Segfaults during production of documentation

classic Classic list List threaded Threaded
9 messages Options
Reply | Threaded
Open this post in threaded view
|

Segfaults during production of documentation

Rik-4
Just an FYI, I've been seeing the occasional segfault during the production
of the documentation.

  GEN      doc/interpreter/interpft.pdf
fatal: caught signal Segmentation fault -- stopping myself...
octave exited with signal 11
  GEN      doc/interpreter/interpn.pdf

The actual file that causes the problem varies, but I'm pretty sure it is
always a pdf.  It seems to have started in the last week or so.

--Rik

Reply | Threaded
Open this post in threaded view
|

Re: Segfaults during production of documentation

Dmitri A. Sergatskov


On Sun, Jan 7, 2018 at 7:11 PM, Rik <[hidden email]> wrote:
Just an FYI, I've been seeing the occasional segfault during the production
of the documentation.

  GEN      doc/interpreter/interpft.pdf
fatal: caught signal Segmentation fault -- stopping myself...
octave exited with signal 11
  GEN      doc/interpreter/interpn.pdf

The actual file that causes the problem varies, but I'm pretty sure it is
always a pdf.  It seems to have started in the last week or so.

​I have noticed those too...​

 

--Rik


​Dmitri.
--

Reply | Threaded
Open this post in threaded view
|

Re: Segfaults during production of documentation

Daniel Sebald
On 01/07/2018 08:14 PM, Dmitri A. Sergatskov wrote:

>
>
> On Sun, Jan 7, 2018 at 7:11 PM, Rik <[hidden email]
> <mailto:[hidden email]>> wrote:
>
>     Just an FYI, I've been seeing the occasional segfault during the
>     production
>     of the documentation.
>
>        GEN      doc/interpreter/interpft.pdf
>     fatal: caught signal Segmentation fault -- stopping myself...
>     octave exited with signal 11
>        GEN      doc/interpreter/interpn.pdf
>
>     The actual file that causes the problem varies, but I'm pretty sure
>     it is
>     always a pdf.  It seems to have started in the last week or so.
>
>
> ​I have noticed those too...​

You've run without the -jX option of make?  The make process can be
deceiving when there is an error and -j8 option.

Anyway, I suspect this is a figure generation problem.  The file
interpft.pdf is a figure, not a bigger document.  That pdf-print is
using Mesa on your system?   (I.e., first creates an EPS file that is
used to create the PDF file?  On my system the EPS is just prior to the PDF:

-rw-r--r-- 1 root root 14496 Dec 29 18:16 interpft.eps
-rw-r--r-- 1 root root  5686 Dec 29 18:17 interpft.pdf

so I'm guessing that's how it works.)  This looks like the long-standing
occasional-crash-during-print issue of yester-year.

I noticed recently when reviewing this bug report:

https://savannah.gnu.org/bugs/?52781#comment3

that rescaling of the Qt and FLTK windows (i.e., drag the lower-right
corner around via mouse) looks very poor, flashing between white and
black backgrounds.  It's supposed to be white background all the time,
but I'm recalling that flashing black background a trait similar to the
bug I noted above--just a vague memory at this point.  How does the
rescaling of Qt and FLTK figures look for you?

Dan

Reply | Threaded
Open this post in threaded view
|

Re: Segfaults during production of documentation

Dmitri A. Sergatskov


On Sun, Jan 7, 2018 at 10:43 PM, Daniel J Sebald <[hidden email]> wrote:
On 01/07/2018 08:14 PM, Dmitri A. Sergatskov wrote:


On Sun, Jan 7, 2018 at 7:11 PM, Rik <[hidden email] <mailto:[hidden email]>> wrote:

    Just an FYI, I've been seeing the occasional segfault during the
    production
    of the documentation.

       GEN      doc/interpreter/interpft.pdf
    fatal: caught signal Segmentation fault -- stopping myself...
    octave exited with signal 11
       GEN      doc/interpreter/interpn.pdf

    The actual file that causes the problem varies, but I'm pretty sure
    it is
    always a pdf.  It seems to have started in the last week or so.


​I have noticed those too...​

You've run without the -jX option of make?  The make process can be deceiving when there is an error and -j8 option.

