RE: [PATCH 2/2 ] PowerPC: fix for gdb.reverse/finish-precsave.exp and gdb.reverse/finish-reverse.exp

[Carl Love <cel@us.ibm.com>]

Bruno:


On Wed, 2023-08-02 at 11:29 +0200, Bruno Larsen wrote:
> Hi Carl!
> 
> Thanks for the thorough explanation and test. You did understood what
> I 
> meant perfectly :)
> 
> This is a little besides the point, but I don't see any other
> patches 
> that touch this area and haven't been committed yet. Am I missing 
> something obvious?

The patch that I was talking about is patch 2/2 for the reverse
execution that I posted a few weeks ago.  Patch 1 was for checking the
gcc version.  But yea, the discussion thread is getting long enough
that some of this is getting lost.  
> 
> Regardless, I feel like this one change can propagate enough that it 
> should be its own commit anyway.
> 
> > If you were to apply the following patch on top of the currently
> > proposed patch, does gdb then execute as you expect on your
> > system?  I
> > have tested the change on IBM Power 10.  It would be good to make
> > sure
> > it works on your X86 system as well.  I also ran the full gdb
> > regression tests on Power 10 with no additional regression
> > failures.
> > Thanks for looking at the patch, proposed fix and the additional
> > testing.
> 
> The patch didn't apply on current master, but I added the changes 
> manually and tested it. It does fix the issue, and adds no
> regressions.

I think this little fix needs to be added to patch 2/2 for fixing the
reverse execution as this fix and the patch are trying to fix the same
things.  I was also thinking this fix needs to be inside a if (reverse-
execution) statement as it is really for the reverse execution.
> 
> I feel like it could be worth sending an official version this patch
> as 
> a new thread, instead of a reply to this one, to get people who
> might 
> not be interested in reverse debugging but may have thoughts on this 
> patch, but maybe wait a bit to see if someone in here says anything
> :)

I have been thinking it is time to start a fresh thread for the patches
as I think the length of this thread is losing people.  Lets see if
anyone else responds to this thread in a week or so.  I will add the if
(reverse execution) to the little fix as mentioned, update the tests,
refresh the patch and then send out a new version in a new thread.

Thanks for the help.

                            Carl