wait, no. It's GetThreadBacktraceContext's job to suspend the thread.

This commit is contained in:
Glenn Maynard
2004-09-08 21:17:43 +00:00
parent 3e7f55ae30
commit 4d829e3d29
@@ -364,13 +364,6 @@ void ForceCrashHandler( const char *reason )
void ForceCrashHandlerDeadlock( CString reason, uint64_t iID )
{
/* How can this possibly work without suspending the thread first? What
* stops the thread from returning from the function it was in when
* GetThreadBacktraceContext was called and then changing the stack?
*/
/* There's just no good way to pause other threads, that doesn't result in us
* being paused, weird stuff happening to the TTY, or kernel hung processes.
* We just hope that we backtrace in time (which we almost always do). */
#if defined(DARWIN)
SuspendThread(iID);
#endif