[LabPlot2] [Bug 398342] ProjectImportTest segfaults while loading Excel file

[Alexander Semke]

https://bugs.kde.org/show_bug.cgi?id=398342

Alexander Semke  changed:

   What|Removed |Added

 Resolution|--- |FIXED
   Version Fixed In||2.6
  Latest Commit||https://commits.kde.org/lab
   ||plot/9d8af4646a2b42d4324830
   ||76a2045125a84ecf49
 Status|CONFIRMED   |RESOLVED

--- Comment #11 from Alexander Semke  ---
Git commit 9d8af4646a2b42d432483076a2045125a84ecf49 by Alexander Semke.
Committed on 20/09/2018 at 20:56.
Pushed by asemke into branch 'master'.

Correctly handle big endians on ppc64. Fix contributed by Miquel.
FIXED-IN: 2.6

M  +11   -3liborigin/OriginAnyParser.cpp

https://commits.kde.org/labplot/9d8af4646a2b42d432483076a2045125a84ecf49

-- 
You are receiving this mail because:
You are watching all bug changes.

[LabPlot2] [Bug 398342] ProjectImportTest segfaults while loading Excel file

[awilfox]

https://bugs.kde.org/show_bug.cgi?id=398342

--- Comment #10 from A. Wilcox (awilfox)  ---
I'm terribly sorry, for some reason I didn't get any bugmail that there was a
second patch.

Yes, this patch does fix it!  The project import test passes on ppc64 now. 
Thank you so much for your effort.

-- 
You are receiving this mail because:
You are watching all bug changes.

[LabPlot2] [Bug 398342] ProjectImportTest segfaults while loading Excel file

[Alexander Semke]

https://bugs.kde.org/show_bug.cgi?id=398342

--- Comment #9 from Alexander Semke  ---
can you please confirm the second version of the patch solves the problem for
you, too?

-- 
You are receiving this mail because:
You are watching all bug changes.

[LabPlot2] [Bug 398342] ProjectImportTest segfaults while loading Excel file

[Alexander Semke]

https://bugs.kde.org/show_bug.cgi?id=398342

--- Comment #8 from Alexander Semke  ---
Created attachment 114936
  --> https://bugs.kde.org/attachment.cgi?id=114936&action=edit
patch, second version

-- 
You are receiving this mail because:
You are watching all bug changes.

[LabPlot2] [Bug 398342] ProjectImportTest segfaults while loading Excel file

[Alexander Semke]

https://bugs.kde.org/show_bug.cgi?id=398342

Alexander Semke  changed:

   What|Removed |Added

 Ever confirmed|0   |1
 Status|UNCONFIRMED |CONFIRMED

--- Comment #7 from Alexander Semke  ---
(In reply to A. Wilcox (awilfox) from comment #6)
> It is not possible to "mix" endians, unfortunately, so there is no way for
> us to compile LabPlot in LE mode.
> 
> I applied the attached patch to our tree and rebuilt.  The exact same stack
> trace was present; only the memory addresses were different, due to ASLR.
Ok. We'll have a more deeper look into this.


> Attempting to pass -DENABLE_LIBORIGIN=OFF to CMake resulted in a build error:
> [...]
I fixed the build issue with
https://phabricator.kde.org/R262:d67faf643b2e69c017c21e9d9aaa1fe2d3de6b0e . Can
you please try with this fix?

-- 
You are receiving this mail because:
You are watching all bug changes.

[LabPlot2] [Bug 398342] ProjectImportTest segfaults while loading Excel file

[awilfox]

https://bugs.kde.org/show_bug.cgi?id=398342

--- Comment #6 from A. Wilcox (awilfox)  ---
It is not possible to "mix" endians, unfortunately, so there is no way for us
to compile LabPlot in LE mode.

I applied the attached patch to our tree and rebuilt.  The exact same stack
trace was present; only the memory addresses were different, due to ASLR.

