From: Jeff Layton <jlayton@poochiereds.net>
Date: Thu, 19 Nov 2015 19:30:26 +0000 (-0500)
Subject: nfs: use sliding delay when LAYOUTGET gets NFS4ERR_DELAY
X-Git-Tag: v4.4-rc3~9^2~5
X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=91ab4b4d16e6649fbbf65f303c0c4e20ed680bd1;p=platform%2Fkernel%2Flinux-exynos.git

nfs: use sliding delay when LAYOUTGET gets NFS4ERR_DELAY

When LAYOUTGET gets NFS4ERR_DELAY, we currently will wait 15s before
retrying the call. That is a _very_ long time, so add a timeout value to
struct nfs4_layoutget and pass nfs4_async_handle_error a pointer to it.
This allows the RPC engine to use a sliding delay window, instead of a
15s delay.

Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
---

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 765a035..8981803 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -7866,7 +7866,7 @@ static void nfs4_layoutget_done(struct rpc_task *task, void *calldata)
 			spin_unlock(&inode->i_lock);
 		goto out_restart;
 	}
-	if (nfs4_async_handle_error(task, server, state, NULL) == -EAGAIN)
+	if (nfs4_async_handle_error(task, server, state, &lgp->timeout) == -EAGAIN)
 		goto out_restart;
 out:
 	dprintk("<-- %s\n", __func__);
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 570d630..11bbae4 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -251,6 +251,7 @@ struct nfs4_layoutget {
 	struct nfs4_layoutget_res res;
 	struct rpc_cred *cred;
 	gfp_t gfp_flags;
+	long timeout;
 };
 
 struct nfs4_getdeviceinfo_args {