(32|64)Zrmk(z?)$",
"^VPBROADCAST(D|Q)Zrmk(z?)$")>;
def : InstRW<[SPRWriteResGroup131, ReadAfterVecLd], (instregex "^MMX_P(ADD|SUB)(B|D|Q|W)rm$")>;
def : InstRW<[SPRWriteResGroup131, ReadAfterVecYLd], (instregex "^VINSERT(F|I)(32|64)x4Zrmi((k|kz)?)$",
"^VINSERT(F|I)(32x8|64x2)Zrmi((k|kz)?)$",
def : InstRW<[SPRWriteResGroup131, ReadAfterVecYLd], (instregex "^VINSERT(F|I)(32|64)X4Zrmi((k|kz)?)$",
"^VINSERT(F|I)(32X8|64X2)Zrmi((k|kz)?)$",
"^VP(ADD|SUB)(B|D|Q|W)Zrm$",
"^VP(ADD|SUB)(D|Q)Zrm(b|k|kz)$",
"^VP(ADD|SUB)(D|Q)Zrmbk(z?)$",
@ -2710,7 +2710,7 @@ def : InstRW<[SPRWriteResGroup262], (instregex "^VBROADCAST(F|I)32X(2|4)Z256rmk(
"^VMOVDQ(A|U)(32|64)Z256rmk(z?)$",
"^VPBROADCAST(D|Q)Z256rmk(z?)$")>;
def : InstRW<[SPRWriteResGroup262, ReadAfterVecYLd], (instregex "^VINSERT(F|I)128rmi$",
"^VINSERT(F|I)(32x4|64x2)Z256rmi((k|kz)?)$",
"^VINSERT(F|I)(32X4|64X2)Z256rmi((k|kz)?)$",
"^VP(ADD|SUB)(B|D|Q|W)(Y|Z256)rm$",
"^VP(ADD|SUB)(D|Q)Z256rm(b|k|kz)$",
"^VP(ADD|SUB)(D|Q)Z256rmbk(z?)$",

16
llvm/lib/Target/X86/X86SchedSkylakeServer.td
View File

@ -1562,14 +1562,14 @@ def: InstRW<[SKXWriteResGroup121, ReadAfterVecYLd],
"VBROADCASTI64X4Zrm(b?)",
"VBROADCASTSD(Z|Z256)rm(b?)",
"VBROADCASTSS(Z|Z256)rm(b?)",
"VINSERTF32x4(Z|Z256)rm(b?)",
"VINSERTF32x8Zrm(b?)",
"VINSERTF64x2(Z|Z256)rm(b?)",
"VINSERTF64x4Zrm(b?)",
"VINSERTI32x4(Z|Z256)rm(b?)",
"VINSERTI32x8Zrm(b?)",
"VINSERTI64x2(Z|Z256)rm(b?)",
"VINSERTI64x4Zrm(b?)",
"VINSERTF32X4(Z|Z256)rm(b?)",
"VINSERTF32X8Zrm(b?)",
"VINSERTF64X2(Z|Z256)rm(b?)",
"VINSERTF64X4Zrm(b?)",
"VINSERTI32X4(Z|Z256)rm(b?)",
"VINSERTI32X8Zrm(b?)",
"VINSERTI64X2(Z|Z256)rm(b?)",
"VINSERTI64X4Zrm(b?)",
"VMOVAPD(Z|Z256)rm(b?)",
"VMOVAPS(Z|Z256)rm(b?)",
"VMOVDDUP(Z|Z256)rm(b?)",

2
llvm/test/CodeGen/X86/GlobalISel/select-extract-vec256.mir
View File

@ -65,7 +65,7 @@ registers:
# AVX-NEXT: RET 0, implicit $xmm0
#
# AVX512VL: %0:vr256x = COPY $ymm1
# AVX512VL-NEXT: %1:vr128x = VEXTRACTF32x4Z256rri %0, 1
# AVX512VL-NEXT: %1:vr128x = VEXTRACTF32X4Z256rri %0, 1
# AVX512VL-NEXT: $xmm0 = COPY %1
# AVX512VL-NEXT: RET 0, implicit $xmm0
body: |

4
llvm/test/CodeGen/X86/GlobalISel/select-extract-vec512.mir
View File

@ -59,7 +59,7 @@ registers:
- { id: 0, class: vecr }
- { id: 1, class: vecr }
# ALL: %0:vr512 = COPY $zmm1
# ALL-NEXT: %1:vr128x = VEXTRACTF32x4Zrri %0, 1
# ALL-NEXT: %1:vr128x = VEXTRACTF32X4Zrri %0, 1
# ALL-NEXT: $xmm0 = COPY %1
# ALL-NEXT: RET 0, implicit $xmm0
body: |
@ -111,7 +111,7 @@ registers:
- { id: 0, class: vecr }
- { id: 1, class: vecr }
# ALL: %0:vr512 = COPY $zmm1
# ALL-NEXT: %1:vr256x = VEXTRACTF64x4Zrri %0, 1
# ALL-NEXT: %1:vr256x = VEXTRACTF64X4Zrri %0, 1
# ALL-NEXT: $ymm0 = COPY %1
# ALL-NEXT: RET 0, implicit $ymm0
body: |

