8:42 a.m.
Sorry for being MIA.... answers below... or attempted.
Dan Kalowsky
SSG OTC - Android Power and Performance
IRDA and StarPeak Architect
-----Original Message-----
From: Sergey Senozhatsky [mailto:sergey.senozhatsky@gmail.com]
Sent: Friday, May 9, 2014 1:28 AM
To: Sergey Senozhatsky
Cc: powertop(a)ml01.01.org; Kalowsky, Daniel; Joerg Mayer; Alexandra Yates
Subject: Re: [Powertop] segmentation fault when shutting down
On (05/09/14 11:18), Sergey Senozhatsky wrote:
On (05/09/14 08:14), Joerg Mayer wrote:
> I also get a segfault.
>
> Ciao
> Jörg
>
> powertop (git describe: v2.6-rc2-1-gb162f5d)
>
> (gdb) bt
> #0 0x00000010 in ?? ()
> #1 0x0805339b in close_display () at ../../src/display.cpp:94
> #2 0x08090c30 in clean_shutdown () at ../../src/main.cpp:365
> #3 0x0804e575 in main (argc=1, argv=0xbff65814) at
> ../../src/main.cpp:500
> (gdb) quit
>
ok... I took a quick look, and I didn't understand some things.
commit eb08d4480969c8e48347c7621bd96dc248ee2138
Author: Dan Kalowsky <daniel.kalowsky(a)intel.com>
Date: Mon Apr 28 13:52:05 2014 -0700
Remove another memory leak
how did you spot that?
Valgrind was a good start to this on a Fedora Core 20 build. After that the initial focus
was on Android OS and adding a LOT of printf's to see what happened where.
Pointer gets over written without ever being properly cleaned up. This
should solve that problem.
and my next silly question is -- can someone please explain to me what pointer is
overwritten (in original code) in
tuning_window::window_refresh() -> clear_tuning(), init_tuning()
chain?
-ss
+class tuning_window *tune_window;
+
class tuning_window: public tab_window {
public:
virtual void repaint(void);
@@ -80,6 +82,11 @@ void initialize_tuning(void)
init_tuning();
w->cursor_max = all_tunables.size() - 1;
+
+ if (tune_window)
+ delete tune_window;
+
+ tune_window = w;
}
@@ -305,6 +312,10 @@ void clear_tuning()
delete all_untunables[i];
}
all_untunables.clear();
+
+ if (tune_window)
+ delete tune_window;
+ tune_window = NULL;
^^^^^^^^^
}
void auto_toggle_tuning()
tuning_window::window_refresh() calls clear_tuning(). which deletes
tune_window and sets it to NULL. then init_tuning() called.
the fishy part is that it all happens within that `if (w)' in
display.cpp window_refresh()
if (w) {
w->ypad_pos = 0;
w->xpad_pos = 0;
w->window_refresh();
^^^^^ here we kill tuning_window
w->repaint();
^^^^^ and here we try to execute its function?
}
can someone check (or prove me wrong) if we actually have problems here?
if we do -- then was it actually tested?
All of this was tested on our internal Android builds and on my Fedora Core system. It
helped remove the segfault we were encountering on Android at regular intervals.
-ss
> jmayer@egg:~/work/power/powertop/git/build(master)> lsb_release -a
> LSB Version:
> core-2.0-noarch:core-3.2-noarch:core-4.0-noarch:core-2.0-ia32:core-3
> .2-ia32:core-4.0-ia32:desktop-4.0-ia32:desktop-4.0-noarch:graphics-2
> .0-ia32:graphics-2.0-noarch:graphics-3.2-ia32:graphics-3.2-noarch:gr
> aphics-4.0-ia32:graphics-4.0-noarch
> Distributor ID: openSUSE project
> Description: openSUSE 13.1 (Bottle) (i586)
> Release: 13.1
> Codename: Bottle
>
> jmayer@egg:~/work/power/powertop/git/build(master)> uname -a Linux
> egg 3.14.2-1.g1474ea5-desktop #1 SMP PREEMPT Sun Apr 27 14:35:22 UTC
> 2014 (1474ea5) i686 i686 i386 GNU/Linux
>
>
> jmayer@egg:~/work/power/powertop/git/build(master)> cat /proc/cpuinfo
> processor : 0
> vendor_id : GenuineIntel
> cpu family : 6
> model : 23
> model name : Intel(R) Core(TM)2 Duo CPU T8300 @ 2.40GHz
> stepping : 6
> microcode : 0x60f
> cpu MHz : 2400.000
> cache size : 3072 KB
> physical id : 0
> siblings : 2
> core id : 0
> cpu cores : 2
> apicid : 0
> initial apicid : 0
> fdiv_bug : no
> f00f_bug : no
> coma_bug : no
> fpu : yes
> fpu_exception : yes
> cpuid level : 10
> wp : yes
> flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat
> +pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx lm
> +constant_tsc arch_perfmon pebs bts aperfmperf pni dtes64 monitor
> +ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm sse4_1 lahf_lm ida dtherm
> +tpr_shadow vnmi flexpriority
> bogomips : 4788.29
> clflush size : 64
> cache_alignment : 64
> address sizes : 36 bits physical, 48 bits virtual
> power management:
> [output for processor 1 deleted]
>
> _______________________________________________
> PowerTop mailing list
> PowerTop(a)lists.01.org
> https://lists.01.org/mailman/listinfo/powertop