linux-kernel - Re: [PATCH v2] perf report: fix off-by-one for non-activation frames

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, 16 May 2017 18:26:47 +0200
From:   Milian Wolff <milian.wolff@...b.com>
To:     Namhyung Kim <namhyung@...nel.org>
Cc:     Linux-kernel@...r.kernel.org, linux-perf-users@...r.kernel.org,
        Arnaldo Carvalho de Melo <acme@...hat.com>,
        David Ahern <dsahern@...il.com>,
        Peter Zijlstra <a.p.zijlstra@...llo.nl>,
        Yao Jin <yao.jin@...ux.intel.com>, kernel-team@....com
Subject: Re: [PATCH v2] perf report: fix off-by-one for non-activation frames

On Dienstag, 16. Mai 2017 18:17:26 CEST Milian Wolff wrote:
> On Dienstag, 16. Mai 2017 16:38:29 CEST Namhyung Kim wrote:
> > On Tue, May 16, 2017 at 10:59:51AM +0200, Milian Wolff wrote:
> > > As the documentation for dwfl_frame_pc says, frames that
> > > are no activation frames need to have their program counter
> > > decremented by one to properly find the function of the caller.
> > > 
> > > This fixes many cases where perf report currently attributes
> > > the cost to the next line. I.e. I have code like this:
> > > 
> > > ~~~~~~~~~~~~~~~
> > > 
> > >   #include <thread>
> > >   #include <chrono>
> > >   
> > >   using namespace std;
> > >   
> > >   int main()
> > >   {
> > >   
> > >     this_thread::sleep_for(chrono::milliseconds(1000));
> > >     this_thread::sleep_for(chrono::milliseconds(100));
> > >     this_thread::sleep_for(chrono::milliseconds(10));
> > >     
> > >     return 0;
> > >   
> > >   }
> > > 
> > > ~~~~~~~~~~~~~~~
> > 
> > It'd be nice if the test program has a signal frame for verification.
> 
> I have pretty much zero experience about signals. Would it be enough to add
> a signal handler for, say, SIGUSR1 to my test application and then trigger
> a sleep when that signal is delivered? If that should be enough, I'll write
> and test it out.

~~~~~~~~~~~~~~~
#include <thread>
#include <chrono>
#include <signal.h>

using namespace std;

volatile bool run_loop = true;

void my_handler(int signum)
{
    this_thread::sleep_for(chrono::milliseconds(1000));
    this_thread::sleep_for(chrono::milliseconds(100));
    this_thread::sleep_for(chrono::milliseconds(10));
    run_loop = false;
}

int main()
{
    signal(SIGUSR1, my_handler);

    while (run_loop) {}

    return 0;
}
~~~~~~~~~~~~~~~

This does not properly unwind neither before nor after this patch. I only ever 
get:

   100.00%  core.c:0
            |
            ---__schedule core.c:0
               schedule
               do_nanosleep hrtimer.c:0
               hrtimer_nanosleep
               sys_nanosleep
               entry_SYSCALL_64_fastpath .tmp_entry_64.o:0
               __nanosleep_nocancel .:0
               std::this_thread::sleep_for<long, std::ratio<1l, 1000l> > 
thread:323

So... should this work? Please tell me how to test this properly.

Thanks
-- 
Milian Wolff | milian.wolff@...b.com | Software Engineer
KDAB (Deutschland) GmbH&Co KG, a KDAB Group company
Tel: +49-30-521325470
KDAB - The Qt Experts
Download attachment "smime.p7s" of type "application/pkcs7-signature" (5903 bytes)