6
llvm/test/CodeGen/X86/GlobalISel/select-insert-vec256.mir
View File

@ -36,7 +36,7 @@ registers:
#
# AVX512VL: %0:vr256x = COPY $ymm0
# AVX512VL-NEXT: %1:vr128x = COPY $xmm1
# AVX512VL-NEXT: %2:vr256x = VINSERTF32x4Z256rri %0, %1, 0
# AVX512VL-NEXT: %2:vr256x = VINSERTF32X4Z256rri %0, %1, 0
# AVX512VL-NEXT: $ymm0 = COPY %2
# AVX512VL-NEXT: RET 0, implicit $ymm0
body: |
@ -98,7 +98,7 @@ registers:
#
# AVX512VL: %0:vr256x = COPY $ymm0
# AVX512VL-NEXT: %1:vr128x = COPY $xmm1
# AVX512VL-NEXT: %2:vr256x = VINSERTF32x4Z256rri %0, %1, 1
# AVX512VL-NEXT: %2:vr256x = VINSERTF32X4Z256rri %0, %1, 1
# AVX512VL-NEXT: $ymm0 = COPY %2
# AVX512VL-NEXT: RET 0, implicit $ymm0
body: |
@ -129,7 +129,7 @@ registers:
#
# AVX512VL: %0:vr256x = IMPLICIT_DEF
# AVX512VL-NEXT: %1:vr128x = COPY $xmm1
# AVX512VL-NEXT: %2:vr256x = VINSERTF32x4Z256rri %0, %1, 1
# AVX512VL-NEXT: %2:vr256x = VINSERTF32X4Z256rri %0, %1, 1
# AVX512VL-NEXT: $ymm0 = COPY %2
# AVX512VL-NEXT: RET 0, implicit $ymm0
body: |

24
llvm/test/CodeGen/X86/GlobalISel/select-insert-vec512.mir
View File

@ -51,8 +51,8 @@ body: |
; ALL-LABEL: name: test_insert_128_idx0
; ALL: [[COPY:%[0-9]+]]:vr512 = COPY $zmm0
; ALL: [[COPY1:%[0-9]+]]:vr128x = COPY $xmm1
; ALL: [[VINSERTF32x4Zrri:%[0-9]+]]:vr512 = VINSERTF32x4Zrri [[COPY]], [[COPY1]], 0
; ALL: $zmm0 = COPY [[VINSERTF32x4Zrri]]
; ALL: [[VINSERTF32X4Zrri:%[0-9]+]]:vr512 = VINSERTF32X4Zrri [[COPY]], [[COPY1]], 0
; ALL: $zmm0 = COPY [[VINSERTF32X4Zrri]]
; ALL: RET 0, implicit $ymm0
%0(<16 x s32>) = COPY $zmm0
%1(<4 x s32>) = COPY $xmm1
@ -102,8 +102,8 @@ body: |
; ALL-LABEL: name: test_insert_128_idx1
; ALL: [[COPY:%[0-9]+]]:vr512 = COPY $zmm0
; ALL: [[COPY1:%[0-9]+]]:vr128x = COPY $xmm1
; ALL: [[VINSERTF32x4Zrri:%[0-9]+]]:vr512 = VINSERTF32x4Zrri [[COPY]], [[COPY1]], 1
; ALL: $zmm0 = COPY [[VINSERTF32x4Zrri]]
; ALL: [[VINSERTF32X4Zrri:%[0-9]+]]:vr512 = VINSERTF32X4Zrri [[COPY]], [[COPY1]], 1
; ALL: $zmm0 = COPY [[VINSERTF32X4Zrri]]
; ALL: RET 0, implicit $ymm0
%0(<16 x s32>) = COPY $zmm0
%1(<4 x s32>) = COPY $xmm1
@ -127,8 +127,8 @@ body: |
; ALL-LABEL: name: test_insert_128_idx1_undef
; ALL: [[DEF:%[0-9]+]]:vr512 = IMPLICIT_DEF
; ALL: [[COPY:%[0-9]+]]:vr128x = COPY $xmm1
; ALL: [[VINSERTF32x4Zrri:%[0-9]+]]:vr512 = VINSERTF32x4Zrri [[DEF]], [[COPY]], 1
; ALL: $zmm0 = COPY [[VINSERTF32x4Zrri]]
; ALL: [[VINSERTF32X4Zrri:%[0-9]+]]:vr512 = VINSERTF32X4Zrri [[DEF]], [[COPY]], 1
; ALL: $zmm0 = COPY [[VINSERTF32X4Zrri]]
; ALL: RET 0, implicit $ymm0
%0(<16 x s32>) = IMPLICIT_DEF
%1(<4 x s32>) = COPY $xmm1
@ -152,8 +152,8 @@ body: |
; ALL-LABEL: name: test_insert_256_idx0
; ALL: [[COPY:%[0-9]+]]:vr512 = COPY $zmm0
; ALL: [[COPY1:%[0-9]+]]:vr256x = COPY $ymm1
; ALL: [[VINSERTF64x4Zrri:%[0-9]+]]:vr512 = VINSERTF64x4Zrri [[COPY]], [[COPY1]], 0
; ALL: $zmm0 = COPY [[VINSERTF64x4Zrri]]
; ALL: [[VINSERTF64X4Zrri:%[0-9]+]]:vr512 = VINSERTF64X4Zrri [[COPY]], [[COPY1]], 0
; ALL: $zmm0 = COPY [[VINSERTF64X4Zrri]]
; ALL: RET 0, implicit $ymm0
%0(<16 x s32>) = COPY $zmm0
%1(<8 x s32>) = COPY $ymm1
@ -203,8 +203,8 @@ body: |
; ALL-LABEL: name: test_insert_256_idx1
; ALL: [[COPY:%[0-9]+]]:vr512 = COPY $zmm0
; ALL: [[COPY1:%[0-9]+]]:vr256x = COPY $ymm1
; ALL: [[VINSERTF64x4Zrri:%[0-9]+]]:vr512 = VINSERTF64x4Zrri [[COPY]], [[COPY1]], 1
; ALL: $zmm0 = COPY [[VINSERTF64x4Zrri]]
; ALL: [[VINSERTF64X4Zrri:%[0-9]+]]:vr512 = VINSERTF64X4Zrri [[COPY]], [[COPY1]], 1
; ALL: $zmm0 = COPY [[VINSERTF64X4Zrri]]
; ALL: RET 0, implicit $ymm0
%0(<16 x s32>) = COPY $zmm0
%1(<8 x s32>) = COPY $ymm1
@ -228,8 +228,8 @@ body: |
; ALL-LABEL: name: test_insert_256_idx1_undef
; ALL: [[DEF:%[0-9]+]]:vr512 = IMPLICIT_DEF
; ALL: [[COPY:%[0-9]+]]:vr256x = COPY $ymm1
; ALL: [[VINSERTF64x4Zrri:%[0-9]+]]:vr512 = VINSERTF64x4Zrri [[DEF]], [[COPY]], 1
; ALL: $zmm0 = COPY [[VINSERTF64x4Zrri]]
; ALL: [[VINSERTF64X4Zrri:%[0-9]+]]:vr512 = VINSERTF64X4Zrri [[DEF]], [[COPY]], 1
; ALL: $zmm0 = COPY [[VINSERTF64X4Zrri]]
; ALL: RET 0, implicit $ymm0
%0(<16 x s32>) = IMPLICIT_DEF
%1(<8 x s32>) = COPY $ymm1

