Author: pfg
Date: Tue Jun 7 14:37:43 2016
New Revision: 301548
URL: https://svnweb.freebsd.org/changeset/base/301548
Log:
ext2fs: cleanup generation number management.
Ext2/3/4 manages generation numbers differently than UFS so adopt
some rules that should work well. When allocating a new inode,
make sure we generate a "good" random value specifically avoiding
zero.
Don't interfere with the numbers that are already generated in
the filesystem: ext2fs doesn't have the backwards compatibility
issues where there were no generation numbers.
Reviewed by: kevlo
MFC after: 1 week
Modified:
head/sys/fs/ext2fs/ext2_alloc.c
head/sys/fs/ext2fs/ext2_vfsops.c
Modified: head/sys/fs/ext2fs/ext2_alloc.c
==============================================================================
--- head/sys/fs/ext2fs/ext2_alloc.c Tue Jun 7 14:16:10 2016
(r301547)
+++ head/sys/fs/ext2fs/ext2_alloc.c Tue Jun 7 14:37:43 2016
(r301548)
@@ -407,9 +407,11 @@ ext2_valloc(struct vnode *pvp, int mode,
/*
* Set up a new generation number for this inode.
+ * Avoid zero values.
*/
- while (ip->i_gen == 0 || ++ip->i_gen == 0)
+ do {
ip->i_gen = arc4random();
+ } while ( ip->i_gen == 0);
vfs_timestamp(&ts);
ip->i_birthtime = ts.tv_sec;
Modified: head/sys/fs/ext2fs/ext2_vfsops.c
==============================================================================
--- head/sys/fs/ext2fs/ext2_vfsops.c Tue Jun 7 14:16:10 2016
(r301547)
+++ head/sys/fs/ext2fs/ext2_vfsops.c Tue Jun 7 14:37:43 2016
(r301548)
@@ -993,16 +993,6 @@ ext2_vget(struct mount *mp, ino_t ino, i
* Finish inode initialization.
*/
- /*
- * Set up a generation number for this inode if it does not
- * already have one. This should only happen on old filesystems.
- */
- if (ip->i_gen == 0) {
- while (ip->i_gen == 0)
- ip->i_gen = arc4random();
- if ((vp->v_mount->mnt_flag & MNT_RDONLY) == 0)
- ip->i_flag |= IN_MODIFIED;
- }
*vpp = vp;
return (0);
}
_______________________________________________
[email protected] mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "[email protected]"
