Re: posix_spawn

"Kirill A. Korinsky" <[email protected]> Sun, 24 Dec 2023 15:25:08 +0100

Newsgroups	gmane.lisp.gcl.devel
Message-ID	<[email protected]>

Hello!

> On 24. Dec 2023, at 15:17, Camm Maguire <[email protected]> wrote:
> 
> "Kirill A. Korinsky" <[email protected]> writes:
> 
>> Thus, additionaly I've added patch below to catch the real place where it crashes:
>> 
>> --- o/main.c
>> +++ o/main.c
>> @@ -665,10 +665,10 @@ main(int argc, char **argv, char **envp) {
>> /* catch certain signals */
>> void install_segmentation_catcher(void)
>> {
>> -  unblock_signals(SIGSEGV,SIGSEGV);
>> -  unblock_signals(SIGBUS,SIGBUS);
>> -  (void) gcl_signal(SIGSEGV,segmentation_catcher);
>> -  (void) gcl_signal(SIGBUS,segmentation_catcher);
>> +/*   unblock_signals(SIGSEGV,SIGSEGV); */
>> +/*   unblock_signals(SIGBUS,SIGBUS); */
>> +/*   (void) gcl_signal(SIGSEGV,segmentation_catcher); */
>> +/*   (void) gcl_signal(SIGBUS,segmentation_catcher); */
>> }
>> 
> 
> A new issue specific to arm?  Where does it fail without this?

Not as far as I recall. I have the same behaviour on macOS 12 x86_64.

This patch allows to ignore GCL's error handler which masks real stacktrace and simple says "Segmentation violation" without any trace, instead an application failed and OS catches it and makes nice crash report which allows to investigate where.

signature.asc (application/pgp-signature, 833 B)