* [Bug target/100885] [12 Regression] ICE: in extract_constrain_insn, at recog.c:2671: insn does not satisfy its constraints: {sse4_1_zero_extendv8qiv8hi2}
2021-06-02 20:51 [Bug target/100885] New: [12 Regression] ICE: in extract_constrain_insn, at recog.c:2671: insn does not satisfy its constraints: {sse4_1_zero_extendv8qiv8hi2} zsojka at seznam dot cz
@ 2021-06-03 3:33 ` crazylht at gmail dot com
2021-06-03 3:49 ` crazylht at gmail dot com
` (9 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: crazylht at gmail dot com @ 2021-06-03 3:33 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100885
--- Comment #1 from Hongtao.liu <crazylht at gmail dot com> ---
152(define_register_constraint "Yw"
153 "TARGET_AVX512BW && TARGET_AVX512VL ? ALL_SSE_REGS : TARGET_SSE ? SSE_REGS
: NO_REGS"
154 "@internal Any EVEX encodable SSE register (@code{%xmm0-%xmm31}) for
AVX512BW with TARGET_AVX512VL target, otherwise any \
SSE register.")
19076(define_insn "sse4_1_<code>v8qiv8hi2<mask_name>"
19077 [(set (match_operand:V8HI 0 "register_operand" "=Yr,*x,Yw")
19078 (any_extend:V8HI
19079 (vec_select:V8QI
19080 (match_operand:V16QI 1 "register_operand" "Yr,*x,Yw")
19081 (parallel [(const_int 0) (const_int 1)
19082 (const_int 2) (const_int 3)
19083 (const_int 4) (const_int 5)
19084 (const_int 6) (const_int 7)]))))]
19085 "TARGET_SSE4_1 && <mask_avx512bw_condition> &&
<mask_avx512vl_condition>"
19086 "%vpmov<extsuffix>bw\t{%1, %0<mask_operand2>|%0<mask_operand2>, %1}"
19087 [(set_attr "isa" "noavx,noavx,avx")
19088 (set_attr "type" "ssemov")
19089 (set_attr "prefix_extra" "1")
19090 (set_attr "prefix" "orig,orig,maybe_evex")
19091 (set_attr "mode" "TI")])
With avx512vl Yw should be matched, and w/o avx512bw, only SSE_REGS should be
matched, why xmm16 is allocated?
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug target/100885] [12 Regression] ICE: in extract_constrain_insn, at recog.c:2671: insn does not satisfy its constraints: {sse4_1_zero_extendv8qiv8hi2}
2021-06-02 20:51 [Bug target/100885] New: [12 Regression] ICE: in extract_constrain_insn, at recog.c:2671: insn does not satisfy its constraints: {sse4_1_zero_extendv8qiv8hi2} zsojka at seznam dot cz
2021-06-03 3:33 ` [Bug target/100885] " crazylht at gmail dot com
@ 2021-06-03 3:49 ` crazylht at gmail dot com
2021-06-03 5:10 ` crazylht at gmail dot com
` (8 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: crazylht at gmail dot com @ 2021-06-03 3:49 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100885
--- Comment #2 from Hongtao.liu <crazylht at gmail dot com> ---
> With avx512vl Yw should be matched, and w/o avx512bw, only SSE_REGS should
> be matched, why xmm16 is allocated?
It didn't come from RA, but post_reload splitter.
