* [Cluster-devel] [GFS2] Pre-pull patch posting
@ 2008-01-21 9:21 swhiteho
0 siblings, 0 replies; 34+ messages in thread
From: swhiteho @ 2008-01-21 9:21 UTC (permalink / raw)
To: cluster-devel.redhat.com
Hi,
Here is the current GFS2 patch queue. You'll notice that this time
there are no DLM patches in this list. That is because the DLM team
are setting up their own git tree and this future DLM patches will
be sent directly by them rather than via the GFS2 tree.
Most of this set of patches is clean up and bug fixes, there is really
not a lot new this time. I guess the most significant thing is the
patch to use ->page_mkwrite which will greatly increase efficiency
when files opened r/w are mostly only accessed for reading across a
cluster.
There are a number of cleanups related to journalling which is really
where the largest number of changes in terms of code lines is. The
indirect blocks for the journal are now scanned once only at mount
time and the bmap information is retained in the form of an extent
list. Since we expect journals to consist of only a single extent
in the normal case, this should generally be quite a short list :-)
In addition some of the tunables relating to the journal have been
removed in favour of autotuning those variables.
Steve.
^ permalink raw reply [flat|nested] 34+ messages in thread
* [Cluster-devel] [GFS2] Pre-pull patch posting
@ 2008-04-17 8:37 swhiteho
0 siblings, 0 replies; 34+ messages in thread
From: swhiteho @ 2008-04-17 8:37 UTC (permalink / raw)
To: cluster-devel.redhat.com
This is the current content of the GFS2 -nmw git tree. Mostly bug
fixes, there are some changes relating to block mapping which are
working towards cleaning up this code and allowing more efficient
block mapping. There is a second part to that work which is not
included in this patch set - the plan is that it will be in the
next patch set and its currently undergoing testing.
There are a number of clean up patches in the series too. We have
been continuing the work of gradually reducing the fields in the
various ..._host structures with a view to eventually eliminating
them completely. They were introduced as a stop-gap measure to fix
the endianess annotation and the fields are now gradually being moved
to other structures (or eliminated).
Bob Peterson's new improved "bitfit" algorithm provides a nice
speed up when we are allocating blocks as well as cleaning up
that area of the code.
Steve.
^ permalink raw reply [flat|nested] 34+ messages in thread
* [Cluster-devel] [GFS2] Pre-pull patch posting
@ 2008-07-11 10:11 swhiteho
0 siblings, 0 replies; 34+ messages in thread
From: swhiteho @ 2008-07-11 10:11 UTC (permalink / raw)
To: cluster-devel.redhat.com
So, although the merge window isn't yet open, I'm guessing that its
probably not too far away, hence this posting of the contents of
the current GFS2 -nmw git tree.
This time the big news is locking changes, although having said that,
there are far fewer queued patches in total than I've had for previous
merge windows and I believe that this is an indication of the
growing maturity of GFS2.
The first patch in the series is really the main change and is
a big clean up of the core of the glocks, which are really the
core of GFS2 in a lot of ways. Further through the series is
a documentation patch, which explains the fine detail of how
glocks work and the assumptions made by the glock core when
calling the functions relating to individual glock types.
Other notable changes include merging the lock_nolock module
into the core of GFS2 since there is little point in retaining
it separately. There is a plan to do the same to lock_dlm as
well in the future (not the DLM itself obviously, just the
interface module thats part of GFS2).
Most of the remaing changes are bug fixes or futher optimisations
over the initial glock changes, plus one or two minor clean ups
along the way.
Steve.
^ permalink raw reply [flat|nested] 34+ messages in thread
* [Cluster-devel] GFS2: Pre-pull patch posting
@ 2008-09-26 12:00 Steven Whitehouse
0 siblings, 0 replies; 34+ messages in thread
From: Steven Whitehouse @ 2008-09-26 12:00 UTC (permalink / raw)
To: cluster-devel.redhat.com
I'm guessing that the merge window opening might not be too far
away now, and in any case, I won't have quite my normal internet
access next week. So I'm pushing out the current GFS2 tree so
that (I hope) there will be time to fix any issues.
Again, there are fewer patches here. A lot of them are fairly small
too. The most noteable item deals with the meta filesystem which
was in response to Al Viro's suggestions concerning a better way
to structure that code. It certainly results in a much cleaner
implementation, so thanks go to Al for pointing that out.
A couple of new features: I/O barrier support (needs no user
configuration, see the patch for details) and UUID support
(no code changes, its all userland but we reserve space in
the super block, again details in the patch itself).
I know that I hardly need say, but please let me know if you have any
comments :-)
Steve.
^ permalink raw reply [flat|nested] 34+ messages in thread
* [Cluster-devel] GFS2: Pre-pull patch posting
@ 2008-12-17 11:29 swhiteho
0 siblings, 0 replies; 34+ messages in thread
From: swhiteho @ 2008-12-17 11:29 UTC (permalink / raw)
To: cluster-devel.redhat.com
In preparation for the next merge window, here is the current content
of the GFS2 git tree. Firstly, we have one new feature, which is
support for the FIEMAP ioctl. That patch does touch some code
outside of GFS2 itself, but its the only patch in this series which
does so.
The remaining patches are mostly clean up and bug fixes, as usual.
They are working towards a point where I can submit a patch to finally
merge the lock_dlm module (not dlm itself I should emphasise) into
GFS2. That is a large patch and a preliminary version has already been
posted to cluster-devel. My plan is to put that patch into my -nmw
git tree as the first patch for the following merge window to give
it maximum exposure.
The other highlight of this patch series, is a patch which removes
the two daemons (gfs2_scand and gfs2_glockd) and replaces them with
a "shrinker" routine registered with the VM. As expected, this also
reduces the code size. We are also expecting do do a similar thing
with the GFS2 quota data strucutures at some point in the future.
Steve.
^ permalink raw reply [flat|nested] 34+ messages in thread
* [Cluster-devel] [GFS2] Pre-pull patch posting
@ 2009-03-18 12:23 swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 01/18] GFS2: Fix remount argument parsing swhiteho
0 siblings, 1 reply; 34+ messages in thread
From: swhiteho @ 2009-03-18 12:23 UTC (permalink / raw)
To: cluster-devel.redhat.com
Hi,
So as the merge window draws closer, here is the current content of the
GFS2 git tree. The major item this time is patch 5 in the series. This
contians by far the majority of the changes, and the majority of those
changes are actually removal of code. The patch merges the lock_dlm
module (not the dlm itself, but GFS2's interface to the dlm) into
GFS2 itself. This means that a number of optimisations are then
possible in terms of merging strucutures resulting in a considerable
saving in memory.
Since that patch is so large (I'm afraid that it really doesn't make
any sense to split it up) its been in the -nmw git tree for the whole
period since the last merge window and has also been posted for review
on cluster-devel on a number of occasions before that. We've run a
number of tests on it as well in that period, so I believe that its
pretty stable now. It certainly makes the code a lot cleaner and
easier to follow in that area.
The remainder of the patches are mostly bug fixes, but there are one
or two other interesting features, those being:
o GFS2 now supports the discard I/O requests for thin provisioning, etc
o A new "demote a glock" interface is added to sysfs to help in
testing GFS2
o With a new mkfs.gfs2 which writes UUIDs, the UUID is now included
in uevent messages (with older filesystems which don't have UUIDs,
we just don't send that information)
The GFS2 tracing patches which I posted a little while back are not
included in this patch set. I think I can see what I need to do in
order to avoid patching blktrace now, so my plan is to look at those
patches again after this merge window, and when all the queued patches
for the tracing subsystem have been merged.
As always, please let us know if you spot any issues in the patches,
Steve.
^ permalink raw reply [flat|nested] 34+ messages in thread
* [Cluster-devel] [PATCH 01/18] GFS2: Fix remount argument parsing
2009-03-18 12:23 [Cluster-devel] [GFS2] Pre-pull patch posting swhiteho
@ 2009-03-18 12:23 ` swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 02/18] GFS2: Bring back lvb-related stuff to lock_nolock to support quotas swhiteho
0 siblings, 1 reply; 34+ messages in thread
From: swhiteho @ 2009-03-18 12:23 UTC (permalink / raw)
To: cluster-devel.redhat.com
From: Steven Whitehouse <swhiteho@redhat.com>
The following patch fixes an issue relating to remount and argument
parsing. After this fix is applied, remount becomes atomic in that
it either succeeds changing the mount to the new state, or it fails
and leaves it in the old state. Previously it was possible for the
parsing of options to fail part way though and for the fs to be left
in a state where some of the new arguments had been applied, but some
had not.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
diff --git a/fs/gfs2/mount.c b/fs/gfs2/mount.c
index 3cb0a44..3524ae8 100644
--- a/fs/gfs2/mount.c
+++ b/fs/gfs2/mount.c
@@ -17,7 +17,7 @@
#include "gfs2.h"
#include "incore.h"
-#include "mount.h"
+#include "super.h"
#include "sys.h"
#include "util.h"
@@ -77,101 +77,46 @@ static const match_table_t tokens = {
* Return: errno
*/
-int gfs2_mount_args(struct gfs2_sbd *sdp, char *data_arg, int remount)
+int gfs2_mount_args(struct gfs2_sbd *sdp, struct gfs2_args *args, char *options)
{
- struct gfs2_args *args = &sdp->sd_args;
- char *data = data_arg;
- char *options, *o, *v;
- int error = 0;
-
- if (!remount) {
- /* Set some defaults */
- args->ar_quota = GFS2_QUOTA_DEFAULT;
- args->ar_data = GFS2_DATA_DEFAULT;
- }
+ char *o;
+ int token;
+ substring_t tmp[MAX_OPT_ARGS];
/* Split the options into tokens with the "," character and
process them */
- for (options = data; (o = strsep(&options, ",")); ) {
- int token;
- substring_t tmp[MAX_OPT_ARGS];
-
- if (!*o)
+ while (1) {
+ o = strsep(&options, ",");
+ if (o == NULL)
+ break;
+ if (*o == '\0')
continue;
token = match_token(o, tokens, tmp);
switch (token) {
case Opt_lockproto:
- v = match_strdup(&tmp[0]);
- if (!v) {
- fs_info(sdp, "no memory for lockproto\n");
- error = -ENOMEM;
- goto out_error;
- }
-
- if (remount && strcmp(v, args->ar_lockproto)) {
- kfree(v);
- goto cant_remount;
- }
-
- strncpy(args->ar_lockproto, v, GFS2_LOCKNAME_LEN);
- args->ar_lockproto[GFS2_LOCKNAME_LEN - 1] = 0;
- kfree(v);
+ match_strlcpy(args->ar_lockproto, &tmp[0],
+ GFS2_LOCKNAME_LEN);
break;
case Opt_locktable:
- v = match_strdup(&tmp[0]);
- if (!v) {
- fs_info(sdp, "no memory for locktable\n");
- error = -ENOMEM;
- goto out_error;
- }
-
- if (remount && strcmp(v, args->ar_locktable)) {
- kfree(v);
- goto cant_remount;
- }
-
- strncpy(args->ar_locktable, v, GFS2_LOCKNAME_LEN);
- args->ar_locktable[GFS2_LOCKNAME_LEN - 1] = 0;
- kfree(v);
+ match_strlcpy(args->ar_locktable, &tmp[0],
+ GFS2_LOCKNAME_LEN);
break;
case Opt_hostdata:
- v = match_strdup(&tmp[0]);
- if (!v) {
- fs_info(sdp, "no memory for hostdata\n");
- error = -ENOMEM;
- goto out_error;
- }
-
- if (remount && strcmp(v, args->ar_hostdata)) {
- kfree(v);
- goto cant_remount;
- }
-
- strncpy(args->ar_hostdata, v, GFS2_LOCKNAME_LEN);
- args->ar_hostdata[GFS2_LOCKNAME_LEN - 1] = 0;
- kfree(v);
+ match_strlcpy(args->ar_hostdata, &tmp[0],
+ GFS2_LOCKNAME_LEN);
break;
case Opt_spectator:
- if (remount && !args->ar_spectator)
- goto cant_remount;
args->ar_spectator = 1;
- sdp->sd_vfs->s_flags |= MS_RDONLY;
break;
case Opt_ignore_local_fs:
- if (remount && !args->ar_ignore_local_fs)
- goto cant_remount;
args->ar_ignore_local_fs = 1;
break;
case Opt_localflocks:
- if (remount && !args->ar_localflocks)
- goto cant_remount;
args->ar_localflocks = 1;
break;
case Opt_localcaching:
- if (remount && !args->ar_localcaching)
- goto cant_remount;
args->ar_localcaching = 1;
break;
case Opt_debug:
@@ -181,17 +126,13 @@ int gfs2_mount_args(struct gfs2_sbd *sdp, char *data_arg, int remount)
args->ar_debug = 0;
break;
case Opt_upgrade:
- if (remount && !args->ar_upgrade)
- goto cant_remount;
args->ar_upgrade = 1;
break;
case Opt_acl:
args->ar_posix_acl = 1;
- sdp->sd_vfs->s_flags |= MS_POSIXACL;
break;
case Opt_noacl:
args->ar_posix_acl = 0;
- sdp->sd_vfs->s_flags &= ~MS_POSIXACL;
break;
case Opt_quota_off:
args->ar_quota = GFS2_QUOTA_OFF;
@@ -215,29 +156,15 @@ int gfs2_mount_args(struct gfs2_sbd *sdp, char *data_arg, int remount)
args->ar_data = GFS2_DATA_ORDERED;
break;
case Opt_meta:
- if (remount && args->ar_meta != 1)
- goto cant_remount;
args->ar_meta = 1;
break;
case Opt_err:
default:
- fs_info(sdp, "unknown option: %s\n", o);
- error = -EINVAL;
- goto out_error;
+ fs_info(sdp, "invalid mount option: %s\n", o);
+ return -EINVAL;
}
}
-out_error:
- if (error)
- fs_info(sdp, "invalid mount option(s)\n");
-
- if (data != data_arg)
- kfree(data);
-
- return error;
-
-cant_remount:
- fs_info(sdp, "can't remount with option %s\n", o);
- return -EINVAL;
+ return 0;
}
diff --git a/fs/gfs2/mount.h b/fs/gfs2/mount.h
deleted file mode 100644
index 401288a..0000000
--- a/fs/gfs2/mount.h
+++ /dev/null
@@ -1,17 +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 __MOUNT_DOT_H__
-#define __MOUNT_DOT_H__
-
-struct gfs2_sbd;
-
-int gfs2_mount_args(struct gfs2_sbd *sdp, char *data_arg, int remount);
-
-#endif /* __MOUNT_DOT_H__ */
diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index f91eebd..3eb49ed 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -25,7 +25,6 @@
#include "glock.h"
#include "glops.h"
#include "inode.h"
-#include "mount.h"
#include "recovery.h"
#include "rgrp.h"
#include "super.h"
@@ -1116,12 +1115,20 @@ static int fill_super(struct super_block *sb, void *data, int silent)
return -ENOMEM;
}
- error = gfs2_mount_args(sdp, (char *)data, 0);
+ sdp->sd_args.ar_quota = GFS2_QUOTA_DEFAULT;
+ sdp->sd_args.ar_data = GFS2_DATA_DEFAULT;
+
+ error = gfs2_mount_args(sdp, &sdp->sd_args, data);
if (error) {
printk(KERN_WARNING "GFS2: can't parse mount arguments\n");
goto fail;
}
+ if (sdp->sd_args.ar_spectator)
+ sb->s_flags |= MS_RDONLY;
+ if (sdp->sd_args.ar_posix_acl)
+ sb->s_flags |= MS_POSIXACL;
+
sb->s_magic = GFS2_MAGIC;
sb->s_op = &gfs2_super_ops;
sb->s_export_op = &gfs2_export_ops;
diff --git a/fs/gfs2/ops_super.c b/fs/gfs2/ops_super.c
index 320323d..f0699ac 100644
--- a/fs/gfs2/ops_super.c
+++ b/fs/gfs2/ops_super.c
@@ -27,7 +27,6 @@
#include "glock.h"
#include "inode.h"
#include "log.h"
-#include "mount.h"
#include "quota.h"
#include "recovery.h"
#include "rgrp.h"
@@ -40,6 +39,8 @@
#include "bmap.h"
#include "meta_io.h"
+#define args_neq(a1, a2, x) ((a1)->ar_##x != (a2)->ar_##x)
+
/**
* gfs2_write_inode - Make sure the inode is stable on the disk
* @inode: The inode
@@ -435,25 +436,45 @@ static int gfs2_statfs(struct dentry *dentry, struct kstatfs *buf)
static int gfs2_remount_fs(struct super_block *sb, int *flags, char *data)
{
struct gfs2_sbd *sdp = sb->s_fs_info;
+ struct gfs2_args args = sdp->sd_args; /* Default to current settings */
int error;
- error = gfs2_mount_args(sdp, data, 1);
+ error = gfs2_mount_args(sdp, &args, data);
if (error)
return error;
+ /* Not allowed to change locking details */
+ if (strcmp(args.ar_lockproto, sdp->sd_args.ar_lockproto) ||
+ strcmp(args.ar_locktable, sdp->sd_args.ar_locktable) ||
+ strcmp(args.ar_hostdata, sdp->sd_args.ar_hostdata))
+ return -EINVAL;
+
+ /* Some flags must not be changed */
+ if (args_neq(&args, &sdp->sd_args, spectator) ||
+ args_neq(&args, &sdp->sd_args, ignore_local_fs) ||
+ args_neq(&args, &sdp->sd_args, localflocks) ||
+ args_neq(&args, &sdp->sd_args, localcaching) ||
+ args_neq(&args, &sdp->sd_args, meta))
+ return -EINVAL;
+
if (sdp->sd_args.ar_spectator)
*flags |= MS_RDONLY;
- else {
- if (*flags & MS_RDONLY) {
- if (!(sb->s_flags & MS_RDONLY))
- error = gfs2_make_fs_ro(sdp);
- } else if (!(*flags & MS_RDONLY) &&
- (sb->s_flags & MS_RDONLY)) {
+
+ if ((sb->s_flags ^ *flags) & MS_RDONLY) {
+ if (*flags & MS_RDONLY)
+ error = gfs2_make_fs_ro(sdp);
+ else
error = gfs2_make_fs_rw(sdp);
- }
+ if (error)
+ return error;
}
- return error;
+ sdp->sd_args = args;
+ if (sdp->sd_args.ar_posix_acl)
+ sb->s_flags |= MS_POSIXACL;
+ else
+ sb->s_flags &= ~MS_POSIXACL;
+ return 0;
}
/**
diff --git a/fs/gfs2/super.h b/fs/gfs2/super.h
index f6b8b00..91abdbe 100644
--- a/fs/gfs2/super.h
+++ b/fs/gfs2/super.h
@@ -14,7 +14,7 @@
#include <linux/dcache.h>
#include "incore.h"
-void gfs2_lm_unmount(struct gfs2_sbd *sdp);
+extern void gfs2_lm_unmount(struct gfs2_sbd *sdp);
static inline unsigned int gfs2_jindex_size(struct gfs2_sbd *sdp)
{
@@ -27,21 +27,23 @@ static inline unsigned int gfs2_jindex_size(struct gfs2_sbd *sdp)
void gfs2_jindex_free(struct gfs2_sbd *sdp);
-struct gfs2_jdesc *gfs2_jdesc_find(struct gfs2_sbd *sdp, unsigned int jid);
-int gfs2_jdesc_check(struct gfs2_jdesc *jd);
+extern int gfs2_mount_args(struct gfs2_sbd *sdp, struct gfs2_args *args, char *data);
-int gfs2_lookup_in_master_dir(struct gfs2_sbd *sdp, char *filename,
- struct gfs2_inode **ipp);
+extern struct gfs2_jdesc *gfs2_jdesc_find(struct gfs2_sbd *sdp, unsigned int jid);
+extern int gfs2_jdesc_check(struct gfs2_jdesc *jd);
-int gfs2_make_fs_rw(struct gfs2_sbd *sdp);
+extern int gfs2_lookup_in_master_dir(struct gfs2_sbd *sdp, char *filename,
+ struct gfs2_inode **ipp);
-int gfs2_statfs_init(struct gfs2_sbd *sdp);
-void gfs2_statfs_change(struct gfs2_sbd *sdp,
- s64 total, s64 free, s64 dinodes);
-int gfs2_statfs_sync(struct gfs2_sbd *sdp);
+extern int gfs2_make_fs_rw(struct gfs2_sbd *sdp);
-int gfs2_freeze_fs(struct gfs2_sbd *sdp);
-void gfs2_unfreeze_fs(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);
+extern int gfs2_statfs_sync(struct gfs2_sbd *sdp);
+
+extern int gfs2_freeze_fs(struct gfs2_sbd *sdp);
+extern void gfs2_unfreeze_fs(struct gfs2_sbd *sdp);
extern struct file_system_type gfs2_fs_type;
extern struct file_system_type gfs2meta_fs_type;
--
1.6.0.3
^ permalink raw reply related [flat|nested] 34+ messages in thread
* [Cluster-devel] [PATCH 02/18] GFS2: Bring back lvb-related stuff to lock_nolock to support quotas
2009-03-18 12:23 ` [Cluster-devel] [PATCH 01/18] GFS2: Fix remount argument parsing swhiteho
@ 2009-03-18 12:23 ` swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 03/18] GFS2: change gfs2_quota_scan into a shrinker swhiteho
0 siblings, 1 reply; 34+ messages in thread
From: swhiteho @ 2009-03-18 12:23 UTC (permalink / raw)
To: cluster-devel.redhat.com
From: Abhijith Das <adas@redhat.com>
The quota code uses lvbs and this is currently not implemented in
lock_nolock, thereby causing panics when quota is enabled with
lock_nolock. This patch adds the relevant bits.
Signed-off-by: Abhijith Das <adas@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
diff --git a/fs/gfs2/locking.c b/fs/gfs2/locking.c
index 523243a..d3657bc 100644
--- a/fs/gfs2/locking.c
+++ b/fs/gfs2/locking.c
@@ -23,11 +23,74 @@ struct lmh_wrapper {
const struct lm_lockops *lw_ops;
};
+struct nolock_lockspace {
+ unsigned int nl_lvb_size;
+};
+
+/**
+ * nolock_get_lock - get a lm_lock_t given a descripton of the lock
+ * @lockspace: the lockspace the lock lives in
+ * @name: the name of the lock
+ * @lockp: return the lm_lock_t here
+ *
+ * Returns: 0 on success, -EXXX on failure
+ */
+
+static int nolock_get_lock(void *lockspace, struct lm_lockname *name,
+ void **lockp)
+{
+ *lockp = lockspace;
+ return 0;
+}
+
+/**
+ * nolock_put_lock - get rid of a lock structure
+ * @lock: the lock to throw away
+ *
+ */
+
+static void nolock_put_lock(void *lock)
+{
+}
+
+/**
+ * nolock_hold_lvb - hold on to a lock value block
+ * @lock: the lock the LVB is associated with
+ * @lvbp: return the lm_lvb_t here
+ *
+ * Returns: 0 on success, -EXXX on failure
+ */
+
+static int nolock_hold_lvb(void *lock, char **lvbp)
+{
+ struct nolock_lockspace *nl = lock;
+ int error = 0;
+
+ *lvbp = kzalloc(nl->nl_lvb_size, GFP_KERNEL);
+ if (!*lvbp)
+ error = -ENOMEM;
+
+ return error;
+}
+
+/**
+ * nolock_unhold_lvb - release a LVB
+ * @lock: the lock the LVB is associated with
+ * @lvb: the lock value block
+ *
+ */
+
+static void nolock_unhold_lvb(void *lock, char *lvb)
+{
+ kfree(lvb);
+}
+
static int nolock_mount(char *table_name, char *host_data,
lm_callback_t cb, void *cb_data,
unsigned int min_lvb_size, int flags,
struct lm_lockstruct *lockstruct,
struct kobject *fskobj);
+static void nolock_unmount(void *lockspace);
/* List of registered low-level locking protocols. A file system selects one
of them by name at mount time, e.g. lock_nolock, lock_dlm. */
@@ -35,6 +98,11 @@ static int nolock_mount(char *table_name, char *host_data,
static const struct lm_lockops nolock_ops = {
.lm_proto_name = "lock_nolock",
.lm_mount = nolock_mount,
+ .lm_unmount = nolock_unmount,
+ .lm_get_lock = nolock_get_lock,
+ .lm_put_lock = nolock_put_lock,
+ .lm_hold_lvb = nolock_hold_lvb,
+ .lm_unhold_lvb = nolock_unhold_lvb,
};
static struct lmh_wrapper nolock_proto = {
@@ -53,6 +121,7 @@ static int nolock_mount(char *table_name, char *host_data,
{
char *c;
unsigned int jid;
+ struct nolock_lockspace *nl;
c = strstr(host_data, "jid=");
if (!c)
@@ -62,15 +131,28 @@ static int nolock_mount(char *table_name, char *host_data,
sscanf(c, "%u", &jid);
}
+ nl = kzalloc(sizeof(struct nolock_lockspace), GFP_KERNEL);
+ if (!nl)
+ return -ENOMEM;
+
+ nl->nl_lvb_size = min_lvb_size;
+
lockstruct->ls_jid = jid;
lockstruct->ls_first = 1;
lockstruct->ls_lvb_size = min_lvb_size;
+ lockstruct->ls_lockspace = nl;
lockstruct->ls_ops = &nolock_ops;
lockstruct->ls_flags = LM_LSFLAG_LOCAL;
return 0;
}
+static void nolock_unmount(void *lockspace)
+{
+ struct nolock_lockspace *nl = lockspace;
+ kfree(nl);
+}
+
/**
* gfs2_register_lockproto - Register a low-level locking protocol
* @proto: the protocol definition
--
1.6.0.3
^ permalink raw reply related [flat|nested] 34+ messages in thread
* [Cluster-devel] [PATCH 03/18] GFS2: change gfs2_quota_scan into a shrinker
2009-03-18 12:23 ` [Cluster-devel] [PATCH 02/18] GFS2: Bring back lvb-related stuff to lock_nolock to support quotas swhiteho
@ 2009-03-18 12:23 ` swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 04/18] GFS2: Remove "double" locking in quota swhiteho
0 siblings, 1 reply; 34+ messages in thread
From: swhiteho @ 2009-03-18 12:23 UTC (permalink / raw)
To: cluster-devel.redhat.com
From: Abhijith Das <adas@redhat.com>
Deallocation of gfs2_quota_data objects now happens on-demand through a
shrinker instead of routinely deallocating through the quotad daemon.
