blosc/blosc2.c: optimize the mutex lock with spin lock #1
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
guowangy
reviewed
Oct 19, 2022
| int rc2; | ||
|
|
||
| /* Initialize mutex and condition variable objects */ | ||
| pthread_spin_init(&context->count_spin, PTHREAD_PROCESS_SHARED); |
There was a problem hiding this comment.
PTHREAD_PROCESS_SHARED why we need this flag?
Owner
Author
There was a problem hiding this comment.
PTHREAD_PROCESS_SHAREDwhy we need this flag?
int pthread_spin_init(pthread_spinlock_t *lock, int pshared);
The pshared must be one of the following.
PTHREAD_PROCESS_PRIVATE/PTHREAD_PROCESS_SHARED
The 1st one has limited to the threads that are forked from the same process.
The 2nd one could be obtained by all the threads.
Reference Link: https://man7.org/linux/man-pages/man3/pthread_spin_init.3.html
blosc/context.h
Outdated
| pthread_t *threads; | ||
| struct thread_context *thread_contexts; /* only for user-managed threads */ | ||
| pthread_spinlock_t count_spin; | ||
| pthread_mutex_t count_mutex; |
There was a problem hiding this comment.
if count_mutex is unused, we should remove it.
Owner
Author
There was a problem hiding this comment.
if
count_mutexis unused, we should remove it.
Done.
jiebinn
force-pushed
the
mitigate_lock_contention
branch
from
bff61e0 to
5dffc7c
Compare
jiebinn
force-pushed
the
mitigate_lock_contention
branch
from
5dffc7c to
51fda93
Compare
The mutex lock cost lots of cycles in t_blosc_do_job when the thread number is large. The updating to global values of protected part is not that heavy. It will save lots of cycles if we Replace the mutex lock with spin lock. Apply the patch and test the c-blosc2/bench. $ ./bench/b2bench blosclz noshuffle single 160 83886080 4 Score gain: 7.4% CPU: ICX 8380 x 2 sockets Core number: 160 threads
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The mutex lock cost lots of cycles in t_blosc_do_job when the thread number is large. The updating to global values of protected part is not that heavy. It will save lots of cycles if we Replace the mutex lock with spin lock.
Apply the patch and test the c-blosc2/bench.
$ ./bench/b2bench blosclz noshuffle single 160 83886080 4
Score gain: 7.4%
CPU: ICX 8380 x 2 sockets
Core number: 160 threads