Attempting to pass -DENABLE_LIBORIGIN=OFF to CMake resulted in a build error:

In file included from
/usr/src/packages/user/labplot/src/labplot-2.5.0/src/backend/datasources/projects/OriginProjectParser.h:34:0,
 from
/usr/src/packages/user/labplot/src/labplot-2.5.0/src/kdefrontend/MainWin.cpp:40:
/usr/src/packages/user/labplot/src/labplot-2.5.0/liborigin/OriginFile.h:34:30:
fatal error: liborigin/config.h: No such file or directory
 #include "liborigin/config.h"
  ^
compilation terminated.
make[2]: *** [src/CMakeFiles/labplot2lib.dir/build.make:333:
src/CMakeFiles/labplot2lib.dir/kdefrontend/MainWin.cpp.o] Error 1
make[2]: *** Waiting for unfinished jobs
In file included from
/usr/src/packages/user/labplot/src/labplot-2.5.0/src/backend/datasources/projects/OriginProjectParser.h:34:0,
 from
/usr/src/packages/user/labplot/src/labplot-2.5.0/src/kdefrontend/datasources/ImportProjectDialog.cpp:33:
/usr/src/packages/user/labplot/src/labplot-2.5.0/liborigin/OriginFile.h:34:30:
fatal error: liborigin/config.h: No such file or directory
 #include "liborigin/config.h"
  ^
compilation terminated.
make[2]: *** [src/CMakeFiles/labplot2lib.dir/build.make:541:
src/CMakeFiles/labplot2lib.dir/kdefrontend/datasources/ImportProjectDialog.cpp.o]
Error 1

-- 
You are receiving this mail because:
You are watching all bug changes.

[LabPlot2] [Bug 398342] ProjectImportTest segfaults while loading Excel file

[Alexander Semke]

https://bugs.kde.org/show_bug.cgi?id=398342

--- Comment #5 from Alexander Semke  ---
Can you please apply the attached patch and run the tests again?

-- 
You are receiving this mail because:
You are watching all bug changes.

[LabPlot2] [Bug 398342] ProjectImportTest segfaults while loading Excel file

[Alexander Semke]

https://bugs.kde.org/show_bug.cgi?id=398342

--- Comment #4 from Alexander Semke  ---
Created attachment 114901
  --> https://bugs.kde.org/attachment.cgi?id=114901&action=edit
fix big endian import in liborigin

-- 
You are receiving this mail because:
You are watching all bug changes.

[LabPlot2] [Bug 398342] ProjectImportTest segfaults while loading Excel file

[Alexander Semke]

https://bugs.kde.org/show_bug.cgi?id=398342

--- Comment #3 from Alexander Semke  ---
Looks like our code that parses Origin's binary files is not ready for
big-endian ppc64 architecture. I'm not sure we can easily and quickly fix this
now. A pragmatic solution would be now to switch off the import of Origin's
files on ppc64. This would switch off one single feature and would unblock you
from shipping LabPlot in Adélie Linux. In parallel we'll look for a solution of
this problem in our code. Alternatively, compile and LabPlot in the LE-mode -
not sure this is possible, I'm not an expert here... What do you think about
this?

-- 
You are receiving this mail because:
You are watching all bug changes.

[LabPlot2] [Bug 398342] ProjectImportTest segfaults while loading Excel file

[awilfox]

https://bugs.kde.org/show_bug.cgi?id=398342

--- Comment #2 from A. Wilcox (awilfox)  ---
awilcox on gwyn [pts/15 Sat 8 1:57] labplot-2.5.0:
tests/import_export/project/projectimporttest
* Start testing of ProjectImportTest *
Config: Using QtTest library 5.9.6, Qt 5.9.6 (power64-big_endian-lp64 shared
(dynamic) debug build; by GCC 6.4.0)
PASS   : ProjectImportTest::initTestCase()
QDEBUG : ProjectImportTest::testOrigin01() "Starting the import of
/usr/src/packages/user/labplot/src/labplot-2.5.0/tests/import_export/project/data/origin8_test_tree_import.opj"
OriginProjectParser::load()

= Received signal, dumping stack ==
GNU gdb (GDB) 8.2
Copyright (C) 2018 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later 
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.
Type "show copying" and "show warranty" for details.
This GDB was configured as "powerpc64-foxkit-linux-musl".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
.
Find the GDB manual and other documentation resources online at:
.