Signed-off-by: Abhijith Das <adas@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h
index 608849d..592aa50 100644
--- a/fs/gfs2/incore.h
+++ b/fs/gfs2/incore.h
@@ -283,7 +283,9 @@ enum {
struct gfs2_quota_data {
struct list_head qd_list;
- unsigned int qd_count;
+ struct list_head qd_reclaim;
+
+ atomic_t qd_count;
u32 qd_id;
unsigned long qd_flags; /* QDF_... */
@@ -303,7 +305,6 @@ struct gfs2_quota_data {
u64 qd_sync_gen;
unsigned long qd_last_warn;
- unsigned long qd_last_touched;
};
struct gfs2_trans {
@@ -406,7 +407,6 @@ struct gfs2_tune {
unsigned int gt_quota_warn_period; /* Secs between quota warn msgs */
unsigned int gt_quota_scale_num; /* Numerator */
unsigned int gt_quota_scale_den; /* Denominator */
- unsigned int gt_quota_cache_secs;
unsigned int gt_quota_quantum; /* Secs between syncs to quota file */
unsigned int gt_new_files_jdata;
unsigned int gt_max_readahead; /* Max bytes to read-ahead from disk */
diff --git a/fs/gfs2/main.c b/fs/gfs2/main.c
index 7cacfde..86fe067 100644
--- a/fs/gfs2/main.c
+++ b/fs/gfs2/main.c
@@ -23,6 +23,12 @@
#include "sys.h"
#include "util.h"
#include "glock.h"
+#include "quota.h"
+
+static struct shrinker qd_shrinker = {
+ .shrink = gfs2_shrink_qd_memory,
+ .seeks = DEFAULT_SEEKS,
+};
static void gfs2_init_inode_once(void *foo)
{
@@ -100,6 +106,8 @@ static int __init init_gfs2_fs(void)
if (!gfs2_quotad_cachep)
goto fail;
+ register_shrinker(&qd_shrinker);
+
error = register_filesystem(&gfs2_fs_type);
if (error)
goto fail;
@@ -117,6 +125,7 @@ static int __init init_gfs2_fs(void)
fail_unregister:
unregister_filesystem(&gfs2_fs_type);
fail:
+ unregister_shrinker(&qd_shrinker);
gfs2_glock_exit();
if (gfs2_quotad_cachep)
@@ -145,6 +154,7 @@ fail:
static void __exit exit_gfs2_fs(void)
{
+ unregister_shrinker(&qd_shrinker);
gfs2_glock_exit();
gfs2_unregister_debugfs();
unregister_filesystem(&gfs2_fs_type);
diff --git a/fs/gfs2/ops_address.c b/fs/gfs2/ops_address.c
index 4ddab67..dde4ead 100644
--- a/fs/gfs2/ops_address.c
+++ b/fs/gfs2/ops_address.c
@@ -442,6 +442,7 @@ static int stuffed_readpage(struct gfs2_inode *ip, struct page *page)
*/
if (unlikely(page->index)) {
zero_user(page, 0, PAGE_CACHE_SIZE);
+ SetPageUptodate(page);
return 0;
}
diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index 3eb49ed..530d3f6 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -63,7 +63,6 @@ static void gfs2_tune_init(struct gfs2_tune *gt)
gt->gt_quota_warn_period = 10;
gt->gt_quota_scale_num = 1;
gt->gt_quota_scale_den = 1;
- gt->gt_quota_cache_secs = 300;
gt->gt_quota_quantum = 60;
gt->gt_new_files_jdata = 0;
gt->gt_max_readahead = 1 << 18;
diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c
index b08d096..2ada6e1 100644
--- a/fs/gfs2/quota.c
+++ b/fs/gfs2/quota.c
@@ -80,6 +80,53 @@ struct gfs2_quota_change_host {
u32 qc_id;
};
+static LIST_HEAD(qd_lru_list);
+static atomic_t qd_lru_count = ATOMIC_INIT(0);
+static spinlock_t qd_lru_lock = SPIN_LOCK_UNLOCKED;
+
+int gfs2_shrink_qd_memory(int nr, gfp_t gfp_mask)
+{
+ struct gfs2_quota_data *qd;
+ struct gfs2_sbd *sdp;
+
+ if (nr == 0)
+ goto out;
+
+ if (!(gfp_mask & __GFP_FS))
+ return -1;
+
+ spin_lock(&qd_lru_lock);
+ while (nr && !list_empty(&qd_lru_list)) {
+ qd = list_entry(qd_lru_list.next,
+ struct gfs2_quota_data, qd_reclaim);
+ sdp = qd->qd_gl->gl_sbd;
+
+ /* Free from the filesystem-specific list */
+ list_del(&qd->qd_list);
+
+ spin_lock(&sdp->sd_quota_spin);
+ gfs2_assert_warn(sdp, !qd->qd_change);
+ gfs2_assert_warn(sdp, !qd->qd_slot_count);
+ gfs2_assert_warn(sdp, !qd->qd_bh_count);
+
+ gfs2_lvb_unhold(qd->qd_gl);
+ spin_unlock(&sdp->sd_quota_spin);
+ atomic_dec(&sdp->sd_quota_count);
+
+ /* Delete it from the common reclaim list */
+ list_del_init(&qd->qd_reclaim);
+ atomic_dec(&qd_lru_count);
+ spin_unlock(&qd_lru_lock);
+ kmem_cache_free(gfs2_quotad_cachep, qd);
+ spin_lock(&qd_lru_lock);
+ nr--;
+ }
+ spin_unlock(&qd_lru_lock);
+
+out:
+ return (atomic_read(&qd_lru_count) * sysctl_vfs_cache_pressure) / 100;
+}
+
static u64 qd2offset(struct gfs2_quota_data *qd)
{
u64 offset;
@@ -100,11 +147,12 @@ static int qd_alloc(struct gfs2_sbd *sdp, int user, u32 id,
if (!qd)
return -ENOMEM;
- qd->qd_count = 1;
+ atomic_set(&qd->qd_count, 1);
qd->qd_id = id;
if (user)
set_bit(QDF_USER, &qd->qd_flags);
qd->qd_slot = -1;
+ INIT_LIST_HEAD(&qd->qd_reclaim);
error = gfs2_glock_get(sdp, 2 * (u64)id + !user,
&gfs2_quota_glops, CREATE, &qd->qd_gl);
@@ -135,11 +183,17 @@ static int qd_get(struct gfs2_sbd *sdp, int user, u32 id, int create,
for (;;) {
found = 0;
- spin_lock(&sdp->sd_quota_spin);
+ spin_lock(&qd_lru_lock);
list_for_each_entry(qd, &sdp->sd_quota_list, qd_list) {
if (qd->qd_id == id &&
!test_bit(QDF_USER, &qd->qd_flags) == !user) {
- qd->qd_count++;
+ if (!atomic_read(&qd->qd_count) &&
+ !list_empty(&qd->qd_reclaim)) {
+ /* Remove it from reclaim list */
+ list_del_init(&qd->qd_reclaim);
+ atomic_dec(&qd_lru_count);
+ }
+ atomic_inc(&qd->qd_count);
found = 1;
break;
}
@@ -155,7 +209,7 @@ static int qd_get(struct gfs2_sbd *sdp, int user, u32 id, int create,
new_qd = NULL;
}
- spin_unlock(&sdp->sd_quota_spin);
+ spin_unlock(&qd_lru_lock);
if (qd || !create) {
if (new_qd) {
@@ -175,21 +229,18 @@ static int qd_get(struct gfs2_sbd *sdp, int user, u32 id, int create,
static void qd_hold(struct gfs2_quota_data *qd)
{
struct gfs2_sbd *sdp = qd->qd_gl->gl_sbd;
-
- spin_lock(&sdp->sd_quota_spin);
- gfs2_assert(sdp, qd->qd_count);
- qd->qd_count++;
- spin_unlock(&sdp->sd_quota_spin);
+ gfs2_assert(sdp, atomic_read(&qd->qd_count));
+ atomic_inc(&qd->qd_count);
}
static void qd_put(struct gfs2_quota_data *qd)
{
- struct gfs2_sbd *sdp = qd->qd_gl->gl_sbd;
- spin_lock(&sdp->sd_quota_spin);
- gfs2_assert(sdp, qd->qd_count);
- if (!--qd->qd_count)
- qd->qd_last_touched = jiffies;
- spin_unlock(&sdp->sd_quota_spin);
+ if (atomic_dec_and_lock(&qd->qd_count, &qd_lru_lock)) {
+ /* Add to the reclaim list */
+ list_add_tail(&qd->qd_reclaim, &qd_lru_list);
+ atomic_inc(&qd_lru_count);
+ spin_unlock(&qd_lru_lock);
+ }
}
static int slot_get(struct gfs2_quota_data *qd)
@@ -330,6 +381,7 @@ static int qd_fish(struct gfs2_sbd *sdp, struct gfs2_quota_data **qdp)
if (sdp->sd_vfs->s_flags & MS_RDONLY)
return 0;
+ spin_lock(&qd_lru_lock);
spin_lock(&sdp->sd_quota_spin);
list_for_each_entry(qd, &sdp->sd_quota_list, qd_list) {
@@ -341,8 +393,8 @@ static int qd_fish(struct gfs2_sbd *sdp, struct gfs2_quota_data **qdp)
list_move_tail(&qd->qd_list, &sdp->sd_quota_list);
set_bit(QDF_LOCKED, &qd->qd_flags);
- gfs2_assert_warn(sdp, qd->qd_count);
- qd->qd_count++;
+ gfs2_assert_warn(sdp, atomic_read(&qd->qd_count));
+ atomic_inc(&qd->qd_count);
qd->qd_change_sync = qd->qd_change;
gfs2_assert_warn(sdp, qd->qd_slot_count);
qd->qd_slot_count++;
@@ -355,6 +407,7 @@ static int qd_fish(struct gfs2_sbd *sdp, struct gfs2_quota_data **qdp)
qd = NULL;
spin_unlock(&sdp->sd_quota_spin);
+ spin_unlock(&qd_lru_lock);
if (qd) {
gfs2_assert_warn(sdp, qd->qd_change_sync);
@@ -379,24 +432,27 @@ static int qd_trylock(struct gfs2_quota_data *qd)
if (sdp->sd_vfs->s_flags & MS_RDONLY)
return 0;
+ spin_lock(&qd_lru_lock);
spin_lock(&sdp->sd_quota_spin);
if (test_bit(QDF_LOCKED, &qd->qd_flags) ||
!test_bit(QDF_CHANGE, &qd->qd_flags)) {
spin_unlock(&sdp->sd_quota_spin);
+ spin_unlock(&qd_lru_lock);
return 0;
}
list_move_tail(&qd->qd_list, &sdp->sd_quota_list);
set_bit(QDF_LOCKED, &qd->qd_flags);
- gfs2_assert_warn(sdp, qd->qd_count);
- qd->qd_count++;
+ gfs2_assert_warn(sdp, atomic_read(&qd->qd_count));
+ atomic_inc(&qd->qd_count);
qd->qd_change_sync = qd->qd_change;
gfs2_assert_warn(sdp, qd->qd_slot_count);
qd->qd_slot_count++;
spin_unlock(&sdp->sd_quota_spin);
+ spin_unlock(&qd_lru_lock);
gfs2_assert_warn(sdp, qd->qd_change_sync);
if (bh_get(qd)) {
@@ -802,8 +858,8 @@ restart:
loff_t pos;
gfs2_glock_dq_uninit(q_gh);
error = gfs2_glock_nq_init(qd->qd_gl,
- LM_ST_EXCLUSIVE, GL_NOCACHE,
- q_gh);
+ LM_ST_EXCLUSIVE, GL_NOCACHE,
+ q_gh);
if (error)
return error;
@@ -820,7 +876,6 @@ restart:
gfs2_glock_dq_uninit(&i_gh);
-
gfs2_quota_in(&q, buf);
qlvb = (struct gfs2_quota_lvb *)qd->qd_gl->gl_lvb;
qlvb->qb_magic = cpu_to_be32(GFS2_MAGIC);
@@ -1171,13 +1226,14 @@ int gfs2_quota_init(struct gfs2_sbd *sdp)
qd->qd_change = qc.qc_change;
qd->qd_slot = slot;
qd->qd_slot_count = 1;
- qd->qd_last_touched = jiffies;
+ spin_lock(&qd_lru_lock);
spin_lock(&sdp->sd_quota_spin);
gfs2_icbit_munge(sdp, sdp->sd_quota_bitmap, slot, 1);
+ spin_unlock(&sdp->sd_quota_spin);
list_add(&qd->qd_list, &sdp->sd_quota_list);
atomic_inc(&sdp->sd_quota_count);
- spin_unlock(&sdp->sd_quota_spin);
+ spin_unlock(&qd_lru_lock);
found++;
}
@@ -1197,61 +1253,39 @@ fail:
return error;
}
-static void gfs2_quota_scan(struct gfs2_sbd *sdp)
-{
- struct gfs2_quota_data *qd, *safe;
- LIST_HEAD(dead);
-
- spin_lock(&sdp->sd_quota_spin);
- list_for_each_entry_safe(qd, safe, &sdp->sd_quota_list, qd_list) {
- if (!qd->qd_count &&
- time_after_eq(jiffies, qd->qd_last_touched +
- gfs2_tune_get(sdp, gt_quota_cache_secs) * HZ)) {
- list_move(&qd->qd_list, &dead);
- gfs2_assert_warn(sdp,
- atomic_read(&sdp->sd_quota_count) > 0);
- atomic_dec(&sdp->sd_quota_count);
- }
- }
- spin_unlock(&sdp->sd_quota_spin);
-
- while (!list_empty(&dead)) {
- qd = list_entry(dead.next, struct gfs2_quota_data, qd_list);
- list_del(&qd->qd_list);
-
- gfs2_assert_warn(sdp, !qd->qd_change);
- gfs2_assert_warn(sdp, !qd->qd_slot_count);
- gfs2_assert_warn(sdp, !qd->qd_bh_count);
-
- gfs2_lvb_unhold(qd->qd_gl);
- kmem_cache_free(gfs2_quotad_cachep, qd);
- }
-}
-
void gfs2_quota_cleanup(struct gfs2_sbd *sdp)
{
struct list_head *head = &sdp->sd_quota_list;
struct gfs2_quota_data *qd;
unsigned int x;
- spin_lock(&sdp->sd_quota_spin);
+ spin_lock(&qd_lru_lock);
while (!list_empty(head)) {
qd = list_entry(head->prev, struct gfs2_quota_data, qd_list);
- if (qd->qd_count > 1 ||
- (qd->qd_count && !test_bit(QDF_CHANGE, &qd->qd_flags))) {
- list_move(&qd->qd_list, head);
+ spin_lock(&sdp->sd_quota_spin);
+ if (atomic_read(&qd->qd_count) > 1 ||
+ (atomic_read(&qd->qd_count) &&
+ !test_bit(QDF_CHANGE, &qd->qd_flags))) {
spin_unlock(&sdp->sd_quota_spin);
+ list_move(&qd->qd_list, head);
+ spin_unlock(&qd_lru_lock);
schedule();
- spin_lock(&sdp->sd_quota_spin);
+ spin_lock(&qd_lru_lock);
continue;
}
+ spin_unlock(&sdp->sd_quota_spin);
list_del(&qd->qd_list);
+ /* Also remove if this qd exists in the reclaim list */
+ if (!list_empty(&qd->qd_reclaim)) {
+ list_del_init(&qd->qd_reclaim);
+ atomic_dec(&qd_lru_count);
+ }
atomic_dec(&sdp->sd_quota_count);
- spin_unlock(&sdp->sd_quota_spin);
+ spin_unlock(&qd_lru_lock);
- if (!qd->qd_count) {
+ if (!atomic_read(&qd->qd_count)) {
gfs2_assert_warn(sdp, !qd->qd_change);
gfs2_assert_warn(sdp, !qd->qd_slot_count);
} else
@@ -1261,9 +1295,9 @@ void gfs2_quota_cleanup(struct gfs2_sbd *sdp)
gfs2_lvb_unhold(qd->qd_gl);
kmem_cache_free(gfs2_quotad_cachep, qd);
- spin_lock(&sdp->sd_quota_spin);
+ spin_lock(&qd_lru_lock);
}
- spin_unlock(&sdp->sd_quota_spin);
+ spin_unlock(&qd_lru_lock);
gfs2_assert_warn(sdp, !atomic_read(&sdp->sd_quota_count));
@@ -1341,9 +1375,6 @@ int gfs2_quotad(void *data)
quotad_check_timeo(sdp, "sync", gfs2_quota_sync, t,
"ad_timeo, &tune->gt_quota_quantum);
- /* FIXME: This should be turned into a shrinker */
- gfs2_quota_scan(sdp);
-
/* Check for & recover partially truncated inodes */
quotad_check_trunc_list(sdp);
diff --git a/fs/gfs2/quota.h b/fs/gfs2/quota.h
index cec9032..0fa5fa6 100644
--- a/fs/gfs2/quota.h
+++ b/fs/gfs2/quota.h
@@ -49,4 +49,6 @@ static inline int gfs2_quota_lock_check(struct gfs2_inode *ip)
return ret;
}
+extern int gfs2_shrink_qd_memory(int nr, gfp_t gfp_mask);
+
#endif /* __QUOTA_DOT_H__ */
diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c
index 26c1fa7..a58a120 100644
--- a/fs/gfs2/sys.c
+++ b/fs/gfs2/sys.c
@@ -373,7 +373,6 @@ TUNE_ATTR(complain_secs, 0);
TUNE_ATTR(statfs_slow, 0);
TUNE_ATTR(new_files_jdata, 0);
TUNE_ATTR(quota_simul_sync, 1);
-TUNE_ATTR(quota_cache_secs, 1);
TUNE_ATTR(stall_secs, 1);
TUNE_ATTR(statfs_quantum, 1);
TUNE_ATTR_DAEMON(recoverd_secs, recoverd_process);
@@ -389,7 +388,6 @@ static struct attribute *tune_attrs[] = {
&tune_attr_complain_secs.attr,
&tune_attr_statfs_slow.attr,
&tune_attr_quota_simul_sync.attr,
- &tune_attr_quota_cache_secs.attr,
&tune_attr_stall_secs.attr,
&tune_attr_statfs_quantum.attr,
&tune_attr_recoverd_secs.attr,
--
1.6.0.3
^ permalink raw reply related [flat|nested] 34+ messages in thread
* [Cluster-devel] [PATCH 04/18] GFS2: Remove "double" locking in quota
2009-03-18 12:23 ` [Cluster-devel] [PATCH 03/18] GFS2: change gfs2_quota_scan into a shrinker swhiteho
@ 2009-03-18 12:23 ` swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 05/18] GFS2: Merge lock_dlm module into GFS2 swhiteho
0 siblings, 1 reply; 34+ messages in thread
From: swhiteho @ 2009-03-18 12:23 UTC (permalink / raw)
To: cluster-devel.redhat.com
From: Steven Whitehouse <swhiteho@redhat.com>
We only really need a single spin lock for the quota data, so
lets just use the lru lock for now.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Cc: Abhijith Das <adas@redhat.com>
diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h
index 592aa50..a0117d6 100644
--- a/fs/gfs2/incore.h
+++ b/fs/gfs2/incore.h
@@ -540,7 +540,6 @@ struct gfs2_sbd {
struct list_head sd_quota_list;
atomic_t sd_quota_count;
- spinlock_t sd_quota_spin;
struct mutex sd_quota_mutex;
wait_queue_head_t sd_quota_wait;
struct list_head sd_trunc_list;
diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index 530d3f6..402b6a2 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -98,7 +98,6 @@ static struct gfs2_sbd *init_sbd(struct super_block *sb)
mutex_init(&sdp->sd_jindex_mutex);
INIT_LIST_HEAD(&sdp->sd_quota_list);
- spin_lock_init(&sdp->sd_quota_spin);
mutex_init(&sdp->sd_quota_mutex);
init_waitqueue_head(&sdp->sd_quota_wait);
INIT_LIST_HEAD(&sdp->sd_trunc_list);
diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c
index 2ada6e1..e8ef0f8 100644
--- a/fs/gfs2/quota.c
+++ b/fs/gfs2/quota.c
@@ -104,13 +104,11 @@ int gfs2_shrink_qd_memory(int nr, gfp_t gfp_mask)
/* Free from the filesystem-specific list */
list_del(&qd->qd_list);
- spin_lock(&sdp->sd_quota_spin);
gfs2_assert_warn(sdp, !qd->qd_change);
gfs2_assert_warn(sdp, !qd->qd_slot_count);
gfs2_assert_warn(sdp, !qd->qd_bh_count);
gfs2_lvb_unhold(qd->qd_gl);
- spin_unlock(&sdp->sd_quota_spin);
atomic_dec(&sdp->sd_quota_count);
/* Delete it from the common reclaim list */
@@ -249,10 +247,10 @@ static int slot_get(struct gfs2_quota_data *qd)
unsigned int c, o = 0, b;
unsigned char byte = 0;
- spin_lock(&sdp->sd_quota_spin);
+ spin_lock(&qd_lru_lock);
if (qd->qd_slot_count++) {
- spin_unlock(&sdp->sd_quota_spin);
+ spin_unlock(&qd_lru_lock);
return 0;
}
@@ -276,13 +274,13 @@ found:
sdp->sd_quota_bitmap[c][o] |= 1 << b;
- spin_unlock(&sdp->sd_quota_spin);
+ spin_unlock(&qd_lru_lock);
return 0;
fail:
qd->qd_slot_count--;
- spin_unlock(&sdp->sd_quota_spin);
+ spin_unlock(&qd_lru_lock);
return -ENOSPC;
}
@@ -290,23 +288,23 @@ static void slot_hold(struct gfs2_quota_data *qd)
{
struct gfs2_sbd *sdp = qd->qd_gl->gl_sbd;
- spin_lock(&sdp->sd_quota_spin);
+ spin_lock(&qd_lru_lock);
gfs2_assert(sdp, qd->qd_slot_count);
qd->qd_slot_count++;
- spin_unlock(&sdp->sd_quota_spin);
+ spin_unlock(&qd_lru_lock);
}
static void slot_put(struct gfs2_quota_data *qd)
{
struct gfs2_sbd *sdp = qd->qd_gl->gl_sbd;
- spin_lock(&sdp->sd_quota_spin);
+ spin_lock(&qd_lru_lock);
gfs2_assert(sdp, qd->qd_slot_count);
if (!--qd->qd_slot_count) {
gfs2_icbit_munge(sdp, sdp->sd_quota_bitmap, qd->qd_slot, 0);
qd->qd_slot = -1;
}
- spin_unlock(&sdp->sd_quota_spin);
+ spin_unlock(&qd_lru_lock);
}
static int bh_get(struct gfs2_quota_data *qd)
@@ -382,7 +380,6 @@ static int qd_fish(struct gfs2_sbd *sdp, struct gfs2_quota_data **qdp)
return 0;
spin_lock(&qd_lru_lock);
- spin_lock(&sdp->sd_quota_spin);
list_for_each_entry(qd, &sdp->sd_quota_list, qd_list) {
if (test_bit(QDF_LOCKED, &qd->qd_flags) ||
@@ -406,7 +403,6 @@ static int qd_fish(struct gfs2_sbd *sdp, struct gfs2_quota_data **qdp)
if (!found)
qd = NULL;
- spin_unlock(&sdp->sd_quota_spin);
spin_unlock(&qd_lru_lock);
if (qd) {
@@ -433,11 +429,9 @@ static int qd_trylock(struct gfs2_quota_data *qd)
return 0;
spin_lock(&qd_lru_lock);
- spin_lock(&sdp->sd_quota_spin);
if (test_bit(QDF_LOCKED, &qd->qd_flags) ||
!test_bit(QDF_CHANGE, &qd->qd_flags)) {
- spin_unlock(&sdp->sd_quota_spin);
spin_unlock(&qd_lru_lock);
return 0;
}
@@ -451,7 +445,6 @@ static int qd_trylock(struct gfs2_quota_data *qd)
gfs2_assert_warn(sdp, qd->qd_slot_count);
qd->qd_slot_count++;
- spin_unlock(&sdp->sd_quota_spin);
spin_unlock(&qd_lru_lock);
gfs2_assert_warn(sdp, qd->qd_change_sync);
@@ -612,9 +605,9 @@ static void do_qc(struct gfs2_quota_data *qd, s64 change)
x = be64_to_cpu(qc->qc_change) + change;
qc->qc_change = cpu_to_be64(x);
- spin_lock(&sdp->sd_quota_spin);
+ spin_lock(&qd_lru_lock);
qd->qd_change = x;
- spin_unlock(&sdp->sd_quota_spin);
+ spin_unlock(&qd_lru_lock);
if (!x) {
gfs2_assert_warn(sdp, test_bit(QDF_CHANGE, &qd->qd_flags));
@@ -945,9 +938,9 @@ static int need_sync(struct gfs2_quota_data *qd)
if (!qd->qd_qb.qb_limit)
return 0;
- spin_lock(&sdp->sd_quota_spin);
+ spin_lock(&qd_lru_lock);
value = qd->qd_change;
- spin_unlock(&sdp->sd_quota_spin);
+ spin_unlock(&qd_lru_lock);
spin_lock(>->gt_spin);
num = gt->gt_quota_scale_num;
@@ -1040,9 +1033,9 @@ int gfs2_quota_check(struct gfs2_inode *ip, u32 uid, u32 gid)
continue;
value = (s64)be64_to_cpu(qd->qd_qb.qb_value);
- spin_lock(&sdp->sd_quota_spin);
+ spin_lock(&qd_lru_lock);
value += qd->qd_change;
- spin_unlock(&sdp->sd_quota_spin);
+ spin_unlock(&qd_lru_lock);
if (be64_to_cpu(qd->qd_qb.qb_limit) && (s64)be64_to_cpu(qd->qd_qb.qb_limit) < value) {
print_message(qd, "exceeded");
@@ -1228,9 +1221,7 @@ int gfs2_quota_init(struct gfs2_sbd *sdp)
qd->qd_slot_count = 1;
spin_lock(&qd_lru_lock);
- spin_lock(&sdp->sd_quota_spin);
gfs2_icbit_munge(sdp, sdp->sd_quota_bitmap, slot, 1);
- spin_unlock(&sdp->sd_quota_spin);
list_add(&qd->qd_list, &sdp->sd_quota_list);
atomic_inc(&sdp->sd_quota_count);
spin_unlock(&qd_lru_lock);
@@ -1263,18 +1254,15 @@ void gfs2_quota_cleanup(struct gfs2_sbd *sdp)
while (!list_empty(head)) {
qd = list_entry(head->prev, struct gfs2_quota_data, qd_list);
- spin_lock(&sdp->sd_quota_spin);
if (atomic_read(&qd->qd_count) > 1 ||
(atomic_read(&qd->qd_count) &&
!test_bit(QDF_CHANGE, &qd->qd_flags))) {
- spin_unlock(&sdp->sd_quota_spin);
list_move(&qd->qd_list, head);
spin_unlock(&qd_lru_lock);
schedule();
spin_lock(&qd_lru_lock);
continue;
}
- spin_unlock(&sdp->sd_quota_spin);
list_del(&qd->qd_list);
/* Also remove if this qd exists in the reclaim list */
--
1.6.0.3
^ permalink raw reply related [flat|nested] 34+ messages in thread
* [Cluster-devel] [PATCH 05/18] GFS2: Merge lock_dlm module into GFS2
2009-03-18 12:23 ` [Cluster-devel] [PATCH 04/18] GFS2: Remove "double" locking in quota swhiteho
@ 2009-03-18 12:23 ` swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 06/18] GFS2: Remove unused field from glock swhiteho
0 siblings, 1 reply; 34+ messages in thread
From: swhiteho @ 2009-03-18 12:23 UTC (permalink / raw)
To: cluster-devel.redhat.com
From: Steven Whitehouse <swhiteho@redhat.com>
This is the big patch that I've been working on for some time
now. There are many reasons for wanting to make this change
such as:
o Reducing overhead by eliminating duplicated fields between structures
o Simplifcation of the code (reduces the code size by a fair bit)
o The locking interface is now the DLM interface itself as proposed
some time ago.
o Fewer lookups of glocks when processing replies from the DLM
o Fewer memory allocations/deallocations for each glock
o Scope to do further optimisations in the future (but this patch is
more than big enough for now!)
Please note that (a) this patch relates to the lock_dlm module and
not the DLM itself, that is still a separate module; and (b) that
we retain the ability to build GFS2 as a standalone single node
filesystem with out requiring the DLM.
This patch needs a lot of testing, hence my keeping it I restarted
my -git tree after the last merge window. That way, this has the maximum
exposure before its merged. This is (modulo a few minor bug fixes) the
same patch that I've been posting on and off the the last three months
and its passed a number of different tests so far.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
diff --git a/fs/gfs2/Kconfig b/fs/gfs2/Kconfig
index e563a64..3a981b7 100644
--- a/fs/gfs2/Kconfig
+++ b/fs/gfs2/Kconfig
@@ -1,6 +1,10 @@
config GFS2_FS
tristate "GFS2 file system support"
depends on EXPERIMENTAL && (64BIT || LBD)
+ select DLM if GFS2_FS_LOCKING_DLM
+ select CONFIGFS_FS if GFS2_FS_LOCKING_DLM
+ select SYSFS if GFS2_FS_LOCKING_DLM
+ select IP_SCTP if DLM_SCTP
select FS_POSIX_ACL
select CRC32
help
@@ -18,17 +22,16 @@ config GFS2_FS
the locking module below. Documentation and utilities for GFS2 can
be found here: http://sources.redhat.com/cluster
- The "nolock" lock module is now built in to GFS2 by default.
+ The "nolock" lock module is now built in to GFS2 by default. If
+ you want to use the DLM, be sure to enable HOTPLUG and IPv4/6
+ networking.
config GFS2_FS_LOCKING_DLM
- tristate "GFS2 DLM locking module"
- depends on GFS2_FS && SYSFS && NET && INET && (IPV6 || IPV6=n)
- select IP_SCTP if DLM_SCTP
- select CONFIGFS_FS
- select DLM
+ bool "GFS2 DLM locking"
+ depends on (GFS2_FS!=n) && NET && INET && (IPV6 || IPV6=n) && HOTPLUG
help
Multiple node locking module for GFS2
- Most users of GFS2 will require this module. It provides the locking
+ Most users of GFS2 will require this. It provides the locking
interface between GFS2 and the DLM, which is required to use GFS2
in a cluster environment.
