2009-09-10 10:29:39

by Steven Whitehouse

[permalink] [raw]
Subject: GFS2: Pre-pull patch posting

As merge time is approaching, here is the current content of the
GFS2 -nmw git tree. I'm not expecting to take any more patches
now for the current merge window unless any last minute bugs
are discovered.

There is not a huge amount new this time. Some extra context for
uevent messages, better error handling during block allocation,
and a clean up of extended attribute support. There is still more
to do on the extended attribute side of things, but this is a good
start I think.

There are a few bug fixes as well. Once these patches are merged
I'm intending to start off the next -nmw tree with a patch to
remove some of the (now unused) sysfs files as per the message
on cluster-devel a few weeks back.

Steve.


2009-09-10 10:29:43

by Steven Whitehouse

[permalink] [raw]
Subject: [PATCH 01/15] GFS2: Add online uevent to GFS2

We already have an offline uevent (used when a withdraw occurs)
but no online uevent. This adds an online uevent so that userspace
will be able to detect a successful mount by means other than
not receiving a remove event after the add & recovery (change)
uevents.

It has also been added to the remount path as well - we can't use
a change uevent there as older GFS2 userspace acts on change uevents
according to the state that it thinks the fs is in, so we can't
easily add any new ones.

Signed-off-by: Steven Whitehouse <[email protected]>
---
fs/gfs2/ops_fstype.c | 13 ++++++++++++-
fs/gfs2/super.c | 1 +
fs/gfs2/super.h | 4 ++--
3 files changed, 15 insertions(+), 3 deletions(-)

diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index 7bc3c45..39021c0 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -1133,6 +1133,17 @@ void gfs2_lm_unmount(struct gfs2_sbd *sdp)
lm->lm_unmount(sdp);
}

+void gfs2_online_uevent(struct gfs2_sbd *sdp)
+{
+ struct super_block *sb = sdp->sd_vfs;
+ char ro[20];
+ char spectator[20];
+ char *envp[] = { ro, spectator, NULL };
+ sprintf(ro, "RDONLY=%d", (sb->s_flags & MS_RDONLY) ? 1 : 0);
+ sprintf(spectator, "SPECTATOR=%d", sdp->sd_args.ar_spectator ? 1 : 0);
+ kobject_uevent_env(&sdp->sd_kobj, KOBJ_ONLINE, envp);
+}
+
/**
* fill_super - Read in superblock
* @sb: The VFS superblock
@@ -1236,7 +1247,7 @@ static int fill_super(struct super_block *sb, void *data, int silent)
}

gfs2_glock_dq_uninit(&mount_gh);
-
+ gfs2_online_uevent(sdp);
return 0;

fail_threads:
diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c
index f522bb0..85bd2bc 100644
--- a/fs/gfs2/super.c
+++ b/fs/gfs2/super.c
@@ -1084,6 +1084,7 @@ static int gfs2_remount_fs(struct super_block *sb, int *flags, char *data)
gt->gt_log_flush_secs = args.ar_commit;
spin_unlock(&gt->gt_spin);

+ gfs2_online_uevent(sdp);
return 0;
}

diff --git a/fs/gfs2/super.h b/fs/gfs2/super.h
index 22e0417..911c954 100644
--- a/fs/gfs2/super.h
+++ b/fs/gfs2/super.h
@@ -25,7 +25,7 @@ static inline unsigned int gfs2_jindex_size(struct gfs2_sbd *sdp)
return x;
}

-void gfs2_jindex_free(struct gfs2_sbd *sdp);
+extern void gfs2_jindex_free(struct gfs2_sbd *sdp);

extern int gfs2_mount_args(struct gfs2_sbd *sdp, struct gfs2_args *args, char *data);

@@ -36,7 +36,7 @@ extern int gfs2_lookup_in_master_dir(struct gfs2_sbd *sdp, char *filename,
struct gfs2_inode **ipp);

extern int gfs2_make_fs_rw(struct gfs2_sbd *sdp);
-
+extern void gfs2_online_uevent(struct gfs2_sbd *sdp);
extern int gfs2_statfs_init(struct gfs2_sbd *sdp);
extern void gfs2_statfs_change(struct gfs2_sbd *sdp, s64 total, s64 free,
s64 dinodes);
--
1.6.2.5

2009-09-10 10:29:45

by Steven Whitehouse

[permalink] [raw]
Subject: [PATCH 02/15] GFS2: Add some more info to uevents

With each uevent, we now always include the journal ID. We
can't call it JID since that is already in use by some of
the individual events relating to recovery, so we use
JOURNALID instead. We don't send the JOURNALID for spectator
mounts, since there isn't one.

Also the ADD event now has both RDONLY and SPECTATOR information
to match that of the ONLINE event.

Signed-off-by: Steven Whitehouse <[email protected]>
---
fs/gfs2/sys.c | 13 ++++++++++---
1 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c
index a7cbfbd..dda4bf1 100644
--- a/fs/gfs2/sys.c
+++ b/fs/gfs2/sys.c
@@ -519,7 +519,14 @@ static struct attribute_group lock_module_group = {

int gfs2_sys_fs_add(struct gfs2_sbd *sdp)
{
+ struct super_block *sb = sdp->sd_vfs;
int error;
+ char ro[20];
+ char spectator[20];
+ char *envp[] = { ro, spectator, NULL };
+
+ sprintf(ro, "RDONLY=%d", (sb->s_flags & MS_RDONLY) ? 1 : 0);
+ sprintf(spectator, "SPECTATOR=%d", sdp->sd_args.ar_spectator ? 1 : 0);

sdp->sd_kobj.kset = gfs2_kset;
error = kobject_init_and_add(&sdp->sd_kobj, &gfs2_ktype, NULL,
@@ -535,7 +542,7 @@ int gfs2_sys_fs_add(struct gfs2_sbd *sdp)
if (error)
goto fail_tune;

- kobject_uevent(&sdp->sd_kobj, KOBJ_ADD);
+ kobject_uevent_env(&sdp->sd_kobj, KOBJ_ADD, envp);
return 0;

fail_tune:
@@ -554,7 +561,6 @@ void gfs2_sys_fs_del(struct gfs2_sbd *sdp)
kobject_put(&sdp->sd_kobj);
}

-
static int gfs2_uevent(struct kset *kset, struct kobject *kobj,
struct kobj_uevent_env *env)
{
@@ -563,6 +569,8 @@ static int gfs2_uevent(struct kset *kset, struct kobject *kobj,

add_uevent_var(env, "LOCKTABLE=%s", sdp->sd_table_name);
add_uevent_var(env, "LOCKPROTO=%s", sdp->sd_proto_name);
+ if (!sdp->sd_args.ar_spectator)
+ add_uevent_var(env, "JOURNALID=%u", sdp->sd_lockstruct.ls_jid);
if (gfs2_uuid_valid(uuid)) {
add_uevent_var(env, "UUID=%02X%02X%02X%02X-%02X%02X-%02X%02X-"
"%02X%02X-%02X%02X%02X%02X%02X%02X",
@@ -578,7 +586,6 @@ static struct kset_uevent_ops gfs2_uevent_ops = {
.uevent = gfs2_uevent,
};

-
int gfs2_sys_init(void)
{
gfs2_kset = kset_create_and_add("gfs2", &gfs2_uevent_ops, fs_kobj);
--
1.6.2.5

2009-09-10 10:32:47

by Steven Whitehouse

[permalink] [raw]
Subject: [PATCH 03/15] GFS2: Improve error handling in inode allocation

A little while back, block allocation was given some improved
error handling which meant that -EIO was returned in the case
of there being a problem in the resource group data. In addition
a message is printed explaning what went wrong and how to fix it.
This extends that error handling so that it also covers inode
allocation too.

Signed-off-by: Steven Whitehouse <[email protected]>
---
fs/gfs2/inode.c | 2 +-
fs/gfs2/rgrp.c | 36 +++++++++++++++++++++++++-----------
fs/gfs2/rgrp.h | 2 +-
3 files changed, 27 insertions(+), 13 deletions(-)

diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c
index 2f94bd7..f9b4fe8 100644
--- a/fs/gfs2/inode.c
+++ b/fs/gfs2/inode.c
@@ -731,7 +731,7 @@ static int alloc_dinode(struct gfs2_inode *dip, u64 *no_addr, u64 *generation)
if (error)
goto out_ipreserv;

- *no_addr = gfs2_alloc_di(dip, generation);
+ error = gfs2_alloc_di(dip, no_addr, generation);

gfs2_trans_end(sdp);

diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
index fba7957..3d0193a 100644
--- a/fs/gfs2/rgrp.c
+++ b/fs/gfs2/rgrp.c
@@ -1459,6 +1459,16 @@ int gfs2_rgrp_dump(struct seq_file *seq, const struct gfs2_glock *gl)
return 0;
}

+static void gfs2_rgrp_error(struct gfs2_rgrpd *rgd)
+{
+ struct gfs2_sbd *sdp = rgd->rd_sbd;
+ fs_warn(sdp, "rgrp %llu has an error, marking it readonly until umount\n",
+ (unsigned long long)rgd->rd_addr);
+ fs_warn(sdp, "umount on all nodes and run fsck.gfs2 to fix the error\n");
+ gfs2_rgrp_dump(NULL, rgd->rd_gl);
+ rgd->rd_flags |= GFS2_RDF_ERROR;
+}
+
/**
* gfs2_alloc_block - Allocate one or more blocks
* @ip: the inode to allocate the block for
@@ -1520,22 +1530,20 @@ int gfs2_alloc_block(struct gfs2_inode *ip, u64 *bn, unsigned int *n)
return 0;

rgrp_error:
- fs_warn(sdp, "rgrp %llu has an error, marking it readonly until umount\n",
- (unsigned long long)rgd->rd_addr);
- fs_warn(sdp, "umount on all nodes and run fsck.gfs2 to fix the error\n");
- gfs2_rgrp_dump(NULL, rgd->rd_gl);
- rgd->rd_flags |= GFS2_RDF_ERROR;
+ gfs2_rgrp_error(rgd);
return -EIO;
}

/**
* gfs2_alloc_di - Allocate a dinode
* @dip: the directory that the inode is going in
+ * @bn: the block number which is allocated
+ * @generation: the generation number of the inode
*
- * Returns: the block allocated
+ * Returns: 0 on success or error
*/

-u64 gfs2_alloc_di(struct gfs2_inode *dip, u64 *generation)
+int gfs2_alloc_di(struct gfs2_inode *dip, u64 *bn, u64 *generation)
{
struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode);
struct gfs2_alloc *al = dip->i_alloc;
@@ -1546,12 +1554,13 @@ u64 gfs2_alloc_di(struct gfs2_inode *dip, u64 *generation)

blk = rgblk_search(rgd, rgd->rd_last_alloc,
GFS2_BLKST_FREE, GFS2_BLKST_DINODE, &n);
- BUG_ON(blk == BFITNOENT);

- rgd->rd_last_alloc = blk;
+ /* Since all blocks are reserved in advance, this shouldn't happen */
+ if (blk == BFITNOENT)
+ goto rgrp_error;

+ rgd->rd_last_alloc = blk;
block = rgd->rd_data0 + blk;
-
gfs2_assert_withdraw(sdp, rgd->rd_free);
rgd->rd_free--;
rgd->rd_dinodes++;
@@ -1568,7 +1577,12 @@ u64 gfs2_alloc_di(struct gfs2_inode *dip, u64 *generation)
rgd->rd_free_clone--;
spin_unlock(&sdp->sd_rindex_spin);
trace_gfs2_block_alloc(dip, block, 1, GFS2_BLKST_DINODE);
- return block;
+ *bn = block;
+ return 0;
+
+rgrp_error:
+ gfs2_rgrp_error(rgd);
+ return -EIO;
}

/**
diff --git a/fs/gfs2/rgrp.h b/fs/gfs2/rgrp.h
index 1e76ff0..a8dedd7 100644
--- a/fs/gfs2/rgrp.h
+++ b/fs/gfs2/rgrp.h
@@ -47,7 +47,7 @@ extern void gfs2_inplace_release(struct gfs2_inode *ip);
extern unsigned char gfs2_get_block_type(struct gfs2_rgrpd *rgd, u64 block);

extern int gfs2_alloc_block(struct gfs2_inode *ip, u64 *bn, unsigned int *n);
-extern u64 gfs2_alloc_di(struct gfs2_inode *ip, u64 *generation);
+extern int gfs2_alloc_di(struct gfs2_inode *ip, u64 *bn, u64 *generation);

extern void gfs2_free_data(struct gfs2_inode *ip, u64 bstart, u32 blen);
extern void gfs2_free_meta(struct gfs2_inode *ip, u64 bstart, u32 blen);
--
1.6.2.5

2009-09-10 10:32:35

by Steven Whitehouse

[permalink] [raw]
Subject: [PATCH 04/15] GFS2: Replace assertion with proper error handling

One fewer assert, one more place we can recover gracefully
if there is an error.

Signed-off-by: Steven Whitehouse <[email protected]>
---
fs/gfs2/rgrp.c | 4 +++-
1 files changed, 3 insertions(+), 1 deletions(-)

diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
index 3d0193a..c681c54 100644
--- a/fs/gfs2/rgrp.c
+++ b/fs/gfs2/rgrp.c
@@ -1561,7 +1561,9 @@ int gfs2_alloc_di(struct gfs2_inode *dip, u64 *bn, u64 *generation)

rgd->rd_last_alloc = blk;
block = rgd->rd_data0 + blk;
- gfs2_assert_withdraw(sdp, rgd->rd_free);
+ if (rgd->rd_free == 0)
+ goto rgrp_error;
+
rgd->rd_free--;
rgd->rd_dinodes++;
*generation = rgd->rd_igeneration++;
--
1.6.2.5

2009-09-10 10:29:50

by Steven Whitehouse

[permalink] [raw]
Subject: [PATCH 05/15] GFS2: Add sysfs link to device

This adds a link from the per-gfs2 sb sysfs directory to
the block device upon which the filesystem is mounted. The
link is called "device", strangely enough :-)

Signed-off-by: Steven Whitehouse <[email protected]>
---
fs/gfs2/sys.c | 10 ++++++++++
1 files changed, 10 insertions(+), 0 deletions(-)

diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c
index dda4bf1..0d4f7e9 100644
--- a/fs/gfs2/sys.c
+++ b/fs/gfs2/sys.c
@@ -16,6 +16,7 @@
#include <linux/kobject.h>
#include <asm/uaccess.h>
#include <linux/gfs2_ondisk.h>
+#include <linux/genhd.h>

#include "gfs2.h"
#include "incore.h"
@@ -542,9 +543,17 @@ int gfs2_sys_fs_add(struct gfs2_sbd *sdp)
if (error)
goto fail_tune;

+ error = sysfs_create_link(&sdp->sd_kobj,
+ &disk_to_dev(sb->s_bdev->bd_disk)->kobj,
+ "device");
+ if (error)
+ goto fail_lock_module;
+
kobject_uevent_env(&sdp->sd_kobj, KOBJ_ADD, envp);
return 0;

+fail_lock_module:
+ sysfs_remove_group(&sdp->sd_kobj, &lock_module_group);
fail_tune:
sysfs_remove_group(&sdp->sd_kobj, &tune_group);
fail_reg:
@@ -556,6 +565,7 @@ fail:

void gfs2_sys_fs_del(struct gfs2_sbd *sdp)
{
+ sysfs_remove_link(&sdp->sd_kobj, "device");
sysfs_remove_group(&sdp->sd_kobj, &tune_group);
sysfs_remove_group(&sdp->sd_kobj, &lock_module_group);
kobject_put(&sdp->sd_kobj);
--
1.6.2.5

2009-09-10 10:29:52

by Steven Whitehouse

[permalink] [raw]
Subject: [PATCH 06/15] GFS2: Add a document explaining GFS2's uevents

This will be essential reading for anybody who wants to
understand how GFS2 interacts with the userland gfs_controld,
and the details of recovery.

Signed-off-by: Steven Whitehouse <[email protected]>
Signed-off-by: Bob Peterson <[email protected]>
---
Documentation/filesystems/gfs2-uevents.txt | 100 ++++++++++++++++++++++++++++
1 files changed, 100 insertions(+), 0 deletions(-)
create mode 100644 Documentation/filesystems/gfs2-uevents.txt

diff --git a/Documentation/filesystems/gfs2-uevents.txt b/Documentation/filesystems/gfs2-uevents.txt
new file mode 100644
index 0000000..fd966dc
--- /dev/null
+++ b/Documentation/filesystems/gfs2-uevents.txt
@@ -0,0 +1,100 @@
+ uevents and GFS2
+ ==================
+
+During the lifetime of a GFS2 mount, a number of uevents are generated.
+This document explains what the events are and what they are used
+for (by gfs_controld in gfs2-utils).
+
+A list of GFS2 uevents
+-----------------------
+
+1. ADD
+
+The ADD event occurs at mount time. It will always be the first
+uevent generated by the newly created filesystem. If the mount
+is successful, an ONLINE uevent will follow. If it is not successful
+then a REMOVE uevent will follow.
+
+The ADD uevent has two environment variables: SPECTATOR=[0|1]
+and RDONLY=[0|1] that specify the spectator status (a read-only mount
+with no journal assigned), and read-only (with journal assigned) status
+of the filesystem respectively.
+
+2. ONLINE
+
+The ONLINE uevent is generated after a successful mount or remount. It
+has the same environment variables as the ADD uevent. The ONLINE
+uevent, along with the two environment variables for spectator and
+RDONLY are a relatively recent addition (2.6.32-rc+) and will not
+be generated by older kernels.
+
+3. CHANGE
+
+The CHANGE uevent is used in two places. One is when reporting the
+successful mount of the filesystem by the first node (FIRSTMOUNT=Done).
+This is used as a signal by gfs_controld that it is then ok for other
+nodes in the cluster to mount the filesystem.
+
+The other CHANGE uevent is used to inform of the completion
+of journal recovery for one of the filesystems journals. It has
+two environment variables, JID= which specifies the journal id which
+has just been recovered, and RECOVERY=[Done|Failed] to indicate the
+success (or otherwise) of the operation. These uevents are generated
+for every journal recovered, whether it is during the initial mount
+process or as the result of gfs_controld requesting a specific journal
+recovery via the /sys/fs/gfs2/<fsname>/lock_module/recovery file.
+
+Because the CHANGE uevent was used (in early versions of gfs_controld)
+without checking the environment variables to discover the state, we
+cannot add any more functions to it without running the risk of
+someone using an older version of the user tools and breaking their
+cluster. For this reason the ONLINE uevent was used when adding a new
+uevent for a successful mount or remount.
+
+4. OFFLINE
+
+The OFFLINE uevent is only generated due to filesystem errors and is used
+as part of the "withdraw" mechanism. Currently this doesn't give any
+information about what the error is, which is something that needs to
+be fixed.
+
+5. REMOVE
+
+The REMOVE uevent is generated at the end of an unsuccessful mount
+or at the end of a umount of the filesystem. All REMOVE uevents will
+have been preceeded by at least an ADD uevent for the same fileystem,
+and unlike the other uevents is generated automatically by the kernel's
+kobject subsystem.
+
+
+Information common to all GFS2 uevents (uevent environment variables)
+----------------------------------------------------------------------
+
+1. LOCKTABLE=
+
+The LOCKTABLE is a string, as supplied on the mount command
+line (locktable=) or via fstab. It is used as a filesystem label
+as well as providing the information for a lock_dlm mount to be
+able to join the cluster.
+
+2. LOCKPROTO=
+
+The LOCKPROTO is a string, and its value depends on what is set
+on the mount command line, or via fstab. It will be either
+lock_nolock or lock_dlm. In the future other lock managers
+may be supported.
+
+3. JOURNALID=
+
+If a journal is in use by the filesystem (journals are not
+assigned for spectator mounts) then this will give the
+numeric journal id in all GFS2 uevents.
+
+4. UUID=
+
+With recent versions of gfs2-utils, mkfs.gfs2 writes a UUID
+into the filesystem superblock. If it exists, this will
+be included in every uevent relating to the filesystem.
+
+
+
--
1.6.2.5

2009-09-10 10:30:04

by Steven Whitehouse

[permalink] [raw]
Subject: [PATCH 07/15] GFS2: free disk inode which is deleted by remote node -V2

From: Wengang Wang <[email protected]>

this patch is for the same problem that Benjamin Marzinski fixes at commit
b94a170e96dc416828af9d350ae2e34b70ae7347

quotation of the original problem:

---cut here---
When a file is deleted from a gfs2 filesystem on one node, a dcache
entry for it may still exist on other nodes in the cluster. If this
happens, gfs2 will be unable to free this file on disk. Because of this,
it's possible to have a gfs2 filesystem with no files on it and no free
space. With this patch, when a node receives a callback notifying it
that the file is being deleted on another node, it schedules a new
workqueue thread to remove the file's dcache entry.
---end cut---

after applying Benjamin's patch, I think there is still a case in which the disk
inode remains even when "no space" is hit. the case is that when running
d_prune_aliases() against the inode, there are one or more dentries(aliases)
which have reference count number > 0. in this case the dentries won't be pruned.
and even later, the reference count becomes to 0, the dentries can still be
cached in memory. unfortunately, no callback come again, things come back to
the state before the callback runs. thus the on disk inode remains there until
in memoryinode is removed for some other reason(shrinking inode cache or unmount
the volume..).

this patch is to remove those dentries when their reference count becomes to 0 and
the inode is deleted by remote node. for implementation, gfs2_dentry_delete() is
added as dentry_operations.d_delete. the function returns true when the inode is
deleted by remote node. in dput(), gfs2_dentry_delete() is called and since it
returns true, the dentry is unhashed from dcache and then removed. when all dentries
are removed, the in memory inode get removed so that the on disk inode is freed.

Signed-off-by: Wengang Wang <[email protected]>
Signed-off-by: Steven Whitehouse <[email protected]>
---
fs/gfs2/dentry.c | 18 ++++++++++++++++++
1 files changed, 18 insertions(+), 0 deletions(-)

diff --git a/fs/gfs2/dentry.c b/fs/gfs2/dentry.c
index 022c66c..91bedda 100644
--- a/fs/gfs2/dentry.c
+++ b/fs/gfs2/dentry.c
@@ -107,8 +107,26 @@ static int gfs2_dhash(struct dentry *dentry, struct qstr *str)
return 0;
}

+static int gfs2_dentry_delete(struct dentry *dentry)
+{
+ struct gfs2_inode *ginode;
+
+ if (!dentry->d_inode)
+ return 0;
+
+ ginode = GFS2_I(dentry->d_inode);
+ if (!ginode->i_iopen_gh.gh_gl)
+ return 0;
+
+ if (test_bit(GLF_DEMOTE, &ginode->i_iopen_gh.gh_gl->gl_flags))
+ return 1;
+
+ return 0;
+}
+
const struct dentry_operations gfs2_dops = {
.d_revalidate = gfs2_drevalidate,
.d_hash = gfs2_dhash,
+ .d_delete = gfs2_dentry_delete,
};

--
1.6.2.5

2009-09-10 10:31:49

by Steven Whitehouse

[permalink] [raw]
Subject: [PATCH 08/15] GFS2: jumping to wrong label?

From: Roel Kluin <[email protected]>

Also a gfs2_glock_dq() is required here.

Signed-off-by: Roel Kluin <[email protected]>
Signed-off-by: Steven Whitehouse <[email protected]>
---
fs/gfs2/ops_inode.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c
index f8bd20b..ce551f8 100644
--- a/fs/gfs2/ops_inode.c
+++ b/fs/gfs2/ops_inode.c
@@ -349,7 +349,7 @@ static int gfs2_unlink(struct inode *dir, struct dentry *dentry)

error = gfs2_trans_begin(sdp, 2*RES_DINODE + RES_LEAF + RES_RG_BIT, 0);
if (error)
- goto out_rgrp;
+ goto out_gunlock;

error = gfs2_dir_del(dip, &dentry->d_name);
if (error)
--
1.6.2.5

2009-09-10 10:29:58

by Steven Whitehouse

[permalink] [raw]
Subject: [PATCH 09/15] GFS2: Add "-o errors=panic|withdraw" mount options

From: Bob Peterson <[email protected]>

This patch adds "-o errors=panic" and "-o errors=withdraw" to the
gfs2 mount options. The "errors=withdraw" option is today's
current behaviour, meaning to withdraw from the file system if a
non-serious gfs2 error occurs. The new "errors=panic" option
tells gfs2 to force a kernel panic if a non-serious gfs2 file
system error occurs. This may be useful, for example, where
fabric-level fencing is used that has no way to reboot (such as
fence_scsi).

