This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [patch] Replace reread_symbols by load+free calls
- From: Tom Tromey <tromey at redhat dot com>
- To: Jan Kratochvil <jan dot kratochvil at redhat dot com>
- Cc: gdb-patches at sources dot redhat dot com
- Date: Fri, 14 Aug 2009 16:34:31 -0600
- Subject: Re: [patch] Replace reread_symbols by load+free calls
- References: <8ac60eac0906101839t4d3978fyc1c6d3b3e2eccb6e@mail.gmail.com> <8ac60eac0906101842y2d2fc9fco331cb4336d9508d0@mail.gmail.com> <m3zlc80yxc.fsf@fleche.redhat.com> <8ac60eac0906161342m67c30151lc7f685e0eb36e41f@mail.gmail.com> <m3y6rrzyi7.fsf@fleche.redhat.com> <8ac60eac0906161407s6498913cu5e50e930d599be77@mail.gmail.com> <20090623150322.GA7654@host0.dyn.jankratochvil.net> <m3ab3ylt5i.fsf@fleche.redhat.com> <20090623182139.GA30969@host0.dyn.jankratochvil.net> <m3y6rik9eu.fsf@fleche.redhat.com> <20090625192120.GA29930@host0.dyn.jankratochvil.net>
- Reply-to: tromey at redhat dot com
>>>>> "Jan" == Jan Kratochvil <jan.kratochvil@redhat.com> writes:
Sorry about the long delay on this. I try not to just drop threads I'm
involved in, but sometimes I get distracted.
Jan> IMO the whole `struct objfile' should replaced by a new one
Jan> (instead of its in-place patching). Which will probably have
Jan> other dependencies...
Tom> I'm interested to hear your idea and reasoning.
Jan> Sometimes there are differentiations between complete objects
Jan> destructors (objfile_free_data) and in-place object deletions
Jan> (clear_objfile_data); OK did not briefly found more such cases now.
Jan> Expecting it was there for some performance improvements which not
Jan> worth it nowadays.
I misunderstood something you were saying earlier in the thread -- I
thought you were talking about dropping 'struct objfile' itself in favor
of some new datastructure (as opposed to slowly refactoring it). But,
now I understand that you really meant that reread_symbols should be
rewritten.
I realize you pulled this patch from consideration. But, FWIW, I tend
to think it is a good idea, because it ensures that all the hooks and
observers will be run identically in both cases.
Tom