This is the mail archive of the gdb@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: [ANNOUNCEMENT] GDB 7.0 release process created


> Ah, I see.  It's true that overlapping sections are already filtered
> out, but it is not guaranteed which ones, and in fact the .text from
> a.exe is more likely to be discarded then the .text from (some) foo.o,
> because the latter will generally have lower section start.

Hmmm, I knew that it was not guaranteed that the .o sections would
be discarded, but I was hoping that it just happened to be the case.
ISTR that this was the case when I looked at the warnings that were
printed. I assumed that, after relocation, the .text section of the
.o files should always be equal or higher than the .text section of
the executable...

> Presumably you did a fair amount of testing on that.
> 
> If so, should I commit that patch on the 7.0 branch (and perhaps on
> the trunk) as a temporary workaround, until Tristan comes up with a
> better fix?

That's what I suggested we do a few days ago, before we cut the branch.
But I haven't had a chance to test this much. Right now, AdaCore has
a separate branch for x86-darwin GDB, that we used to develop the port.
We're trying to move to gdb-7.0, but we haven't had time to complete
the integration and until then, testing is difficult. I definitely
plan on completing this part before 7.0 is out, however.

In the meantime, I'll see if I can reproduce Jack's issue with the
latest sources.

-- 
Joel


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