linux-2.6
15 years agoocfs2/quota: sparse fixes for quota
Tao Ma [Mon, 24 Nov 2008 23:21:36 +0000 (07:21 +0800)] 
ocfs2/quota: sparse fixes for quota

Fix 2 minor things in quota. They are both found by sparse check.
1. an endian bug in ocfs2_local_quota_add_chunk.
2. change olq_alloc_dquot to static.

Signed-off-by: Tao Ma <tao.ma@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: fix indendation in ocfs2_dquot_drop_slow
Tao Ma [Thu, 27 Nov 2008 00:20:19 +0000 (16:20 -0800)] 
ocfs2: fix indendation in ocfs2_dquot_drop_slow

Signed-off-by: Tao Ma <tao.ma@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoext4: Add default allocation routines for quota structures
Jan Kara [Tue, 25 Nov 2008 14:31:35 +0000 (15:31 +0100)] 
ext4: Add default allocation routines for quota structures

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoext3: Add default allocation routines for quota structures
Jan Kara [Tue, 25 Nov 2008 14:31:34 +0000 (15:31 +0100)] 
ext3: Add default allocation routines for quota structures

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoreiserfs: Add default allocation routines for quota structures
Jan Kara [Tue, 25 Nov 2008 14:31:33 +0000 (15:31 +0100)] 
reiserfs: Add default allocation routines for quota structures

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoquota: Export dquot_alloc() and dquot_destroy() functions
Jan Kara [Tue, 25 Nov 2008 14:31:32 +0000 (15:31 +0100)] 
quota: Export dquot_alloc() and dquot_destroy() functions

These are default functions for creating and destroying quota structures
and they should be used from filesystems.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoquota: Unexport dqblk_v1.h and dqblk_v2.h
Jan Kara [Tue, 25 Nov 2008 14:31:31 +0000 (15:31 +0100)] 
quota: Unexport dqblk_v1.h and dqblk_v2.h

Unexport header files dqblk_v[12].h since except for quota format ID they
don't contain information userspace should be interested in. Move ID
definitions to quota.h.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Fix build warnings (64-bit types vs long long)
Jan Kara [Tue, 25 Nov 2008 14:31:30 +0000 (15:31 +0100)] 
ocfs2: Fix build warnings (64-bit types vs long long)

fs/ocfs2/quota_local.c: In function 'olq_set_dquot':
fs/ocfs2/quota_local.c:844: warning: format '%lld' expects type 'long long int', but argument 7 has type '__le64'
fs/ocfs2/quota_local.c:844: warning: format '%lld' expects type 'long long int', but argument 8 has type '__le64'
fs/ocfs2/quota_local.c:844: warning: format '%lld' expects type 'long long int', but argument 7 has type '__le64'
fs/ocfs2/quota_local.c:844: warning: format '%lld' expects type 'long long int', but argument 8 has type '__le64'
fs/ocfs2/quota_local.c:844: warning: format '%lld' expects type 'long long int', but argument 7 has type '__le64'
fs/ocfs2/quota_local.c:844: warning: format '%lld' expects type 'long long int', but argument 8 has type '__le64'
fs/ocfs2/quota_global.c: In function '__ocfs2_sync_dquot':
fs/ocfs2/quota_global.c:457: warning: format '%lld' expects type 'long long int', but argument 8 has type 's64'
fs/ocfs2/quota_global.c:457: warning: format '%lld' expects type 'long long int', but argument 10 has type 's64'
fs/ocfs2/quota_global.c:457: warning: format '%lld' expects type 'long long int', but argument 8 has type 's64'
fs/ocfs2/quota_global.c:457: warning: format '%lld' expects type 'long long int', but argument 10 has type 's64'
fs/ocfs2/quota_global.c:457: warning: format '%lld' expects type 'long long int', but argument 8 has type 's64'
fs/ocfs2/quota_global.c:457: warning: format '%lld' expects type 'long long int', but argument 10 has type 's64'

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Make ocfs2_get_quota_block() consistent with ocfs2_read_quota_block()
Jan Kara [Tue, 25 Nov 2008 14:31:29 +0000 (15:31 +0100)] 
ocfs2: Make ocfs2_get_quota_block() consistent with ocfs2_read_quota_block()

Make function return error status and not buffer pointer so that it's
consistent with ocfs2_read_quota_block().

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Fix oops when extending quota files
Jan Kara [Tue, 25 Nov 2008 14:31:28 +0000 (15:31 +0100)] 
ocfs2: Fix oops when extending quota files

We have to mark buffer as uptodate before calling ocfs2_journal_access() and
ocfs2_set_buffer_uptodate() does not do this for us.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Fix ocfs2_read_quota_block() error handling.
Joel Becker [Tue, 25 Nov 2008 14:31:27 +0000 (15:31 +0100)] 
ocfs2: Fix ocfs2_read_quota_block() error handling.

ocfs2_bread() has become ocfs2_read_virt_blocks(), with a prototype to
match ocfs2_read_blocks().  The quota code, converting from
ocfs2_bread(), wraps the call to ocfs2_read_virt_blocks() in
ocfs2_read_quota_block().  Unfortunately, the prototype of
ocfs2_read_quota_block() matches the old prototype of ocfs2_bread().

The problem is that ocfs2_bread() returned the buffer head, and callers
assumed that a NULL pointer was indicative of error.  It wasn't.  This
is why ocfs2_bread() took an int*err argument as well.

The new prototype of ocfs2_read_virt_blocks() avoids this error handling
confusion.  Let's change ocfs2_read_quota_block() to match.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Acked-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Add missing initialization
Jan Kara [Tue, 25 Nov 2008 14:31:26 +0000 (15:31 +0100)] 
ocfs2: Add missing initialization

Add missing variable initialization to ocfs2_dquot_drop_slow().

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Use BH_JBDPrivateStart instead of BH_Unshadow
Mark Fasheh [Wed, 19 Nov 2008 01:16:47 +0000 (17:16 -0800)] 
ocfs2: Use BH_JBDPrivateStart instead of BH_Unshadow

This is safer. We no longer have to worry about tracking changes to
jbd_state_bits.

Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agojbd2: Add BH_JBDPrivateStart
Mark Fasheh [Wed, 19 Nov 2008 01:15:24 +0000 (17:15 -0800)] 
jbd2: Add BH_JBDPrivateStart

Add this so that file systems using JBD2 can safely allocate unused b_state
bits.

In this case, we add it so that Ocfs2 can define a single bit for tracking
the validation state of a buffer.

Acked-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Enable quota accounting on mount, disable on umount
Jan Kara [Thu, 21 Aug 2008 18:13:17 +0000 (20:13 +0200)] 
ocfs2: Enable quota accounting on mount, disable on umount

Enable quota usage tracking on mount and disable it on umount. Also
add support for quota on and quota off quotactls and usrquota and
grpquota mount options. Add quota features among supported ones.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Implement quota recovery
Jan Kara [Mon, 20 Oct 2008 21:50:38 +0000 (23:50 +0200)] 
ocfs2: Implement quota recovery

Implement functions for recovery after a crash. Functions just
read local quota file and sync info to global quota file.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Periodic quota syncing
Mark Fasheh [Mon, 20 Oct 2008 13:36:47 +0000 (15:36 +0200)] 
ocfs2: Periodic quota syncing

This patch creates a work queue for periodic syncing of locally cached quota
information to the global quota files. We constantly queue a delayed work
item, to get the periodic behavior.

Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Acked-by: Jan Kara <jack@suse.cz>
15 years agoocfs2: Add quota calls for allocation and freeing of inodes and space
Jan Kara [Thu, 9 Oct 2008 17:38:40 +0000 (19:38 +0200)] 
ocfs2: Add quota calls for allocation and freeing of inodes and space

Add quota calls for allocation and freeing of inodes and space, also update
estimates on number of needed credits for a transaction. Move out inode
allocation from ocfs2_mknod_locked() because vfs_dq_init() must be called
outside of a transaction.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Implementation of local and global quota file handling
Jan Kara [Mon, 25 Aug 2008 17:56:50 +0000 (19:56 +0200)] 
ocfs2: Implementation of local and global quota file handling

For each quota type each node has local quota file. In this file it stores
changes users have made to disk usage via this node. Once in a while this
information is synced to global file (and thus with other nodes) so that
limits enforcement at least aproximately works.

Global quota files contain all the information about usage and limits. It's
mostly handled by the generic VFS code (which implements a trie of structures
inside a quota file). We only have to provide functions to convert structures
from on-disk format to in-memory one. We also have to provide wrappers for
various quota functions starting transactions and acquiring necessary cluster
locks before the actual IO is really started.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Mark system files as not subject to quota accounting
Jan Kara [Thu, 21 Aug 2008 16:22:30 +0000 (18:22 +0200)] 
ocfs2: Mark system files as not subject to quota accounting

Mark system files as not subject to quota accounting. This prevents
possible recursions into quota code and thus deadlocks.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Assign feature bits and system inodes to quota feature and quota files
Jan Kara [Wed, 20 Aug 2008 13:43:36 +0000 (15:43 +0200)] 
ocfs2: Assign feature bits and system inodes to quota feature and quota files

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Support nested transactions
Jan Kara [Wed, 27 Aug 2008 20:30:28 +0000 (22:30 +0200)] 
ocfs2: Support nested transactions

OCFS2 can easily support nested transactions. We just have to
take care and not spoil statistics acquire semaphore unnecessarily.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoquota: Implement function for scanning active dquots
Jan Kara [Mon, 20 Oct 2008 15:05:00 +0000 (17:05 +0200)] 
quota: Implement function for scanning active dquots

OCFS2 needs to scan all active dquots once in a while and sync quota
information among cluster nodes. Provide a helper function for it so
that it does not have to reimplement internally a list which VFS
already has. Moreover this function is probably going to be useful
for other clustered filesystems if they decide to use VFS quotas.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoquota: Add helpers to allow ocfs2 specific quota initialization, freeing and recovery
Jan Kara [Fri, 10 Oct 2008 14:12:23 +0000 (16:12 +0200)] 
quota: Add helpers to allow ocfs2 specific quota initialization, freeing and recovery

OCFS2 needs to peek whether quota structure is already in memory so
that it can avoid expensive cluster locking in that case. Similarly
when freeing dquots, it checks whether it is the last quota structure
user or not. Finally, it needs to get reference to dquot structure for
specified id and quota type when recovering quota file after crash.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoquota: Update version number
Jan Kara [Thu, 30 Oct 2008 08:17:52 +0000 (09:17 +0100)] 
quota: Update version number

Increase reported version number of quota support since quota core has changed
significantly. Also remove __DQUOT_NUM_VERSION__ since nobody uses it.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoquota: Keep which entries were set by SETQUOTA quotactl
Jan Kara [Thu, 2 Oct 2008 14:48:10 +0000 (16:48 +0200)] 
quota: Keep which entries were set by SETQUOTA quotactl

Quota in a clustered environment needs to synchronize quota information
among cluster nodes. This means we have to occasionally update some
information in dquot from disk / network. On the other hand we have to
be careful not to overwrite changes administrator did via SETQUOTA.
So indicate in dquot->dq_flags which entries have been set by SETQUOTA
and quota format can clear these flags when it properly propagated
the changes.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoquota: Allow negative usage of space and inodes
Jan Kara [Wed, 1 Oct 2008 16:21:39 +0000 (18:21 +0200)] 
quota: Allow negative usage of space and inodes

For clustered filesystems, it can happen that space / inode usage goes
negative temporarily (because some node is allocating another node
is freeing and they are not completely in sync). So let quota code
allow this and change qsize_t so a signed type so that we don't
underflow the variables.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoquota: Convert union in mem_dqinfo to a pointer
Jan Kara [Thu, 2 Oct 2008 16:44:14 +0000 (18:44 +0200)] 
quota: Convert union in mem_dqinfo to a pointer