18103(define_insn_and_split "*sse4_1_zero_extendv8qiv8hi2_3"
18104 [(set (match_operand:V16QI 0 "register_operand" "=Yr,*x,v")
18105 (vec_select:V16QI
18106 (vec_concat:V32QI
18107 (match_operand:V16QI 1 "vector_operand" "YrBm,*xBm,vm")
18108 (match_operand:V16QI 2 "const0_operand" "C,C,C"))
18109 (match_parallel 3 "pmovzx_parallel"
18110 [(match_operand 4 "const_int_operand" "n,n,n")])))]
18111 "TARGET_SSE4_1"
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug target/100885] [12 Regression] ICE: in extract_constrain_insn, at recog.c:2671: insn does not satisfy its constraints: {sse4_1_zero_extendv8qiv8hi2}
2021-06-02 20:51 [Bug target/100885] New: [12 Regression] ICE: in extract_constrain_insn, at recog.c:2671: insn does not satisfy its constraints: {sse4_1_zero_extendv8qiv8hi2} zsojka at seznam dot cz
2021-06-03 3:33 ` [Bug target/100885] " crazylht at gmail dot com
2021-06-03 3:49 ` crazylht at gmail dot com
@ 2021-06-03 5:10 ` crazylht at gmail dot com
2021-06-03 7:14 ` crazylht at gmail dot com
` (7 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: crazylht at gmail dot com @ 2021-06-03 5:10 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100885
--- Comment #3 from Hongtao.liu <crazylht at gmail dot com> ---
(In reply to Hongtao.liu from comment #2)
> > With avx512vl Yw should be matched, and w/o avx512bw, only SSE_REGS should
> > be matched, why xmm16 is allocated?
>
> It didn't come from RA, but post_reload splitter.
>
> 18103(define_insn_and_split "*sse4_1_zero_extendv8qiv8hi2_3"
> 18104 [(set (match_operand:V16QI 0 "register_operand" "=Yr,*x,v")
> 18105 (vec_select:V16QI
> 18106 (vec_concat:V32QI
> 18107 (match_operand:V16QI 1 "vector_operand" "YrBm,*xBm,vm")
> 18108 (match_operand:V16QI 2 "const0_operand" "C,C,C"))
> 18109 (match_parallel 3 "pmovzx_parallel"
> 18110 [(match_operand 4 "const_int_operand" "n,n,n")])))]
> 18111 "TARGET_SSE4_1"
Need to adjust the third constraint from (v, vm, C, n) to (Yw, Ywm, C, n)
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug target/100885] [12 Regression] ICE: in extract_constrain_insn, at recog.c:2671: insn does not satisfy its constraints: {sse4_1_zero_extendv8qiv8hi2}
2021-06-02 20:51 [Bug target/100885] New: [12 Regression] ICE: in extract_constrain_insn, at recog.c:2671: insn does not satisfy its constraints: {sse4_1_zero_extendv8qiv8hi2} zsojka at seznam dot cz
` (2 preceding siblings ...)
2021-06-03 5:10 ` crazylht at gmail dot com
@ 2021-06-03 7:14 ` crazylht at gmail dot com
2021-06-03 10:40 ` crazylht at gmail dot com
` (6 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: crazylht at gmail dot com @ 2021-06-03 7:14 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100885
--- Comment #4 from Hongtao.liu <crazylht at gmail dot com> ---
> Need to adjust the third constraint from (v, vm, C, n) to (Yw, Ywm, C, n)
Also for those extend instructions which need AVX512VL is needed for
xmm16-xmm32.
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug target/100885] [12 Regression] ICE: in extract_constrain_insn, at recog.c:2671: insn does not satisfy its constraints: {sse4_1_zero_extendv8qiv8hi2}
2021-06-02 20:51 [Bug target/100885] New: [12 Regression] ICE: in extract_constrain_insn, at recog.c:2671: insn does not satisfy its constraints: {sse4_1_zero_extendv8qiv8hi2} zsojka at seznam dot cz
` (3 preceding siblings ...)
2021-06-03 7:14 ` crazylht at gmail dot com
@ 2021-06-03 10:40 ` crazylht at gmail dot com
2021-06-07 3:28 ` cvs-commit at gcc dot gnu.org
` (5 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: crazylht at gmail dot com @ 2021-06-03 10:40 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100885
--- Comment #5 from Hongtao.liu <crazylht at gmail dot com> ---
A patch is posted at
https://gcc.gnu.org/pipermail/gcc-patches/2021-June/571815.html
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug target/100885] [12 Regression] ICE: in extract_constrain_insn, at recog.c:2671: insn does not satisfy its constraints: {sse4_1_zero_extendv8qiv8hi2}
2021-06-02 20:51 [Bug target/100885] New: [12 Regression] ICE: in extract_constrain_insn, at recog.c:2671: insn does not satisfy its constraints: {sse4_1_zero_extendv8qiv8hi2} zsojka at seznam dot cz
` (4 preceding siblings ...)
