mirror/freebsd
1
0
Fork 0
mirror of https://git.FreeBSD.org/src.git synced 2024-12-23 11:18:54 +00:00
Code Issues Releases Activity

Avoid starvation of the server crash recovery thread for the NFSv4 client.

Browse Source 
This patch gives a requestor of the exclusive lock on the client state
in the NFSv4 client priority over shared lock requestors. This avoids
the server crash recovery thread being starved out by other threads doing
RPCs.

Tested by:	cperciva
PR:		216087
MFC after:	2 weeks
This commit is contained in:
Rick Macklem 2017-04-10 01:28:01 +00:00
parent 4e7dcfab2d
commit 8efffe80ba
Notes: svn2git 2020-12-20 02:59:44 +00:00 
svn path=/head/; revision=316669
1 changed files with 11 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
sys/fs/nfsclient/nfs_clstate.c
View File

@ -797,8 +797,18 @@ nfscl_getcl(struct mount *mp, struct ucred *cred, NFSPROC_T *p,
(mp->mnt_kern_flag & MNTK_UNMOUNTF) == 0)
igotlock = nfsv4_lock(&clp->nfsc_lock, 1, NULL,
NFSCLSTATEMUTEXPTR, mp);
if (!igotlock)
if (igotlock == 0) {
/*
* Call nfsv4_lock() with "iwantlock == 0" so that it will
* wait for a pending exclusive lock request. This gives the
* exclusive lock request priority over this shared lock
* request.
* An exclusive lock on nfsc_lock is used mainly for server
* crash recoveries.
*/
nfsv4_lock(&clp->nfsc_lock, 0, NULL, NFSCLSTATEMUTEXPTR, mp);
nfsv4_getref(&clp->nfsc_lock, NULL, NFSCLSTATEMUTEXPTR, mp);
}
if (igotlock == 0 && (mp->mnt_kern_flag & MNTK_UNMOUNTF) != 0) {
/*
* Both nfsv4_lock() and nfsv4_getref() know to check