For help, type "help".
Type "apropos word" to search for commands related to "word".
Attaching to process 25556
[New LWP 25557]
[New LWP 25558]
__cp_end () at src/thread/powerpc64/syscall_cp.s:32
(gdb) 
Thread 3 (LWP 25558):
#0  __cp_end () at src/thread/powerpc64/syscall_cp.s:32
No locals.
#1  0x3fffbd7642e4 in __syscall_cp_c (nr=128, u=2, v=70367527851920, w=0,
x=0, y=0, z=0) at src/thread/pthread_cancel.c:35
self = 
r = 
st = 0
#2  0x3fffbd7630d8 in __syscall_cp (nr=, u=,
v=, w=, x=, y=,
z=)
at src/thread/__syscall_cp.c:20
No locals.
#3  0x3fffbd7631f4 in __timedwait_cp (addr=0x3fffb7805804, val=, clk=, at=, priv=128) at
src/thread/__timedwait.c:31
r = 
to = {tv_sec = 299, tv_nsec = 99168}
top = 
#4  0x3fffbd70d028 in __pthread_cond_timedwait (c=0x3fffb99a4c28,
m=0x3fffb99a4c00, ts=0x3fffb78058d8) at src/thread/pthread_cond_timedwait.c:105
node = {prev = 0x0, next = 0x0, state = 0, barrier = 2, notify = 0x0}
e = 
seq = 2
clock = 1
cs = 0
shared = 0
oldstate = 
tmp = 
fut = 0x3fffb7805804
#5  0x3fffbb4816f8 in QWaitConditionPrivate::wait_relative (time=30,
this=0x3fffb99a4c00) at thread/qwaitcondition_unix.cpp:133
ti = {tv_sec = 5797241, tv_nsec = 374265993}
ti = 
#6  QWaitConditionPrivate::wait (time=30, this=0x3fffb99a4c00) at
thread/qwaitcondition_unix.cpp:141
code = 
code = 
#7  QWaitCondition::wait (this=, mutex=0x3fffb99a49f8,
time=30) at thread/qwaitcondition_unix.cpp:215
__PRETTY_FUNCTION__ = "bool QWaitCondition::wait(QMutex*, long unsigned
int)"
returnValue = 
#8  0x3fffbd6234d0 in QTest::WatchDog::run (this=0x3fffb99a49e0) at
qtestcase.cpp:988
t = 
locker = {val = 70367563106809}
__PRETTY_FUNCTION__ = "virtual void QTest::WatchDog::run()"
#9  0x3fffbb480c88 in QThreadPrivate::start (arg=0x3fffb99a49e0) at
thread/qthread_unix.cpp:368
__cb = {__f = 0x3fffbb47f04c , __x =
0x3fffb99a49e0, __next = 0x0}
thr = 0x3fffb99a49e0
data = 
#10 0x3fffbd76483c in start (p=0x3fffb7805a88) at
src/thread/pthread_create.c:147
self = 0x3fffb7805a88
#11 0x3fffbd7705a8 in __clone () at src/thread/powerpc64/clone.s:43
No locals.

Thread 2 (LWP 25557):
#0  __cp_end () at src/thread/powerpc64/syscall_cp.s:32
No locals.
#1  0x3fffbd764374 in __syscall_cp_c (nr=167, u=-1, v=0, w=0, x=0, y=0,
z=0) at src/thread/pthread_cancel.c:35
self = 
r = 
st = 0
#2  0x3fffbd7630d8 in __syscall_cp (nr=, u=,
v=, w=, x=, y=,
z=)
at src/thread/__syscall_cp.c:20
No locals.
#3  0x3fffbd7520b4 in poll (fds=, n=,
timeout=) at src/select/poll.c:10
No locals.
#4  0x3fffb9bec6e0 in _xcb_conn_wait (c=0x12d29bec0, cond=,
vector=0x0, count=0x0) at xcb_conn.c:479
ret = 
fd = {fd = 3, events = 1, revents = 0}
#5  0x3fffb9bef074 in xcb_wait_for_event (c=) at
xcb_in.c:697
ret = 0x0
#6  0x3fffb7bb9a98 in QXcbEventReader::run (this=0x3fffbd607a60) at
qxcbconnection.cpp:1330
event = 0x0
#7  0x3fffbb480c88 in QThreadPrivate::start (arg=0x3fffbd607a60) at
thread/qthread_unix.cpp:368
__cb = {__f = 0x3fffbb47f04c , __x =
0x3fffbd607a60, __next = 0x0}
thr = 0x3fffbd607a60
data = 
#8  0x3fffbd76483c in start (p=0x3fffb786ea88) at
src/thread/pthread_create.c:147
self = 0x3fffb786ea88
#9  0x3fffbd7705a8 in __clone () at src/thread/powerpc64/clone.s:43
No locals.

Thread 1 (LWP 25556):
#0  __cp_end () at src/thread/powerpc64/syscall_cp.s:32
No locals.
#1  0x3fffbd764374 in __syscall_c

[LabPlot2] [Bug 398342] ProjectImportTest segfaults while loading Excel file

[Alexander Semke]

https://bugs.kde.org/show_bug.cgi?id=398342

--- Comment #1 from Alexander Semke  ---
Can you please do a debug build and execute the tests again?

-- 
You are receiving this mail because:
You are watching all bug changes.