4
llvm/test/CodeGen/X86/GlobalISel/select-merge-vec256.mir
View File

@ -29,8 +29,8 @@ body: |
; AVX512VL-LABEL: name: test_merge
; AVX512VL: [[DEF:%[0-9]+]]:vr128x = IMPLICIT_DEF
; AVX512VL: undef %2.sub_xmm:vr256x = COPY [[DEF]]
; AVX512VL: [[VINSERTF32x4Z256rri:%[0-9]+]]:vr256x = VINSERTF32x4Z256rri %2, [[DEF]], 1
; AVX512VL: $ymm0 = COPY [[VINSERTF32x4Z256rri]]
; AVX512VL: [[VINSERTF32X4Z256rri:%[0-9]+]]:vr256x = VINSERTF32X4Z256rri %2, [[DEF]], 1
; AVX512VL: $ymm0 = COPY [[VINSERTF32X4Z256rri]]
; AVX512VL: RET 0, implicit $ymm0
%0(<4 x s32>) = IMPLICIT_DEF
%1(<8 x s32>) = G_CONCAT_VECTORS %0(<4 x s32>), %0(<4 x s32>)

12
llvm/test/CodeGen/X86/GlobalISel/select-merge-vec512.mir
View File

@ -24,10 +24,10 @@ body: |
; ALL-LABEL: name: test_merge_v128
; ALL: [[DEF:%[0-9]+]]:vr128x = IMPLICIT_DEF
; ALL: undef %2.sub_xmm:vr512 = COPY [[DEF]]
; ALL: [[VINSERTF32x4Zrri:%[0-9]+]]:vr512 = VINSERTF32x4Zrri %2, [[DEF]], 1
; ALL: [[VINSERTF32x4Zrri1:%[0-9]+]]:vr512 = VINSERTF32x4Zrri [[VINSERTF32x4Zrri]], [[DEF]], 2
; ALL: [[VINSERTF32x4Zrri2:%[0-9]+]]:vr512 = VINSERTF32x4Zrri [[VINSERTF32x4Zrri1]], [[DEF]], 3
; ALL: $zmm0 = COPY [[VINSERTF32x4Zrri2]]
; ALL: [[VINSERTF32X4Zrri:%[0-9]+]]:vr512 = VINSERTF32X4Zrri %2, [[DEF]], 1
; ALL: [[VINSERTF32X4Zrri1:%[0-9]+]]:vr512 = VINSERTF32X4Zrri [[VINSERTF32X4Zrri]], [[DEF]], 2
; ALL: [[VINSERTF32X4Zrri2:%[0-9]+]]:vr512 = VINSERTF32X4Zrri [[VINSERTF32X4Zrri1]], [[DEF]], 3
; ALL: $zmm0 = COPY [[VINSERTF32X4Zrri2]]
; ALL: RET 0, implicit $zmm0
%0(<4 x s32>) = IMPLICIT_DEF
%1(<16 x s32>) = G_CONCAT_VECTORS %0(<4 x s32>), %0(<4 x s32>), %0(<4 x s32>), %0(<4 x s32>)
@ -49,8 +49,8 @@ body: |
; ALL-LABEL: name: test_merge_v256
; ALL: [[DEF:%[0-9]+]]:vr256x = IMPLICIT_DEF
; ALL: undef %2.sub_ymm:vr512 = COPY [[DEF]]
; ALL: [[VINSERTF64x4Zrri:%[0-9]+]]:vr512 = VINSERTF64x4Zrri %2, [[DEF]], 1
; ALL: $zmm0 = COPY [[VINSERTF64x4Zrri]]
; ALL: [[VINSERTF64X4Zrri:%[0-9]+]]:vr512 = VINSERTF64X4Zrri %2, [[DEF]], 1
; ALL: $zmm0 = COPY [[VINSERTF64X4Zrri]]
; ALL: RET 0, implicit $zmm0
%0(<8 x s32>) = IMPLICIT_DEF
%1(<16 x s32>) = G_CONCAT_VECTORS %0(<8 x s32>), %0(<8 x s32>)

