This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: RFC: Updates support for breakpoints that generate SIGILL
- From: Daniel Jacobowitz <dan at codesourcery dot com>
- To: Ulrich Weigand <uweigand at de dot ibm dot com>
- Cc: gdb-patches at sourceware dot org
- Date: Mon, 8 Feb 2010 14:45:45 -0500
- Subject: Re: RFC: Updates support for breakpoints that generate SIGILL
- References: <20100128215914.GB2813@caradoc.them.org> <201002081935.o18JZIMA023833@d12av02.megacenter.de.ibm.com>
On Mon, Feb 08, 2010 at 08:35:17PM +0100, Ulrich Weigand wrote:
> Unfortunately this change broke Cell/B.E. debugging.
>
> The problem is that with the change, handle_inferior_event will now
> *always* look up a regcache, even when the process has exited.
>
> Before the patch, the code has always taken care to avoid that; while
> some targets don't mind, others will break (in particular, the Cell/B.E.
> multi-arch target does).
>
> The patch below changes this to only look up the regcache when needed
> (in particular, this implies the process is still there).
>
> Tested on Cell/B.E. (ppc32 and ppc64) with no regressions.
>
> Does this look OK to you?
Sorry for the breakage. Yes, your fix looks good to me.
--
Daniel Jacobowitz
CodeSourcery