mir/libreoffice
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

python3: disable check in PyThreadState_Swap

Browse Source 
This check is triggered by nested pyuno PyThreadAttach instances.
The assertion is basically about having multiple PyThreadState instances
per OS thread.  Hopefully this is not a "real" problem and the other
checks in PyEval_ReleaseThread/PyEval_AcquireThread will find all "real"
problems.

http://www.mail-archive.com/libreoffice@lists.freedesktop.org/msg62195.html

Change-Id: Ia82135f37f55ea69b545a83098619939869cb7c5
Reviewed-on: https://gerrit.libreoffice.org/3453
Reviewed-by: David Ostrovsky <David.Ostrovsky@gmx.de>
Tested-by: David Ostrovsky <David.Ostrovsky@gmx.de>
This commit is contained in:
Michael Stahl
2013-04-18 12:23:22 +02:00
committed by David Ostrovsky
parent 1fb53a6375
commit e83990a872
2 changed files with 16 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

15
python3/python-3.3.0-pythreadstate.patch.1 Normal file
View File

@@ -0,0 +1,15 @@
disable a check in PyThreadState_Swap that is hopefully not a "real" problem
--- python3/Python/pystate.c 2013-04-17 22:45:00.799800000 +0200
+++ python3/Python/pystate.c 2013-04-17 22:46:53.743800000 +0200
@@ -437,7 +437,9 @@
to be used for a thread. Check this the best we can in debug
builds.
*/
-#if defined(Py_DEBUG) && defined(WITH_THREAD)
+#if defined(Py_DEBUG) && defined(WITH_THREAD) && 0
+ /* disable this for LO - it is triggered by nested PyThreadAttach
+ which do not appear to be a real problem */
if (newts) {
/* This can be called from PyEval_RestoreThread(). Similar
to it, we need to ensure errno doesn't change.