diff --git a/fs/gfs2/Makefile b/fs/gfs2/Makefile
index c1b4ec6..a851ea4 100644
--- a/fs/gfs2/Makefile
+++ b/fs/gfs2/Makefile
@@ -1,9 +1,9 @@
obj-$(CONFIG_GFS2_FS) += gfs2.o
gfs2-y := acl.o bmap.o dir.o eaops.o eattr.o glock.o \
- glops.o inode.o log.o lops.o locking.o main.o meta_io.o \
+ glops.o inode.o log.o lops.o main.o meta_io.o \
mount.o ops_address.o ops_dentry.o ops_export.o ops_file.o \
ops_fstype.o ops_inode.o ops_super.o quota.o \
recovery.o rgrp.o super.o sys.o trans.o util.o
-obj-$(CONFIG_GFS2_FS_LOCKING_DLM) += locking/dlm/
+gfs2-$(CONFIG_GFS2_FS_LOCKING_DLM) += lock_dlm.o
diff --git a/fs/gfs2/acl.c b/fs/gfs2/acl.c
index e335dce..43764f4 100644
--- a/fs/gfs2/acl.c
+++ b/fs/gfs2/acl.c
@@ -15,7 +15,6 @@
#include <linux/posix_acl.h>
#include <linux/posix_acl_xattr.h>
#include <linux/gfs2_ondisk.h>
-#include <linux/lm_interface.h>
#include "gfs2.h"
#include "incore.h"
diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c
index 11ffc56..3a5d3f8 100644
--- a/fs/gfs2/bmap.c
+++ b/fs/gfs2/bmap.c
@@ -13,7 +13,6 @@
#include <linux/buffer_head.h>
#include <linux/gfs2_ondisk.h>
#include <linux/crc32.h>
-#include <linux/lm_interface.h>
#include "gfs2.h"
#include "incore.h"
diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c
index b7c8e5c..aef4d0c 100644
--- a/fs/gfs2/dir.c
+++ b/fs/gfs2/dir.c
@@ -60,7 +60,6 @@
#include <linux/gfs2_ondisk.h>
#include <linux/crc32.h>
#include <linux/vmalloc.h>
-#include <linux/lm_interface.h>
#include "gfs2.h"
#include "incore.h"
diff --git a/fs/gfs2/eaops.c b/fs/gfs2/eaops.c
index f114ba2..dee9b03 100644
--- a/fs/gfs2/eaops.c
+++ b/fs/gfs2/eaops.c
@@ -14,7 +14,6 @@
#include <linux/capability.h>
#include <linux/xattr.h>
#include <linux/gfs2_ondisk.h>
-#include <linux/lm_interface.h>
#include <asm/uaccess.h>
#include "gfs2.h"
diff --git a/fs/gfs2/eattr.c b/fs/gfs2/eattr.c
index 0d1c76d..899763a 100644
--- a/fs/gfs2/eattr.c
+++ b/fs/gfs2/eattr.c
@@ -13,7 +13,6 @@
#include <linux/buffer_head.h>
#include <linux/xattr.h>
#include <linux/gfs2_ondisk.h>
-#include <linux/lm_interface.h>
#include <asm/uaccess.h>
#include "gfs2.h"
diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c
index 6b983ae..cd200a5 100644
--- a/fs/gfs2/glock.c
+++ b/fs/gfs2/glock.c
@@ -10,7 +10,6 @@
#include <linux/sched.h>
#include <linux/slab.h>
#include <linux/spinlock.h>
-#include <linux/completion.h>
#include <linux/buffer_head.h>
#include <linux/delay.h>
#include <linux/sort.h>
@@ -18,7 +17,6 @@
#include <linux/kallsyms.h>
#include <linux/gfs2_ondisk.h>
#include <linux/list.h>
-#include <linux/lm_interface.h>
#include <linux/wait.h>
#include <linux/module.h>
#include <linux/rwsem.h>
@@ -155,13 +153,10 @@ static void glock_free(struct gfs2_glock *gl)
struct gfs2_sbd *sdp = gl->gl_sbd;
struct inode *aspace = gl->gl_aspace;
- if (sdp->sd_lockstruct.ls_ops->lm_put_lock)
- sdp->sd_lockstruct.ls_ops->lm_put_lock(gl->gl_lock);
-
if (aspace)
gfs2_aspace_put(aspace);
- kmem_cache_free(gfs2_glock_cachep, gl);
+ sdp->sd_lockstruct.ls_ops->lm_put_lock(gfs2_glock_cachep, gl);
}
/**
@@ -211,7 +206,6 @@ int gfs2_glock_put(struct gfs2_glock *gl)
atomic_dec(&lru_count);
}
spin_unlock(&lru_lock);
- GLOCK_BUG_ON(gl, gl->gl_state != LM_ST_UNLOCKED);
GLOCK_BUG_ON(gl, !list_empty(&gl->gl_lru));
GLOCK_BUG_ON(gl, !list_empty(&gl->gl_holders));
glock_free(gl);
@@ -256,27 +250,6 @@ static struct gfs2_glock *search_bucket(unsigned int hash,
}
/**
- * gfs2_glock_find() - Find glock by lock number
- * @sdp: The GFS2 superblock
- * @name: The lock name
- *
- * Returns: NULL, or the struct gfs2_glock with the requested number
- */
-
-static struct gfs2_glock *gfs2_glock_find(const struct gfs2_sbd *sdp,
- const struct lm_lockname *name)
-{
- unsigned int hash = gl_hash(sdp, name);
- struct gfs2_glock *gl;
-
- read_lock(gl_lock_addr(hash));
- gl = search_bucket(hash, sdp, name);
- read_unlock(gl_lock_addr(hash));
-
- return gl;
-}
-
-/**
* may_grant - check if its ok to grant a new lock
* @gl: The glock
* @gh: The lock request which we wish to grant
@@ -523,7 +496,7 @@ out_locked:
}
static unsigned int gfs2_lm_lock(struct gfs2_sbd *sdp, void *lock,
- unsigned int cur_state, unsigned int req_state,
+ unsigned int req_state,
unsigned int flags)
{
int ret = LM_OUT_ERROR;
@@ -532,7 +505,7 @@ static unsigned int gfs2_lm_lock(struct gfs2_sbd *sdp, void *lock,
return req_state == LM_ST_UNLOCKED ? 0 : req_state;
if (likely(!test_bit(SDF_SHUTDOWN, &sdp->sd_flags)))
- ret = sdp->sd_lockstruct.ls_ops->lm_lock(lock, cur_state,
+ ret = sdp->sd_lockstruct.ls_ops->lm_lock(lock,
req_state, flags);
return ret;
}
@@ -575,7 +548,7 @@ __acquires(&gl->gl_spin)
gl->gl_state == LM_ST_DEFERRED) &&
!(lck_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB)))
lck_flags |= LM_FLAG_TRY_1CB;
- ret = gfs2_lm_lock(sdp, gl->gl_lock, gl->gl_state, target, lck_flags);
+ ret = gfs2_lm_lock(sdp, gl, target, lck_flags);
if (!(ret & LM_OUT_ASYNC)) {
finish_xmote(gl, ret);
@@ -681,18 +654,6 @@ static void glock_work_func(struct work_struct *work)
gfs2_glock_put(gl);
}
-static int gfs2_lm_get_lock(struct gfs2_sbd *sdp, struct lm_lockname *name,
- void **lockp)
-{
- int error = -EIO;
- if (!sdp->sd_lockstruct.ls_ops->lm_get_lock)
- return 0;
- if (likely(!test_bit(SDF_SHUTDOWN, &sdp->sd_flags)))
- error = sdp->sd_lockstruct.ls_ops->lm_get_lock(
- sdp->sd_lockstruct.ls_lockspace, name, lockp);
- return error;
-}
-
/**
* gfs2_glock_get() - Get a glock, or create one if one doesn't exist
* @sdp: The GFS2 superblock
@@ -736,6 +697,9 @@ int gfs2_glock_get(struct gfs2_sbd *sdp, u64 number,
gl->gl_demote_state = LM_ST_EXCLUSIVE;
gl->gl_hash = hash;
gl->gl_ops = glops;
+ snprintf(gl->gl_strname, GDLM_STRNAME_BYTES, "%8x%16llx", name.ln_type, (unsigned long long)number);
+ memset(&gl->gl_lksb, 0, sizeof(struct dlm_lksb));
+ gl->gl_lksb.sb_lvbptr = gl->gl_lvb;
gl->gl_stamp = jiffies;
gl->gl_tchange = jiffies;
gl->gl_object = NULL;
@@ -753,10 +717,6 @@ int gfs2_glock_get(struct gfs2_sbd *sdp, u64 number,
}
}
- error = gfs2_lm_get_lock(sdp, &name, &gl->gl_lock);
- if (error)
- goto fail_aspace;
-
write_lock(gl_lock_addr(hash));
tmp = search_bucket(hash, sdp, &name);
if (tmp) {
@@ -772,9 +732,6 @@ int gfs2_glock_get(struct gfs2_sbd *sdp, u64 number,
return 0;
-fail_aspace:
- if (gl->gl_aspace)
- gfs2_aspace_put(gl->gl_aspace);
fail:
kmem_cache_free(gfs2_glock_cachep, gl);
return error;
@@ -966,7 +923,7 @@ do_cancel:
if (!(gh->gh_flags & LM_FLAG_PRIORITY)) {
spin_unlock(&gl->gl_spin);
if (sdp->sd_lockstruct.ls_ops->lm_cancel)
- sdp->sd_lockstruct.ls_ops->lm_cancel(gl->gl_lock);
+ sdp->sd_lockstruct.ls_ops->lm_cancel(gl);
spin_lock(&gl->gl_spin);
}
return;
@@ -1240,70 +1197,13 @@ void gfs2_glock_dq_uninit_m(unsigned int num_gh, struct gfs2_holder *ghs)
gfs2_glock_dq_uninit(&ghs[x]);
}
-static int gfs2_lm_hold_lvb(struct gfs2_sbd *sdp, void *lock, char **lvbp)
+void gfs2_glock_cb(struct gfs2_glock *gl, unsigned int state)
{
- int error = -EIO;
- if (!sdp->sd_lockstruct.ls_ops->lm_hold_lvb)
- return 0;
- if (likely(!test_bit(SDF_SHUTDOWN, &sdp->sd_flags)))
- error = sdp->sd_lockstruct.ls_ops->lm_hold_lvb(lock, lvbp);
- return error;
-}
-
-/**
- * gfs2_lvb_hold - attach a LVB from a glock
- * @gl: The glock in question
- *
- */
-
-int gfs2_lvb_hold(struct gfs2_glock *gl)
-{
- int error;
-
- if (!atomic_read(&gl->gl_lvb_count)) {
- error = gfs2_lm_hold_lvb(gl->gl_sbd, gl->gl_lock, &gl->gl_lvb);
- if (error)
- return error;
- gfs2_glock_hold(gl);
- }
- atomic_inc(&gl->gl_lvb_count);
-
- return 0;
-}
-
-/**
- * gfs2_lvb_unhold - detach a LVB from a glock
- * @gl: The glock in question
- *
- */
-
-void gfs2_lvb_unhold(struct gfs2_glock *gl)
-{
- struct gfs2_sbd *sdp = gl->gl_sbd;
-
- gfs2_glock_hold(gl);
- gfs2_assert(gl->gl_sbd, atomic_read(&gl->gl_lvb_count) > 0);
- if (atomic_dec_and_test(&gl->gl_lvb_count)) {
- if (sdp->sd_lockstruct.ls_ops->lm_unhold_lvb)
- sdp->sd_lockstruct.ls_ops->lm_unhold_lvb(gl->gl_lock, gl->gl_lvb);
- gl->gl_lvb = NULL;
- gfs2_glock_put(gl);
- }
- gfs2_glock_put(gl);
-}
-
-static void blocking_cb(struct gfs2_sbd *sdp, struct lm_lockname *name,
- unsigned int state)
-{
- struct gfs2_glock *gl;
unsigned long delay = 0;
unsigned long holdtime;
unsigned long now = jiffies;
- gl = gfs2_glock_find(sdp, name);
- if (!gl)
- return;
-
+ gfs2_glock_hold(gl);
holdtime = gl->gl_tchange + gl->gl_ops->go_min_hold_time;
if (time_before(now, holdtime))
delay = holdtime - now;
@@ -1317,74 +1217,37 @@ static void blocking_cb(struct gfs2_sbd *sdp, struct lm_lockname *name,
gfs2_glock_put(gl);
}
-static void gfs2_jdesc_make_dirty(struct gfs2_sbd *sdp, unsigned int jid)
-{
- struct gfs2_jdesc *jd;
-
- spin_lock(&sdp->sd_jindex_spin);
- list_for_each_entry(jd, &sdp->sd_jindex_list, jd_list) {
- if (jd->jd_jid != jid)
- continue;
- jd->jd_dirty = 1;
- break;
- }
- spin_unlock(&sdp->sd_jindex_spin);
-}
-
/**
- * gfs2_glock_cb - Callback used by locking module
- * @sdp: Pointer to the superblock
- * @type: Type of callback
- * @data: Type dependent data pointer
+ * gfs2_glock_complete - Callback used by locking
+ * @gl: Pointer to the glock
+ * @ret: The return value from the dlm
*
- * Called by the locking module when it wants to tell us something.
- * Either we need to drop a lock, one of our ASYNC requests completed, or
- * a journal from another client needs to be recovered.
*/
-void gfs2_glock_cb(void *cb_data, unsigned int type, void *data)
+void gfs2_glock_complete(struct gfs2_glock *gl, int ret)
{
- struct gfs2_sbd *sdp = cb_data;
-
- switch (type) {
- case LM_CB_NEED_E:
- blocking_cb(sdp, data, LM_ST_UNLOCKED);
- return;
-
- case LM_CB_NEED_D:
- blocking_cb(sdp, data, LM_ST_DEFERRED);
- return;
-
- case LM_CB_NEED_S:
- blocking_cb(sdp, data, LM_ST_SHARED);
- return;
-
- case LM_CB_ASYNC: {
- struct lm_async_cb *async = data;
- struct gfs2_glock *gl;
-
- down_read(&gfs2_umount_flush_sem);
- gl = gfs2_glock_find(sdp, &async->lc_name);
- if (gfs2_assert_warn(sdp, gl))
+ struct lm_lockstruct *ls = &gl->gl_sbd->sd_lockstruct;
+ down_read(&gfs2_umount_flush_sem);
+ gl->gl_reply = ret;
+ if (unlikely(test_bit(DFL_BLOCK_LOCKS, &ls->ls_flags))) {
+ struct gfs2_holder *gh;
+ spin_lock(&gl->gl_spin);
+ gh = find_first_waiter(gl);
+ if ((!(gh && (gh->gh_flags & LM_FLAG_NOEXP)) &&
+ (gl->gl_target != LM_ST_UNLOCKED)) ||
+ ((ret & ~LM_OUT_ST_MASK) != 0))
+ set_bit(GLF_FROZEN, &gl->gl_flags);
+ spin_unlock(&gl->gl_spin);
+ if (test_bit(GLF_FROZEN, &gl->gl_flags)) {
+ up_read(&gfs2_umount_flush_sem);
return;
- gl->gl_reply = async->lc_ret;
- set_bit(GLF_REPLY_PENDING, &gl->gl_flags);
- if (queue_delayed_work(glock_workqueue, &gl->gl_work, 0) == 0)
- gfs2_glock_put(gl);
- up_read(&gfs2_umount_flush_sem);
- return;
- }
-
- case LM_CB_NEED_RECOVERY:
- gfs2_jdesc_make_dirty(sdp, *(unsigned int *)data);
- if (sdp->sd_recoverd_process)
- wake_up_process(sdp->sd_recoverd_process);
- return;
-
- default:
- gfs2_assert_warn(sdp, 0);
- return;
+ }
}
+ set_bit(GLF_REPLY_PENDING, &gl->gl_flags);
+ gfs2_glock_hold(gl);
+ if (queue_delayed_work(glock_workqueue, &gl->gl_work, 0) == 0)
+ gfs2_glock_put(gl);
+ up_read(&gfs2_umount_flush_sem);
}
/**
@@ -1515,6 +1378,27 @@ out:
return has_entries;
}
+
+/**
+ * thaw_glock - thaw out a glock which has an unprocessed reply waiting
+ * @gl: The glock to thaw
+ *
+ * N.B. When we freeze a glock, we leave a ref to the glock outstanding,
+ * so this has to result in the ref count being dropped by one.
+ */
+
+static void thaw_glock(struct gfs2_glock *gl)
+{
+ if (!test_and_clear_bit(GLF_FROZEN, &gl->gl_flags))
+ return;
+ down_read(&gfs2_umount_flush_sem);
+ set_bit(GLF_REPLY_PENDING, &gl->gl_flags);
+ gfs2_glock_hold(gl);
+ if (queue_delayed_work(glock_workqueue, &gl->gl_work, 0) == 0)
+ gfs2_glock_put(gl);
+ up_read(&gfs2_umount_flush_sem);
+}
+
/**
* clear_glock - look at a glock and see if we can free it from glock cache
* @gl: the glock to look at
@@ -1540,6 +1424,20 @@ static void clear_glock(struct gfs2_glock *gl)
}
/**
+ * gfs2_glock_thaw - Thaw any frozen glocks
+ * @sdp: The super block
+ *
+ */
+
+void gfs2_glock_thaw(struct gfs2_sbd *sdp)
+{
+ unsigned x;
+
+ for (x = 0; x < GFS2_GL_HASH_SIZE; x++)
+ examine_bucket(thaw_glock, sdp, x);
+}
+
+/**
* gfs2_gl_hash_clear - Empty out the glock hash table
* @sdp: the filesystem
* @wait: wait until it's all gone
@@ -1619,7 +1517,7 @@ static const char *hflags2str(char *buf, unsigned flags, unsigned long iflags)
if (flags & LM_FLAG_NOEXP)
*p++ = 'e';
if (flags & LM_FLAG_ANY)
- *p++ = 'a';
+ *p++ = 'A';
if (flags & LM_FLAG_PRIORITY)
*p++ = 'p';
if (flags & GL_ASYNC)
@@ -1683,6 +1581,10 @@ static const char *gflags2str(char *buf, const unsigned long *gflags)
*p++ = 'i';
if (test_bit(GLF_REPLY_PENDING, gflags))
*p++ = 'r';
+ if (test_bit(GLF_INITIAL, gflags))
+ *p++ = 'i';
+ if (test_bit(GLF_FROZEN, gflags))
+ *p++ = 'F';
*p = 0;
return buf;
}
@@ -1717,14 +1619,13 @@ static int __dump_glock(struct seq_file *seq, const struct gfs2_glock *gl)
dtime *= 1000000/HZ; /* demote time in uSec */
if (!test_bit(GLF_DEMOTE, &gl->gl_flags))
dtime = 0;
- gfs2_print_dbg(seq, "G: s:%s n:%u/%llu f:%s t:%s d:%s/%llu l:%d a:%d r:%d\n",
+ gfs2_print_dbg(seq, "G: s:%s n:%u/%llu f:%s t:%s d:%s/%llu a:%d r:%d\n",
state2str(gl->gl_state),
gl->gl_name.ln_type,
(unsigned long long)gl->gl_name.ln_number,
gflags2str(gflags_buf, &gl->gl_flags),
state2str(gl->gl_target),
state2str(gl->gl_demote_state), dtime,
- atomic_read(&gl->gl_lvb_count),
atomic_read(&gl->gl_ail_count),
atomic_read(&gl->gl_ref));
diff --git a/fs/gfs2/glock.h b/fs/gfs2/glock.h
index 543ec7e..a602a28 100644
--- a/fs/gfs2/glock.h
+++ b/fs/gfs2/glock.h
@@ -11,15 +11,130 @@
#define __GLOCK_DOT_H__
#include <linux/sched.h>
+#include <linux/parser.h>
#include "incore.h"
-/* Flags for lock requests; used in gfs2_holder gh_flag field.
- From lm_interface.h:
+/* Options for hostdata parser */
+
+enum {
+ Opt_jid,
+ Opt_id,
+ Opt_first,
+ Opt_nodir,
+ Opt_err,
+};
+
+/*
+ * lm_lockname types
+ */
+
+#define LM_TYPE_RESERVED 0x00
+#define LM_TYPE_NONDISK 0x01
+#define LM_TYPE_INODE 0x02
+#define LM_TYPE_RGRP 0x03
+#define LM_TYPE_META 0x04
+#define LM_TYPE_IOPEN 0x05
+#define LM_TYPE_FLOCK 0x06
+#define LM_TYPE_PLOCK 0x07
+#define LM_TYPE_QUOTA 0x08
+#define LM_TYPE_JOURNAL 0x09
+
+/*
+ * lm_lock() states
+ *
+ * SHARED is compatible with SHARED, not with DEFERRED or EX.
+ * DEFERRED is compatible with DEFERRED, not with SHARED or EX.
+ */
+
+#define LM_ST_UNLOCKED 0
+#define LM_ST_EXCLUSIVE 1
+#define LM_ST_DEFERRED 2
+#define LM_ST_SHARED 3
+
+/*
+ * lm_lock() flags
+ *
+ * LM_FLAG_TRY
+ * Don't wait to acquire the lock if it can't be granted immediately.
+ *
+ * LM_FLAG_TRY_1CB
+ * Send one blocking callback if TRY is set and the lock is not granted.
+ *
+ * LM_FLAG_NOEXP
+ * GFS sets this flag on lock requests it makes while doing journal recovery.
+ * These special requests should not be blocked due to the recovery like
+ * ordinary locks would be.
+ *
+ * LM_FLAG_ANY
+ * A SHARED request may also be granted in DEFERRED, or a DEFERRED request may
+ * also be granted in SHARED. The preferred state is whichever is compatible
+ * with other granted locks, or the specified state if no other locks exist.
+ *
+ * LM_FLAG_PRIORITY
+ * Override fairness considerations. Suppose a lock is held in a shared state
+ * and there is a pending request for the deferred state. A shared lock
+ * request with the priority flag would be allowed to bypass the deferred
+ * request and directly join the other shared lock. A shared lock request
+ * without the priority flag might be forced to wait until the deferred
+ * requested had acquired and released the lock.
+ */
+
#define LM_FLAG_TRY 0x00000001
#define LM_FLAG_TRY_1CB 0x00000002
#define LM_FLAG_NOEXP 0x00000004
#define LM_FLAG_ANY 0x00000008
-#define LM_FLAG_PRIORITY 0x00000010 */
+#define LM_FLAG_PRIORITY 0x00000010
+#define GL_ASYNC 0x00000040
+#define GL_EXACT 0x00000080
+#define GL_SKIP 0x00000100
+#define GL_ATIME 0x00000200
+#define GL_NOCACHE 0x00000400
+
+/*
+ * lm_lock() and lm_async_cb return flags
+ *
+ * LM_OUT_ST_MASK
+ * Masks the lower two bits of lock state in the returned value.
+ *
+ * LM_OUT_CANCELED
+ * The lock request was canceled.
+ *
+ * LM_OUT_ASYNC
+ * The result of the request will be returned in an LM_CB_ASYNC callback.
+ *
+ */
+
+#define LM_OUT_ST_MASK 0x00000003
+#define LM_OUT_CANCELED 0x00000008
+#define LM_OUT_ASYNC 0x00000080
+#define LM_OUT_ERROR 0x00000100
+
+/*
+ * lm_recovery_done() messages
+ */
+
+#define LM_RD_GAVEUP 308
+#define LM_RD_SUCCESS 309
+
+#define GLR_TRYFAILED 13
+
+struct lm_lockops {
+ const char *lm_proto_name;
+ int (*lm_mount) (struct gfs2_sbd *sdp, const char *fsname);
+ void (*lm_unmount) (struct gfs2_sbd *sdp);
+ void (*lm_withdraw) (struct gfs2_sbd *sdp);
+ void (*lm_put_lock) (struct kmem_cache *cachep, void *gl);
+ unsigned int (*lm_lock) (struct gfs2_glock *gl,
+ unsigned int req_state, unsigned int flags);
+ void (*lm_cancel) (struct gfs2_glock *gl);
+ const match_table_t *lm_tokens;
+};
+
+#define LM_FLAG_TRY 0x00000001
+#define LM_FLAG_TRY_1CB 0x00000002
+#define LM_FLAG_NOEXP 0x00000004
+#define LM_FLAG_ANY 0x00000008
+#define LM_FLAG_PRIORITY 0x00000010
#define GL_ASYNC 0x00000040
#define GL_EXACT 0x00000080
@@ -128,10 +243,12 @@ static inline int gfs2_glock_nq_init(struct gfs2_glock *gl,
int gfs2_lvb_hold(struct gfs2_glock *gl);
void gfs2_lvb_unhold(struct gfs2_glock *gl);
-void gfs2_glock_cb(void *cb_data, unsigned int type, void *data);
+void gfs2_glock_cb(struct gfs2_glock *gl, unsigned int state);
+void gfs2_glock_complete(struct gfs2_glock *gl, int ret);
void gfs2_reclaim_glock(struct gfs2_sbd *sdp);
void gfs2_gl_hash_clear(struct gfs2_sbd *sdp);
void gfs2_glock_finish_truncate(struct gfs2_inode *ip);
+void gfs2_glock_thaw(struct gfs2_sbd *sdp);
int __init gfs2_glock_init(void);
void gfs2_glock_exit(void);
@@ -141,4 +258,6 @@ void gfs2_delete_debugfs_file(struct gfs2_sbd *sdp);
int gfs2_register_debugfs(void);
void gfs2_unregister_debugfs(void);
+extern const struct lm_lockops gfs2_dlm_ops;
+
#endif /* __GLOCK_DOT_H__ */
diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c
index 8522d3a..f07ede8 100644
--- a/fs/gfs2/glops.c
+++ b/fs/gfs2/glops.c
@@ -12,7 +12,6 @@
#include <linux/completion.h>
#include <linux/buffer_head.h>
#include <linux/gfs2_ondisk.h>
-#include <linux/lm_interface.h>
#include <linux/bio.h>
#include "gfs2.h"
@@ -390,18 +389,6 @@ static int trans_go_demote_ok(const struct gfs2_glock *gl)
return 0;
}
-/**
- * quota_go_demote_ok - Check to see if it's ok to unlock a quota glock
- * @gl: the glock
- *
- * Returns: 1 if it's ok
- */
-
-static int quota_go_demote_ok(const struct gfs2_glock *gl)
-{
- return !atomic_read(&gl->gl_lvb_count);
-}
-
const struct gfs2_glock_operations gfs2_meta_glops = {
.go_xmote_th = meta_go_sync,
.go_type = LM_TYPE_META,
@@ -448,7 +435,6 @@ const struct gfs2_glock_operations gfs2_nondisk_glops = {
};
const struct gfs2_glock_operations gfs2_quota_glops = {
- .go_demote_ok = quota_go_demote_ok,
.go_type = LM_TYPE_QUOTA,
};
diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h
index a0117d6..0af7c24 100644
--- a/fs/gfs2/incore.h
+++ b/fs/gfs2/incore.h
@@ -12,6 +12,8 @@
#include <linux/fs.h>
#include <linux/workqueue.h>
+#include <linux/dlm.h>
+#include <linux/buffer_head.h>
#define DIO_WAIT 0x00000010
#define DIO_METADATA 0x00000020
@@ -26,6 +28,7 @@ struct gfs2_trans;
struct gfs2_ail;
struct gfs2_jdesc;
struct gfs2_sbd;
+struct lm_lockops;
typedef void (*gfs2_glop_bh_t) (struct gfs2_glock *gl, unsigned int ret);
@@ -121,6 +124,28 @@ struct gfs2_bufdata {
struct list_head bd_ail_gl_list;
};
+/*
+ * Internally, we prefix things with gdlm_ and GDLM_ (for gfs-dlm) since a
+ * prefix of lock_dlm_ gets awkward.
+ */
+
+#define GDLM_STRNAME_BYTES 25
+#define GDLM_LVB_SIZE 32
+
+enum {
+ DFL_BLOCK_LOCKS = 0,
+};
+
+struct lm_lockname {
+ u64 ln_number;
+ unsigned int ln_type;
+};
+
+#define lm_name_equal(name1, name2) \
+ (((name1)->ln_number == (name2)->ln_number) && \
+ ((name1)->ln_type == (name2)->ln_type))
+
+
struct gfs2_glock_operations {
void (*go_xmote_th) (struct gfs2_glock *gl);
int (*go_xmote_bh) (struct gfs2_glock *gl, struct gfs2_holder *gh);
@@ -162,6 +187,8 @@ enum {
GLF_LFLUSH = 7,
GLF_INVALIDATE_IN_PROGRESS = 8,
GLF_REPLY_PENDING = 9,
+ GLF_INITIAL = 10,
+ GLF_FROZEN = 11,
};
struct gfs2_glock {
@@ -181,10 +208,9 @@ struct gfs2_glock {
struct list_head gl_holders;
const struct gfs2_glock_operations *gl_ops;
- void *gl_lock;
- char *gl_lvb;
- atomic_t gl_lvb_count;
-
+ char gl_strname[GDLM_STRNAME_BYTES];
+ struct dlm_lksb gl_lksb;
+ char gl_lvb[32];
unsigned long gl_stamp;
unsigned long gl_tchange;
void *gl_object;
@@ -447,6 +473,30 @@ struct gfs2_sb_host {
char sb_locktable[GFS2_LOCKNAME_LEN];
};
+/*
+ * lm_mount() return values
+ *
+ * ls_jid - the journal ID this node should use
+ * ls_first - this node is the first to mount the file system
+ * ls_lockspace - lock module's context for this file system
+ * ls_ops - lock module's functions
+ */
+
+struct lm_lockstruct {
+ u32 ls_id;
+ unsigned int ls_jid;
+ unsigned int ls_first;
+ unsigned int ls_first_done;
+ unsigned int ls_nodir;
+ const struct lm_lockops *ls_ops;
+ unsigned long ls_flags;
+ dlm_lockspace_t *ls_dlm;
+
+ int ls_recover_jid;
+ int ls_recover_jid_done;
+ int ls_recover_jid_status;
+};
+
struct gfs2_sbd {
struct super_block *sd_vfs;
struct kobject sd_kobj;
@@ -520,7 +570,6 @@ struct gfs2_sbd {
spinlock_t sd_jindex_spin;
struct mutex sd_jindex_mutex;
unsigned int sd_journals;
- unsigned long sd_jindex_refresh_time;
struct gfs2_jdesc *sd_jdesc;
struct gfs2_holder sd_journal_gh;
diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c
index 3b87c18..7b277d4 100644
--- a/fs/gfs2/inode.c
+++ b/fs/gfs2/inode.c
@@ -16,7 +16,6 @@
#include <linux/sort.h>
#include <linux/gfs2_ondisk.h>
#include <linux/crc32.h>
-#include <linux/lm_interface.h>
#include <linux/security.h>
#include <linux/time.h>
@@ -137,16 +136,16 @@ void gfs2_set_iop(struct inode *inode)
if (S_ISREG(mode)) {
inode->i_op = &gfs2_file_iops;
- if (sdp->sd_args.ar_localflocks)
- inode->i_fop = &gfs2_file_fops_nolock;
+ if (gfs2_localflocks(sdp))
+ inode->i_fop = gfs2_file_fops_nolock;
else
- inode->i_fop = &gfs2_file_fops;
+ inode->i_fop = gfs2_file_fops;
} else if (S_ISDIR(mode)) {
inode->i_op = &gfs2_dir_iops;
- if (sdp->sd_args.ar_localflocks)
- inode->i_fop = &gfs2_dir_fops_nolock;
+ if (gfs2_localflocks(sdp))
+ inode->i_fop = gfs2_dir_fops_nolock;
else
- inode->i_fop = &gfs2_dir_fops;
+ inode->i_fop = gfs2_dir_fops;
} else if (S_ISLNK(mode)) {
inode->i_op = &gfs2_symlink_iops;
} else {
diff --git a/fs/gfs2/inode.h b/fs/gfs2/inode.h
index d532936..dca4fee 100644
--- a/fs/gfs2/inode.h
+++ b/fs/gfs2/inode.h
@@ -101,12 +101,26 @@ void gfs2_dinode_print(const struct gfs2_inode *ip);
extern const struct inode_operations gfs2_file_iops;
extern const struct inode_operations gfs2_dir_iops;
extern const struct inode_operations gfs2_symlink_iops;
-extern const struct file_operations gfs2_file_fops;
-extern const struct file_operations gfs2_dir_fops;
-extern const struct file_operations gfs2_file_fops_nolock;
-extern const struct file_operations gfs2_dir_fops_nolock;
+extern const struct file_operations *gfs2_file_fops_nolock;
+extern const struct file_operations *gfs2_dir_fops_nolock;
extern void gfs2_set_inode_flags(struct inode *inode);
+
+#ifdef CONFIG_GFS2_FS_LOCKING_DLM
+extern const struct file_operations *gfs2_file_fops;
+extern const struct file_operations *gfs2_dir_fops;
+static inline int gfs2_localflocks(const struct gfs2_sbd *sdp)
+{
+ return sdp->sd_args.ar_localflocks;
+}
+#else /* Single node only */
+#define gfs2_file_fops NULL
+#define gfs2_dir_fops NULL
+static inline int gfs2_localflocks(const struct gfs2_sbd *sdp)
+{
+ return 1;
+}
+#endif /* CONFIG_GFS2_FS_LOCKING_DLM */
#endif /* __INODE_DOT_H__ */
diff --git a/fs/gfs2/lock_dlm.c b/fs/gfs2/lock_dlm.c
new file mode 100644
index 0000000..a0bb7d2
--- /dev/null
+++ b/fs/gfs2/lock_dlm.c
@@ -0,0 +1,240 @@
+/*
+ * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved.
+ * Copyright (C) 2004-2009 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/fs.h>
+#include <linux/dlm.h>
+#include <linux/types.h>
+#include <linux/gfs2_ondisk.h>
+
+#include "incore.h"
+#include "glock.h"
+#include "util.h"
+
+
+static void gdlm_ast(void *arg)
+{
+ struct gfs2_glock *gl = arg;
+ unsigned ret = gl->gl_state;
+
+ BUG_ON(gl->gl_lksb.sb_flags & DLM_SBF_DEMOTED);
+
+ if (gl->gl_lksb.sb_flags & DLM_SBF_VALNOTVALID)
+ memset(gl->gl_lvb, 0, GDLM_LVB_SIZE);
+
+ switch (gl->gl_lksb.sb_status) {
+ case -DLM_EUNLOCK: /* Unlocked, so glock can be freed */
+ kmem_cache_free(gfs2_glock_cachep, gl);
+ return;
+ case -DLM_ECANCEL: /* Cancel while getting lock */
+ ret |= LM_OUT_CANCELED;
+ goto out;
+ case -EAGAIN: /* Try lock fails */
+ goto out;
+ case -EINVAL: /* Invalid */
+ case -ENOMEM: /* Out of memory */
+ ret |= LM_OUT_ERROR;
+ goto out;
+ case 0: /* Success */
+ break;
+ default: /* Something unexpected */
+ BUG();
+ }
+
+ ret = gl->gl_target;
+ if (gl->gl_lksb.sb_flags & DLM_SBF_ALTMODE) {
+ if (gl->gl_target == LM_ST_SHARED)
+ ret = LM_ST_DEFERRED;
+ else if (gl->gl_target == LM_ST_DEFERRED)
+ ret = LM_ST_SHARED;
+ else
+ BUG();
+ }
+
+ set_bit(GLF_INITIAL, &gl->gl_flags);
+ gfs2_glock_complete(gl, ret);
+ return;
+out:
+ if (!test_bit(GLF_INITIAL, &gl->gl_flags))
+ gl->gl_lksb.sb_lkid = 0;
+ gfs2_glock_complete(gl, ret);
+}
+
+static void gdlm_bast(void *arg, int mode)
+{
+ struct gfs2_glock *gl = arg;
+
+ switch (mode) {
+ case DLM_LOCK_EX:
+ gfs2_glock_cb(gl, LM_ST_UNLOCKED);
+ break;
+ case DLM_LOCK_CW:
+ gfs2_glock_cb(gl, LM_ST_DEFERRED);
+ break;
+ case DLM_LOCK_PR:
+ gfs2_glock_cb(gl, LM_ST_SHARED);
+ break;
+ default:
+ printk(KERN_ERR "unknown bast mode %d", mode);
+ BUG();
+ }
+}
+
+/* convert gfs lock-state to dlm lock-mode */
+
+static int make_mode(const unsigned int lmstate)
+{
+ switch (lmstate) {
+ case LM_ST_UNLOCKED:
+ return DLM_LOCK_NL;
+ case LM_ST_EXCLUSIVE:
+ return DLM_LOCK_EX;
+ case LM_ST_DEFERRED:
+ return DLM_LOCK_CW;
+ case LM_ST_SHARED:
+ return DLM_LOCK_PR;
+ }
+ printk(KERN_ERR "unknown LM state %d", lmstate);
+ BUG();
+ return -1;
+}
+
+static u32 make_flags(const u32 lkid, const unsigned int gfs_flags,
+ const int req)
+{
+ u32 lkf = 0;
+
+ if (gfs_flags & LM_FLAG_TRY)
+ lkf |= DLM_LKF_NOQUEUE;
+
+ if (gfs_flags & LM_FLAG_TRY_1CB) {
+ lkf |= DLM_LKF_NOQUEUE;
+ lkf |= DLM_LKF_NOQUEUEBAST;
+ }
+
+ if (gfs_flags & LM_FLAG_PRIORITY) {
+ lkf |= DLM_LKF_NOORDER;
+ lkf |= DLM_LKF_HEADQUE;
+ }
+
+ if (gfs_flags & LM_FLAG_ANY) {
+ if (req == DLM_LOCK_PR)
+ lkf |= DLM_LKF_ALTCW;
+ else if (req == DLM_LOCK_CW)
+ lkf |= DLM_LKF_ALTPR;
+ else
+ BUG();
+ }
+
+ if (lkid != 0)
+ lkf |= DLM_LKF_CONVERT;
+
+ lkf |= DLM_LKF_VALBLK;
+
+ return lkf;
+}
+
+static unsigned int gdlm_lock(struct gfs2_glock *gl,
+ unsigned int req_state, unsigned int flags)
+{
+ struct lm_lockstruct *ls = &gl->gl_sbd->sd_lockstruct;
+ int error;
+ int req;
+ u32 lkf;
+
+ req = make_mode(req_state);
+ lkf = make_flags(gl->gl_lksb.sb_lkid, flags, req);
+
+ /*
+ * Submit the actual lock request.