Coming quota support for OCFS2 is going to need quite a bit
of additional per-sb quota information. Moreover having fs.h
include all the types needed for this structure would be a
pain in the a**. So remove the union from mem_dqinfo and add
a private pointer for filesystem's use.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoquota: Split off quota tree handling into a separate file
Jan Kara [Mon, 22 Sep 2008 03:54:49 +0000 (05:54 +0200)] 
quota: Split off quota tree handling into a separate file

There is going to be a new version of quota format having 64-bit
quota limits and a new quota format for OCFS2. They are both
going to use the same tree structure as VFSv0 quota format. So
split out tree handling into a separate file and make size of
leaf blocks, amount of space usable in each block (needed for
checksumming) and structures contained in them configurable
so that the code can be shared.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoquota: Move quotaio_v[12].h from include/linux/ to fs/
Jan Kara [Sun, 21 Sep 2008 21:17:53 +0000 (23:17 +0200)] 
quota: Move quotaio_v[12].h from include/linux/ to fs/

Since these include files are used only by implementation of quota formats,
there's no need to have them in include/linux/.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoquota: Introduce DQUOT_QUOTA_SYS_FILE flag
Jan Kara [Tue, 30 Sep 2008 15:53:37 +0000 (17:53 +0200)] 
quota: Introduce DQUOT_QUOTA_SYS_FILE flag

If filesystem can handle quota files as system files hidden from users, we can
skip a lot of cache invalidation, syncing, inode flags setting etc. when
turning quotas on, off and quota_sync. Allow filesystem to indicate that it is
hiding quota files from users by DQUOT_QUOTA_SYS_FILE flag.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoquota: Remove compatibility function sb_any_quota_enabled()
Jan Kara [Wed, 20 Aug 2008 16:30:40 +0000 (18:30 +0200)] 
quota: Remove compatibility function sb_any_quota_enabled()

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoreiserfs: Use sb_any_quota_loaded() instead of sb_any_quota_enabled().
Jan Kara [Wed, 20 Aug 2008 16:16:36 +0000 (18:16 +0200)] 
reiserfs: Use sb_any_quota_loaded() instead of sb_any_quota_enabled().

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoext4: Use sb_any_quota_loaded() instead of sb_any_quota_enabled()
Jan Kara [Wed, 20 Aug 2008 16:14:35 +0000 (18:14 +0200)] 
ext4: Use sb_any_quota_loaded() instead of sb_any_quota_enabled()

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoext3: Use sb_any_quota_loaded() instead of sb_any_quota_enabled()
Jan Kara [Wed, 20 Aug 2008 16:11:50 +0000 (18:11 +0200)] 
ext3: Use sb_any_quota_loaded() instead of sb_any_quota_enabled()

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoquota: Allow to separately enable quota accounting and enforcing limits
Jan Kara [Wed, 20 Aug 2008 15:50:32 +0000 (17:50 +0200)] 
quota: Allow to separately enable quota accounting and enforcing limits

Split DQUOT_USR_ENABLED (and DQUOT_GRP_ENABLED) into DQUOT_USR_USAGE_ENABLED
and DQUOT_USR_LIMITS_ENABLED. This way we are able to separately enable /
disable whether we should:
1) ignore quotas completely
2) just keep uptodate information about usage
3) actually enforce quota limits

This is going to be useful when quota is treated as filesystem metadata - we
then want to keep quota information uptodate all the time and just enable /
disable limits enforcement.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoquota: Make _SUSPENDED just a flag
Jan Kara [Wed, 20 Aug 2008 14:21:01 +0000 (16:21 +0200)] 
quota: Make _SUSPENDED just a flag

Upto now, DQUOT_USR_SUSPENDED behaved like a state - i.e., either quota
was enabled or suspended or none. Now allowed states are 0, ENABLED,
ENABLED | SUSPENDED. This will be useful later when we implement separate
enabling of quota usage tracking and limits enforcement because we need to
keep track of a state which has been suspended.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoquota: Remove bogus 'optimization' in check_idq() and check_bdq()
Jan Kara [Wed, 20 Aug 2008 13:49:59 +0000 (15:49 +0200)] 
quota: Remove bogus 'optimization' in check_idq() and check_bdq()

Checks like <= 0 for an unsigned type do not make much sence. The value
could be only 0 and that does not happen often enough for the check
to be worth it.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoquota: Increase size of variables for limits and inode usage
Jan Kara [Wed, 20 Aug 2008 12:45:12 +0000 (14:45 +0200)] 
quota: Increase size of variables for limits and inode usage

So far quota was fine with quota block limits and inode limits/numbers in
a 32-bit type. Now with rapid increase in storage sizes there are coming
requests to be able to handle quota limits above 4TB / more that 2^32 inodes.
So bump up sizes of types in mem_dqblk structure to 64-bits to be able to
handle this. Also update inode allocation / checking functions to use qsize_t
and make global structure keep quota limits in bytes so that things are
consistent.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoquota: Add callbacks for allocating and destroying dquot structures
Jan Kara [Tue, 19 Aug 2008 12:51:22 +0000 (14:51 +0200)] 
quota: Add callbacks for allocating and destroying dquot structures

Some filesystems would like to keep private information together with each
dquot. Add callbacks alloc_dquot and destroy_dquot allowing filesystem to
allocate larger dquots from their private slab in a similar fashion we
currently allocate inodes.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2/xattr: Restore not_found in xis
Tao Ma [Wed, 19 Nov 2008 08:48:42 +0000 (16:48 +0800)] 
ocfs2/xattr: Restore not_found in xis

During an xattr set, when we move a xattr which was stored in inode to the
outside bucket, we have to delete it and it will use the old value of
xis->not_found. xis->not_found is removed by ocfs2_calc_xattr_set_need
though, so we must restore it.