Signed-off-by: Bob Peterson <[email protected]>
Signed-off-by: Steven Whitehouse <[email protected]>
---
fs/gfs2/incore.h | 7 +++++++
fs/gfs2/ops_fstype.c | 1 +
fs/gfs2/super.c | 36 ++++++++++++++++++++++++++++++++++++
fs/gfs2/util.c | 41 +++++++++++++++++++++++++++--------------
4 files changed, 71 insertions(+), 14 deletions(-)

diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h
index 61801ad..1d11e6e 100644
--- a/fs/gfs2/incore.h
+++ b/fs/gfs2/incore.h
@@ -406,6 +406,12 @@ struct gfs2_statfs_change_host {
#define GFS2_DATA_WRITEBACK 1
#define GFS2_DATA_ORDERED 2

+#define GFS2_ERRORS_DEFAULT GFS2_ERRORS_WITHDRAW
+#define GFS2_ERRORS_WITHDRAW 0
+#define GFS2_ERRORS_CONTINUE 1 /* place holder for future feature */
+#define GFS2_ERRORS_RO 2 /* place holder for future feature */
+#define GFS2_ERRORS_PANIC 3
+
struct gfs2_args {
char ar_lockproto[GFS2_LOCKNAME_LEN]; /* Name of the Lock Protocol */
char ar_locktable[GFS2_LOCKNAME_LEN]; /* Name of the Lock Table */
@@ -422,6 +428,7 @@ struct gfs2_args {
unsigned int ar_data:2; /* ordered/writeback */
unsigned int ar_meta:1; /* mount metafs */
unsigned int ar_discard:1; /* discard requests */
+ unsigned int ar_errors:2; /* errors=withdraw | panic */
int ar_commit; /* Commit interval */
};

diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index 39021c0..165518a 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -1168,6 +1168,7 @@ static int fill_super(struct super_block *sb, void *data, int silent)
sdp->sd_args.ar_quota = GFS2_QUOTA_DEFAULT;
sdp->sd_args.ar_data = GFS2_DATA_DEFAULT;
sdp->sd_args.ar_commit = 60;
+ sdp->sd_args.ar_errors = GFS2_ERRORS_DEFAULT;

error = gfs2_mount_args(sdp, &sdp->sd_args, data);
if (error) {
diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c
index 85bd2bc..7a5c128 100644
--- a/fs/gfs2/super.c
+++ b/fs/gfs2/super.c
@@ -68,6 +68,8 @@ enum {
Opt_discard,
Opt_nodiscard,
Opt_commit,
+ Opt_err_withdraw,
+ Opt_err_panic,
Opt_error,
};

@@ -97,6 +99,8 @@ static const match_table_t tokens = {
{Opt_discard, "discard"},
{Opt_nodiscard, "nodiscard"},
{Opt_commit, "commit=%d"},
+ {Opt_err_withdraw, "errors=withdraw"},
+ {Opt_err_panic, "errors=panic"},
{Opt_error, NULL}
};

@@ -152,6 +156,11 @@ int gfs2_mount_args(struct gfs2_sbd *sdp, struct gfs2_args *args, char *options)
args->ar_localcaching = 1;
break;
case Opt_debug:
+ if (args->ar_errors == GFS2_ERRORS_PANIC) {
+ fs_info(sdp, "-o debug and -o errors=panic "
+ "are mutually exclusive.\n");
+ return -EINVAL;
+ }
args->ar_debug = 1;
break;
case Opt_nodebug:
@@ -205,6 +214,17 @@ int gfs2_mount_args(struct gfs2_sbd *sdp, struct gfs2_args *args, char *options)
return rv ? rv : -EINVAL;
}
break;
+ case Opt_err_withdraw:
+ args->ar_errors = GFS2_ERRORS_WITHDRAW;
+ break;
+ case Opt_err_panic:
+ if (args->ar_debug) {
+ fs_info(sdp, "-o debug and -o errors=panic "
+ "are mutually exclusive.\n");
+ return -EINVAL;
+ }
+ args->ar_errors = GFS2_ERRORS_PANIC;
+ break;
case Opt_error:
default:
fs_info(sdp, "invalid mount option: %s\n", o);
@@ -1226,6 +1246,22 @@ static int gfs2_show_options(struct seq_file *s, struct vfsmount *mnt)
lfsecs = sdp->sd_tune.gt_log_flush_secs;
if (lfsecs != 60)
seq_printf(s, ",commit=%d", lfsecs);
+ if (args->ar_errors != GFS2_ERRORS_DEFAULT) {
+ const char *state;
+
+ switch (args->ar_errors) {
+ case GFS2_ERRORS_WITHDRAW:
+ state = "withdraw";
+ break;
+ case GFS2_ERRORS_PANIC:
+ state = "panic";
+ break;
+ default:
+ state = "unknown";
+ break;
+ }
+ seq_printf(s, ",errors=%s", state);
+ }
return 0;
}

diff --git a/fs/gfs2/util.c b/fs/gfs2/util.c
index 9d12b11..f6a7efa 100644
--- a/fs/gfs2/util.c
+++ b/fs/gfs2/util.c
@@ -38,24 +38,30 @@ int gfs2_lm_withdraw(struct gfs2_sbd *sdp, char *fmt, ...)
const struct lm_lockops *lm = ls->ls_ops;
va_list args;

- if (test_and_set_bit(SDF_SHUTDOWN, &sdp->sd_flags))
+ if (sdp->sd_args.ar_errors == GFS2_ERRORS_WITHDRAW &&
+ test_and_set_bit(SDF_SHUTDOWN, &sdp->sd_flags))
return 0;

va_start(args, fmt);
vprintk(fmt, args);
va_end(args);

- fs_err(sdp, "about to withdraw this file system\n");
- BUG_ON(sdp->sd_args.ar_debug);
+ if (sdp->sd_args.ar_errors == GFS2_ERRORS_WITHDRAW) {
+ fs_err(sdp, "about to withdraw this file system\n");
+ BUG_ON(sdp->sd_args.ar_debug);

- kobject_uevent(&sdp->sd_kobj, KOBJ_OFFLINE);
+ kobject_uevent(&sdp->sd_kobj, KOBJ_OFFLINE);

- if (lm->lm_unmount) {
- fs_err(sdp, "telling LM to unmount\n");
- lm->lm_unmount(sdp);
+ if (lm->lm_unmount) {
+ fs_err(sdp, "telling LM to unmount\n");
+ lm->lm_unmount(sdp);
+ }
+ fs_err(sdp, "withdrawn\n");
+ dump_stack();
}
- fs_err(sdp, "withdrawn\n");
- dump_stack();
+
+ if (sdp->sd_args.ar_errors == GFS2_ERRORS_PANIC)
+ panic("GFS2: fsid=%s: panic requested.\n", sdp->sd_fsname);

return -1;
}
@@ -93,17 +99,24 @@ int gfs2_assert_warn_i(struct gfs2_sbd *sdp, char *assertion,
gfs2_tune_get(sdp, gt_complain_secs) * HZ))
return -2;

- printk(KERN_WARNING
- "GFS2: fsid=%s: warning: assertion \"%s\" failed\n"
- "GFS2: fsid=%s: function = %s, file = %s, line = %u\n",
- sdp->sd_fsname, assertion,
- sdp->sd_fsname, function, file, line);
+ if (sdp->sd_args.ar_errors == GFS2_ERRORS_WITHDRAW)
+ printk(KERN_WARNING
+ "GFS2: fsid=%s: warning: assertion \"%s\" failed\n"
+ "GFS2: fsid=%s: function = %s, file = %s, line = %u\n",
+ sdp->sd_fsname, assertion,
+ sdp->sd_fsname, function, file, line);

if (sdp->sd_args.ar_debug)
BUG();
else
dump_stack();

+ if (sdp->sd_args.ar_errors == GFS2_ERRORS_PANIC)
+ panic("GFS2: fsid=%s: warning: assertion \"%s\" failed\n"
+ "GFS2: fsid=%s: function = %s, file = %s, line = %u\n",
+ sdp->sd_fsname, assertion,
+ sdp->sd_fsname, function, file, line);
+
sdp->sd_last_warning = jiffies;

return -1;
--
1.6.2.5

2009-09-10 10:30:01

by Steven Whitehouse

[permalink] [raw]
Subject: [PATCH 10/15] GFS2: Add explanation of extended attr on-disk format

Some useful info regarding the on-disk representation of
GFS2 extended attributes.

Signed-off-by: Steven Whitehouse <[email protected]>
---
include/linux/gfs2_ondisk.h | 22 ++++++++++++++++++++++
1 files changed, 22 insertions(+), 0 deletions(-)

diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h
index c56b4bc..b80c88d 100644
--- a/include/linux/gfs2_ondisk.h
+++ b/include/linux/gfs2_ondisk.h
@@ -333,6 +333,28 @@ struct gfs2_leaf {

/*
* Extended attribute header format
+ *
+ * This works in a similar way to dirents. There is a fixed size header
+ * followed by a variable length section made up of the name and the
+ * associated data. In the case of a "stuffed" entry, the value is
+ * inline directly after the name, the ea_num_ptrs entry will be
+ * zero in that case. For non-"stuffed" entries, there will be
+ * a set of pointers (aligned to 8 byte boundary) to the block(s)
+ * containing the value.
+ *
+ * The blocks containing the values and the blocks containing the
+ * extended attribute headers themselves all start with the common
+ * metadata header. Each inode, if it has extended attributes, will
+ * have either a single block containing the extended attribute headers
+ * or a single indirect block pointing to blocks containing the
+ * extended attribure headers.
+ *
+ * The maximim size of the data part of an extended attribute is 64k
+ * so the number of blocks required depends upon block size. Since the
+ * block size also determines the number of pointers in an indirect
+ * block, its a fairly complicated calculation to work out the maximum
+ * number of blocks that an inode may have relating to extended attributes.
+ *
*/

#define GFS2_EA_MAX_NAME_LEN 255
--
1.6.2.5

2009-09-10 10:30:47

by Steven Whitehouse

[permalink] [raw]
Subject: [PATCH 11/15] GFS2: Clean up of extended attribute support

This has been on my list for some time. We need to change the way
in which we handle extended attributes to allow faster file creation
times (by reducing the number of transactions required) and the
extended attribute code is the main obstacle to this.

In addition to that, the VFS provides a way to demultiplex the xattr
calls which we ought to be using, rather than rolling our own. This
patch changes the GFS2 code to use that VFS feature and as a result
the code shrinks by a couple of hundred lines or so, and becomes
easier to read.

I'm planning on doing further clean up work in this area, but this
patch is a good start. The cleaned up code also uses the more usual
"xattr" shorthand, I plan to eliminate the use of "eattr" eventually
and in the mean time it serves as a flag as to which bits of the code
have been updated.

Signed-off-by: Steven Whitehouse <[email protected]>
---
fs/gfs2/Makefile | 2 +-
fs/gfs2/acl.c | 104 ++++++-------
fs/gfs2/eaops.c | 157 -------------------
fs/gfs2/eaops.h | 30 ----
fs/gfs2/eattr.c | 421 ++++++++++++++++++++++++++++----------------------
fs/gfs2/eattr.h | 52 ++-----
fs/gfs2/file.c | 1 -
fs/gfs2/inode.c | 12 +--
fs/gfs2/ops_fstype.c | 1 +
fs/gfs2/ops_inode.c | 78 ++++-----
fs/gfs2/super.h | 1 +
11 files changed, 333 insertions(+), 526 deletions(-)
delete mode 100644 fs/gfs2/eaops.c
delete mode 100644 fs/gfs2/eaops.h

diff --git a/fs/gfs2/Makefile b/fs/gfs2/Makefile
index 3da2f1f..2a4fd89 100644
--- a/fs/gfs2/Makefile
+++ b/fs/gfs2/Makefile
@@ -1,6 +1,6 @@
EXTRA_CFLAGS := -I$(src)
obj-$(CONFIG_GFS2_FS) += gfs2.o
-gfs2-y := acl.o bmap.o dir.o eaops.o eattr.o glock.o \
+gfs2-y := acl.o bmap.o dir.o eattr.o glock.o \
glops.o inode.o log.o lops.o main.o meta_io.o \
aops.o dentry.o export.o file.o \
ops_fstype.o ops_inode.o quota.o \
diff --git a/fs/gfs2/acl.c b/fs/gfs2/acl.c
index fa881bd..f6777f1 100644
--- a/fs/gfs2/acl.c
+++ b/fs/gfs2/acl.c
@@ -19,7 +19,6 @@
#include "gfs2.h"
#include "incore.h"
#include "acl.h"
-#include "eaops.h"
#include "eattr.h"
#include "glock.h"
#include "inode.h"
@@ -31,8 +30,7 @@
#define ACL_DEFAULT 0

int gfs2_acl_validate_set(struct gfs2_inode *ip, int access,
- struct gfs2_ea_request *er,
- int *remove, mode_t *mode)
+ struct gfs2_ea_request *er, int *remove, mode_t *mode)
{
struct posix_acl *acl;
int error;
@@ -83,30 +81,20 @@ int gfs2_acl_validate_remove(struct gfs2_inode *ip, int access)
return 0;
}

-static int acl_get(struct gfs2_inode *ip, int access, struct posix_acl **acl,
- struct gfs2_ea_location *el, char **data, unsigned int *len)
+static int acl_get(struct gfs2_inode *ip, const char *name,
+ struct posix_acl **acl, struct gfs2_ea_location *el,
+ char **datap, unsigned int *lenp)
{
- struct gfs2_ea_request er;
- struct gfs2_ea_location el_this;
+ char *data;
+ unsigned int len;
int error;

+ el->el_bh = NULL;
+
if (!ip->i_eattr)
return 0;

- memset(&er, 0, sizeof(struct gfs2_ea_request));
- if (access) {
- er.er_name = GFS2_POSIX_ACL_ACCESS;
- er.er_name_len = GFS2_POSIX_ACL_ACCESS_LEN;
- } else {
- er.er_name = GFS2_POSIX_ACL_DEFAULT;
- er.er_name_len = GFS2_POSIX_ACL_DEFAULT_LEN;
- }
- er.er_type = GFS2_EATYPE_SYS;
-
- if (!el)
- el = &el_this;
-
- error = gfs2_ea_find(ip, &er, el);
+ error = gfs2_ea_find(ip, GFS2_EATYPE_SYS, name, el);
if (error)
return error;
if (!el->el_ea)
@@ -114,32 +102,31 @@ static int acl_get(struct gfs2_inode *ip, int access, struct posix_acl **acl,
if (!GFS2_EA_DATA_LEN(el->el_ea))
goto out;

- er.er_data_len = GFS2_EA_DATA_LEN(el->el_ea);
- er.er_data = kmalloc(er.er_data_len, GFP_NOFS);
+ len = GFS2_EA_DATA_LEN(el->el_ea);
+ data = kmalloc(len, GFP_NOFS);
error = -ENOMEM;
- if (!er.er_data)
+ if (!data)
goto out;

- error = gfs2_ea_get_copy(ip, el, er.er_data);
- if (error)
+ error = gfs2_ea_get_copy(ip, el, data, len);
+ if (error < 0)
goto out_kfree;
+ error = 0;

if (acl) {
- *acl = posix_acl_from_xattr(er.er_data, er.er_data_len);
+ *acl = posix_acl_from_xattr(data, len);
if (IS_ERR(*acl))
error = PTR_ERR(*acl);
}

out_kfree:
- if (error || !data)
- kfree(er.er_data);
- else {
- *data = er.er_data;
- *len = er.er_data_len;
+ if (error || !datap) {
+ kfree(data);
+ } else {
+ *datap = data;
+ *lenp = len;
}
out:
- if (error || el == &el_this)
- brelse(el->el_bh);
return error;
}

@@ -153,10 +140,12 @@ out:

int gfs2_check_acl(struct inode *inode, int mask)
{
+ struct gfs2_ea_location el;
struct posix_acl *acl = NULL;
int error;

- error = acl_get(GFS2_I(inode), ACL_ACCESS, &acl, NULL, NULL, NULL);
+ error = acl_get(GFS2_I(inode), GFS2_POSIX_ACL_ACCESS, &acl, &el, NULL, NULL);
+ brelse(el.el_bh);
if (error)
return error;

@@ -196,10 +185,12 @@ static int munge_mode(struct gfs2_inode *ip, mode_t mode)

int gfs2_acl_create(struct gfs2_inode *dip, struct gfs2_inode *ip)
{
+ struct gfs2_ea_location el;
struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode);
struct posix_acl *acl = NULL, *clone;
- struct gfs2_ea_request er;
mode_t mode = ip->i_inode.i_mode;
+ char *data = NULL;
+ unsigned int len;
int error;

if (!sdp->sd_args.ar_posix_acl)
@@ -207,11 +198,8 @@ int gfs2_acl_create(struct gfs2_inode *dip, struct gfs2_inode *ip)
if (S_ISLNK(ip->i_inode.i_mode))
return 0;

- memset(&er, 0, sizeof(struct gfs2_ea_request));
- er.er_type = GFS2_EATYPE_SYS;
-
- error = acl_get(dip, ACL_DEFAULT, &acl, NULL,
- &er.er_data, &er.er_data_len);
+ error = acl_get(dip, GFS2_POSIX_ACL_DEFAULT, &acl, &el, &data, &len);
+ brelse(el.el_bh);
if (error)
return error;
if (!acl) {
@@ -229,9 +217,8 @@ int gfs2_acl_create(struct gfs2_inode *dip, struct gfs2_inode *ip)
acl = clone;

if (S_ISDIR(ip->i_inode.i_mode)) {
- er.er_name = GFS2_POSIX_ACL_DEFAULT;
- er.er_name_len = GFS2_POSIX_ACL_DEFAULT_LEN;
- error = gfs2_system_eaops.eo_set(ip, &er);
+ error = gfs2_xattr_set(&ip->i_inode, GFS2_EATYPE_SYS,
+ GFS2_POSIX_ACL_DEFAULT, data, len, 0);
if (error)
goto out;
}
@@ -239,21 +226,19 @@ int gfs2_acl_create(struct gfs2_inode *dip, struct gfs2_inode *ip)
error = posix_acl_create_masq(acl, &mode);
if (error < 0)
goto out;
- if (error > 0) {
- er.er_name = GFS2_POSIX_ACL_ACCESS;
- er.er_name_len = GFS2_POSIX_ACL_ACCESS_LEN;
- posix_acl_to_xattr(acl, er.er_data, er.er_data_len);
- er.er_mode = mode;
- er.er_flags = GFS2_ERF_MODE;
- error = gfs2_system_eaops.eo_set(ip, &er);
- if (error)
- goto out;
- } else
- munge_mode(ip, mode);
+ if (error == 0)
+ goto munge;

+ posix_acl_to_xattr(acl, data, len);
+ error = gfs2_xattr_set(&ip->i_inode, GFS2_EATYPE_SYS,
+ GFS2_POSIX_ACL_ACCESS, data, len, 0);
+ if (error)
+ goto out;
+munge:
+ error = munge_mode(ip, mode);
out:
posix_acl_release(acl);
- kfree(er.er_data);
+ kfree(data);
return error;
}

@@ -265,9 +250,9 @@ int gfs2_acl_chmod(struct gfs2_inode *ip, struct iattr *attr)
unsigned int len;
int error;

- error = acl_get(ip, ACL_ACCESS, &acl, &el, &data, &len);
+ error = acl_get(ip, GFS2_POSIX_ACL_ACCESS, &acl, &el, &data, &len);
if (error)
- return error;
+ goto out_brelse;
if (!acl)
return gfs2_setattr_simple(ip, attr);

@@ -286,8 +271,9 @@ int gfs2_acl_chmod(struct gfs2_inode *ip, struct iattr *attr)

out:
posix_acl_release(acl);
- brelse(el.el_bh);
kfree(data);
+out_brelse:
+ brelse(el.el_bh);
return error;
}

diff --git a/fs/gfs2/eaops.c b/fs/gfs2/eaops.c
deleted file mode 100644
index dee9b03..0000000
--- a/fs/gfs2/eaops.c
+++ /dev/null
@@ -1,157 +0,0 @@
-/*
- * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved.
- * Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved.
- *
- * This copyrighted material is made available to anyone wishing to use,
- * modify, copy, or redistribute it subject to the terms and conditions
- * of the GNU General Public License version 2.
- */
-
-#include <linux/slab.h>
-#include <linux/spinlock.h>
-#include <linux/completion.h>
-#include <linux/buffer_head.h>
-#include <linux/capability.h>
-#include <linux/xattr.h>
-#include <linux/gfs2_ondisk.h>
-#include <asm/uaccess.h>
-
-#include "gfs2.h"
-#include "incore.h"
-#include "acl.h"
-#include "eaops.h"
-#include "eattr.h"
-#include "util.h"
-
-/**
- * gfs2_ea_name2type - get the type of the ea, and truncate type from the name
- * @namep: ea name, possibly with type appended
- *
- * Returns: GFS2_EATYPE_XXX
- */
-
-unsigned int gfs2_ea_name2type(const char *name, const char **truncated_name)
-{
- unsigned int type;
-
- if (strncmp(name, "system.", 7) == 0) {
- type = GFS2_EATYPE_SYS;
- if (truncated_name)
- *truncated_name = name + sizeof("system.") - 1;
- } else if (strncmp(name, "user.", 5) == 0) {
- type = GFS2_EATYPE_USR;
- if (truncated_name)
- *truncated_name = name + sizeof("user.") - 1;
- } else if (strncmp(name, "security.", 9) == 0) {
- type = GFS2_EATYPE_SECURITY;
- if (truncated_name)
- *truncated_name = name + sizeof("security.") - 1;
- } else {
- type = GFS2_EATYPE_UNUSED;
- if (truncated_name)
- *truncated_name = NULL;
- }
-
- return type;
-}
-
-static int system_eo_get(struct gfs2_inode *ip, struct gfs2_ea_request *er)
-{
- if (!GFS2_ACL_IS_ACCESS(er->er_name, er->er_name_len) &&
- !GFS2_ACL_IS_DEFAULT(er->er_name, er->er_name_len) &&
- !capable(CAP_SYS_ADMIN))
- return -EPERM;
-
- if (GFS2_SB(&ip->i_inode)->sd_args.ar_posix_acl == 0 &&
- (GFS2_ACL_IS_ACCESS(er->er_name, er->er_name_len) ||
- GFS2_ACL_IS_DEFAULT(er->er_name, er->er_name_len)))
- return -EOPNOTSUPP;
-
- return gfs2_ea_get_i(ip, er);
-}
-
-static int system_eo_set(struct gfs2_inode *ip, struct gfs2_ea_request *er)
-{
- int remove = 0;
- int error;
-
- if (GFS2_ACL_IS_ACCESS(er->er_name, er->er_name_len)) {
- if (!(er->er_flags & GFS2_ERF_MODE)) {
- er->er_mode = ip->i_inode.i_mode;
- er->er_flags |= GFS2_ERF_MODE;
- }
- error = gfs2_acl_validate_set(ip, 1, er,
- &remove, &er->er_mode);
- if (error)
- return error;
- error = gfs2_ea_set_i(ip, er);
- if (error)
- return error;
- if (remove)
- gfs2_ea_remove_i(ip, er);
- return 0;
-
- } else if (GFS2_ACL_IS_DEFAULT(er->er_name, er->er_name_len)) {
- error = gfs2_acl_validate_set(ip, 0, er,
- &remove, NULL);
- if (error)
- return error;
- if (!remove)
- error = gfs2_ea_set_i(ip, er);
- else {
- error = gfs2_ea_remove_i(ip, er);
- if (error == -ENODATA)
- error = 0;
- }
- return error;
- }
-
- return -EPERM;
-}
-
-static int system_eo_remove(struct gfs2_inode *ip, struct gfs2_ea_request *er)
-{
- if (GFS2_ACL_IS_ACCESS(er->er_name, er->er_name_len)) {
- int error = gfs2_acl_validate_remove(ip, 1);
- if (error)
- return error;
-
- } else if (GFS2_ACL_IS_DEFAULT(er->er_name, er->er_name_len)) {
- int error = gfs2_acl_validate_remove(ip, 0);
- if (error)
- return error;
-
- } else
- return -EPERM;
-
- return gfs2_ea_remove_i(ip, er);
-}
-
-static const struct gfs2_eattr_operations gfs2_user_eaops = {
- .eo_get = gfs2_ea_get_i,
- .eo_set = gfs2_ea_set_i,
- .eo_remove = gfs2_ea_remove_i,
- .eo_name = "user",
-};
-
-const struct gfs2_eattr_operations gfs2_system_eaops = {
- .eo_get = system_eo_get,
- .eo_set = system_eo_set,
- .eo_remove = system_eo_remove,
- .eo_name = "system",
-};
-
-static const struct gfs2_eattr_operations gfs2_security_eaops = {
- .eo_get = gfs2_ea_get_i,
- .eo_set = gfs2_ea_set_i,
- .eo_remove = gfs2_ea_remove_i,
- .eo_name = "security",
-};
-
-const struct gfs2_eattr_operations *gfs2_ea_ops[] = {
- NULL,
- &gfs2_user_eaops,
- &gfs2_system_eaops,
- &gfs2_security_eaops,
-};
-
diff --git a/fs/gfs2/eaops.h b/fs/gfs2/eaops.h
deleted file mode 100644
index da2f7fb..0000000
--- a/fs/gfs2/eaops.h
+++ /dev/null
@@ -1,30 +0,0 @@
-/*
- * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved.
- * Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved.
- *
- * This copyrighted material is made available to anyone wishing to use,
- * modify, copy, or redistribute it subject to the terms and conditions
- * of the GNU General Public License version 2.
- */
-
-#ifndef __EAOPS_DOT_H__
-#define __EAOPS_DOT_H__
-
-struct gfs2_ea_request;
-struct gfs2_inode;
-
-struct gfs2_eattr_operations {
- int (*eo_get) (struct gfs2_inode *ip, struct gfs2_ea_request *er);
- int (*eo_set) (struct gfs2_inode *ip, struct gfs2_ea_request *er);
- int (*eo_remove) (struct gfs2_inode *ip, struct gfs2_ea_request *er);
- char *eo_name;
-};
-
-unsigned int gfs2_ea_name2type(const char *name, const char **truncated_name);
-
-extern const struct gfs2_eattr_operations gfs2_system_eaops;
-
-extern const struct gfs2_eattr_operations *gfs2_ea_ops[];
-
-#endif /* __EAOPS_DOT_H__ */
-
diff --git a/fs/gfs2/eattr.c b/fs/gfs2/eattr.c
index 07ea952..d9a9e26 100644
--- a/fs/gfs2/eattr.c
+++ b/fs/gfs2/eattr.c
@@ -18,7 +18,6 @@
#include "gfs2.h"
#include "incore.h"
#include "acl.h"
-#include "eaops.h"
#include "eattr.h"
#include "glock.h"
#include "inode.h"
@@ -38,26 +37,32 @@
* Returns: 1 if the EA should be stuffed
*/

-static int ea_calc_size(struct gfs2_sbd *sdp, struct gfs2_ea_request *er,
+static int ea_calc_size(struct gfs2_sbd *sdp, unsigned int nsize, size_t dsize,
unsigned int *size)
{
- *size = GFS2_EAREQ_SIZE_STUFFED(er);
- if (*size <= sdp->sd_jbsize)
+ unsigned int jbsize = sdp->sd_jbsize;
+
+ /* Stuffed */
+ *size = ALIGN(sizeof(struct gfs2_ea_header) + nsize + dsize, 8);
+
+ if (*size <= jbsize)
return 1;

- *size = GFS2_EAREQ_SIZE_UNSTUFFED(sdp, er);
+ /* Unstuffed */
+ *size = ALIGN(sizeof(struct gfs2_ea_header) + nsize +
+ (sizeof(__be64) * DIV_ROUND_UP(dsize, jbsize)), 8);

return 0;
}

-static int ea_check_size(struct gfs2_sbd *sdp, struct gfs2_ea_request *er)
+static int ea_check_size(struct gfs2_sbd *sdp, unsigned int nsize, size_t dsize)
{
unsigned int size;

- if (er->er_data_len > GFS2_EA_MAX_DATA_LEN)
+ if (dsize > GFS2_EA_MAX_DATA_LEN)
return -ERANGE;

- ea_calc_size(sdp, er, &size);
+ ea_calc_size(sdp, nsize, dsize, &size);

/* This can only happen with 512 byte blocks */
if (size > sdp->sd_jbsize)
@@ -151,7 +156,9 @@ out:
}

struct ea_find {
- struct gfs2_ea_request *ef_er;
+ int type;
+ const char *name;
+ size_t namel;
struct gfs2_ea_location *ef_el;
};

@@ -160,14 +167,13 @@ static int ea_find_i(struct gfs2_inode *ip, struct buffer_head *bh,
void *private)
{
struct ea_find *ef = private;
- struct gfs2_ea_request *er = ef->ef_er;

if (ea->ea_type == GFS2_EATYPE_UNUSED)
return 0;

- if (ea->ea_type == er->er_type) {
- if (ea->ea_name_len == er->er_name_len &&
- !memcmp(GFS2_EA2NAME(ea), er->er_name, ea->ea_name_len)) {
+ if (ea->ea_type == ef->type) {
+ if (ea->ea_name_len == ef->namel &&
+ !memcmp(GFS2_EA2NAME(ea), ef->name, ea->ea_name_len)) {
struct gfs2_ea_location *el = ef->ef_el;
get_bh(bh);
el->el_bh = bh;
@@ -180,13 +186,15 @@ static int ea_find_i(struct gfs2_inode *ip, struct buffer_head *bh,
return 0;
}

-int gfs2_ea_find(struct gfs2_inode *ip, struct gfs2_ea_request *er,
+int gfs2_ea_find(struct gfs2_inode *ip, int type, const char *name,
struct gfs2_ea_location *el)
{
struct ea_find ef;
int error;

- ef.ef_er = er;
+ ef.type = type;
+ ef.name = name;
+ ef.namel = strlen(name);
ef.ef_el = el;

memset(el, 0, sizeof(struct gfs2_ea_location));
@@ -344,6 +352,20 @@ struct ea_list {
unsigned int ei_size;
};

+static inline unsigned int gfs2_ea_strlen(struct gfs2_ea_header *ea)
+{
+ switch (ea->ea_type) {
+ case GFS2_EATYPE_USR:
+ return 5 + ea->ea_name_len + 1;
+ case GFS2_EATYPE_SYS:
+ return 7 + ea->ea_name_len + 1;
+ case GFS2_EATYPE_SECURITY:
+ return 9 + ea->ea_name_len + 1;
+ default:
+ return 0;
+ }
+}
+
static int ea_list_i(struct gfs2_inode *ip, struct buffer_head *bh,
struct gfs2_ea_header *ea, struct gfs2_ea_header *prev,
void *private)
@@ -392,21 +414,25 @@ static int ea_list_i(struct gfs2_inode *ip, struct buffer_head *bh,
}

/**
- * gfs2_ea_list -
- * @ip:
- * @er:
+ * gfs2_listxattr - List gfs2 extended attributes
+ * @dentry: The dentry whose inode we are interested in
+ * @buffer: The buffer to write the results
+ * @size: The size of the buffer
*
* Returns: actual size of data on success, -errno on error
*/

-int gfs2_ea_list(struct gfs2_inode *ip, struct gfs2_ea_request *er)
+ssize_t gfs2_listxattr(struct dentry *dentry, char *buffer, size_t size)
{
+ struct gfs2_inode *ip = GFS2_I(dentry->d_inode);
+ struct gfs2_ea_request er;
struct gfs2_holder i_gh;
int error;

