linux-kernel - Re: [PATCH 8/9] perf annotate: fix it for non-prelinked *.so

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:	Fri, 05 Feb 2010 07:54:38 +0100
From:	Mike Galbraith <efault@....de>
To:	Arnaldo Carvalho de Melo <acme@...radead.org>
Cc:	Kirill Smelkov <kirr@...dau.phys.spbu.ru>,
	Ingo Molnar <mingo@...e.hu>, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 8/9] perf annotate: fix it for non-prelinked *.so

On Thu, 2010-02-04 at 17:48 -0200, Arnaldo Carvalho de Melo wrote:
> Em Thu, Feb 04, 2010 at 10:34:04PM +0300, Kirill Smelkov escreveu:
> > On Thu, Feb 04, 2010 at 07:31:46AM +0100, Mike Galbraith wrote:
> > > Monkey see monkey do.
> > > 
> > > perf tools: fix perf top module symbol annotation.
> > >  		"objdump --start-address=0x%016Lx "
> > >  			 "--stop-address=0x%016Lx -dS %s",
> > > -		map->unmap_ip(map, sym->start),
> > > -		map->unmap_ip(map, sym->end), path);
> > > +		map__rip_2objdump(map, sym->start),
> > > +		map__rip_2objdump(map, sym->end), path);
> > 
> > If I recall correctly, that's not enough.
> > 
> > The problem is top code is also wrong at mapping objdump addresses to
> > absolute ip. That is another part of builtin-top.c which does
> > map->unmap_ip(), and I've already suggested a fix back at holidays:
> > 
> > http://marc.info/?l=linux-kernel&m=126295508002536&w=2
> 
> Yeah, I was looking at Mike's report about 'perf annotate' not working
> with modules, debugged, wrote a patch that looed at the ELF header obj
> type as the key to apply or not the unmap_ip operation and when it was
> working I thought I saw that patch somewhere, looked at yours and
> applied it instead.
> 
> Then I was testing your subsequent patches but had to call it a day,
> sent what I had that at least fixed 'annotate' and Mike took it from
> there.
> 
> Looking at this now.

Hm, seems to work fine.

(piddle)

Aha, the rest is needed for userland annotation to work.

	-Mike

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/