Anyway, I suspect this is a figure generation problem.  The file interpft.pdf is a figure, not a bigger document.  That pdf-print is using Mesa on your system?   (I.e., first creates an EPS file that is used to create the PDF file?  On my system the EPS is just prior to the PDF:

-rw-r--r-- 1 root root 14496 Dec 29 18:16 interpft.eps
-rw-r--r-- 1 root root  5686 Dec 29 18:17 interpft.pdf

so I'm guessing that's how it works.)  This looks like the long-standing occasional-crash-during-print issue of yester-year.

I noticed recently when reviewing this bug report:

https://savannah.gnu.org/bugs/?52781#comment3

that rescaling of the Qt and FLTK windows (i.e., drag the lower-right corner around via mouse) looks very poor, flashing between white and black backgrounds.  It's supposed to be white background all the time, but I'm recalling that flashing black background a trait similar to the bug I noted above--just a vague memory at this point.  How does the rescaling of Qt and FLTK figures look for you?

Dan



​Scaling looks fine. I do run make as:

LIBGL_ALWAYS_SOFTWARE=1 LD_PRELOAD=/usr/lib64/libGLX_mesa.so.0
​make -j8​


The old ​crashed due to​ bad GL required restarting the make. Now it just goes through.

Dmitri.
--


Reply | Threaded
Open this post in threaded view
|

Re: Segfaults during production of documentation

Dmitri A. Sergatskov
On Sun, Jan 7, 2018 at 10:55 PM, Dmitri A. Sergatskov <[hidden email]> wrote:


On Sun, Jan 7, 2018 at 10:43 PM, Daniel J Sebald <[hidden email]> wrote:
On 01/07/2018 08:14 PM, Dmitri A. Sergatskov wrote:


On Sun, Jan 7, 2018 at 7:11 PM, Rik <[hidden email] <mailto:[hidden email]>> wrote:

    Just an FYI, I've been seeing the occasional segfault during the
    production
    of the documentation.

       GEN      doc/interpreter/interpft.pdf
    fatal: caught signal Segmentation fault -- stopping myself...
    octave exited with signal 11
       GEN      doc/interpreter/interpn.pdf

    The actual file that causes the problem varies, but I'm pretty sure
    it is
    always a pdf.  It seems to have started in the last week or so.


​I have noticed those too...​

You've run without the -jX option of make?  The make process can be deceiving when there is an error and -j8 option.

Anyway, I suspect this is a figure generation problem.  The file interpft.pdf is a figure, not a bigger document.  That pdf-print is using Mesa on your system?   (I.e., first creates an EPS file that is used to create the PDF file?  On my system the EPS is just prior to the PDF:

-rw-r--r-- 1 root root 14496 Dec 29 18:16 interpft.eps
-rw-r--r-- 1 root root  5686 Dec 29 18:17 interpft.pdf

so I'm guessing that's how it works.)  This looks like the long-standing occasional-crash-during-print issue of yester-year.

I noticed recently when reviewing this bug report:

https://savannah.gnu.org/bugs/?52781#comment3

that rescaling of the Qt and FLTK windows (i.e., drag the lower-right corner around via mouse) looks very poor, flashing between white and black backgrounds.  It's supposed to be white background all the time, but I'm recalling that flashing black background a trait similar to the bug I noted above--just a vague memory at this point.  How does the rescaling of Qt and FLTK figures look for you?

Dan



​Scaling looks fine. I do run make as:

LIBGL_ALWAYS_SOFTWARE=1 LD_PRELOAD=/usr/lib64/libGLX_mesa.so.0
​make -j8​


The old ​crashed due to​ bad GL required restarting the make. Now it just goes through.

Dmitri.
--


​OK I was able to get plot windows to some funny state by shrinking it first to 0 (so no plot area is visible)
and then *quickly​* stretching it out. So it is not perfected yet...