- if (!er->er_data || !er->er_data_len) {
- er->er_data = NULL;
- er->er_data_len = 0;
+ memset(&er, 0, sizeof(struct gfs2_ea_request));
+ if (size) {
+ er.er_data = buffer;
+ er.er_data_len = size;
}

error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &i_gh);
@@ -414,7 +440,7 @@ int gfs2_ea_list(struct gfs2_inode *ip, struct gfs2_ea_request *er)
return error;

if (ip->i_eattr) {
- struct ea_list ei = { .ei_er = er, .ei_size = 0 };
+ struct ea_list ei = { .ei_er = &er, .ei_size = 0 };

error = ea_foreach(ip, ea_list_i, &ei);
if (!error)
@@ -491,84 +517,61 @@ out:
}

int gfs2_ea_get_copy(struct gfs2_inode *ip, struct gfs2_ea_location *el,
- char *data)
+ char *data, size_t size)
{
+ int ret;
+ size_t len = GFS2_EA_DATA_LEN(el->el_ea);
+ if (len > size)
+ return -ERANGE;
+
if (GFS2_EA_IS_STUFFED(el->el_ea)) {
- memcpy(data, GFS2_EA2DATA(el->el_ea), GFS2_EA_DATA_LEN(el->el_ea));
- return 0;
- } else
- return ea_get_unstuffed(ip, el->el_ea, data);
+ memcpy(data, GFS2_EA2DATA(el->el_ea), len);
+ return len;
+ }
+ ret = ea_get_unstuffed(ip, el->el_ea, data);
+ if (ret < 0)
+ return ret;
+ return len;
}

/**
- * gfs2_ea_get_i -
- * @ip: The GFS2 inode
- * @er: The request structure
+ * gfs2_xattr_get - Get a GFS2 extended attribute
+ * @inode: The inode
+ * @type: The type of extended attribute
+ * @name: The name of the extended attribute
+ * @buffer: The buffer to write the result into
+ * @size: The size of the buffer
*
* Returns: actual size of data on success, -errno on error
*/

-int gfs2_ea_get_i(struct gfs2_inode *ip, struct gfs2_ea_request *er)
+int gfs2_xattr_get(struct inode *inode, int type, const char *name,
+ void *buffer, size_t size)
{
+ struct gfs2_inode *ip = GFS2_I(inode);
struct gfs2_ea_location el;
int error;

if (!ip->i_eattr)
return -ENODATA;
+ if (strlen(name) > GFS2_EA_MAX_NAME_LEN)
+ return -EINVAL;

- error = gfs2_ea_find(ip, er, &el);
+ error = gfs2_ea_find(ip, type, name, &el);
if (error)
return error;
if (!el.el_ea)
return -ENODATA;
-
- if (er->er_data_len) {
- if (GFS2_EA_DATA_LEN(el.el_ea) > er->er_data_len)
- error = -ERANGE;
- else
- error = gfs2_ea_get_copy(ip, &el, er->er_data);
- }
- if (!error)
+ if (size)
+ error = gfs2_ea_get_copy(ip, &el, buffer, size);
+ else
error = GFS2_EA_DATA_LEN(el.el_ea);
-
brelse(el.el_bh);

return error;
}

/**
- * gfs2_ea_get -
- * @ip: The GFS2 inode
- * @er: The request structure
- *
- * Returns: actual size of data on success, -errno on error
- */
-
-int gfs2_ea_get(struct gfs2_inode *ip, struct gfs2_ea_request *er)
-{
- struct gfs2_holder i_gh;
- int error;
-
- if (!er->er_name_len ||
- er->er_name_len > GFS2_EA_MAX_NAME_LEN)
- return -EINVAL;
- if (!er->er_data || !er->er_data_len) {
- er->er_data = NULL;
- er->er_data_len = 0;
- }
-
- error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &i_gh);
- if (error)
- return error;
-
- error = gfs2_ea_ops[er->er_type]->eo_get(ip, er);
-
- gfs2_glock_dq_uninit(&i_gh);
-
- return error;
-}
-
-/**
* ea_alloc_blk - allocates a new block for extended attributes.
* @ip: A pointer to the inode that's getting extended attributes
* @bhp: Pointer to pointer to a struct buffer_head
@@ -713,12 +716,6 @@ static int ea_alloc_skeleton(struct gfs2_inode *ip, struct gfs2_ea_request *er,

error = gfs2_meta_inode_buffer(ip, &dibh);
if (!error) {
- if (er->er_flags & GFS2_ERF_MODE) {
- gfs2_assert_withdraw(GFS2_SB(&ip->i_inode),
- (ip->i_inode.i_mode & S_IFMT) ==
- (er->er_mode & S_IFMT));
- ip->i_inode.i_mode = er->er_mode;
- }
ip->i_inode.i_ctime = CURRENT_TIME;
gfs2_trans_add_bh(ip->i_gl, dibh, 1);
gfs2_dinode_out(ip, dibh->b_data);
@@ -762,15 +759,23 @@ static int ea_init_i(struct gfs2_inode *ip, struct gfs2_ea_request *er,
* Returns: errno
*/

-static int ea_init(struct gfs2_inode *ip, struct gfs2_ea_request *er)
+static int ea_init(struct gfs2_inode *ip, int type, const char *name,
+ const void *data, size_t size)
{
+ struct gfs2_ea_request er;
unsigned int jbsize = GFS2_SB(&ip->i_inode)->sd_jbsize;
unsigned int blks = 1;

- if (GFS2_EAREQ_SIZE_STUFFED(er) > jbsize)
- blks += DIV_ROUND_UP(er->er_data_len, jbsize);
+ er.er_type = type;
+ er.er_name = name;
+ er.er_name_len = strlen(name);
+ er.er_data = (void *)data;
+ er.er_data_len = size;
+
+ if (GFS2_EAREQ_SIZE_STUFFED(&er) > jbsize)
+ blks += DIV_ROUND_UP(er.er_data_len, jbsize);

- return ea_alloc_skeleton(ip, er, blks, ea_init_i, NULL);
+ return ea_alloc_skeleton(ip, &er, blks, ea_init_i, NULL);
}

static struct gfs2_ea_header *ea_split_ea(struct gfs2_ea_header *ea)
@@ -848,12 +853,6 @@ static int ea_set_simple_noalloc(struct gfs2_inode *ip, struct buffer_head *bh,
error = gfs2_meta_inode_buffer(ip, &dibh);
if (error)
goto out;
-
- if (er->er_flags & GFS2_ERF_MODE) {
- gfs2_assert_withdraw(GFS2_SB(&ip->i_inode),
- (ip->i_inode.i_mode & S_IFMT) == (er->er_mode & S_IFMT));
- ip->i_inode.i_mode = er->er_mode;
- }
ip->i_inode.i_ctime = CURRENT_TIME;
gfs2_trans_add_bh(ip->i_gl, dibh, 1);
gfs2_dinode_out(ip, dibh->b_data);
@@ -894,7 +893,8 @@ static int ea_set_simple(struct gfs2_inode *ip, struct buffer_head *bh,
int stuffed;
int error;

- stuffed = ea_calc_size(GFS2_SB(&ip->i_inode), es->es_er, &size);
+ stuffed = ea_calc_size(GFS2_SB(&ip->i_inode), es->es_er->er_name_len,
+ es->es_er->er_data_len, &size);

if (ea->ea_type == GFS2_EATYPE_UNUSED) {
if (GFS2_EA_REC_LEN(ea) < size)
@@ -1005,15 +1005,22 @@ out:
return error;
}

-static int ea_set_i(struct gfs2_inode *ip, struct gfs2_ea_request *er,
- struct gfs2_ea_location *el)
+static int ea_set_i(struct gfs2_inode *ip, int type, const char *name,
+ const void *value, size_t size, struct gfs2_ea_location *el)
{
+ struct gfs2_ea_request er;
struct ea_set es;
unsigned int blks = 2;
int error;

+ er.er_type = type;
+ er.er_name = name;
+ er.er_data = (void *)value;
+ er.er_name_len = strlen(name);
+ er.er_data_len = size;
+
memset(&es, 0, sizeof(struct ea_set));
- es.es_er = er;
+ es.es_er = &er;
es.es_el = el;

error = ea_foreach(ip, ea_set_simple, &es);
@@ -1024,10 +1031,10 @@ static int ea_set_i(struct gfs2_inode *ip, struct gfs2_ea_request *er,

if (!(ip->i_diskflags & GFS2_DIF_EA_INDIRECT))
blks++;
- if (GFS2_EAREQ_SIZE_STUFFED(er) > GFS2_SB(&ip->i_inode)->sd_jbsize)
- blks += DIV_ROUND_UP(er->er_data_len, GFS2_SB(&ip->i_inode)->sd_jbsize);
+ if (GFS2_EAREQ_SIZE_STUFFED(&er) > GFS2_SB(&ip->i_inode)->sd_jbsize)
+ blks += DIV_ROUND_UP(er.er_data_len, GFS2_SB(&ip->i_inode)->sd_jbsize);

- return ea_alloc_skeleton(ip, er, blks, ea_set_block, el);
+ return ea_alloc_skeleton(ip, &er, blks, ea_set_block, el);
}

static int ea_set_remove_unstuffed(struct gfs2_inode *ip,
@@ -1042,74 +1049,6 @@ static int ea_set_remove_unstuffed(struct gfs2_inode *ip,
return ea_remove_unstuffed(ip, el->el_bh, el->el_ea, el->el_prev,0);
}

-int gfs2_ea_set_i(struct gfs2_inode *ip, struct gfs2_ea_request *er)
-{
- struct gfs2_ea_location el;
- int error;
-
- if (!ip->i_eattr) {
- if (er->er_flags & XATTR_REPLACE)
- return -ENODATA;
- return ea_init(ip, er);
- }
-
- error = gfs2_ea_find(ip, er, &el);
- if (error)
- return error;
-
- if (el.el_ea) {
- if (ip->i_diskflags & GFS2_DIF_APPENDONLY) {
- brelse(el.el_bh);
- return -EPERM;
- }
-
- error = -EEXIST;
- if (!(er->er_flags & XATTR_CREATE)) {
- int unstuffed = !GFS2_EA_IS_STUFFED(el.el_ea);
- error = ea_set_i(ip, er, &el);
- if (!error && unstuffed)
- ea_set_remove_unstuffed(ip, &el);
- }
-
- brelse(el.el_bh);
- } else {
- error = -ENODATA;
- if (!(er->er_flags & XATTR_REPLACE))
- error = ea_set_i(ip, er, NULL);
- }
-
- return error;
-}
-
-int gfs2_ea_set(struct gfs2_inode *ip, struct gfs2_ea_request *er)
-{
- struct gfs2_holder i_gh;
- int error;
-
- if (!er->er_name_len || er->er_name_len > GFS2_EA_MAX_NAME_LEN)
- return -EINVAL;
- if (!er->er_data || !er->er_data_len) {
- er->er_data = NULL;
- er->er_data_len = 0;
- }
- error = ea_check_size(GFS2_SB(&ip->i_inode), er);
- if (error)
- return error;
-
- error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &i_gh);
- if (error)
- return error;
-
- if (IS_IMMUTABLE(&ip->i_inode))
- error = -EPERM;
- else
- error = gfs2_ea_ops[er->er_type]->eo_set(ip, er);
-
- gfs2_glock_dq_uninit(&i_gh);
-
- return error;
-}
-
static int ea_remove_stuffed(struct gfs2_inode *ip, struct gfs2_ea_location *el)
{
struct gfs2_ea_header *ea = el->el_ea;
@@ -1131,8 +1070,9 @@ static int ea_remove_stuffed(struct gfs2_inode *ip, struct gfs2_ea_location *el)

if (GFS2_EA_IS_LAST(ea))
prev->ea_flags |= GFS2_EAFLAG_LAST;
- } else
+ } else {
ea->ea_type = GFS2_EATYPE_UNUSED;
+ }

error = gfs2_meta_inode_buffer(ip, &dibh);
if (!error) {
@@ -1147,15 +1087,29 @@ static int ea_remove_stuffed(struct gfs2_inode *ip, struct gfs2_ea_location *el)
return error;
}

-int gfs2_ea_remove_i(struct gfs2_inode *ip, struct gfs2_ea_request *er)
+/**
+ * gfs2_xattr_remove - Remove a GFS2 extended attribute
+ * @inode: The inode
+ * @type: The type of the extended attribute
+ * @name: The name of the extended attribute
+ *
+ * This is not called directly by the VFS since we use the (common)
+ * scheme of making a "set with NULL data" mean a remove request. Note
+ * that this is different from a set with zero length data.
+ *
+ * Returns: 0, or errno on failure
+ */
+
+static int gfs2_xattr_remove(struct inode *inode, int type, const char *name)
{
+ struct gfs2_inode *ip = GFS2_I(inode);
struct gfs2_ea_location el;
int error;

if (!ip->i_eattr)
return -ENODATA;

- error = gfs2_ea_find(ip, er, &el);
+ error = gfs2_ea_find(ip, type, name, &el);
if (error)
return error;
if (!el.el_ea)
@@ -1164,8 +1118,7 @@ int gfs2_ea_remove_i(struct gfs2_inode *ip, struct gfs2_ea_request *er)
if (GFS2_EA_IS_STUFFED(el.el_ea))
error = ea_remove_stuffed(ip, &el);
else
- error = ea_remove_unstuffed(ip, el.el_bh, el.el_ea, el.el_prev,
- 0);
+ error = ea_remove_unstuffed(ip, el.el_bh, el.el_ea, el.el_prev, 0);

brelse(el.el_bh);

@@ -1173,31 +1126,70 @@ int gfs2_ea_remove_i(struct gfs2_inode *ip, struct gfs2_ea_request *er)
}

/**
- * gfs2_ea_remove - sets (or creates or replaces) an extended attribute
- * @ip: pointer to the inode of the target file
- * @er: request information
+ * gfs2_xattr_set - Set (or remove) a GFS2 extended attribute
+ * @inode: The inode
+ * @type: The type of the extended attribute
+ * @name: The name of the extended attribute
+ * @value: The value of the extended attribute (NULL for remove)
+ * @size: The size of the @value argument
+ * @flags: Create or Replace
*
- * Returns: errno
+ * See gfs2_xattr_remove() for details of the removal of xattrs.
+ *
+ * Returns: 0 or errno on failure
*/

-int gfs2_ea_remove(struct gfs2_inode *ip, struct gfs2_ea_request *er)
+int gfs2_xattr_set(struct inode *inode, int type, const char *name,
+ const void *value, size_t size, int flags)
{
- struct gfs2_holder i_gh;
+ struct gfs2_sbd *sdp = GFS2_SB(inode);
+ struct gfs2_inode *ip = GFS2_I(inode);
+ struct gfs2_ea_location el;
+ unsigned int namel = strlen(name);
int error;

- if (!er->er_name_len || er->er_name_len > GFS2_EA_MAX_NAME_LEN)
- return -EINVAL;
+ if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
+ return -EPERM;
+ if (namel > GFS2_EA_MAX_NAME_LEN)
+ return -ERANGE;

- error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &i_gh);
+ if (value == NULL)
+ return gfs2_xattr_remove(inode, type, name);
+
+ if (ea_check_size(sdp, namel, size))
+ return -ERANGE;
+
+ if (!ip->i_eattr) {
+ if (flags & XATTR_REPLACE)
+ return -ENODATA;
+ return ea_init(ip, type, name, value, size);
+ }
+
+ error = gfs2_ea_find(ip, type, name, &el);
if (error)
return error;

- if (IS_IMMUTABLE(&ip->i_inode) || IS_APPEND(&ip->i_inode))
- error = -EPERM;
- else
- error = gfs2_ea_ops[er->er_type]->eo_remove(ip, er);
+ if (el.el_ea) {
+ if (ip->i_diskflags & GFS2_DIF_APPENDONLY) {
+ brelse(el.el_bh);
+ return -EPERM;
+ }

- gfs2_glock_dq_uninit(&i_gh);
+ error = -EEXIST;
+ if (!(flags & XATTR_CREATE)) {
+ int unstuffed = !GFS2_EA_IS_STUFFED(el.el_ea);
+ error = ea_set_i(ip, type, name, value, size, &el);
+ if (!error && unstuffed)
+ ea_set_remove_unstuffed(ip, &el);
+ }
+
+ brelse(el.el_bh);
+ return error;
+ }
+
+ error = -ENODATA;
+ if (!(flags & XATTR_REPLACE))
+ error = ea_set_i(ip, type, name, value, size, NULL);

return error;
}
@@ -1503,3 +1495,64 @@ out_alloc:
return error;
}

+static int gfs2_xattr_user_get(struct inode *inode, const char *name,
+ void *buffer, size_t size)
+{
+ return gfs2_xattr_get(inode, GFS2_EATYPE_USR, name, buffer, size);
+}
+
+static int gfs2_xattr_user_set(struct inode *inode, const char *name,
+ const void *value, size_t size, int flags)
+{
+ return gfs2_xattr_set(inode, GFS2_EATYPE_USR, name, value, size, flags);
+}
+
+static int gfs2_xattr_system_get(struct inode *inode, const char *name,
+ void *buffer, size_t size)
+{
+ return gfs2_xattr_get(inode, GFS2_EATYPE_SYS, name, buffer, size);
+}
+
+static int gfs2_xattr_system_set(struct inode *inode, const char *name,
+ const void *value, size_t size, int flags)
+{
+ return gfs2_xattr_set(inode, GFS2_EATYPE_SYS, name, value, size, flags);
+}
+
+static int gfs2_xattr_security_get(struct inode *inode, const char *name,
+ void *buffer, size_t size)
+{
+ return gfs2_xattr_get(inode, GFS2_EATYPE_SECURITY, name, buffer, size);
+}
+
+static int gfs2_xattr_security_set(struct inode *inode, const char *name,
+ const void *value, size_t size, int flags)
+{
+ return gfs2_xattr_set(inode, GFS2_EATYPE_SECURITY, name, value, size, flags);
+}
+
+static struct xattr_handler gfs2_xattr_user_handler = {
+ .prefix = XATTR_USER_PREFIX,
+ .get = gfs2_xattr_user_get,
+ .set = gfs2_xattr_user_set,
+};
+
+static struct xattr_handler gfs2_xattr_security_handler = {
+ .prefix = XATTR_SECURITY_PREFIX,
+ .get = gfs2_xattr_security_get,
+ .set = gfs2_xattr_security_set,
+};
+
+static struct xattr_handler gfs2_xattr_system_handler = {
+ .prefix = XATTR_SYSTEM_PREFIX,
+ .get = gfs2_xattr_system_get,
+ .set = gfs2_xattr_system_set,
+};
+
+struct xattr_handler *gfs2_xattr_handlers[] = {
+ &gfs2_xattr_user_handler,
+ &gfs2_xattr_security_handler,
+ &gfs2_xattr_system_handler,
+ NULL,
+};
+
diff --git a/fs/gfs2/eattr.h b/fs/gfs2/eattr.h
index c82dbe0..4040a18 100644
--- a/fs/gfs2/eattr.h
+++ b/fs/gfs2/eattr.h
@@ -27,10 +27,6 @@ ALIGN(sizeof(struct gfs2_ea_header) + (ea)->ea_name_len + \
#define GFS2_EAREQ_SIZE_STUFFED(er) \
ALIGN(sizeof(struct gfs2_ea_header) + (er)->er_name_len + (er)->er_data_len, 8)

-#define GFS2_EAREQ_SIZE_UNSTUFFED(sdp, er) \
-ALIGN(sizeof(struct gfs2_ea_header) + (er)->er_name_len + \
- sizeof(__be64) * DIV_ROUND_UP((er)->er_data_len, (sdp)->sd_jbsize), 8)
-
#define GFS2_EA2NAME(ea) ((char *)((struct gfs2_ea_header *)(ea) + 1))
#define GFS2_EA2DATA(ea) (GFS2_EA2NAME(ea) + (ea)->ea_name_len)

@@ -43,16 +39,12 @@ ALIGN(sizeof(struct gfs2_ea_header) + (er)->er_name_len + \
#define GFS2_EA_BH2FIRST(bh) \
((struct gfs2_ea_header *)((bh)->b_data + sizeof(struct gfs2_meta_header)))

-#define GFS2_ERF_MODE 0x80000000
-
struct gfs2_ea_request {
const char *er_name;
char *er_data;
unsigned int er_name_len;
unsigned int er_data_len;
unsigned int er_type; /* GFS2_EATYPE_... */
- int er_flags;
- mode_t er_mode;
};

struct gfs2_ea_location {
@@ -61,40 +53,20 @@ struct gfs2_ea_location {
struct gfs2_ea_header *el_prev;
};

-int gfs2_ea_get_i(struct gfs2_inode *ip, struct gfs2_ea_request *er);
-int gfs2_ea_set_i(struct gfs2_inode *ip, struct gfs2_ea_request *er);
-int gfs2_ea_remove_i(struct gfs2_inode *ip, struct gfs2_ea_request *er);
-
-int gfs2_ea_list(struct gfs2_inode *ip, struct gfs2_ea_request *er);
-int gfs2_ea_get(struct gfs2_inode *ip, struct gfs2_ea_request *er);
-int gfs2_ea_set(struct gfs2_inode *ip, struct gfs2_ea_request *er);
-int gfs2_ea_remove(struct gfs2_inode *ip, struct gfs2_ea_request *er);
-
-int gfs2_ea_dealloc(struct gfs2_inode *ip);
+extern int gfs2_xattr_get(struct inode *inode, int type, const char *name,
+ void *buffer, size_t size);
+extern int gfs2_xattr_set(struct inode *inode, int type, const char *name,
+ const void *value, size_t size, int flags);
+extern ssize_t gfs2_listxattr(struct dentry *dentry, char *buffer, size_t size);
+extern int gfs2_ea_dealloc(struct gfs2_inode *ip);

/* Exported to acl.c */

-int gfs2_ea_find(struct gfs2_inode *ip,
- struct gfs2_ea_request *er,
- struct gfs2_ea_location *el);
-int gfs2_ea_get_copy(struct gfs2_inode *ip,
- struct gfs2_ea_location *el,
- char *data);
-int gfs2_ea_acl_chmod(struct gfs2_inode *ip, struct gfs2_ea_location *el,
- struct iattr *attr, char *data);
-
-static inline unsigned int gfs2_ea_strlen(struct gfs2_ea_header *ea)
-{
- switch (ea->ea_type) {
- case GFS2_EATYPE_USR:
- return 5 + ea->ea_name_len + 1;
- case GFS2_EATYPE_SYS:
- return 7 + ea->ea_name_len + 1;
- case GFS2_EATYPE_SECURITY:
- return 9 + ea->ea_name_len + 1;
- default:
- return 0;
- }
-}
+extern int gfs2_ea_find(struct gfs2_inode *ip, int type, const char *name,
+ struct gfs2_ea_location *el);
+extern int gfs2_ea_get_copy(struct gfs2_inode *ip, struct gfs2_ea_location *el,
+ char *data, size_t size);
+extern int gfs2_ea_acl_chmod(struct gfs2_inode *ip, struct gfs2_ea_location *el,
+ struct iattr *attr, char *data);

#endif /* __EATTR_DOT_H__ */
diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c
index 73318a3..166f38f 100644
--- a/fs/gfs2/file.c
+++ b/fs/gfs2/file.c
@@ -38,7 +38,6 @@
#include "rgrp.h"
#include "trans.h"
#include "util.h"
-#include "eaops.h"

/**
* gfs2_llseek - seek to a location in a file
diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c
index f9b4fe8..f936d2d 100644
--- a/fs/gfs2/inode.c
+++ b/fs/gfs2/inode.c
@@ -924,7 +924,6 @@ static int gfs2_security_init(struct gfs2_inode *dip, struct gfs2_inode *ip)
size_t len;
void *value;
char *name;
- struct gfs2_ea_request er;

err = security_inode_init_security(&ip->i_inode, &dip->i_inode,
&name, &value, &len);
@@ -935,16 +934,7 @@ static int gfs2_security_init(struct gfs2_inode *dip, struct gfs2_inode *ip)
return err;
}

- memset(&er, 0, sizeof(struct gfs2_ea_request));
-
- er.er_type = GFS2_EATYPE_SECURITY;
- er.er_name = name;
- er.er_data = value;
- er.er_name_len = strlen(name);
- er.er_data_len = len;
-
- err = gfs2_ea_set_i(ip, &er);
-
+ err = gfs2_xattr_set(&ip->i_inode, GFS2_EATYPE_SECURITY, name, value, len, 0);
kfree(value);
kfree(name);

diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index 165518a..1ec69e6 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -1186,6 +1186,7 @@ static int fill_super(struct super_block *sb, void *data, int silent)
sb->s_magic = GFS2_MAGIC;
sb->s_op = &gfs2_super_ops;
sb->s_export_op = &gfs2_export_ops;
+ sb->s_xattr = gfs2_xattr_handlers;
sb->s_time_gran = 1;
sb->s_maxbytes = MAX_LFS_FILESIZE;

diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c
index ce551f8..1ee0c26 100644
--- a/fs/gfs2/ops_inode.c
+++ b/fs/gfs2/ops_inode.c
@@ -26,7 +26,6 @@
#include "acl.h"
#include "bmap.h"
#include "dir.h"
-#include "eaops.h"
#include "eattr.h"
#include "glock.h"
#include "inode.h"
@@ -1302,60 +1301,53 @@ static int gfs2_setxattr(struct dentry *dentry, const char *name,
const void *data, size_t size, int flags)
{
struct inode *inode = dentry->d_inode;
- struct gfs2_ea_request er;
-
- memset(&er, 0, sizeof(struct gfs2_ea_request));
- er.er_type = gfs2_ea_name2type(name, &er.er_name);
- if (er.er_type == GFS2_EATYPE_UNUSED)
- return -EOPNOTSUPP;
- er.er_data = (char *)data;
- er.er_name_len = strlen(er.er_name);
- er.er_data_len = size;
- er.er_flags = flags;
-
- gfs2_assert_warn(GFS2_SB(inode), !(er.er_flags & GFS2_ERF_MODE));
+ struct gfs2_inode *ip = GFS2_I(inode);
+ struct gfs2_holder gh;
+ int ret;

- return gfs2_ea_set(GFS2_I(inode), &er);
+ gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh);
+ ret = gfs2_glock_nq(&gh);
+ if (ret == 0) {
+ ret = generic_setxattr(dentry, name, data, size, flags);
+ gfs2_glock_dq(&gh);
+ }
+ gfs2_holder_uninit(&gh);
+ return ret;
}

static ssize_t gfs2_getxattr(struct dentry *dentry, const char *name,
void *data, size_t size)
{
- struct gfs2_ea_request er;
-
- memset(&er, 0, sizeof(struct gfs2_ea_request));
- er.er_type = gfs2_ea_name2type(name, &er.er_name);
- if (er.er_type == GFS2_EATYPE_UNUSED)
- return -EOPNOTSUPP;
- er.er_data = data;
- er.er_name_len = strlen(er.er_name);
- er.er_data_len = size;
-
- return gfs2_ea_get(GFS2_I(dentry->d_inode), &er);
-}
-
-static ssize_t gfs2_listxattr(struct dentry *dentry, char *buffer, size_t size)
-{
- struct gfs2_ea_request er;
-
- memset(&er, 0, sizeof(struct gfs2_ea_request));
- er.er_data = (size) ? buffer : NULL;
- er.er_data_len = size;
+ struct inode *inode = dentry->d_inode;
+ struct gfs2_inode *ip = GFS2_I(inode);
+ struct gfs2_holder gh;
+ int ret;

- return gfs2_ea_list(GFS2_I(dentry->d_inode), &er);
+ gfs2_holder_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &gh);
+ ret = gfs2_glock_nq(&gh);
+ if (ret == 0) {
+ ret = generic_getxattr(dentry, name, data, size);
+ gfs2_glock_dq(&gh);
+ }
+ gfs2_holder_uninit(&gh);
+ return ret;
}

static int gfs2_removexattr(struct dentry *dentry, const char *name)
{
- struct gfs2_ea_request er;
-
- memset(&er, 0, sizeof(struct gfs2_ea_request));
- er.er_type = gfs2_ea_name2type(name, &er.er_name);
- if (er.er_type == GFS2_EATYPE_UNUSED)
- return -EOPNOTSUPP;
- er.er_name_len = strlen(er.er_name);
+ struct inode *inode = dentry->d_inode;
+ struct gfs2_inode *ip = GFS2_I(inode);
+ struct gfs2_holder gh;
+ int ret;

- return gfs2_ea_remove(GFS2_I(dentry->d_inode), &er);
+ gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh);
+ ret = gfs2_glock_nq(&gh);
+ if (ret == 0) {
+ ret = generic_removexattr(dentry, name);
+ gfs2_glock_dq(&gh);
+ }
+ gfs2_holder_uninit(&gh);
+ return ret;
}

static int gfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
diff --git a/fs/gfs2/super.h b/fs/gfs2/super.h
index 911c954..235db36 100644
--- a/fs/gfs2/super.h
+++ b/fs/gfs2/super.h
@@ -54,6 +54,7 @@ extern struct file_system_type gfs2meta_fs_type;
extern const struct export_operations gfs2_export_ops;
extern const struct super_operations gfs2_super_ops;
extern const struct dentry_operations gfs2_dops;
+extern struct xattr_handler *gfs2_xattr_handlers[];

#endif /* __SUPER_DOT_H__ */

--
1.6.2.5

2009-09-10 10:31:07

by Steven Whitehouse

[permalink] [raw]
Subject: [PATCH 12/15] GFS2: Rename eattr.[ch] as xattr.[ch]

Use the more conventional name for the extended attribute
support code. Update all the places which care.

