Hi Changwei Ge,
On Thu, Aug 22, 2019 at 5:57 PM Changwei Ge <chge@xxxxxxxxxxxxxxxxx> wrote:
Hi,
Now inode_table_t:lock is type of mutex which I think we can use
‘pthread_rwlock' to replace it for a better concurrency.
Because phread_rwlock allows more than one thread accessing inode table
at the same time.
Moreover, the critical section the lock is protecting won't take many
CPU cycles and no I/O and CPU fault/exception involved after a quick
glance at glusterfs code.
I hope I didn't miss something.
If I would get an ACK from major glusterfs developer, I will try to do it.
You are right. I believe this is possible. No harm in trying this out.
Xavier, Raghavendra, Pranith, Nithya, do you think this is possible?
Regards,
Thanks.
_______________________________________________
Community Meeting Calendar:
APAC Schedule -
Every 2nd and 4th Tuesday at 11:30 AM IST
Bridge: https://bluejeans.com/836554017
NA/EMEA Schedule -
Every 1st and 3rd Tuesday at 01:00 PM EDT
Bridge: https://bluejeans.com/486278655
Gluster-devel mailing list
Gluster-devel@xxxxxxxxxxx
https://lists.gluster.org/mailman/listinfo/gluster-devel
Amar Tumballi (amarts)
_______________________________________________ Community Meeting Calendar: APAC Schedule - Every 2nd and 4th Tuesday at 11:30 AM IST Bridge: https://bluejeans.com/836554017 NA/EMEA Schedule - Every 1st and 3rd Tuesday at 01:00 PM EDT Bridge: https://bluejeans.com/486278655 Gluster-devel mailing list Gluster-devel@xxxxxxxxxxx https://lists.gluster.org/mailman/listinfo/gluster-devel
