aboutsummaryrefslogtreecommitdiff
path: root/pthreads/BUGS
diff options
context:
space:
mode:
authormarha <marha@users.sourceforge.net>2011-07-18 10:33:05 +0200
committermarha <marha@users.sourceforge.net>2011-07-18 10:33:05 +0200
commit88101146f2ec7d53ffb793e365f05097ffd35fd3 (patch)
treedb77eef572a00c62c8d9485c4786f682e20a2fd0 /pthreads/BUGS
parent772373ccc8da2b2019555228f4972cc0c5a885a4 (diff)
downloadvcxsrv-88101146f2ec7d53ffb793e365f05097ffd35fd3.tar.gz
vcxsrv-88101146f2ec7d53ffb793e365f05097ffd35fd3.tar.bz2
vcxsrv-88101146f2ec7d53ffb793e365f05097ffd35fd3.zip
cvs version of pthreads
Diffstat (limited to 'pthreads/BUGS')
-rw-r--r--pthreads/BUGS10
1 files changed, 9 insertions, 1 deletions
diff --git a/pthreads/BUGS b/pthreads/BUGS
index 29cee0033..285ba4eb9 100644
--- a/pthreads/BUGS
+++ b/pthreads/BUGS
@@ -22,9 +22,17 @@ Known bugs
Workaround: avoid using pthread_exit() in C++ applications. Exit
threads by dropping through the end of the thread routine.
-2. Cancellation problems in optimised code
+2. Cancellation problems in C++ builds
- Milan Gardian
+ [Note: It's not clear if this problem isn't simply due to the context
+ switch in pthread_cancel() which occurs unless the QueueUserAPCEx
+ library and driver are installed and used. Just like setjmp/longjmp,
+ this is probably not going to work well in C++. In any case, unless for
+ some very unusual reason you really must use the C++ build then please
+ use the C build pthreadVC2.dll or pthreadGC2.dll, i.e. for C++
+ applications.]
+
This is suspected to be a compiler bug in VC6.0, and also seen in
VC7.0 and VS .NET 2003. The GNU C++ compiler does not have a problem
with this, and it has been reported that the Intel C++ 8.1 compiler