4
llvm/test/CodeGen/X86/GlobalISel/select-unmerge-vec256.mir
View File

@ -33,9 +33,9 @@ body: |
; AVX512VL-LABEL: name: test_unmerge
; AVX512VL: [[DEF:%[0-9]+]]:vr256x = IMPLICIT_DEF
; AVX512VL-NEXT: [[COPY:%[0-9]+]]:vr128x = COPY [[DEF]].sub_xmm
; AVX512VL-NEXT: [[VEXTRACTF32x4Z256rri:%[0-9]+]]:vr128x = VEXTRACTF32x4Z256rri [[DEF]], 1
; AVX512VL-NEXT: [[VEXTRACTF32X4Z256rri:%[0-9]+]]:vr128x = VEXTRACTF32X4Z256rri [[DEF]], 1
; AVX512VL-NEXT: $xmm0 = COPY [[COPY]]
; AVX512VL-NEXT: $xmm1 = COPY [[VEXTRACTF32x4Z256rri]]
; AVX512VL-NEXT: $xmm1 = COPY [[VEXTRACTF32X4Z256rri]]
; AVX512VL-NEXT: RET 0, implicit $xmm0, implicit $xmm1
%0(<8 x s32>) = IMPLICIT_DEF
%1(<4 x s32>), %2(<4 x s32>) = G_UNMERGE_VALUES %0(<8 x s32>)

8
llvm/test/CodeGen/X86/GlobalISel/select-unmerge-vec512.mir
View File

@ -27,9 +27,9 @@ body: |
; ALL-LABEL: name: test_unmerge_v128
; ALL: [[DEF:%[0-9]+]]:vr512 = IMPLICIT_DEF
; ALL: [[COPY:%[0-9]+]]:vr128x = COPY [[DEF]].sub_xmm
; ALL: [[VEXTRACTF32x4Zrri:%[0-9]+]]:vr128x = VEXTRACTF32x4Zrri [[DEF]], 1
; ALL: [[VEXTRACTF32x4Zrri1:%[0-9]+]]:vr128x = VEXTRACTF32x4Zrri [[DEF]], 2
; ALL: [[VEXTRACTF32x4Zrri2:%[0-9]+]]:vr128x = VEXTRACTF32x4Zrri [[DEF]], 3
; ALL: [[VEXTRACTF32X4Zrri:%[0-9]+]]:vr128x = VEXTRACTF32X4Zrri [[DEF]], 1
; ALL: [[VEXTRACTF32X4Zrri1:%[0-9]+]]:vr128x = VEXTRACTF32X4Zrri [[DEF]], 2
; ALL: [[VEXTRACTF32X4Zrri2:%[0-9]+]]:vr128x = VEXTRACTF32X4Zrri [[DEF]], 3
; ALL: $xmm0 = COPY [[COPY]]
; ALL: RET 0, implicit $xmm0
%0(<16 x s32>) = IMPLICIT_DEF
@ -53,7 +53,7 @@ body: |
; ALL-LABEL: name: test_unmerge_v256
; ALL: [[DEF:%[0-9]+]]:vr512 = IMPLICIT_DEF
; ALL: [[COPY:%[0-9]+]]:vr256x = COPY [[DEF]].sub_ymm
; ALL: [[VEXTRACTF64x4Zrri:%[0-9]+]]:vr256x = VEXTRACTF64x4Zrri [[DEF]], 1
; ALL: [[VEXTRACTF64X4Zrri:%[0-9]+]]:vr256x = VEXTRACTF64X4Zrri [[DEF]], 1
; ALL: $ymm0 = COPY [[COPY]]
; ALL: RET 0, implicit $ymm0
%0(<16 x s32>) = IMPLICIT_DEF