+ */
+
+ error = dlm_lock(ls->ls_dlm, req, &gl->gl_lksb, lkf, gl->gl_strname,
+ GDLM_STRNAME_BYTES - 1, 0, gdlm_ast, gl, gdlm_bast);
+ if (error == -EAGAIN)
+ return 0;
+ if (error)
+ return LM_OUT_ERROR;
+ return LM_OUT_ASYNC;
+}
+
+static void gdlm_put_lock(struct kmem_cache *cachep, void *ptr)
+{
+ struct gfs2_glock *gl = ptr;
+ struct lm_lockstruct *ls = &gl->gl_sbd->sd_lockstruct;
+ int error;
+
+ if (gl->gl_lksb.sb_lkid == 0) {
+ kmem_cache_free(cachep, gl);
+ return;
+ }
+
+ error = dlm_unlock(ls->ls_dlm, gl->gl_lksb.sb_lkid, DLM_LKF_VALBLK,
+ NULL, gl);
+ if (error) {
+ printk(KERN_ERR "gdlm_unlock %x,%llx err=%d\n",
+ gl->gl_name.ln_type,
+ (unsigned long long)gl->gl_name.ln_number, error);
+ return;
+ }
+}
+
+static void gdlm_cancel(struct gfs2_glock *gl)
+{
+ struct lm_lockstruct *ls = &gl->gl_sbd->sd_lockstruct;
+ dlm_unlock(ls->ls_dlm, gl->gl_lksb.sb_lkid, DLM_LKF_CANCEL, NULL, gl);
+}
+
+static int gdlm_mount(struct gfs2_sbd *sdp, const char *fsname)
+{
+ struct lm_lockstruct *ls = &sdp->sd_lockstruct;
+ int error;
+
+ if (fsname == NULL) {
+ fs_info(sdp, "no fsname found\n");
+ return -EINVAL;
+ }
+
+ error = dlm_new_lockspace(fsname, strlen(fsname), &ls->ls_dlm,
+ DLM_LSFL_FS | DLM_LSFL_NEWEXCL |
+ (ls->ls_nodir ? DLM_LSFL_NODIR : 0),
+ GDLM_LVB_SIZE);
+ if (error)
+ printk(KERN_ERR "dlm_new_lockspace error %d", error);
+
+ return error;
+}
+
+static void gdlm_unmount(struct gfs2_sbd *sdp)
+{
+ struct lm_lockstruct *ls = &sdp->sd_lockstruct;
+
+ if (ls->ls_dlm) {
+ dlm_release_lockspace(ls->ls_dlm, 2);
+ ls->ls_dlm = NULL;
+ }
+}
+
+static const match_table_t dlm_tokens = {
+ { Opt_jid, "jid=%d"},
+ { Opt_id, "id=%d"},
+ { Opt_first, "first=%d"},
+ { Opt_nodir, "nodir=%d"},
+ { Opt_err, NULL },
+};
+
+const struct lm_lockops gfs2_dlm_ops = {
+ .lm_proto_name = "lock_dlm",
+ .lm_mount = gdlm_mount,
+ .lm_unmount = gdlm_unmount,
+ .lm_put_lock = gdlm_put_lock,
+ .lm_lock = gdlm_lock,
+ .lm_cancel = gdlm_cancel,
+ .lm_tokens = &dlm_tokens,
+};
+
diff --git a/fs/gfs2/locking.c b/fs/gfs2/locking.c
deleted file mode 100644
index d3657bc..0000000
--- a/fs/gfs2/locking.c
+++ /dev/null
@@ -1,314 +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/module.h>
-#include <linux/init.h>
-#include <linux/string.h>
-#include <linux/slab.h>
-#include <linux/wait.h>
-#include <linux/sched.h>
-#include <linux/kmod.h>
-#include <linux/fs.h>
-#include <linux/delay.h>
-#include <linux/lm_interface.h>
-
-struct lmh_wrapper {
- struct list_head lw_list;
- const struct lm_lockops *lw_ops;
-};
-
-struct nolock_lockspace {
- unsigned int nl_lvb_size;
-};
-
-/**
- * nolock_get_lock - get a lm_lock_t given a descripton of the lock
- * @lockspace: the lockspace the lock lives in
- * @name: the name of the lock
- * @lockp: return the lm_lock_t here
- *
- * Returns: 0 on success, -EXXX on failure
- */
-
-static int nolock_get_lock(void *lockspace, struct lm_lockname *name,
- void **lockp)
-{
- *lockp = lockspace;
- return 0;
-}
-
-/**
- * nolock_put_lock - get rid of a lock structure
- * @lock: the lock to throw away
- *
- */
-
-static void nolock_put_lock(void *lock)
-{
-}
-
-/**
- * nolock_hold_lvb - hold on to a lock value block
- * @lock: the lock the LVB is associated with
- * @lvbp: return the lm_lvb_t here
- *
- * Returns: 0 on success, -EXXX on failure
- */
-
-static int nolock_hold_lvb(void *lock, char **lvbp)
-{
- struct nolock_lockspace *nl = lock;
- int error = 0;
-
- *lvbp = kzalloc(nl->nl_lvb_size, GFP_KERNEL);
- if (!*lvbp)
- error = -ENOMEM;
-
- return error;
-}
-
-/**
- * nolock_unhold_lvb - release a LVB
- * @lock: the lock the LVB is associated with
- * @lvb: the lock value block
- *
- */
-
-static void nolock_unhold_lvb(void *lock, char *lvb)
-{
- kfree(lvb);
-}
-
-static int nolock_mount(char *table_name, char *host_data,
- lm_callback_t cb, void *cb_data,
- unsigned int min_lvb_size, int flags,
- struct lm_lockstruct *lockstruct,
- struct kobject *fskobj);
-static void nolock_unmount(void *lockspace);
-
-/* List of registered low-level locking protocols. A file system selects one
- of them by name at mount time, e.g. lock_nolock, lock_dlm. */
-
-static const struct lm_lockops nolock_ops = {
- .lm_proto_name = "lock_nolock",
- .lm_mount = nolock_mount,
- .lm_unmount = nolock_unmount,
- .lm_get_lock = nolock_get_lock,
- .lm_put_lock = nolock_put_lock,
- .lm_hold_lvb = nolock_hold_lvb,
- .lm_unhold_lvb = nolock_unhold_lvb,
-};
-
-static struct lmh_wrapper nolock_proto = {
- .lw_list = LIST_HEAD_INIT(nolock_proto.lw_list),
- .lw_ops = &nolock_ops,
-};
-
-static LIST_HEAD(lmh_list);
-static DEFINE_MUTEX(lmh_lock);
-
-static int nolock_mount(char *table_name, char *host_data,
- lm_callback_t cb, void *cb_data,
- unsigned int min_lvb_size, int flags,
- struct lm_lockstruct *lockstruct,
- struct kobject *fskobj)
-{
- char *c;
- unsigned int jid;
- struct nolock_lockspace *nl;
-
- c = strstr(host_data, "jid=");
- if (!c)
- jid = 0;
- else {
- c += 4;
- sscanf(c, "%u", &jid);
- }
-
- nl = kzalloc(sizeof(struct nolock_lockspace), GFP_KERNEL);
- if (!nl)
- return -ENOMEM;
-
- nl->nl_lvb_size = min_lvb_size;
-
- lockstruct->ls_jid = jid;
- lockstruct->ls_first = 1;
- lockstruct->ls_lvb_size = min_lvb_size;
- lockstruct->ls_lockspace = nl;
- lockstruct->ls_ops = &nolock_ops;
- lockstruct->ls_flags = LM_LSFLAG_LOCAL;
-
- return 0;
-}
-
-static void nolock_unmount(void *lockspace)
-{
- struct nolock_lockspace *nl = lockspace;
- kfree(nl);
-}
-
-/**
- * gfs2_register_lockproto - Register a low-level locking protocol
- * @proto: the protocol definition
- *
- * Returns: 0 on success, -EXXX on failure
- */
-
-int gfs2_register_lockproto(const struct lm_lockops *proto)
-{
- struct lmh_wrapper *lw;
-
- mutex_lock(&lmh_lock);
-
- list_for_each_entry(lw, &lmh_list, lw_list) {
- if (!strcmp(lw->lw_ops->lm_proto_name, proto->lm_proto_name)) {
- mutex_unlock(&lmh_lock);
- printk(KERN_INFO "GFS2: protocol %s already exists\n",
- proto->lm_proto_name);
- return -EEXIST;
- }
- }
-
- lw = kzalloc(sizeof(struct lmh_wrapper), GFP_KERNEL);
- if (!lw) {
- mutex_unlock(&lmh_lock);
- return -ENOMEM;
- }
-
- lw->lw_ops = proto;
- list_add(&lw->lw_list, &lmh_list);
-
- mutex_unlock(&lmh_lock);
-
- return 0;
-}
-
-/**
- * gfs2_unregister_lockproto - Unregister a low-level locking protocol
- * @proto: the protocol definition
- *
- */
-
-void gfs2_unregister_lockproto(const struct lm_lockops *proto)
-{
- struct lmh_wrapper *lw;
-
- mutex_lock(&lmh_lock);
-
- list_for_each_entry(lw, &lmh_list, lw_list) {
- if (!strcmp(lw->lw_ops->lm_proto_name, proto->lm_proto_name)) {
- list_del(&lw->lw_list);
- mutex_unlock(&lmh_lock);
- kfree(lw);
- return;
- }
- }
-
- mutex_unlock(&lmh_lock);
-
- printk(KERN_WARNING "GFS2: can't unregister lock protocol %s\n",
- proto->lm_proto_name);
-}
-
-/**
- * gfs2_mount_lockproto - Mount a lock protocol
- * @proto_name - the name of the protocol
- * @table_name - the name of the lock space
- * @host_data - data specific to this host
- * @cb - the callback to the code using the lock module
- * @sdp - The GFS2 superblock
- * @min_lvb_size - the mininum LVB size that the caller can deal with
- * @flags - LM_MFLAG_*
- * @lockstruct - a structure returned describing the mount
- *
- * Returns: 0 on success, -EXXX on failure
- */
-
-int gfs2_mount_lockproto(char *proto_name, char *table_name, char *host_data,
- lm_callback_t cb, void *cb_data,
- unsigned int min_lvb_size, int flags,
- struct lm_lockstruct *lockstruct,
- struct kobject *fskobj)
-{
- struct lmh_wrapper *lw = NULL;
- int try = 0;
- int error, found;
-
-
-retry:
- mutex_lock(&lmh_lock);
-
- if (list_empty(&nolock_proto.lw_list))
- list_add(&nolock_proto.lw_list, &lmh_list);
-
- found = 0;
- list_for_each_entry(lw, &lmh_list, lw_list) {
- if (!strcmp(lw->lw_ops->lm_proto_name, proto_name)) {
- found = 1;
- break;
- }
- }
-
- if (!found) {
- if (!try && capable(CAP_SYS_MODULE)) {
- try = 1;
- mutex_unlock(&lmh_lock);
- request_module(proto_name);
- goto retry;
- }
- printk(KERN_INFO "GFS2: can't find protocol %s\n", proto_name);
- error = -ENOENT;
- goto out;
- }
-
- if (lw->lw_ops->lm_owner &&
- !try_module_get(lw->lw_ops->lm_owner)) {
- try = 0;
- mutex_unlock(&lmh_lock);
- msleep(1000);
- goto retry;
- }
-
- error = lw->lw_ops->lm_mount(table_name, host_data, cb, cb_data,
- min_lvb_size, flags, lockstruct, fskobj);
- if (error)
- module_put(lw->lw_ops->lm_owner);
-out:
- mutex_unlock(&lmh_lock);
- return error;
-}
-
-void gfs2_unmount_lockproto(struct lm_lockstruct *lockstruct)
-{
- mutex_lock(&lmh_lock);
- if (lockstruct->ls_ops->lm_unmount)
- lockstruct->ls_ops->lm_unmount(lockstruct->ls_lockspace);
- if (lockstruct->ls_ops->lm_owner)
- module_put(lockstruct->ls_ops->lm_owner);
- mutex_unlock(&lmh_lock);
-}
-
-/**
- * gfs2_withdraw_lockproto - abnormally unmount a lock module
- * @lockstruct: the lockstruct passed into mount
- *
- */
-
-void gfs2_withdraw_lockproto(struct lm_lockstruct *lockstruct)
-{
- mutex_lock(&lmh_lock);
- lockstruct->ls_ops->lm_withdraw(lockstruct->ls_lockspace);
- if (lockstruct->ls_ops->lm_owner)
- module_put(lockstruct->ls_ops->lm_owner);
- mutex_unlock(&lmh_lock);
-}
-
-EXPORT_SYMBOL_GPL(gfs2_register_lockproto);
-EXPORT_SYMBOL_GPL(gfs2_unregister_lockproto);
-
diff --git a/fs/gfs2/locking/dlm/Makefile b/fs/gfs2/locking/dlm/Makefile
deleted file mode 100644
index 2609bb6..0000000
--- a/fs/gfs2/locking/dlm/Makefile
+++ /dev/null
@@ -1,3 +0,0 @@
-obj-$(CONFIG_GFS2_FS_LOCKING_DLM) += lock_dlm.o
-lock_dlm-y := lock.o main.o mount.o sysfs.o thread.o
-
diff --git a/fs/gfs2/locking/dlm/lock.c b/fs/gfs2/locking/dlm/lock.c
deleted file mode 100644
index 2482c90..0000000
--- a/fs/gfs2/locking/dlm/lock.c
+++ /dev/null
@@ -1,708 +0,0 @@
-/*
- * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved.
- * Copyright (C) 2004-2005 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 "lock_dlm.h"
-
-static char junk_lvb[GDLM_LVB_SIZE];
-
-
-/* convert dlm lock-mode to gfs lock-state */
-
-static s16 gdlm_make_lmstate(s16 dlmmode)
-{
- switch (dlmmode) {
- case DLM_LOCK_IV:
- case DLM_LOCK_NL:
- return LM_ST_UNLOCKED;
- case DLM_LOCK_EX:
- return LM_ST_EXCLUSIVE;
- case DLM_LOCK_CW:
- return LM_ST_DEFERRED;
- case DLM_LOCK_PR:
- return LM_ST_SHARED;
- }
- gdlm_assert(0, "unknown DLM mode %d", dlmmode);
- return -1;
-}
-
-/* A lock placed on this queue is re-submitted to DLM as soon as the lock_dlm
- thread gets to it. */
-
-static void queue_submit(struct gdlm_lock *lp)
-{
- struct gdlm_ls *ls = lp->ls;
-
- spin_lock(&ls->async_lock);
- list_add_tail(&lp->delay_list, &ls->submit);
- spin_unlock(&ls->async_lock);
- wake_up(&ls->thread_wait);
-}
-
-static void wake_up_ast(struct gdlm_lock *lp)
-{
- clear_bit(LFL_AST_WAIT, &lp->flags);
- smp_mb__after_clear_bit();
- wake_up_bit(&lp->flags, LFL_AST_WAIT);
-}
-
-static void gdlm_delete_lp(struct gdlm_lock *lp)
-{
- struct gdlm_ls *ls = lp->ls;
-
- spin_lock(&ls->async_lock);
- if (!list_empty(&lp->delay_list))
- list_del_init(&lp->delay_list);
- ls->all_locks_count--;
- spin_unlock(&ls->async_lock);
-
- kfree(lp);
-}
-
-static void gdlm_queue_delayed(struct gdlm_lock *lp)
-{
- struct gdlm_ls *ls = lp->ls;
-
- spin_lock(&ls->async_lock);
- list_add_tail(&lp->delay_list, &ls->delayed);
- spin_unlock(&ls->async_lock);
-}
-
-static void process_complete(struct gdlm_lock *lp)
-{
- struct gdlm_ls *ls = lp->ls;
- struct lm_async_cb acb;
-
- memset(&acb, 0, sizeof(acb));
-
- if (lp->lksb.sb_status == -DLM_ECANCEL) {
- log_info("complete dlm cancel %x,%llx flags %lx",
- lp->lockname.ln_type,
- (unsigned long long)lp->lockname.ln_number,
- lp->flags);
-
- lp->req = lp->cur;
- acb.lc_ret |= LM_OUT_CANCELED;
- if (lp->cur == DLM_LOCK_IV)
- lp->lksb.sb_lkid = 0;
- goto out;
- }
-
- if (test_and_clear_bit(LFL_DLM_UNLOCK, &lp->flags)) {
- if (lp->lksb.sb_status != -DLM_EUNLOCK) {
- log_info("unlock sb_status %d %x,%llx flags %lx",
- lp->lksb.sb_status, lp->lockname.ln_type,
- (unsigned long long)lp->lockname.ln_number,
- lp->flags);
- return;
- }
-
- lp->cur = DLM_LOCK_IV;
- lp->req = DLM_LOCK_IV;
- lp->lksb.sb_lkid = 0;
-
- if (test_and_clear_bit(LFL_UNLOCK_DELETE, &lp->flags)) {
- gdlm_delete_lp(lp);
- return;
- }
- goto out;
- }
-
- if (lp->lksb.sb_flags & DLM_SBF_VALNOTVALID)
- memset(lp->lksb.sb_lvbptr, 0, GDLM_LVB_SIZE);
-
- if (lp->lksb.sb_flags & DLM_SBF_ALTMODE) {
- if (lp->req == DLM_LOCK_PR)
- lp->req = DLM_LOCK_CW;
- else if (lp->req == DLM_LOCK_CW)
- lp->req = DLM_LOCK_PR;
- }
-
- /*
- * A canceled lock request. The lock was just taken off the delayed
- * list and was never even submitted to dlm.
- */
-
- if (test_and_clear_bit(LFL_CANCEL, &lp->flags)) {
- log_info("complete internal cancel %x,%llx",
- lp->lockname.ln_type,
- (unsigned long long)lp->lockname.ln_number);
- lp->req = lp->cur;
- acb.lc_ret |= LM_OUT_CANCELED;
- goto out;
- }
-
- /*
- * An error occured.
- */
-
- if (lp->lksb.sb_status) {
- /* a "normal" error */
- if ((lp->lksb.sb_status == -EAGAIN) &&
- (lp->lkf & DLM_LKF_NOQUEUE)) {
- lp->req = lp->cur;
- if (lp->cur == DLM_LOCK_IV)
- lp->lksb.sb_lkid = 0;
- goto out;
- }
-
- /* this could only happen with cancels I think */
- log_info("ast sb_status %d %x,%llx flags %lx",
- lp->lksb.sb_status, lp->lockname.ln_type,
- (unsigned long long)lp->lockname.ln_number,
- lp->flags);
- return;
- }
-
- /*
- * This is an AST for an EX->EX conversion for sync_lvb from GFS.
- */
-
- if (test_and_clear_bit(LFL_SYNC_LVB, &lp->flags)) {
- wake_up_ast(lp);
- return;
- }
-
- /*
- * A lock has been demoted to NL because it initially completed during
- * BLOCK_LOCKS. Now it must be requested in the originally requested
- * mode.
- */
-
- if (test_and_clear_bit(LFL_REREQUEST, &lp->flags)) {
- gdlm_assert(lp->req == DLM_LOCK_NL, "%x,%llx",
- lp->lockname.ln_type,
- (unsigned long long)lp->lockname.ln_number);
- gdlm_assert(lp->prev_req > DLM_LOCK_NL, "%x,%llx",
- lp->lockname.ln_type,
- (unsigned long long)lp->lockname.ln_number);
-
- lp->cur = DLM_LOCK_NL;
- lp->req = lp->prev_req;
- lp->prev_req = DLM_LOCK_IV;
- lp->lkf &= ~DLM_LKF_CONVDEADLK;
-
- set_bit(LFL_NOCACHE, &lp->flags);
-
- if (test_bit(DFL_BLOCK_LOCKS, &ls->flags) &&
- !test_bit(LFL_NOBLOCK, &lp->flags))
- gdlm_queue_delayed(lp);
- else
- queue_submit(lp);
- return;
- }
-
- /*
- * A request is granted during dlm recovery. It may be granted
- * because the locks of a failed node were cleared. In that case,
- * there may be inconsistent data beneath this lock and we must wait
- * for recovery to complete to use it. When gfs recovery is done this
- * granted lock will be converted to NL and then reacquired in this
- * granted state.
- */
-
- if (test_bit(DFL_BLOCK_LOCKS, &ls->flags) &&
- !test_bit(LFL_NOBLOCK, &lp->flags) &&
- lp->req != DLM_LOCK_NL) {
-
- lp->cur = lp->req;
- lp->prev_req = lp->req;
- lp->req = DLM_LOCK_NL;
- lp->lkf |= DLM_LKF_CONVERT;
- lp->lkf &= ~DLM_LKF_CONVDEADLK;
-
- log_debug("rereq %x,%llx id %x %d,%d",
- lp->lockname.ln_type,
- (unsigned long long)lp->lockname.ln_number,
- lp->lksb.sb_lkid, lp->cur, lp->req);
-
- set_bit(LFL_REREQUEST, &lp->flags);
- queue_submit(lp);
- return;
- }
-
- /*
- * DLM demoted the lock to NL before it was granted so GFS must be
- * told it cannot cache data for this lock.
- */
-
- if (lp->lksb.sb_flags & DLM_SBF_DEMOTED)
- set_bit(LFL_NOCACHE, &lp->flags);
-
-out:
- /*
- * This is an internal lock_dlm lock
- */
-
- if (test_bit(LFL_INLOCK, &lp->flags)) {
- clear_bit(LFL_NOBLOCK, &lp->flags);
- lp->cur = lp->req;
- wake_up_ast(lp);
- return;
- }
-
- /*
- * Normal completion of a lock request. Tell GFS it now has the lock.
- */
-
- clear_bit(LFL_NOBLOCK, &lp->flags);
- lp->cur = lp->req;
-
- acb.lc_name = lp->lockname;
- acb.lc_ret |= gdlm_make_lmstate(lp->cur);
-
- ls->fscb(ls->sdp, LM_CB_ASYNC, &acb);
-}
-
-static void gdlm_ast(void *astarg)
-{
- struct gdlm_lock *lp = astarg;
- clear_bit(LFL_ACTIVE, &lp->flags);
- process_complete(lp);
-}
-
-static void process_blocking(struct gdlm_lock *lp, int bast_mode)
-{
- struct gdlm_ls *ls = lp->ls;
- unsigned int cb = 0;
-
- switch (gdlm_make_lmstate(bast_mode)) {
- case LM_ST_EXCLUSIVE:
- cb = LM_CB_NEED_E;
- break;
- case LM_ST_DEFERRED:
- cb = LM_CB_NEED_D;
- break;
- case LM_ST_SHARED:
- cb = LM_CB_NEED_S;
- break;
- default:
- gdlm_assert(0, "unknown bast mode %u", bast_mode);
- }
-
- ls->fscb(ls->sdp, cb, &lp->lockname);
-}
-
-
-static void gdlm_bast(void *astarg, int mode)
-{
- struct gdlm_lock *lp = astarg;
-
- if (!mode) {
- printk(KERN_INFO "lock_dlm: bast mode zero %x,%llx\n",
- lp->lockname.ln_type,
- (unsigned long long)lp->lockname.ln_number);
- return;
- }
-
- process_blocking(lp, mode);
-}
-
-/* convert gfs lock-state to dlm lock-mode */
-
-static s16 make_mode(s16 lmstate)
-{
- switch (lmstate) {
- case LM_ST_UNLOCKED:
- return DLM_LOCK_NL;
- case LM_ST_EXCLUSIVE:
- return DLM_LOCK_EX;
- case LM_ST_DEFERRED:
- return DLM_LOCK_CW;
- case LM_ST_SHARED:
- return DLM_LOCK_PR;
- }
- gdlm_assert(0, "unknown LM state %d", lmstate);
- return -1;
-}
-
-
-/* verify agreement with GFS on the current lock state, NB: DLM_LOCK_NL and
- DLM_LOCK_IV are both considered LM_ST_UNLOCKED by GFS. */
-
-static void check_cur_state(struct gdlm_lock *lp, unsigned int cur_state)
-{
- s16 cur = make_mode(cur_state);
- if (lp->cur != DLM_LOCK_IV)
- gdlm_assert(lp->cur == cur, "%d, %d", lp->cur, cur);
-}
-
-static inline unsigned int make_flags(struct gdlm_lock *lp,
- unsigned int gfs_flags,
- s16 cur, s16 req)
-{
- unsigned int lkf = 0;
-
- if (gfs_flags & LM_FLAG_TRY)
- lkf |= DLM_LKF_NOQUEUE;
-
- if (gfs_flags & LM_FLAG_TRY_1CB) {
- lkf |= DLM_LKF_NOQUEUE;
- lkf |= DLM_LKF_NOQUEUEBAST;
- }
-
- if (gfs_flags & LM_FLAG_PRIORITY) {
- lkf |= DLM_LKF_NOORDER;
- lkf |= DLM_LKF_HEADQUE;
- }
-
- if (gfs_flags & LM_FLAG_ANY) {
- if (req == DLM_LOCK_PR)
- lkf |= DLM_LKF_ALTCW;
- else if (req == DLM_LOCK_CW)
- lkf |= DLM_LKF_ALTPR;
- }
-
- if (lp->lksb.sb_lkid != 0) {
- lkf |= DLM_LKF_CONVERT;
- }
-
- if (lp->lvb)
- lkf |= DLM_LKF_VALBLK;
-
- return lkf;
-}
-
-/* make_strname - convert GFS lock numbers to a string */
-
-static inline void make_strname(const struct lm_lockname *lockname,
- struct gdlm_strname *str)
-{
- sprintf(str->name, "%8x%16llx", lockname->ln_type,
- (unsigned long long)lockname->ln_number);
- str->namelen = GDLM_STRNAME_BYTES;
-}
-
-static int gdlm_create_lp(struct gdlm_ls *ls, struct lm_lockname *name,
- struct gdlm_lock **lpp)
-{
- struct gdlm_lock *lp;
-
- lp = kzalloc(sizeof(struct gdlm_lock), GFP_NOFS);
- if (!lp)
- return -ENOMEM;
-
- lp->lockname = *name;
- make_strname(name, &lp->strname);
- lp->ls = ls;
- lp->cur = DLM_LOCK_IV;
- INIT_LIST_HEAD(&lp->delay_list);
-
- spin_lock(&ls->async_lock);
- ls->all_locks_count++;
- spin_unlock(&ls->async_lock);
-
- *lpp = lp;
- return 0;
-}
-
-int gdlm_get_lock(void *lockspace, struct lm_lockname *name,
- void **lockp)
-{
- struct gdlm_lock *lp;
- int error;
-
- error = gdlm_create_lp(lockspace, name, &lp);
-
- *lockp = lp;
- return error;
-}
-
-void gdlm_put_lock(void *lock)
-{
- gdlm_delete_lp(lock);
-}
-
-unsigned int gdlm_do_lock(struct gdlm_lock *lp)
-{
- struct gdlm_ls *ls = lp->ls;
- int error, bast = 1;
-
- /*
- * When recovery is in progress, delay lock requests for submission
- * once recovery is done. Requests for recovery (NOEXP) and unlocks
- * can pass.
- */
-
- if (test_bit(DFL_BLOCK_LOCKS, &ls->flags) &&
- !test_bit(LFL_NOBLOCK, &lp->flags) && lp->req != DLM_LOCK_NL) {
- gdlm_queue_delayed(lp);
- return LM_OUT_ASYNC;
- }
-
- /*
- * Submit the actual lock request.