Signed-off-by: Steven Whitehouse <[email protected]>
---
fs/gfs2/Makefile | 2 +-
fs/gfs2/acl.c | 2 +-
fs/gfs2/eattr.c | 1558 ---------------------------------------------------
fs/gfs2/eattr.h | 72 ---
fs/gfs2/inode.c | 2 +-
fs/gfs2/ops_inode.c | 2 +-
fs/gfs2/super.c | 2 +-
fs/gfs2/xattr.c | 1558 +++++++++++++++++++++++++++++++++++++++++++++++++++
fs/gfs2/xattr.h | 72 +++
9 files changed, 1635 insertions(+), 1635 deletions(-)
delete mode 100644 fs/gfs2/eattr.c
delete mode 100644 fs/gfs2/eattr.h
create mode 100644 fs/gfs2/xattr.c
create mode 100644 fs/gfs2/xattr.h

diff --git a/fs/gfs2/Makefile b/fs/gfs2/Makefile
index 2a4fd89..21f7e46 100644
--- a/fs/gfs2/Makefile
+++ b/fs/gfs2/Makefile
@@ -1,6 +1,6 @@
EXTRA_CFLAGS := -I$(src)
obj-$(CONFIG_GFS2_FS) += gfs2.o
-gfs2-y := acl.o bmap.o dir.o eattr.o glock.o \
+gfs2-y := acl.o bmap.o dir.o xattr.o glock.o \
glops.o inode.o log.o lops.o main.o meta_io.o \
aops.o dentry.o export.o file.o \
ops_fstype.o ops_inode.o quota.o \
diff --git a/fs/gfs2/acl.c b/fs/gfs2/acl.c
index f6777f1..3fc4e3a 100644
--- a/fs/gfs2/acl.c
+++ b/fs/gfs2/acl.c
@@ -19,7 +19,7 @@
#include "gfs2.h"
#include "incore.h"
#include "acl.h"
-#include "eattr.h"
+#include "xattr.h"
#include "glock.h"
#include "inode.h"
#include "meta_io.h"
diff --git a/fs/gfs2/eattr.c b/fs/gfs2/eattr.c
deleted file mode 100644
index d9a9e26..0000000
--- a/fs/gfs2/eattr.c
+++ /dev/null
@@ -1,1558 +0,0 @@
-/*
- * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved.
- * Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved.
- *
- * This copyrighted material is made available to anyone wishing to use,
- * modify, copy, or redistribute it subject to the terms and conditions
- * of the GNU General Public License version 2.
- */
-
-#include <linux/slab.h>
-#include <linux/spinlock.h>
-#include <linux/completion.h>
-#include <linux/buffer_head.h>
-#include <linux/xattr.h>
-#include <linux/gfs2_ondisk.h>
-#include <asm/uaccess.h>
-
-#include "gfs2.h"
-#include "incore.h"
-#include "acl.h"
-#include "eattr.h"
-#include "glock.h"
-#include "inode.h"
-#include "meta_io.h"
-#include "quota.h"
-#include "rgrp.h"
-#include "trans.h"
-#include "util.h"
-
-/**
- * ea_calc_size - returns the acutal number of bytes the request will take up
- * (not counting any unstuffed data blocks)
- * @sdp:
- * @er:
- * @size:
- *
- * Returns: 1 if the EA should be stuffed
- */
-
-static int ea_calc_size(struct gfs2_sbd *sdp, unsigned int nsize, size_t dsize,
- unsigned int *size)
-{
- unsigned int jbsize = sdp->sd_jbsize;
-
- /* Stuffed */
- *size = ALIGN(sizeof(struct gfs2_ea_header) + nsize + dsize, 8);
-
- if (*size <= jbsize)
- return 1;
-
- /* Unstuffed */
- *size = ALIGN(sizeof(struct gfs2_ea_header) + nsize +
- (sizeof(__be64) * DIV_ROUND_UP(dsize, jbsize)), 8);
-
- return 0;
-}
-
-static int ea_check_size(struct gfs2_sbd *sdp, unsigned int nsize, size_t dsize)
-{
- unsigned int size;
-
- if (dsize > GFS2_EA_MAX_DATA_LEN)
- return -ERANGE;
-
- ea_calc_size(sdp, nsize, dsize, &size);
-
- /* This can only happen with 512 byte blocks */
- if (size > sdp->sd_jbsize)
- return -ERANGE;
-
- return 0;
-}
-
-typedef int (*ea_call_t) (struct gfs2_inode *ip, struct buffer_head *bh,
- struct gfs2_ea_header *ea,
- struct gfs2_ea_header *prev, void *private);
-
-static int ea_foreach_i(struct gfs2_inode *ip, struct buffer_head *bh,
- ea_call_t ea_call, void *data)
-{
- struct gfs2_ea_header *ea, *prev = NULL;
- int error = 0;
-
- if (gfs2_metatype_check(GFS2_SB(&ip->i_inode), bh, GFS2_METATYPE_EA))
- return -EIO;
-
- for (ea = GFS2_EA_BH2FIRST(bh);; prev = ea, ea = GFS2_EA2NEXT(ea)) {
- if (!GFS2_EA_REC_LEN(ea))
- goto fail;
- if (!(bh->b_data <= (char *)ea && (char *)GFS2_EA2NEXT(ea) <=
- bh->b_data + bh->b_size))
- goto fail;
- if (!GFS2_EATYPE_VALID(ea->ea_type))
- goto fail;
-
- error = ea_call(ip, bh, ea, prev, data);
- if (error)
- return error;
-
- if (GFS2_EA_IS_LAST(ea)) {
- if ((char *)GFS2_EA2NEXT(ea) !=
- bh->b_data + bh->b_size)
- goto fail;
- break;
- }
- }
-
- return error;
-
-fail:
- gfs2_consist_inode(ip);
- return -EIO;
-}
-
-static int ea_foreach(struct gfs2_inode *ip, ea_call_t ea_call, void *data)
-{
- struct buffer_head *bh, *eabh;
- __be64 *eablk, *end;
- int error;
-
- error = gfs2_meta_read(ip->i_gl, ip->i_eattr, DIO_WAIT, &bh);
- if (error)
- return error;
-
- if (!(ip->i_diskflags & GFS2_DIF_EA_INDIRECT)) {
- error = ea_foreach_i(ip, bh, ea_call, data);
- goto out;
- }
-
- if (gfs2_metatype_check(GFS2_SB(&ip->i_inode), bh, GFS2_METATYPE_IN)) {
- error = -EIO;
- goto out;
- }
-
- eablk = (__be64 *)(bh->b_data + sizeof(struct gfs2_meta_header));
- end = eablk + GFS2_SB(&ip->i_inode)->sd_inptrs;
-
- for (; eablk < end; eablk++) {
- u64 bn;
-
- if (!*eablk)
- break;
- bn = be64_to_cpu(*eablk);
-
- error = gfs2_meta_read(ip->i_gl, bn, DIO_WAIT, &eabh);
- if (error)
- break;
- error = ea_foreach_i(ip, eabh, ea_call, data);
- brelse(eabh);
- if (error)
- break;
- }
-out:
- brelse(bh);
- return error;
-}
-
-struct ea_find {
- int type;
- const char *name;
- size_t namel;
- struct gfs2_ea_location *ef_el;
-};
-
-static int ea_find_i(struct gfs2_inode *ip, struct buffer_head *bh,
- struct gfs2_ea_header *ea, struct gfs2_ea_header *prev,
- void *private)
-{
- struct ea_find *ef = private;
-
- if (ea->ea_type == GFS2_EATYPE_UNUSED)
- return 0;
-
- if (ea->ea_type == ef->type) {
- if (ea->ea_name_len == ef->namel &&
- !memcmp(GFS2_EA2NAME(ea), ef->name, ea->ea_name_len)) {
- struct gfs2_ea_location *el = ef->ef_el;
- get_bh(bh);
- el->el_bh = bh;
- el->el_ea = ea;
- el->el_prev = prev;
- return 1;
- }
- }
-
- return 0;
-}
-
-int gfs2_ea_find(struct gfs2_inode *ip, int type, const char *name,
- struct gfs2_ea_location *el)
-{
- struct ea_find ef;
- int error;
-
- ef.type = type;
- ef.name = name;
- ef.namel = strlen(name);
- ef.ef_el = el;
-
- memset(el, 0, sizeof(struct gfs2_ea_location));
-
- error = ea_foreach(ip, ea_find_i, &ef);
- if (error > 0)
- return 0;
-
- return error;
-}
-
-/**
- * ea_dealloc_unstuffed -
- * @ip:
- * @bh:
- * @ea:
- * @prev:
- * @private:
- *
- * Take advantage of the fact that all unstuffed blocks are
- * allocated from the same RG. But watch, this may not always
- * be true.
- *
- * Returns: errno
- */
-
-static int ea_dealloc_unstuffed(struct gfs2_inode *ip, struct buffer_head *bh,
- struct gfs2_ea_header *ea,
- struct gfs2_ea_header *prev, void *private)
-{
- int *leave = private;
- struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
- struct gfs2_rgrpd *rgd;
- struct gfs2_holder rg_gh;
- struct buffer_head *dibh;
- __be64 *dataptrs;
- u64 bn = 0;
- u64 bstart = 0;
- unsigned int blen = 0;
- unsigned int blks = 0;
- unsigned int x;
- int error;
-
- if (GFS2_EA_IS_STUFFED(ea))
- return 0;
-
- dataptrs = GFS2_EA2DATAPTRS(ea);
- for (x = 0; x < ea->ea_num_ptrs; x++, dataptrs++) {
- if (*dataptrs) {
- blks++;
- bn = be64_to_cpu(*dataptrs);
- }
- }
- if (!blks)
- return 0;
-
- rgd = gfs2_blk2rgrpd(sdp, bn);
- if (!rgd) {
- gfs2_consist_inode(ip);
- return -EIO;
- }
-
- error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, &rg_gh);
- if (error)
- return error;
-
- error = gfs2_trans_begin(sdp, rgd->rd_length + RES_DINODE +
- RES_EATTR + RES_STATFS + RES_QUOTA, blks);
- if (error)
- goto out_gunlock;
-
- gfs2_trans_add_bh(ip->i_gl, bh, 1);
-
- dataptrs = GFS2_EA2DATAPTRS(ea);
- for (x = 0; x < ea->ea_num_ptrs; x++, dataptrs++) {
- if (!*dataptrs)
- break;
- bn = be64_to_cpu(*dataptrs);
-
- if (bstart + blen == bn)
- blen++;
- else {
- if (bstart)
- gfs2_free_meta(ip, bstart, blen);
- bstart = bn;
- blen = 1;
- }
-
- *dataptrs = 0;
- gfs2_add_inode_blocks(&ip->i_inode, -1);
- }
- if (bstart)
- gfs2_free_meta(ip, bstart, blen);
-
- if (prev && !leave) {
- u32 len;
-
- len = GFS2_EA_REC_LEN(prev) + GFS2_EA_REC_LEN(ea);
- prev->ea_rec_len = cpu_to_be32(len);
-
- if (GFS2_EA_IS_LAST(ea))
- prev->ea_flags |= GFS2_EAFLAG_LAST;
- } else {
- ea->ea_type = GFS2_EATYPE_UNUSED;
- ea->ea_num_ptrs = 0;
- }
-
- error = gfs2_meta_inode_buffer(ip, &dibh);
- if (!error) {
- ip->i_inode.i_ctime = CURRENT_TIME;
- gfs2_trans_add_bh(ip->i_gl, dibh, 1);
- gfs2_dinode_out(ip, dibh->b_data);
- brelse(dibh);
- }
-
- gfs2_trans_end(sdp);
-
-out_gunlock:
- gfs2_glock_dq_uninit(&rg_gh);
- return error;
-}
-
-static int ea_remove_unstuffed(struct gfs2_inode *ip, struct buffer_head *bh,
- struct gfs2_ea_header *ea,
- struct gfs2_ea_header *prev, int leave)
-{
- struct gfs2_alloc *al;
- int error;
-
- al = gfs2_alloc_get(ip);
- if (!al)
- return -ENOMEM;
-
- error = gfs2_quota_hold(ip, NO_QUOTA_CHANGE, NO_QUOTA_CHANGE);
- if (error)
- goto out_alloc;
-
- error = gfs2_rindex_hold(GFS2_SB(&ip->i_inode), &al->al_ri_gh);
- if (error)
- goto out_quota;
-
- error = ea_dealloc_unstuffed(ip, bh, ea, prev, (leave) ? &error : NULL);
-
- gfs2_glock_dq_uninit(&al->al_ri_gh);
-
-out_quota:
- gfs2_quota_unhold(ip);
-out_alloc:
- gfs2_alloc_put(ip);
- return error;
-}
-
-struct ea_list {
- struct gfs2_ea_request *ei_er;
- unsigned int ei_size;
-};
-
-static inline unsigned int gfs2_ea_strlen(struct gfs2_ea_header *ea)
-{
- switch (ea->ea_type) {
- case GFS2_EATYPE_USR:
- return 5 + ea->ea_name_len + 1;
- case GFS2_EATYPE_SYS:
- return 7 + ea->ea_name_len + 1;
- case GFS2_EATYPE_SECURITY:
- return 9 + ea->ea_name_len + 1;
- default:
- return 0;
- }
-}
-
-static int ea_list_i(struct gfs2_inode *ip, struct buffer_head *bh,
- struct gfs2_ea_header *ea, struct gfs2_ea_header *prev,
- void *private)
-{
- struct ea_list *ei = private;
- struct gfs2_ea_request *er = ei->ei_er;
- unsigned int ea_size = gfs2_ea_strlen(ea);
-
- if (ea->ea_type == GFS2_EATYPE_UNUSED)
- return 0;
-
- if (er->er_data_len) {
- char *prefix = NULL;
- unsigned int l = 0;
- char c = 0;
-
- if (ei->ei_size + ea_size > er->er_data_len)
- return -ERANGE;
-
- switch (ea->ea_type) {
- case GFS2_EATYPE_USR:
- prefix = "user.";
- l = 5;
- break;
- case GFS2_EATYPE_SYS:
- prefix = "system.";
- l = 7;
- break;
- case GFS2_EATYPE_SECURITY:
- prefix = "security.";
- l = 9;
- break;
- }
-
- BUG_ON(l == 0);
-
- memcpy(er->er_data + ei->ei_size, prefix, l);
- memcpy(er->er_data + ei->ei_size + l, GFS2_EA2NAME(ea),
- ea->ea_name_len);
- memcpy(er->er_data + ei->ei_size + ea_size - 1, &c, 1);
- }
-
- ei->ei_size += ea_size;
-
- return 0;
-}
-
-/**
- * gfs2_listxattr - List gfs2 extended attributes
- * @dentry: The dentry whose inode we are interested in
- * @buffer: The buffer to write the results
- * @size: The size of the buffer
- *
- * Returns: actual size of data on success, -errno on error
- */
-
-ssize_t gfs2_listxattr(struct dentry *dentry, char *buffer, size_t size)
-{
- struct gfs2_inode *ip = GFS2_I(dentry->d_inode);
- struct gfs2_ea_request er;
- struct gfs2_holder i_gh;
- int error;
-
- memset(&er, 0, sizeof(struct gfs2_ea_request));
- if (size) {
- er.er_data = buffer;
- er.er_data_len = size;
- }
-
- error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &i_gh);
- if (error)
- return error;
-
- if (ip->i_eattr) {
- struct ea_list ei = { .ei_er = &er, .ei_size = 0 };
-
- error = ea_foreach(ip, ea_list_i, &ei);
- if (!error)
- error = ei.ei_size;
- }
-
- gfs2_glock_dq_uninit(&i_gh);
-
- return error;
-}
-
-/**
- * ea_get_unstuffed - actually copies the unstuffed data into the
- * request buffer
- * @ip: The GFS2 inode
- * @ea: The extended attribute header structure
- * @data: The data to be copied
- *
- * Returns: errno
- */
-
-static int ea_get_unstuffed(struct gfs2_inode *ip, struct gfs2_ea_header *ea,
- char *data)
-{
- struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
- struct buffer_head **bh;
- unsigned int amount = GFS2_EA_DATA_LEN(ea);
- unsigned int nptrs = DIV_ROUND_UP(amount, sdp->sd_jbsize);
- __be64 *dataptrs = GFS2_EA2DATAPTRS(ea);
- unsigned int x;
- int error = 0;
-
- bh = kcalloc(nptrs, sizeof(struct buffer_head *), GFP_NOFS);
- if (!bh)
- return -ENOMEM;
-
- for (x = 0; x < nptrs; x++) {
- error = gfs2_meta_read(ip->i_gl, be64_to_cpu(*dataptrs), 0,
- bh + x);
- if (error) {
- while (x--)
- brelse(bh[x]);
- goto out;
- }
- dataptrs++;
- }
-
- for (x = 0; x < nptrs; x++) {
- error = gfs2_meta_wait(sdp, bh[x]);
- if (error) {
- for (; x < nptrs; x++)
- brelse(bh[x]);
- goto out;
- }
- if (gfs2_metatype_check(sdp, bh[x], GFS2_METATYPE_ED)) {
- for (; x < nptrs; x++)
- brelse(bh[x]);
- error = -EIO;
- goto out;
- }
-
- memcpy(data, bh[x]->b_data + sizeof(struct gfs2_meta_header),
- (sdp->sd_jbsize > amount) ? amount : sdp->sd_jbsize);
-
- amount -= sdp->sd_jbsize;
- data += sdp->sd_jbsize;
-
- brelse(bh[x]);
- }
-
-out:
- kfree(bh);
- return error;
-}
-
-int gfs2_ea_get_copy(struct gfs2_inode *ip, struct gfs2_ea_location *el,
- char *data, size_t size)
-{
- int ret;
- size_t len = GFS2_EA_DATA_LEN(el->el_ea);
- if (len > size)
- return -ERANGE;
-
- if (GFS2_EA_IS_STUFFED(el->el_ea)) {
- memcpy(data, GFS2_EA2DATA(el->el_ea), len);
- return len;
- }
- ret = ea_get_unstuffed(ip, el->el_ea, data);
- if (ret < 0)
- return ret;
- return len;
-}
-
-/**
- * gfs2_xattr_get - Get a GFS2 extended attribute
- * @inode: The inode
- * @type: The type of extended attribute
- * @name: The name of the extended attribute
- * @buffer: The buffer to write the result into
- * @size: The size of the buffer
- *
- * Returns: actual size of data on success, -errno on error
- */
-
-int gfs2_xattr_get(struct inode *inode, int type, const char *name,
- void *buffer, size_t size)
-{
- struct gfs2_inode *ip = GFS2_I(inode);
- struct gfs2_ea_location el;
- int error;
-
- if (!ip->i_eattr)
- return -ENODATA;
- if (strlen(name) > GFS2_EA_MAX_NAME_LEN)
- return -EINVAL;
-
- error = gfs2_ea_find(ip, type, name, &el);
- if (error)
- return error;
- if (!el.el_ea)
- return -ENODATA;
- if (size)
- error = gfs2_ea_get_copy(ip, &el, buffer, size);
- else
- error = GFS2_EA_DATA_LEN(el.el_ea);
- brelse(el.el_bh);
-
- return error;
-}
-
-/**
- * ea_alloc_blk - allocates a new block for extended attributes.
- * @ip: A pointer to the inode that's getting extended attributes
- * @bhp: Pointer to pointer to a struct buffer_head
- *
- * Returns: errno
- */
-
-static int ea_alloc_blk(struct gfs2_inode *ip, struct buffer_head **bhp)
-{
- struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
- struct gfs2_ea_header *ea;
- unsigned int n = 1;
- u64 block;
- int error;
-
- error = gfs2_alloc_block(ip, &block, &n);
- if (error)
- return error;
- gfs2_trans_add_unrevoke(sdp, block, 1);
- *bhp = gfs2_meta_new(ip->i_gl, block);
- gfs2_trans_add_bh(ip->i_gl, *bhp, 1);
- gfs2_metatype_set(*bhp, GFS2_METATYPE_EA, GFS2_FORMAT_EA);
- gfs2_buffer_clear_tail(*bhp, sizeof(struct gfs2_meta_header));
-
- ea = GFS2_EA_BH2FIRST(*bhp);
- ea->ea_rec_len = cpu_to_be32(sdp->sd_jbsize);
- ea->ea_type = GFS2_EATYPE_UNUSED;
- ea->ea_flags = GFS2_EAFLAG_LAST;
- ea->ea_num_ptrs = 0;
-
- gfs2_add_inode_blocks(&ip->i_inode, 1);
-
- return 0;
-}
-
-/**
- * ea_write - writes the request info to an ea, creating new blocks if
- * necessary
- * @ip: inode that is being modified
- * @ea: the location of the new ea in a block
- * @er: the write request
- *
- * Note: does not update ea_rec_len or the GFS2_EAFLAG_LAST bin of ea_flags
- *
- * returns : errno
- */
-
-static int ea_write(struct gfs2_inode *ip, struct gfs2_ea_header *ea,
- struct gfs2_ea_request *er)
-{
- struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
- int error;
-
- ea->ea_data_len = cpu_to_be32(er->er_data_len);
- ea->ea_name_len = er->er_name_len;
- ea->ea_type = er->er_type;
- ea->__pad = 0;
-
- memcpy(GFS2_EA2NAME(ea), er->er_name, er->er_name_len);
-
- if (GFS2_EAREQ_SIZE_STUFFED(er) <= sdp->sd_jbsize) {
- ea->ea_num_ptrs = 0;
- memcpy(GFS2_EA2DATA(ea), er->er_data, er->er_data_len);
- } else {
- __be64 *dataptr = GFS2_EA2DATAPTRS(ea);
- const char *data = er->er_data;
- unsigned int data_len = er->er_data_len;
- unsigned int copy;
- unsigned int x;
-
- ea->ea_num_ptrs = DIV_ROUND_UP(er->er_data_len, sdp->sd_jbsize);
- for (x = 0; x < ea->ea_num_ptrs; x++) {
- struct buffer_head *bh;
- u64 block;
- int mh_size = sizeof(struct gfs2_meta_header);
- unsigned int n = 1;
-
- error = gfs2_alloc_block(ip, &block, &n);
- if (error)
- return error;
- gfs2_trans_add_unrevoke(sdp, block, 1);
- bh = gfs2_meta_new(ip->i_gl, block);
- gfs2_trans_add_bh(ip->i_gl, bh, 1);
- gfs2_metatype_set(bh, GFS2_METATYPE_ED, GFS2_FORMAT_ED);
-
- gfs2_add_inode_blocks(&ip->i_inode, 1);
-
- copy = data_len > sdp->sd_jbsize ? sdp->sd_jbsize :
- data_len;
- memcpy(bh->b_data + mh_size, data, copy);
- if (copy < sdp->sd_jbsize)
- memset(bh->b_data + mh_size + copy, 0,
- sdp->sd_jbsize - copy);
-
- *dataptr++ = cpu_to_be64(bh->b_blocknr);
- data += copy;
- data_len -= copy;
-
- brelse(bh);
- }
-
- gfs2_assert_withdraw(sdp, !data_len);
- }
-
- return 0;
-}
-
-typedef int (*ea_skeleton_call_t) (struct gfs2_inode *ip,
- struct gfs2_ea_request *er, void *private);
-
-static int ea_alloc_skeleton(struct gfs2_inode *ip, struct gfs2_ea_request *er,
- unsigned int blks,
- ea_skeleton_call_t skeleton_call, void *private)
-{
- struct gfs2_alloc *al;
- struct buffer_head *dibh;
- int error;
-
- al = gfs2_alloc_get(ip);
- if (!al)
- return -ENOMEM;
-
- error = gfs2_quota_lock_check(ip);
- if (error)
- goto out;
-
- al->al_requested = blks;
-
- error = gfs2_inplace_reserve(ip);
- if (error)
- goto out_gunlock_q;
-
- error = gfs2_trans_begin(GFS2_SB(&ip->i_inode),
- blks + al->al_rgd->rd_length +
- RES_DINODE + RES_STATFS + RES_QUOTA, 0);
- if (error)
- goto out_ipres;
-
- error = skeleton_call(ip, er, private);
- if (error)
- goto out_end_trans;
-
- error = gfs2_meta_inode_buffer(ip, &dibh);
- if (!error) {
- ip->i_inode.i_ctime = CURRENT_TIME;
- gfs2_trans_add_bh(ip->i_gl, dibh, 1);
- gfs2_dinode_out(ip, dibh->b_data);
- brelse(dibh);
- }
-
-out_end_trans:
- gfs2_trans_end(GFS2_SB(&ip->i_inode));
-out_ipres:
- gfs2_inplace_release(ip);
-out_gunlock_q:
- gfs2_quota_unlock(ip);
-out:
- gfs2_alloc_put(ip);
- return error;
-}
-
-static int ea_init_i(struct gfs2_inode *ip, struct gfs2_ea_request *er,
- void *private)
-{
- struct buffer_head *bh;
- int error;
-
- error = ea_alloc_blk(ip, &bh);
- if (error)
- return error;
-
- ip->i_eattr = bh->b_blocknr;
- error = ea_write(ip, GFS2_EA_BH2FIRST(bh), er);
-
- brelse(bh);
-
- return error;
-}
-
-/**
- * ea_init - initializes a new eattr block
- * @ip:
- * @er:
- *
- * Returns: errno
- */
-
-static int ea_init(struct gfs2_inode *ip, int type, const char *name,
- const void *data, size_t size)
-{
- struct gfs2_ea_request er;
- unsigned int jbsize = GFS2_SB(&ip->i_inode)->sd_jbsize;
- unsigned int blks = 1;
-
- er.er_type = type;
- er.er_name = name;
- er.er_name_len = strlen(name);
- er.er_data = (void *)data;
- er.er_data_len = size;
-
- if (GFS2_EAREQ_SIZE_STUFFED(&er) > jbsize)
- blks += DIV_ROUND_UP(er.er_data_len, jbsize);
-
- return ea_alloc_skeleton(ip, &er, blks, ea_init_i, NULL);
-}
-
-static struct gfs2_ea_header *ea_split_ea(struct gfs2_ea_header *ea)
-{
- u32 ea_size = GFS2_EA_SIZE(ea);
- struct gfs2_ea_header *new = (struct gfs2_ea_header *)((char *)ea +
- ea_size);
- u32 new_size = GFS2_EA_REC_LEN(ea) - ea_size;
- int last = ea->ea_flags & GFS2_EAFLAG_LAST;
-
- ea->ea_rec_len = cpu_to_be32(ea_size);
- ea->ea_flags ^= last;
-
- new->ea_rec_len = cpu_to_be32(new_size);
- new->ea_flags = last;
-
- return new;
-}
-
-static void ea_set_remove_stuffed(struct gfs2_inode *ip,
- struct gfs2_ea_location *el)
-{
- struct gfs2_ea_header *ea = el->el_ea;
- struct gfs2_ea_header *prev = el->el_prev;
- u32 len;
-
- gfs2_trans_add_bh(ip->i_gl, el->el_bh, 1);
-
- if (!prev || !GFS2_EA_IS_STUFFED(ea)) {
- ea->ea_type = GFS2_EATYPE_UNUSED;
- return;
- } else if (GFS2_EA2NEXT(prev) != ea) {
- prev = GFS2_EA2NEXT(prev);
- gfs2_assert_withdraw(GFS2_SB(&ip->i_inode), GFS2_EA2NEXT(prev) == ea);
- }
-
- len = GFS2_EA_REC_LEN(prev) + GFS2_EA_REC_LEN(ea);
- prev->ea_rec_len = cpu_to_be32(len);
-
- if (GFS2_EA_IS_LAST(ea))
- prev->ea_flags |= GFS2_EAFLAG_LAST;
-}
-
-struct ea_set {
- int ea_split;
-
- struct gfs2_ea_request *es_er;
- struct gfs2_ea_location *es_el;
-
- struct buffer_head *es_bh;
- struct gfs2_ea_header *es_ea;
-};
-