96
llvm/test/TableGen/x86-fold-tables.inc
View File

@ -487,19 +487,19 @@ static const X86FoldTableEntry Table0[] = {
{X86::VCVTPS2PHZ256rr, X86::VCVTPS2PHZ256mr, TB_FOLDED_STORE},
{X86::VCVTPS2PHZrr, X86::VCVTPS2PHZmr, TB_FOLDED_STORE},
{X86::VEXTRACTF128rri, X86::VEXTRACTF128mri, TB_FOLDED_STORE},
{X86::VEXTRACTF32x4Z256rri, X86::VEXTRACTF32x4Z256mri, TB_FOLDED_STORE},
{X86::VEXTRACTF32x4Zrri, X86::VEXTRACTF32x4Zmri, TB_FOLDED_STORE},
{X86::VEXTRACTF32x8Zrri, X86::VEXTRACTF32x8Zmri, TB_FOLDED_STORE},
{X86::VEXTRACTF64x2Z256rri, X86::VEXTRACTF64x2Z256mri, TB_FOLDED_STORE},
{X86::VEXTRACTF64x2Zrri, X86::VEXTRACTF64x2Zmri, TB_FOLDED_STORE},
{X86::VEXTRACTF64x4Zrri, X86::VEXTRACTF64x4Zmri, TB_FOLDED_STORE},
{X86::VEXTRACTF32X4Z256rri, X86::VEXTRACTF32X4Z256mri, TB_FOLDED_STORE},
{X86::VEXTRACTF32X4Zrri, X86::VEXTRACTF32X4Zmri, TB_FOLDED_STORE},
{X86::VEXTRACTF32X8Zrri, X86::VEXTRACTF32X8Zmri, TB_FOLDED_STORE},
{X86::VEXTRACTF64X2Z256rri, X86::VEXTRACTF64X2Z256mri, TB_FOLDED_STORE},
{X86::VEXTRACTF64X2Zrri, X86::VEXTRACTF64X2Zmri, TB_FOLDED_STORE},
{X86::VEXTRACTF64X4Zrri, X86::VEXTRACTF64X4Zmri, TB_FOLDED_STORE},
{X86::VEXTRACTI128rri, X86::VEXTRACTI128mri, TB_FOLDED_STORE},
{X86::VEXTRACTI32x4Z256rri, X86::VEXTRACTI32x4Z256mri, TB_FOLDED_STORE},
{X86::VEXTRACTI32x4Zrri, X86::VEXTRACTI32x4Zmri, TB_FOLDED_STORE},
{X86::VEXTRACTI32x8Zrri, X86::VEXTRACTI32x8Zmri, TB_FOLDED_STORE},
{X86::VEXTRACTI64x2Z256rri, X86::VEXTRACTI64x2Z256mri, TB_FOLDED_STORE},
{X86::VEXTRACTI64x2Zrri, X86::VEXTRACTI64x2Zmri, TB_FOLDED_STORE},
{X86::VEXTRACTI64x4Zrri, X86::VEXTRACTI64x4Zmri, TB_FOLDED_STORE},
{X86::VEXTRACTI32X4Z256rri, X86::VEXTRACTI32X4Z256mri, TB_FOLDED_STORE},
{X86::VEXTRACTI32X4Zrri, X86::VEXTRACTI32X4Zmri, TB_FOLDED_STORE},
{X86::VEXTRACTI32X8Zrri, X86::VEXTRACTI32X8Zmri, TB_FOLDED_STORE},
{X86::VEXTRACTI64X2Z256rri, X86::VEXTRACTI64X2Z256mri, TB_FOLDED_STORE},
{X86::VEXTRACTI64X2Zrri, X86::VEXTRACTI64X2Zmri, TB_FOLDED_STORE},
{X86::VEXTRACTI64X4Zrri, X86::VEXTRACTI64X4Zmri, TB_FOLDED_STORE},
{X86::VEXTRACTPSZrri, X86::VEXTRACTPSZmri, TB_FOLDED_STORE},
{X86::VEXTRACTPSrri, X86::VEXTRACTPSmri, TB_FOLDED_STORE},
{X86::VMOV64toSDZrr, X86::MOV64mr, TB_FOLDED_STORE|TB_NO_REVERSE},
@ -2998,19 +2998,19 @@ static const X86FoldTableEntry Table2[] = {
{X86::VHSUBPSYrr, X86::VHSUBPSYrm, 0},
{X86::VHSUBPSrr, X86::VHSUBPSrm, 0},
{X86::VINSERTF128rri, X86::VINSERTF128rmi, 0},
{X86::VINSERTF32x4Z256rri, X86::VINSERTF32x4Z256rmi, 0},
{X86::VINSERTF32x4Zrri, X86::VINSERTF32x4Zrmi, 0},
{X86::VINSERTF32x8Zrri, X86::VINSERTF32x8Zrmi, 0},
{X86::VINSERTF64x2Z256rri, X86::VINSERTF64x2Z256rmi, 0},
{X86::VINSERTF64x2Zrri, X86::VINSERTF64x2Zrmi, 0},
{X86::VINSERTF64x4Zrri, X86::VINSERTF64x4Zrmi, 0},
{X86::VINSERTF32X4Z256rri, X86::VINSERTF32X4Z256rmi, 0},
{X86::VINSERTF32X4Zrri, X86::VINSERTF32X4Zrmi, 0},
{X86::VINSERTF32X8Zrri, X86::VINSERTF32X8Zrmi, 0},
{X86::VINSERTF64X2Z256rri, X86::VINSERTF64X2Z256rmi, 0},
{X86::VINSERTF64X2Zrri, X86::VINSERTF64X2Zrmi, 0},
{X86::VINSERTF64X4Zrri, X86::VINSERTF64X4Zrmi, 0},
{X86::VINSERTI128rri, X86::VINSERTI128rmi, 0},
{X86::VINSERTI32x4Z256rri, X86::VINSERTI32x4Z256rmi, 0},
{X86::VINSERTI32x4Zrri, X86::VINSERTI32x4Zrmi, 0},
{X86::VINSERTI32x8Zrri, X86::VINSERTI32x8Zrmi, 0},
{X86::VINSERTI64x2Z256rri, X86::VINSERTI64x2Z256rmi, 0},
{X86::VINSERTI64x2Zrri, X86::VINSERTI64x2Zrmi, 0},
{X86::VINSERTI64x4Zrri, X86::VINSERTI64x4Zrmi, 0},
{X86::VINSERTI32X4Z256rri, X86::VINSERTI32X4Z256rmi, 0},
{X86::VINSERTI32X4Zrri, X86::VINSERTI32X4Zrmi, 0},
{X86::VINSERTI32X8Zrri, X86::VINSERTI32X8Zrmi, 0},
{X86::VINSERTI64X2Z256rri, X86::VINSERTI64X2Z256rmi, 0},
