Aren
/
NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Always unlock the "fnode", even if we're about to free it, to keep 

lock counts consistent. DIAGNOSTIC checks will barf otherwise.

XXX ntfs_ntput should not free the node, the reclaim vop should do it
XXX why does the ntfs code do its own internal locking?
This commit is contained in:
fvdl 1999-12-20 22:11:57 +00:00
parent bc8f4e2f46
commit 655a7bf795
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
sys/ntfs/ntfs_subr.c
View File

@ -1,4 +1,4 @@
/* $NetBSD: ntfs_subr.c,v 1.23 1999/10/31 19:45:26 jdolecek Exp $ */
/* $NetBSD: ntfs_subr.c,v 1.24 1999/12/20 22:11:57 fvdl Exp $ */
/*-
* Copyright (c) 1998, 1999 Semen Ustimenko (semenu@FreeBSD.org)
@ -450,6 +450,8 @@ ntfs_ntput(ip)
}
#endif
lockmgr(&ip->i_lock, LK_RELEASE|LK_INTERLOCK, &ip->i_interlock);
if (ip->i_usecount == 0) {
dprintf(("ntfs_ntput: deallocating ntnode: %d\n",
ip->i_number));
@ -465,8 +467,6 @@ ntfs_ntput(ip)
ntfs_freentvattr(vap);
}
FREE(ip, M_NTFSNTNODE);
} else {
lockmgr(&ip->i_lock, LK_RELEASE|LK_INTERLOCK, &ip->i_interlock);
}
}