Rojin George has uploaded this change for review.

View Change

Ease the lock contention by not rw locking entire tree

Huge directories, for validation when multiple
parallel clients perform operations like md5sum, lsattr etc,
we could see lock contention between ReadDir and Lookup operations
on the content_lock, especially when backend is slow.
As part of this CR, easing the lock contention by not RW locking
entire tree. RW lock would be taken back once the chunk is
to be updated in the callback post subcall.

Change-Id: I800d903152111ba271b09021be9e8b5acc873c04
Signed-off-by: Rojin George <itsmerojin@gmail.com>
---
M src/FSAL/Stackable_FSALs/FSAL_MDCACHE/mdcache_helpers.c
1 file changed, 8 insertions(+), 1 deletion(-)

git pull ssh://review.gerrithub.io:29418/ffilz/nfs-ganesha refs/changes/99/556799/1

To view, visit change 556799. To unsubscribe, or for help writing mail filters, visit settings.