{X86::VINSERTI64X2Zrri, X86::VINSERTI64X2Zrmi, 0},
{X86::VINSERTI64X4Zrri, X86::VINSERTI64X4Zrmi, 0},
{X86::VMAXCPDYrr, X86::VMAXCPDYrm, 0},
{X86::VMAXCPDZ128rr, X86::VMAXCPDZ128rm, 0},
{X86::VMAXCPDZ256rr, X86::VMAXCPDZ256rm, 0},
@ -5074,18 +5074,18 @@ static const X86FoldTableEntry Table3[] = {
{X86::VGF2P8MULBZ128rrkz, X86::VGF2P8MULBZ128rmkz, 0},
{X86::VGF2P8MULBZ256rrkz, X86::VGF2P8MULBZ256rmkz, 0},
{X86::VGF2P8MULBZrrkz, X86::VGF2P8MULBZrmkz, 0},
{X86::VINSERTF32x4Z256rrikz, X86::VINSERTF32x4Z256rmikz, 0},
{X86::VINSERTF32x4Zrrikz, X86::VINSERTF32x4Zrmikz, 0},
{X86::VINSERTF32x8Zrrikz, X86::VINSERTF32x8Zrmikz, 0},
{X86::VINSERTF64x2Z256rrikz, X86::VINSERTF64x2Z256rmikz, 0},
{X86::VINSERTF64x2Zrrikz, X86::VINSERTF64x2Zrmikz, 0},
{X86::VINSERTF64x4Zrrikz, X86::VINSERTF64x4Zrmikz, 0},
{X86::VINSERTI32x4Z256rrikz, X86::VINSERTI32x4Z256rmikz, 0},
{X86::VINSERTI32x4Zrrikz, X86::VINSERTI32x4Zrmikz, 0},
{X86::VINSERTI32x8Zrrikz, X86::VINSERTI32x8Zrmikz, 0},
{X86::VINSERTI64x2Z256rrikz, X86::VINSERTI64x2Z256rmikz, 0},
{X86::VINSERTI64x2Zrrikz, X86::VINSERTI64x2Zrmikz, 0},
{X86::VINSERTI64x4Zrrikz, X86::VINSERTI64x4Zrmikz, 0},
{X86::VINSERTF32X4Z256rrikz, X86::VINSERTF32X4Z256rmikz, 0},
{X86::VINSERTF32X4Zrrikz, X86::VINSERTF32X4Zrmikz, 0},
{X86::VINSERTF32X8Zrrikz, X86::VINSERTF32X8Zrmikz, 0},
{X86::VINSERTF64X2Z256rrikz, X86::VINSERTF64X2Z256rmikz, 0},
{X86::VINSERTF64X2Zrrikz, X86::VINSERTF64X2Zrmikz, 0},
{X86::VINSERTF64X4Zrrikz, X86::VINSERTF64X4Zrmikz, 0},
{X86::VINSERTI32X4Z256rrikz, X86::VINSERTI32X4Z256rmikz, 0},
{X86::VINSERTI32X4Zrrikz, X86::VINSERTI32X4Zrmikz, 0},
{X86::VINSERTI32X8Zrrikz, X86::VINSERTI32X8Zrmikz, 0},
{X86::VINSERTI64X2Z256rrikz, X86::VINSERTI64X2Z256rmikz, 0},
{X86::VINSERTI64X2Zrrikz, X86::VINSERTI64X2Zrmikz, 0},
{X86::VINSERTI64X4Zrrikz, X86::VINSERTI64X4Zrmikz, 0},
{X86::VMAXCPDZ128rrkz, X86::VMAXCPDZ128rmkz, 0},
{X86::VMAXCPDZ256rrkz, X86::VMAXCPDZ256rmkz, 0},
{X86::VMAXCPDZrrkz, X86::VMAXCPDZrmkz, 0},
@ -6696,18 +6696,18 @@ static const X86FoldTableEntry Table4[] = {
{X86::VGF2P8MULBZ128rrk, X86::VGF2P8MULBZ128rmk, 0},
{X86::VGF2P8MULBZ256rrk, X86::VGF2P8MULBZ256rmk, 0},
{X86::VGF2P8MULBZrrk, X86::VGF2P8MULBZrmk, 0},
{X86::VINSERTF32x4Z256rrik, X86::VINSERTF32x4Z256rmik, 0},
{X86::VINSERTF32x4Zrrik, X86::VINSERTF32x4Zrmik, 0},
{X86::VINSERTF32x8Zrrik, X86::VINSERTF32x8Zrmik, 0},
{X86::VINSERTF64x2Z256rrik, X86::VINSERTF64x2Z256rmik, 0},
{X86::VINSERTF64x2Zrrik, X86::VINSERTF64x2Zrmik, 0},
{X86::VINSERTF64x4Zrrik, X86::VINSERTF64x4Zrmik, 0},
{X86::VINSERTI32x4Z256rrik, X86::VINSERTI32x4Z256rmik, 0},
{X86::VINSERTI32x4Zrrik, X86::VINSERTI32x4Zrmik, 0},
{X86::VINSERTI32x8Zrrik, X86::VINSERTI32x8Zrmik, 0},
{X86::VINSERTI64x2Z256rrik, X86::VINSERTI64x2Z256rmik, 0},
{X86::VINSERTI64x2Zrrik, X86::VINSERTI64x2Zrmik, 0},
{X86::VINSERTI64x4Zrrik, X86::VINSERTI64x4Zrmik, 0},
{X86::VINSERTF32X4Z256rrik, X86::VINSERTF32X4Z256rmik, 0},
{X86::VINSERTF32X4Zrrik, X86::VINSERTF32X4Zrmik, 0},
{X86::VINSERTF32X8Zrrik, X86::VINSERTF32X8Zrmik, 0},
{X86::VINSERTF64X2Z256rrik, X86::VINSERTF64X2Z256rmik, 0},
{X86::VINSERTF64X2Zrrik, X86::VINSERTF64X2Zrmik, 0},
{X86::VINSERTF64X4Zrrik, X86::VINSERTF64X4Zrmik, 0},
{X86::VINSERTI32X4Z256rrik, X86::VINSERTI32X4Z256rmik, 0},
{X86::VINSERTI32X4Zrrik, X86::VINSERTI32X4Zrmik, 0},
{X86::VINSERTI32X8Zrrik, X86::VINSERTI32X8Zrmik, 0},
{X86::VINSERTI64X2Z256rrik, X86::VINSERTI64X2Z256rmik, 0},
{X86::VINSERTI64X2Zrrik, X86::VINSERTI64X2Zrmik, 0},
{X86::VINSERTI64X4Zrrik, X86::VINSERTI64X4Zrmik, 0},
{X86::VMAXCPDZ128rrk, X86::VMAXCPDZ128rmk, 0},
{X86::VMAXCPDZ256rrk, X86::VMAXCPDZ256rmk, 0},
{X86::VMAXCPDZrrk, X86::VMAXCPDZrmk, 0},