- */
-
- if (test_bit(LFL_NOBAST, &lp->flags))
- bast = 0;
-
- set_bit(LFL_ACTIVE, &lp->flags);
-
- log_debug("lk %x,%llx id %x %d,%d %x", lp->lockname.ln_type,
- (unsigned long long)lp->lockname.ln_number, lp->lksb.sb_lkid,
- lp->cur, lp->req, lp->lkf);
-
- error = dlm_lock(ls->dlm_lockspace, lp->req, &lp->lksb, lp->lkf,
- lp->strname.name, lp->strname.namelen, 0, gdlm_ast,
- lp, bast ? gdlm_bast : NULL);
-
- if ((error == -EAGAIN) && (lp->lkf & DLM_LKF_NOQUEUE)) {
- lp->lksb.sb_status = -EAGAIN;
- gdlm_ast(lp);
- error = 0;
- }
-
- if (error) {
- log_error("%s: gdlm_lock %x,%llx err=%d cur=%d req=%d lkf=%x "
- "flags=%lx", ls->fsname, lp->lockname.ln_type,
- (unsigned long long)lp->lockname.ln_number, error,
- lp->cur, lp->req, lp->lkf, lp->flags);
- return LM_OUT_ERROR;
- }
- return LM_OUT_ASYNC;
-}
-
-static unsigned int gdlm_do_unlock(struct gdlm_lock *lp)
-{
- struct gdlm_ls *ls = lp->ls;
- unsigned int lkf = 0;
- int error;
-
- set_bit(LFL_DLM_UNLOCK, &lp->flags);
- set_bit(LFL_ACTIVE, &lp->flags);
-
- if (lp->lvb)
- lkf = DLM_LKF_VALBLK;
-
- log_debug("un %x,%llx %x %d %x", lp->lockname.ln_type,
- (unsigned long long)lp->lockname.ln_number,
- lp->lksb.sb_lkid, lp->cur, lkf);
-
- error = dlm_unlock(ls->dlm_lockspace, lp->lksb.sb_lkid, lkf, NULL, lp);
-
- if (error) {
- log_error("%s: gdlm_unlock %x,%llx err=%d cur=%d req=%d lkf=%x "
- "flags=%lx", ls->fsname, lp->lockname.ln_type,
- (unsigned long long)lp->lockname.ln_number, error,
- lp->cur, lp->req, lp->lkf, lp->flags);
- return LM_OUT_ERROR;
- }
- return LM_OUT_ASYNC;
-}
-
-unsigned int gdlm_lock(void *lock, unsigned int cur_state,
- unsigned int req_state, unsigned int flags)
-{
- struct gdlm_lock *lp = lock;
-
- if (req_state == LM_ST_UNLOCKED)
- return gdlm_unlock(lock, cur_state);
-
- if (req_state == LM_ST_UNLOCKED)
- return gdlm_unlock(lock, cur_state);
-
- clear_bit(LFL_DLM_CANCEL, &lp->flags);
- if (flags & LM_FLAG_NOEXP)
- set_bit(LFL_NOBLOCK, &lp->flags);
-
- check_cur_state(lp, cur_state);
- lp->req = make_mode(req_state);
- lp->lkf = make_flags(lp, flags, lp->cur, lp->req);
-
- return gdlm_do_lock(lp);
-}
-
-unsigned int gdlm_unlock(void *lock, unsigned int cur_state)
-{
- struct gdlm_lock *lp = lock;
-
- clear_bit(LFL_DLM_CANCEL, &lp->flags);
- if (lp->cur == DLM_LOCK_IV)
- return 0;
- return gdlm_do_unlock(lp);
-}
-
-void gdlm_cancel(void *lock)
-{
- struct gdlm_lock *lp = lock;
- struct gdlm_ls *ls = lp->ls;
- int error, delay_list = 0;
-
- if (test_bit(LFL_DLM_CANCEL, &lp->flags))
- return;
-
- log_info("gdlm_cancel %x,%llx flags %lx", lp->lockname.ln_type,
- (unsigned long long)lp->lockname.ln_number, lp->flags);
-
- spin_lock(&ls->async_lock);
- if (!list_empty(&lp->delay_list)) {
- list_del_init(&lp->delay_list);
- delay_list = 1;
- }
- spin_unlock(&ls->async_lock);
-
- if (delay_list) {
- set_bit(LFL_CANCEL, &lp->flags);
- set_bit(LFL_ACTIVE, &lp->flags);
- gdlm_ast(lp);
- return;
- }
-
- if (!test_bit(LFL_ACTIVE, &lp->flags) ||
- test_bit(LFL_DLM_UNLOCK, &lp->flags)) {
- log_info("gdlm_cancel skip %x,%llx flags %lx",
- lp->lockname.ln_type,
- (unsigned long long)lp->lockname.ln_number, lp->flags);
- return;
- }
-
- /* the lock is blocked in the dlm */
-
- set_bit(LFL_DLM_CANCEL, &lp->flags);
- set_bit(LFL_ACTIVE, &lp->flags);
-
- error = dlm_unlock(ls->dlm_lockspace, lp->lksb.sb_lkid, DLM_LKF_CANCEL,
- NULL, lp);
-
- log_info("gdlm_cancel rv %d %x,%llx flags %lx", error,
- lp->lockname.ln_type,
- (unsigned long long)lp->lockname.ln_number, lp->flags);
-
- if (error == -EBUSY)
- clear_bit(LFL_DLM_CANCEL, &lp->flags);
-}
-
-static int gdlm_add_lvb(struct gdlm_lock *lp)
-{
- char *lvb;
-
- lvb = kzalloc(GDLM_LVB_SIZE, GFP_NOFS);
- if (!lvb)
- return -ENOMEM;
-
- lp->lksb.sb_lvbptr = lvb;
- lp->lvb = lvb;
- return 0;
-}
-
-static void gdlm_del_lvb(struct gdlm_lock *lp)
-{
- kfree(lp->lvb);
- lp->lvb = NULL;
- lp->lksb.sb_lvbptr = NULL;
-}
-
-static int gdlm_ast_wait(void *word)
-{
- schedule();
- return 0;
-}
-
-/* This can do a synchronous dlm request (requiring a lock_dlm thread to get
- the completion) because gfs won't call hold_lvb() during a callback (from
- the context of a lock_dlm thread). */
-
-static int hold_null_lock(struct gdlm_lock *lp)
-{
- struct gdlm_lock *lpn = NULL;
- int error;
-
- if (lp->hold_null) {
- printk(KERN_INFO "lock_dlm: lvb already held\n");
- return 0;
- }
-
- error = gdlm_create_lp(lp->ls, &lp->lockname, &lpn);
- if (error)
- goto out;
-
- lpn->lksb.sb_lvbptr = junk_lvb;
- lpn->lvb = junk_lvb;
-
- lpn->req = DLM_LOCK_NL;
- lpn->lkf = DLM_LKF_VALBLK | DLM_LKF_EXPEDITE;
- set_bit(LFL_NOBAST, &lpn->flags);
- set_bit(LFL_INLOCK, &lpn->flags);
- set_bit(LFL_AST_WAIT, &lpn->flags);
-
- gdlm_do_lock(lpn);
- wait_on_bit(&lpn->flags, LFL_AST_WAIT, gdlm_ast_wait, TASK_UNINTERRUPTIBLE);
- error = lpn->lksb.sb_status;
- if (error) {
- printk(KERN_INFO "lock_dlm: hold_null_lock dlm error %d\n",
- error);
- gdlm_delete_lp(lpn);
- lpn = NULL;
- }
-out:
- lp->hold_null = lpn;
- return error;
-}
-
-/* This cannot do a synchronous dlm request (requiring a lock_dlm thread to get
- the completion) because gfs may call unhold_lvb() during a callback (from
- the context of a lock_dlm thread) which could cause a deadlock since the
- other lock_dlm thread could be engaged in recovery. */
-
-static void unhold_null_lock(struct gdlm_lock *lp)
-{
- struct gdlm_lock *lpn = lp->hold_null;
-
- gdlm_assert(lpn, "%x,%llx", lp->lockname.ln_type,
- (unsigned long long)lp->lockname.ln_number);
- lpn->lksb.sb_lvbptr = NULL;
- lpn->lvb = NULL;
- set_bit(LFL_UNLOCK_DELETE, &lpn->flags);
- gdlm_do_unlock(lpn);
- lp->hold_null = NULL;
-}
-
-/* Acquire a NL lock because gfs requires the value block to remain
- intact on the resource while the lvb is "held" even if it's holding no locks
- on the resource. */
-
-int gdlm_hold_lvb(void *lock, char **lvbp)
-{
- struct gdlm_lock *lp = lock;
- int error;
-
- error = gdlm_add_lvb(lp);
- if (error)
- return error;
-
- *lvbp = lp->lvb;
-
- error = hold_null_lock(lp);
- if (error)
- gdlm_del_lvb(lp);
-
- return error;
-}
-
-void gdlm_unhold_lvb(void *lock, char *lvb)
-{
- struct gdlm_lock *lp = lock;
-
- unhold_null_lock(lp);
- gdlm_del_lvb(lp);
-}
-
-void gdlm_submit_delayed(struct gdlm_ls *ls)
-{
- struct gdlm_lock *lp, *safe;
-
- spin_lock(&ls->async_lock);
- list_for_each_entry_safe(lp, safe, &ls->delayed, delay_list) {
- list_del_init(&lp->delay_list);
- list_add_tail(&lp->delay_list, &ls->submit);
- }
- spin_unlock(&ls->async_lock);
- wake_up(&ls->thread_wait);
-}
-
diff --git a/fs/gfs2/locking/dlm/lock_dlm.h b/fs/gfs2/locking/dlm/lock_dlm.h
deleted file mode 100644
index 3c98e7c..0000000
--- a/fs/gfs2/locking/dlm/lock_dlm.h
+++ /dev/null
@@ -1,166 +0,0 @@
-/*
- * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved.
- * Copyright (C) 2004-2005 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 LOCK_DLM_DOT_H
-#define LOCK_DLM_DOT_H
-
-#include <linux/module.h>
-#include <linux/slab.h>
-#include <linux/spinlock.h>
-#include <linux/types.h>
-#include <linux/string.h>
-#include <linux/list.h>
-#include <linux/socket.h>
-#include <linux/delay.h>
-#include <linux/kthread.h>
-#include <linux/kobject.h>
-#include <linux/fcntl.h>
-#include <linux/wait.h>
-#include <net/sock.h>
-
-#include <linux/dlm.h>
-#include <linux/dlm_plock.h>
-#include <linux/lm_interface.h>
-
-/*
- * Internally, we prefix things with gdlm_ and GDLM_ (for gfs-dlm) since a
- * prefix of lock_dlm_ gets awkward. Externally, GFS refers to this module
- * as "lock_dlm".
- */
-
-#define GDLM_STRNAME_BYTES 24
-#define GDLM_LVB_SIZE 32
-#define GDLM_DROP_COUNT 0
-#define GDLM_DROP_PERIOD 60
-#define GDLM_NAME_LEN 128
-
-/* GFS uses 12 bytes to identify a resource (32 bit type + 64 bit number).
- We sprintf these numbers into a 24 byte string of hex values to make them
- human-readable (to make debugging simpler.) */
-
-struct gdlm_strname {
- unsigned char name[GDLM_STRNAME_BYTES];
- unsigned short namelen;
-};
-
-enum {
- DFL_BLOCK_LOCKS = 0,
- DFL_SPECTATOR = 1,
- DFL_WITHDRAW = 2,
-};
-
-struct gdlm_ls {
- u32 id;
- int jid;
- int first;
- int first_done;
- unsigned long flags;
- struct kobject kobj;
- char clustername[GDLM_NAME_LEN];
- char fsname[GDLM_NAME_LEN];
- int fsflags;
- dlm_lockspace_t *dlm_lockspace;
- lm_callback_t fscb;
- struct gfs2_sbd *sdp;
- int recover_jid;
- int recover_jid_done;
- int recover_jid_status;
- spinlock_t async_lock;
- struct list_head delayed;
- struct list_head submit;
- u32 all_locks_count;
- wait_queue_head_t wait_control;
- struct task_struct *thread;
- wait_queue_head_t thread_wait;
-};
-
-enum {
- LFL_NOBLOCK = 0,
- LFL_NOCACHE = 1,
- LFL_DLM_UNLOCK = 2,
- LFL_DLM_CANCEL = 3,
- LFL_SYNC_LVB = 4,
- LFL_FORCE_PROMOTE = 5,
- LFL_REREQUEST = 6,
- LFL_ACTIVE = 7,
- LFL_INLOCK = 8,
- LFL_CANCEL = 9,
- LFL_NOBAST = 10,
- LFL_HEADQUE = 11,
- LFL_UNLOCK_DELETE = 12,
- LFL_AST_WAIT = 13,
-};
-
-struct gdlm_lock {
- struct gdlm_ls *ls;
- struct lm_lockname lockname;
- struct gdlm_strname strname;
- char *lvb;
- struct dlm_lksb lksb;
-
- s16 cur;
- s16 req;
- s16 prev_req;
- u32 lkf; /* dlm flags DLM_LKF_ */
- unsigned long flags; /* lock_dlm flags LFL_ */
-
- struct list_head delay_list; /* delayed */
- struct gdlm_lock *hold_null; /* NL lock for hold_lvb */
-};
-
-#define gdlm_assert(assertion, fmt, args...) \
-do { \
- if (unlikely(!(assertion))) { \
- printk(KERN_EMERG "lock_dlm: fatal assertion failed \"%s\"\n" \
- "lock_dlm: " fmt "\n", \
- #assertion, ##args); \
- BUG(); \
- } \
-} while (0)
-
-#define log_print(lev, fmt, arg...) printk(lev "lock_dlm: " fmt "\n" , ## arg)
-#define log_info(fmt, arg...) log_print(KERN_INFO , fmt , ## arg)
-#define log_error(fmt, arg...) log_print(KERN_ERR , fmt , ## arg)
-#ifdef LOCK_DLM_LOG_DEBUG
-#define log_debug(fmt, arg...) log_print(KERN_DEBUG , fmt , ## arg)
-#else
-#define log_debug(fmt, arg...)
-#endif
-
-/* sysfs.c */
-
-int gdlm_sysfs_init(void);
-void gdlm_sysfs_exit(void);
-int gdlm_kobject_setup(struct gdlm_ls *, struct kobject *);
-void gdlm_kobject_release(struct gdlm_ls *);
-
-/* thread.c */
-
-int gdlm_init_threads(struct gdlm_ls *);
-void gdlm_release_threads(struct gdlm_ls *);
-
-/* lock.c */
-
-void gdlm_submit_delayed(struct gdlm_ls *);
-unsigned int gdlm_do_lock(struct gdlm_lock *);
-
-int gdlm_get_lock(void *, struct lm_lockname *, void **);
-void gdlm_put_lock(void *);
-unsigned int gdlm_lock(void *, unsigned int, unsigned int, unsigned int);
-unsigned int gdlm_unlock(void *, unsigned int);
-void gdlm_cancel(void *);
-int gdlm_hold_lvb(void *, char **);
-void gdlm_unhold_lvb(void *, char *);
-
-/* mount.c */
-
-extern const struct lm_lockops gdlm_ops;
-
-#endif
-
diff --git a/fs/gfs2/locking/dlm/main.c b/fs/gfs2/locking/dlm/main.c
deleted file mode 100644
index b9a03a7..0000000
--- a/fs/gfs2/locking/dlm/main.c
+++ /dev/null
@@ -1,48 +0,0 @@
-/*
- * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved.
- * Copyright (C) 2004-2005 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/init.h>
-
-#include "lock_dlm.h"
-
-static int __init init_lock_dlm(void)
-{
- int error;
-
- error = gfs2_register_lockproto(&gdlm_ops);
- if (error) {
- printk(KERN_WARNING "lock_dlm: can't register protocol: %d\n",
- error);
- return error;
- }
-
- error = gdlm_sysfs_init();
- if (error) {
- gfs2_unregister_lockproto(&gdlm_ops);
- return error;
- }
-
- printk(KERN_INFO
- "Lock_DLM (built %s %s) installed\n", __DATE__, __TIME__);
- return 0;
-}
-
-static void __exit exit_lock_dlm(void)
-{
- gdlm_sysfs_exit();
- gfs2_unregister_lockproto(&gdlm_ops);
-}
-
-module_init(init_lock_dlm);
-module_exit(exit_lock_dlm);
-
-MODULE_DESCRIPTION("GFS DLM Locking Module");
-MODULE_AUTHOR("Red Hat, Inc.");
-MODULE_LICENSE("GPL");
-
diff --git a/fs/gfs2/locking/dlm/mount.c b/fs/gfs2/locking/dlm/mount.c
deleted file mode 100644
index 1aa7eb6..0000000
--- a/fs/gfs2/locking/dlm/mount.c
+++ /dev/null
@@ -1,276 +0,0 @@
-/*
- * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved.
- * Copyright (C) 2004-2005 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 "lock_dlm.h"
-
-const struct lm_lockops gdlm_ops;
-
-
-static struct gdlm_ls *init_gdlm(lm_callback_t cb, struct gfs2_sbd *sdp,
- int flags, char *table_name)
-{
- struct gdlm_ls *ls;
- char buf[256], *p;
-
- ls = kzalloc(sizeof(struct gdlm_ls), GFP_KERNEL);
- if (!ls)
- return NULL;
-
- ls->fscb = cb;
- ls->sdp = sdp;
- ls->fsflags = flags;
- spin_lock_init(&ls->async_lock);
- INIT_LIST_HEAD(&ls->delayed);
- INIT_LIST_HEAD(&ls->submit);
- init_waitqueue_head(&ls->thread_wait);
- init_waitqueue_head(&ls->wait_control);
- ls->jid = -1;
-
- strncpy(buf, table_name, 256);
- buf[255] = '\0';
-
- p = strchr(buf, ':');
- if (!p) {
- log_info("invalid table_name \"%s\"", table_name);
- kfree(ls);
- return NULL;
- }
- *p = '\0';
- p++;
-
- strncpy(ls->clustername, buf, GDLM_NAME_LEN);
- strncpy(ls->fsname, p, GDLM_NAME_LEN);
-
- return ls;
-}
-
-static int make_args(struct gdlm_ls *ls, char *data_arg, int *nodir)
-{
- char data[256];
- char *options, *x, *y;
- int error = 0;
-
- memset(data, 0, 256);
- strncpy(data, data_arg, 255);
-
- if (!strlen(data)) {
- log_error("no mount options, (u)mount helpers not installed");
- return -EINVAL;
- }
-
- for (options = data; (x = strsep(&options, ":")); ) {
- if (!*x)
- continue;
-
- y = strchr(x, '=');
- if (y)
- *y++ = 0;
-
- if (!strcmp(x, "jid")) {
- if (!y) {
- log_error("need argument to jid");
- error = -EINVAL;
- break;
- }
- sscanf(y, "%u", &ls->jid);
-
- } else if (!strcmp(x, "first")) {
- if (!y) {
- log_error("need argument to first");
- error = -EINVAL;
- break;
- }
- sscanf(y, "%u", &ls->first);
-
- } else if (!strcmp(x, "id")) {
- if (!y) {
- log_error("need argument to id");
- error = -EINVAL;
- break;
- }
- sscanf(y, "%u", &ls->id);
-
- } else if (!strcmp(x, "nodir")) {
- if (!y) {
- log_error("need argument to nodir");
- error = -EINVAL;
- break;
- }
- sscanf(y, "%u", nodir);
-
- } else {
- log_error("unkonwn option: %s", x);
- error = -EINVAL;
- break;
- }
- }
-
- return error;
-}
-
-static int gdlm_mount(char *table_name, char *host_data,
- lm_callback_t cb, void *cb_data,
- unsigned int min_lvb_size, int flags,
- struct lm_lockstruct *lockstruct,
- struct kobject *fskobj)
-{
- struct gdlm_ls *ls;
- int error = -ENOMEM, nodir = 0;
-
- if (min_lvb_size > GDLM_LVB_SIZE)
- goto out;
-
- ls = init_gdlm(cb, cb_data, flags, table_name);
- if (!ls)
- goto out;
-
- error = make_args(ls, host_data, &nodir);
- if (error)
- goto out;
-
- error = gdlm_init_threads(ls);
- if (error)
- goto out_free;
-
- error = gdlm_kobject_setup(ls, fskobj);
- if (error)
- goto out_thread;
-
- error = dlm_new_lockspace(ls->fsname, strlen(ls->fsname),
- &ls->dlm_lockspace,
- DLM_LSFL_FS | DLM_LSFL_NEWEXCL |
- (nodir ? DLM_LSFL_NODIR : 0),
- GDLM_LVB_SIZE);
- if (error) {
- log_error("dlm_new_lockspace error %d", error);
- goto out_kobj;
- }
-
- lockstruct->ls_jid = ls->jid;
- lockstruct->ls_first = ls->first;
- lockstruct->ls_lockspace = ls;
- lockstruct->ls_ops = &gdlm_ops;
- lockstruct->ls_flags = 0;
- lockstruct->ls_lvb_size = GDLM_LVB_SIZE;
- return 0;
-
-out_kobj:
- gdlm_kobject_release(ls);
-out_thread:
- gdlm_release_threads(ls);
-out_free:
- kfree(ls);
-out:
- return error;
-}
-
-static void gdlm_unmount(void *lockspace)
-{
- struct gdlm_ls *ls = lockspace;
-
- log_debug("unmount flags %lx", ls->flags);
-
- /* FIXME: serialize unmount and withdraw in case they
- happen at once. Also, if unmount follows withdraw,
- wait for withdraw to finish. */
-
- if (test_bit(DFL_WITHDRAW, &ls->flags))
- goto out;
-
- gdlm_kobject_release(ls);
- dlm_release_lockspace(ls->dlm_lockspace, 2);
- gdlm_release_threads(ls);
- BUG_ON(ls->all_locks_count);
-out:
- kfree(ls);
-}
-
-static void gdlm_recovery_done(void *lockspace, unsigned int jid,
- unsigned int message)
-{
- char env_jid[20];
- char env_status[20];
- char *envp[] = { env_jid, env_status, NULL };
- struct gdlm_ls *ls = lockspace;
- ls->recover_jid_done = jid;
- ls->recover_jid_status = message;
- sprintf(env_jid, "JID=%d", jid);
- sprintf(env_status, "RECOVERY=%s",
- message == LM_RD_SUCCESS ? "Done" : "Failed");
- kobject_uevent_env(&ls->kobj, KOBJ_CHANGE, envp);
-}
-
-static void gdlm_others_may_mount(void *lockspace)
-{
- char *message = "FIRSTMOUNT=Done";
- char *envp[] = { message, NULL };
- struct gdlm_ls *ls = lockspace;
- ls->first_done = 1;
- kobject_uevent_env(&ls->kobj, KOBJ_CHANGE, envp);
-}
-
-/* Userspace gets the offline uevent, blocks new gfs locks on
- other mounters, and lets us know (sets WITHDRAW flag). Then,
- userspace leaves the mount group while we leave the lockspace. */
-
-static void gdlm_withdraw(void *lockspace)
-{
- struct gdlm_ls *ls = lockspace;
-
- kobject_uevent(&ls->kobj, KOBJ_OFFLINE);
-
- wait_event_interruptible(ls->wait_control,
- test_bit(DFL_WITHDRAW, &ls->flags));
-
- dlm_release_lockspace(ls->dlm_lockspace, 2);
- gdlm_release_threads(ls);
- gdlm_kobject_release(ls);
-}
-
-static int gdlm_plock(void *lockspace, struct lm_lockname *name,
- struct file *file, int cmd, struct file_lock *fl)
-{
- struct gdlm_ls *ls = lockspace;
- return dlm_posix_lock(ls->dlm_lockspace, name->ln_number, file, cmd, fl);
-}
-
-static int gdlm_punlock(void *lockspace, struct lm_lockname *name,
- struct file *file, struct file_lock *fl)
-{
- struct gdlm_ls *ls = lockspace;
- return dlm_posix_unlock(ls->dlm_lockspace, name->ln_number, file, fl);
-}
-
-static int gdlm_plock_get(void *lockspace, struct lm_lockname *name,
- struct file *file, struct file_lock *fl)
-{
- struct gdlm_ls *ls = lockspace;
- return dlm_posix_get(ls->dlm_lockspace, name->ln_number, file, fl);
-}
-
-const struct lm_lockops gdlm_ops = {
- .lm_proto_name = "lock_dlm",
- .lm_mount = gdlm_mount,
- .lm_others_may_mount = gdlm_others_may_mount,
- .lm_unmount = gdlm_unmount,
- .lm_withdraw = gdlm_withdraw,
- .lm_get_lock = gdlm_get_lock,
- .lm_put_lock = gdlm_put_lock,
- .lm_lock = gdlm_lock,
- .lm_unlock = gdlm_unlock,
- .lm_plock = gdlm_plock,
- .lm_punlock = gdlm_punlock,
- .lm_plock_get = gdlm_plock_get,
- .lm_cancel = gdlm_cancel,
- .lm_hold_lvb = gdlm_hold_lvb,
- .lm_unhold_lvb = gdlm_unhold_lvb,
- .lm_recovery_done = gdlm_recovery_done,
- .lm_owner = THIS_MODULE,
-};
-
diff --git a/fs/gfs2/locking/dlm/sysfs.c b/fs/gfs2/locking/dlm/sysfs.c
deleted file mode 100644
index 9b7edcf..0000000
--- a/fs/gfs2/locking/dlm/sysfs.c
+++ /dev/null
@@ -1,226 +0,0 @@
-/*
- * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved.
- * Copyright (C) 2004-2005 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/ctype.h>
-#include <linux/stat.h>
-
-#include "lock_dlm.h"
-
-static ssize_t proto_name_show(struct gdlm_ls *ls, char *buf)
-{
- return sprintf(buf, "%s\n", gdlm_ops.lm_proto_name);
-}
-
-static ssize_t block_show(struct gdlm_ls *ls, char *buf)
-{
- ssize_t ret;
- int val = 0;
-
- if (test_bit(DFL_BLOCK_LOCKS, &ls->flags))
- val = 1;
- ret = sprintf(buf, "%d\n", val);
- return ret;
-}
-
-static ssize_t block_store(struct gdlm_ls *ls, const char *buf, size_t len)
-{
- ssize_t ret = len;
- int val;
-
- val = simple_strtol(buf, NULL, 0);
-
- if (val == 1)
- set_bit(DFL_BLOCK_LOCKS, &ls->flags);
- else if (val == 0) {
- clear_bit(DFL_BLOCK_LOCKS, &ls->flags);
- gdlm_submit_delayed(ls);
- } else {
- ret = -EINVAL;
- }
- return ret;
-}
-
-static ssize_t withdraw_show(struct gdlm_ls *ls, char *buf)
-{
- ssize_t ret;
- int val = 0;
-
- if (test_bit(DFL_WITHDRAW, &ls->flags))
- val = 1;
- ret = sprintf(buf, "%d\n", val);
- return ret;
-}
-
-static ssize_t withdraw_store(struct gdlm_ls *ls, const char *buf, size_t len)
-{
- ssize_t ret = len;
- int val;
-
- val = simple_strtol(buf, NULL, 0);
-
- if (val == 1)
- set_bit(DFL_WITHDRAW, &ls->flags);
- else
- ret = -EINVAL;
- wake_up(&ls->wait_control);
- return ret;
-}
-
-static ssize_t id_show(struct gdlm_ls *ls, char *buf)
-{
- return sprintf(buf, "%u\n", ls->id);
-}
-
-static ssize_t jid_show(struct gdlm_ls *ls, char *buf)
-{
- return sprintf(buf, "%d\n", ls->jid);
-}
-
-static ssize_t first_show(struct gdlm_ls *ls, char *buf)
-{
- return sprintf(buf, "%d\n", ls->first);
-}
-
-static ssize_t first_done_show(struct gdlm_ls *ls, char *buf)
-{
- return sprintf(buf, "%d\n", ls->first_done);
-}
-
-static ssize_t recover_show(struct gdlm_ls *ls, char *buf)
-{
- return sprintf(buf, "%d\n", ls->recover_jid);
-}
-
-static ssize_t recover_store(struct gdlm_ls *ls, const char *buf, size_t len)
-{
- ls->recover_jid = simple_strtol(buf, NULL, 0);
- ls->fscb(ls->sdp, LM_CB_NEED_RECOVERY, &ls->recover_jid);
- return len;
-}
-
-static ssize_t recover_done_show(struct gdlm_ls *ls, char *buf)
-{
- return sprintf(buf, "%d\n", ls->recover_jid_done);
-}
-
-static ssize_t recover_status_show(struct gdlm_ls *ls, char *buf)
-{
- return sprintf(buf, "%d\n", ls->recover_jid_status);
-}
-
-struct gdlm_attr {
- struct attribute attr;
- ssize_t (*show)(struct gdlm_ls *, char *);
- ssize_t (*store)(struct gdlm_ls *, const char *, size_t);
-};
-
-#define GDLM_ATTR(_name,_mode,_show,_store) \
-static struct gdlm_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, id_show, NULL);
-GDLM_ATTR(jid, 0444, jid_show, NULL);
-GDLM_ATTR(first, 0444, first_show, NULL);
-GDLM_ATTR(first_done, 0444, first_done_show, NULL);
-GDLM_ATTR(recover, 0644, recover_show, recover_store);
-GDLM_ATTR(recover_done, 0444, recover_done_show, NULL);
-GDLM_ATTR(recover_status, 0444, recover_status_show, NULL);
-
-static struct attribute *gdlm_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,
- &gdlm_attr_recover.attr,
- &gdlm_attr_recover_done.attr,
- &gdlm_attr_recover_status.attr,
- NULL,
-};
-
-static ssize_t gdlm_attr_show(struct kobject *kobj, struct attribute *attr,
- char *buf)
-{
- struct gdlm_ls *ls = container_of(kobj, struct gdlm_ls, kobj);
- struct gdlm_attr *a = container_of(attr, struct gdlm_attr, attr);
- return a->show ? a->show(ls, buf) : 0;
-}
-
-static ssize_t gdlm_attr_store(struct kobject *kobj, struct attribute *attr,
- const char *buf, size_t len)
-{
- struct gdlm_ls *ls = container_of(kobj, struct gdlm_ls, kobj);
- struct gdlm_attr *a = container_of(attr, struct gdlm_attr, attr);
- return a->store ? a->store(ls, buf, len) : len;
-}
-
-static struct sysfs_ops gdlm_attr_ops = {
- .show = gdlm_attr_show,
- .store = gdlm_attr_store,
-};
-
-static struct kobj_type gdlm_ktype = {
- .default_attrs = gdlm_attrs,
- .sysfs_ops = &gdlm_attr_ops,
-};
-
-static struct kset *gdlm_kset;
-
-int gdlm_kobject_setup(struct gdlm_ls *ls, struct kobject *fskobj)
-{
- int error;
-
- ls->kobj.kset = gdlm_kset;
- error = kobject_init_and_add(&ls->kobj, &gdlm_ktype, fskobj,
- "lock_module");
- if (error)
- log_error("can't register kobj %d", error);
- kobject_uevent(&ls->kobj, KOBJ_ADD);
-
- return error;
-}
-
-void gdlm_kobject_release(struct gdlm_ls *ls)
-{
- kobject_put(&ls->kobj);
-}
-
-static int gdlm_uevent(struct kset *kset, struct kobject *kobj,
- struct kobj_uevent_env *env)
-{
- struct gdlm_ls *ls = container_of(kobj, struct gdlm_ls, kobj);
- add_uevent_var(env, "LOCKTABLE=%s:%s", ls->clustername, ls->fsname);
- add_uevent_var(env, "LOCKPROTO=lock_dlm");
- return 0;
-}
-
-static struct kset_uevent_ops gdlm_uevent_ops = {
- .uevent = gdlm_uevent,
-};
-
-
-int gdlm_sysfs_init(void)
-{
- gdlm_kset = kset_create_and_add("lock_dlm", &gdlm_uevent_ops, kernel_kobj);
- if (!gdlm_kset) {
- printk(KERN_WARNING "%s: can not create kset\n", __func__);
- return -ENOMEM;
- }
- return 0;
-}
-
-void gdlm_sysfs_exit(void)
-{
- kset_unregister(gdlm_kset);
-}
-
diff --git a/fs/gfs2/locking/dlm/thread.c b/fs/gfs2/locking/dlm/thread.c
deleted file mode 100644
index 38823ef..0000000
--- a/fs/gfs2/locking/dlm/thread.c
+++ /dev/null
@@ -1,68 +0,0 @@
-/*
- * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved.