2021-06-03 10:40 ` crazylht at gmail dot com
@ 2021-06-07 3:28 ` cvs-commit at gcc dot gnu.org
2021-06-07 4:38 ` cvs-commit at gcc dot gnu.org
` (4 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: cvs-commit at gcc dot gnu.org @ 2021-06-07 3:28 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100885
--- Comment #6 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by hongtao Liu <liuhongt@gcc.gnu.org>:
https://gcc.gnu.org/g:be5efe9c12cb852c788f74f8555e6ab8d755479b
commit r12-1254-gbe5efe9c12cb852c788f74f8555e6ab8d755479b
Author: liuhongt <hongtao.liu@intel.com>
Date: Thu Jun 3 16:38:32 2021 +0800
Fix ICE of insn does not satisfy its constraints.
evex encoding vpmovzxbx needs both AVX512BW and AVX512VL which means
constraint "Yw" should be used instead of constraint "v".
gcc/ChangeLog:
PR target/100885
* config/i386/sse.md (*sse4_1_zero_extendv8qiv8hi2_3): Refine
constraints.
(<insn>v4siv4di2): Delete constraints for define_expand.
gcc/testsuite/ChangeLog:
PR target/100885
* g++.target/i386/pr100885.C: New test.
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug target/100885] [12 Regression] ICE: in extract_constrain_insn, at recog.c:2671: insn does not satisfy its constraints: {sse4_1_zero_extendv8qiv8hi2}
2021-06-02 20:51 [Bug target/100885] New: [12 Regression] ICE: in extract_constrain_insn, at recog.c:2671: insn does not satisfy its constraints: {sse4_1_zero_extendv8qiv8hi2} zsojka at seznam dot cz
` (5 preceding siblings ...)
2021-06-07 3:28 ` cvs-commit at gcc dot gnu.org
@ 2021-06-07 4:38 ` cvs-commit at gcc dot gnu.org
2021-06-07 4:41 ` crazylht at gmail dot com
` (3 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: cvs-commit at gcc dot gnu.org @ 2021-06-07 4:38 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100885
--- Comment #7 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-11 branch has been updated by hongtao Liu
<liuhongt@gcc.gnu.org>:
https://gcc.gnu.org/g:c064e787b10069e3de56bd3d0d1a34a1a09086ea
commit r11-8517-gc064e787b10069e3de56bd3d0d1a34a1a09086ea
Author: liuhongt <hongtao.liu@intel.com>
Date: Thu Jun 3 16:38:32 2021 +0800
Fix ICE of insn does not satisfy its constraints.
evex encoding vpmovzxbx needs both AVX512BW and AVX512VL which means
constraint "Yw" should be used instead of constraint "v".
gcc/ChangeLog:
PR target/100885
* config/i386/sse.md (*sse4_1_zero_extendv8qiv8hi2_3): Refine
constraints.
(<insn>v4siv4di2): Delete constraints for define_expand.
gcc/testsuite/ChangeLog:
PR target/100885
* g++.target/i386/pr100885.C: New test.
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug target/100885] [12 Regression] ICE: in extract_constrain_insn, at recog.c:2671: insn does not satisfy its constraints: {sse4_1_zero_extendv8qiv8hi2}
2021-06-02 20:51 [Bug target/100885] New: [12 Regression] ICE: in extract_constrain_insn, at recog.c:2671: insn does not satisfy its constraints: {sse4_1_zero_extendv8qiv8hi2} zsojka at seznam dot cz
` (6 preceding siblings ...)
2021-06-07 4:38 ` cvs-commit at gcc dot gnu.org
@ 2021-06-07 4:41 ` crazylht at gmail dot com
2021-06-07 9:04 ` rguenth at gcc dot gnu.org
` (2 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: crazylht at gmail dot com @ 2021-06-07 4:41 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100885
--- Comment #8 from Hongtao.liu <crazylht at gmail dot com> ---
Fixed in trunk.
