8:22 a.m.
The bisection resulted in
commit 5f5976f8b42e4e4717f2d573cf43cb87ccd88fa2
Author: Dan Kalowsky <daniel.kalowsky(a)intel.com>
Date: Tue Apr 29 15:55:31 2014 -0700
Adding a clean_shutdown function
Function is designed to be a on stop place to clean up
the entire application at shutdown time.
as the first bad commit.
Regards
Martin
8:06 p.m.
New subject: [Powertop] segmentation fault when shutting
On (05/09/14 00:22), Martin Ziegler wrote:
The bisection resulted in
commit 5f5976f8b42e4e4717f2d573cf43cb87ccd88fa2
Author: Dan Kalowsky <daniel.kalowsky(a)intel.com>
Date: Tue Apr 29 15:55:31 2014 -0700
Adding a clean_shutdown function
Function is designed to be a on stop place to clean up
the entire application at shutdown time.
as the first bad commit.
thanks, Martin.
yes... well
a) clean_shutdown() calls close_display() which iterates tab_windows[tab_names[i]]
and deletes every tab_window. the problem is that prior to
close_display() main executes clear_tuning(), which [surprise surprise]
does `delete tune_window' since commit eb08d4480969c8e48347c7621bd96dc248ee2138
("Remove another memory leak") , whilst tab_windows[tab_names[i]] still holds
a pointer which we just secretly free()-d. so this is double free().
and it has no other option (and will never have) than to crash.
this obviously has not been tested before push.
b) there is no point in free()-ing memory which will be free()-d anyway,
because we exit() the program.
-ss
9:22 a.m.
New subject: [Powertop] segmentation fault when shutting
On (05/09/14 00:22), Martin Ziegler wrote:
> The bisection resulted in
>
> commit 5f5976f8b42e4e4717f2d573cf43cb87ccd88fa2
> Author: Dan Kalowsky <daniel.kalowsky(a)intel.com>
> Date: Tue Apr 29 15:55:31 2014 -0700
>
> Adding a clean_shutdown function
>
> Function is designed to be a on stop place to clean up
> the entire application at shutdown time.
>
> as the first bad commit.
>
thanks, Martin.
yes... well
a) clean_shutdown() calls close_display() which iterates
tab_windows[tab_names[i]]
and deletes every tab_window. the problem is that prior to
close_display() main executes clear_tuning(), which [surprise surprise]
does `delete tune_window' since commit
eb08d4480969c8e48347c7621bd96dc248ee2138
("Remove another memory leak") , whilst tab_windows[tab_names[i]] still
holds
a pointer which we just secretly free()-d. so this is double free().
and it has no other option (and will never have) than to crash.
this obviously has not been tested before push.
b) there is no point in free()-ing memory which will be free()-d anyway,
because we exit() the program.
-ss
Martin, thank you for reporting and bisecting the culprit.
Sergey, thank you for poking the issues.
Dan, please fix the patches and resend them. They will be included after
release 2.6 is completed.
I reverted Dan Kalowsky's patches.
Thank you,
Alexandra.
5:26 a.m.
New subject: [Powertop] segmentation fault when shutting
Not all patches are at fault here so reverting the entire series seems short sighted.
Even the patch in question isn't causing issues on all platforms.
-----Original Message-----
From: Alexandra Yates [mailto:alexandra.yates@linux.intel.com]
Sent: Monday, May 12, 2014 4:23 PM
To: powertop(a)lists.01.org
Cc: Kalowsky, Daniel
Subject: Re: [Powertop] segmentation fault when shutting
On (05/09/14 00:22), Martin Ziegler wrote:
> The bisection resulted in
>
> commit 5f5976f8b42e4e4717f2d573cf43cb87ccd88fa2
> Author: Dan Kalowsky <daniel.kalowsky(a)intel.com>
> Date: Tue Apr 29 15:55:31 2014 -0700
>
> Adding a clean_shutdown function
>
> Function is designed to be a on stop place to clean up
> the entire application at shutdown time.
>
> as the first bad commit.
>
thanks, Martin.
yes... well
a) clean_shutdown() calls close_display() which iterates
tab_windows[tab_names[i]] and deletes every tab_window. the problem is
that prior to
close_display() main executes clear_tuning(), which [surprise
surprise] does `delete tune_window' since commit
eb08d4480969c8e48347c7621bd96dc248ee2138
("Remove another memory leak") , whilst tab_windows[tab_names[i]]
still holds a pointer which we just secretly free()-d. so this is
double free().
and it has no other option (and will never have) than to crash.
this obviously has not been tested before push.
b) there is no point in free()-ing memory which will be free()-d
anyway, because we exit() the program.
-ss
Martin, thank you for reporting and bisecting the culprit.
Sergey, thank you for poking the issues.
Dan, please fix the patches and resend them. They will be included after release 2.6 is
completed.
I reverted Dan Kalowsky's patches.
Thank you,
Alexandra.
6:16 a.m.
New subject: [Powertop] segmentation fault when shutting
On 5/14/2014 12:26 PM, Kalowsky, Daniel wrote:
Not all patches are at fault here so reverting the entire series
seems short sighted. Even the patch in question isn't causing issues on all
platforms.
reverting is I think the right thing.
Lets revert, and then let these patches go through the normal mailing list review cycle
and then they can easily go back in ;-)
3049
days inactive
3055
days old
4 comments
5 participants
participants (5)
-
Alexandra Yates -
Arjan van de Ven -
Kalowsky, Daniel -
Martin Ziegler -
Sergey Senozhatsky