-static int ea_set_simple_noalloc(struct gfs2_inode *ip, struct buffer_head *bh,
- struct gfs2_ea_header *ea, struct ea_set *es)
-{
- struct gfs2_ea_request *er = es->es_er;
- struct buffer_head *dibh;
- int error;
-
- error = gfs2_trans_begin(GFS2_SB(&ip->i_inode), RES_DINODE + 2 * RES_EATTR, 0);
- if (error)
- return error;
-
- gfs2_trans_add_bh(ip->i_gl, bh, 1);
-
- if (es->ea_split)
- ea = ea_split_ea(ea);
-
- ea_write(ip, ea, er);
-
- if (es->es_el)
- ea_set_remove_stuffed(ip, es->es_el);
-
- error = gfs2_meta_inode_buffer(ip, &dibh);
- if (error)
- goto out;
- ip->i_inode.i_ctime = CURRENT_TIME;
- gfs2_trans_add_bh(ip->i_gl, dibh, 1);
- gfs2_dinode_out(ip, dibh->b_data);
- brelse(dibh);
-out:
- gfs2_trans_end(GFS2_SB(&ip->i_inode));
- return error;
-}
-
-static int ea_set_simple_alloc(struct gfs2_inode *ip,
- struct gfs2_ea_request *er, void *private)
-{
- struct ea_set *es = private;
- struct gfs2_ea_header *ea = es->es_ea;
- int error;
-
- gfs2_trans_add_bh(ip->i_gl, es->es_bh, 1);
-
- if (es->ea_split)
- ea = ea_split_ea(ea);
-
- error = ea_write(ip, ea, er);
- if (error)
- return error;
-
- if (es->es_el)
- ea_set_remove_stuffed(ip, es->es_el);
-
- return 0;
-}
-
-static int ea_set_simple(struct gfs2_inode *ip, struct buffer_head *bh,
- struct gfs2_ea_header *ea, struct gfs2_ea_header *prev,
- void *private)
-{
- struct ea_set *es = private;
- unsigned int size;
- int stuffed;
- int error;
-
- stuffed = ea_calc_size(GFS2_SB(&ip->i_inode), es->es_er->er_name_len,
- es->es_er->er_data_len, &size);
-
- if (ea->ea_type == GFS2_EATYPE_UNUSED) {
- if (GFS2_EA_REC_LEN(ea) < size)
- return 0;
- if (!GFS2_EA_IS_STUFFED(ea)) {
- error = ea_remove_unstuffed(ip, bh, ea, prev, 1);
- if (error)
- return error;
- }
- es->ea_split = 0;
- } else if (GFS2_EA_REC_LEN(ea) - GFS2_EA_SIZE(ea) >= size)
- es->ea_split = 1;
- else
- return 0;
-
- if (stuffed) {
- error = ea_set_simple_noalloc(ip, bh, ea, es);
- if (error)
- return error;
- } else {
- unsigned int blks;
-
- es->es_bh = bh;
- es->es_ea = ea;
- blks = 2 + DIV_ROUND_UP(es->es_er->er_data_len,
- GFS2_SB(&ip->i_inode)->sd_jbsize);
-
- error = ea_alloc_skeleton(ip, es->es_er, blks,
- ea_set_simple_alloc, es);
- if (error)
- return error;
- }
-
- return 1;
-}
-
-static int ea_set_block(struct gfs2_inode *ip, struct gfs2_ea_request *er,
- void *private)
-{
- struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
- struct buffer_head *indbh, *newbh;
- __be64 *eablk;
- int error;
- int mh_size = sizeof(struct gfs2_meta_header);
-
- if (ip->i_diskflags & GFS2_DIF_EA_INDIRECT) {
- __be64 *end;
-
- error = gfs2_meta_read(ip->i_gl, ip->i_eattr, DIO_WAIT,
- &indbh);
- if (error)
- return error;
-
- if (gfs2_metatype_check(sdp, indbh, GFS2_METATYPE_IN)) {
- error = -EIO;
- goto out;
- }
-
- eablk = (__be64 *)(indbh->b_data + mh_size);
- end = eablk + sdp->sd_inptrs;
-
- for (; eablk < end; eablk++)
- if (!*eablk)
- break;
-
- if (eablk == end) {
- error = -ENOSPC;
- goto out;
- }
-
- gfs2_trans_add_bh(ip->i_gl, indbh, 1);
- } else {
- u64 blk;
- unsigned int n = 1;
- error = gfs2_alloc_block(ip, &blk, &n);
- if (error)
- return error;
- gfs2_trans_add_unrevoke(sdp, blk, 1);
- indbh = gfs2_meta_new(ip->i_gl, blk);
- gfs2_trans_add_bh(ip->i_gl, indbh, 1);
- gfs2_metatype_set(indbh, GFS2_METATYPE_IN, GFS2_FORMAT_IN);
- gfs2_buffer_clear_tail(indbh, mh_size);
-
- eablk = (__be64 *)(indbh->b_data + mh_size);
- *eablk = cpu_to_be64(ip->i_eattr);
- ip->i_eattr = blk;
- ip->i_diskflags |= GFS2_DIF_EA_INDIRECT;
- gfs2_add_inode_blocks(&ip->i_inode, 1);
-
- eablk++;
- }
-
- error = ea_alloc_blk(ip, &newbh);
- if (error)
- goto out;
-
- *eablk = cpu_to_be64((u64)newbh->b_blocknr);
- error = ea_write(ip, GFS2_EA_BH2FIRST(newbh), er);
- brelse(newbh);
- if (error)
- goto out;
-
- if (private)
- ea_set_remove_stuffed(ip, private);
-
-out:
- brelse(indbh);
- return error;
-}
-
-static int ea_set_i(struct gfs2_inode *ip, int type, const char *name,
- const void *value, size_t size, struct gfs2_ea_location *el)
-{
- struct gfs2_ea_request er;
- struct ea_set es;
- unsigned int blks = 2;
- int error;
-
- er.er_type = type;
- er.er_name = name;
- er.er_data = (void *)value;
- er.er_name_len = strlen(name);
- er.er_data_len = size;
-
- memset(&es, 0, sizeof(struct ea_set));
- es.es_er = &er;
- es.es_el = el;
-
- error = ea_foreach(ip, ea_set_simple, &es);
- if (error > 0)
- return 0;
- if (error)
- return error;
-
- if (!(ip->i_diskflags & GFS2_DIF_EA_INDIRECT))
- blks++;
- if (GFS2_EAREQ_SIZE_STUFFED(&er) > GFS2_SB(&ip->i_inode)->sd_jbsize)
- blks += DIV_ROUND_UP(er.er_data_len, GFS2_SB(&ip->i_inode)->sd_jbsize);
-
- return ea_alloc_skeleton(ip, &er, blks, ea_set_block, el);
-}
-
-static int ea_set_remove_unstuffed(struct gfs2_inode *ip,
- struct gfs2_ea_location *el)
-{
- if (el->el_prev && GFS2_EA2NEXT(el->el_prev) != el->el_ea) {
- el->el_prev = GFS2_EA2NEXT(el->el_prev);
- gfs2_assert_withdraw(GFS2_SB(&ip->i_inode),
- GFS2_EA2NEXT(el->el_prev) == el->el_ea);
- }
-
- return ea_remove_unstuffed(ip, el->el_bh, el->el_ea, el->el_prev,0);
-}
-
-static int ea_remove_stuffed(struct gfs2_inode *ip, struct gfs2_ea_location *el)
-{
- struct gfs2_ea_header *ea = el->el_ea;
- struct gfs2_ea_header *prev = el->el_prev;
- struct buffer_head *dibh;
- int error;
-
- error = gfs2_trans_begin(GFS2_SB(&ip->i_inode), RES_DINODE + RES_EATTR, 0);
- if (error)
- return error;
-
- gfs2_trans_add_bh(ip->i_gl, el->el_bh, 1);
-
- if (prev) {
- u32 len;
-
- len = GFS2_EA_REC_LEN(prev) + GFS2_EA_REC_LEN(ea);
- prev->ea_rec_len = cpu_to_be32(len);
-
- if (GFS2_EA_IS_LAST(ea))
- prev->ea_flags |= GFS2_EAFLAG_LAST;
- } else {
- ea->ea_type = GFS2_EATYPE_UNUSED;
- }
-
- error = gfs2_meta_inode_buffer(ip, &dibh);
- if (!error) {
- ip->i_inode.i_ctime = CURRENT_TIME;
- gfs2_trans_add_bh(ip->i_gl, dibh, 1);
- gfs2_dinode_out(ip, dibh->b_data);
- brelse(dibh);
- }
-
- gfs2_trans_end(GFS2_SB(&ip->i_inode));
-
- return error;
-}
-
-/**
- * gfs2_xattr_remove - Remove a GFS2 extended attribute
- * @inode: The inode
- * @type: The type of the extended attribute
- * @name: The name of the extended attribute
- *
- * This is not called directly by the VFS since we use the (common)
- * scheme of making a "set with NULL data" mean a remove request. Note
- * that this is different from a set with zero length data.
- *
- * Returns: 0, or errno on failure
- */
-
-static int gfs2_xattr_remove(struct inode *inode, int type, const char *name)
-{
- struct gfs2_inode *ip = GFS2_I(inode);
- struct gfs2_ea_location el;
- int error;
-
- if (!ip->i_eattr)
- return -ENODATA;
-
- error = gfs2_ea_find(ip, type, name, &el);
- if (error)
- return error;
- if (!el.el_ea)
- return -ENODATA;
-
- if (GFS2_EA_IS_STUFFED(el.el_ea))
- error = ea_remove_stuffed(ip, &el);
- else
- error = ea_remove_unstuffed(ip, el.el_bh, el.el_ea, el.el_prev, 0);
-
- brelse(el.el_bh);
-
- return error;
-}
-
-/**
- * gfs2_xattr_set - Set (or remove) a GFS2 extended attribute
- * @inode: The inode
- * @type: The type of the extended attribute
- * @name: The name of the extended attribute
- * @value: The value of the extended attribute (NULL for remove)
- * @size: The size of the @value argument
- * @flags: Create or Replace
- *
- * See gfs2_xattr_remove() for details of the removal of xattrs.
- *
- * Returns: 0 or errno on failure
- */
-
-int gfs2_xattr_set(struct inode *inode, int type, const char *name,
- const void *value, size_t size, int flags)
-{
- struct gfs2_sbd *sdp = GFS2_SB(inode);
- struct gfs2_inode *ip = GFS2_I(inode);
- struct gfs2_ea_location el;
- unsigned int namel = strlen(name);
- int error;
-
- if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
- return -EPERM;
- if (namel > GFS2_EA_MAX_NAME_LEN)
- return -ERANGE;
-
- if (value == NULL)
- return gfs2_xattr_remove(inode, type, name);
-
- if (ea_check_size(sdp, namel, size))
- return -ERANGE;
-
- if (!ip->i_eattr) {
- if (flags & XATTR_REPLACE)
- return -ENODATA;
- return ea_init(ip, type, name, value, size);
- }
-
- error = gfs2_ea_find(ip, type, name, &el);
- if (error)
- return error;
-
- if (el.el_ea) {
- if (ip->i_diskflags & GFS2_DIF_APPENDONLY) {
- brelse(el.el_bh);
- return -EPERM;
- }
-
- error = -EEXIST;
- if (!(flags & XATTR_CREATE)) {
- int unstuffed = !GFS2_EA_IS_STUFFED(el.el_ea);
- error = ea_set_i(ip, type, name, value, size, &el);
- if (!error && unstuffed)
- ea_set_remove_unstuffed(ip, &el);
- }
-
- brelse(el.el_bh);
- return error;
- }
-
- error = -ENODATA;
- if (!(flags & XATTR_REPLACE))
- error = ea_set_i(ip, type, name, value, size, NULL);
-
- return error;
-}
-
-static int ea_acl_chmod_unstuffed(struct gfs2_inode *ip,
- struct gfs2_ea_header *ea, char *data)
-{
- struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
- struct buffer_head **bh;
- unsigned int amount = GFS2_EA_DATA_LEN(ea);
- unsigned int nptrs = DIV_ROUND_UP(amount, sdp->sd_jbsize);
- __be64 *dataptrs = GFS2_EA2DATAPTRS(ea);
- unsigned int x;
- int error;
-
- bh = kcalloc(nptrs, sizeof(struct buffer_head *), GFP_NOFS);
- if (!bh)
- return -ENOMEM;
-
- error = gfs2_trans_begin(sdp, nptrs + RES_DINODE, 0);
- if (error)
- goto out;
-
- for (x = 0; x < nptrs; x++) {
- error = gfs2_meta_read(ip->i_gl, be64_to_cpu(*dataptrs), 0,
- bh + x);
- if (error) {
- while (x--)
- brelse(bh[x]);
- goto fail;
- }
- dataptrs++;
- }
-
- for (x = 0; x < nptrs; x++) {
- error = gfs2_meta_wait(sdp, bh[x]);
- if (error) {
- for (; x < nptrs; x++)
- brelse(bh[x]);
- goto fail;
- }
- if (gfs2_metatype_check(sdp, bh[x], GFS2_METATYPE_ED)) {
- for (; x < nptrs; x++)
- brelse(bh[x]);
- error = -EIO;
- goto fail;
- }
-
- gfs2_trans_add_bh(ip->i_gl, bh[x], 1);
-
- memcpy(bh[x]->b_data + sizeof(struct gfs2_meta_header), data,
- (sdp->sd_jbsize > amount) ? amount : sdp->sd_jbsize);
-
- amount -= sdp->sd_jbsize;
- data += sdp->sd_jbsize;
-
- brelse(bh[x]);
- }
-
-out:
- kfree(bh);
- return error;
-
-fail:
- gfs2_trans_end(sdp);
- kfree(bh);
- return error;
-}
-
-int gfs2_ea_acl_chmod(struct gfs2_inode *ip, struct gfs2_ea_location *el,
- struct iattr *attr, char *data)
-{
- struct buffer_head *dibh;
- int error;
-
- if (GFS2_EA_IS_STUFFED(el->el_ea)) {
- error = gfs2_trans_begin(GFS2_SB(&ip->i_inode), RES_DINODE + RES_EATTR, 0);
- if (error)
- return error;
-
- gfs2_trans_add_bh(ip->i_gl, el->el_bh, 1);
- memcpy(GFS2_EA2DATA(el->el_ea), data,
- GFS2_EA_DATA_LEN(el->el_ea));
- } else
- error = ea_acl_chmod_unstuffed(ip, el->el_ea, data);
-
- if (error)
- return error;
-
- error = gfs2_meta_inode_buffer(ip, &dibh);
- if (!error) {
- error = inode_setattr(&ip->i_inode, attr);
- gfs2_assert_warn(GFS2_SB(&ip->i_inode), !error);
- gfs2_trans_add_bh(ip->i_gl, dibh, 1);
- gfs2_dinode_out(ip, dibh->b_data);
- brelse(dibh);
- }
-
- gfs2_trans_end(GFS2_SB(&ip->i_inode));
-
- return error;
-}
-
-static int ea_dealloc_indirect(struct gfs2_inode *ip)
-{
- struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
- struct gfs2_rgrp_list rlist;
- struct buffer_head *indbh, *dibh;
- __be64 *eablk, *end;
- unsigned int rg_blocks = 0;
- u64 bstart = 0;
- unsigned int blen = 0;
- unsigned int blks = 0;
- unsigned int x;
- int error;
-
- memset(&rlist, 0, sizeof(struct gfs2_rgrp_list));
-
- error = gfs2_meta_read(ip->i_gl, ip->i_eattr, DIO_WAIT, &indbh);
- if (error)
- return error;
-
- if (gfs2_metatype_check(sdp, indbh, GFS2_METATYPE_IN)) {
- error = -EIO;
- goto out;
- }
-
- eablk = (__be64 *)(indbh->b_data + sizeof(struct gfs2_meta_header));
- end = eablk + sdp->sd_inptrs;
-
- for (; eablk < end; eablk++) {
- u64 bn;
-
- if (!*eablk)
- break;
- bn = be64_to_cpu(*eablk);
-
- if (bstart + blen == bn)
- blen++;
- else {
- if (bstart)
- gfs2_rlist_add(sdp, &rlist, bstart);
- bstart = bn;
- blen = 1;
- }
- blks++;
- }
- if (bstart)
- gfs2_rlist_add(sdp, &rlist, bstart);
- else
- goto out;
-
- gfs2_rlist_alloc(&rlist, LM_ST_EXCLUSIVE);
-
- for (x = 0; x < rlist.rl_rgrps; x++) {
- struct gfs2_rgrpd *rgd;
- rgd = rlist.rl_ghs[x].gh_gl->gl_object;
- rg_blocks += rgd->rd_length;
- }
-
- error = gfs2_glock_nq_m(rlist.rl_rgrps, rlist.rl_ghs);
- if (error)
- goto out_rlist_free;
-
- error = gfs2_trans_begin(sdp, rg_blocks + RES_DINODE + RES_INDIRECT +
- RES_STATFS + RES_QUOTA, blks);
- if (error)
- goto out_gunlock;
-
- gfs2_trans_add_bh(ip->i_gl, indbh, 1);
-
- eablk = (__be64 *)(indbh->b_data + sizeof(struct gfs2_meta_header));
- bstart = 0;
- blen = 0;
-
- for (; eablk < end; eablk++) {
- u64 bn;
-
- if (!*eablk)
- break;
- bn = be64_to_cpu(*eablk);
-
- if (bstart + blen == bn)
- blen++;
- else {
- if (bstart)
- gfs2_free_meta(ip, bstart, blen);
- bstart = bn;
- blen = 1;
- }
-
- *eablk = 0;
- gfs2_add_inode_blocks(&ip->i_inode, -1);
- }
- if (bstart)
- gfs2_free_meta(ip, bstart, blen);
-
- ip->i_diskflags &= ~GFS2_DIF_EA_INDIRECT;
-
- error = gfs2_meta_inode_buffer(ip, &dibh);
- if (!error) {
- gfs2_trans_add_bh(ip->i_gl, dibh, 1);
- gfs2_dinode_out(ip, dibh->b_data);
- brelse(dibh);
- }
-
- gfs2_trans_end(sdp);
-
-out_gunlock:
- gfs2_glock_dq_m(rlist.rl_rgrps, rlist.rl_ghs);
-out_rlist_free:
- gfs2_rlist_free(&rlist);
-out:
- brelse(indbh);
- return error;
-}
-
-static int ea_dealloc_block(struct gfs2_inode *ip)
-{
- struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
- struct gfs2_alloc *al = ip->i_alloc;
- struct gfs2_rgrpd *rgd;
- struct buffer_head *dibh;
- int error;
-
- rgd = gfs2_blk2rgrpd(sdp, ip->i_eattr);
- if (!rgd) {
- gfs2_consist_inode(ip);
- return -EIO;
- }
-
- error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0,
- &al->al_rgd_gh);
- if (error)
- return error;
-
- error = gfs2_trans_begin(sdp, RES_RG_BIT + RES_DINODE + RES_STATFS +
- RES_QUOTA, 1);
- if (error)
- goto out_gunlock;
-
- gfs2_free_meta(ip, ip->i_eattr, 1);
-
- ip->i_eattr = 0;
- gfs2_add_inode_blocks(&ip->i_inode, -1);
-
- error = gfs2_meta_inode_buffer(ip, &dibh);
- if (!error) {
- gfs2_trans_add_bh(ip->i_gl, dibh, 1);
- gfs2_dinode_out(ip, dibh->b_data);
- brelse(dibh);
- }
-
- gfs2_trans_end(sdp);
-
-out_gunlock:
- gfs2_glock_dq_uninit(&al->al_rgd_gh);
- return error;
-}
-
-/**
- * gfs2_ea_dealloc - deallocate the extended attribute fork
- * @ip: the inode
- *
- * Returns: errno
- */
-
-int gfs2_ea_dealloc(struct gfs2_inode *ip)
-{
- struct gfs2_alloc *al;
- int error;
-
- al = gfs2_alloc_get(ip);
- if (!al)
- return -ENOMEM;
-
- error = gfs2_quota_hold(ip, NO_QUOTA_CHANGE, NO_QUOTA_CHANGE);
- if (error)
- goto out_alloc;
-
- error = gfs2_rindex_hold(GFS2_SB(&ip->i_inode), &al->al_ri_gh);
- if (error)
- goto out_quota;
-
- error = ea_foreach(ip, ea_dealloc_unstuffed, NULL);
- if (error)
- goto out_rindex;
-
- if (ip->i_diskflags & GFS2_DIF_EA_INDIRECT) {
- error = ea_dealloc_indirect(ip);
- if (error)
- goto out_rindex;
- }
-
- error = ea_dealloc_block(ip);
-
-out_rindex:
- gfs2_glock_dq_uninit(&al->al_ri_gh);
-out_quota:
- gfs2_quota_unhold(ip);
-out_alloc:
- gfs2_alloc_put(ip);
- return error;
-}
-
-static int gfs2_xattr_user_get(struct inode *inode, const char *name,
- void *buffer, size_t size)
-{
- return gfs2_xattr_get(inode, GFS2_EATYPE_USR, name, buffer, size);
-}
-
-static int gfs2_xattr_user_set(struct inode *inode, const char *name,
- const void *value, size_t size, int flags)
-{
- return gfs2_xattr_set(inode, GFS2_EATYPE_USR, name, value, size, flags);
-}
-
-static int gfs2_xattr_system_get(struct inode *inode, const char *name,
- void *buffer, size_t size)
-{
- return gfs2_xattr_get(inode, GFS2_EATYPE_SYS, name, buffer, size);
-}
-
-static int gfs2_xattr_system_set(struct inode *inode, const char *name,
- const void *value, size_t size, int flags)
-{
- return gfs2_xattr_set(inode, GFS2_EATYPE_SYS, name, value, size, flags);
-}
-
-static int gfs2_xattr_security_get(struct inode *inode, const char *name,
- void *buffer, size_t size)
-{
- return gfs2_xattr_get(inode, GFS2_EATYPE_SECURITY, name, buffer, size);
-}
-
-static int gfs2_xattr_security_set(struct inode *inode, const char *name,
- const void *value, size_t size, int flags)
-{
- return gfs2_xattr_set(inode, GFS2_EATYPE_SECURITY, name, value, size, flags);
-}
-
-static struct xattr_handler gfs2_xattr_user_handler = {
- .prefix = XATTR_USER_PREFIX,
- .get = gfs2_xattr_user_get,
- .set = gfs2_xattr_user_set,
-};
-
-static struct xattr_handler gfs2_xattr_security_handler = {
- .prefix = XATTR_SECURITY_PREFIX,
- .get = gfs2_xattr_security_get,
- .set = gfs2_xattr_security_set,
-};
-
-static struct xattr_handler gfs2_xattr_system_handler = {
- .prefix = XATTR_SYSTEM_PREFIX,
- .get = gfs2_xattr_system_get,
- .set = gfs2_xattr_system_set,
-};
-
-struct xattr_handler *gfs2_xattr_handlers[] = {
- &gfs2_xattr_user_handler,
- &gfs2_xattr_security_handler,
- &gfs2_xattr_system_handler,
- NULL,
-};
-
diff --git a/fs/gfs2/eattr.h b/fs/gfs2/eattr.h
deleted file mode 100644
index 4040a18..0000000
--- a/fs/gfs2/eattr.h
+++ /dev/null
@@ -1,72 +0,0 @@
-/*
- * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved.
- * Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved.
- *
- * This copyrighted material is made available to anyone wishing to use,
- * modify, copy, or redistribute it subject to the terms and conditions
- * of the GNU General Public License version 2.
- */
-
-#ifndef __EATTR_DOT_H__
-#define __EATTR_DOT_H__
-
-struct gfs2_inode;
-struct iattr;
-
-#define GFS2_EA_REC_LEN(ea) be32_to_cpu((ea)->ea_rec_len)
-#define GFS2_EA_DATA_LEN(ea) be32_to_cpu((ea)->ea_data_len)
-
-#define GFS2_EA_SIZE(ea) \
-ALIGN(sizeof(struct gfs2_ea_header) + (ea)->ea_name_len + \
- ((GFS2_EA_IS_STUFFED(ea)) ? GFS2_EA_DATA_LEN(ea) : \
- (sizeof(__be64) * (ea)->ea_num_ptrs)), 8)
-
-#define GFS2_EA_IS_STUFFED(ea) (!(ea)->ea_num_ptrs)
-#define GFS2_EA_IS_LAST(ea) ((ea)->ea_flags & GFS2_EAFLAG_LAST)
-
-#define GFS2_EAREQ_SIZE_STUFFED(er) \
-ALIGN(sizeof(struct gfs2_ea_header) + (er)->er_name_len + (er)->er_data_len, 8)
-
-#define GFS2_EA2NAME(ea) ((char *)((struct gfs2_ea_header *)(ea) + 1))
-#define GFS2_EA2DATA(ea) (GFS2_EA2NAME(ea) + (ea)->ea_name_len)
-
-#define GFS2_EA2DATAPTRS(ea) \
-((__be64 *)(GFS2_EA2NAME(ea) + ALIGN((ea)->ea_name_len, 8)))
-
-#define GFS2_EA2NEXT(ea) \
-((struct gfs2_ea_header *)((char *)(ea) + GFS2_EA_REC_LEN(ea)))
-
-#define GFS2_EA_BH2FIRST(bh) \
-((struct gfs2_ea_header *)((bh)->b_data + sizeof(struct gfs2_meta_header)))
-
-struct gfs2_ea_request {
- const char *er_name;
- char *er_data;
- unsigned int er_name_len;
- unsigned int er_data_len;
- unsigned int er_type; /* GFS2_EATYPE_... */
-};
-
-struct gfs2_ea_location {
- struct buffer_head *el_bh;
- struct gfs2_ea_header *el_ea;
- struct gfs2_ea_header *el_prev;
-};
-
-extern int gfs2_xattr_get(struct inode *inode, int type, const char *name,
- void *buffer, size_t size);
-extern int gfs2_xattr_set(struct inode *inode, int type, const char *name,
- const void *value, size_t size, int flags);
-extern ssize_t gfs2_listxattr(struct dentry *dentry, char *buffer, size_t size);
-extern int gfs2_ea_dealloc(struct gfs2_inode *ip);
-
-/* Exported to acl.c */
-
-extern int gfs2_ea_find(struct gfs2_inode *ip, int type, const char *name,
- struct gfs2_ea_location *el);
-extern int gfs2_ea_get_copy(struct gfs2_inode *ip, struct gfs2_ea_location *el,
- char *data, size_t size);
-extern int gfs2_ea_acl_chmod(struct gfs2_inode *ip, struct gfs2_ea_location *el,
- struct iattr *attr, char *data);
-
-#endif /* __EATTR_DOT_H__ */
diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c
index f936d2d..4f5e442 100644
--- a/fs/gfs2/inode.c
+++ b/fs/gfs2/inode.c
@@ -24,7 +24,7 @@
#include "acl.h"
#include "bmap.h"
#include "dir.h"
-#include "eattr.h"
+#include "xattr.h"
#include "glock.h"
#include "glops.h"
#include "inode.h"
diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c
index 1ee0c26..c3ac180 100644
--- a/fs/gfs2/ops_inode.c
+++ b/fs/gfs2/ops_inode.c
@@ -26,7 +26,7 @@
#include "acl.h"
#include "bmap.h"
#include "dir.h"
-#include "eattr.h"
+#include "xattr.h"
#include "glock.h"
#include "inode.h"
#include "meta_io.h"
diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c
index 7a5c128..2e78a3f 100644
--- a/fs/gfs2/super.c
+++ b/fs/gfs2/super.c
@@ -38,7 +38,7 @@
#include "trans.h"
#include "util.h"
#include "sys.h"
-#include "eattr.h"
+#include "xattr.h"

