2.3.3_rc4 - cpu goes up to 100% - sampling process
Hi Jeremy, I did disable all pasteboard options, killed klipper, no proxy in .xinitrc - but still the same pattern, here are the actual sampling process: :-( Mery Christmas :-) christof Sampling process 1748 for 3 seconds with 1 millisecond of run time between samples Sampling completed, processing symbols... Analysis of sampling X11.bin (pid 1748) every 1 millisecond Call graph: 1910 Thread_2503 1910 start 1910 main 1910 mach_msg_server 1910 mach_startup_server 1910 _Xstart_x11_server 1910 do_start_x11_server 1910 server_main 1910 X11ApplicationMain 1910 -[NSApplication run] 1910 -[NSApplication nextEventMatchingMask:untilDate:inMode:dequeue:] 1910 _DPSNextEvent 1910 BlockUntilNextEventMatchingListInMode 1910 ReceiveNextEventCommon 1910 RunCurrentEventLoopInMode 1910 CFRunLoopRunInMode 1910 CFRunLoopRunSpecific 1861 mach_msg 1861 mach_msg_trap 1861 mach_msg_trap 49 0xffffffff 49 _sigtramp 49 _sigtramp 1910 Thread_2603 1910 thread_start 1910 _pthread_start 1910 CAPThread::Entry(CAPThread*) 1910 HALRunLoop::OwnThread(void*) 1910 CFRunLoopRunInMode 1910 CFRunLoopRunSpecific 1910 mach_msg 1910 mach_msg_trap 1910 mach_msg_trap 1910 Thread_2703 1910 thread_start 1910 _pthread_start 1910 server_thread 1910 dix_main 1910 Dispatch 1763 select$DARWIN_EXTSN 1763 select$DARWIN_EXTSN 143 WaitForSomething 83 BlockHandler 59 QuartzBlockHandler 32 NSPopAutoreleasePool 18 NSPopAutoreleasePool 4 objc_assign_strongCast 4 objc_assign_strongCast 3 NSClassFromObject 2 -[NSObject class] 2 -[NSObject class] 1 object_getClass 1 object_getClass 3 objc_msgSend 3 objc_msgSend 2 objc_collecting_enabled 2 objc_collecting_enabled 1 _CFExecutableLinkedOnOrAfter 1 _CFExecutableLinkedOnOrAfter 1 object_getClass 1 object_getClass 9 objc_msgSend 9 objc_msgSend 6 -[NSAutoreleasePool init] 5 -[NSAutoreleasePool initWithCapacity:] 3 NSPushAutoreleasePool 3 NSPushAutoreleasePool 1 -[NSAutoreleasePool initWithCapacity:] 1 -[NSObject class] 1 -[NSObject class] 1 -[NSAutoreleasePool init] 4 objc_collecting_enabled 4 objc_collecting_enabled 3 QuartzBlockHandler 2 +[NSObject alloc] 2 +[NSObject alloc] 2 -[NSAutoreleasePool release] 2 -[NSAutoreleasePool release] 1 dyld_stub_objc_collecting_enabled 1 dyld_stub_objc_collecting_enabled 12 AnimCurScreenBlockHandler 10 AnimCurScreenBlockHandler 2 miSpriteBlockHandler 2 miSpriteBlockHandler 7 objc_msgSend 7 objc_msgSend 3 BlockHandler 1 RootlessBlockHandler 1 RootlessBlockHandler 1 dyld_stub_objc_msgSend 1 dyld_stub_objc_msgSend 19 WaitForSomething 18 GetTimeInMillis 17 gettimeofday 14 __gettimeofday 8 __nanotime 8 __nanotime 6 __gettimeofday 3 gettimeofday 1 GetTimeInMillis 11 WakeupHandler 8 WakeupHandler 2 QuartzWakeupHandler 2 QuartzWakeupHandler 1 NoopDDA 1 NoopDDA 6 select$UNIX2003 6 select$UNIX2003 5 __bzero 5 __bzero 1 select$DARWIN_EXTSN 1 select$DARWIN_EXTSN 2 Dispatch 1 __error 1 __error 1 memset 1 memset 1910 Thread_2803 1910 thread_start 1910 _pthread_start 1910 glvmDoWork 1910 pthread_cond_wait$UNIX2003 1910 __semwait_signal 1910 __semwait_signal 1910 Thread_2903 1910 thread_start 1910 _pthread_start 1910 DarwinProcessFDAdditionQueue_thread 1910 pthread_cond_wait$UNIX2003 1910 __semwait_signal 1910 __semwait_signal 1910 Thread_2a03 1910 thread_start 1910 _pthread_start 1910 xpbproxy_input_thread 1910 pthread_cond_wait$UNIX2003 1910 __semwait_signal 1910 __semwait_signal 1910 Thread_2b03 1910 thread_start 1910 _pthread_start 1910 select$DARWIN_EXTSN 1910 select$DARWIN_EXTSN 1910 Thread_2c03 1910 thread_start 1910 _pthread_start 1910 _xp_async_thread 1910 _xp_async_dequeue 1910 pthread_cond_wait$UNIX2003 1910 __semwait_signal 1910 __semwait_signal Total number in stack (recursive counted multiple, when >=5): 7 _pthread_start 7 thread_start Sort by top of stack, same collapsed (when >= 5): __semwait_signal 7640 mach_msg_trap 3771 select$DARWIN_EXTSN 3674 _sigtramp 49 WaitForSomething 19 objc_msgSend 19 NSPopAutoreleasePool 18 AnimCurScreenBlockHandler 10 WakeupHandler 8 __nanotime 8 __gettimeofday 6 objc_collecting_enabled 6 select$UNIX2003 6 __bzero 5 Sample analysis of process 1748 written to file /dev/stdout -- public key www.hfph.mwn.de/~chwolf/chwolf.asc Please respect my privacy and do not make my contact information available to third parties.
Christof, Does this condition arise randomly, or do you know what to do to provoke it? I am still unable to reproduce it, but I swear I saw the same thing 3-4 times after I installed rc3. It was before I knew how to sample it, so I cannot claim it was the same, but I saw the 100% cpu load by the idle process -- by idle I mean undisturbed by user events. It is interesting that it did go down substantially while I used it, and returned to 100% as soon as I left it alone. Also, I suspect that it is caused by something outside X11, because whenever this condition arose, restarting the server did not help; it went away only after system restart. --Gene
Is it X11.bin that is using the CPU? The sample provided in the previous email shows it waiting almost 99% of the time... is it another process that's using up the CPU? Or am I missing something here... Sort by top of stack, same collapsed (when >= 5): __semwait_signal 7640 mach_msg_trap 3771 select$DARWIN_EXTSN 3674 _sigtramp 49 WaitForSomething 19 objc_msgSend 19 NSPopAutoreleasePool 18 AnimCurScreenBlockHandler 10 WakeupHandler 8 __nanotime 8 __gettimeofday 6 objc_collecting_enabled 6 select$UNIX2003 6 __bzero 5 On Dec 25, 2008, at 06:36, Gene Selkov wrote:
Christof,
Does this condition arise randomly, or do you know what to do to provoke it? I am still unable to reproduce it, but I swear I saw the same thing 3-4 times after I installed rc3. It was before I knew how to sample it, so I cannot claim it was the same, but I saw the 100% cpu load by the idle process -- by idle I mean undisturbed by user events. It is interesting that it did go down substantially while I used it, and returned to 100% as soon as I left it alone.
Also, I suspect that it is caused by something outside X11, because whenever this condition arose, restarting the server did not help; it went away only after system restart.
--Gene _______________________________________________ Xquartz-dev mailing list Xquartz-dev@lists.macosforge.org http://lists.macosforge.org/mailman/listinfo.cgi/xquartz-dev
participants (3)
-
Christof Wolf
-
Gene Selkov
-
Jeremy Huddleston