diff options
| author | dfr <dfr@FreeBSD.org> | 1996-10-11 10:15:33 +0000 |
|---|---|---|
| committer | dfr <dfr@FreeBSD.org> | 1996-10-11 10:15:33 +0000 |
| commit | de60fb9205631ede6a176edd5d010140cfa266d3 (patch) | |
| tree | 7d089c2d16ff47871a314141686d876a406a3741 /sys/nfsclient/nfs_bio.c | |
| parent | 2d340b514c77a95e6ab924510db5f79b7d3f448a (diff) | |
| download | FreeBSD-src-de60fb9205631ede6a176edd5d010140cfa266d3.zip FreeBSD-src-de60fb9205631ede6a176edd5d010140cfa266d3.tar.gz | |
This fixes a problem with the nfs socket handling code which happens
if a single process is performing a large number of requests (in this
case writing a large file). The writing process could monopolise the
recieve lock and prevent any other processes from recieving their
replies.
It also adds a new sysctl variable 'vfs.nfs.dwrite' which controls the
behaviour which originally pointed out the problem. When a process
writes to a file over NFS, it usually arranges for another process
(the 'iod') to perform the request. If no iods are available, then it
turns the write into a 'delayed write' which is later picked up by the
next iod to do a write request for that file. This can cause that
particular iod to do a disproportionate number of requests from a
single process which can harm performance on some NFS servers. The
alternative is to perform the write synchronously in the context of
the original writing process if no iod is avaiable for asynchronous
writing.
The 'delayed write' behaviour is selected when vfs.nfs.dwrite=1 and
the non-delayed behaviour is selected when vfs.nfs.dwrite=0. The
default is vfs.nfs.dwrite=1; if many people tell me that performance
is better if vfs.nfs.dwrite=0 then I will change the default.
Submitted by: Hidetoshi Shimokawa <simokawa@sat.t.u-tokyo.ac.jp>
Diffstat (limited to 'sys/nfsclient/nfs_bio.c')
| -rw-r--r-- | sys/nfsclient/nfs_bio.c | 14 |
1 files changed, 13 insertions, 1 deletions
diff --git a/sys/nfsclient/nfs_bio.c b/sys/nfsclient/nfs_bio.c index 61e6d00..ce4b896 100644 --- a/sys/nfsclient/nfs_bio.c +++ b/sys/nfsclient/nfs_bio.c @@ -34,7 +34,7 @@ * SUCH DAMAGE. * * @(#)nfs_bio.c 8.5 (Berkeley) 1/4/94 - * $Id: nfs_bio.c,v 1.24 1996/07/16 10:19:43 dfr Exp $ + * $Id: nfs_bio.c,v 1.25 1996/09/19 18:20:54 nate Exp $ */ #include <sys/param.h> @@ -46,6 +46,7 @@ #include <sys/vnode.h> #include <sys/mount.h> #include <sys/kernel.h> +#include <sys/sysctl.h> #include <vm/vm.h> #include <vm/vm_param.h> @@ -65,6 +66,9 @@ extern struct proc *nfs_iodwant[NFS_MAXASYNCDAEMON]; extern int nfs_numasync; extern struct nfsstats nfsstats; +int nfs_dwrite = 1; +SYSCTL_INT(_vfs_nfs, OID_AUTO, dwrite, CTLFLAG_RW, &nfs_dwrite, 0, ""); + /* * Ifdefs for FreeBSD-current's merged VM/buffer cache. It is unfortunate * that this isn't done inside getblk() and brelse() so these calls @@ -753,6 +757,14 @@ nfs_asyncio(bp, cred) return (EIO); /* + * Allow the administrator to override the choice of using a delayed + * write since it is a pessimization for some servers, notably some + * Solaris servers. + */ + if (!nfs_dwrite) + return (EIO); + + /* * Just turn the async write into a delayed write, instead of * doing in synchronously. Hopefully, at least one of the nfsiods * is currently doing a write for this file and will pick up the |