Dmitri.
--



Reply | Threaded
Open this post in threaded view
|

Re: Segfaults during production of documentation

Daniel Sebald
On 01/07/2018 11:02 PM, Dmitri A. Sergatskov wrote:
> On Sun, Jan 7, 2018 at 10:55 PM, Dmitri A. Sergatskov
> <[hidden email] <mailto:[hidden email]>> wrote:
[snip]

>     ​Scaling looks fine. I do run make as:
>
>     LIBGL_ALWAYS_SOFTWARE=1 LD_PRELOAD=/usr/lib64/libGLX_mesa.so.0
>     ​make -j8​
>
>
>     The old ​crashed due to​ bad GL required restarting the make. Now it
>     just goes through.
>
>     Dmitri.
>     --
>
>
> ​OK I was able to get plot windows to some funny state by shrinking it
> first to 0 (so no plot area is visible)
> and then *quickly​* stretching it out. So it is not perfected yet...

That's a known bug:

https://savannah.gnu.org/bugs/?func=detailitem&item_id=52398

Please try again to produce a black background, in particular the FLTK
terminal.  After creating the plot, keep dragging the lower right corner
around for 30 s.  It seems about every seven or eight seconds there is a
black background.

Now that I try Qt graphics toolkit again, maybe the Qt g.t. terminal is
not flashing to a black background.  It's only the outside edges that
appear black as the window is expanded.  But the refresh on the graphics
routines should not be so slow that the black edges are so noticeable.
The Qt graphics is going directly to the OpenGL routines.  gnuplot I
could understand being slower because there is a lot going on between
its refresh...but gnuplot is pretty responsive.

Just speculation, but I wonder if there was some graphics bug for which
the solution was to introduce a delay in the refresh, and maybe in
offline mode there is no delay and the bug is re-emerging?

Dan

Reply | Threaded
Open this post in threaded view
|

Re: Segfaults during production of documentation

Dmitri A. Sergatskov
In reply to this post by Rik-4


On Sun, Jan 7, 2018 at 7:11 PM, Rik <[hidden email]> wrote:
Just an FYI, I've been seeing the occasional segfault during the production
of the documentation.

  GEN      doc/interpreter/interpft.pdf
fatal: caught signal Segmentation fault -- stopping myself...
octave exited with signal 11
  GEN      doc/interpreter/interpn.pdf

The actual file that causes the problem varies, but I'm pretty sure it is
always a pdf.  It seems to have started in the last week or so.

​Looking at buildbot logs I see those:

​Jan 06 17:01:22 amd systemd-coredump[28960]: Process 28416 (lt-octave-gui) of user 1002 dumped core.
                                             
                                             Stack trace of thread 28495:
                                             #0  0x00007febd276b641 _ZN16QCoreApplication9postEventEP7QObjectP6QEventi (libQt5Core.so.5)
                                             #1  0x00007febd279085f _ZN11QMetaObject8activateEP7QObjectiiPPv (libQt5Core.so.5)
                                             #2  0x00007febd57176e1 n/a (/home/buildbotu/fc25-x86_64/gcc-fedora/build/libgui/.libs/liboctgui.so.2.0.0)


 

--Rik

​Dmitri.

Reply | Threaded
Open this post in threaded view
|

Re: Segfaults during production of documentation

Dmitri A. Sergatskov
In reply to this post by Daniel Sebald


On Sun, Jan 7, 2018 at 11:27 PM, Daniel J Sebald <[hidden email]> wrote:

That's a known bug:

https://savannah.gnu.org/bugs/?func=detailitem&item_id=52398

​Yes. This is it. Thanks.


Please try again to produce a black background, in particular the FLTK terminal.  After creating the plot, keep dragging the lower right corner around for 30 s.  It seems about every seven or eight seconds there is a black background.

​I get black stripes (like the plot windows is lagging behind the canvas). I cannot get enire plot black.
The "lagging" looks worse with hardware driver than with llvmpipe.​

​fltk looks more responsive to me than qt,

Dan

​Dmitri.
--

Reply | Threaded
Open this post in threaded view
|

Re: Segfaults during production of documentation

Dmitri A. Sergatskov
​Here is somewhat a longer trace (running as normal user):

​Process 23933 (lt-octave-gui) of user 1001 dumped core.
                                                   
