Move the setting of the default value for nm_wcommitsize to

before the nfs_decode_args() call in the new NFS client, so
that a specfied command line value won't be overwritten.
Also, modify the calculation for small values of desiredvnodes
to avoid an unusually large value or a divide by zero crash.
It seems that the default value for nm_wcommitsize is very
conservative and may need to change at some time.

PR:		kern/159351
Submitted by:	onwahe at gmail.com (earlier version)
Reviewed by:	jhb
MFC after:	2 weeks
This commit is contained in:
Rick Macklem 2011-11-15 01:39:02 +00:00
parent a38f2acf11
commit e42a8d7e24

View file

@ -1231,6 +1231,10 @@ mountnfs(struct nfs_args *argp, struct mount *mp, struct sockaddr *nam,
nmp->nm_timeo = NFS_TIMEO;
nmp->nm_retry = NFS_RETRANS;
nmp->nm_readahead = NFS_DEFRAHEAD;
if (desiredvnodes >= 11000)
nmp->nm_wcommitsize = hibufspace / (desiredvnodes / 1000);
else
nmp->nm_wcommitsize = hibufspace / 10;
nfs_decode_args(mp, nmp, argp, hst, cred, td);
@ -1252,7 +1256,6 @@ mountnfs(struct nfs_args *argp, struct mount *mp, struct sockaddr *nam,
nmp->nm_rsize = NFS_RSIZE;
nmp->nm_readdirsize = NFS_READDIRSIZE;
}
nmp->nm_wcommitsize = hibufspace / (desiredvnodes / 1000);
nmp->nm_numgrps = NFS_MAXGRPS;
nmp->nm_tprintf_delay = nfs_tprintf_delay;
if (nmp->nm_tprintf_delay < 0)