On 12/19/25 11:06 AM, Jakub Jelinek wrote:
On Fri, Dec 19, 2025 at 04:33:04PM +0100, Tobias Burnus wrote:
Vladimir Makarov wrote:
So the patch has sense. You can commit it into the trunk. Thank you
for the patch, Alex.
This patch unfortunately breaks GCN bootstrap,
cf. https://gcc.gnu.org/PR123223
for gcc/ChangeLog
PR target/55212
* lra-constraints.cc (match_reload): Treat former scratch
regs as implicit unused output reloads.
(process_alt_operands): Likewise.
(curr_insn_transform): Likewise.
x86_64 and i686-linux bootstrap as well, all with
--enable-checking=yes,rtl,extra
REGNO is used multiple times in the patch without making sure
it is applied on REG_P only.
Sorry for the troubles. I'll fix it as soon as possible or revert
Alex's patch if I can not do it today.
E.g. when building
bid128_fma.o
bid64_div.o
bid_binarydecimal.o
_divhc3.o
divtf3.o
emutls.o
floatbitinttd.o
_gcov_execle.o
_gcov_execl.o
_gcov_execlp.o
_gcov_info_to_gcda.o
generic-morestack.o
_mulhc3.o
unwind-dw2-fde-dip.o
unwind-dw2.o
in libgcc but probably pretty much anything larger.
Jakub
