Re: synchronous signal in the blocked signal context

From: Linus Torvalds
Date: Tue Aug 01 2006 - 00:52:43 EST




On Mon, 31 Jul 2006, Siddha, Suresh B wrote:
>
> This patch (b0423a0d9cc836b2c3d796623cd19236bfedfe63)
>
> [PATCH] Remove duplicate code in signal.c
>
> reverts a patch introduced by Linus long time back.

Good catch.

> Was this intentional?
>
> With the current mainline code, SIGSEGV inside a SIGSEGV handler will endup
> in linux handling endless recursive faults.
>
> Just wondering if this has been discussed before and is intentional.

It certainly wasn't discussed, and I don't think it was intentional. We
should _not_ just unblock a blocked signal. We should kill the process,
because sending the signal is actually very very wrong.

Paul? Should I just revert, or did you have some deeper reason for it?

Linus
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/