[Crash-utility] [PATCH] GDB started inside crash should not read any .gdbinit

HAGIO KAZUHITO(萩尾 一仁) k-hagio-ab at nec.com
Wed Mar 10 04:10:49 UTC 2021


-----Original Message-----
> I don't think we want .gdbinit files to impact functions in
> crash. Example of bad behavior, a .gdbinit file with:
> set logging on
> 
> then when starting crash there is segmentation fault.

Thanks for the report and patch.

Reading .gdbinit is the intended functionality of crash:

crash> define func
crash: restricted gdb command: define
       "define" may only be used in a .gdbinit file or in a command file.
       The .gdbinit file is read automatically during crash initialization.
...

and I would like to avoid changing the default behavior, so how about
adding --no_gdbinit option like --no_crashrc ?

(btw, the crash mailing list accepts only emails posted by member.
I accepted this email but if someone replies only to the list,
you will miss it.  Please subscribe to the list if possible.)

Thanks,
Kazu

> ---
>  gdb_interface.c | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/gdb_interface.c b/gdb_interface.c
> index f4f4dd3..cfa9f1e 100644
> --- a/gdb_interface.c
> +++ b/gdb_interface.c
> @@ -42,6 +42,7 @@ gdb_main_loop(int argc, char **argv)
>  {
>         argc = 1;
> 
> +       argv[argc++] = "--nx";
>         if (pc->flags & SILENT) {
>                 if (pc->flags & READNOW)
>                         argv[argc++] = "--readnow";
> --
> 2.17.1
> 
> 
> 
> --
> Crash-utility mailing list
> Crash-utility at redhat.com
> https://listman.redhat.com/mailman/listinfo/crash-utility





More information about the Crash-utility mailing list