This is the mail archive of the gdb-patches@sourceware.org mailing list for the GDB project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Fix regression in always-inserted breakpoints mode.


On Friday 20 November 2009 11:38:21, Jan Kratochvil wrote:
> On Fri, 20 Nov 2009 03:11:18 +0100, Pedro Alves wrote:
> > There's for loop a bit below, that starts iterating on
> > `locp', and that would misses a few locations,
> 
> I agree I brought in such regression, thanks for catching it.
> 
> I will check-in the [obv] patch below after you check it in.

Good idea.  I've checked mine in.  Go ahead.

> > There's another problem with this for loop.  The
> > breakpoint_address_match is too strict when debugging
> > more than one inferior.  Even if breakpoint_address_match
> > fails due to address space mismatch, there could be other
> > locations with the same address following loc2p.
> 
> This is unrelated to the regression from me.

We were always looping over all locations previously,
and breakpoint_address_match wasn't a part of the for
loop conditional, so it is.  But that's okay, it doesn't
matter, regressions happen.  I was just looking for an extra
pair of eyes that knows this new code.

-- 
Pedro Alves


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]