- * Copyright (C) 2004-2005 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 "lock_dlm.h"
-
-static inline int no_work(struct gdlm_ls *ls)
-{
- int ret;
-
- spin_lock(&ls->async_lock);
- ret = list_empty(&ls->submit);
- spin_unlock(&ls->async_lock);
-
- return ret;
-}
-
-static int gdlm_thread(void *data)
-{
- struct gdlm_ls *ls = (struct gdlm_ls *) data;
- struct gdlm_lock *lp = NULL;
-
- while (!kthread_should_stop()) {
- wait_event_interruptible(ls->thread_wait,
- !no_work(ls) || kthread_should_stop());
-
- spin_lock(&ls->async_lock);
-
- if (!list_empty(&ls->submit)) {
- lp = list_entry(ls->submit.next, struct gdlm_lock,
- delay_list);
- list_del_init(&lp->delay_list);
- spin_unlock(&ls->async_lock);
- gdlm_do_lock(lp);
- spin_lock(&ls->async_lock);
- }
- spin_unlock(&ls->async_lock);
- }
-
- return 0;
-}
-
-int gdlm_init_threads(struct gdlm_ls *ls)
-{
- struct task_struct *p;
- int error;
-
- p = kthread_run(gdlm_thread, ls, "lock_dlm");
- error = IS_ERR(p);
- if (error) {
- log_error("can't start lock_dlm thread %d", error);
- return error;
- }
- ls->thread = p;
-
- return 0;
-}
-
-void gdlm_release_threads(struct gdlm_ls *ls)
-{
- kthread_stop(ls->thread);
-}
-
diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c
index ad30585..98918a7 100644
--- a/fs/gfs2/log.c
+++ b/fs/gfs2/log.c
@@ -14,7 +14,6 @@
#include <linux/buffer_head.h>
#include <linux/gfs2_ondisk.h>
#include <linux/crc32.h>
-#include <linux/lm_interface.h>
#include <linux/delay.h>
#include <linux/kthread.h>
#include <linux/freezer.h>
diff --git a/fs/gfs2/lops.c b/fs/gfs2/lops.c
index 4390f6f..80e4f5f 100644
--- a/fs/gfs2/lops.c
+++ b/fs/gfs2/lops.c
@@ -13,7 +13,6 @@
#include <linux/completion.h>
#include <linux/buffer_head.h>
#include <linux/gfs2_ondisk.h>
-#include <linux/lm_interface.h>
#include "gfs2.h"
#include "incore.h"
diff --git a/fs/gfs2/main.c b/fs/gfs2/main.c
index 86fe067..a6892ed 100644
--- a/fs/gfs2/main.c
+++ b/fs/gfs2/main.c
@@ -14,7 +14,6 @@
#include <linux/module.h>
#include <linux/init.h>
#include <linux/gfs2_ondisk.h>
-#include <linux/lm_interface.h>
#include <asm/atomic.h>
#include "gfs2.h"
@@ -47,8 +46,6 @@ static void gfs2_init_glock_once(void *foo)
INIT_HLIST_NODE(&gl->gl_list);
spin_lock_init(&gl->gl_spin);
INIT_LIST_HEAD(&gl->gl_holders);
- gl->gl_lvb = NULL;
- atomic_set(&gl->gl_lvb_count, 0);
INIT_LIST_HEAD(&gl->gl_lru);
INIT_LIST_HEAD(&gl->gl_ail_list);
atomic_set(&gl->gl_ail_count, 0);
diff --git a/fs/gfs2/meta_io.c b/fs/gfs2/meta_io.c
index 0985362..870d65a 100644
--- a/fs/gfs2/meta_io.c
+++ b/fs/gfs2/meta_io.c
@@ -19,7 +19,6 @@
#include <linux/delay.h>
#include <linux/bio.h>
#include <linux/gfs2_ondisk.h>
-#include <linux/lm_interface.h>
#include "gfs2.h"
#include "incore.h"
diff --git a/fs/gfs2/mount.c b/fs/gfs2/mount.c
index 3524ae8..fba502a 100644
--- a/fs/gfs2/mount.c
+++ b/fs/gfs2/mount.c
@@ -12,7 +12,6 @@
#include <linux/completion.h>
#include <linux/buffer_head.h>
#include <linux/gfs2_ondisk.h>
-#include <linux/lm_interface.h>
#include <linux/parser.h>
#include "gfs2.h"
diff --git a/fs/gfs2/ops_address.c b/fs/gfs2/ops_address.c
index dde4ead..a6d00e8 100644
--- a/fs/gfs2/ops_address.c
+++ b/fs/gfs2/ops_address.c
@@ -19,7 +19,6 @@
#include <linux/writeback.h>
#include <linux/swap.h>
#include <linux/gfs2_ondisk.h>
-#include <linux/lm_interface.h>
#include <linux/backing-dev.h>
#include "gfs2.h"
diff --git a/fs/gfs2/ops_dentry.c b/fs/gfs2/ops_dentry.c
index c2ad363..5eb57b0 100644
--- a/fs/gfs2/ops_dentry.c
+++ b/fs/gfs2/ops_dentry.c
@@ -13,7 +13,6 @@
#include <linux/buffer_head.h>
#include <linux/gfs2_ondisk.h>
#include <linux/crc32.h>
-#include <linux/lm_interface.h>
#include "gfs2.h"
#include "incore.h"
diff --git a/fs/gfs2/ops_export.c b/fs/gfs2/ops_export.c
index 7fdeb14..9200ef2 100644
--- a/fs/gfs2/ops_export.c
+++ b/fs/gfs2/ops_export.c
@@ -14,7 +14,6 @@
#include <linux/exportfs.h>
#include <linux/gfs2_ondisk.h>
#include <linux/crc32.h>
-#include <linux/lm_interface.h>
#include "gfs2.h"
#include "incore.h"
diff --git a/fs/gfs2/ops_file.c b/fs/gfs2/ops_file.c
index 93fe41b..99d726f 100644
--- a/fs/gfs2/ops_file.c
+++ b/fs/gfs2/ops_file.c
@@ -20,9 +20,10 @@
#include <linux/gfs2_ondisk.h>
#include <linux/ext2_fs.h>
#include <linux/crc32.h>
-#include <linux/lm_interface.h>
#include <linux/writeback.h>
#include <asm/uaccess.h>
+#include <linux/dlm.h>
+#include <linux/dlm_plock.h>
#include "gfs2.h"
#include "incore.h"
@@ -560,57 +561,24 @@ static int gfs2_fsync(struct file *file, struct dentry *dentry, int datasync)
return ret;
}
+#ifdef CONFIG_GFS2_FS_LOCKING_DLM
+
/**
* gfs2_setlease - acquire/release a file lease
* @file: the file pointer
* @arg: lease type
* @fl: file lock
*
+ * We don't currently have a way to enforce a lease across the whole
+ * cluster; until we do, disable leases (by just returning -EINVAL),
+ * unless the administrator has requested purely local locking.
+ *
* Returns: errno
*/
static int gfs2_setlease(struct file *file, long arg, struct file_lock **fl)
{
- struct gfs2_sbd *sdp = GFS2_SB(file->f_mapping->host);
-
- /*
- * We don't currently have a way to enforce a lease across the whole
- * cluster; until we do, disable leases (by just returning -EINVAL),
- * unless the administrator has requested purely local locking.
- */
- if (!sdp->sd_args.ar_localflocks)
- return -EINVAL;
- return generic_setlease(file, arg, fl);
-}
-
-static int gfs2_lm_plock_get(struct gfs2_sbd *sdp, struct lm_lockname *name,
- struct file *file, struct file_lock *fl)
-{
- int error = -EIO;
- if (likely(!test_bit(SDF_SHUTDOWN, &sdp->sd_flags)))
- error = sdp->sd_lockstruct.ls_ops->lm_plock_get(
- sdp->sd_lockstruct.ls_lockspace, name, file, fl);
- return error;
-}
-
-static int gfs2_lm_plock(struct gfs2_sbd *sdp, struct lm_lockname *name,
- struct file *file, int cmd, struct file_lock *fl)
-{
- int error = -EIO;
- if (likely(!test_bit(SDF_SHUTDOWN, &sdp->sd_flags)))
- error = sdp->sd_lockstruct.ls_ops->lm_plock(
- sdp->sd_lockstruct.ls_lockspace, name, file, cmd, fl);
- return error;
-}
-
-static int gfs2_lm_punlock(struct gfs2_sbd *sdp, struct lm_lockname *name,
- struct file *file, struct file_lock *fl)
-{
- int error = -EIO;
- if (likely(!test_bit(SDF_SHUTDOWN, &sdp->sd_flags)))
- error = sdp->sd_lockstruct.ls_ops->lm_punlock(
- sdp->sd_lockstruct.ls_lockspace, name, file, fl);
- return error;
+ return -EINVAL;
}
/**
@@ -626,9 +594,7 @@ static int gfs2_lock(struct file *file, int cmd, struct file_lock *fl)
{
struct gfs2_inode *ip = GFS2_I(file->f_mapping->host);
struct gfs2_sbd *sdp = GFS2_SB(file->f_mapping->host);
- struct lm_lockname name =
- { .ln_number = ip->i_no_addr,
- .ln_type = LM_TYPE_PLOCK };
+ struct lm_lockstruct *ls = &sdp->sd_lockstruct;
if (!(fl->fl_flags & FL_POSIX))
return -ENOLCK;
@@ -640,12 +606,14 @@ static int gfs2_lock(struct file *file, int cmd, struct file_lock *fl)
cmd = F_SETLK;
fl->fl_type = F_UNLCK;
}
+ if (unlikely(test_bit(SDF_SHUTDOWN, &sdp->sd_flags)))
+ return -EIO;
if (IS_GETLK(cmd))
- return gfs2_lm_plock_get(sdp, &name, file, fl);
+ return dlm_posix_get(ls->ls_dlm, ip->i_no_addr, file, fl);
else if (fl->fl_type == F_UNLCK)
- return gfs2_lm_punlock(sdp, &name, file, fl);
+ return dlm_posix_unlock(ls->ls_dlm, ip->i_no_addr, file, fl);
else
- return gfs2_lm_plock(sdp, &name, file, cmd, fl);
+ return dlm_posix_lock(ls->ls_dlm, ip->i_no_addr, file, cmd, fl);
}
static int do_flock(struct file *file, int cmd, struct file_lock *fl)
@@ -732,7 +700,7 @@ static int gfs2_flock(struct file *file, int cmd, struct file_lock *fl)
}
}
-const struct file_operations gfs2_file_fops = {
+const struct file_operations *gfs2_file_fops = &(const struct file_operations){
.llseek = gfs2_llseek,
.read = do_sync_read,
.aio_read = generic_file_aio_read,
@@ -750,7 +718,7 @@ const struct file_operations gfs2_file_fops = {
.setlease = gfs2_setlease,
};
-const struct file_operations gfs2_dir_fops = {
+const struct file_operations *gfs2_dir_fops = &(const struct file_operations){
.readdir = gfs2_readdir,
.unlocked_ioctl = gfs2_ioctl,
.open = gfs2_open,
@@ -760,7 +728,9 @@ const struct file_operations gfs2_dir_fops = {
.flock = gfs2_flock,
};
-const struct file_operations gfs2_file_fops_nolock = {
+#endif /* CONFIG_GFS2_FS_LOCKING_DLM */
+
+const struct file_operations *gfs2_file_fops_nolock = &(const struct file_operations){
.llseek = gfs2_llseek,
.read = do_sync_read,
.aio_read = generic_file_aio_read,
@@ -773,10 +743,10 @@ const struct file_operations gfs2_file_fops_nolock = {
.fsync = gfs2_fsync,
.splice_read = generic_file_splice_read,
.splice_write = generic_file_splice_write,
- .setlease = gfs2_setlease,
+ .setlease = generic_setlease,
};
-const struct file_operations gfs2_dir_fops_nolock = {
+const struct file_operations *gfs2_dir_fops_nolock = &(const struct file_operations){
.readdir = gfs2_readdir,
.unlocked_ioctl = gfs2_ioctl,
.open = gfs2_open,
diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index 402b6a2..95bb33e 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -17,7 +17,6 @@
#include <linux/namei.h>
#include <linux/mount.h>
#include <linux/gfs2_ondisk.h>
-#include <linux/lm_interface.h>
#include "gfs2.h"
#include "incore.h"
@@ -627,13 +626,13 @@ static int map_journal_extents(struct gfs2_sbd *sdp)
return rc;
}
-static void gfs2_lm_others_may_mount(struct gfs2_sbd *sdp)
+static void gfs2_others_may_mount(struct gfs2_sbd *sdp)
{
- if (!sdp->sd_lockstruct.ls_ops->lm_others_may_mount)
- return;
- if (likely(!test_bit(SDF_SHUTDOWN, &sdp->sd_flags)))
- sdp->sd_lockstruct.ls_ops->lm_others_may_mount(
- sdp->sd_lockstruct.ls_lockspace);
+ char *message = "FIRSTMOUNT=Done";
+ char *envp[] = { message, NULL };
+ struct lm_lockstruct *ls = &sdp->sd_lockstruct;
+ ls->ls_first_done = 1;
+ kobject_uevent_env(&sdp->sd_kobj, KOBJ_CHANGE, envp);
}
/**
@@ -793,7 +792,7 @@ static int init_journal(struct gfs2_sbd *sdp, int undo)
}
}
- gfs2_lm_others_may_mount(sdp);
+ gfs2_others_may_mount(sdp);
} else if (!sdp->sd_args.ar_spectator) {
error = gfs2_recover_journal(sdp->sd_jdesc);
if (error) {
@@ -1002,7 +1001,6 @@ static int init_threads(struct gfs2_sbd *sdp, int undo)
goto fail_quotad;
sdp->sd_log_flush_time = jiffies;
- sdp->sd_jindex_refresh_time = jiffies;
p = kthread_run(gfs2_logd, sdp, "gfs2_logd");
error = IS_ERR(p);
@@ -1030,6 +1028,17 @@ fail:
return error;
}
+static const match_table_t nolock_tokens = {
+ { Opt_jid, "jid=%d\n", },
+ { Opt_err, NULL },
+};
+
+static const struct lm_lockops nolock_ops = {
+ .lm_proto_name = "lock_nolock",
+ .lm_put_lock = kmem_cache_free,
+ .lm_tokens = &nolock_tokens,
+};
+
/**
* gfs2_lm_mount - mount a locking protocol
* @sdp: the filesystem
@@ -1041,31 +1050,73 @@ fail:
static int gfs2_lm_mount(struct gfs2_sbd *sdp, int silent)
{
- char *proto = sdp->sd_proto_name;
- char *table = sdp->sd_table_name;
- int flags = LM_MFLAG_CONV_NODROP;
- int error;
+ const struct lm_lockops *lm;
+ struct lm_lockstruct *ls = &sdp->sd_lockstruct;
+ struct gfs2_args *args = &sdp->sd_args;
+ const char *proto = sdp->sd_proto_name;
+ const char *table = sdp->sd_table_name;
+ const char *fsname;
+ char *o, *options;
+ int ret;
- if (sdp->sd_args.ar_spectator)
- flags |= LM_MFLAG_SPECTATOR;
+ if (!strcmp("lock_nolock", proto)) {
+ lm = &nolock_ops;
+ sdp->sd_args.ar_localflocks = 1;
+ sdp->sd_args.ar_localcaching = 1;
+#ifdef CONFIG_GFS2_FS_LOCKING_DLM
+ } else if (!strcmp("lock_dlm", proto)) {
+ lm = &gfs2_dlm_ops;
+#endif
+ } else {
+ printk(KERN_INFO "GFS2: can't find protocol %s\n", proto);
+ return -ENOENT;
+ }
fs_info(sdp, "Trying to join cluster \"%s\", \"%s\"\n", proto, table);
- error = gfs2_mount_lockproto(proto, table, sdp->sd_args.ar_hostdata,
- gfs2_glock_cb, sdp,
- GFS2_MIN_LVB_SIZE, flags,
- &sdp->sd_lockstruct, &sdp->sd_kobj);
- if (error) {
- fs_info(sdp, "can't mount proto=%s, table=%s, hostdata=%s\n",
- proto, table, sdp->sd_args.ar_hostdata);
- goto out;
- }
+ ls->ls_ops = lm;
+ ls->ls_first = 1;
+ ls->ls_id = 0;
- if (gfs2_assert_warn(sdp, sdp->sd_lockstruct.ls_ops) ||
- gfs2_assert_warn(sdp, sdp->sd_lockstruct.ls_lvb_size >=
- GFS2_MIN_LVB_SIZE)) {
- gfs2_unmount_lockproto(&sdp->sd_lockstruct);
- goto out;
+ for (options = args->ar_hostdata; (o = strsep(&options, ":")); ) {
+ substring_t tmp[MAX_OPT_ARGS];
+ int token, option;
+
+ if (!o || !*o)
+ continue;
+
+ token = match_token(o, *lm->lm_tokens, tmp);
+ switch (token) {
+ case Opt_jid:
+ ret = match_int(&tmp[0], &option);
+ if (ret || option < 0)
+ goto hostdata_error;
+ ls->ls_jid = option;
+ break;
+ case Opt_id:
+ ret = match_int(&tmp[0], &option);
+ if (ret)
+ goto hostdata_error;
+ ls->ls_id = option;
+ break;
+ case Opt_first:
+ ret = match_int(&tmp[0], &option);
+ if (ret || (option != 0 && option != 1))
+ goto hostdata_error;
+ ls->ls_first = option;
+ break;
+ case Opt_nodir:
+ ret = match_int(&tmp[0], &option);
+ if (ret || (option != 0 && option != 1))
+ goto hostdata_error;
+ ls->ls_nodir = option;
+ break;
+ case Opt_err:
+ default:
+hostdata_error:
+ fs_info(sdp, "unknown hostdata (%s)\n", o);
+ return -EINVAL;
+ }
}
if (sdp->sd_args.ar_spectator)
@@ -1074,22 +1125,25 @@ static int gfs2_lm_mount(struct gfs2_sbd *sdp, int silent)
snprintf(sdp->sd_fsname, GFS2_FSNAME_LEN, "%s.%u", table,
sdp->sd_lockstruct.ls_jid);
- fs_info(sdp, "Joined cluster. Now mounting FS...\n");
-
- if ((sdp->sd_lockstruct.ls_flags & LM_LSFLAG_LOCAL) &&
- !sdp->sd_args.ar_ignore_local_fs) {
- sdp->sd_args.ar_localflocks = 1;
- sdp->sd_args.ar_localcaching = 1;
+ fsname = strchr(table, ':');
+ if (fsname)
+ fsname++;
+ if (lm->lm_mount == NULL) {
+ fs_info(sdp, "Now mounting FS...\n");
+ return 0;
}
-
-out:
- return error;
+ ret = lm->lm_mount(sdp, fsname);
+ if (ret == 0)
+ fs_info(sdp, "Joined cluster. Now mounting FS...\n");
+ return ret;
}
void gfs2_lm_unmount(struct gfs2_sbd *sdp)
{
- if (likely(!test_bit(SDF_SHUTDOWN, &sdp->sd_flags)))
- gfs2_unmount_lockproto(&sdp->sd_lockstruct);
+ const struct lm_lockops *lm = sdp->sd_lockstruct.ls_ops;
+ if (likely(!test_bit(SDF_SHUTDOWN, &sdp->sd_flags)) &&
+ lm->lm_unmount)
+ lm->lm_unmount(sdp);
}
/**
diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c
index 4987754..abd5429 100644
--- a/fs/gfs2/ops_inode.c
+++ b/fs/gfs2/ops_inode.c
@@ -18,7 +18,6 @@
#include <linux/posix_acl.h>
#include <linux/gfs2_ondisk.h>
#include <linux/crc32.h>
-#include <linux/lm_interface.h>
#include <linux/fiemap.h>
#include <asm/uaccess.h>
diff --git a/fs/gfs2/ops_super.c b/fs/gfs2/ops_super.c
index f0699ac..4ecdad0 100644
--- a/fs/gfs2/ops_super.c
+++ b/fs/gfs2/ops_super.c
@@ -19,7 +19,6 @@
#include <linux/delay.h>
#include <linux/gfs2_ondisk.h>
#include <linux/crc32.h>
-#include <linux/lm_interface.h>
#include <linux/time.h>
#include "gfs2.h"
diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c
index e8ef0f8..8d53f66 100644
--- a/fs/gfs2/quota.c
+++ b/fs/gfs2/quota.c
@@ -45,7 +45,6 @@
#include <linux/fs.h>
#include <linux/bio.h>
#include <linux/gfs2_ondisk.h>
-#include <linux/lm_interface.h>
#include <linux/kthread.h>
#include <linux/freezer.h>
@@ -108,7 +107,7 @@ int gfs2_shrink_qd_memory(int nr, gfp_t gfp_mask)
gfs2_assert_warn(sdp, !qd->qd_slot_count);
gfs2_assert_warn(sdp, !qd->qd_bh_count);
- gfs2_lvb_unhold(qd->qd_gl);
+ gfs2_glock_put(qd->qd_gl);
atomic_dec(&sdp->sd_quota_count);
/* Delete it from the common reclaim list */
@@ -157,11 +156,6 @@ static int qd_alloc(struct gfs2_sbd *sdp, int user, u32 id,
if (error)
goto fail;
- error = gfs2_lvb_hold(qd->qd_gl);
- gfs2_glock_put(qd->qd_gl);
- if (error)
- goto fail;
-
*qdp = qd;
return 0;
@@ -211,7 +205,7 @@ static int qd_get(struct gfs2_sbd *sdp, int user, u32 id, int create,
if (qd || !create) {
if (new_qd) {
- gfs2_lvb_unhold(new_qd->qd_gl);
+ gfs2_glock_put(new_qd->qd_gl);
kmem_cache_free(gfs2_quotad_cachep, new_qd);
}
*qdp = qd;
@@ -1280,7 +1274,7 @@ void gfs2_quota_cleanup(struct gfs2_sbd *sdp)
gfs2_assert_warn(sdp, qd->qd_slot_count == 1);
gfs2_assert_warn(sdp, !qd->qd_bh_count);
- gfs2_lvb_unhold(qd->qd_gl);
+ gfs2_glock_put(qd->qd_gl);
kmem_cache_free(gfs2_quotad_cachep, qd);
spin_lock(&qd_lru_lock);
diff --git a/fs/gfs2/recovery.c b/fs/gfs2/recovery.c
index efd09c3..247e8f7 100644
--- a/fs/gfs2/recovery.c
+++ b/fs/gfs2/recovery.c
@@ -13,7 +13,6 @@
#include <linux/buffer_head.h>
#include <linux/gfs2_ondisk.h>
#include <linux/crc32.h>
-#include <linux/lm_interface.h>
#include <linux/kthread.h>
#include <linux/freezer.h>
@@ -427,20 +426,23 @@ static int clean_journal(struct gfs2_jdesc *jd, struct gfs2_log_header_host *hea
}
-static void gfs2_lm_recovery_done(struct gfs2_sbd *sdp, unsigned int jid,
- unsigned int message)
+static void gfs2_recovery_done(struct gfs2_sbd *sdp, unsigned int jid,
+ unsigned int message)
{
- if (!sdp->sd_lockstruct.ls_ops->lm_recovery_done)
- return;
-
- if (likely(!test_bit(SDF_SHUTDOWN, &sdp->sd_flags)))
- sdp->sd_lockstruct.ls_ops->lm_recovery_done(
- sdp->sd_lockstruct.ls_lockspace, jid, message);
+ char env_jid[20];
+ char env_status[20];
+ char *envp[] = { env_jid, env_status, NULL };
+ struct lm_lockstruct *ls = &sdp->sd_lockstruct;
+ ls->ls_recover_jid_done = jid;
+ ls->ls_recover_jid_status = message;
+ sprintf(env_jid, "JID=%d", jid);
+ sprintf(env_status, "RECOVERY=%s",
+ message == LM_RD_SUCCESS ? "Done" : "Failed");
+ kobject_uevent_env(&sdp->sd_kobj, KOBJ_CHANGE, envp);
}
-
/**
- * gfs2_recover_journal - recovery a given journal
+ * gfs2_recover_journal - recover a given journal
* @jd: the struct gfs2_jdesc describing the journal
*
* Acquire the journal's lock, check to see if the journal is clean, and
@@ -561,7 +563,7 @@ int gfs2_recover_journal(struct gfs2_jdesc *jd)
if (jd->jd_jid != sdp->sd_lockstruct.ls_jid)
gfs2_glock_dq_uninit(&ji_gh);
- gfs2_lm_recovery_done(sdp, jd->jd_jid, LM_RD_SUCCESS);
+ gfs2_recovery_done(sdp, jd->jd_jid, LM_RD_SUCCESS);
if (jd->jd_jid != sdp->sd_lockstruct.ls_jid)
gfs2_glock_dq_uninit(&j_gh);
@@ -581,7 +583,7 @@ fail_gunlock_j:
fs_info(sdp, "jid=%u: %s\n", jd->jd_jid, (error) ? "Failed" : "Done");
fail:
- gfs2_lm_recovery_done(sdp, jd->jd_jid, LM_RD_GAVEUP);
+ gfs2_recovery_done(sdp, jd->jd_jid, LM_RD_GAVEUP);
return error;
}
diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
index 8b01c63..ba5a021 100644
--- a/fs/gfs2/rgrp.c
+++ b/fs/gfs2/rgrp.c
@@ -13,7 +13,6 @@
#include <linux/buffer_head.h>
#include <linux/fs.h>
#include <linux/gfs2_ondisk.h>
-#include <linux/lm_interface.h>
#include <linux/prefetch.h>
#include "gfs2.h"
diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c
index 141b781..7cf302b 100644
--- a/fs/gfs2/super.c
+++ b/fs/gfs2/super.c
@@ -15,7 +15,6 @@
#include <linux/crc32.h>
#include <linux/gfs2_ondisk.h>
#include <linux/bio.h>
-#include <linux/lm_interface.h>
#include "gfs2.h"
#include "incore.h"
diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c
index a58a120..a78997e 100644
--- a/fs/gfs2/sys.c
+++ b/fs/gfs2/sys.c
@@ -14,9 +14,8 @@
#include <linux/buffer_head.h>
#include <linux/module.h>
#include <linux/kobject.h>
-#include <linux/gfs2_ondisk.h>
-#include <linux/lm_interface.h>
#include <asm/uaccess.h>
+#include <linux/gfs2_ondisk.h>
#include "gfs2.h"
#include "incore.h"
@@ -224,14 +223,145 @@ static struct lockstruct_attr lockstruct_attr_##name = __ATTR_RO(name)
LOCKSTRUCT_ATTR(jid, "%u\n");
LOCKSTRUCT_ATTR(first, "%u\n");
-LOCKSTRUCT_ATTR(lvb_size, "%u\n");
-LOCKSTRUCT_ATTR(flags, "%d\n");
static struct attribute *lockstruct_attrs[] = {
&lockstruct_attr_jid.attr,
&lockstruct_attr_first.attr,
- &lockstruct_attr_lvb_size.attr,
- &lockstruct_attr_flags.attr,
+ NULL,
+};
+
+/*
+ * lock_module. Originally from lock_dlm
+ */
+
+static ssize_t proto_name_show(struct gfs2_sbd *sdp, char *buf)
+{
+ const struct lm_lockops *ops = sdp->sd_lockstruct.ls_ops;
+ return sprintf(buf, "%s\n", ops->lm_proto_name);
+}
+
+static ssize_t block_show(struct gfs2_sbd *sdp, char *buf)
+{
+ struct lm_lockstruct *ls = &sdp->sd_lockstruct;
+ ssize_t ret;
+ int val = 0;
+
+ if (test_bit(DFL_BLOCK_LOCKS, &ls->ls_flags))
+ val = 1;
+ ret = sprintf(buf, "%d\n", val);
+ return ret;
+}
+
+static ssize_t block_store(struct gfs2_sbd *sdp, const char *buf, size_t len)
+{
+ struct lm_lockstruct *ls = &sdp->sd_lockstruct;
+ ssize_t ret = len;
+ int val;
+
+ val = simple_strtol(buf, NULL, 0);
+
+ if (val == 1)
+ set_bit(DFL_BLOCK_LOCKS, &ls->ls_flags);
+ else if (val == 0) {
+ clear_bit(DFL_BLOCK_LOCKS, &ls->ls_flags);
+ smp_mb__after_clear_bit();
+ gfs2_glock_thaw(sdp);
+ } else {
+ ret = -EINVAL;
+ }
+ 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;
+ return sprintf(buf, "%d\n", ls->ls_first);
+}
+
+static ssize_t first_done_show(struct gfs2_sbd *sdp, char *buf)
+{
+ struct lm_lockstruct *ls = &sdp->sd_lockstruct;
+ return sprintf(buf, "%d\n", ls->ls_first_done);
+}
+
+static ssize_t recover_show(struct gfs2_sbd *sdp, char *buf)
+{
+ struct lm_lockstruct *ls = &sdp->sd_lockstruct;
+ return sprintf(buf, "%d\n", ls->ls_recover_jid);
+}
+
+static void gfs2_jdesc_make_dirty(struct gfs2_sbd *sdp, unsigned int jid)
+{
+ struct gfs2_jdesc *jd;
+
+ spin_lock(&sdp->sd_jindex_spin);
+ list_for_each_entry(jd, &sdp->sd_jindex_list, jd_list) {
+ if (jd->jd_jid != jid)
+ continue;
+ jd->jd_dirty = 1;
+ break;
+ }
+ spin_unlock(&sdp->sd_jindex_spin);
+}
+
+static ssize_t recover_store(struct gfs2_sbd *sdp, const char *buf, size_t len)
+{
+ struct lm_lockstruct *ls = &sdp->sd_lockstruct;
+ ls->ls_recover_jid = simple_strtol(buf, NULL, 0);
+ gfs2_jdesc_make_dirty(sdp, ls->ls_recover_jid);
+ if (sdp->sd_recoverd_process)
+ wake_up_process(sdp->sd_recoverd_process);
+ return len;
+}
+
+static ssize_t recover_done_show(struct gfs2_sbd *sdp, char *buf)
+{
+ struct lm_lockstruct *ls = &sdp->sd_lockstruct;
+ return sprintf(buf, "%d\n", ls->ls_recover_jid_done);
+}
+
+static ssize_t recover_status_show(struct gfs2_sbd *sdp, char *buf)
+{
+ struct lm_lockstruct *ls = &sdp->sd_lockstruct;
+ return sprintf(buf, "%d\n", ls->ls_recover_jid_status);
+}
+
+struct gdlm_attr {
+ struct attribute attr;
+ ssize_t (*show)(struct gfs2_sbd *sdp, char *);
+ ssize_t (*store)(struct gfs2_sbd *sdp, const char *, size_t);
+};
+
+#define GDLM_ATTR(_name,_mode,_show,_store) \
+static struct gdlm_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(first, 0444, lkfirst_show, NULL);
+GDLM_ATTR(first_done, 0444, first_done_show, NULL);
+GDLM_ATTR(recover, 0644, recover_show, recover_store);
+GDLM_ATTR(recover_done, 0444, recover_done_show, NULL);
+GDLM_ATTR(recover_status, 0444, recover_status_show, NULL);
+
+static struct attribute *lock_module_attrs[] = {
+ &gdlm_attr_proto_name.attr,
+ &gdlm_attr_block.attr,
+ &gdlm_attr_withdraw.attr,
+ &gdlm_attr_id.attr,
+ &lockstruct_attr_jid.attr,
+ &gdlm_attr_first.attr,
+ &gdlm_attr_first_done.attr,
+ &gdlm_attr_recover.attr,
+ &gdlm_attr_recover_done.attr,
+ &gdlm_attr_recover_status.attr,
NULL,
};
@@ -412,6 +542,11 @@ static struct attribute_group tune_group = {
.attrs = tune_attrs,
};
+static struct attribute_group lock_module_group = {
+ .name = "lock_module",
+ .attrs = lock_module_attrs,
+};
+
int gfs2_sys_fs_add(struct gfs2_sbd *sdp)
{
int error;
@@ -434,9 +569,15 @@ int gfs2_sys_fs_add(struct gfs2_sbd *sdp)
if (error)
goto fail_args;
+ error = sysfs_create_group(&sdp->sd_kobj, &lock_module_group);
+ if (error)
+ goto fail_tune;
+
kobject_uevent(&sdp->sd_kobj, KOBJ_ADD);
return 0;
+fail_tune:
+ sysfs_remove_group(&sdp->sd_kobj, &tune_group);
fail_args:
sysfs_remove_group(&sdp->sd_kobj, &args_group);
fail_lockstruct:
@@ -453,6 +594,7 @@ void gfs2_sys_fs_del(struct gfs2_sbd *sdp)
sysfs_remove_group(&sdp->sd_kobj, &tune_group);
sysfs_remove_group(&sdp->sd_kobj, &args_group);
sysfs_remove_group(&sdp->sd_kobj, &lockstruct_group);
+ sysfs_remove_group(&sdp->sd_kobj, &lock_module_group);
kobject_put(&sdp->sd_kobj);
}
diff --git a/fs/gfs2/trans.c b/fs/gfs2/trans.c
index f677b8a..33cd523 100644
--- a/fs/gfs2/trans.c
+++ b/fs/gfs2/trans.c
@@ -12,9 +12,8 @@
#include <linux/spinlock.h>
#include <linux/completion.h>
#include <linux/buffer_head.h>
-#include <linux/gfs2_ondisk.h>
#include <linux/kallsyms.h>
-#include <linux/lm_interface.h>
+#include <linux/gfs2_ondisk.h>
#include "gfs2.h"
#include "incore.h"
diff --git a/fs/gfs2/util.c b/fs/gfs2/util.c
index 374f50e..9d12b11 100644
--- a/fs/gfs2/util.c
+++ b/fs/gfs2/util.c
@@ -13,7 +13,6 @@
#include <linux/buffer_head.h>
#include <linux/crc32.h>
#include <linux/gfs2_ondisk.h>
-#include <linux/lm_interface.h>
#include <asm/uaccess.h>
#include "gfs2.h"
@@ -35,6 +34,8 @@ void gfs2_assert_i(struct gfs2_sbd *sdp)
int gfs2_lm_withdraw(struct gfs2_sbd *sdp, char *fmt, ...)
