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: [patch+7.4] reread.exp 7.3->7.4 regression


Jan Kratochvil wrote:
> On Mon, 19 Dec 2011 11:30:54 +0100, Ulrich Weigand wrote:
> > I can test on ARM if you want.
> 
> I have tested it on fedora13beta.  Just on
> 
> Hardware        : OMAP3 Beagle Board
> Revision        : 0020
> Processor       : ARMv7 Processor rev 2 (v7l)
> BogoMIPS        : 506.27
> 
> it takes 57m28.760s to build GDB (although with full symbols) and 51m13.889s
> to run the testsuite, is it normal to develop on ARM with this performance?

I guess so.  A Panda is a bit faster than the Beagle, but still ...

> > Your initial patch already moved the callbacks calls up a bit;
> 
> My patch was using free_objfile, I do not understand which patch of mine do
> you refer to now.
> 	Re: [patch] Replace reread_symbols by load+free calls
> 	http://sourceware.org/ml/gdb-patches/2009-06/msg00696.html

Sorry, I was refering to the first patch in list you had in the mail I replied to.
Specifically, I was talking about this patch:
http://sourceware.org/ml/gdb-patches/2009-06/msg00606.html
which seems to have gone upstream 2009-06-23.

> > In addition, we should probably call observer_notify_new_objfile so that new
> > tables can be built up for the re-read file ...
> 
> I can post it afterwards, that is probably not for 7.4, I will have to run the
> ARM testsuite again.

OK, thanks.

> No regressions on {x86_64,x86_64-m32,i686}-fedora16-linux-gnu and on
> arm-fedora13beta-linux-gnu.  OK to check it in and for 7.4?
> 
> gdb/
> 2011-12-19  Jan Kratochvil  <jan.kratochvil@redhat.com>
> 
> 	* symfile.c (reread_symbols): Move free_objfile_separate_debug,
> 	preserve_values, sym_finish and clear_objfile_data calls before BFD
> 	close.  Move free_objfile_separate_debug as the very first call.  New
> 	comment on the ordering.

This looks reasonable to me, and seems preferable to the ARM-specific hack ...


Thanks,
Ulrich

-- 
  Dr. Ulrich Weigand
  GNU Toolchain for Linux on System z and Cell BE
  Ulrich.Weigand@de.ibm.com


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