#define args_neq(a1, a2, x) ((a1)->ar_##x != (a2)->ar_##x)

diff --git a/fs/gfs2/xattr.c b/fs/gfs2/xattr.c
new file mode 100644
index 0000000..34ae9ba
--- /dev/null
+++ b/fs/gfs2/xattr.c
@@ -0,0 +1,1558 @@
+/*
+ * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved.
+ * Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved.
+ *
+ * This copyrighted material is made available to anyone wishing to use,
+ * modify, copy, or redistribute it subject to the terms and conditions
+ * of the GNU General Public License version 2.
+ */
+
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <linux/completion.h>
+#include <linux/buffer_head.h>
+#include <linux/xattr.h>
+#include <linux/gfs2_ondisk.h>
+#include <asm/uaccess.h>
+
+#include "gfs2.h"
+#include "incore.h"
+#include "acl.h"
+#include "xattr.h"
+#include "glock.h"
+#include "inode.h"
+#include "meta_io.h"
+#include "quota.h"
+#include "rgrp.h"
+#include "trans.h"
+#include "util.h"
+
+/**
+ * ea_calc_size - returns the acutal number of bytes the request will take up
+ * (not counting any unstuffed data blocks)
+ * @sdp:
+ * @er:
+ * @size:
+ *
+ * Returns: 1 if the EA should be stuffed
+ */
+
+static int ea_calc_size(struct gfs2_sbd *sdp, unsigned int nsize, size_t dsize,
+ unsigned int *size)
+{
+ unsigned int jbsize = sdp->sd_jbsize;
+
+ /* Stuffed */
+ *size = ALIGN(sizeof(struct gfs2_ea_header) + nsize + dsize, 8);
+
+ if (*size <= jbsize)
+ return 1;
+
+ /* Unstuffed */
+ *size = ALIGN(sizeof(struct gfs2_ea_header) + nsize +
+ (sizeof(__be64) * DIV_ROUND_UP(dsize, jbsize)), 8);
+
+ return 0;
+}
+
+static int ea_check_size(struct gfs2_sbd *sdp, unsigned int nsize, size_t dsize)
+{
+ unsigned int size;
+
+ if (dsize > GFS2_EA_MAX_DATA_LEN)
+ return -ERANGE;
+
+ ea_calc_size(sdp, nsize, dsize, &size);
+
+ /* This can only happen with 512 byte blocks */
+ if (size > sdp->sd_jbsize)
+ return -ERANGE;
+
+ return 0;
+}
+
+typedef int (*ea_call_t) (struct gfs2_inode *ip, struct buffer_head *bh,
+ struct gfs2_ea_header *ea,
+ struct gfs2_ea_header *prev, void *private);
+
+static int ea_foreach_i(struct gfs2_inode *ip, struct buffer_head *bh,
+ ea_call_t ea_call, void *data)
+{
+ struct gfs2_ea_header *ea, *prev = NULL;
+ int error = 0;
+
+ if (gfs2_metatype_check(GFS2_SB(&ip->i_inode), bh, GFS2_METATYPE_EA))
+ return -EIO;
+
+ for (ea = GFS2_EA_BH2FIRST(bh);; prev = ea, ea = GFS2_EA2NEXT(ea)) {
+ if (!GFS2_EA_REC_LEN(ea))
+ goto fail;
+ if (!(bh->b_data <= (char *)ea && (char *)GFS2_EA2NEXT(ea) <=
+ bh->b_data + bh->b_size))
+ goto fail;
+ if (!GFS2_EATYPE_VALID(ea->ea_type))
+ goto fail;
+
+ error = ea_call(ip, bh, ea, prev, data);
+ if (error)
+ return error;
+
+ if (GFS2_EA_IS_LAST(ea)) {
+ if ((char *)GFS2_EA2NEXT(ea) !=
+ bh->b_data + bh->b_size)
+ goto fail;
+ break;
+ }
+ }
+
+ return error;
+
+fail:
+ gfs2_consist_inode(ip);
+ return -EIO;
+}
+
+static int ea_foreach(struct gfs2_inode *ip, ea_call_t ea_call, void *data)
+{
+ struct buffer_head *bh, *eabh;
+ __be64 *eablk, *end;
+ int error;
+
+ error = gfs2_meta_read(ip->i_gl, ip->i_eattr, DIO_WAIT, &bh);
+ if (error)
+ return error;
+
+ if (!(ip->i_diskflags & GFS2_DIF_EA_INDIRECT)) {
+ error = ea_foreach_i(ip, bh, ea_call, data);
+ goto out;
+ }
+
+ if (gfs2_metatype_check(GFS2_SB(&ip->i_inode), bh, GFS2_METATYPE_IN)) {
+ error = -EIO;
+ goto out;
+ }
+
+ eablk = (__be64 *)(bh->b_data + sizeof(struct gfs2_meta_header));
+ end = eablk + GFS2_SB(&ip->i_inode)->sd_inptrs;
+
+ for (; eablk < end; eablk++) {
+ u64 bn;
+
+ if (!*eablk)
+ break;
+ bn = be64_to_cpu(*eablk);
+
+ error = gfs2_meta_read(ip->i_gl, bn, DIO_WAIT, &eabh);
+ if (error)
+ break;
+ error = ea_foreach_i(ip, eabh, ea_call, data);
+ brelse(eabh);
+ if (error)
+ break;
+ }
+out:
+ brelse(bh);
+ return error;
+}
+
+struct ea_find {
+ int type;
+ const char *name;
+ size_t namel;
+ struct gfs2_ea_location *ef_el;
+};
+
+static int ea_find_i(struct gfs2_inode *ip, struct buffer_head *bh,
+ struct gfs2_ea_header *ea, struct gfs2_ea_header *prev,
+ void *private)
+{
+ struct ea_find *ef = private;
+
+ if (ea->ea_type == GFS2_EATYPE_UNUSED)
+ return 0;
+
+ if (ea->ea_type == ef->type) {
+ if (ea->ea_name_len == ef->namel &&
+ !memcmp(GFS2_EA2NAME(ea), ef->name, ea->ea_name_len)) {
+ struct gfs2_ea_location *el = ef->ef_el;
+ get_bh(bh);
+ el->el_bh = bh;
+ el->el_ea = ea;
+ el->el_prev = prev;
+ return 1;
+ }
+ }
+
+ return 0;
+}
+
+int gfs2_ea_find(struct gfs2_inode *ip, int type, const char *name,
+ struct gfs2_ea_location *el)
+{
+ struct ea_find ef;
+ int error;
+
+ ef.type = type;
+ ef.name = name;
+ ef.namel = strlen(name);
+ ef.ef_el = el;
+
+ memset(el, 0, sizeof(struct gfs2_ea_location));
+
+ error = ea_foreach(ip, ea_find_i, &ef);
+ if (error > 0)
+ return 0;
+
+ return error;
+}
+
+/**
+ * ea_dealloc_unstuffed -
+ * @ip:
+ * @bh:
+ * @ea:
+ * @prev:
+ * @private:
+ *
+ * Take advantage of the fact that all unstuffed blocks are
+ * allocated from the same RG. But watch, this may not always
+ * be true.
+ *
+ * Returns: errno
+ */
+
+static int ea_dealloc_unstuffed(struct gfs2_inode *ip, struct buffer_head *bh,
+ struct gfs2_ea_header *ea,
+ struct gfs2_ea_header *prev, void *private)
+{
+ int *leave = private;
+ struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
+ struct gfs2_rgrpd *rgd;
+ struct gfs2_holder rg_gh;
+ struct buffer_head *dibh;
+ __be64 *dataptrs;
+ u64 bn = 0;
+ u64 bstart = 0;
+ unsigned int blen = 0;
+ unsigned int blks = 0;
+ unsigned int x;
+ int error;
+
+ if (GFS2_EA_IS_STUFFED(ea))
+ return 0;
+
+ dataptrs = GFS2_EA2DATAPTRS(ea);
+ for (x = 0; x < ea->ea_num_ptrs; x++, dataptrs++) {
+ if (*dataptrs) {
+ blks++;
+ bn = be64_to_cpu(*dataptrs);
+ }
+ }
+ if (!blks)
+ return 0;
+
+ rgd = gfs2_blk2rgrpd(sdp, bn);
+ if (!rgd) {
+ gfs2_consist_inode(ip);
+ return -EIO;
+ }
+
+ error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, &rg_gh);
+ if (error)
+ return error;
+
+ error = gfs2_trans_begin(sdp, rgd->rd_length + RES_DINODE +
+ RES_EATTR + RES_STATFS + RES_QUOTA, blks);
+ if (error)
+ goto out_gunlock;
+
+ gfs2_trans_add_bh(ip->i_gl, bh, 1);
+
+ dataptrs = GFS2_EA2DATAPTRS(ea);
+ for (x = 0; x < ea->ea_num_ptrs; x++, dataptrs++) {
+ if (!*dataptrs)
+ break;
+ bn = be64_to_cpu(*dataptrs);
+
+ if (bstart + blen == bn)
+ blen++;
+ else {
+ if (bstart)
+ gfs2_free_meta(ip, bstart, blen);
+ bstart = bn;
+ blen = 1;
+ }
+
+ *dataptrs = 0;
+ gfs2_add_inode_blocks(&ip->i_inode, -1);
+ }
+ if (bstart)
+ gfs2_free_meta(ip, bstart, blen);
+
+ if (prev && !leave) {
+ u32 len;
+
+ len = GFS2_EA_REC_LEN(prev) + GFS2_EA_REC_LEN(ea);
+ prev->ea_rec_len = cpu_to_be32(len);
+
+ if (GFS2_EA_IS_LAST(ea))
+ prev->ea_flags |= GFS2_EAFLAG_LAST;
+ } else {
+ ea->ea_type = GFS2_EATYPE_UNUSED;
+ ea->ea_num_ptrs = 0;
+ }
+
+ error = gfs2_meta_inode_buffer(ip, &dibh);
+ if (!error) {
+ ip->i_inode.i_ctime = CURRENT_TIME;
+ gfs2_trans_add_bh(ip->i_gl, dibh, 1);
+ gfs2_dinode_out(ip, dibh->b_data);
+ brelse(dibh);
+ }
+
+ gfs2_trans_end(sdp);
+
+out_gunlock:
+ gfs2_glock_dq_uninit(&rg_gh);
+ return error;
+}
+
+static int ea_remove_unstuffed(struct gfs2_inode *ip, struct buffer_head *bh,
+ struct gfs2_ea_header *ea,
+ struct gfs2_ea_header *prev, int leave)
+{
+ struct gfs2_alloc *al;
+ int error;
+
+ al = gfs2_alloc_get(ip);
+ if (!al)
+ return -ENOMEM;
+
+ error = gfs2_quota_hold(ip, NO_QUOTA_CHANGE, NO_QUOTA_CHANGE);
+ if (error)
+ goto out_alloc;
+
+ error = gfs2_rindex_hold(GFS2_SB(&ip->i_inode), &al->al_ri_gh);
+ if (error)
+ goto out_quota;
+
+ error = ea_dealloc_unstuffed(ip, bh, ea, prev, (leave) ? &error : NULL);
+
+ gfs2_glock_dq_uninit(&al->al_ri_gh);
+
+out_quota:
+ gfs2_quota_unhold(ip);
+out_alloc:
+ gfs2_alloc_put(ip);
+ return error;
+}
+
+struct ea_list {
+ struct gfs2_ea_request *ei_er;
+ unsigned int ei_size;
+};
+
+static inline unsigned int gfs2_ea_strlen(struct gfs2_ea_header *ea)
+{
+ switch (ea->ea_type) {
+ case GFS2_EATYPE_USR:
+ return 5 + ea->ea_name_len + 1;
+ case GFS2_EATYPE_SYS:
+ return 7 + ea->ea_name_len + 1;
+ case GFS2_EATYPE_SECURITY:
+ return 9 + ea->ea_name_len + 1;
+ default:
+ return 0;
+ }
+}
+
+static int ea_list_i(struct gfs2_inode *ip, struct buffer_head *bh,
+ struct gfs2_ea_header *ea, struct gfs2_ea_header *prev,
+ void *private)
+{
+ struct ea_list *ei = private;
+ struct gfs2_ea_request *er = ei->ei_er;
+ unsigned int ea_size = gfs2_ea_strlen(ea);
+
+ if (ea->ea_type == GFS2_EATYPE_UNUSED)
+ return 0;
+
+ if (er->er_data_len) {
+ char *prefix = NULL;
+ unsigned int l = 0;
+ char c = 0;
+
+ if (ei->ei_size + ea_size > er->er_data_len)
+ return -ERANGE;
+
+ switch (ea->ea_type) {
+ case GFS2_EATYPE_USR:
+ prefix = "user.";
+ l = 5;
+ break;
+ case GFS2_EATYPE_SYS:
+ prefix = "system.";
+ l = 7;
+ break;
+ case GFS2_EATYPE_SECURITY:
+ prefix = "security.";
+ l = 9;
+ break;
+ }
+
+ BUG_ON(l == 0);
+
+ memcpy(er->er_data + ei->ei_size, prefix, l);
+ memcpy(er->er_data + ei->ei_size + l, GFS2_EA2NAME(ea),
+ ea->ea_name_len);
+ memcpy(er->er_data + ei->ei_size + ea_size - 1, &c, 1);
+ }
+
+ ei->ei_size += ea_size;
+
+ return 0;
+}
+
+/**
+ * gfs2_listxattr - List gfs2 extended attributes
+ * @dentry: The dentry whose inode we are interested in
+ * @buffer: The buffer to write the results
+ * @size: The size of the buffer
+ *
+ * Returns: actual size of data on success, -errno on error
+ */
+
+ssize_t gfs2_listxattr(struct dentry *dentry, char *buffer, size_t size)
+{
+ struct gfs2_inode *ip = GFS2_I(dentry->d_inode);
+ struct gfs2_ea_request er;
+ struct gfs2_holder i_gh;
+ int error;
+
+ memset(&er, 0, sizeof(struct gfs2_ea_request));
+ if (size) {
+ er.er_data = buffer;
+ er.er_data_len = size;
+ }
+
+ error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &i_gh);
+ if (error)
+ return error;
+
+ if (ip->i_eattr) {
+ struct ea_list ei = { .ei_er = &er, .ei_size = 0 };
+
+ error = ea_foreach(ip, ea_list_i, &ei);
+ if (!error)
+ error = ei.ei_size;
+ }
+
+ gfs2_glock_dq_uninit(&i_gh);
+
+ return error;
+}
+
+/**
+ * ea_get_unstuffed - actually copies the unstuffed data into the
+ * request buffer
+ * @ip: The GFS2 inode
+ * @ea: The extended attribute header structure
+ * @data: The data to be copied
+ *
+ * Returns: errno
+ */
+
+static int ea_get_unstuffed(struct gfs2_inode *ip, struct gfs2_ea_header *ea,
+ char *data)
+{
+ struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
+ struct buffer_head **bh;
+ unsigned int amount = GFS2_EA_DATA_LEN(ea);
+ unsigned int nptrs = DIV_ROUND_UP(amount, sdp->sd_jbsize);
+ __be64 *dataptrs = GFS2_EA2DATAPTRS(ea);
+ unsigned int x;
+ int error = 0;
+
+ bh = kcalloc(nptrs, sizeof(struct buffer_head *), GFP_NOFS);
+ if (!bh)
+ return -ENOMEM;
+
+ for (x = 0; x < nptrs; x++) {
+ error = gfs2_meta_read(ip->i_gl, be64_to_cpu(*dataptrs), 0,
+ bh + x);
+ if (error) {
+ while (x--)
+ brelse(bh[x]);
+ goto out;
+ }
+ dataptrs++;
+ }
+
+ for (x = 0; x < nptrs; x++) {
+ error = gfs2_meta_wait(sdp, bh[x]);
+ if (error) {
+ for (; x < nptrs; x++)
+ brelse(bh[x]);
+ goto out;
+ }
+ if (gfs2_metatype_check(sdp, bh[x], GFS2_METATYPE_ED)) {
+ for (; x < nptrs; x++)
+ brelse(bh[x]);
+ error = -EIO;
+ goto out;
+ }
+
+ memcpy(data, bh[x]->b_data + sizeof(struct gfs2_meta_header),
+ (sdp->sd_jbsize > amount) ? amount : sdp->sd_jbsize);
+
+ amount -= sdp->sd_jbsize;
+ data += sdp->sd_jbsize;
+
+ brelse(bh[x]);
+ }
+
+out:
+ kfree(bh);
+ return error;
+}
+
+int gfs2_ea_get_copy(struct gfs2_inode *ip, struct gfs2_ea_location *el,
+ char *data, size_t size)
+{
+ int ret;
+ size_t len = GFS2_EA_DATA_LEN(el->el_ea);
+ if (len > size)
+ return -ERANGE;
+
+ if (GFS2_EA_IS_STUFFED(el->el_ea)) {
+ memcpy(data, GFS2_EA2DATA(el->el_ea), len);
+ return len;
+ }
+ ret = ea_get_unstuffed(ip, el->el_ea, data);
+ if (ret < 0)
+ return ret;
+ return len;
+}
+
+/**
+ * gfs2_xattr_get - Get a GFS2 extended attribute
+ * @inode: The inode
+ * @type: The type of extended attribute
+ * @name: The name of the extended attribute
+ * @buffer: The buffer to write the result into
+ * @size: The size of the buffer
+ *
+ * Returns: actual size of data on success, -errno on error
+ */
+
+int gfs2_xattr_get(struct inode *inode, int type, const char *name,
+ void *buffer, size_t size)
+{
+ struct gfs2_inode *ip = GFS2_I(inode);
+ struct gfs2_ea_location el;
+ int error;
+
+ if (!ip->i_eattr)
+ return -ENODATA;
+ if (strlen(name) > GFS2_EA_MAX_NAME_LEN)
+ return -EINVAL;
+
+ error = gfs2_ea_find(ip, type, name, &el);
+ if (error)
+ return error;
+ if (!el.el_ea)
+ return -ENODATA;
+ if (size)
+ error = gfs2_ea_get_copy(ip, &el, buffer, size);
+ else
+ error = GFS2_EA_DATA_LEN(el.el_ea);
+ brelse(el.el_bh);
+
+ return error;
+}
+
+/**
+ * ea_alloc_blk - allocates a new block for extended attributes.
+ * @ip: A pointer to the inode that's getting extended attributes
+ * @bhp: Pointer to pointer to a struct buffer_head
+ *
+ * Returns: errno
+ */
+
+static int ea_alloc_blk(struct gfs2_inode *ip, struct buffer_head **bhp)
+{
+ struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
+ struct gfs2_ea_header *ea;
+ unsigned int n = 1;
+ u64 block;
+ int error;
+
+ error = gfs2_alloc_block(ip, &block, &n);
+ if (error)
+ return error;
+ gfs2_trans_add_unrevoke(sdp, block, 1);
+ *bhp = gfs2_meta_new(ip->i_gl, block);
+ gfs2_trans_add_bh(ip->i_gl, *bhp, 1);
+ gfs2_metatype_set(*bhp, GFS2_METATYPE_EA, GFS2_FORMAT_EA);
+ gfs2_buffer_clear_tail(*bhp, sizeof(struct gfs2_meta_header));
+
+ ea = GFS2_EA_BH2FIRST(*bhp);
+ ea->ea_rec_len = cpu_to_be32(sdp->sd_jbsize);
+ ea->ea_type = GFS2_EATYPE_UNUSED;
+ ea->ea_flags = GFS2_EAFLAG_LAST;
+ ea->ea_num_ptrs = 0;
+
+ gfs2_add_inode_blocks(&ip->i_inode, 1);
+
+ return 0;
+}
+
+/**
+ * ea_write - writes the request info to an ea, creating new blocks if
+ * necessary
+ * @ip: inode that is being modified
+ * @ea: the location of the new ea in a block
+ * @er: the write request
+ *
+ * Note: does not update ea_rec_len or the GFS2_EAFLAG_LAST bin of ea_flags
+ *
+ * returns : errno
+ */
+
+static int ea_write(struct gfs2_inode *ip, struct gfs2_ea_header *ea,
+ struct gfs2_ea_request *er)
+{
+ struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
+ int error;
+
+ ea->ea_data_len = cpu_to_be32(er->er_data_len);
+ ea->ea_name_len = er->er_name_len;
+ ea->ea_type = er->er_type;
+ ea->__pad = 0;
+
+ memcpy(GFS2_EA2NAME(ea), er->er_name, er->er_name_len);
+
+ if (GFS2_EAREQ_SIZE_STUFFED(er) <= sdp->sd_jbsize) {
+ ea->ea_num_ptrs = 0;
+ memcpy(GFS2_EA2DATA(ea), er->er_data, er->er_data_len);
+ } else {
+ __be64 *dataptr = GFS2_EA2DATAPTRS(ea);
+ const char *data = er->er_data;
+ unsigned int data_len = er->er_data_len;
+ unsigned int copy;
+ unsigned int x;
+
+ ea->ea_num_ptrs = DIV_ROUND_UP(er->er_data_len, sdp->sd_jbsize);
+ for (x = 0; x < ea->ea_num_ptrs; x++) {
+ struct buffer_head *bh;
+ u64 block;
+ int mh_size = sizeof(struct gfs2_meta_header);
+ unsigned int n = 1;
+
+ error = gfs2_alloc_block(ip, &block, &n);
+ if (error)
+ return error;
+ gfs2_trans_add_unrevoke(sdp, block, 1);
+ bh = gfs2_meta_new(ip->i_gl, block);
+ gfs2_trans_add_bh(ip->i_gl, bh, 1);
+ gfs2_metatype_set(bh, GFS2_METATYPE_ED, GFS2_FORMAT_ED);
+
+ gfs2_add_inode_blocks(&ip->i_inode, 1);
+
+ copy = data_len > sdp->sd_jbsize ? sdp->sd_jbsize :
+ data_len;
+ memcpy(bh->b_data + mh_size, data, copy);
+ if (copy < sdp->sd_jbsize)
+ memset(bh->b_data + mh_size + copy, 0,
+ sdp->sd_jbsize - copy);
+
+ *dataptr++ = cpu_to_be64(bh->b_blocknr);
+ data += copy;
+ data_len -= copy;
+
+ brelse(bh);
+ }
+
+ gfs2_assert_withdraw(sdp, !data_len);
+ }
+
+ return 0;
+}
+
+typedef int (*ea_skeleton_call_t) (struct gfs2_inode *ip,
+ struct gfs2_ea_request *er, void *private);
+
+static int ea_alloc_skeleton(struct gfs2_inode *ip, struct gfs2_ea_request *er,
+ unsigned int blks,
+ ea_skeleton_call_t skeleton_call, void *private)
+{
+ struct gfs2_alloc *al;
+ struct buffer_head *dibh;
+ int error;
+
+ al = gfs2_alloc_get(ip);
+ if (!al)
+ return -ENOMEM;
+
+ error = gfs2_quota_lock_check(ip);
+ if (error)
+ goto out;
+
+ al->al_requested = blks;
+
+ error = gfs2_inplace_reserve(ip);
+ if (error)
+ goto out_gunlock_q;
+
+ error = gfs2_trans_begin(GFS2_SB(&ip->i_inode),
+ blks + al->al_rgd->rd_length +
+ RES_DINODE + RES_STATFS + RES_QUOTA, 0);
+ if (error)
+ goto out_ipres;
+
+ error = skeleton_call(ip, er, private);
+ if (error)
+ goto out_end_trans;
+
+ error = gfs2_meta_inode_buffer(ip, &dibh);
+ if (!error) {
+ ip->i_inode.i_ctime = CURRENT_TIME;
+ gfs2_trans_add_bh(ip->i_gl, dibh, 1);
+ gfs2_dinode_out(ip, dibh->b_data);
+ brelse(dibh);
+ }
+
+out_end_trans:
+ gfs2_trans_end(GFS2_SB(&ip->i_inode));
+out_ipres:
+ gfs2_inplace_release(ip);
+out_gunlock_q:
+ gfs2_quota_unlock(ip);
+out:
+ gfs2_alloc_put(ip);
+ return error;
+}
+
+static int ea_init_i(struct gfs2_inode *ip, struct gfs2_ea_request *er,
+ void *private)
+{
+ struct buffer_head *bh;
+ int error;
+
+ error = ea_alloc_blk(ip, &bh);
+ if (error)
+ return error;
+
+ ip->i_eattr = bh->b_blocknr;
+ error = ea_write(ip, GFS2_EA_BH2FIRST(bh), er);
+
+ brelse(bh);
+
+ return error;
+}
+
+/**
+ * ea_init - initializes a new eattr block
+ * @ip:
+ * @er:
+ *
+ * Returns: errno
+ */
+
+static int ea_init(struct gfs2_inode *ip, int type, const char *name,
+ const void *data, size_t size)
+{
+ struct gfs2_ea_request er;
+ unsigned int jbsize = GFS2_SB(&ip->i_inode)->sd_jbsize;
+ unsigned int blks = 1;
+
+ er.er_type = type;
+ er.er_name = name;
+ er.er_name_len = strlen(name);
+ er.er_data = (void *)data;
+ er.er_data_len = size;
+
+ if (GFS2_EAREQ_SIZE_STUFFED(&er) > jbsize)
+ blks += DIV_ROUND_UP(er.er_data_len, jbsize);
+
+ return ea_alloc_skeleton(ip, &er, blks, ea_init_i, NULL);
+}
+
+static struct gfs2_ea_header *ea_split_ea(struct gfs2_ea_header *ea)
+{
+ u32 ea_size = GFS2_EA_SIZE(ea);
+ struct gfs2_ea_header *new = (struct gfs2_ea_header *)((char *)ea +
+ ea_size);
+ u32 new_size = GFS2_EA_REC_LEN(ea) - ea_size;
+ int last = ea->ea_flags & GFS2_EAFLAG_LAST;
+
+ ea->ea_rec_len = cpu_to_be32(ea_size);
+ ea->ea_flags ^= last;
+
+ new->ea_rec_len = cpu_to_be32(new_size);
+ new->ea_flags = last;
+
+ return new;
+}
+
+static void ea_set_remove_stuffed(struct gfs2_inode *ip,
+ struct gfs2_ea_location *el)
+{
+ struct gfs2_ea_header *ea = el->el_ea;
+ struct gfs2_ea_header *prev = el->el_prev;
+ u32 len;
+
+ gfs2_trans_add_bh(ip->i_gl, el->el_bh, 1);
+
+ if (!prev || !GFS2_EA_IS_STUFFED(ea)) {
+ ea->ea_type = GFS2_EATYPE_UNUSED;
+ return;
+ } else if (GFS2_EA2NEXT(prev) != ea) {
+ prev = GFS2_EA2NEXT(prev);
+ gfs2_assert_withdraw(GFS2_SB(&ip->i_inode), GFS2_EA2NEXT(prev) == ea);
+ }
+
+ len = GFS2_EA_REC_LEN(prev) + GFS2_EA_REC_LEN(ea);
+ prev->ea_rec_len = cpu_to_be32(len);
+
+ if (GFS2_EA_IS_LAST(ea))
+ prev->ea_flags |= GFS2_EAFLAG_LAST;
+}
+
+struct ea_set {
+ int ea_split;
+
+ struct gfs2_ea_request *es_er;
+ struct gfs2_ea_location *es_el;
+
+ struct buffer_head *es_bh;
+ struct gfs2_ea_header *es_ea;
+};
+
+static int ea_set_simple_noalloc(struct gfs2_inode *ip, struct buffer_head *bh,
+ struct gfs2_ea_header *ea, struct ea_set *es)
+{
+ struct gfs2_ea_request *er = es->es_er;
+ struct buffer_head *dibh;
+ int error;
