diff options
| author | Jakub Jelinek <jakub@redhat.com> | 2022-04-22 13:38:11 +0200 |
|---|---|---|
| committer | Jakub Jelinek <jakub@redhat.com> | 2022-04-22 13:38:11 +0200 |
| commit | 7092b7aea122a91824d048aeb23834cf1d19b1a1 (patch) | |
| tree | eb6712fec07d8cb32129652226f5be2eade6d0c4 /libphobos/libdruntime | |
| parent | c1a9cf679153719d066f1ec79154c406e1e48cdf (diff) | |
| download | gcc-7092b7aea122a91824d048aeb23834cf1d19b1a1.zip gcc-7092b7aea122a91824d048aeb23834cf1d19b1a1.tar.gz gcc-7092b7aea122a91824d048aeb23834cf1d19b1a1.tar.bz2 | |
rtlanal: Fix up replace_rtx [PR105333]
The following testcase FAILs, because replace_rtx replaces a REG with
CONST_WIDE_INT inside of a SUBREG, which is an invalid transformation
because a SUBREG relies on SUBREG_REG having non-VOIDmode but
CONST_WIDE_INT has VOIDmode.
replace_rtx already has code to deal with it, but it was doing
it only for CONST_INTs. The following patch does it also for
VOIDmode CONST_DOUBLE or CONST_WIDE_INT.
2022-04-22 Jakub Jelinek <jakub@redhat.com>
PR rtl-optimization/105333
* rtlanal.cc (replace_rtx): Use simplify_subreg or
simplify_unary_operation if CONST_SCALAR_INT_P rather than just
CONST_INT_P.
* gcc.dg/pr105333.c: New test.
Diffstat (limited to 'libphobos/libdruntime')
0 files changed, 0 insertions, 0 deletions