32
llvm/test/TableGen/x86-instr-mapping.inc
View File

@ -360,14 +360,14 @@ static const X86TableEntry X86CompressEVEXTable[] = {
{ X86::VDIVSSZrm_Int, X86::VDIVSSrm_Int },
{ X86::VDIVSSZrr, X86::VDIVSSrr },
{ X86::VDIVSSZrr_Int, X86::VDIVSSrr_Int },
{ X86::VEXTRACTF32x4Z256mri, X86::VEXTRACTF128mri },
{ X86::VEXTRACTF32x4Z256rri, X86::VEXTRACTF128rri },
{ X86::VEXTRACTF64x2Z256mri, X86::VEXTRACTF128mri },
{ X86::VEXTRACTF64x2Z256rri, X86::VEXTRACTF128rri },
{ X86::VEXTRACTI32x4Z256mri, X86::VEXTRACTI128mri },
{ X86::VEXTRACTI32x4Z256rri, X86::VEXTRACTI128rri },
{ X86::VEXTRACTI64x2Z256mri, X86::VEXTRACTI128mri },
{ X86::VEXTRACTI64x2Z256rri, X86::VEXTRACTI128rri },
{ X86::VEXTRACTF32X4Z256mri, X86::VEXTRACTF128mri },
{ X86::VEXTRACTF32X4Z256rri, X86::VEXTRACTF128rri },
{ X86::VEXTRACTF64X2Z256mri, X86::VEXTRACTF128mri },
{ X86::VEXTRACTF64X2Z256rri, X86::VEXTRACTF128rri },
{ X86::VEXTRACTI32X4Z256mri, X86::VEXTRACTI128mri },
{ X86::VEXTRACTI32X4Z256rri, X86::VEXTRACTI128rri },
{ X86::VEXTRACTI64X2Z256mri, X86::VEXTRACTI128mri },
{ X86::VEXTRACTI64X2Z256rri, X86::VEXTRACTI128rri },
{ X86::VEXTRACTPSZmri, X86::VEXTRACTPSmri },
{ X86::VEXTRACTPSZrri, X86::VEXTRACTPSrri },
{ X86::VFMADD132PDZ128m, X86::VFMADD132PDm },
@ -622,14 +622,14 @@ static const X86TableEntry X86CompressEVEXTable[] = {
{ X86::VGF2P8MULBZ128rr, X86::VGF2P8MULBrr },
{ X86::VGF2P8MULBZ256rm, X86::VGF2P8MULBYrm },
{ X86::VGF2P8MULBZ256rr, X86::VGF2P8MULBYrr },
{ X86::VINSERTF32x4Z256rmi, X86::VINSERTF128rmi },
{ X86::VINSERTF32x4Z256rri, X86::VINSERTF128rri },
{ X86::VINSERTF64x2Z256rmi, X86::VINSERTF128rmi },
{ X86::VINSERTF64x2Z256rri, X86::VINSERTF128rri },
{ X86::VINSERTI32x4Z256rmi, X86::VINSERTI128rmi },
{ X86::VINSERTI32x4Z256rri, X86::VINSERTI128rri },
{ X86::VINSERTI64x2Z256rmi, X86::VINSERTI128rmi },
{ X86::VINSERTI64x2Z256rri, X86::VINSERTI128rri },
{ X86::VINSERTF32X4Z256rmi, X86::VINSERTF128rmi },
{ X86::VINSERTF32X4Z256rri, X86::VINSERTF128rri },
{ X86::VINSERTF64X2Z256rmi, X86::VINSERTF128rmi },
{ X86::VINSERTF64X2Z256rri, X86::VINSERTF128rri },
{ X86::VINSERTI32X4Z256rmi, X86::VINSERTI128rmi },
{ X86::VINSERTI32X4Z256rri, X86::VINSERTI128rri },
{ X86::VINSERTI64X2Z256rmi, X86::VINSERTI128rmi },
{ X86::VINSERTI64X2Z256rri, X86::VINSERTI128rri },
{ X86::VINSERTPSZrmi, X86::VINSERTPSrmi },
{ X86::VINSERTPSZrri, X86::VINSERTPSrri },
{ X86::VMAXCPDZ128rm, X86::VMAXCPDrm },

