From: Jakub Jelinek Date: Mon, 10 Feb 2020 21:44:40 +0000 (+0100) Subject: i386: Fix -mavx -mno-mavx2 ICE with VEC_COND_EXPR [PR93637] X-Git-Tag: releases/gcc-9.3.0~138 X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=b7cbce7a174292adc7c9d6db81bba6922a591d69;p=thirdparty%2Fgcc.git i386: Fix -mavx -mno-mavx2 ICE with VEC_COND_EXPR [PR93637] As mentioned in the PR, for -mavx -mno-avx2 the backend does support vcondv4div4df and vcondv8siv8sf optabs (while generally 32-byte vectors aren't much supported in that case, it is performed using vandps/vandnps/vorps). The problem is that after the last generic vector lowering (where the VEC_COND_EXPR still compares two V4DF vectors and has two V4DI last operands and V4DI result and so is considered ok) fre4 folds the condition into constant, at which point the middle-end during expansion will try vcond_mask_optab and fall back to trying to expand it as the constant vector < 0 vcondv4div4di, but neither of them is supported for -mavx -mno-avx2 and thus we ICE. So, the options I see is either what the following patch does, also support vcond_mask_v4div4di and vcond_mask_v4siv4si already for TARGET_AVX, or require for vcondv4div4df and vcondv8siv8sf TARGET_AVX2 rather than current TARGET_AVX. 2020-02-10 Jakub Jelinek PR target/93637 * config/i386/sse.md (VI_256_AVX2): New mode iterator. (vcond_mask_): Use it instead of VI_256. Change condition from TARGET_AVX2 to TARGET_AVX. * gcc.target/i386/avx-pr93637.c: New test. --- diff --git a/gcc/ChangeLog b/gcc/ChangeLog index 72c8ee6bd677..34091354380c 100644 --- a/gcc/ChangeLog +++ b/gcc/ChangeLog @@ -1,6 +1,13 @@ 2020-02-13 Jakub Jelinek Backported from mainline + 2020-02-10 Jakub Jelinek + + PR target/93637 + * config/i386/sse.md (VI_256_AVX2): New mode iterator. + (vcond_mask_): Use it instead of VI_256. + Change condition from TARGET_AVX2 to TARGET_AVX. + 2020-02-08 Uroš Bizjak Jakub Jelinek diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md index 659cbff22f37..c0fe0eefd50d 100644 --- a/gcc/config/i386/sse.md +++ b/gcc/config/i386/sse.md @@ -3180,13 +3180,19 @@ (match_operand: 3 "register_operand")))] "TARGET_AVX512BW") +;; As vcondv4div4df and vcondv8siv8sf are enabled already with TARGET_AVX, +;; and their condition can be folded late into a constant, we need to +;; support vcond_mask_v4div4di and vcond_mask_v8siv8si for TARGET_AVX. +(define_mode_iterator VI_256_AVX2 [(V32QI "TARGET_AVX2") (V16HI "TARGET_AVX2") + V8SI V4DI]) + (define_expand "vcond_mask_" - [(set (match_operand:VI_256 0 "register_operand") - (vec_merge:VI_256 - (match_operand:VI_256 1 "nonimmediate_operand") - (match_operand:VI_256 2 "nonimm_or_0_operand") + [(set (match_operand:VI_256_AVX2 0 "register_operand") + (vec_merge:VI_256_AVX2 + (match_operand:VI_256_AVX2 1 "nonimmediate_operand") + (match_operand:VI_256_AVX2 2 "nonimm_or_0_operand") (match_operand: 3 "register_operand")))] - "TARGET_AVX2" + "TARGET_AVX" { ix86_expand_sse_movcc (operands[0], operands[3], operands[1], operands[2]); diff --git a/gcc/testsuite/ChangeLog b/gcc/testsuite/ChangeLog index c7b8e6a585ae..9ec4d50dac37 100644 --- a/gcc/testsuite/ChangeLog +++ b/gcc/testsuite/ChangeLog @@ -1,6 +1,11 @@ 2020-02-13 Jakub Jelinek Backported from mainline + 2020-02-10 Jakub Jelinek + + PR target/93637 + * gcc.target/i386/avx-pr93637.c: New test. + 2020-02-08 Uroš Bizjak Jakub Jelinek diff --git a/gcc/testsuite/gcc.target/i386/avx-pr93637.c b/gcc/testsuite/gcc.target/i386/avx-pr93637.c new file mode 100644 index 000000000000..9e7a0a7c9c1a --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/avx-pr93637.c @@ -0,0 +1,17 @@ +/* PR target/93637 */ +/* { dg-do compile } */ +/* { dg-options "-mavx -mno-avx2 -O3 --param sccvn-max-alias-queries-per-access=3" } */ + +double +foo (void) +{ + int i; + double r = 7.0; + double a[] = { 0.0, 0.0, -0.0, 0.0, 0.0, -0.0, 1.0, 0.0, 0.0, -0.0, 1.0, 0.0, 1.0, 1.0 }; + + for (i = 0; i < sizeof (a) / sizeof (a[0]); ++i) + if (a[i] == 0.0) + r = a[i]; + + return r; +}