                                                   Stack trace of thread 24145:
                                                   #0  0x00007fb291116641 _ZN16QCoreApplication9postEventEP7QObjectP6QEventi (libQt5Core.so.5)
                                                   #1  0x00007fb29113b85f _ZN11QMetaObject8activateEP7QObjectiiPPv (libQt5Core.so.5)
                                                   #2  0x00007fb293ed32e1 _ZN18octave_interpreter22octave_finished_signalEi (liboctgui.so.2)
                                                   #3  0x00007fb293ea3391 _ZN18octave_interpreter7executeEv (liboctgui.so.2)
                                                   #4  0x00007fb29113c91a _ZN7QObject5eventEP6QEvent (libQt5Core.so.5)
                                                   #5  0x00007fb291d7c6ac _ZN19QApplicationPrivate13notify_helperEP7QObjectP6QEvent (libQt5Widgets.so.5)
                                                   #6  0x00007fb291d83ef4 _ZN12QApplication6notifyEP7QObjectP6QEvent (libQt5Widgets.so.5)
                                                   #7  0x00007fb291113367 _ZN16QCoreApplication15notifyInternal2EP7QObjectP6QEvent (libQt5Core.so.5)
                                                   #8  0x00007fb291115aeb _ZN23QCoreApplicationPrivate16sendPostedEventsEP7QObjectiP11QThreadData (libQt5Core.so.5)
                                                   #9  0x00007fb291164553 _ZL23postEventSourceDispatchP8_GSourcePFiPvES1_ (libQt5Core.so.5)
                                                   #10 0x00007fb282a2bbb7 g_main_context_dispatch (libglib-2.0.so.0)
                                                   #11 0x00007fb282a2bf60 g_main_context_iterate.isra.25 (libglib-2.0.so.0)
                                                   #12 0x00007fb282a2bfec g_main_context_iteration (libglib-2.0.so.0)
                                                   #13 0x00007fb29116433f _ZN20QEventDispatcherGlib13processEventsE6QFlagsIN10QEventLoop17ProcessEventsFlagEE (libQt5Core.so.5)
                                                   #14 0x00007fb2911120ea _ZN10QEventLoop4execE6QFlagsINS_17ProcessEventsFlagEE (libQt5Core.so.5)
                                                   #15 0x00007fb290f638ba _ZN7QThread4execEv (libQt5Core.so.5)
                                                   #16 0x00007fb290f67b92 _ZN14QThreadPrivate5startEPv (libQt5Core.so.5)
                                                   #17 0x00007fb28895761b start_thread (libpthread.so.0)
                                                   #18 0x00007fb28868491f __clone (libc.so.6)
                                                   
                                                   Stack trace of thread 23933:
                                                   #0  0x00007fb28895dc4b pthread_cond_wait@@GLIBC_2.3.2 (libpthread.so.0)
                                                   #1  0x00007fb290f682eb _ZN14QWaitCondition4waitEP6QMutexm (libQt5Core.so.5)
                                                   #2  0x00007fb290f6231b _ZN10QSemaphore7acquireEi (libQt5Core.so.5)
                                                   #3  0x00007fb29113ba3b _ZN11QMetaObject8activateEP7QObjectiiPPv (libQt5Core.so.5)
                                                   #4  0x00007fb29113c2df _ZN7QObject9destroyedEPS_ (libQt5Core.so.5)
                                                   #5  0x00007fb291142ef9 _ZN7QObjectD1Ev (libQt5Core.so.5)
                                                   #6  0x00007fb25d421bb9 _ZN7Adwaita5StyleD0Ev (adwaita.so)
                                                   #7  0x00007fb291d7d033 _ZN12QApplicationD1Ev (libQt5Widgets.so.5)
                                                   #8  0x00007fb293ef0fd8 _ZN6octave15gui_application7executeEv (liboctgui.so.2)
                                                   #9  0x000000000040134c main (lt-octave-gui)
                                                   #10 0x00007fb28858e00a __libc_start_main (libc.so.6)
                                                   #11 0x000000000040160a _start (lt-octave-gui)
                                                   
