Re: [PATCH] kdb: include header in signal handling code

[Daniel Thompson]

On Fri, Jun 30, 2023 at 05:31:01PM +0200, Arnd Bergmann wrote:
> On Fri, Jun 30, 2023, at 17:24, Daniel Thompson wrote:
> > On Wed, May 17, 2023 at 02:54:09PM +0200, Arnd Bergmann wrote:
> >> diff --git a/kernel/signal.c b/kernel/signal.c
> >> index 8f6330f0e9ca..d38df14f71ac 100644
> >> --- a/kernel/signal.c
> >> +++ b/kernel/signal.c
> >> @@ -4780,6 +4780,8 @@ void __init signals_init(void)
> >>
> >>  #ifdef CONFIG_KGDB_KDB
> >>  #include <linux/kdb.h>
> >> +#include "debug/kdb/kdb_private.h"
> >> +
> >
> > Isn't is better to move the prototype for kdb_send_sig() into
> > linux/kdb.h instead?
> >
> > That's what other kdb helpers spread across the kernel do
> > (kdb_walk_kallsyms() for example).
>
> Right, that is probably better here. Not sure if it's worth
> reworking the branch if you already merged it, the difference
> seems rather minor.

I figure it will take me as long to rework the branch as it will to
write the covering letter on the pull-request to explain why kgdb/kdb
is messing around in kernel/signal.c ;-) .


Daniel.