> On Aug 23, 2021, at 05:49, Bruno Haible <br...@clisp.org> wrote:
>
> That's what I suspected, for the reasons mentioned in [1].
>
> Now, to check that with libsigsegv it works fine:
>
> make distclean
>
> ./configure CPPFLAGS=-Wall --with-libsigsegv
> --with-libsigsegv-prefix=/opt/local
> (use the prefix where you actually have installed libsigsegv)
>
> make
>
> Comment out the failing/hanging tests (test-nanosleep, test-pthread,
> test-pthread-rwlock) from gltests/Makefile again.
>
> make check
>
Linking against libsigsegv, there are 6 failing tests. Consolidating console
and test-suite.log output:
./test-c-stack.sh: line 7: 24173 Illegal instruction ${CHECKER}
./test-c-stack${EXEEXT} 2>t-c-stack.tmp
FAIL test-c-stack.sh (exit status: 1)
FAIL test-c-stack2.sh (exit status: 1)
../build-aux/test-driver: line 112: 24614 Illegal instruction "$@"
>$log_file 2>&1
FAIL: test-sigsegv-catch-segv1 (exit status: 132)
../build-aux/test-driver: line 112: 24621 Illegal instruction "$@"
>$log_file 2>&1
FAIL: test-sigsegv-catch-segv2 (exit status: 132)
../build-aux/test-driver: line 112: 24628 Illegal instruction "$@"
>$log_file 2>&1
FAIL: test-sigsegv-catch-stackoverflow1 (exit status: 132)
../build-aux/test-driver: line 112: 24635 Illegal instruction "$@"
>$log_file 2>&1
FAIL: test-sigsegv-catch-stackoverflow2 (exit status: 132)
So perhaps we're better off with the c-stack module for now.