Signed-off-by: Tao Ma <tao.ma@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2/xattr: Fix a bug in xattr allocation estimation
Tao Ma [Wed, 19 Nov 2008 08:48:41 +0000 (16:48 +0800)] 
ocfs2/xattr: Fix a bug in xattr allocation estimation

When we extend one xattr's value to a large size, the old value size might
be smaller than the size of a value root. In those cases, we still need to
guess the metadata allocation.

Reported-by: Tiger Yang <tiger.yang@oracle.com>
Signed-off-by: Tao Ma <tao.ma@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Remove JBD compatibility layer
Mark Fasheh [Wed, 19 Nov 2008 00:53:43 +0000 (16:53 -0800)] 
ocfs2: Remove JBD compatibility layer

JBD2 is fully backwards compatible with JBD and it's been tested enough with
Ocfs2 that we can clean this code up now.

Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Convert ocfs2_read_dir_block() to ocfs2_read_virt_blocks()
Joel Becker [Thu, 13 Nov 2008 22:49:21 +0000 (14:49 -0800)] 
ocfs2: Convert ocfs2_read_dir_block() to ocfs2_read_virt_blocks()

Now that we've centralized the ocfs2_read_virt_blocks() code, let's use
it in ocfs2_read_dir_block().

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Wrap virtual block reads in ocfs2_read_virt_blocks()
Joel Becker [Thu, 13 Nov 2008 22:49:20 +0000 (14:49 -0800)] 
ocfs2: Wrap virtual block reads in ocfs2_read_virt_blocks()

The ocfs2_read_dir_block() function really maps an inode's virtual
blocks to physical ones before calling ocfs2_read_blocks().  Let's
extract that to common code, because other places might want to do that.

Other than the block number being virtual, ocfs2_read_virt_blocks()
takes the same arguments as ocfs2_read_blocks().  It converts those
virtual block numbers to physical before calling ocfs2_read_blocks()
directly.  If the blocks asked for are discontiguous, this can mean
multiple calls to ocfs2_read_blocks(), but this is mostly hidden from
the caller.

Like ocfs2_read_blocks(), the caller can pass in an existing
buffer_head.  This is usually done to pick up some readahead I/O.
ocfs2_read_virt_blocks() checks the buffer_head's block number
against the extent map - it must match.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Validate metadata only when it's read from disk.
Joel Becker [Thu, 13 Nov 2008 22:49:19 +0000 (14:49 -0800)] 
ocfs2: Validate metadata only when it's read from disk.

Add an optional validation hook to ocfs2_read_blocks().  Now the
validation function is only called when a block was actually read off of
disk.  It is not called when the buffer was in cache.

We add a buffer state bit BH_NeedsValidate to flag these buffers.  It
must always be one higher than the last JBD2 buffer state bit.

The dinode, dirblock, extent_block, and xattr_block validators are
lifted to this scheme directly.  The group_descriptor validator needs to
be split into two pieces.  The first part only needs the gd buffer and
is passed to ocfs2_read_block().  The second part requires the dinode as
well, and is called every time.  It's only 3 compares, so it's tiny.
This also allows us to clean up the non-fatal gd check used by resize.c.
It now has no magic argument.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Wrap xattr block reads in a dedicated function
Joel Becker [Thu, 13 Nov 2008 22:49:18 +0000 (14:49 -0800)] 
ocfs2: Wrap xattr block reads in a dedicated function

We weren't consistently checking xattr blocks after we read them.
Most places checked the signature, but none checked xb_blkno or
xb_fs_signature.  Create a toplevel ocfs2_read_xattr_block() that does
the read and the validation.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Wrap dirblock reads in a dedicated function.
Joel Becker [Thu, 13 Nov 2008 22:49:17 +0000 (14:49 -0800)] 
ocfs2: Wrap dirblock reads in a dedicated function.

We have ocfs2_bread() as a vestige of the original ext-based dir code.
It's only used by directories, though.  Turn it into
ocfs2_read_dir_block(), with a prototype matching the other metadata
read functions.  It's set up to validate dirblocks when the time comes.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Wrap extent block reads in a dedicated function.
Joel Becker [Thu, 13 Nov 2008 22:49:16 +0000 (14:49 -0800)] 
ocfs2: Wrap extent block reads in a dedicated function.

We weren't consistently checking extent blocks after we read them.
Most places checked the signature, but none checked h_blkno or
h_fs_signature.  Create a toplevel ocfs2_read_extent_block() that does
the read and the validation.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Morph the haphazard OCFS2_IS_VALID_GROUP_DESC() checks.
Joel Becker [Thu, 13 Nov 2008 22:49:15 +0000 (14:49 -0800)] 
ocfs2: Morph the haphazard OCFS2_IS_VALID_GROUP_DESC() checks.

Random places in the code would check a group descriptor bh to see if it
was valid. The previous commit unified descriptor block reads,
validating all block reads in the same place.  Thus, these checks are no
longer necessary.  Rather than eliminate them, however, we change them
to BUG_ON() checks.  This ensures the assumptions remain true.  All of
the code paths to these checks have been audited to ensure they come
from a validated descriptor read.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Wrap group descriptor reads in a dedicated function.
Joel Becker [Thu, 13 Nov 2008 22:49:14 +0000 (14:49 -0800)] 
ocfs2: Wrap group descriptor reads in a dedicated function.

We have a clean call for validating group descriptors, but every place
that wants the always does a read_block()+validate() call pair.  Create
a toplevel ocfs2_read_group_descriptor() that does the right
thing.  This allows us to leverage the single call point later for
fancier handling.  We also add validation of gd->bg_generation against
the superblock and gd->bg_blkno against the block we thought we read.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Consolidate validation of group descriptors.
Joel Becker [Thu, 13 Nov 2008 22:49:13 +0000 (14:49 -0800)] 
ocfs2: Consolidate validation of group descriptors.

Currently the validation of group descriptors is directly duplicated so
that one version can error the filesystem and the other (resize) can
just report the problem.  Consolidate to one function that takes a
boolean.  Wrap that function with the old call for the old users.

