From 77a50c772771f681085922b493922516c3c03e9a Mon Sep 17 00:00:00 2001 From: liuhongt Date: Sun, 25 Jun 2023 11:35:09 +0800 Subject: Don't use intermiediate type for FIX_TRUNC_EXPR when ftrapping-math. > > Hmm, good question. GENERIC has a direct truncation to unsigned char > > for example, the C standard generally says if the integral part cannot > > be represented then the behavior is undefined. So I think we should be > > safe here (0x1.0p32 doesn't fit an int). > > We should be following Annex F (unspecified value plus "invalid" exception > for out-of-range floating-to-integer conversions rather than undefined > behavior). But we don't achieve that very well at present (see bug 93806 > comments 27-29 for examples of how such conversions produce wobbly > values). That would mean guarding this with !flag_trapping_math would be the appropriate thing to do. gcc/ChangeLog: PR tree-optimization/110371 PR tree-optimization/110018 * tree-vect-stmts.cc (vectorizable_conversion): Don't use intermiediate type for FIX_TRUNC_EXPR when ftrapping-math. gcc/testsuite/ChangeLog: * gcc.target/i386/pr110018-1.c: Add -fno-trapping-math to dg-options. * gcc.target/i386/pr110018-2.c: Ditto. --- gcc/testsuite/gcc.target/i386/pr110018-1.c | 2 +- gcc/testsuite/gcc.target/i386/pr110018-2.c | 2 +- gcc/tree-vect-stmts.cc | 3 ++- 3 files changed, 4 insertions(+), 3 deletions(-) (limited to 'gcc') diff --git a/gcc/testsuite/gcc.target/i386/pr110018-1.c b/gcc/testsuite/gcc.target/i386/pr110018-1.c index b6a3be7..24eeca6 100644 --- a/gcc/testsuite/gcc.target/i386/pr110018-1.c +++ b/gcc/testsuite/gcc.target/i386/pr110018-1.c @@ -1,5 +1,5 @@ /* { dg-do compile } */ -/* { dg-options "-mavx512fp16 -mavx512vl -O2 -mavx512dq" } */ +/* { dg-options "-mavx512fp16 -mavx512vl -O2 -mavx512dq -fno-trapping-math" } */ /* { dg-final { scan-assembler-times {(?n)vcvttp[dsh]2[dqw]} 5 } } */ /* { dg-final { scan-assembler-times {(?n)vcvt[dqw]*2p[dsh]} 5 } } */ diff --git a/gcc/testsuite/gcc.target/i386/pr110018-2.c b/gcc/testsuite/gcc.target/i386/pr110018-2.c index a663e07..9a2d9e1 100644 --- a/gcc/testsuite/gcc.target/i386/pr110018-2.c +++ b/gcc/testsuite/gcc.target/i386/pr110018-2.c @@ -1,5 +1,5 @@ /* { dg-do compile } */ -/* { dg-options "-mavx512fp16 -mavx512vl -O2 -mavx512dq" } */ +/* { dg-options "-mavx512fp16 -mavx512vl -O2 -mavx512dq -fno-trapping-math" } */ /* { dg-final { scan-assembler-times {(?n)vcvttp[dsh]2[dqw]} 5 } } */ /* { dg-final { scan-assembler-times {(?n)vcvt[dqw]*2p[dsh]} 5 } } */ diff --git a/gcc/tree-vect-stmts.cc b/gcc/tree-vect-stmts.cc index 85d1f3a..7d24bbe 100644 --- a/gcc/tree-vect-stmts.cc +++ b/gcc/tree-vect-stmts.cc @@ -5263,7 +5263,8 @@ vectorizable_conversion (vec_info *vinfo, if ((code == FLOAT_EXPR && GET_MODE_SIZE (lhs_mode) > GET_MODE_SIZE (rhs_mode)) || (code == FIX_TRUNC_EXPR - && GET_MODE_SIZE (rhs_mode) > GET_MODE_SIZE (lhs_mode))) + && GET_MODE_SIZE (rhs_mode) > GET_MODE_SIZE (lhs_mode) + && !flag_trapping_math)) { bool float_expr_p = code == FLOAT_EXPR; scalar_mode imode = float_expr_p ? rhs_mode : lhs_mode; -- cgit v1.1