24
llvm/utils/TableGen/X86ManualFoldTables.def
View File

@ -31,18 +31,18 @@ NOFOLD(VCOMPRESSPSZrrk)
NOFOLD(VCVTPS2PHZ128rrk)
NOFOLD(VCVTPS2PHZ256rrk)
NOFOLD(VCVTPS2PHZrrk)
NOFOLD(VEXTRACTF32x4Z256rrik)
NOFOLD(VEXTRACTF32x4Zrrik)
NOFOLD(VEXTRACTF32x8Zrrik)
NOFOLD(VEXTRACTF64x2Z256rrik)
NOFOLD(VEXTRACTF64x2Zrrik)
NOFOLD(VEXTRACTF64x4Zrrik)
NOFOLD(VEXTRACTI32x4Z256rrik)
NOFOLD(VEXTRACTI32x4Zrrik)
NOFOLD(VEXTRACTI32x8Zrrik)
NOFOLD(VEXTRACTI64x2Z256rrik)
NOFOLD(VEXTRACTI64x2Zrrik)
NOFOLD(VEXTRACTI64x4Zrrik)
NOFOLD(VEXTRACTF32X4Z256rrik)
NOFOLD(VEXTRACTF32X4Zrrik)
NOFOLD(VEXTRACTF32X8Zrrik)
NOFOLD(VEXTRACTF64X2Z256rrik)
NOFOLD(VEXTRACTF64X2Zrrik)
NOFOLD(VEXTRACTF64X4Zrrik)
NOFOLD(VEXTRACTI32X4Z256rrik)
NOFOLD(VEXTRACTI32X4Zrrik)
NOFOLD(VEXTRACTI32X8Zrrik)
NOFOLD(VEXTRACTI64X2Z256rrik)
NOFOLD(VEXTRACTI64X2Zrrik)
NOFOLD(VEXTRACTI64X4Zrrik)
NOFOLD(VMOVAPDZ128mrk)
NOFOLD(VMOVAPDZ256mrk)
NOFOLD(VMOVAPDZmrk)

16
llvm/utils/TableGen/X86ManualInstrMapping.def
View File

@ -246,14 +246,14 @@ ENTRY(VCVTTPD2DQZ256rm, VCVTTPD2DQYrm)
ENTRY(VCVTTPD2DQZ256rr, VCVTTPD2DQYrr)
ENTRY(VDIVPDZ256rm, VDIVPDYrm)
ENTRY(VDIVPDZ256rr, VDIVPDYrr)
ENTRY(VEXTRACTF64x2Z256mri, VEXTRACTF128mri)
ENTRY(VEXTRACTF64x2Z256rri, VEXTRACTF128rri)
ENTRY(VEXTRACTI64x2Z256mri, VEXTRACTI128mri)
ENTRY(VEXTRACTI64x2Z256rri, VEXTRACTI128rri)
ENTRY(VINSERTF64x2Z256rmi, VINSERTF128rmi)
ENTRY(VINSERTF64x2Z256rri, VINSERTF128rri)
ENTRY(VINSERTI64x2Z256rmi, VINSERTI128rmi)
ENTRY(VINSERTI64x2Z256rri, VINSERTI128rri)
ENTRY(VEXTRACTF64X2Z256mri, VEXTRACTF128mri)
ENTRY(VEXTRACTF64X2Z256rri, VEXTRACTF128rri)
ENTRY(VEXTRACTI64X2Z256mri, VEXTRACTI128mri)
ENTRY(VEXTRACTI64X2Z256rri, VEXTRACTI128rri)
ENTRY(VINSERTF64X2Z256rmi, VINSERTF128rmi)
ENTRY(VINSERTF64X2Z256rri, VINSERTF128rri)
ENTRY(VINSERTI64X2Z256rmi, VINSERTI128rmi)
ENTRY(VINSERTI64X2Z256rri, VINSERTI128rri)
ENTRY(VMAXCPDZ256rm, VMAXCPDYrm)
ENTRY(VMAXCPDZ256rr, VMAXCPDYrr)
ENTRY(VMAXPDZ256rm, VMAXPDYrm)