This is in preparation for lifting the read+validate step into a
single function.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Morph the haphazard OCFS2_IS_VALID_DINODE() checks.
Joel Becker [Thu, 13 Nov 2008 22:49:12 +0000 (14:49 -0800)] 
ocfs2: Morph the haphazard OCFS2_IS_VALID_DINODE() checks.

Random places in the code would check a dinode bh to see if it was
valid.  Not only did they do different levels of validation, they
handled errors in different ways.

The previous commit unified inode block reads, validating all block
reads in the same place.  Thus, these haphazard checks are no longer
necessary.  Rather than eliminate them, however, we change them to
BUG_ON() checks.  This ensures the assumptions remain true.  All of the
code paths to these checks have been audited to ensure they come from a
validated inode read.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Wrap inode block reads in a dedicated function.
Joel Becker [Thu, 13 Nov 2008 22:49:11 +0000 (14:49 -0800)] 
ocfs2: Wrap inode block reads in a dedicated function.

The ocfs2 code currently reads inodes off disk with a simple
ocfs2_read_block() call.  Each place that does this has a different set
of sanity checks it performs.  Some check only the signature.  A couple
validate the block number (the block read vs di->i_blkno).  A couple
others check for VALID_FL.  Only one place validates i_fs_generation.  A
couple check nothing.  Even when an error is found, they don't all do
the same thing.

We wrap inode reading into ocfs2_read_inode_block().  This will validate
all the above fields, going readonly if they are invalid (they never
should be).  ocfs2_read_inode_block_full() is provided for the places
that want to pass read_block flags.  Every caller is passing a struct
inode with a valid ip_blkno, so we don't need a separate blkno argument
either.

We will remove the validation checks from the rest of the code in a
later commit, as they are no longer necessary.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: add mount option and Kconfig option for acl
Tiger Yang [Fri, 14 Nov 2008 03:17:52 +0000 (11:17 +0800)] 
ocfs2: add mount option and Kconfig option for acl

This patch adds the Kconfig option "CONFIG_OCFS2_FS_POSIX_ACL"
and mount options "acl" to enable acls in Ocfs2.

Signed-off-by: Tiger Yang <tiger.yang@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: add ocfs2_init_acl in mknod
Tiger Yang [Fri, 14 Nov 2008 03:17:41 +0000 (11:17 +0800)] 
ocfs2: add ocfs2_init_acl in mknod

We need to get the parent directories acls and let the new child inherit it.
To this, we add additional calculations for data/metadata allocation.

Signed-off-by: Tiger Yang <tiger.yang@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: add ocfs2_acl_chmod
Tiger Yang [Fri, 14 Nov 2008 03:17:29 +0000 (11:17 +0800)] 
ocfs2: add ocfs2_acl_chmod

This function is used to update acl xattrs during file mode changes.

Signed-off-by: Tiger Yang <tiger.yang@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: add ocfs2_check_acl
Tiger Yang [Fri, 14 Nov 2008 03:17:18 +0000 (11:17 +0800)] 
ocfs2: add ocfs2_check_acl

This function is used to enhance permission checking with POSIX ACLs.

Signed-off-by: Tiger Yang <tiger.yang@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: add POSIX ACL API
Tiger Yang [Fri, 14 Nov 2008 03:17:04 +0000 (11:17 +0800)] 
ocfs2: add POSIX ACL API

This patch adds POSIX ACL(access control lists) APIs in ocfs2. We convert
struct posix_acl to many ocfs2_acl_entry and regard them as an extended
attribute entry.

Signed-off-by: Tiger Yang <tiger.yang@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: add ocfs2_xattr_get_nolock
Tiger Yang [Fri, 14 Nov 2008 03:16:53 +0000 (11:16 +0800)] 
ocfs2: add ocfs2_xattr_get_nolock

This function does the work of ocfs2_xattr_get under an open lock.

Signed-off-by: Tiger Yang <tiger.yang@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: add ocfs2_init_security in during file create
Tiger Yang [Fri, 14 Nov 2008 03:16:41 +0000 (11:16 +0800)] 
ocfs2: add ocfs2_init_security in during file create

Security attributes must be set when creating a new inode.

We do this in three steps.

- First, get security xattr's name and value by security_operation

- Calculate and reserve the meta data and clusters needed by this security
  xattr before starting transaction

- Finally, we set it before add_entry

Signed-off-by: Tiger Yang <tiger.yang@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: add security xattr API
Tiger Yang [Fri, 14 Nov 2008 03:16:27 +0000 (11:16 +0800)] 
ocfs2: add security xattr API

This patch add security xattr set/get/list APIs to
support security attributes in Ocfs2.

Signed-off-by: Tiger Yang <tiger.yang@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: add ocfs2_xattr_set_handle
Tiger Yang [Fri, 14 Nov 2008 03:16:03 +0000 (11:16 +0800)] 
ocfs2: add ocfs2_xattr_set_handle

This function is used to set xattr's in a started transaction. It is only
called during inode creation inode for initial security/acl xattrs of the
new inode. These xattrs could be put into ibody or extent block, so xattr
bucket would not be use in this case.

Signed-off-by: Tiger Yang <tiger.yang@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: move new inode allocation out of the transaction
Tiger Yang [Fri, 14 Nov 2008 03:15:44 +0000 (11:15 +0800)] 
ocfs2: move new inode allocation out of the transaction

Move out inode allocation from ocfs2_mknod_locked() because
vfs_dq_init() must be called outside of a transaction.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Tiger Yang <tiger.yang@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: turn __ocfs2_remove_inode_range() into ocfs2_remove_btree_range()
Mark Fasheh [Wed, 12 Nov 2008 23:16:38 +0000 (15:16 -0800)] 
ocfs2: turn __ocfs2_remove_inode_range() into ocfs2_remove_btree_range()

This patch genericizes the high level handling of extent removal.
ocfs2_remove_btree_range() is nearly identical to
__ocfs2_remove_inode_range(), except that extent tree operations have been
used where necessary. We update ocfs2_remove_inode_range() to use the
generic helper. Now extent tree based structures have an easy way to
truncate ranges.

Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Acked-by: Joel Becker <joel.becker@oracle.com>
15 years agoocfs2/xattr: Merge xattr set transaction.
Tao Ma [Wed, 12 Nov 2008 00:27:01 +0000 (08:27 +0800)] 
ocfs2/xattr: Merge xattr set transaction.

In current ocfs2/xattr, the whole xattr set is divided into
many steps are many transaction are used, this make the
xattr set process isn't like a real transaction, so this
patch try to merge all the transaction into one. Another
benefit is that acl can use it easily now.

I don't merge the transaction of deleting xattr when we
remove an inode. The reason is that if we have a large number
of xattrs and every xattrs has large values(large enough
for outside storage), the whole transaction will be very
huge and it looks like jbd can't handle it(I meet with a
jbd complain once). And the old inode removal is also divided
into many steps, so I'd like to leave as it is.

Note:
In xattr set, I try to avoid ocfs2_extend_trans since if
the credits aren't enough for the extension, it will commit
all the dirty blocks and create a new transaction which may
lead to inconsistency in metadata. All ocfs2_extend_trans
remained are safe now.

Signed-off-by: Tao Ma <tao.ma@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2/xattr: Reserve meta/data at the beginning of ocfs2_xattr_set.
Tao Ma [Wed, 12 Nov 2008 00:27:00 +0000 (08:27 +0800)] 
ocfs2/xattr: Reserve meta/data at the beginning of ocfs2_xattr_set.

In ocfs2 xattr set, we reserve metadata and clusters in any place
they are needed. It is time-consuming and ineffective, so this
patch try to reserve metadata and clusters at the beginning of
ocfs2_xattr_set.

Signed-off-by: Tao Ma <tao.ma@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2/xattr: Move clusters free into dealloc.
Tao Ma [Wed, 12 Nov 2008 00:26:59 +0000 (08:26 +0800)] 
ocfs2/xattr: Move clusters free into dealloc.

Move clusters free process into dealloc context so that
they can be freed after the transaction.

Signed-off-by: Tao Ma <tao.ma@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Add clusters free in dealloc_ctxt.
Tao Ma [Wed, 12 Nov 2008 00:26:58 +0000 (08:26 +0800)] 
ocfs2: Add clusters free in dealloc_ctxt.

Now in ocfs2 xattr set, the whole process are divided into many small
parts and they are wrapped into diffrent transactions and it make the
set doesn't look like a real transaction. So we want to integrate it
into a real one.

In some cases we will allocate some clusters and free some in just one
transaction. e.g, one xattr is larger than inline size, so it and its
value root is stored within the inode while the value is outside in a
cluster. Then we try to update it with a smaller value(larger than the
size of root but smaller than inline size), we may need to free the
outside cluster while allocate a new bucket(one cluster) since now the
inode may be full. The old solution will lock the global_bitmap(if the
local alloc failed in stress test) and then the truncate log. This will
cause a ABBA lock with truncate log flush.

This patch add the clusters free in dealloc_ctxt, so that we can record
the free clusters during the transaction and then free it after we
release the global_bitmap in xattr set.

Signed-off-by: Tao Ma <tao.ma@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2/xattr: Only extend xattr bucket in need.
Tao Ma [Wed, 12 Nov 2008 00:26:57 +0000 (08:26 +0800)] 
ocfs2/xattr: Only extend xattr bucket in need.

When the first block of a bucket is filled up with xattr
entries, we normally extend the bucket. But if we are
just replace one xattr with small length, we don't need
to extend it. This is important since we will calculate
what we need before the transaction and in this situation
no resources will be allocated.

Signed-off-by: Tao Ma <tao.ma@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2/xattr: Only set buffer update if it doesn't exist in cache.
Tao Ma [Thu, 6 Nov 2008 00:10:48 +0000 (08:10 +0800)] 
ocfs2/xattr: Only set buffer update if it doesn't exist in cache.

When we call ocfs2_init_xattr_bucket, we deem that the new buffer head
will be written to disk immediately, so we just use sb_getblk. But in
some cases the buffer may have already been in ocfs2 uptodate cache,
so we only call ocfs2_set_buffer_uptodate if the buffer head isn't
in the cache.

Signed-off-by: Tao Ma <tao.ma@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2/xattr: Remove additional bucket allocation in bucket defragment.
Tao Ma [Thu, 6 Nov 2008 00:10:47 +0000 (08:10 +0800)] 
ocfs2/xattr: Remove additional bucket allocation in bucket defragment.

Joel has refactored xattr bucket and make xattr bucket a general
wrapper. So in ocfs2_defrag_xattr_bucket, we have already passed the
bucket in, so there is no need to allocate a new one and read it.

Signed-off-by: Tao Ma <tao.ma@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Use buckets in ocfs2_xattr_set_entry_in_bucket().
Joel Becker [Tue, 28 Oct 2008 01:07:45 +0000 (18:07 -0700)] 
ocfs2: Use buckets in ocfs2_xattr_set_entry_in_bucket().

The ocfs2_xattr_set_entry_in_bucket() function is already working on an
ocfs2_xattr_bucket structure, so let's use the bucket API.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Use buckets in ocfs2_defrag_xattr_bucket().
Joel Becker [Mon, 27 Oct 2008 22:25:18 +0000 (15:25 -0700)] 
ocfs2: Use buckets in ocfs2_defrag_xattr_bucket().

Use the ocfs2_xattr_bucket abstraction for reading and writing the
bucket in ocfs2_defrag_xattr_bucket().

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Use buckets in ocfs2_xattr_create_index_block().
Joel Becker [Mon, 27 Oct 2008 22:18:29 +0000 (15:18 -0700)] 
ocfs2: Use buckets in ocfs2_xattr_create_index_block().

