Re: [PATCH] Multiple locations for tracepoints

[Stan Shebs <stan at codesourcery dot com>]

Joel Brobecker wrote:
> > Committed to the trunk.  I'm not sure if this should go to 7.1 also;
> > it has no effect outside tracepoint-land, but don't want to churn
> > the sources if it's close to release.
> >     
>
> I think we still have a couple of weeks before we decide to release.
> I haven't started calling for known problems that people might think
> is blocking for the release. I'll probably do that early next week.
>
> For your patch, my initial reaction was that it was best to leave it
> out of 7.1. But I'm not 100% sure because, not knowing tracepoints
> all that well, I'm not sure whether you're fixing a bug or just
> enhancing support for multi-location tracepoints. My guideline for
> merging a change to the branch is: Is this a bad bug or regression?;
> and is the fix relatively safe?  Your patch is not all that small or
> obvious, so unless it's a crippling issue, I'd suggest we leave it out
> (everyone is allowed to disagree ;-), as always).
>
>   
It's not a regression, you'll just get surprising results if you hit the 
wrong location. But I don't think people are going to do much with 
tracepoints until gdbserver support is in anyway, and that won't be in 
7.1, so I'm going to restrict myself to the trunk.

Stan