{
+ struct lm_lockstruct *ls = &sdp->sd_lockstruct;
+ const struct lm_lockops *lm = ls->ls_ops;
va_list args;
if (test_and_set_bit(SDF_SHUTDOWN, &sdp->sd_flags))
@@ -47,8 +48,12 @@ int gfs2_lm_withdraw(struct gfs2_sbd *sdp, char *fmt, ...)
fs_err(sdp, "about to withdraw this file system\n");
BUG_ON(sdp->sd_args.ar_debug);
- fs_err(sdp, "telling LM to withdraw\n");
- gfs2_withdraw_lockproto(&sdp->sd_lockstruct);
+ kobject_uevent(&sdp->sd_kobj, KOBJ_OFFLINE);
+
+ if (lm->lm_unmount) {
+ fs_err(sdp, "telling LM to unmount\n");
+ lm->lm_unmount(sdp);
+ }
fs_err(sdp, "withdrawn\n");
dump_stack();
diff --git a/include/linux/lm_interface.h b/include/linux/lm_interface.h
deleted file mode 100644
index 2ed8fa1..0000000
--- a/include/linux/lm_interface.h
+++ /dev/null
@@ -1,277 +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 __LM_INTERFACE_DOT_H__
-#define __LM_INTERFACE_DOT_H__
-
-
-typedef void (*lm_callback_t) (void *ptr, unsigned int type, void *data);
-
-/*
- * lm_mount() flags
- *
- * LM_MFLAG_SPECTATOR
- * GFS is asking to join the filesystem's lockspace, but it doesn't want to
- * modify the filesystem. The lock module shouldn't assign a journal to the FS
- * mount. It shouldn't send recovery callbacks to the FS mount. If the node
- * dies or withdraws, all locks can be wiped immediately.
- *
- * LM_MFLAG_CONV_NODROP
- * Do not allow the dlm to internally resolve conversion deadlocks by demoting
- * the lock to unlocked and then reacquiring it in the requested mode. Instead,
- * it should cancel the request and return LM_OUT_CONV_DEADLK.
- */
-
-#define LM_MFLAG_SPECTATOR 0x00000001
-#define LM_MFLAG_CONV_NODROP 0x00000002
-
-/*
- * lm_lockstruct flags
- *
- * LM_LSFLAG_LOCAL
- * The lock_nolock module returns LM_LSFLAG_LOCAL to GFS, indicating that GFS
- * can make single-node optimizations.
- */
-
-#define LM_LSFLAG_LOCAL 0x00000001
-
-/*
- * lm_lockname types
- */
-
-#define LM_TYPE_RESERVED 0x00
-#define LM_TYPE_NONDISK 0x01
-#define LM_TYPE_INODE 0x02
-#define LM_TYPE_RGRP 0x03
-#define LM_TYPE_META 0x04
-#define LM_TYPE_IOPEN 0x05
-#define LM_TYPE_FLOCK 0x06
-#define LM_TYPE_PLOCK 0x07
-#define LM_TYPE_QUOTA 0x08
-#define LM_TYPE_JOURNAL 0x09
-
-/*
- * lm_lock() states
- *
- * SHARED is compatible with SHARED, not with DEFERRED or EX.
- * DEFERRED is compatible with DEFERRED, not with SHARED or EX.
- */
-
-#define LM_ST_UNLOCKED 0
-#define LM_ST_EXCLUSIVE 1
-#define LM_ST_DEFERRED 2
-#define LM_ST_SHARED 3
-
-/*
- * lm_lock() flags
- *
- * LM_FLAG_TRY
- * Don't wait to acquire the lock if it can't be granted immediately.
- *
- * LM_FLAG_TRY_1CB
- * Send one blocking callback if TRY is set and the lock is not granted.
- *
- * LM_FLAG_NOEXP
- * GFS sets this flag on lock requests it makes while doing journal recovery.
- * These special requests should not be blocked due to the recovery like
- * ordinary locks would be.
- *
- * LM_FLAG_ANY
- * A SHARED request may also be granted in DEFERRED, or a DEFERRED request may
- * also be granted in SHARED. The preferred state is whichever is compatible
- * with other granted locks, or the specified state if no other locks exist.
- *
- * LM_FLAG_PRIORITY
- * Override fairness considerations. Suppose a lock is held in a shared state
- * and there is a pending request for the deferred state. A shared lock
- * request with the priority flag would be allowed to bypass the deferred
- * request and directly join the other shared lock. A shared lock request
- * without the priority flag might be forced to wait until the deferred
- * requested had acquired and released the lock.
- */
-
-#define LM_FLAG_TRY 0x00000001
-#define LM_FLAG_TRY_1CB 0x00000002
-#define LM_FLAG_NOEXP 0x00000004
-#define LM_FLAG_ANY 0x00000008
-#define LM_FLAG_PRIORITY 0x00000010
-
-/*
- * lm_lock() and lm_async_cb return flags
- *
- * LM_OUT_ST_MASK
- * Masks the lower two bits of lock state in the returned value.
- *
- * LM_OUT_CACHEABLE
- * The lock hasn't been released so GFS can continue to cache data for it.
- *
- * LM_OUT_CANCELED
- * The lock request was canceled.
- *
- * LM_OUT_ASYNC
- * The result of the request will be returned in an LM_CB_ASYNC callback.
- *
- * LM_OUT_CONV_DEADLK
- * The lock request was canceled do to a conversion deadlock.
- */
-
-#define LM_OUT_ST_MASK 0x00000003
-#define LM_OUT_CANCELED 0x00000008
-#define LM_OUT_ASYNC 0x00000080
-#define LM_OUT_ERROR 0x00000100
-
-/*
- * lm_callback_t types
- *
- * LM_CB_NEED_E LM_CB_NEED_D LM_CB_NEED_S
- * Blocking callback, a remote node is requesting the given lock in
- * EXCLUSIVE, DEFERRED, or SHARED.
- *
- * LM_CB_NEED_RECOVERY
- * The given journal needs to be recovered.
- *
- * LM_CB_ASYNC
- * The given lock has been granted.
- */
-
-#define LM_CB_NEED_E 257
-#define LM_CB_NEED_D 258
-#define LM_CB_NEED_S 259
-#define LM_CB_NEED_RECOVERY 260
-#define LM_CB_ASYNC 262
-
-/*
- * lm_recovery_done() messages
- */
-
-#define LM_RD_GAVEUP 308
-#define LM_RD_SUCCESS 309
-
-
-struct lm_lockname {
- u64 ln_number;
- unsigned int ln_type;
-};
-
-#define lm_name_equal(name1, name2) \
- (((name1)->ln_number == (name2)->ln_number) && \
- ((name1)->ln_type == (name2)->ln_type)) \
-
-struct lm_async_cb {
- struct lm_lockname lc_name;
- int lc_ret;
-};
-
-struct lm_lockstruct;
-
-struct lm_lockops {
- const char *lm_proto_name;
-
- /*
- * Mount/Unmount
- */
-
- int (*lm_mount) (char *table_name, char *host_data,
- lm_callback_t cb, void *cb_data,
- unsigned int min_lvb_size, int flags,
- struct lm_lockstruct *lockstruct,
- struct kobject *fskobj);
-
- void (*lm_others_may_mount) (void *lockspace);
-
- void (*lm_unmount) (void *lockspace);
-
- void (*lm_withdraw) (void *lockspace);
-
- /*
- * Lock oriented operations
- */
-
- int (*lm_get_lock) (void *lockspace, struct lm_lockname *name, void **lockp);
-
- void (*lm_put_lock) (void *lock);
-
- unsigned int (*lm_lock) (void *lock, unsigned int cur_state,
- unsigned int req_state, unsigned int flags);
-
- unsigned int (*lm_unlock) (void *lock, unsigned int cur_state);
-
- void (*lm_cancel) (void *lock);
-
- int (*lm_hold_lvb) (void *lock, char **lvbp);
- void (*lm_unhold_lvb) (void *lock, char *lvb);
-
- /*
- * Posix Lock oriented operations
- */
-
- int (*lm_plock_get) (void *lockspace, struct lm_lockname *name,
- struct file *file, struct file_lock *fl);
-
- int (*lm_plock) (void *lockspace, struct lm_lockname *name,
- struct file *file, int cmd, struct file_lock *fl);
-
- int (*lm_punlock) (void *lockspace, struct lm_lockname *name,
- struct file *file, struct file_lock *fl);
-
- /*
- * Client oriented operations
- */
-
- void (*lm_recovery_done) (void *lockspace, unsigned int jid,
- unsigned int message);
-
- struct module *lm_owner;
-};
-
-/*
- * lm_mount() return values
- *
- * ls_jid - the journal ID this node should use
- * ls_first - this node is the first to mount the file system
- * ls_lvb_size - size in bytes of lock value blocks
- * ls_lockspace - lock module's context for this file system
- * ls_ops - lock module's functions
- * ls_flags - lock module features
- */
-
-struct lm_lockstruct {
- unsigned int ls_jid;
- unsigned int ls_first;
- unsigned int ls_lvb_size;
- void *ls_lockspace;
- const struct lm_lockops *ls_ops;
- int ls_flags;
-};
-
-/*
- * Lock module bottom interface. A lock module makes itself available to GFS
- * with these functions.
- */
-
-int gfs2_register_lockproto(const struct lm_lockops *proto);
-void gfs2_unregister_lockproto(const struct lm_lockops *proto);
-
-/*
- * Lock module top interface. GFS calls these functions when mounting or
- * unmounting a file system.
- */
-
-int gfs2_mount_lockproto(char *proto_name, char *table_name, char *host_data,
- lm_callback_t cb, void *cb_data,
- unsigned int min_lvb_size, int flags,
- struct lm_lockstruct *lockstruct,
- struct kobject *fskobj);
-
-void gfs2_unmount_lockproto(struct lm_lockstruct *lockstruct);
-
-void gfs2_withdraw_lockproto(struct lm_lockstruct *lockstruct);
-
-#endif /* __LM_INTERFACE_DOT_H__ */
-
--
1.6.0.3
^ permalink raw reply related [flat|nested] 34+ messages in thread
* [Cluster-devel] [PATCH 06/18] GFS2: Remove unused field from glock
2009-03-18 12:23 ` [Cluster-devel] [PATCH 05/18] GFS2: Merge lock_dlm module into GFS2 swhiteho
@ 2009-03-18 12:23 ` swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 07/18] GFS2: Fix error path ref counting for root inode swhiteho
0 siblings, 1 reply; 34+ messages in thread
From: swhiteho @ 2009-03-18 12:23 UTC (permalink / raw)
To: cluster-devel.redhat.com
From: Steven Whitehouse <swhiteho@redhat.com>
The time stamp field is unused in the glock now that we are
using a shrinker, so that we can remove it and save sizeof(unsigned long)
bytes in each glock.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c
index cd200a5..173e59c 100644
--- a/fs/gfs2/glock.c
+++ b/fs/gfs2/glock.c
@@ -700,7 +700,6 @@ int gfs2_glock_get(struct gfs2_sbd *sdp, u64 number,
snprintf(gl->gl_strname, GDLM_STRNAME_BYTES, "%8x%16llx", name.ln_type, (unsigned long long)number);
memset(&gl->gl_lksb, 0, sizeof(struct dlm_lksb));
gl->gl_lksb.sb_lvbptr = gl->gl_lvb;
- gl->gl_stamp = jiffies;
gl->gl_tchange = jiffies;
gl->gl_object = NULL;
gl->gl_sbd = sdp;
@@ -1008,7 +1007,6 @@ void gfs2_glock_dq(struct gfs2_holder *gh)
spin_lock(&gl->gl_spin);
clear_bit(GLF_LOCK, &gl->gl_flags);
}
- gl->gl_stamp = jiffies;
if (list_empty(&gl->gl_holders) &&
!test_bit(GLF_PENDING_DEMOTE, &gl->gl_flags) &&
!test_bit(GLF_DEMOTE, &gl->gl_flags))
diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h
index 0af7c24..8fe0675 100644
--- a/fs/gfs2/incore.h
+++ b/fs/gfs2/incore.h
@@ -211,7 +211,6 @@ struct gfs2_glock {
char gl_strname[GDLM_STRNAME_BYTES];
struct dlm_lksb gl_lksb;
char gl_lvb[32];
- unsigned long gl_stamp;
unsigned long gl_tchange;
void *gl_object;
--
1.6.0.3
^ permalink raw reply related [flat|nested] 34+ messages in thread
* [Cluster-devel] [PATCH 07/18] GFS2: Fix error path ref counting for root inode
2009-03-18 12:23 ` [Cluster-devel] [PATCH 06/18] GFS2: Remove unused field from glock swhiteho
@ 2009-03-18 12:23 ` swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 08/18] GFS2: Fix deadlock on journal flush swhiteho
0 siblings, 1 reply; 34+ messages in thread
From: swhiteho @ 2009-03-18 12:23 UTC (permalink / raw)
To: cluster-devel.redhat.com
From: Steven Whitehouse <swhiteho@redhat.com>
We were keeping hold of an extra ref to the root inode in one
of the error paths, that resulted in a hang.
Reported-by: Nate Straz <nstraz@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Tested-by: Robert Peterson <rpeterso@redhat.com>
diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index 95bb33e..e502b37 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -1258,6 +1258,8 @@ fail_sb:
dput(sdp->sd_root_dir);
if (sdp->sd_master_dir)
dput(sdp->sd_master_dir);
+ if (sb->s_root)
+ dput(sb->s_root);
sb->s_root = NULL;
fail_locking:
init_locking(sdp, &mount_gh, UNDO);
--
1.6.0.3
^ permalink raw reply related [flat|nested] 34+ messages in thread
* [Cluster-devel] [PATCH 08/18] GFS2: Fix deadlock on journal flush
2009-03-18 12:23 ` [Cluster-devel] [PATCH 07/18] GFS2: Fix error path ref counting for root inode swhiteho
@ 2009-03-18 12:23 ` swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 09/18] GFS2: Support generation of discard requests swhiteho
0 siblings, 1 reply; 34+ messages in thread
From: swhiteho @ 2009-03-18 12:23 UTC (permalink / raw)
To: cluster-devel.redhat.com
From: Steven Whitehouse <swhiteho@redhat.com>
This patch fixes a deadlock when the journal is flushed and there
are dirty inodes other than the one which caused the journal flush.
Originally the journal flushing code was trying to obtain the
transaction glock while running the flush code for an inode glock.
We no longer require the transaction glock at this point in time
since we know that any attempt to get the transaction glock from
another node will result in a journal flush. So if we are flushing
the journal, we can be sure that the transaction lock is still
cached from when the transaction was started.
By inlining a version of gfs2_trans_begin() (minus the bit which
gets the transaction glock) we can avoid the deadlock problems
caused if there is a demote request queued up on the transaction
glock.
In addition I've also moved the umount rwsem so that it covers
the glock workqueue, since it all demotions are done by this
workqueue now. That fixes a bug on umount which I came across
while fixing the original problem.
Reported-by: David Teigland <teigland@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c
index 173e59c..ad8e121 100644
--- a/fs/gfs2/glock.c
+++ b/fs/gfs2/glock.c
@@ -167,6 +167,7 @@ static void glock_free(struct gfs2_glock *gl)
static void gfs2_glock_hold(struct gfs2_glock *gl)
{
+ GLOCK_BUG_ON(gl, atomic_read(&gl->gl_ref) == 0);
atomic_inc(&gl->gl_ref);
}
@@ -206,16 +207,15 @@ int gfs2_glock_put(struct gfs2_glock *gl)
atomic_dec(&lru_count);
}
spin_unlock(&lru_lock);
- GLOCK_BUG_ON(gl, !list_empty(&gl->gl_lru));
GLOCK_BUG_ON(gl, !list_empty(&gl->gl_holders));
glock_free(gl);
rv = 1;
goto out;
}
- write_unlock(gl_lock_addr(gl->gl_hash));
/* 1 for being hashed, 1 for having state != LM_ST_UNLOCKED */
if (atomic_read(&gl->gl_ref) == 2)
gfs2_glock_schedule_for_reclaim(gl);
+ write_unlock(gl_lock_addr(gl->gl_hash));
out:
return rv;
}
@@ -597,10 +597,11 @@ __acquires(&gl->gl_spin)
GLOCK_BUG_ON(gl, test_bit(GLF_DEMOTE_IN_PROGRESS, &gl->gl_flags));
+ down_read(&gfs2_umount_flush_sem);
if (test_bit(GLF_DEMOTE, &gl->gl_flags) &&
gl->gl_demote_state != gl->gl_state) {
if (find_first_holder(gl))
- goto out;
+ goto out_unlock;
if (nonblock)
goto out_sched;
set_bit(GLF_DEMOTE_IN_PROGRESS, &gl->gl_flags);
@@ -611,23 +612,26 @@ __acquires(&gl->gl_spin)
gfs2_demote_wake(gl);
ret = do_promote(gl);
if (ret == 0)
- goto out;
+ goto out_unlock;
if (ret == 2)
- return;
+ goto out_sem;
gh = find_first_waiter(gl);
gl->gl_target = gh->gh_state;
if (!(gh->gh_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB)))
do_error(gl, 0); /* Fail queued try locks */
}
do_xmote(gl, gh, gl->gl_target);
+out_sem:
+ up_read(&gfs2_umount_flush_sem);
return;
out_sched:
gfs2_glock_hold(gl);
if (queue_delayed_work(glock_workqueue, &gl->gl_work, 0) == 0)
gfs2_glock_put(gl);
-out:
+out_unlock:
clear_bit(GLF_LOCK, &gl->gl_flags);
+ goto out_sem;
}
static void glock_work_func(struct work_struct *work)
@@ -1225,7 +1229,6 @@ void gfs2_glock_cb(struct gfs2_glock *gl, unsigned int state)
void gfs2_glock_complete(struct gfs2_glock *gl, int ret)
{
struct lm_lockstruct *ls = &gl->gl_sbd->sd_lockstruct;
- down_read(&gfs2_umount_flush_sem);
gl->gl_reply = ret;
if (unlikely(test_bit(DFL_BLOCK_LOCKS, &ls->ls_flags))) {
struct gfs2_holder *gh;
@@ -1236,16 +1239,13 @@ void gfs2_glock_complete(struct gfs2_glock *gl, int ret)
((ret & ~LM_OUT_ST_MASK) != 0))
set_bit(GLF_FROZEN, &gl->gl_flags);
spin_unlock(&gl->gl_spin);
- if (test_bit(GLF_FROZEN, &gl->gl_flags)) {
- up_read(&gfs2_umount_flush_sem);
+ if (test_bit(GLF_FROZEN, &gl->gl_flags))
return;
- }
}
set_bit(GLF_REPLY_PENDING, &gl->gl_flags);
gfs2_glock_hold(gl);
if (queue_delayed_work(glock_workqueue, &gl->gl_work, 0) == 0)
gfs2_glock_put(gl);
- up_read(&gfs2_umount_flush_sem);
}
/**
@@ -1389,12 +1389,10 @@ static void thaw_glock(struct gfs2_glock *gl)
{
if (!test_and_clear_bit(GLF_FROZEN, &gl->gl_flags))
return;
- down_read(&gfs2_umount_flush_sem);
set_bit(GLF_REPLY_PENDING, &gl->gl_flags);
gfs2_glock_hold(gl);
if (queue_delayed_work(glock_workqueue, &gl->gl_work, 0) == 0)
gfs2_glock_put(gl);
- up_read(&gfs2_umount_flush_sem);
}
/**
@@ -1580,7 +1578,7 @@ static const char *gflags2str(char *buf, const unsigned long *gflags)
if (test_bit(GLF_REPLY_PENDING, gflags))
*p++ = 'r';
if (test_bit(GLF_INITIAL, gflags))
- *p++ = 'i';
+ *p++ = 'I';
if (test_bit(GLF_FROZEN, gflags))
*p++ = 'F';
*p = 0;
diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c
index f07ede8..a9b7d3a 100644
--- a/fs/gfs2/glops.c
+++ b/fs/gfs2/glops.c
@@ -37,20 +37,25 @@
static void gfs2_ail_empty_gl(struct gfs2_glock *gl)
{
struct gfs2_sbd *sdp = gl->gl_sbd;
- unsigned int blocks;
struct list_head *head = &gl->gl_ail_list;
struct gfs2_bufdata *bd;
struct buffer_head *bh;
- int error;
+ struct gfs2_trans tr;
- blocks = atomic_read(&gl->gl_ail_count);
- if (!blocks)
- return;
+ memset(&tr, 0, sizeof(tr));
+ tr.tr_revokes = atomic_read(&gl->gl_ail_count);
- error = gfs2_trans_begin(sdp, 0, blocks);
- if (gfs2_assert_withdraw(sdp, !error))
+ if (!tr.tr_revokes)
return;
+ /* A shortened, inline version of gfs2_trans_begin() */
+ tr.tr_reserved = 1 + gfs2_struct2blk(sdp, tr.tr_revokes, sizeof(u64));
+ tr.tr_ip = (unsigned long)__builtin_return_address(0);
+ INIT_LIST_HEAD(&tr.tr_list_buf);
+ gfs2_log_reserve(sdp, tr.tr_reserved);
+ BUG_ON(current->journal_info);
+ current->journal_info = &tr;
+
gfs2_log_lock(sdp);
while (!list_empty(head)) {
bd = list_entry(head->next, struct gfs2_bufdata,
diff --git a/fs/gfs2/trans.c b/fs/gfs2/trans.c
index 33cd523..053752d 100644
--- a/fs/gfs2/trans.c
+++ b/fs/gfs2/trans.c
@@ -87,9 +87,11 @@ void gfs2_trans_end(struct gfs2_sbd *sdp)
if (!tr->tr_touched) {
gfs2_log_release(sdp, tr->tr_reserved);
- gfs2_glock_dq(&tr->tr_t_gh);
- gfs2_holder_uninit(&tr->tr_t_gh);
- kfree(tr);
+ if (tr->tr_t_gh.gh_gl) {
+ gfs2_glock_dq(&tr->tr_t_gh);
+ gfs2_holder_uninit(&tr->tr_t_gh);
+ kfree(tr);
+ }
return;
}
@@ -105,9 +107,11 @@ void gfs2_trans_end(struct gfs2_sbd *sdp)
}
gfs2_log_commit(sdp, tr);
- gfs2_glock_dq(&tr->tr_t_gh);
- gfs2_holder_uninit(&tr->tr_t_gh);
- kfree(tr);
+ if (tr->tr_t_gh.gh_gl) {
+ gfs2_glock_dq(&tr->tr_t_gh);
+ gfs2_holder_uninit(&tr->tr_t_gh);
+ kfree(tr);
+ }
if (sdp->sd_vfs->s_flags & MS_SYNCHRONOUS)
gfs2_log_flush(sdp, NULL);
--
1.6.0.3
^ permalink raw reply related [flat|nested] 34+ messages in thread
* [Cluster-devel] [PATCH 09/18] GFS2: Support generation of discard requests
2009-03-18 12:23 ` [Cluster-devel] [PATCH 08/18] GFS2: Fix deadlock on journal flush swhiteho
@ 2009-03-18 12:23 ` swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 10/18] GFS2: Expose UUID via sysfs/uevent swhiteho
0 siblings, 1 reply; 34+ messages in thread
From: swhiteho @ 2009-03-18 12:23 UTC (permalink / raw)
To: cluster-devel.redhat.com
From: Steven Whitehouse <swhiteho@redhat.com>
This patch allows GFS2 to generate discard requests for blocks which are
no longer useful to the filesystem (i.e. those which have been freed as
the result of an unlink operation). The requests are generated at the
time which those blocks become available for reuse in the filesystem.
In order to use this new feature, you have to specify the "discard"
mount option. The code coalesces adjacent blocks into a single extent
when generating the discard requests, thus generating the minimum
number.
If an error occurs when the request has been sent to the block device,
then it will print a message and turn off the requests for that
filesystem. If the problem is temporary, then you can use remount to
turn the option back on again. There is also a nodiscard mount option
so that you can use remount to turn discard requests off, if required.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h
index 8fe0675..3f29bd2 100644
--- a/fs/gfs2/incore.h
+++ b/fs/gfs2/incore.h
@@ -416,7 +416,7 @@ struct gfs2_args {
unsigned int ar_suiddir:1; /* suiddir support */
unsigned int ar_data:2; /* ordered/writeback */
unsigned int ar_meta:1; /* mount metafs */
- unsigned int ar_num_glockd; /* Number of glockd threads */
+ unsigned int ar_discard:1; /* discard requests */
};
struct gfs2_tune {
diff --git a/fs/gfs2/mount.c b/fs/gfs2/mount.c
index fba502a..ee69701 100644
--- a/fs/gfs2/mount.c
+++ b/fs/gfs2/mount.c
@@ -41,6 +41,8 @@ enum {
Opt_data_writeback,
Opt_data_ordered,
Opt_meta,
+ Opt_discard,
+ Opt_nodiscard,
Opt_err,
};
@@ -65,6 +67,8 @@ static const match_table_t tokens = {
{Opt_data_writeback, "data=writeback"},
{Opt_data_ordered, "data=ordered"},
{Opt_meta, "meta"},
+ {Opt_discard, "discard"},
+ {Opt_nodiscard, "nodiscard"},
{Opt_err, NULL}
};
@@ -157,6 +161,12 @@ int gfs2_mount_args(struct gfs2_sbd *sdp, struct gfs2_args *args, char *options)
case Opt_meta:
args->ar_meta = 1;
break;
+ case Opt_discard:
+ args->ar_discard = 1;
+ break;
+ case Opt_nodiscard:
+ args->ar_discard = 0;
+ break;
case Opt_err:
default:
fs_info(sdp, "invalid mount option: %s\n", o);
diff --git a/fs/gfs2/ops_super.c b/fs/gfs2/ops_super.c
index 4ecdad0..4580195 100644
--- a/fs/gfs2/ops_super.c
+++ b/fs/gfs2/ops_super.c
@@ -608,6 +608,8 @@ static int gfs2_show_options(struct seq_file *s, struct vfsmount *mnt)
}
seq_printf(s, ",data=%s", state);
}
+ if (args->ar_discard)
+ seq_printf(s, ",discard");
return 0;
}
diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
index ba5a021..789953a 100644
--- a/fs/gfs2/rgrp.c
+++ b/fs/gfs2/rgrp.c
@@ -14,6 +14,7 @@
#include <linux/fs.h>
#include <linux/gfs2_ondisk.h>
#include <linux/prefetch.h>
+#include <linux/blkdev.h>
#include "gfs2.h"
#include "incore.h"
@@ -830,6 +831,58 @@ void gfs2_rgrp_bh_put(struct gfs2_rgrpd *rgd)
spin_unlock(&sdp->sd_rindex_spin);
}
+static void gfs2_rgrp_send_discards(struct gfs2_sbd *sdp, u64 offset,
+ const struct gfs2_bitmap *bi)
+{
+ struct super_block *sb = sdp->sd_vfs;
+ struct block_device *bdev = sb->s_bdev;
+ const unsigned int sects_per_blk = sdp->sd_sb.sb_bsize /
+ bdev_hardsect_size(sb->s_bdev);
+ u64 blk;
+ sector_t start;
+ sector_t nr_sects = 0;
+ int rv;
+ unsigned int x;
+
+ for (x = 0; x < bi->bi_len; x++) {
+ const u8 *orig = bi->bi_bh->b_data + bi->bi_offset + x;
+ const u8 *clone = bi->bi_clone + bi->bi_offset + x;
+ u8 diff = ~(*orig | (*orig >> 1)) & (*clone | (*clone >> 1));
+ diff &= 0x55;
+ if (diff == 0)
+ continue;
+ blk = offset + ((bi->bi_start + x) * GFS2_NBBY);
+ blk *= sects_per_blk; /* convert to sectors */
+ while(diff) {
+ if (diff & 1) {
+ if (nr_sects == 0)
+ goto start_new_extent;
+ if ((start + nr_sects) != blk) {
+ rv = blkdev_issue_discard(bdev, start,
+ nr_sects, GFP_NOFS);
+ if (rv)
+ goto fail;
+ nr_sects = 0;
+start_new_extent:
+ start = blk;
+ }
+ nr_sects += sects_per_blk;
+ }
+ diff >>= 2;
+ blk += sects_per_blk;
+ }
+ }
+ if (nr_sects) {
+ rv = blkdev_issue_discard(bdev, start, nr_sects, GFP_NOFS);
+ if (rv)
+ goto fail;
+ }
+ return;
+fail:
+ fs_warn(sdp, "error %d on discard request, turning discards off for this filesystem", rv);
+ sdp->sd_args.ar_discard = 0;
+}
+
void gfs2_rgrp_repolish_clones(struct gfs2_rgrpd *rgd)
{
struct gfs2_sbd *sdp = rgd->rd_sbd;
@@ -840,6 +893,8 @@ void gfs2_rgrp_repolish_clones(struct gfs2_rgrpd *rgd)
struct gfs2_bitmap *bi = rgd->rd_bits + x;
if (!bi->bi_clone)
continue;
+ if (sdp->sd_args.ar_discard)
+ gfs2_rgrp_send_discards(sdp, rgd->rd_data0, bi);
memcpy(bi->bi_clone + bi->bi_offset,
bi->bi_bh->b_data + bi->bi_offset, bi->bi_len);
}
--
1.6.0.3
^ permalink raw reply related [flat|nested] 34+ messages in thread
* [Cluster-devel] [PATCH 10/18] GFS2: Expose UUID via sysfs/uevent
2009-03-18 12:23 ` [Cluster-devel] [PATCH 09/18] GFS2: Support generation of discard requests swhiteho
@ 2009-03-18 12:23 ` swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 11/18] GFS2: Add a "demote a glock" interface to sysfs swhiteho
0 siblings, 1 reply; 34+ messages in thread
From: swhiteho @ 2009-03-18 12:23 UTC (permalink / raw)
To: cluster-devel.redhat.com
From: Steven Whitehouse <swhiteho@redhat.com>
Since we have a UUID, we ought to expose it to the user via sysfs
and uevents. We already have the fs name in both of these places
(a combination of the lock proto and lock table name) so if we add
the UUID as well, we have a full set.