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug target/100885] [12 Regression] ICE: in extract_constrain_insn, at recog.c:2671: insn does not satisfy its constraints: {sse4_1_zero_extendv8qiv8hi2}
2021-06-02 20:51 [Bug target/100885] New: [12 Regression] ICE: in extract_constrain_insn, at recog.c:2671: insn does not satisfy its constraints: {sse4_1_zero_extendv8qiv8hi2} zsojka at seznam dot cz
` (7 preceding siblings ...)
2021-06-07 4:41 ` crazylht at gmail dot com
@ 2021-06-07 9:04 ` rguenth at gcc dot gnu.org
2021-06-07 14:49 ` cvs-commit at gcc dot gnu.org
2021-06-07 14:59 ` cvs-commit at gcc dot gnu.org
10 siblings, 0 replies; 12+ messages in thread
From: rguenth at gcc dot gnu.org @ 2021-06-07 9:04 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100885
Richard Biener <rguenth at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Target Milestone|--- |12.0
Resolution|--- |FIXED
Status|UNCONFIRMED |RESOLVED
--- Comment #9 from Richard Biener <rguenth at gcc dot gnu.org> ---
Fixed.
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug target/100885] [12 Regression] ICE: in extract_constrain_insn, at recog.c:2671: insn does not satisfy its constraints: {sse4_1_zero_extendv8qiv8hi2}
2021-06-02 20:51 [Bug target/100885] New: [12 Regression] ICE: in extract_constrain_insn, at recog.c:2671: insn does not satisfy its constraints: {sse4_1_zero_extendv8qiv8hi2} zsojka at seznam dot cz
` (8 preceding siblings ...)
2021-06-07 9:04 ` rguenth at gcc dot gnu.org
@ 2021-06-07 14:49 ` cvs-commit at gcc dot gnu.org
2021-06-07 14:59 ` cvs-commit at gcc dot gnu.org
10 siblings, 0 replies; 12+ messages in thread
From: cvs-commit at gcc dot gnu.org @ 2021-06-07 14:49 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100885
--- Comment #10 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by H.J. Lu <hjl@gcc.gnu.org>:
https://gcc.gnu.org/g:52730540e83c81ee595a51feb7736ff753c98139
commit r12-1261-g52730540e83c81ee595a51feb7736ff753c98139
Author: H.J. Lu <hjl.tools@gmail.com>
Date: Mon Jun 7 07:29:31 2021 -0700
x86: Update g++.target/i386/pr100885.C
Since long is 32 bits for x32, update g++.target/i386/pr100885.C to cast
__m64 to long long for x32.
PR target/100885
* g++.target/i386/pr100885.C (_mm_set_epi64): Cast __m64 to long
long.
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug target/100885] [12 Regression] ICE: in extract_constrain_insn, at recog.c:2671: insn does not satisfy its constraints: {sse4_1_zero_extendv8qiv8hi2}
2021-06-02 20:51 [Bug target/100885] New: [12 Regression] ICE: in extract_constrain_insn, at recog.c:2671: insn does not satisfy its constraints: {sse4_1_zero_extendv8qiv8hi2} zsojka at seznam dot cz
` (9 preceding siblings ...)
2021-06-07 14:49 ` cvs-commit at gcc dot gnu.org
@ 2021-06-07 14:59 ` cvs-commit at gcc dot gnu.org
10 siblings, 0 replies; 12+ messages in thread
From: cvs-commit at gcc dot gnu.org @ 2021-06-07 14:59 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100885
--- Comment #11 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-11 branch has been updated by H.J. Lu <hjl@gcc.gnu.org>:
https://gcc.gnu.org/g:bcb5e97899e324c092806b49256a515593268a54
commit r11-8518-gbcb5e97899e324c092806b49256a515593268a54
Author: H.J. Lu <hjl.tools@gmail.com>
Date: Mon Jun 7 07:29:31 2021 -0700
x86: Update g++.target/i386/pr100885.C
Since long is 32 bits for x32, update g++.target/i386/pr100885.C to cast
__m64 to long long for x32.
PR target/100885
* g++.target/i386/pr100885.C (_mm_set_epi64): Cast __m64 to long
long.
(cherry picked from commit 52730540e83c81ee595a51feb7736ff753c98139)
^ permalink raw reply [flat|nested] 12+ messages in thread