Don't let the pager_map deadlock avoidance code in lfs_putpages() write

segments containing zero-block FINFO records.  These records cause segments
to become uncleanable, which would eventually result in a "no clean segments"
panic.
This commit is contained in:
perseant 2005-05-04 04:58:22 +00:00
parent 8abde347bd
commit 0d41dd0d46

View File

@ -1,4 +1,4 @@
/* $NetBSD: lfs_vfsops.c,v 1.177 2005/04/23 19:47:51 perseant Exp $ */
/* $NetBSD: lfs_vfsops.c,v 1.178 2005/05/04 04:58:22 perseant Exp $ */
/*-
* Copyright (c) 1999, 2000, 2001, 2002, 2003 The NetBSD Foundation, Inc.
@ -67,7 +67,7 @@
*/
#include <sys/cdefs.h>
__KERNEL_RCSID(0, "$NetBSD: lfs_vfsops.c,v 1.177 2005/04/23 19:47:51 perseant Exp $");
__KERNEL_RCSID(0, "$NetBSD: lfs_vfsops.c,v 1.178 2005/05/04 04:58:22 perseant Exp $");
#if defined(_KERNEL_OPT)
#include "opt_quota.h"
@ -2029,7 +2029,12 @@ lfs_gop_write(struct vnode *vp, struct vm_page **pgs, int npages, int flags)
(int)((SEGSUM *)(sp->segsum))->ss_nfinfo,
(unsigned)fs->lfs_offset));
#endif
lfs_updatemeta(sp);
if (sp->fip->fi_nblocks == 0) {
/* Don't write zero-length finfos */
--((SEGSUM *)(sp->segsum))->ss_nfinfo;
sp->sum_bytes_left += FINFOSIZE;
} else
lfs_updatemeta(sp);
version = sp->fip->fi_version;
(void) lfs_writeseg(fs, sp);