Use the ocfs2_xattr_bucket abstraction in
ocfs2_xattr_create_index_block() and its helpers.  We get more efficient
reads, a lot less buffer_head munging, and nicer code to boot.  While
we're at it, ocfs2_xattr_update_xattr_search() becomes void.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Use buckets in ocfs2_xattr_bucket_find().
Joel Becker [Mon, 27 Oct 2008 22:01:54 +0000 (15:01 -0700)] 
ocfs2: Use buckets in ocfs2_xattr_bucket_find().

Change the ocfs2_xattr_bucket_find() function to use ocfs2_xattr_bucket
as its abstraction.  This makes for more efficient reads, as buckets are
linear blocks, and also has improved caching characteristics.  It also
reads better.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Take ocfs2_xattr_bucket structures off of the stack.
Joel Becker [Sat, 25 Oct 2008 02:13:20 +0000 (19:13 -0700)] 
ocfs2: Take ocfs2_xattr_bucket structures off of the stack.

The ocfs2_xattr_bucket structure is a nice abstraction, but it is a bit
large to have on the stack.  Just like ocfs2_path, let's allocate it
with a ocfs2_xattr_bucket_new() function.

We can now store the inode on the bucket, cleaning up all the other
bucket functions.  While we're here, we catch another place or two that
wasn't using ocfs2_read_xattr_bucket().

Updates:
- No longer allocating xis.bucket, as it will never be used.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Copy xattr buckets with a dedicated function.
Joel Becker [Sat, 25 Oct 2008 01:54:43 +0000 (18:54 -0700)] 
ocfs2: Copy xattr buckets with a dedicated function.

Now that the places that copy whole buckets are using struct
ocfs2_xattr_bucket, we can do the copy in a dedicated function.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Wrap journal_access/journal_dirty for xattr buckets.
Joel Becker [Sat, 25 Oct 2008 01:47:33 +0000 (18:47 -0700)] 
ocfs2: Wrap journal_access/journal_dirty for xattr buckets.

A common action is to call ocfs2_journal_access() and
ocfs2_journal_dirty() on the buffer heads of an xattr bucket.  Let's
create nice wrappers.

While we're there, let's drop the places that try to be smart by writing
only the first and last blocks of a bucket.  A bucket is contiguous, so
writing the whole thing is actually more efficient.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Improve ocfs2_read_xattr_bucket().
Joel Becker [Sat, 25 Oct 2008 00:33:40 +0000 (17:33 -0700)] 
ocfs2: Improve ocfs2_read_xattr_bucket().

The ocfs2_read_xattr_bucket() function would read an xattr bucket into a
list of buffer heads.  However, we have a nice ocfs2_xattr_bucket
structure.  Let's have it fill that out instead.

In addition, ocfs2_read_xattr_bucket() would initialize buffer heads for
a bucket that's never been on disk before.  That's confusing.  Let's
call that functionality ocfs2_init_xattr_bucket().

The functions ocfs2_cp_xattr_bucket() and ocfs2_half_xattr_bucket() are
updated to use the ocfs2_xattr_bucket structure rather than raw bh
lists.  That way they can use the new read/init calls.  In addition,
they drop the wasted read of an existing target bucket.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Provide a wrapper to brelse() xattr bucket buffers.
Joel Becker [Sat, 25 Oct 2008 00:16:48 +0000 (17:16 -0700)] 
ocfs2: Provide a wrapper to brelse() xattr bucket buffers.

A common theme is walking all the buffer heads on an ocfs2_xattr_bucket
and releasing them.  Let's wrap that.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Convenient access to an xattr bucket's header.
Joel Becker [Sat, 25 Oct 2008 00:04:49 +0000 (17:04 -0700)] 
ocfs2: Convenient access to an xattr bucket's header.

The xattr code often wants to access the ocfs2_xattr_header at the start
of an bucket.  Rather than walk the pointer chains, let's just create
another nice macro.  As a side benefit, we can get rid of the mostly
spurious ->bu_xh element on the bucket structure.  The idea is ripped
from the ocfs2_path code.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Convenient access to xattr bucket data blocks.
Joel Becker [Fri, 24 Oct 2008 23:57:21 +0000 (16:57 -0700)] 
ocfs2: Convenient access to xattr bucket data blocks.

The xattr code often wants to access the data pointer for blocks in an
xattr bucket.  This is usually found by dereferencing the bh array
hanging off of the ocfs2_xattr_bucket structure.  Rather than do this
all the time, let's provide a nice little macro.  The idea is ripped
from the ocfs2_path code.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Convenient access to an xattr bucket's block number.
Joel Becker [Fri, 24 Oct 2008 23:21:03 +0000 (16:21 -0700)] 
ocfs2: Convenient access to an xattr bucket's block number.

The xattr code often wants to know the block number of an xattr bucket.
This is usually found by dereferencing the first bh hanging off of the
ocfs2_xattr_bucket structure.  Rather than do this all the time, let's
provide a nice little macro.  The idea is ripped from the ocfs2_path
code.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoocfs2: Field prefixes for the xattr_bucket structure
Joel Becker [Sun, 19 Oct 2008 02:11:42 +0000 (19:11 -0700)] 
ocfs2: Field prefixes for the xattr_bucket structure

The ocfs2_xattr_bucket structure keeps track of the buffers for one
xattr bucket.  Let's prefix the fields for easier code navigation.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
15 years agoASoC: fix davinci-sffsdr buglet
David Brownell [Mon, 5 Jan 2009 10:08:30 +0000 (02:08 -0800)] 
ASoC: fix davinci-sffsdr buglet

Minor bugfix:  now that DaVinci kernels can support multiple
boards, board-specific ASoC components need to verify they're
running on the right board before initializing.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
15 years agoMerge branch 'i2c-next-s3c' into i2c-next
Ben Dooks [Mon, 5 Jan 2009 10:54:50 +0000 (10:54 +0000)] 
Merge branch 'i2c-next-s3c' into i2c-next

15 years agoMerge branch 'topic/usbaudio' into for-linus
Takashi Iwai [Mon, 5 Jan 2009 09:53:58 +0000 (10:53 +0100)] 
Merge branch 'topic/usbaudio' into for-linus

15 years agoMerge branch 'topic/hda' into for-linus
Takashi Iwai [Mon, 5 Jan 2009 09:53:46 +0000 (10:53 +0100)] 
Merge branch 'topic/hda' into for-linus

15 years agoMerge branch 'topic/asoc' into for-linus
Takashi Iwai [Mon, 5 Jan 2009 09:53:43 +0000 (10:53 +0100)] 
Merge branch 'topic/asoc' into for-linus

15 years agotcp: Kill extraneous SPLICE_F_NONBLOCK checks.
David S. Miller [Mon, 5 Jan 2009 08:59:00 +0000 (00:59 -0800)] 
tcp: Kill extraneous SPLICE_F_NONBLOCK checks.

