touch-packages team mailing list archive
-
touch-packages team
-
Mailing list archive
-
Message #70177
[Bug 1444178] Re: Crash when creating a page with sections
Sounds very much much like a memory corruption bug to me. With images,
you have larger memory requirements than with rectangles, meaning that
an over-run is much more likely.
--
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to ubuntu-ui-toolkit in
Ubuntu.
https://bugs.launchpad.net/bugs/1444178
Title:
Crash when creating a page with sections
Status in ubuntu-ui-toolkit package in Ubuntu:
New
Bug description:
See https://github.com/rschroll/beru/tree/section-crash for a test
case. This creates a page with a GridView and populates the gridview
with images loaded in from a javascript file. (This is to stand in
for loading from a database in the original.) This crashes most of
the time it is loaded for me. If the "model" line in the
head.sections block is commented out, it never crashes.
This is seen on a devel r1 image.
When run under gdb, I get this:
Program received signal SIGILL, Illegal instruction.
0xb306e368 in ?? () from /lib/arm-linux-gnueabihf/libcrypto.so.1.0.0
(gdb) bt
#0 0xb306e368 in ?? () from /lib/arm-linux-gnueabihf/libcrypto.so.1.0.0
#1 0xb306bec4 in OPENSSL_cpuid_setup ()
from /lib/arm-linux-gnueabihf/libcrypto.so.1.0.0
#2 0xb6fe222e in ?? () from /lib/ld-linux-armhf.so.3
#3 0x00039e60 in ?? ()
Backtrace stopped: previous frame identical to this frame (corrupt stack?)
However, this seems to occur before anything gets loaded, whereas the
usual crash happens as the images are loaded. Setting "handle SIGILL
nostop" gets it to the usual crash location, where it says:
Program received signal SIGSEGV, Segmentation fault.
0x00000000 in ?? ()
(gdb) bt
#0 0x00000000 in ?? ()
#1 0xb67ee750 in QApplicationPrivate::notify_helper(QObject*, QEvent*) ()
from /usr/lib/arm-linux-gnueabihf/libQt5Widgets.so.5
#2 0xb67f2520 in QApplication::notify(QObject*, QEvent*) ()
from /usr/lib/arm-linux-gnueabihf/libQt5Widgets.so.5
#3 0x00032df8 in ?? ()
Backtrace stopped: previous frame identical to this frame (corrupt stack?)
The fact that images are being loaded seems to be important. If I
replace the images with simple rectangles, the crash goes away. I
can't imagine that they're directly related; perhaps there is a timing
bug exposed by the time it takes to decode and render the images.
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/ubuntu-ui-toolkit/+bug/1444178/+subscriptions
References