diff options
| author | Suresh Jayaraman <sjayaraman@suse.de> | 2010-12-01 14:42:28 +0530 |
|---|---|---|
| committer | Steve French <sfrench@us.ibm.com> | 2010-12-02 19:32:11 +0000 |
| commit | 6d20e8406f0942228a73000663c2b33f488103ea (patch) | |
| tree | 2469267c2ee10c4c723eaa01b1f24c8d0f704870 /fs/cifs/README | |
| parent | 8cb280c90f9cfaab3ba3afbace0b1711dee80d0c (diff) | |
cifs: add attribute cache timeout (actimeo) tunable
Currently, the attribute cache timeout for CIFS is hardcoded to 1 second. This
means that the client might have to issue a QPATHINFO/QFILEINFO call every 1
second to verify if something has changes, which seems too expensive. On the
other hand, if the timeout is hardcoded to a higher value, workloads that
expect strict cache coherency might see unexpected results.
Making attribute cache timeout as a tunable will allow us to make a tradeoff
between performance and cache metadata correctness depending on the
application/workload needs.
Add 'actimeo' tunable that can be used to tune the attribute cache timeout.
The default timeout is set to 1 second. Also, display actimeo option value in
/proc/mounts.
It appears to me that 'actimeo' and the proposed (but not yet merged)
'strictcache' option cannot coexist, so care must be taken that we reset the
other option if one of them is set.
Changes since last post:
- fix option parsing and handle possible values correcly
Reviewed-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Suresh Jayaraman <sjayaraman@suse.de>
Signed-off-by: Steve French <sfrench@us.ibm.com>
Diffstat (limited to 'fs/cifs/README')
| -rw-r--r-- | fs/cifs/README | 9 |
1 files changed, 9 insertions, 0 deletions
diff --git a/fs/cifs/README b/fs/cifs/README index ee68d1036544..46af99ab3614 100644 --- a/fs/cifs/README +++ b/fs/cifs/README @@ -337,6 +337,15 @@ A partial list of the supported mount options follows: wsize default write size (default 57344) maximum wsize currently allowed by CIFS is 57344 (fourteen 4096 byte pages) + actimeo=n attribute cache timeout in seconds (default 1 second). + After this timeout, the cifs client requests fresh attribute + information from the server. This option allows to tune the + attribute cache timeout to suit the workload needs. Shorter + timeouts mean better the cache coherency, but increased number + of calls to the server. Longer timeouts mean reduced number + of calls to the server at the expense of less stricter cache + coherency checks (i.e. incorrect attribute cache for a short + period of time). rw mount the network share read-write (note that the server may still consider the share read-only) ro mount network share read-only |