- Thread::Release() wrote a local variable on the calling thread but read it on
another thread, with no synchronization. Happily it has no non-test callers
so deleting it instead of trying to fix it (see bug for details).
- Thread::started_ similarly was racily being written to; replaced with a
running_ Event, and hid the accessor except for tests & legacy callers,
with a note about why it's a bad idea.
webrtc/base patched with:
git diff origin --relative=talk/base | patch -p1 -dwebrtc/base
followed by manual merge of 3 thunks that ran afoul of naming differences
between talk/base and webrtc/base.
BUG=3388
R=andrew@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/14589005
git-svn-id: http://webrtc.googlecode.com/svn/trunk@6236 4adac7df-926f-26a2-2b94-8c16560cd09d
Update Thread documentation, fix all subclasses that had a problem.
This is to avoid a data racing between the destructor modifying the vtable, and
Thread::PreRun calling virtual method Run at the same time.
For example:
[ RUN ] FileMediaEngineTest.TestGetCapabilities
==================
WARNING: ThreadSanitizer: data race on vptr (ctor/dtor vs virtual call) (pid=2967)
Read of size 8 at 0x7d480000bd00 by thread T1:
#0 talk_base::Thread::PreRun(void*) /mnt/data/b/build/slave/Linux_Tsan_v2/build/src/out/Release/../../talk/base/thread.cc:353 (libjingle_media_unittest+0x000000234da8)
Previous write of size 8 at 0x7d480000bd00 by main thread:
#0 talk_base::Thread::~Thread() /mnt/data/b/build/slave/Linux_Tsan_v2/build/src/out/Release/../../talk/base/thread.cc:158 (libjingle_media_unittest+0x00000023478c)
#1 ~RtpSenderReceiver /mnt/data/b/build/slave/Linux_Tsan_v2/build/src/out/Release/../../talk/media/base/filemediaengine.cc:122 (libjingle_media_unittest+0x0000001b551f)
...
RISK=P2
TESTED=try bots and tsan
BUG=2078,2080
R=fischman@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/2428004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@4999 4adac7df-926f-26a2-2b94-8c16560cd09d