                                                   Stack trace of thread 24083:
                                                   #0  0x00007fb28867836b __poll (libc.so.6)
                                                   #1  0x00007fb282a2bed9 g_main_context_iterate.isra.25 (libglib-2.0.so.0)
                                                   #2  0x00007fb282a2c272 g_main_loop_run (libglib-2.0.so.0)
                                                   #3  0x00007fb263170b36 gdbus_shared_thread_func (libgio-2.0.so.0)
                                                   #4  0x00007fb282a534c6 g_thread_proxy (libglib-2.0.so.0)
                                                   #5  0x00007fb28895761b start_thread (libpthread.so.0)
                                                   #6  0x00007fb28868491f __clone (libc.so.6)
                                                   
                                                   Stack trace of thread 24041:
                                                   #0  0x00007fb28867836b __poll (libc.so.6)
                                                   #1  0x00007fb286a90fe7 _xcb_conn_wait (libxcb.so.1)
                                                   #2  0x00007fb286a92dda xcb_wait_for_event (libxcb.so.1)
                                                   #3  0x00007fb26c065a49 _ZN15QXcbEventReader3runEv (libQt5XcbQpa.so.5)
                                                   #4  0x00007fb290f67b92 _ZN14QThreadPrivate5startEPv (libQt5Core.so.5)
                                                   #5  0x00007fb28895761b start_thread (libpthread.so.0)
                                                   #6  0x00007fb28868491f __clone (libc.so.6)
                                                   
                                                   Stack trace of thread 24144:
                                                   #0  0x00007fb28867836b __poll (libc.so.6)
                                                   #1  0x00007fb282a2bed9 g_main_context_iterate.isra.25 (libglib-2.0.so.0)
                                                   #2  0x00007fb282a2bfec g_main_context_iteration (libglib-2.0.so.0)
                                                   #3  0x00007fb29116433f _ZN20QEventDispatcherGlib13processEventsE6QFlagsIN10QEventLoop17ProcessEventsFlagEE (libQt5Core.so.5)
                                                   #4  0x00007fb2911120ea _ZN10QEventLoop4execE6QFlagsINS_17ProcessEventsFlagEE (libQt5Core.so.5)
                                                   #5  0x00007fb290f638ba _ZN7QThread4execEv (libQt5Core.so.5)
                                                   #6  0x00007fb26bdb7479 _ZN22QDBusConnectionManager3runEv (libQt5DBus.so.5)
                                                   #7  0x00007fb290f67b92 _ZN14QThreadPrivate5startEPv (libQt5Core.so.5)
                                                   #8  0x00007fb28895761b start_thread (libpthread.so.0)
                                                   #9  0x00007fb28868491f __clone (libc.so.6)
                                                   
                                                   Stack trace of thread 24076:
                                                   #0  0x00007fb28867836b __poll (libc.so.6)
                                                   #1  0x00007fb282a2bed9 g_main_context_iterate.isra.25 (libglib-2.0.so.0)
                                                   #2  0x00007fb282a2bfec g_main_context_iteration (libglib-2.0.so.0)
                                                   #3  0x00007fb26055d42d dconf_gdbus_worker_thread (libdconfsettings.so)
                                                   #4  0x00007fb282a534c6 g_thread_proxy (libglib-2.0.so.0)
                                                   #5  0x00007fb28895761b start_thread (libpthread.so.0)
                                                   #6  0x00007fb28868491f __clone (libc.so.6)
                                                   
                                                   Stack trace of thread 24146:
                                                   #0  0x00007fb288962616 sigwait (libpthread.so.0)
                                                   #1  0x00007fb2903a4d12 signal_watcher (liboctave.so.4)
                                                   #2  0x00007fb28895761b start_thread (libpthread.so.0)
                                                   #3  0x00007fb28868491f __clone (libc.so.6)
                                                  
                                                   Stack trace of thread 24078:
                                                   #0  0x00007fb28867836b __poll (libc.so.6)
                                                   #1  0x00007fb282a2bed9 g_main_context_iterate.isra.25 (libglib-2.0.so.0)
                                                   #2  0x00007fb282a2bfec g_main_context_iteration (libglib-2.0.so.0)
                                                   #3  0x00007fb282a2c031 glib_worker_main (libglib-2.0.so.0)
                                                   #4  0x00007fb282a534c6 g_thread_proxy (libglib-2.0.so.0)
                                                   #5  0x00007fb28895761b start_thread (libpthread.so.0)
                                                   #6  0x00007fb28868491f __clone (libc.so.6)


​Dmitri.
--