+
+ error = gfs2_trans_begin(GFS2_SB(&ip->i_inode), RES_DINODE + 2 * RES_EATTR, 0);
+ if (error)
+ return error;
+
+ gfs2_trans_add_bh(ip->i_gl, bh, 1);
+
+ if (es->ea_split)
+ ea = ea_split_ea(ea);
+
+ ea_write(ip, ea, er);
+
+ if (es->es_el)
+ ea_set_remove_stuffed(ip, es->es_el);
+
+ error = gfs2_meta_inode_buffer(ip, &dibh);
+ if (error)
+ goto out;
+ ip->i_inode.i_ctime = CURRENT_TIME;
+ gfs2_trans_add_bh(ip->i_gl, dibh, 1);
+ gfs2_dinode_out(ip, dibh->b_data);
+ brelse(dibh);
+out:
+ gfs2_trans_end(GFS2_SB(&ip->i_inode));
+ return error;
+}
+
+static int ea_set_simple_alloc(struct gfs2_inode *ip,
+ struct gfs2_ea_request *er, void *private)
+{
+ struct ea_set *es = private;
+ struct gfs2_ea_header *ea = es->es_ea;
+ int error;
+
+ gfs2_trans_add_bh(ip->i_gl, es->es_bh, 1);
+
+ if (es->ea_split)
+ ea = ea_split_ea(ea);
+
+ error = ea_write(ip, ea, er);
+ if (error)
+ return error;
+
+ if (es->es_el)
+ ea_set_remove_stuffed(ip, es->es_el);
+
+ return 0;
+}
+
+static int ea_set_simple(struct gfs2_inode *ip, struct buffer_head *bh,
+ struct gfs2_ea_header *ea, struct gfs2_ea_header *prev,
+ void *private)
+{
+ struct ea_set *es = private;
+ unsigned int size;
+ int stuffed;
+ int error;
+
+ stuffed = ea_calc_size(GFS2_SB(&ip->i_inode), es->es_er->er_name_len,
+ es->es_er->er_data_len, &size);
+
+ if (ea->ea_type == GFS2_EATYPE_UNUSED) {
+ if (GFS2_EA_REC_LEN(ea) < size)
+ return 0;
+ if (!GFS2_EA_IS_STUFFED(ea)) {
+ error = ea_remove_unstuffed(ip, bh, ea, prev, 1);
+ if (error)
+ return error;
+ }
+ es->ea_split = 0;
+ } else if (GFS2_EA_REC_LEN(ea) - GFS2_EA_SIZE(ea) >= size)
+ es->ea_split = 1;
+ else
+ return 0;
+
+ if (stuffed) {
+ error = ea_set_simple_noalloc(ip, bh, ea, es);
+ if (error)
+ return error;
+ } else {
+ unsigned int blks;
+
+ es->es_bh = bh;
+ es->es_ea = ea;
+ blks = 2 + DIV_ROUND_UP(es->es_er->er_data_len,
+ GFS2_SB(&ip->i_inode)->sd_jbsize);
+
+ error = ea_alloc_skeleton(ip, es->es_er, blks,
+ ea_set_simple_alloc, es);
+ if (error)
+ return error;
+ }
+
+ return 1;
+}
+
+static int ea_set_block(struct gfs2_inode *ip, struct gfs2_ea_request *er,
+ void *private)
+{
+ struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
+ struct buffer_head *indbh, *newbh;
+ __be64 *eablk;
+ int error;
+ int mh_size = sizeof(struct gfs2_meta_header);
+
+ if (ip->i_diskflags & GFS2_DIF_EA_INDIRECT) {
+ __be64 *end;
+
+ error = gfs2_meta_read(ip->i_gl, ip->i_eattr, DIO_WAIT,
+ &indbh);
+ if (error)
+ return error;
+
+ if (gfs2_metatype_check(sdp, indbh, GFS2_METATYPE_IN)) {
+ error = -EIO;
+ goto out;
+ }
+
+ eablk = (__be64 *)(indbh->b_data + mh_size);
+ end = eablk + sdp->sd_inptrs;
+
+ for (; eablk < end; eablk++)
+ if (!*eablk)
+ break;
+
+ if (eablk == end) {
+ error = -ENOSPC;
+ goto out;
+ }
+
+ gfs2_trans_add_bh(ip->i_gl, indbh, 1);
+ } else {
+ u64 blk;
+ unsigned int n = 1;
+ error = gfs2_alloc_block(ip, &blk, &n);
+ if (error)
+ return error;
+ gfs2_trans_add_unrevoke(sdp, blk, 1);
+ indbh = gfs2_meta_new(ip->i_gl, blk);
+ gfs2_trans_add_bh(ip->i_gl, indbh, 1);
+ gfs2_metatype_set(indbh, GFS2_METATYPE_IN, GFS2_FORMAT_IN);
+ gfs2_buffer_clear_tail(indbh, mh_size);
+
+ eablk = (__be64 *)(indbh->b_data + mh_size);
+ *eablk = cpu_to_be64(ip->i_eattr);
+ ip->i_eattr = blk;
+ ip->i_diskflags |= GFS2_DIF_EA_INDIRECT;
+ gfs2_add_inode_blocks(&ip->i_inode, 1);
+
+ eablk++;
+ }
+
+ error = ea_alloc_blk(ip, &newbh);
+ if (error)
+ goto out;
+
+ *eablk = cpu_to_be64((u64)newbh->b_blocknr);
+ error = ea_write(ip, GFS2_EA_BH2FIRST(newbh), er);
+ brelse(newbh);
+ if (error)
+ goto out;
+
+ if (private)
+ ea_set_remove_stuffed(ip, private);
+
+out:
+ brelse(indbh);
+ return error;
+}
+
+static int ea_set_i(struct gfs2_inode *ip, int type, const char *name,
+ const void *value, size_t size, struct gfs2_ea_location *el)
+{
+ struct gfs2_ea_request er;
+ struct ea_set es;
+ unsigned int blks = 2;
+ int error;
+
+ er.er_type = type;
+ er.er_name = name;
+ er.er_data = (void *)value;
+ er.er_name_len = strlen(name);
+ er.er_data_len = size;
+
+ memset(&es, 0, sizeof(struct ea_set));
+ es.es_er = &er;
+ es.es_el = el;
+
+ error = ea_foreach(ip, ea_set_simple, &es);
+ if (error > 0)
+ return 0;
+ if (error)
+ return error;
+
+ if (!(ip->i_diskflags & GFS2_DIF_EA_INDIRECT))
+ blks++;
+ if (GFS2_EAREQ_SIZE_STUFFED(&er) > GFS2_SB(&ip->i_inode)->sd_jbsize)
+ blks += DIV_ROUND_UP(er.er_data_len, GFS2_SB(&ip->i_inode)->sd_jbsize);
+
+ return ea_alloc_skeleton(ip, &er, blks, ea_set_block, el);
+}
+
+static int ea_set_remove_unstuffed(struct gfs2_inode *ip,
+ struct gfs2_ea_location *el)
+{
+ if (el->el_prev && GFS2_EA2NEXT(el->el_prev) != el->el_ea) {
+ el->el_prev = GFS2_EA2NEXT(el->el_prev);
+ gfs2_assert_withdraw(GFS2_SB(&ip->i_inode),
+ GFS2_EA2NEXT(el->el_prev) == el->el_ea);
+ }
+
+ return ea_remove_unstuffed(ip, el->el_bh, el->el_ea, el->el_prev,0);
+}
+
+static int ea_remove_stuffed(struct gfs2_inode *ip, struct gfs2_ea_location *el)
+{
+ struct gfs2_ea_header *ea = el->el_ea;
+ struct gfs2_ea_header *prev = el->el_prev;
+ struct buffer_head *dibh;
+ int error;
+
+ error = gfs2_trans_begin(GFS2_SB(&ip->i_inode), RES_DINODE + RES_EATTR, 0);
+ if (error)
+ return error;
+
+ gfs2_trans_add_bh(ip->i_gl, el->el_bh, 1);
+
+ if (prev) {
+ u32 len;
+
+ len = GFS2_EA_REC_LEN(prev) + GFS2_EA_REC_LEN(ea);
+ prev->ea_rec_len = cpu_to_be32(len);
+
+ if (GFS2_EA_IS_LAST(ea))
+ prev->ea_flags |= GFS2_EAFLAG_LAST;
+ } else {
+ ea->ea_type = GFS2_EATYPE_UNUSED;
+ }
+
+ error = gfs2_meta_inode_buffer(ip, &dibh);
+ if (!error) {
+ ip->i_inode.i_ctime = CURRENT_TIME;
+ gfs2_trans_add_bh(ip->i_gl, dibh, 1);
+ gfs2_dinode_out(ip, dibh->b_data);
+ brelse(dibh);
+ }
+
+ gfs2_trans_end(GFS2_SB(&ip->i_inode));
+
+ return error;
+}
+
+/**
+ * gfs2_xattr_remove - Remove a GFS2 extended attribute
+ * @inode: The inode
+ * @type: The type of the extended attribute
+ * @name: The name of the extended attribute
+ *
+ * This is not called directly by the VFS since we use the (common)
+ * scheme of making a "set with NULL data" mean a remove request. Note
+ * that this is different from a set with zero length data.
+ *
+ * Returns: 0, or errno on failure
+ */
+
+static int gfs2_xattr_remove(struct inode *inode, int type, const char *name)
+{
+ struct gfs2_inode *ip = GFS2_I(inode);
+ struct gfs2_ea_location el;
+ int error;
+
+ if (!ip->i_eattr)
+ return -ENODATA;
+
+ error = gfs2_ea_find(ip, type, name, &el);
+ if (error)
+ return error;
+ if (!el.el_ea)
+ return -ENODATA;
+
+ if (GFS2_EA_IS_STUFFED(el.el_ea))
+ error = ea_remove_stuffed(ip, &el);
+ else
+ error = ea_remove_unstuffed(ip, el.el_bh, el.el_ea, el.el_prev, 0);
+
+ brelse(el.el_bh);
+
+ return error;
+}
+
+/**
+ * gfs2_xattr_set - Set (or remove) a GFS2 extended attribute
+ * @inode: The inode
+ * @type: The type of the extended attribute
+ * @name: The name of the extended attribute
+ * @value: The value of the extended attribute (NULL for remove)
+ * @size: The size of the @value argument
+ * @flags: Create or Replace
+ *
+ * See gfs2_xattr_remove() for details of the removal of xattrs.
+ *
+ * Returns: 0 or errno on failure
+ */
+
+int gfs2_xattr_set(struct inode *inode, int type, const char *name,
+ const void *value, size_t size, int flags)
+{
+ struct gfs2_sbd *sdp = GFS2_SB(inode);
+ struct gfs2_inode *ip = GFS2_I(inode);
+ struct gfs2_ea_location el;
+ unsigned int namel = strlen(name);
+ int error;
+
+ if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
+ return -EPERM;
+ if (namel > GFS2_EA_MAX_NAME_LEN)
+ return -ERANGE;
+
+ if (value == NULL)
+ return gfs2_xattr_remove(inode, type, name);
+
+ if (ea_check_size(sdp, namel, size))
+ return -ERANGE;
+
+ if (!ip->i_eattr) {
+ if (flags & XATTR_REPLACE)
+ return -ENODATA;
+ return ea_init(ip, type, name, value, size);
+ }
+
+ error = gfs2_ea_find(ip, type, name, &el);
+ if (error)
+ return error;
+
+ if (el.el_ea) {
+ if (ip->i_diskflags & GFS2_DIF_APPENDONLY) {
+ brelse(el.el_bh);
+ return -EPERM;
+ }
+
+ error = -EEXIST;
+ if (!(flags & XATTR_CREATE)) {
+ int unstuffed = !GFS2_EA_IS_STUFFED(el.el_ea);
+ error = ea_set_i(ip, type, name, value, size, &el);
+ if (!error && unstuffed)
+ ea_set_remove_unstuffed(ip, &el);
+ }
+
+ brelse(el.el_bh);
+ return error;
+ }
+
+ error = -ENODATA;
+ if (!(flags & XATTR_REPLACE))
+ error = ea_set_i(ip, type, name, value, size, NULL);
+
+ return error;
+}
+
+static int ea_acl_chmod_unstuffed(struct gfs2_inode *ip,
+ struct gfs2_ea_header *ea, char *data)
+{
+ struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
+ struct buffer_head **bh;
+ unsigned int amount = GFS2_EA_DATA_LEN(ea);
+ unsigned int nptrs = DIV_ROUND_UP(amount, sdp->sd_jbsize);
+ __be64 *dataptrs = GFS2_EA2DATAPTRS(ea);
+ unsigned int x;
+ int error;
+
+ bh = kcalloc(nptrs, sizeof(struct buffer_head *), GFP_NOFS);
+ if (!bh)
+ return -ENOMEM;
+
+ error = gfs2_trans_begin(sdp, nptrs + RES_DINODE, 0);
+ if (error)
+ goto out;
+
+ for (x = 0; x < nptrs; x++) {
+ error = gfs2_meta_read(ip->i_gl, be64_to_cpu(*dataptrs), 0,
+ bh + x);
+ if (error) {
+ while (x--)
+ brelse(bh[x]);
+ goto fail;
+ }
+ dataptrs++;
+ }
+
+ for (x = 0; x < nptrs; x++) {
+ error = gfs2_meta_wait(sdp, bh[x]);
+ if (error) {
+ for (; x < nptrs; x++)
+ brelse(bh[x]);
+ goto fail;
+ }
+ if (gfs2_metatype_check(sdp, bh[x], GFS2_METATYPE_ED)) {
+ for (; x < nptrs; x++)
+ brelse(bh[x]);
+ error = -EIO;
+ goto fail;
+ }
+
+ gfs2_trans_add_bh(ip->i_gl, bh[x], 1);
+
+ memcpy(bh[x]->b_data + sizeof(struct gfs2_meta_header), data,
+ (sdp->sd_jbsize > amount) ? amount : sdp->sd_jbsize);
+
+ amount -= sdp->sd_jbsize;
+ data += sdp->sd_jbsize;
+
+ brelse(bh[x]);
+ }
+
+out:
+ kfree(bh);
+ return error;
+
+fail:
+ gfs2_trans_end(sdp);
+ kfree(bh);
+ return error;
+}
+
+int gfs2_ea_acl_chmod(struct gfs2_inode *ip, struct gfs2_ea_location *el,
+ struct iattr *attr, char *data)
+{
+ struct buffer_head *dibh;
+ int error;
+
+ if (GFS2_EA_IS_STUFFED(el->el_ea)) {
+ error = gfs2_trans_begin(GFS2_SB(&ip->i_inode), RES_DINODE + RES_EATTR, 0);
+ if (error)
+ return error;
+
+ gfs2_trans_add_bh(ip->i_gl, el->el_bh, 1);
+ memcpy(GFS2_EA2DATA(el->el_ea), data,
+ GFS2_EA_DATA_LEN(el->el_ea));
+ } else
+ error = ea_acl_chmod_unstuffed(ip, el->el_ea, data);
+
+ if (error)
+ return error;
+
+ error = gfs2_meta_inode_buffer(ip, &dibh);
+ if (!error) {
+ error = inode_setattr(&ip->i_inode, attr);
+ gfs2_assert_warn(GFS2_SB(&ip->i_inode), !error);
+ gfs2_trans_add_bh(ip->i_gl, dibh, 1);
+ gfs2_dinode_out(ip, dibh->b_data);
+ brelse(dibh);
+ }
+
+ gfs2_trans_end(GFS2_SB(&ip->i_inode));
+
+ return error;
+}
+
+static int ea_dealloc_indirect(struct gfs2_inode *ip)
+{
+ struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
+ struct gfs2_rgrp_list rlist;
+ struct buffer_head *indbh, *dibh;
+ __be64 *eablk, *end;
+ unsigned int rg_blocks = 0;
+ u64 bstart = 0;
+ unsigned int blen = 0;
+ unsigned int blks = 0;
+ unsigned int x;
+ int error;
+
+ memset(&rlist, 0, sizeof(struct gfs2_rgrp_list));
+
+ error = gfs2_meta_read(ip->i_gl, ip->i_eattr, DIO_WAIT, &indbh);
+ if (error)
+ return error;
+
+ if (gfs2_metatype_check(sdp, indbh, GFS2_METATYPE_IN)) {
+ error = -EIO;
+ goto out;
+ }
+
+ eablk = (__be64 *)(indbh->b_data + sizeof(struct gfs2_meta_header));
+ end = eablk + sdp->sd_inptrs;
+
+ for (; eablk < end; eablk++) {
+ u64 bn;
+
+ if (!*eablk)
+ break;
+ bn = be64_to_cpu(*eablk);
+
+ if (bstart + blen == bn)
+ blen++;
+ else {
+ if (bstart)
+ gfs2_rlist_add(sdp, &rlist, bstart);
+ bstart = bn;
+ blen = 1;
+ }
+ blks++;
+ }
+ if (bstart)
+ gfs2_rlist_add(sdp, &rlist, bstart);
+ else
+ goto out;
+
+ gfs2_rlist_alloc(&rlist, LM_ST_EXCLUSIVE);
+
+ for (x = 0; x < rlist.rl_rgrps; x++) {
+ struct gfs2_rgrpd *rgd;
+ rgd = rlist.rl_ghs[x].gh_gl->gl_object;
+ rg_blocks += rgd->rd_length;
+ }
+
+ error = gfs2_glock_nq_m(rlist.rl_rgrps, rlist.rl_ghs);
+ if (error)
+ goto out_rlist_free;
+
+ error = gfs2_trans_begin(sdp, rg_blocks + RES_DINODE + RES_INDIRECT +
+ RES_STATFS + RES_QUOTA, blks);
+ if (error)
+ goto out_gunlock;
+
+ gfs2_trans_add_bh(ip->i_gl, indbh, 1);
+
+ eablk = (__be64 *)(indbh->b_data + sizeof(struct gfs2_meta_header));
+ bstart = 0;
+ blen = 0;
+
+ for (; eablk < end; eablk++) {
+ u64 bn;
+
+ if (!*eablk)
+ break;
+ bn = be64_to_cpu(*eablk);
+
+ if (bstart + blen == bn)
+ blen++;
+ else {
+ if (bstart)
+ gfs2_free_meta(ip, bstart, blen);
+ bstart = bn;
+ blen = 1;
+ }
+
+ *eablk = 0;
+ gfs2_add_inode_blocks(&ip->i_inode, -1);
+ }
+ if (bstart)
+ gfs2_free_meta(ip, bstart, blen);
+
+ ip->i_diskflags &= ~GFS2_DIF_EA_INDIRECT;
+
+ error = gfs2_meta_inode_buffer(ip, &dibh);
+ if (!error) {
+ gfs2_trans_add_bh(ip->i_gl, dibh, 1);
+ gfs2_dinode_out(ip, dibh->b_data);
+ brelse(dibh);
+ }
+
+ gfs2_trans_end(sdp);
+
+out_gunlock:
+ gfs2_glock_dq_m(rlist.rl_rgrps, rlist.rl_ghs);
+out_rlist_free:
+ gfs2_rlist_free(&rlist);
+out:
+ brelse(indbh);
+ return error;
+}
+
+static int ea_dealloc_block(struct gfs2_inode *ip)
+{
+ struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
+ struct gfs2_alloc *al = ip->i_alloc;
+ struct gfs2_rgrpd *rgd;
+ struct buffer_head *dibh;
+ int error;
+
+ rgd = gfs2_blk2rgrpd(sdp, ip->i_eattr);
+ if (!rgd) {
+ gfs2_consist_inode(ip);
+ return -EIO;
+ }
+
+ error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0,
+ &al->al_rgd_gh);
+ if (error)
+ return error;
+
+ error = gfs2_trans_begin(sdp, RES_RG_BIT + RES_DINODE + RES_STATFS +
+ RES_QUOTA, 1);
+ if (error)
+ goto out_gunlock;
+
+ gfs2_free_meta(ip, ip->i_eattr, 1);
+
+ ip->i_eattr = 0;
+ gfs2_add_inode_blocks(&ip->i_inode, -1);
+
+ error = gfs2_meta_inode_buffer(ip, &dibh);
+ if (!error) {
+ gfs2_trans_add_bh(ip->i_gl, dibh, 1);
+ gfs2_dinode_out(ip, dibh->b_data);
+ brelse(dibh);
+ }
+
+ gfs2_trans_end(sdp);
+
+out_gunlock:
+ gfs2_glock_dq_uninit(&al->al_rgd_gh);
+ return error;
+}
+
+/**
+ * gfs2_ea_dealloc - deallocate the extended attribute fork
+ * @ip: the inode
+ *
+ * Returns: errno
+ */
+
+int gfs2_ea_dealloc(struct gfs2_inode *ip)
+{
+ struct gfs2_alloc *al;
+ int error;
+
+ al = gfs2_alloc_get(ip);
+ if (!al)
+ return -ENOMEM;
+
+ error = gfs2_quota_hold(ip, NO_QUOTA_CHANGE, NO_QUOTA_CHANGE);
+ if (error)
+ goto out_alloc;
+
+ error = gfs2_rindex_hold(GFS2_SB(&ip->i_inode), &al->al_ri_gh);
+ if (error)
+ goto out_quota;
+
+ error = ea_foreach(ip, ea_dealloc_unstuffed, NULL);
+ if (error)
+ goto out_rindex;
+
+ if (ip->i_diskflags & GFS2_DIF_EA_INDIRECT) {
+ error = ea_dealloc_indirect(ip);
+ if (error)
+ goto out_rindex;
+ }
+
+ error = ea_dealloc_block(ip);
+
+out_rindex:
+ gfs2_glock_dq_uninit(&al->al_ri_gh);
+out_quota:
+ gfs2_quota_unhold(ip);
+out_alloc:
+ gfs2_alloc_put(ip);
+ return error;
+}
+
+static int gfs2_xattr_user_get(struct inode *inode, const char *name,
+ void *buffer, size_t size)
+{
+ return gfs2_xattr_get(inode, GFS2_EATYPE_USR, name, buffer, size);
+}
+
+static int gfs2_xattr_user_set(struct inode *inode, const char *name,
+ const void *value, size_t size, int flags)
+{
+ return gfs2_xattr_set(inode, GFS2_EATYPE_USR, name, value, size, flags);
+}
+
+static int gfs2_xattr_system_get(struct inode *inode, const char *name,
+ void *buffer, size_t size)
+{
+ return gfs2_xattr_get(inode, GFS2_EATYPE_SYS, name, buffer, size);
+}
+
+static int gfs2_xattr_system_set(struct inode *inode, const char *name,
+ const void *value, size_t size, int flags)
+{
+ return gfs2_xattr_set(inode, GFS2_EATYPE_SYS, name, value, size, flags);
+}
+
+static int gfs2_xattr_security_get(struct inode *inode, const char *name,
+ void *buffer, size_t size)
+{
+ return gfs2_xattr_get(inode, GFS2_EATYPE_SECURITY, name, buffer, size);
+}
+
+static int gfs2_xattr_security_set(struct inode *inode, const char *name,
+ const void *value, size_t size, int flags)
+{
+ return gfs2_xattr_set(inode, GFS2_EATYPE_SECURITY, name, value, size, flags);
+}
+
+static struct xattr_handler gfs2_xattr_user_handler = {
+ .prefix = XATTR_USER_PREFIX,
+ .get = gfs2_xattr_user_get,
+ .set = gfs2_xattr_user_set,
+};
+
+static struct xattr_handler gfs2_xattr_security_handler = {
+ .prefix = XATTR_SECURITY_PREFIX,
+ .get = gfs2_xattr_security_get,
+ .set = gfs2_xattr_security_set,
+};
+
+static struct xattr_handler gfs2_xattr_system_handler = {
+ .prefix = XATTR_SYSTEM_PREFIX,
+ .get = gfs2_xattr_system_get,
+ .set = gfs2_xattr_system_set,
+};
+
+struct xattr_handler *gfs2_xattr_handlers[] = {
+ &gfs2_xattr_user_handler,
+ &gfs2_xattr_security_handler,
+ &gfs2_xattr_system_handler,
+ NULL,
+};
+
diff --git a/fs/gfs2/xattr.h b/fs/gfs2/xattr.h
new file mode 100644
index 0000000..4040a18
--- /dev/null
+++ b/fs/gfs2/xattr.h
@@ -0,0 +1,72 @@
+/*
+ * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved.
+ * Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved.
+ *
+ * This copyrighted material is made available to anyone wishing to use,
+ * modify, copy, or redistribute it subject to the terms and conditions
+ * of the GNU General Public License version 2.
+ */
+
+#ifndef __EATTR_DOT_H__
+#define __EATTR_DOT_H__
+
+struct gfs2_inode;
+struct iattr;
+
+#define GFS2_EA_REC_LEN(ea) be32_to_cpu((ea)->ea_rec_len)
+#define GFS2_EA_DATA_LEN(ea) be32_to_cpu((ea)->ea_data_len)
+
+#define GFS2_EA_SIZE(ea) \
+ALIGN(sizeof(struct gfs2_ea_header) + (ea)->ea_name_len + \
+ ((GFS2_EA_IS_STUFFED(ea)) ? GFS2_EA_DATA_LEN(ea) : \
+ (sizeof(__be64) * (ea)->ea_num_ptrs)), 8)
+
+#define GFS2_EA_IS_STUFFED(ea) (!(ea)->ea_num_ptrs)
+#define GFS2_EA_IS_LAST(ea) ((ea)->ea_flags & GFS2_EAFLAG_LAST)
+
+#define GFS2_EAREQ_SIZE_STUFFED(er) \
+ALIGN(sizeof(struct gfs2_ea_header) + (er)->er_name_len + (er)->er_data_len, 8)
+
+#define GFS2_EA2NAME(ea) ((char *)((struct gfs2_ea_header *)(ea) + 1))
+#define GFS2_EA2DATA(ea) (GFS2_EA2NAME(ea) + (ea)->ea_name_len)
+
+#define GFS2_EA2DATAPTRS(ea) \
+((__be64 *)(GFS2_EA2NAME(ea) + ALIGN((ea)->ea_name_len, 8)))
+
+#define GFS2_EA2NEXT(ea) \
+((struct gfs2_ea_header *)((char *)(ea) + GFS2_EA_REC_LEN(ea)))
+
+#define GFS2_EA_BH2FIRST(bh) \
+((struct gfs2_ea_header *)((bh)->b_data + sizeof(struct gfs2_meta_header)))
+
+struct gfs2_ea_request {
+ const char *er_name;
+ char *er_data;
+ unsigned int er_name_len;
+ unsigned int er_data_len;
+ unsigned int er_type; /* GFS2_EATYPE_... */
+};
+
+struct gfs2_ea_location {
+ struct buffer_head *el_bh;
+ struct gfs2_ea_header *el_ea;
+ struct gfs2_ea_header *el_prev;
+};
+
+extern int gfs2_xattr_get(struct inode *inode, int type, const char *name,
+ void *buffer, size_t size);
+extern int gfs2_xattr_set(struct inode *inode, int type, const char *name,
+ const void *value, size_t size, int flags);
+extern ssize_t gfs2_listxattr(struct dentry *dentry, char *buffer, size_t size);
+extern int gfs2_ea_dealloc(struct gfs2_inode *ip);
+
+/* Exported to acl.c */
+
+extern int gfs2_ea_find(struct gfs2_inode *ip, int type, const char *name,
+ struct gfs2_ea_location *el);
+extern int gfs2_ea_get_copy(struct gfs2_inode *ip, struct gfs2_ea_location *el,
+ char *data, size_t size);
+extern int gfs2_ea_acl_chmod(struct gfs2_inode *ip, struct gfs2_ea_location *el,
+ struct iattr *attr, char *data);
+
+#endif /* __EATTR_DOT_H__ */
--
1.6.2.5

