https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115905
--- Comment #10 from Iain Sandoe <iains at gcc dot gnu.org> ---
unfortunately, (or ...) I Have not succeeded in reproducing this - so will need
some help to identify what's being done differently from you.
I built: r15-3667-gf5448384a213 (also on my Darwin17+ set, which includes a
32b host - which also seems fine).
On cfarm216 :
../src/configure --prefix=/home/iains/gcc-master/gcc-15-0-0
--with-as=/usr/ccs/bin/as --without-gnu-as --enable-languages=c,c++
--build=sparc-sun-solaris2.11 CC='gcc -m32' CXX='g++ -m32'
Thread model: posix
Supported LTO compression algorithms: zlib zstd
gcc version 15.0.0 20240916 (experimental) [master f5448384a21] (GCC)
----
gmake check-gcc-c++ RUNTESTFLAGS="coroutines.exp coro-torture.exp"
Running
/home/iains/gcc-master/src/gcc/testsuite/g++.dg/coroutines/coroutines.exp ...
Running
/home/iains/gcc-master/src/gcc/testsuite/g++.dg/coroutines/torture/coro-torture.exp
...
=== g++ Summary ===
# of expected passes 2275
# of expected failures 1
# of unsupported tests 1
/home/iains/gcc-master/bld/gcc/xg++ version 15.0.0 20240916 (experimental)
[master f5448384a21] (GCC)
====
Your report seems to be a compile-time SIGBUS, which is somewhat surprising for
the changes to address this bug - but always one can be surprised, of course.
