diff options
author | Hongyu Wang <hongyu.wang@intel.com> | 2023-11-09 13:11:41 +0800 |
---|---|---|
committer | Hongyu Wang <hongyu.wang@intel.com> | 2023-11-16 09:58:54 +0800 |
commit | 9251db0dabc8e75c31b49a5b589124e9a2bc8299 (patch) | |
tree | 46f3dc7fdadf225ac4e074f1280765d02868a7e7 /gcc | |
parent | 17cc2e57f41d88a0a828b4239b5ce852e8ec1406 (diff) | |
download | gcc-9251db0dabc8e75c31b49a5b589124e9a2bc8299.zip gcc-9251db0dabc8e75c31b49a5b589124e9a2bc8299.tar.gz gcc-9251db0dabc8e75c31b49a5b589124e9a2bc8299.tar.bz2 |
[i386] APX: Fix EGPR usage in several patterns.
For vextract/insert{if}128 they cannot adopt EGPR in their memory operand, all
related pattern should be adjusted to disable EGPR usage on them.
Also fix a wrong gpr16 attr for insertps.
gcc/ChangeLog:
* config/i386/sse.md (vec_extract_hi_<mode>): Add noavx512vl
alternative with attr addr gpr16 and "jm" constraint.
(vec_extract_hi_<mode>): Likewise for SF vector modes.
(@vec_extract_hi_<mode>): Likewise.
(*vec_extractv2ti): Likewise.
(vec_set_hi_<mode><mask_name>): Likewise.
* config/i386/mmx.md (@sse4_1_insertps_<mode>): Correct gpr16 attr for
each alternative.
Diffstat (limited to 'gcc')
-rw-r--r-- | gcc/config/i386/mmx.md | 2 | ||||
-rw-r--r-- | gcc/config/i386/sse.md | 32 |
2 files changed, 21 insertions, 13 deletions
diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md index a3d08bb..3555387 100644 --- a/gcc/config/i386/mmx.md +++ b/gcc/config/i386/mmx.md @@ -1215,7 +1215,7 @@ } } [(set_attr "isa" "noavx,noavx,avx") - (set_attr "addr" "*,*,gpr16") + (set_attr "addr" "gpr16,gpr16,*") (set_attr "type" "sselog") (set_attr "prefix_data16" "1,1,*") (set_attr "prefix_extra" "1") diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md index af482f2..d250a6c 100644 --- a/gcc/config/i386/sse.md +++ b/gcc/config/i386/sse.md @@ -12049,9 +12049,9 @@ (set_attr "mode" "<sseinsnmode>")]) (define_insn "vec_extract_hi_<mode>" - [(set (match_operand:<ssehalfvecmode> 0 "nonimmediate_operand" "=vm") + [(set (match_operand:<ssehalfvecmode> 0 "nonimmediate_operand" "=xjm,vm") (vec_select:<ssehalfvecmode> - (match_operand:VI8F_256 1 "register_operand" "v") + (match_operand:VI8F_256 1 "register_operand" "x,v") (parallel [(const_int 2) (const_int 3)])))] "TARGET_AVX" { @@ -12065,7 +12065,9 @@ else return "vextract<i128>\t{$0x1, %1, %0|%0, %1, 0x1}"; } - [(set_attr "type" "sselog1") + [(set_attr "isa" "noavx512vl,avx512vl") + (set_attr "addr" "gpr16,*") + (set_attr "type" "sselog1") (set_attr "prefix_extra" "1") (set_attr "length_immediate" "1") (set_attr "prefix" "vex") @@ -12132,7 +12134,7 @@ (set_attr "mode" "<sseinsnmode>")]) (define_insn "vec_extract_hi_<mode>" - [(set (match_operand:<ssehalfvecmode> 0 "nonimmediate_operand" "=xm, vm") + [(set (match_operand:<ssehalfvecmode> 0 "nonimmediate_operand" "=xjm, vm") (vec_select:<ssehalfvecmode> (match_operand:VI4F_256 1 "register_operand" "x, v") (parallel [(const_int 4) (const_int 5) @@ -12141,7 +12143,8 @@ "@ vextract<i128>\t{$0x1, %1, %0|%0, %1, 0x1} vextract<shuffletype>32x4\t{$0x1, %1, %0|%0, %1, 0x1}" - [(set_attr "isa" "*, avx512vl") + [(set_attr "isa" "noavx512vl, avx512vl") + (set_attr "addr" "gpr16,*") (set_attr "prefix" "vex, evex") (set_attr "type" "sselog1") (set_attr "length_immediate" "1") @@ -12222,7 +12225,7 @@ "operands[1] = gen_lowpart (<ssehalfvecmode>mode, operands[1]);") (define_insn "@vec_extract_hi_<mode>" - [(set (match_operand:<ssehalfvecmode> 0 "nonimmediate_operand" "=xm,vm") + [(set (match_operand:<ssehalfvecmode> 0 "nonimmediate_operand" "=xjm,vm") (vec_select:<ssehalfvecmode> (match_operand:V16_256 1 "register_operand" "x,v") (parallel [(const_int 8) (const_int 9) @@ -12236,7 +12239,8 @@ [(set_attr "type" "sselog1") (set_attr "prefix_extra" "1") (set_attr "length_immediate" "1") - (set_attr "isa" "*,avx512vl") + (set_attr "isa" "noavx512vl,avx512vl") + (set_attr "addr" "gpr16,*") (set_attr "prefix" "vex,evex") (set_attr "mode" "OI")]) @@ -20465,7 +20469,7 @@ }) (define_insn "*vec_extractv2ti" - [(set (match_operand:TI 0 "nonimmediate_operand" "=xm,vm") + [(set (match_operand:TI 0 "nonimmediate_operand" "=xjm,vm") (vec_select:TI (match_operand:V2TI 1 "register_operand" "x,v") (parallel @@ -20477,6 +20481,8 @@ [(set_attr "type" "sselog") (set_attr "prefix_extra" "1") (set_attr "length_immediate" "1") + (set_attr "isa" "noavx512vl,avx512vl") + (set_attr "addr" "gpr16,*") (set_attr "prefix" "vex,evex") (set_attr "mode" "OI")]) @@ -27556,12 +27562,12 @@ (set_attr "mode" "<sseinsnmode>")]) (define_insn "vec_set_hi_<mode><mask_name>" - [(set (match_operand:VI8F_256 0 "register_operand" "=v") + [(set (match_operand:VI8F_256 0 "register_operand" "=x,v") (vec_concat:VI8F_256 (vec_select:<ssehalfvecmode> - (match_operand:VI8F_256 1 "register_operand" "v") + (match_operand:VI8F_256 1 "register_operand" "x,v") (parallel [(const_int 0) (const_int 1)])) - (match_operand:<ssehalfvecmode> 2 "nonimmediate_operand" "vm")))] + (match_operand:<ssehalfvecmode> 2 "nonimmediate_operand" "xjm,vm")))] "TARGET_AVX && <mask_avx512dq_condition>" { if (TARGET_AVX512DQ) @@ -27571,7 +27577,9 @@ else return "vinsert<i128>\t{$0x1, %2, %1, %0|%0, %1, %2, 0x1}"; } - [(set_attr "type" "sselog") + [(set_attr "isa" "noavx512vl,avx512vl") + (set_attr "addr" "gpr16,*") + (set_attr "type" "sselog") (set_attr "prefix_extra" "1") (set_attr "length_immediate" "1") (set_attr "prefix" "vex") |