2009-09-10 10:31:01

by Steven Whitehouse

[permalink] [raw]
Subject: [PATCH 13/15] GFS2: Remove no_formal_ino generating code

The inum structure used throughout GFS2 has two fields. One
no_addr is the disk block number of the inode in question and
is used everywhere as the inode number. The other, no_formal_ino,
is used only as the generation number for NFS.

Historically the no_formal_ino field was set using a complicated
system of one global and one per-node file containing inode numbers
in order to ensure that each no_formal_ino was unique. Also this
code made no provision for what would happen when eventually the
(64 bit) numbers ran out. Now I know that is pretty unlikely to
happen given the large space of numbers, but it is possible
nevertheless.

The only guarantee required for no_formal_ino is that, for any
single inode, the same number doesn't get reused too quickly.

We already have a generation number which is kept in the inode
and initialised from a counter in the resource group (almost
no overhead, since we have to touch the resource group anyway
in order to allocate an inode in the first place). Aside from
ensuring that we never use the value 0 in the no_formal_ino
field, we can use that counter directly.

As a result of that change, we lose about 200 lines of code and
also gain about 10 creates/sec on the postmark benchmark (on
my test machine).

Signed-off-by: Steven Whitehouse <[email protected]>
---
fs/gfs2/incore.h | 7 ---
fs/gfs2/inode.c | 143 +------------------------------------------------
fs/gfs2/ops_fstype.c | 45 ++--------------
fs/gfs2/rgrp.c | 2 +
fs/gfs2/super.c | 3 -
5 files changed, 10 insertions(+), 190 deletions(-)

diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h
index 1d11e6e..52436ab 100644
--- a/fs/gfs2/incore.h
+++ b/fs/gfs2/incore.h
@@ -548,18 +548,12 @@ struct gfs2_sbd {
struct dentry *sd_root_dir;

struct inode *sd_jindex;
- struct inode *sd_inum_inode;
struct inode *sd_statfs_inode;
- struct inode *sd_ir_inode;
struct inode *sd_sc_inode;
struct inode *sd_qc_inode;
struct inode *sd_rindex;
struct inode *sd_quota_inode;

- /* Inum stuff */
-
- struct mutex sd_inum_mutex;
-
/* StatFS stuff */

spinlock_t sd_statfs_spin;
@@ -587,7 +581,6 @@ struct gfs2_sbd {
struct gfs2_holder sd_journal_gh;
struct gfs2_holder sd_jinode_gh;

- struct gfs2_holder sd_ir_gh;
struct gfs2_holder sd_sc_gh;
struct gfs2_holder sd_qc_gh;

diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c
index 4f5e442..fb15d3b 100644
--- a/fs/gfs2/inode.c
+++ b/fs/gfs2/inode.c
@@ -519,139 +519,6 @@ out:
return inode ? inode : ERR_PTR(error);
}

-static void gfs2_inum_range_in(struct gfs2_inum_range_host *ir, const void *buf)
-{
- const struct gfs2_inum_range *str = buf;
-
- ir->ir_start = be64_to_cpu(str->ir_start);
- ir->ir_length = be64_to_cpu(str->ir_length);
-}
-
-static void gfs2_inum_range_out(const struct gfs2_inum_range_host *ir, void *buf)
-{
- struct gfs2_inum_range *str = buf;
-
- str->ir_start = cpu_to_be64(ir->ir_start);
- str->ir_length = cpu_to_be64(ir->ir_length);
-}
-
-static int pick_formal_ino_1(struct gfs2_sbd *sdp, u64 *formal_ino)
-{
- struct gfs2_inode *ip = GFS2_I(sdp->sd_ir_inode);
- struct buffer_head *bh;
- struct gfs2_inum_range_host ir;
- int error;
-
- error = gfs2_trans_begin(sdp, RES_DINODE, 0);
- if (error)
- return error;
- mutex_lock(&sdp->sd_inum_mutex);
-
- error = gfs2_meta_inode_buffer(ip, &bh);
- if (error) {
- mutex_unlock(&sdp->sd_inum_mutex);
- gfs2_trans_end(sdp);
- return error;
- }
-
- gfs2_inum_range_in(&ir, bh->b_data + sizeof(struct gfs2_dinode));
-
- if (ir.ir_length) {
- *formal_ino = ir.ir_start++;
- ir.ir_length--;
- gfs2_trans_add_bh(ip->i_gl, bh, 1);
- gfs2_inum_range_out(&ir,
- bh->b_data + sizeof(struct gfs2_dinode));
- brelse(bh);
- mutex_unlock(&sdp->sd_inum_mutex);
- gfs2_trans_end(sdp);
- return 0;
- }
-
- brelse(bh);
-
- mutex_unlock(&sdp->sd_inum_mutex);
- gfs2_trans_end(sdp);
-
- return 1;
-}
-
-static int pick_formal_ino_2(struct gfs2_sbd *sdp, u64 *formal_ino)
-{
- struct gfs2_inode *ip = GFS2_I(sdp->sd_ir_inode);
- struct gfs2_inode *m_ip = GFS2_I(sdp->sd_inum_inode);
- struct gfs2_holder gh;
- struct buffer_head *bh;
- struct gfs2_inum_range_host ir;
- int error;
-
- error = gfs2_glock_nq_init(m_ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh);
- if (error)
- return error;
-
- error = gfs2_trans_begin(sdp, 2 * RES_DINODE, 0);
- if (error)
- goto out;
- mutex_lock(&sdp->sd_inum_mutex);
-
- error = gfs2_meta_inode_buffer(ip, &bh);
- if (error)
- goto out_end_trans;
-
- gfs2_inum_range_in(&ir, bh->b_data + sizeof(struct gfs2_dinode));
-
- if (!ir.ir_length) {
- struct buffer_head *m_bh;
- u64 x, y;
- __be64 z;
-
- error = gfs2_meta_inode_buffer(m_ip, &m_bh);
- if (error)
- goto out_brelse;
-
- z = *(__be64 *)(m_bh->b_data + sizeof(struct gfs2_dinode));
- x = y = be64_to_cpu(z);
- ir.ir_start = x;
- ir.ir_length = GFS2_INUM_QUANTUM;
- x += GFS2_INUM_QUANTUM;
- if (x < y)
- gfs2_consist_inode(m_ip);
- z = cpu_to_be64(x);
- gfs2_trans_add_bh(m_ip->i_gl, m_bh, 1);
- *(__be64 *)(m_bh->b_data + sizeof(struct gfs2_dinode)) = z;
-
- brelse(m_bh);
- }
-
- *formal_ino = ir.ir_start++;
- ir.ir_length--;
-
- gfs2_trans_add_bh(ip->i_gl, bh, 1);
- gfs2_inum_range_out(&ir, bh->b_data + sizeof(struct gfs2_dinode));
-
-out_brelse:
- brelse(bh);
-out_end_trans:
- mutex_unlock(&sdp->sd_inum_mutex);
- gfs2_trans_end(sdp);
-out:
- gfs2_glock_dq_uninit(&gh);
- return error;
-}
-
-static int pick_formal_ino(struct gfs2_sbd *sdp, u64 *inum)
-{
- int error;
-
- error = pick_formal_ino_1(sdp, inum);
- if (error <= 0)
- return error;
-
- error = pick_formal_ino_2(sdp, inum);
-
- return error;
-}
-
/**
* create_ok - OK to create a new on-disk inode here?
* @dip: Directory in which dinode is to be created
@@ -981,13 +848,10 @@ struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name,
if (error)
goto fail_gunlock;

- error = pick_formal_ino(sdp, &inum.no_formal_ino);
- if (error)
- goto fail_gunlock;
-
error = alloc_dinode(dip, &inum.no_addr, &generation);
if (error)
goto fail_gunlock;
+ inum.no_formal_ino = generation;

error = gfs2_glock_nq_num(sdp, inum.no_addr, &gfs2_inode_glops,
LM_ST_EXCLUSIVE, GL_SKIP, ghs + 1);
@@ -998,9 +862,8 @@ struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name,
if (error)
goto fail_gunlock2;

- inode = gfs2_inode_lookup(dir->i_sb, IF2DT(mode),
- inum.no_addr,
- inum.no_formal_ino, 0);
+ inode = gfs2_inode_lookup(dir->i_sb, IF2DT(mode), inum.no_addr,
+ inum.no_formal_ino, 0);
if (IS_ERR(inode))
goto fail_gunlock2;

diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index 1ec69e6..8e3f00e 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -84,7 +84,6 @@ static struct gfs2_sbd *init_sbd(struct super_block *sb)

gfs2_tune_init(&sdp->sd_tune);

- mutex_init(&sdp->sd_inum_mutex);
spin_lock_init(&sdp->sd_statfs_spin);

spin_lock_init(&sdp->sd_rindex_spin);
@@ -833,21 +832,12 @@ static int init_inodes(struct gfs2_sbd *sdp, int undo)
if (error)
goto fail;

- /* Read in the master inode number inode */
- sdp->sd_inum_inode = gfs2_lookup_simple(master, "inum");
- if (IS_ERR(sdp->sd_inum_inode)) {
- error = PTR_ERR(sdp->sd_inum_inode);
- fs_err(sdp, "can't read in inum inode: %d\n", error);
- goto fail_journal;
- }
-
-
/* Read in the master statfs inode */
sdp->sd_statfs_inode = gfs2_lookup_simple(master, "statfs");
if (IS_ERR(sdp->sd_statfs_inode)) {
error = PTR_ERR(sdp->sd_statfs_inode);
fs_err(sdp, "can't read in statfs inode: %d\n", error);
- goto fail_inum;
+ goto fail_journal;
}

/* Read in the resource index inode */
@@ -876,8 +866,6 @@ fail_rindex:
iput(sdp->sd_rindex);
fail_statfs:
iput(sdp->sd_statfs_inode);
-fail_inum:
- iput(sdp->sd_inum_inode);
fail_journal:
init_journal(sdp, UNDO);
fail:
@@ -905,20 +893,12 @@ static int init_per_node(struct gfs2_sbd *sdp, int undo)
return error;
}

- sprintf(buf, "inum_range%u", sdp->sd_jdesc->jd_jid);
- sdp->sd_ir_inode = gfs2_lookup_simple(pn, buf);
- if (IS_ERR(sdp->sd_ir_inode)) {
- error = PTR_ERR(sdp->sd_ir_inode);
- fs_err(sdp, "can't find local \"ir\" file: %d\n", error);
- goto fail;
- }
-
sprintf(buf, "statfs_change%u", sdp->sd_jdesc->jd_jid);
sdp->sd_sc_inode = gfs2_lookup_simple(pn, buf);
if (IS_ERR(sdp->sd_sc_inode)) {
error = PTR_ERR(sdp->sd_sc_inode);
fs_err(sdp, "can't find local \"sc\" file: %d\n", error);
- goto fail_ir_i;
+ goto fail;
}

sprintf(buf, "quota_change%u", sdp->sd_jdesc->jd_jid);
@@ -932,27 +912,16 @@ static int init_per_node(struct gfs2_sbd *sdp, int undo)
iput(pn);
pn = NULL;

- ip = GFS2_I(sdp->sd_ir_inode);
- error = gfs2_glock_nq_init(ip->i_gl,
- LM_ST_EXCLUSIVE, 0,
- &sdp->sd_ir_gh);
- if (error) {
- fs_err(sdp, "can't lock local \"ir\" file: %d\n", error);
- goto fail_qc_i;
- }
-
ip = GFS2_I(sdp->sd_sc_inode);
- error = gfs2_glock_nq_init(ip->i_gl,
- LM_ST_EXCLUSIVE, 0,
+ error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0,
&sdp->sd_sc_gh);
if (error) {
fs_err(sdp, "can't lock local \"sc\" file: %d\n", error);
- goto fail_ir_gh;
+ goto fail_qc_i;
}

ip = GFS2_I(sdp->sd_qc_inode);
- error = gfs2_glock_nq_init(ip->i_gl,
- LM_ST_EXCLUSIVE, 0,
+ error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0,
&sdp->sd_qc_gh);
if (error) {
fs_err(sdp, "can't lock local \"qc\" file: %d\n", error);
@@ -965,14 +934,10 @@ fail_qc_gh:
gfs2_glock_dq_uninit(&sdp->sd_qc_gh);
fail_ut_gh:
gfs2_glock_dq_uninit(&sdp->sd_sc_gh);
-fail_ir_gh:
- gfs2_glock_dq_uninit(&sdp->sd_ir_gh);
fail_qc_i:
iput(sdp->sd_qc_inode);
fail_ut_i:
iput(sdp->sd_sc_inode);
-fail_ir_i:
- iput(sdp->sd_ir_inode);
fail:
if (pn)
iput(pn);
diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
index c681c54..388a61d 100644
--- a/fs/gfs2/rgrp.c
+++ b/fs/gfs2/rgrp.c
@@ -1567,6 +1567,8 @@ int gfs2_alloc_di(struct gfs2_inode *dip, u64 *bn, u64 *generation)
rgd->rd_free--;
rgd->rd_dinodes++;
*generation = rgd->rd_igeneration++;
+ if (*generation == 0)
+ *generation = rgd->rd_igeneration++;
gfs2_trans_add_bh(rgd->rd_gl, rgd->rd_bits[0].bi_bh, 1);
gfs2_rgrp_out(rgd, rgd->rd_bits[0].bi_bh->b_data);

diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c
index 2e78a3f..d95cf77 100644
--- a/fs/gfs2/super.c
+++ b/fs/gfs2/super.c
@@ -788,7 +788,6 @@ restart:
/* Release stuff */

iput(sdp->sd_jindex);
- iput(sdp->sd_inum_inode);
iput(sdp->sd_statfs_inode);
iput(sdp->sd_rindex);
iput(sdp->sd_quota_inode);
@@ -799,10 +798,8 @@ restart:
if (!sdp->sd_args.ar_spectator) {
gfs2_glock_dq_uninit(&sdp->sd_journal_gh);
gfs2_glock_dq_uninit(&sdp->sd_jinode_gh);
- gfs2_glock_dq_uninit(&sdp->sd_ir_gh);
gfs2_glock_dq_uninit(&sdp->sd_sc_gh);
gfs2_glock_dq_uninit(&sdp->sd_qc_gh);
- iput(sdp->sd_ir_inode);
iput(sdp->sd_sc_inode);
iput(sdp->sd_qc_inode);
}
--
1.6.2.5

2009-09-10 10:30:12

by Steven Whitehouse

[permalink] [raw]
Subject: [PATCH 14/15] GFS2: Be extra careful about deallocating inodes

There is a potential race in the inode deallocation code if two
nodes try to deallocate the same inode at the same time. Most of
the issue is solved by the iopen locking. There is still a small
window which is not covered by the iopen lock. This patches fixes
that and also makes the deallocation code more robust in the face of
any errors in the rgrp bitmaps, or erroneous iopen callbacks from
other nodes.

This does introduce one extra disk read, but that is generally not
an issue since its the same block that must be written to later
in the deallocation process. The total disk accesses therefore stay
the same,

Signed-off-by: Steven Whitehouse <[email protected]>
---
fs/gfs2/export.c | 36 ++++--------------------------------
fs/gfs2/rgrp.c | 42 +++++++++++++++++++++++++++++++++++++++++-
fs/gfs2/rgrp.h | 4 ++--
fs/gfs2/super.c | 4 ++++
4 files changed, 51 insertions(+), 35 deletions(-)

diff --git a/fs/gfs2/export.c b/fs/gfs2/export.c
index 9200ef2..d15876e 100644
--- a/fs/gfs2/export.c
+++ b/fs/gfs2/export.c
@@ -143,17 +143,14 @@ static struct dentry *gfs2_get_parent(struct dentry *child)
}

static struct dentry *gfs2_get_dentry(struct super_block *sb,
- struct gfs2_inum_host *inum)
+ struct gfs2_inum_host *inum)
{
struct gfs2_sbd *sdp = sb->s_fs_info;
- struct gfs2_holder i_gh, ri_gh, rgd_gh;
- struct gfs2_rgrpd *rgd;
+ struct gfs2_holder i_gh;
struct inode *inode;
struct dentry *dentry;
int error;

- /* System files? */
-
inode = gfs2_ilookup(sb, inum->no_addr);
if (inode) {
if (GFS2_I(inode)->i_no_formal_ino != inum->no_formal_ino) {
@@ -168,29 +165,11 @@ static struct dentry *gfs2_get_dentry(struct super_block *sb,
if (error)
return ERR_PTR(error);

- error = gfs2_rindex_hold(sdp, &ri_gh);
+ error = gfs2_check_blk_type(sdp, inum->no_addr, GFS2_BLKST_DINODE);
if (error)
goto fail;

- error = -EINVAL;
- rgd = gfs2_blk2rgrpd(sdp, inum->no_addr);
- if (!rgd)
- goto fail_rindex;
-
- error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_SHARED, 0, &rgd_gh);
- if (error)
- goto fail_rindex;
-
- error = -ESTALE;
- if (gfs2_get_block_type(rgd, inum->no_addr) != GFS2_BLKST_DINODE)
- goto fail_rgd;
-
- gfs2_glock_dq_uninit(&rgd_gh);
- gfs2_glock_dq_uninit(&ri_gh);
-
- inode = gfs2_inode_lookup(sb, DT_UNKNOWN,
- inum->no_addr,
- 0, 0);
+ inode = gfs2_inode_lookup(sb, DT_UNKNOWN, inum->no_addr, 0, 0);
if (IS_ERR(inode)) {
error = PTR_ERR(inode);
goto fail;
@@ -224,13 +203,6 @@ out_inode:
if (!IS_ERR(dentry))
dentry->d_op = &gfs2_dops;
return dentry;
-
-fail_rgd:
- gfs2_glock_dq_uninit(&rgd_gh);
-
-fail_rindex:
- gfs2_glock_dq_uninit(&ri_gh);
-
fail:
gfs2_glock_dq_uninit(&i_gh);
return ERR_PTR(error);
diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
index 388a61d..caaa665 100644
--- a/fs/gfs2/rgrp.c
+++ b/fs/gfs2/rgrp.c
@@ -1256,7 +1256,7 @@ void gfs2_inplace_release(struct gfs2_inode *ip)
* Returns: The block type (GFS2_BLKST_*)
*/

-unsigned char gfs2_get_block_type(struct gfs2_rgrpd *rgd, u64 block)
+static unsigned char gfs2_get_block_type(struct gfs2_rgrpd *rgd, u64 block)
{
struct gfs2_bitmap *bi = NULL;
u32 length, rgrp_block, buf_block;
@@ -1694,6 +1694,46 @@ void gfs2_free_di(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip)
}

/**
+ * gfs2_check_blk_type - Check the type of a block
+ * @sdp: The superblock
+ * @no_addr: The block number to check
+ * @type: The block type we are looking for
+ *
+ * Returns: 0 if the block type matches the expected type
+ * -ESTALE if it doesn't match
+ * or -ve errno if something went wrong while checking
+ */
+
+int gfs2_check_blk_type(struct gfs2_sbd *sdp, u64 no_addr, unsigned int type)
+{
+ struct gfs2_rgrpd *rgd;
+ struct gfs2_holder ri_gh, rgd_gh;
+ int error;
+
+ error = gfs2_rindex_hold(sdp, &ri_gh);
+ if (error)
+ goto fail;
+
+ error = -EINVAL;
+ rgd = gfs2_blk2rgrpd(sdp, no_addr);
+ if (!rgd)
+ goto fail_rindex;
+
+ error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_SHARED, 0, &rgd_gh);
+ if (error)
+ goto fail_rindex;
+
+ if (gfs2_get_block_type(rgd, no_addr) != type)
+ error = -ESTALE;
+
+ gfs2_glock_dq_uninit(&rgd_gh);
+fail_rindex:
+ gfs2_glock_dq_uninit(&ri_gh);
+fail:
+ return error;
+}
+
+/**
* gfs2_rlist_add - add a RG to a list of RGs
* @sdp: the filesystem
* @rlist: the list of resource groups
diff --git a/fs/gfs2/rgrp.h b/fs/gfs2/rgrp.h
index a8dedd7..b4106dd 100644
--- a/fs/gfs2/rgrp.h
+++ b/fs/gfs2/rgrp.h
@@ -44,8 +44,6 @@ gfs2_inplace_reserve_i((ip), __FILE__, __LINE__)

extern void gfs2_inplace_release(struct gfs2_inode *ip);

-extern unsigned char gfs2_get_block_type(struct gfs2_rgrpd *rgd, u64 block);
-
extern int gfs2_alloc_block(struct gfs2_inode *ip, u64 *bn, unsigned int *n);
extern int gfs2_alloc_di(struct gfs2_inode *ip, u64 *bn, u64 *generation);

@@ -53,6 +51,8 @@ extern void gfs2_free_data(struct gfs2_inode *ip, u64 bstart, u32 blen);
extern void gfs2_free_meta(struct gfs2_inode *ip, u64 bstart, u32 blen);
extern void gfs2_free_di(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip);
extern void gfs2_unlink_di(struct inode *inode);
+extern int gfs2_check_blk_type(struct gfs2_sbd *sdp, u64 no_addr,
+ unsigned int type);

struct gfs2_rgrp_list {
unsigned int rl_rgrps;
diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c
index d95cf77..0ec3ec6 100644
--- a/fs/gfs2/super.c
+++ b/fs/gfs2/super.c
@@ -1286,6 +1286,10 @@ static void gfs2_delete_inode(struct inode *inode)
goto out;
}

+ error = gfs2_check_blk_type(sdp, ip->i_no_addr, GFS2_BLKST_UNLINKED);
+ if (error)
+ goto out_truncate;
+
gfs2_glock_dq_wait(&ip->i_iopen_gh);
gfs2_holder_reinit(LM_ST_EXCLUSIVE, LM_FLAG_TRY_1CB | GL_NOCACHE, &ip->i_iopen_gh);
error = gfs2_glock_nq(&ip->i_iopen_gh);
--
1.6.2.5

2009-09-10 10:30:11

by Steven Whitehouse

[permalink] [raw]
Subject: [PATCH 15/15] GFS2: Remove unused sysfs file

The /sys/fs/gfs2/<fsname>/lock_module/id file has been unused for
some time now, so we can remove it. We still accept the mount option
though, as userspace still sends that.

Signed-off-by: Steven Whitehouse <[email protected]>
---
fs/gfs2/incore.h | 1 -
fs/gfs2/ops_fstype.c | 6 +-----
fs/gfs2/sys.c | 8 --------
3 files changed, 1 insertions(+), 14 deletions(-)

diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h
index 52436ab..6edb423 100644
--- a/fs/gfs2/incore.h
+++ b/fs/gfs2/incore.h
@@ -496,7 +496,6 @@ struct gfs2_sb_host {
*/

struct lm_lockstruct {
- u32 ls_id;
unsigned int ls_jid;
unsigned int ls_first;
unsigned int ls_first_done;
diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index 8e3f00e..52fb6c0 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -1028,7 +1028,6 @@ static int gfs2_lm_mount(struct gfs2_sbd *sdp, int silent)

ls->ls_ops = lm;
ls->ls_first = 1;
- ls->ls_id = 0;

for (options = args->ar_hostdata; (o = strsep(&options, ":")); ) {
substring_t tmp[MAX_OPT_ARGS];
@@ -1046,10 +1045,7 @@ static int gfs2_lm_mount(struct gfs2_sbd *sdp, int silent)
ls->ls_jid = option;
break;
case Opt_id:
- ret = match_int(&tmp[0], &option);
- if (ret)
- goto hostdata_error;
- ls->ls_id = option;
+ /* Obsolete, but left for backward compat purposes */
break;
case Opt_first:
ret = match_int(&tmp[0], &option);
diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c
index 0d4f7e9..4463297 100644
--- a/fs/gfs2/sys.c
+++ b/fs/gfs2/sys.c
@@ -320,12 +320,6 @@ static ssize_t block_store(struct gfs2_sbd *sdp, const char *buf, size_t len)
return ret;
}

-static ssize_t lkid_show(struct gfs2_sbd *sdp, char *buf)
-{
- struct lm_lockstruct *ls = &sdp->sd_lockstruct;
- return sprintf(buf, "%u\n", ls->ls_id);
-}
-
static ssize_t lkfirst_show(struct gfs2_sbd *sdp, char *buf)
{
struct lm_lockstruct *ls = &sdp->sd_lockstruct;
@@ -390,7 +384,6 @@ static struct gfs2_attr gdlm_attr_##_name = __ATTR(_name,_mode,_show,_store)
GDLM_ATTR(proto_name, 0444, proto_name_show, NULL);
GDLM_ATTR(block, 0644, block_show, block_store);
GDLM_ATTR(withdraw, 0644, withdraw_show, withdraw_store);
-GDLM_ATTR(id, 0444, lkid_show, NULL);
GDLM_ATTR(jid, 0444, jid_show, NULL);
GDLM_ATTR(first, 0444, lkfirst_show, NULL);
GDLM_ATTR(first_done, 0444, first_done_show, NULL);
@@ -402,7 +395,6 @@ static struct attribute *lock_module_attrs[] = {
&gdlm_attr_proto_name.attr,
&gdlm_attr_block.attr,
&gdlm_attr_withdraw.attr,
- &gdlm_attr_id.attr,
&gdlm_attr_jid.attr,
&gdlm_attr_first.attr,
&gdlm_attr_first_done.attr,
--
1.6.2.5

2009-09-10 14:46:07

by Daniel Walker

[permalink] [raw]
Subject: Re: GFS2: Pre-pull patch posting

On Thu, 2009-09-10 at 12:27 +0100, Steven Whitehouse wrote:
> As merge time is approaching, here is the current content of the
> GFS2 -nmw git tree. I'm not expecting to take any more patches
> now for the current merge window unless any last minute bugs
> are discovered.
>
> There is not a huge amount new this time. Some extra context for
> uevent messages, better error handling during block allocation,
> and a clean up of extended attribute support. There is still more
> to do on the extended attribute side of things, but this is a good
> start I think.
>
> There are a few bug fixes as well. Once these patches are merged
> I'm intending to start off the next -nmw tree with a patch to
> remove some of the (now unused) sysfs files as per the message
> on cluster-devel a few weeks back.

This patch set has a couple of checkpatch issues..

Patch 3,

ERROR: code indent should use tabs where possible
#83: FILE: fs/gfs2/rgrp.c:1466:
+^I (unsigned long long)rgd->rd_addr);$


Patch 11,

ERROR: trailing whitespace
#741: FILE: fs/gfs2/eattr.c:565:
+^Iif (size) $

and patch 12,

ERROR: trailing whitespace
#2349: FILE: fs/gfs2/xattr.c:565:
+^Iif (size) $

ERROR: space required after that ',' (ctx:VxV)
#2833: FILE: fs/gfs2/xattr.c:1049:
+ return ea_remove_unstuffed(ip, el->el_bh, el->el_ea, el->el_prev,0);

ERROR: code indent should use tabs where possible
#3370: FILE: fs/gfs2/xattr.h:22:
+ (sizeof(__be64) * (ea)->ea_num_ptrs)), 8)$

Could you correct these prior to you pull request?

Daniel

2009-09-14 07:57:30

by Steven Whitehouse

[permalink] [raw]
Subject: Re: GFS2: Pre-pull patch posting

Hi,

On Thu, 2009-09-10 at 07:46 -0700, Daniel Walker wrote:
> On Thu, 2009-09-10 at 12:27 +0100, Steven Whitehouse wrote:
> > As merge time is approaching, here is the current content of the
> > GFS2 -nmw git tree. I'm not expecting to take any more patches
> > now for the current merge window unless any last minute bugs
> > are discovered.
> >
> > There is not a huge amount new this time. Some extra context for
> > uevent messages, better error handling during block allocation,
> > and a clean up of extended attribute support. There is still more
> > to do on the extended attribute side of things, but this is a good
> > start I think.
> >
> > There are a few bug fixes as well. Once these patches are merged
> > I'm intending to start off the next -nmw tree with a patch to
> > remove some of the (now unused) sysfs files as per the message
> > on cluster-devel a few weeks back.
>
> This patch set has a couple of checkpatch issues..
>
> Patch 3,
>
> ERROR: code indent should use tabs where possible
> #83: FILE: fs/gfs2/rgrp.c:1466:
> +^I (unsigned long long)rgd->rd_addr);$
>
>
Now fixed in the latest -nmw tree.

> Patch 11,
>
> ERROR: trailing whitespace
> #741: FILE: fs/gfs2/eattr.c:565:
> +^Iif (size) $
>
> and patch 12,
>
> ERROR: trailing whitespace
> #2349: FILE: fs/gfs2/xattr.c:565:
> +^Iif (size) $
>
This is the same line of code. There is no need to report it twice.

> ERROR: space required after that ',' (ctx:VxV)
> #2833: FILE: fs/gfs2/xattr.c:1049:
> + return ea_remove_unstuffed(ip, el->el_bh, el->el_ea, el->el_prev,0);
>
> ERROR: code indent should use tabs where possible
> #3370: FILE: fs/gfs2/xattr.h:22:
> + (sizeof(__be64) * (ea)->ea_num_ptrs)), 8)$
>
> Could you correct these prior to you pull request?
>
> Daniel
>
The above two are both part of the context rather than something I've
added. I've fixed them anyway though,

Steve.

2009-09-14 14:15:00

by Daniel Walker

[permalink] [raw]
Subject: Re: GFS2: Pre-pull patch posting

On Mon, 2009-09-14 at 08:57 +0100, Steven Whitehouse wrote:

> > ERROR: code indent should use tabs where possible
> > #83: FILE: fs/gfs2/rgrp.c:1466:
> > +^I (unsigned long long)rgd->rd_addr);$
> >
> >
> Now fixed in the latest -nmw tree.
>

Ok, Thank you.

Daniel