mongo.pl: 2.6.9-cko3, xattrs + journaling

2004-11-20 Thread Christian Mayrhuber
Hi, I did some runs of mongo to see how the journaling filesystems compare to each other on a recent kernel. My focus has been how the fs's compare with enabled acls and xattrs. Summary: 1) data=ordered + acl: reiserfs is best for writing, ext3 for reading. I guess reiserfs becomes

Re: mongo.pl: 2.6.9-cko3, xattrs + journaling

2004-11-20 Thread Hans Reiser
Christian Mayrhuber wrote: Hi, I did some runs of mongo to see how the journaling filesystems compare to each other on a recent kernel. My focus has been how the fs's compare with enabled acls and xattrs. Summary: 1) data=ordered + acl: reiserfs is best for writing, ext3 for reading.

Re: fsck.reiserfs an image file ?

2004-11-20 Thread daniel.poelzleithner
Helge Hielscher wrote: after saving a partition with dd_rhelp to an image file I wonder if is possible to repair the image to a state where it can be mounted? You have to create a loopback device for it with losetup /dev/loop[number] /myfile and reiserfsck /dev/loop[number] kindly regards

[PATCH 0/5] reiserfs/selinux: patches to allow reiserfs and selinux to work together

2004-11-20 Thread Jeffrey Mahoney
Hey all - The following 5 patches allow reiserfs and selinux to work together. 01-selinux-load_policy.diff - Allows load_policy to fail and to not BUG on the second attempt. 02-selinux-private-inode.diff - Allows private inodes that aren't tracked by selinux xattrs

[PATCH 1/5] selinux: cache not freed if load_policy fails.

2004-11-20 Thread Jeffrey Mahoney
If security_load_policy() fails on the first try, the cache is never cleaned up. When the policy is fixed and a reload is attempted, the old cache will still exist, causing a BUG() in kmem_cache_create(). This patch adds a destroy operation to clean up the cache on failure. Signed-off-by: Jeff

[PATCH 2/5] selinux: adds a private inode operation

2004-11-20 Thread Jeffrey Mahoney
ReiserFS implements extended attributes by backing them with regular files. This means that selinux would create an infinite series of xattrs-on-xattrs-on-xattrs etc. However, due to the locking of xattrs, this series never happens. Instead, a locking loop occurs. This patch allows SElinux to

[PATCH 3/5] reiserfs: private inode abstracted to static inline

2004-11-20 Thread Jeffrey Mahoney
This patch moves the assignment of i_priv_object to a static inline. This is in preparation for selinux support in reiserfs. Signed-off-by: Jeff Mahoney [EMAIL PROTECTED] diff -ruNpX dontdiff linux-2.6.9/fs/reiserfs/inode.c linux-2.6.9.selinux/fs/reiserfs/inode.c ---

[PATCH 4/5] reiserfs: fixes to allow reiserfs to use selinux attributes

2004-11-20 Thread Jeffrey Mahoney
This patch adds the selinux private flag to inodes marked as reiserfs-private. It also avoids the use of vfs_rmdir, since it will use the selinux permissions check. The permission checks aren't valid in this context. Also, some of the checks performed are superfluous for this case.

[PATCH 5/5] reiserfs: cleaning up const checks

2004-11-20 Thread Jeffrey Mahoney
This patch cleans up some warnings in the const qualifiers introduced by the reiserfs/selinux patches. Signed-off-by: Jeff Mahoney [EMAIL PROTECTED] diff -ruNpX dontdiff linux-2.6.9/fs/reiserfs/xattr.c linux-2.6.9.selinux/fs/reiserfs/xattr.c --- linux-2.6.9/fs/reiserfs/xattr.c 2004-11-19

AMD64/reiser4 sync mode test, documented corruption

2004-11-20 Thread Jake Maciejewski
Vladimir, I did another test, my typical dd and make, this time with kernel 2.6.10-rc2-mm2 and reiser4 mounted in sync mode. I'm still having problems with bitmap.c. Details can be found at http://people.msoe.edu/~maciejej/patches/AMD64_reiser4_debug/11-20-04/sync_mount/ Should I repeat the test