In splice TCP receive, the SPLICE_F_NONBLOCK flag is used
to compute the "timeo" value.  So checking it again inside
of the main receive loop to trigger -EAGAIN processing is
entirely unnecessary.

Noticed by Jarek P. and Lennert Buytenhek.

Signed-off-by: David S. Miller <davem@davemloft.net>
15 years agosparc: Remove reg*.h from Kbuild
David S. Miller [Mon, 5 Jan 2009 08:55:24 +0000 (00:55 -0800)] 
sparc: Remove reg*.h from Kbuild

Forgot to commit this in previous change, noticed by
Sam.

Signed-off-by: David S. Miller <davem@davemloft.net>
15 years agosparc: Clean arch-specific code in prom_common.c
Julian Calaby [Mon, 5 Jan 2009 08:07:18 +0000 (00:07 -0800)] 
sparc: Clean arch-specific code in prom_common.c

prom_nextprop() and prom_firstprop() have slightly different calling
conventions in 32 and 64 bit SPARC.

prom_common.c uses a ifdef guard to ensure that these functions are
called correctly.

Adjust code to eliminate this ifdef by using a calling convention that
is compatible with both 32 and 64 bit SPARC.

Signed-off-by: Julian Calaby <julian.calaby@gmail.com>
Reviewed-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
15 years agotcp: don't mask EOF and socket errors on nonblocking splice receive
Lennert Buytenhek [Mon, 5 Jan 2009 08:00:12 +0000 (00:00 -0800)] 
tcp: don't mask EOF and socket errors on nonblocking splice receive

Currently, setting SPLICE_F_NONBLOCK on splice from a TCP socket
results in masking of EOF (RDHUP) and error conditions on the socket
by an -EAGAIN return.  Move the NONBLOCK check in tcp_splice_read()
to be after the EOF and error checks to fix this.

Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
15 years agoGFS2: Use DEFINE_SPINLOCK
Julia Lawall [Thu, 25 Dec 2008 14:35:27 +0000 (15:35 +0100)] 
GFS2: Use DEFINE_SPINLOCK

SPIN_LOCK_UNLOCKED is deprecated.  The following makes the change suggested
in Documentation/spinlocks.txt

The semantic patch that makes this change is as follows:
(http://www.emn.fr/x-info/coccinelle/)

// <smpl>
@@
declarer name DEFINE_SPINLOCK;
identifier xxx_lock;
@@

- spinlock_t xxx_lock = SPIN_LOCK_UNLOCKED;
+ DEFINE_SPINLOCK(xxx_lock);
// </smpl>

Signed-off-by: Julia Lawall <julia@diku.dk>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
15 years agoGFS2: Fix use-after-free bug on umount (try #2)
Steven Whitehouse [Fri, 19 Dec 2008 15:43:05 +0000 (15:43 +0000)] 
GFS2: Fix use-after-free bug on umount (try #2)

This should solve the issue with the previous attempt at fixing this.

Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
15 years agoRevert "GFS2: Fix use-after-free bug on umount"
Steven Whitehouse [Fri, 19 Dec 2008 15:32:06 +0000 (15:32 +0000)] 
Revert "GFS2: Fix use-after-free bug on umount"

This reverts commit 78802499912f1ba31ce83a94c55b5a980f250a43.

The original patch is causing problems in relation to order of
operations at umount in relation to jdata files. I need to fix
this a different way.

Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
15 years agoGFS2: Streamline alloc calculations for writes
Steven Whitehouse [Wed, 10 Dec 2008 10:28:10 +0000 (10:28 +0000)] 
GFS2: Streamline alloc calculations for writes

This patch removes some unused code, and make the calculation
of the number of blocks required conditional in order to reduce
the number of times this (potentially expensive) calculation
is done.

Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
15 years agoGFS2: Send useful information with uevent messages
Steven Whitehouse [Thu, 27 Nov 2008 09:42:51 +0000 (09:42 +0000)] 
GFS2: Send useful information with uevent messages

In order to distinguish between two differing uevent messages
and to avoid using the (racy) method of reading status from
sysfs in future, this adds some status information to our
uevent messages.

Btw, before anybody says "sysfs isn't racy", I'm aware of that,
but the way that GFS2 was using it (send an ambiugous uevent and
then expect the receiver to read sysfs to find out the status
of the reported operation) was.

The additional benefit of using the new interface is that it
should be possible for a node to recover multiple journals
at the same time, since there is no longer any confusion as
to which journal the status belongs to.

At some future stage, when all the userland programs have been
converted, I intend to remove the old interface.

Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
15 years agoGFS2: Fix use-after-free bug on umount
Steven Whitehouse [Thu, 27 Nov 2008 08:27:28 +0000 (08:27 +0000)] 
GFS2: Fix use-after-free bug on umount

There was a use-after-free with the GFS2 super block during
umount. This patch moves almost all of the umount code from
->put_super into ->kill_sb, the only bit that cannot be moved
being the glock hash clearing which has to remain as ->put_super
due to umount ordering requirements. As a result its now obvious
that the kfree is the final operation, whereas before it was
hidden in ->put_super.

Also gfs2_jindex_free is then only referenced from a single file
so thats moved and marked static too.

Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>