cvs commit: src/lib/libthr/thread thr_mutex.c src/lib/libkse/thread thr_mutex.c src/include pthread.h

[Kris Kennaway]

David Xu wrote:
> Kris Kennaway wrote:
> 
>>
>> PTHREAD_ADAPTIVE_MUTEX_INITIALIZER_NP is just in <pthread.h>.
>>
>> /*
>>  * Static initialization values.
>>  */
>> #define PTHREAD_MUTEX_INITIALIZER       NULL
>> #define PTHREAD_ADAPTIVE_MUTEX_INITIALIZER_NP   NULL
>> #define PTHREAD_COND_INITIALIZER        NULL
>> #define PTHREAD_RWLOCK_INITIALIZER      NULL
>>
>>>>> I remembered mysql uses this macro to initialize spin mutex, and you
>>>>> indead needs a patch to let it work
>>>>
>>>>
>>>> No, with the code I committed mysql detects and uses it out of the 
>>>> box, without requiring any patches.  It is easy to measure the 
>>>> resulting 30% performance improvement at high loads ;-)
>>>>
>>> see above, I didn't see any code set PTHREAD_MUTEX_ADAPTIVE_NP type.
>>
>> The code is already in mysql for use with glibc.  It basically does
>>
>> #ifdef PTHREAD_ADAPTIVE_MUTEX_INITIALIZER_NP
>> /*
>>  * Use PTHREAD_MUTEX_ADAPTIVE_NP for the mutexes we know will benefit
>>  * from it
>>  */
>> ...
>> #endif
>>
>> so it just works.
>>
> 
> I can not find code in libthr set mutex's member field m_type to
> PTHREAD_MUTEX_ADAPTIVE_NP by your change, so how can it work as
> expected ?

mysql does:

#ifdef PTHREAD_ADAPTIVE_MUTEX_INITIALIZER_NP
   /*
     Set mutex type to "fast" a.k.a "adaptive"

     In this case the thread may steal the mutex from some other thread
     that is waiting for the same mutex.  This will save us some
     context switches but may cause a thread to 'starve forever' while
     waiting for the mutex (not likely if the code within the mutex is
     short).
   */
   pthread_mutexattr_init(&my_fast_mutexattr);
   pthread_mutexattr_settype(&my_fast_mutexattr,
                             PTHREAD_MUTEX_ADAPTIVE_NP);
#endif

Kris