linux-kernel - Re: [PATCH] perf report: auto-detect branch stack sampling mode

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: Windows password security audit tool. GUI, reports in PDF.

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

Date:	Tue, 6 Mar 2012 00:06:35 -0300
From:	Arnaldo Carvalho de Melo <acme@...hat.com>
To:	Arun Sharma <asharma@...com>
Cc:	Peter Zijlstra <peterz@...radead.org>, Ingo Molnar <mingo@...e.hu>,
	Stephane Eranian <eranian@...gle.com>,
	linux-kernel@...r.kernel.org, dsahern@...il.com, ravitillo@....gov,
	khandual@...ux.vnet.ibm.com, robert.richter@....com,
	ming.m.lin@...el.com, vweaver1@...s.utk.edu, andi@...stfloor.org
Subject: Re: [PATCH] perf report: auto-detect branch stack sampling mode

Em Mon, Mar 05, 2012 at 03:35:35PM -0800, Arun Sharma escreveu:
> Like you probably figured from my other mail, we deal with deeply
> nested callchains with unwieldy function names a lot -- thanks to C++
> and template programming. --tui's collapsing/expanding functionality
> is quite useful to navigate that mess. I'm just taking this
> opportunity to get some attention focused on improving it :)

Excellent!

I can think about other Zoom operations, like zooming into just the
entries where some specific function in its callchains, say the one
under the cursor, appears, etc.
 
> On 3/5/12 2:26 PM, Arnaldo Carvalho de Melo wrote:
> 
> >>Here's an example:
> >>
> >># perf record -ag -- sleep 3
> >># perf report -G -s pid --tui # SIGSEGV
> >
> >Ok, now this is a good report, I managed to reproduce and will work on a
> >fix, thanks,
> 
> Something like this seems to do it for me.
> 
>         offset = addr - sym->start;
> +       len = sym->end - sym->start;
> +       if (offset >= len)
> +               return 0;
> +

That is my fault, I should have added a BUG_ON() spitting out a
callchain in this case, as that function shouldn't be called if the
address is not within its range :-\
 
> The other problem area seems to be callchains when using -p regexp
> -x options. I'll try to summarize problems there in a separate
> thread.

Please do,

Thanks a lot!

- Arnaldo
--
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/