This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: Oh dear. I regret to inform you that commit 053af8c9034f92d6e36a1180655ba22a65c56437 might be unfortunate
- From: Nick Clifton <nickc at redhat dot com>
- To: Simon Marchi <simon dot marchi at polymtl dot ca>, gdb-buildbot at sergiodj dot net
- Cc: gdb-patches at sourceware dot org
- Date: Tue, 8 Jan 2019 11:24:32 +0000
- Subject: Re: Oh dear. I regret to inform you that commit 053af8c9034f92d6e36a1180655ba22a65c56437 might be unfortunate
- References: <053af8c9034f92d6e36a1180655ba22a65c56437-master-breakage@gdb-build> <c9d33a75a00d4ceb8e126a1dbcbcb16c@polymtl.ca> <3f3e538c74685b15579bf913aa764067@polymtl.ca>
Hi Simon,
>> As you can see, GDB needs to be adjusted to stop using
Sorry - I totally forgot to check the effect that the change would have on GDB. Mea culpa.
>> test it, get it reviewed). Would you mind reverting this patch, just
>> the time we adjust the GDB code?
Not at all. In fact I see that you have now done so.
> I ended up reverting this patch to unbreak the build. I'll try to update the GDB code shortly to unblock it (or if somebody else is motived, I don't mind).
>
> When re-syncing the code with GCC, note that there are corresponding changes to include/demangle.h that should be imported too.
Yes, I missed that too. :-( There is also an update to the binutils/testsuite/binutils-all/cxxfilt.exp file that needs to go in to fix the tests for targets that use an
underscore prefix to their symbol names. Also, once demangle.h is synchronized the
binutils/cxxfilt.c source file will fail to build because it checks for redacted demangling
styles. I have patches locally to fix both of these problems, which I can supply to
a committer or else check in myself.
Cheers
Nick