diff mbox

nfs41: change PNFS_LAYOUTRET_ON_SETATTR to only return on chown/truncate

Message ID 1410447752-11288-1-git-send-email-tao.peng@primarydata.com (mailing list archive)
State New, archived
Headers show

Commit Message

Peng Tao Sept. 11, 2014, 3:02 p.m. UTC
Both blocks layout and objects layout want to use it to avoid CB_LAYOUTRECALL
but that should only happen if client is doing chown or truncate. So change
PNFS_LAYOUTRET_ON_SETATTR to follow the logic and not to send layoutreturn
unnecessarily.

Cc: Christoph Hellwig <hch@lst.de>
Cc: Boaz Harrosh <boaz@plexistor.com>
Signed-off-by: Peng Tao <tao.peng@primarydata.com>
---
 fs/nfs/nfs4proc.c | 3 ++-
 fs/nfs/pnfs.h     | 3 ++-
 2 files changed, 4 insertions(+), 2 deletions(-)

Comments

Christoph Hellwig Sept. 11, 2014, 8:11 p.m. UTC | #1
On Thu, Sep 11, 2014 at 11:02:32PM +0800, Peng Tao wrote:
> Both blocks layout and objects layout want to use it to avoid CB_LAYOUTRECALL
> but that should only happen if client is doing chown or truncate. So change
> PNFS_LAYOUTRET_ON_SETATTR to follow the logic and not to send layoutreturn
> unnecessarily.

We should defintively reduce the amount of returns, but why do we even
bother and chown and a truncate to a larger size?

--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Peng Tao Sept. 12, 2014, 2:23 a.m. UTC | #2
On Fri, Sep 12, 2014 at 4:11 AM, Christoph Hellwig <hch@infradead.org> wrote:
> On Thu, Sep 11, 2014 at 11:02:32PM +0800, Peng Tao wrote:
>> Both blocks layout and objects layout want to use it to avoid CB_LAYOUTRECALL
>> but that should only happen if client is doing chown or truncate. So change
>> PNFS_LAYOUTRET_ON_SETATTR to follow the logic and not to send layoutreturn
>> unnecessarily.
>
> We should defintively reduce the amount of returns, but why do we even
> bother and chown and a truncate to a larger size?
hmm, I agree. Let's only layoutreturn on truncate to a smaller size
and let server decide if it needs CB_LAYOUTRECALL in chown.
--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
diff mbox

Patch

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 7dd8aca..e55bd0f 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -3213,7 +3213,8 @@  nfs4_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr,
 	struct nfs4_label *label = NULL;
 	int status;
 
-	if (pnfs_ld_layoutret_on_setattr(inode))
+	if (pnfs_ld_layoutret_on_setattr(inode) &&
+	    sattr->ia_valid & (ATTR_UID|ATTR_GID|ATTR_SIZE))
 		pnfs_commit_and_return_layout(inode);
 
 	nfs_fattr_init(fattr);
diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h
index aca3dff..1765ea8 100644
--- a/fs/nfs/pnfs.h
+++ b/fs/nfs/pnfs.h
@@ -68,7 +68,8 @@  enum {
 };
 
 enum layoutdriver_policy_flags {
-	/* Should the pNFS client commit and return the layout upon a setattr */
+	/* Should the pNFS client commit and return the layout upon a setattr
+	 * that changes uid/gid/size. */
 	PNFS_LAYOUTRET_ON_SETATTR	= 1 << 0,
 	PNFS_LAYOUTRET_ON_ERROR		= 1 << 1,
 };