For older filesystems (i.e. those created before mkfs.gfs2 was writing
UUIDs by default) the sysfs file will appear zero length, and no UUID
env var will be added to the uevents.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h
index 3f29bd2..980a086 100644
--- a/fs/gfs2/incore.h
+++ b/fs/gfs2/incore.h
@@ -470,6 +470,7 @@ struct gfs2_sb_host {
char sb_lockproto[GFS2_LOCKNAME_LEN];
char sb_locktable[GFS2_LOCKNAME_LEN];
+ u8 sb_uuid[16];
};
/*
diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index e502b37..804ca72 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -234,6 +234,7 @@ static void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf)
memcpy(sb->sb_lockproto, str->sb_lockproto, GFS2_LOCKNAME_LEN);
memcpy(sb->sb_locktable, str->sb_locktable, GFS2_LOCKNAME_LEN);
+ memcpy(sb->sb_uuid, str->sb_uuid, 16);
}
/**
diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c
index a78997e..4d284d1 100644
--- a/fs/gfs2/sys.c
+++ b/fs/gfs2/sys.c
@@ -36,6 +36,30 @@ static ssize_t fsname_show(struct gfs2_sbd *sdp, char *buf)
return snprintf(buf, PAGE_SIZE, "%s\n", sdp->sd_fsname);
}
+static int gfs2_uuid_valid(const u8 *uuid)
+{
+ int i;
+
+ for (i = 0; i < 16; i++) {
+ if (uuid[i])
+ return 1;
+ }
+ return 0;
+}
+
+static ssize_t uuid_show(struct gfs2_sbd *sdp, char *buf)
+{
+ const u8 *uuid = sdp->sd_sb.sb_uuid;
+ buf[0] = '\0';
+ if (!gfs2_uuid_valid(uuid))
+ return 0;
+ return snprintf(buf, PAGE_SIZE, "%02X%02X%02X%02X-%02X%02X-"
+ "%02X%02X-%02X%02X-%02X%02X%02X%02X%02X%02X\n",
+ uuid[0], uuid[1], uuid[2], uuid[3], uuid[4], uuid[5],
+ uuid[6], uuid[7], uuid[8], uuid[9], uuid[10], uuid[11],
+ uuid[12], uuid[13], uuid[14], uuid[15]);
+}
+
static ssize_t freeze_show(struct gfs2_sbd *sdp, char *buf)
{
unsigned int count;
@@ -158,6 +182,7 @@ static struct gfs2_attr gfs2_attr_##name = __ATTR(name, mode, show, store)
GFS2_ATTR(id, 0444, id_show, NULL);
GFS2_ATTR(fsname, 0444, fsname_show, NULL);
+GFS2_ATTR(uuid, 0444, uuid_show, NULL);
GFS2_ATTR(freeze, 0644, freeze_show, freeze_store);
GFS2_ATTR(withdraw, 0644, withdraw_show, withdraw_store);
GFS2_ATTR(statfs_sync, 0200, NULL, statfs_sync_store);
@@ -168,6 +193,7 @@ GFS2_ATTR(quota_refresh_group, 0200, NULL, quota_refresh_group_store);
static struct attribute *gfs2_attrs[] = {
&gfs2_attr_id.attr,
&gfs2_attr_fsname.attr,
+ &gfs2_attr_uuid.attr,
&gfs2_attr_freeze.attr,
&gfs2_attr_withdraw.attr,
&gfs2_attr_statfs_sync.attr,
@@ -598,12 +624,23 @@ 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)
{
struct gfs2_sbd *sdp = container_of(kobj, struct gfs2_sbd, sd_kobj);
+ const u8 *uuid = sdp->sd_sb.sb_uuid;
+
add_uevent_var(env, "LOCKTABLE=%s", sdp->sd_table_name);
add_uevent_var(env, "LOCKPROTO=%s", sdp->sd_proto_name);
+ 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",
+ uuid[0], uuid[1], uuid[2], uuid[3], uuid[4],
+ uuid[5], uuid[6], uuid[7], uuid[8], uuid[9],
+ uuid[10], uuid[11], uuid[12], uuid[13],
+ uuid[14], uuid[15]);
+ }
return 0;
}
--
1.6.0.3
^ permalink raw reply related [flat|nested] 34+ messages in thread
* [Cluster-devel] [PATCH 11/18] GFS2: Add a "demote a glock" interface to sysfs
2009-03-18 12:23 ` [Cluster-devel] [PATCH 10/18] GFS2: Expose UUID via sysfs/uevent swhiteho
@ 2009-03-18 12:23 ` swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 12/18] GFS2: Fix alignment issue and tidy gfs2_bitfit swhiteho
0 siblings, 1 reply; 34+ messages in thread
From: swhiteho @ 2009-03-18 12:23 UTC (permalink / raw)
To: cluster-devel.redhat.com
From: Steven Whitehouse <swhiteho@redhat.com>
This adds a sysfs file called demote_rq to GFS2's
per filesystem directory. Its possible to use this
file to demote arbitrary glocks in exactly the same
way as if a request had come in from a remote node.
This is intended for testing issues relating to caching
of data under glocks. Despite that, the interface is
generic enough to send requests to any type of glock,
but be careful as its not always safe to send an
arbitrary message to an arbitrary glock. For that reason
and to prevent DoS, this interface is restricted to root
only.
The messages look like this:
<type>:<glocknumber> <mode>
Example:
echo -n "2:13324 EX" >/sys/fs/gfs2/unity:myfs/demote_rq
Which means "please demote inode glock (type 2) number 13324 so that
I can get an EX (exclusive) lock". The lock modes are those which
would normally be sent by a remote node in its callback so if you
want to unlock a glock, you use EX, to demote to shared, use SH or PR
(depending on whether you like GFS2 or DLM lock modes better!).
If the glock doesn't exist, you'll get -ENOENT returned. If the
arguments don't make sense, you'll get -EINVAL returned.
The plan is that this interface will be used in combination with
the blktrace patch which I recently posted for comments although
it is, of course, still useful in its own right.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c
index ad8e121..3984e47 100644
--- a/fs/gfs2/glock.c
+++ b/fs/gfs2/glock.c
@@ -684,10 +684,11 @@ int gfs2_glock_get(struct gfs2_sbd *sdp, u64 number,
gl = search_bucket(hash, sdp, &name);
read_unlock(gl_lock_addr(hash));
- if (gl || !create) {
- *glp = gl;
+ *glp = gl;
+ if (gl)
return 0;
- }
+ if (!create)
+ return -ENOENT;
gl = kmem_cache_alloc(gfs2_glock_cachep, GFP_KERNEL);
if (!gl)
diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c
index a9b7d3a..f34bc70 100644
--- a/fs/gfs2/glops.c
+++ b/fs/gfs2/glops.c
@@ -447,3 +447,15 @@ const struct gfs2_glock_operations gfs2_journal_glops = {
.go_type = LM_TYPE_JOURNAL,
};
+const struct gfs2_glock_operations *gfs2_glops_list[] = {
+ [LM_TYPE_META] = &gfs2_meta_glops,
+ [LM_TYPE_INODE] = &gfs2_inode_glops,
+ [LM_TYPE_RGRP] = &gfs2_rgrp_glops,
+ [LM_TYPE_NONDISK] = &gfs2_trans_glops,
+ [LM_TYPE_IOPEN] = &gfs2_iopen_glops,
+ [LM_TYPE_FLOCK] = &gfs2_flock_glops,
+ [LM_TYPE_NONDISK] = &gfs2_nondisk_glops,
+ [LM_TYPE_QUOTA] = &gfs2_quota_glops,
+ [LM_TYPE_JOURNAL] = &gfs2_journal_glops,
+};
+
diff --git a/fs/gfs2/glops.h b/fs/gfs2/glops.h
index a1d9b5b..b3aa2e3 100644
--- a/fs/gfs2/glops.h
+++ b/fs/gfs2/glops.h
@@ -21,5 +21,6 @@ extern const struct gfs2_glock_operations gfs2_flock_glops;
extern const struct gfs2_glock_operations gfs2_nondisk_glops;
extern const struct gfs2_glock_operations gfs2_quota_glops;
extern const struct gfs2_glock_operations gfs2_journal_glops;
+extern const struct gfs2_glock_operations *gfs2_glops_list[];
#endif /* __GLOPS_DOT_H__ */
diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
index 789953a..a068ac9 100644
--- a/fs/gfs2/rgrp.c
+++ b/fs/gfs2/rgrp.c
@@ -839,7 +839,7 @@ static void gfs2_rgrp_send_discards(struct gfs2_sbd *sdp, u64 offset,
const unsigned int sects_per_blk = sdp->sd_sb.sb_bsize /
bdev_hardsect_size(sb->s_bdev);
u64 blk;
- sector_t start;
+ sector_t start = 0;
sector_t nr_sects = 0;
int rv;
unsigned int x;
diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c
index 4d284d1..7655f50 100644
--- a/fs/gfs2/sys.c
+++ b/fs/gfs2/sys.c
@@ -24,6 +24,7 @@
#include "glock.h"
#include "quota.h"
#include "util.h"
+#include "glops.h"
static ssize_t id_show(struct gfs2_sbd *sdp, char *buf)
{
@@ -171,6 +172,46 @@ static ssize_t quota_refresh_group_store(struct gfs2_sbd *sdp, const char *buf,
return len;
}
+static ssize_t demote_rq_store(struct gfs2_sbd *sdp, const char *buf, size_t len)
+{
+ struct gfs2_glock *gl;
+ const struct gfs2_glock_operations *glops;
+ unsigned int glmode;
+ unsigned int gltype;
+ unsigned long long glnum;
+ char mode[16];
+ int rv;
+
+ if (!capable(CAP_SYS_ADMIN))
+ return -EACCES;
+
+ rv = sscanf(buf, "%u:%llu %15s", &gltype, &glnum,
+ mode);
+ if (rv != 3)
+ return -EINVAL;
+
+ if (strcmp(mode, "EX") == 0)
+ glmode = LM_ST_UNLOCKED;
+ else if ((strcmp(mode, "CW") == 0) || (strcmp(mode, "DF") == 0))
+ glmode = LM_ST_DEFERRED;
+ else if ((strcmp(mode, "PR") == 0) || (strcmp(mode, "SH") == 0))
+ glmode = LM_ST_SHARED;
+ else
+ return -EINVAL;
+
+ if (gltype > LM_TYPE_JOURNAL)
+ return -EINVAL;
+ glops = gfs2_glops_list[gltype];
+ if (glops == NULL)
+ return -EINVAL;
+ rv = gfs2_glock_get(sdp, glnum, glops, 0, &gl);
+ if (rv)
+ return rv;
+ gfs2_glock_cb(gl, glmode);
+ gfs2_glock_put(gl);
+ return len;
+}
+
struct gfs2_attr {
struct attribute attr;
ssize_t (*show)(struct gfs2_sbd *, char *);
@@ -189,6 +230,7 @@ GFS2_ATTR(statfs_sync, 0200, NULL, statfs_sync_store);
GFS2_ATTR(quota_sync, 0200, NULL, quota_sync_store);
GFS2_ATTR(quota_refresh_user, 0200, NULL, quota_refresh_user_store);
GFS2_ATTR(quota_refresh_group, 0200, NULL, quota_refresh_group_store);
+GFS2_ATTR(demote_rq, 0200, NULL, demote_rq_store);
static struct attribute *gfs2_attrs[] = {
&gfs2_attr_id.attr,
@@ -200,6 +242,7 @@ static struct attribute *gfs2_attrs[] = {
&gfs2_attr_quota_sync.attr,
&gfs2_attr_quota_refresh_user.attr,
&gfs2_attr_quota_refresh_group.attr,
+ &gfs2_attr_demote_rq.attr,
NULL,
};
--
1.6.0.3
^ permalink raw reply related [flat|nested] 34+ messages in thread
* [Cluster-devel] [PATCH 12/18] GFS2: Fix alignment issue and tidy gfs2_bitfit
2009-03-18 12:23 ` [Cluster-devel] [PATCH 11/18] GFS2: Add a "demote a glock" interface to sysfs swhiteho
@ 2009-03-18 12:23 ` swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 13/18] GFS2: Support quota/noquota mount arguments swhiteho
0 siblings, 1 reply; 34+ messages in thread
From: swhiteho @ 2009-03-18 12:23 UTC (permalink / raw)
To: cluster-devel.redhat.com
From: Steven Whitehouse <swhiteho@redhat.com>
An alignment issue with the existing bitfit algorithm was reported
on IA64. This patch attempts to fix that, and also to tidy up the
code a bit. There is now more documentation about how this works
and it has survived a number of different tests.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
index a068ac9..c0abe69 100644
--- a/fs/gfs2/rgrp.c
+++ b/fs/gfs2/rgrp.c
@@ -132,81 +132,89 @@ static inline unsigned char gfs2_testbit(struct gfs2_rgrpd *rgd,
}
/**
+ * gfs2_bit_search
+ * @ptr: Pointer to bitmap data
+ * @mask: Mask to use (normally 0x55555.... but adjusted for search start)
+ * @state: The state we are searching for
+ *
+ * We xor the bitmap data with a patter which is the bitwise opposite
+ * of what we are looking for, this gives rise to a pattern of ones
+ * wherever there is a match. Since we have two bits per entry, we
+ * take this pattern, shift it down by one place and then and it with
+ * the original. All the even bit positions (0,2,4, etc) then represent
+ * successful matches, so we mask with 0x55555..... to remove the unwanted
+ * odd bit positions.
+ *
+ * This allows searching of a whole u64 at once (32 blocks) with a
+ * single test (on 64 bit arches).
+ */
+
+static inline u64 gfs2_bit_search(const __le64 *ptr, u64 mask, u8 state)
+{
+ u64 tmp;
+ static const u64 search[] = {
+ [0] = 0xffffffffffffffff,
+ [1] = 0xaaaaaaaaaaaaaaaa,
+ [2] = 0x5555555555555555,
+ [3] = 0x0000000000000000,
+ };
+ tmp = le64_to_cpu(*ptr) ^ search[state];
+ tmp &= (tmp >> 1);
+ tmp &= mask;
+ return tmp;
+}
+
+/**
* gfs2_bitfit - Search an rgrp's bitmap buffer to find a bit-pair representing
* a block in a given allocation state.
* @buffer: the buffer that holds the bitmaps
- * @buflen: the length (in bytes) of the buffer
+ * @len: the length (in bytes) of the buffer
* @goal: start search at this block's bit-pair (within @buffer)
- * @old_state: GFS2_BLKST_XXX the state of the block we're looking for.
+ * @state: GFS2_BLKST_XXX the state of the block we're looking for.
*
* Scope of @goal and returned block number is only within this bitmap buffer,
* not entire rgrp or filesystem. @buffer will be offset from the actual
- * beginning of a bitmap block buffer, skipping any header structures.
+ * beginning of a bitmap block buffer, skipping any header structures, but
+ * headers are always a multiple of 64 bits long so that the buffer is
+ * always aligned to a 64 bit boundary.
+ *
+ * The size of the buffer is in bytes, but is it assumed that it is
+ * always ok to to read a complete multiple of 64 bits at the end
+ * of the block in case the end is no aligned to a natural boundary.
*
* Return: the block number (bitmap buffer scope) that was found
*/
-static u32 gfs2_bitfit(const u8 *buffer, unsigned int buflen, u32 goal,
- u8 old_state)
+u32 gfs2_bitfit(const u8 *buf, const unsigned int len, u32 goal, u8 state)
{
- const u8 *byte, *start, *end;
- int bit, startbit;
- u32 g1, g2, misaligned;
- unsigned long *plong;
- unsigned long lskipval;
-
- lskipval = (old_state & GFS2_BLKST_USED) ? LBITSKIP00 : LBITSKIP55;
- g1 = (goal / GFS2_NBBY);
- start = buffer + g1;
- byte = start;
- end = buffer + buflen;
- g2 = ALIGN(g1, sizeof(unsigned long));
- plong = (unsigned long *)(buffer + g2);
- startbit = bit = (goal % GFS2_NBBY) * GFS2_BIT_SIZE;
- misaligned = g2 - g1;
- if (!misaligned)
- goto ulong_aligned;
-/* parse the bitmap a byte at a time */
-misaligned:
- while (byte < end) {
- if (((*byte >> bit) & GFS2_BIT_MASK) == old_state) {
- return goal +
- (((byte - start) * GFS2_NBBY) +
- ((bit - startbit) >> 1));
- }
- bit += GFS2_BIT_SIZE;
- if (bit >= GFS2_NBBY * GFS2_BIT_SIZE) {
- bit = 0;
- byte++;
- misaligned--;
- if (!misaligned) {
- plong = (unsigned long *)byte;
- goto ulong_aligned;
- }
- }
- }
- return BFITNOENT;
-
-/* parse the bitmap a unsigned long at a time */
-ulong_aligned:
- /* Stop at "end - 1" or else prefetch can go past the end and segfault.
- We could "if" it but we'd lose some of the performance gained.
- This way will only slow down searching the very last 4/8 bytes
- depending on architecture. I've experimented with several ways
- of writing this section such as using an else before the goto
- but this one seems to be the fastest. */
- while ((unsigned char *)plong < end - sizeof(unsigned long)) {
- prefetch(plong + 1);
- if (((*plong) & LBITMASK) != lskipval)
- break;
- plong++;
- }
- if ((unsigned char *)plong < end) {
- byte = (const u8 *)plong;
- misaligned += sizeof(unsigned long) - 1;
- goto misaligned;
+ u32 spoint = (goal << 1) & ((8*sizeof(u64)) - 1);
+ const __le64 *ptr = ((__le64 *)buf) + (goal >> 5);
+ const __le64 *end = (__le64 *)(buf + ALIGN(len, sizeof(u64)));
+ u64 tmp;
+ u64 mask = 0x5555555555555555;
+ u32 bit;
+
+ BUG_ON(state > 3);
+
+ /* Mask off bits we don't care about at the start of the search */
+ mask <<= spoint;
+ tmp = gfs2_bit_search(ptr, mask, state);
+ ptr++;
+ while(tmp == 0 && ptr < end) {
+ tmp = gfs2_bit_search(ptr, 0x5555555555555555, state);
+ ptr++;
}
- return BFITNOENT;
+ /* Mask off any bits which are more than len bytes from the start */
+ if (ptr == end && (len & (sizeof(u64) - 1)))
+ tmp &= (((u64)~0) >> (64 - 8*(len & (sizeof(u64) - 1))));
+ /* Didn't find anything, so return */
+ if (tmp == 0)
+ return BFITNOENT;
+ ptr--;
+ bit = fls64(tmp);
+ bit--; /* fls64 always adds one to the bit count */
+ bit /= 2; /* two bits per entry in the bitmap */
+ return (((const unsigned char *)ptr - buf) * GFS2_NBBY) + bit;
}
/**
--
1.6.0.3
^ permalink raw reply related [flat|nested] 34+ messages in thread
* [Cluster-devel] [PATCH 13/18] GFS2: Support quota/noquota mount arguments
2009-03-18 12:23 ` [Cluster-devel] [PATCH 12/18] GFS2: Fix alignment issue and tidy gfs2_bitfit swhiteho
@ 2009-03-18 12:23 ` swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 14/18] GFS2: fix sparse warnings: constant is so big it is swhiteho
0 siblings, 1 reply; 34+ messages in thread
From: swhiteho @ 2009-03-18 12:23 UTC (permalink / raw)
To: cluster-devel.redhat.com
From: Steven Whitehouse <swhiteho@redhat.com>
This adds support for "quota" and "noquota" mount options in addition to the
existing "quota=on/off/account" so that we are compatible with the names by
which these options are more generally known.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
diff --git a/fs/gfs2/mount.c b/fs/gfs2/mount.c
index ee69701..f7e8527 100644
--- a/fs/gfs2/mount.c
+++ b/fs/gfs2/mount.c
@@ -36,6 +36,8 @@ enum {
Opt_quota_off,
Opt_quota_account,
Opt_quota_on,
+ Opt_quota,
+ Opt_noquota,
Opt_suiddir,
Opt_nosuiddir,
Opt_data_writeback,
@@ -62,6 +64,8 @@ static const match_table_t tokens = {
{Opt_quota_off, "quota=off"},
{Opt_quota_account, "quota=account"},
{Opt_quota_on, "quota=on"},
+ {Opt_quota, "quota"},
+ {Opt_noquota, "noquota"},
{Opt_suiddir, "suiddir"},
{Opt_nosuiddir, "nosuiddir"},
{Opt_data_writeback, "data=writeback"},
@@ -138,12 +142,14 @@ int gfs2_mount_args(struct gfs2_sbd *sdp, struct gfs2_args *args, char *options)
args->ar_posix_acl = 0;
break;
case Opt_quota_off:
+ case Opt_noquota:
args->ar_quota = GFS2_QUOTA_OFF;
break;
case Opt_quota_account:
args->ar_quota = GFS2_QUOTA_ACCOUNT;
break;
case Opt_quota_on:
+ case Opt_quota:
args->ar_quota = GFS2_QUOTA_ON;
break;
case Opt_suiddir:
--
1.6.0.3
^ permalink raw reply related [flat|nested] 34+ messages in thread
* [Cluster-devel] [PATCH 14/18] GFS2: fix sparse warnings: constant is so big it is ...
2009-03-18 12:23 ` [Cluster-devel] [PATCH 13/18] GFS2: Support quota/noquota mount arguments swhiteho
@ 2009-03-18 12:23 ` swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 15/18] GFS2: fix sparse warning: Should it be static? swhiteho
0 siblings, 1 reply; 34+ messages in thread
From: swhiteho @ 2009-03-18 12:23 UTC (permalink / raw)
To: cluster-devel.redhat.com
From: Hannes Eder <hannes@hanneseder.net>
Fix this sparse warnings:
fs/gfs2/rgrp.c:156:23: warning: constant 0xffffffffffffffff is so big it is unsigned long long
fs/gfs2/rgrp.c:157:23: warning: constant 0xaaaaaaaaaaaaaaaa is so big it is unsigned long long
fs/gfs2/rgrp.c:158:23: warning: constant 0x5555555555555555 is so big it is long long
fs/gfs2/rgrp.c:194:20: warning: constant 0x5555555555555555 is so big it is long long
fs/gfs2/rgrp.c:204:44: warning: constant 0x5555555555555555 is so big it is long long
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
index c0abe69..34691d7 100644
--- a/fs/gfs2/rgrp.c
+++ b/fs/gfs2/rgrp.c
@@ -153,10 +153,10 @@ static inline u64 gfs2_bit_search(const __le64 *ptr, u64 mask, u8 state)
{
u64 tmp;
static const u64 search[] = {
- [0] = 0xffffffffffffffff,
- [1] = 0xaaaaaaaaaaaaaaaa,
- [2] = 0x5555555555555555,
- [3] = 0x0000000000000000,
+ [0] = 0xffffffffffffffffULL,
+ [1] = 0xaaaaaaaaaaaaaaaaULL,
+ [2] = 0x5555555555555555ULL,
+ [3] = 0x0000000000000000ULL,
};
tmp = le64_to_cpu(*ptr) ^ search[state];
tmp &= (tmp >> 1);
@@ -191,7 +191,7 @@ u32 gfs2_bitfit(const u8 *buf, const unsigned int len, u32 goal, u8 state)
const __le64 *ptr = ((__le64 *)buf) + (goal >> 5);
const __le64 *end = (__le64 *)(buf + ALIGN(len, sizeof(u64)));
u64 tmp;
- u64 mask = 0x5555555555555555;
+ u64 mask = 0x5555555555555555ULL;
u32 bit;
BUG_ON(state > 3);
@@ -201,7 +201,7 @@ u32 gfs2_bitfit(const u8 *buf, const unsigned int len, u32 goal, u8 state)
tmp = gfs2_bit_search(ptr, mask, state);
ptr++;
while(tmp == 0 && ptr < end) {
- tmp = gfs2_bit_search(ptr, 0x5555555555555555, state);
+ tmp = gfs2_bit_search(ptr, 0x5555555555555555ULL, state);
ptr++;
}
/* Mask off any bits which are more than len bytes from the start */
--
1.6.0.3
^ permalink raw reply related [flat|nested] 34+ messages in thread
* [Cluster-devel] [PATCH 15/18] GFS2: fix sparse warning: Should it be static?
2009-03-18 12:23 ` [Cluster-devel] [PATCH 14/18] GFS2: fix sparse warnings: constant is so big it is swhiteho
@ 2009-03-18 12:23 ` swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 16/18] GFS2: Pagecache usage optimization on GFS2 swhiteho
0 siblings, 1 reply; 34+ messages in thread
From: swhiteho @ 2009-03-18 12:23 UTC (permalink / raw)
To: cluster-devel.redhat.com
From: Hannes Eder <hannes@hanneseder.net>
Impact: Make symbol static.
Fix this sparse warning:
fs/gfs2/rgrp.c:188:5: warning: symbol 'gfs2_bitfit' was not declared. Should it be static?
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
index 34691d7..f03d024 100644
--- a/fs/gfs2/rgrp.c
+++ b/fs/gfs2/rgrp.c
@@ -185,7 +185,8 @@ static inline u64 gfs2_bit_search(const __le64 *ptr, u64 mask, u8 state)
* Return: the block number (bitmap buffer scope) that was found
*/
-u32 gfs2_bitfit(const u8 *buf, const unsigned int len, u32 goal, u8 state)
+static u32 gfs2_bitfit(const u8 *buf, const unsigned int len,
+ u32 goal, u8 state)
{
u32 spoint = (goal << 1) & ((8*sizeof(u64)) - 1);
const __le64 *ptr = ((__le64 *)buf) + (goal >> 5);
--
1.6.0.3
^ permalink raw reply related [flat|nested] 34+ messages in thread
* [Cluster-devel] [PATCH 16/18] GFS2: Pagecache usage optimization on GFS2
2009-03-18 12:23 ` [Cluster-devel] [PATCH 15/18] GFS2: fix sparse warning: Should it be static? swhiteho
@ 2009-03-18 12:23 ` swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 17/18] GFS2: Fix locking bug in failed shared to exclusive conversion swhiteho
0 siblings, 1 reply; 34+ messages in thread
From: swhiteho @ 2009-03-18 12:23 UTC (permalink / raw)
To: cluster-devel.redhat.com
From: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp>
I introduced "is_partially_uptodate" aops for GFS2.
A page can have multiple buffers and even if a page is not uptodate, some buffers
can be uptodate on pagesize != blocksize environment.
This aops checks that all buffers which correspond to a part of a file
that we want to read are uptodate. If so, we do not have to issue actual
read IO to HDD even if a page is not uptodate because the portion we
want to read are uptodate.
"block_is_partially_uptodate" function is already used by ext2/3/4.
With the following patch random read/write mixed workloads or random read after
random write workloads can be optimized and we can get performance improvement.
I did a performance test using the sysbench.
#sysbench --num-threads=16 --max-requests=200000 --test=fileio --file-num=1
--file-block-size=8K --file-total-size=2G --file-test-mode=rndrw --file-fsync-freq=0
--file-rw-ratio=1 run
-2.6.29-rc6
Test execution summary:
total time: 202.6389s
total number of events: 200000
total time taken by event execution: 2580.0480
per-request statistics:
min: 0.0000s
avg: 0.0129s
max: 49.5852s
approx. 95 percentile: 0.0462s
-2.6.29-rc6-patched
Test execution summary:
total time: 177.8639s
total number of events: 200000
total time taken by event execution: 2419.0199
per-request statistics:
min: 0.0000s
avg: 0.0121s
max: 52.4306s
approx. 95 percentile: 0.0444s
arch: ia64
pagesize: 16k
blocksize: 4k
Signed-off-by: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
diff --git a/fs/gfs2/ops_address.c b/fs/gfs2/ops_address.c
index a6d00e8..a6dde17 100644
--- a/fs/gfs2/ops_address.c
+++ b/fs/gfs2/ops_address.c
@@ -1096,6 +1096,7 @@ static const struct address_space_operations gfs2_writeback_aops = {
.releasepage = gfs2_releasepage,
.direct_IO = gfs2_direct_IO,
.migratepage = buffer_migrate_page,
+ .is_partially_uptodate = block_is_partially_uptodate,
};
static const struct address_space_operations gfs2_ordered_aops = {
@@ -1111,6 +1112,7 @@ static const struct address_space_operations gfs2_ordered_aops = {
.releasepage = gfs2_releasepage,
.direct_IO = gfs2_direct_IO,
.migratepage = buffer_migrate_page,
+ .is_partially_uptodate = block_is_partially_uptodate,
};
static const struct address_space_operations gfs2_jdata_aops = {
@@ -1125,6 +1127,7 @@ static const struct address_space_operations gfs2_jdata_aops = {
.bmap = gfs2_bmap,
.invalidatepage = gfs2_invalidatepage,
.releasepage = gfs2_releasepage,
+ .is_partially_uptodate = block_is_partially_uptodate,
};
void gfs2_set_aops(struct inode *inode)
--
1.6.0.3
^ permalink raw reply related [flat|nested] 34+ messages in thread
* [Cluster-devel] [PATCH 17/18] GFS2: Fix locking bug in failed shared to exclusive conversion
2009-03-18 12:23 ` [Cluster-devel] [PATCH 16/18] GFS2: Pagecache usage optimization on GFS2 swhiteho
@ 2009-03-18 12:23 ` swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 18/18] GFS2: Clean up of glops.c swhiteho
0 siblings, 1 reply; 34+ messages in thread
From: swhiteho @ 2009-03-18 12:23 UTC (permalink / raw)
To: cluster-devel.redhat.com
From: Benjamin Marzinski <bmarzins@redhat.com>
After calling out to the dlm, GFS2 sets the new state of a glock to
gl_target in gdlm_ast(). However, gl_target is not always the lock
state that was requested. If a conversion from shared to exclusive
fails, finish_xmote() will call do_xmote() with LM_ST_UNLOCKED, instead
of gl->gl_target, so that it can reacquire the lock in exlusive the next
time around. In this case, setting the lock to gl_target in gdlm_ast()
will make GFS2 think that it has the glock in exclusive mode, when
really, it doesn't have the glock locked at all. This patch adds a new
field to the gfs2_glock structure, gl_req, to track the mode that was
requested.
Signed-off-by: Benjamin Marzinski <bmarzins@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h
index 980a086..399d1b9 100644
--- a/fs/gfs2/incore.h
+++ b/fs/gfs2/incore.h
@@ -203,6 +203,7 @@ struct gfs2_glock {
unsigned int gl_target;
unsigned int gl_reply;
unsigned int gl_hash;
+ unsigned int gl_req;
unsigned int gl_demote_state; /* state requested by remote node */
unsigned long gl_demote_time; /* time of first demote request */
struct list_head gl_holders;
diff --git a/fs/gfs2/lock_dlm.c b/fs/gfs2/lock_dlm.c
index a0bb7d2..46df988 100644
--- a/fs/gfs2/lock_dlm.c
+++ b/fs/gfs2/lock_dlm.c
@@ -46,11 +46,11 @@ static void gdlm_ast(void *arg)
BUG();
}
- ret = gl->gl_target;
+ ret = gl->gl_req;
if (gl->gl_lksb.sb_flags & DLM_SBF_ALTMODE) {
- if (gl->gl_target == LM_ST_SHARED)
+ if (gl->gl_req == LM_ST_SHARED)
ret = LM_ST_DEFERRED;
- else if (gl->gl_target == LM_ST_DEFERRED)
+ else if (gl->gl_req == LM_ST_DEFERRED)
ret = LM_ST_SHARED;
else
BUG();
@@ -147,6 +147,7 @@ static unsigned int gdlm_lock(struct gfs2_glock *gl,
int req;
u32 lkf;
+ gl->gl_req = req_state;
req = make_mode(req_state);
lkf = make_flags(gl->gl_lksb.sb_lkid, flags, req);
--
1.6.0.3
^ permalink raw reply related [flat|nested] 34+ messages in thread
* [Cluster-devel] [PATCH 18/18] GFS2: Clean up of glops.c
2009-03-18 12:23 ` [Cluster-devel] [PATCH 17/18] GFS2: Fix locking bug in failed shared to exclusive conversion swhiteho
@ 2009-03-18 12:23 ` swhiteho
0 siblings, 0 replies; 34+ messages in thread
From: swhiteho @ 2009-03-18 12:23 UTC (permalink / raw)
To: cluster-devel.redhat.com
From: Steven Whitehouse <swhiteho@redhat.com>
This cleans up a number of bits of code mostly based in glops.c.
A couple of simple functions have been merged into the callers
to make it more obvious what is going on, the mysterious raising
of i_writecount around the truncate_inode_pages() call has been
removed. The meta_go_* operations have been renamed rgrp_go_*
since that is the only lock type that they are used with.
The unused argument of gfs2_read_sb has been removed. Also
a bug has been fixed where a check for the rindex inode was
in the wrong callback. More comments are added, and the
debugging code is improved too.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c
index f34bc70..bf23a62 100644
--- a/fs/gfs2/glops.c
+++ b/fs/gfs2/glops.c
@@ -76,29 +76,7 @@ static void gfs2_ail_empty_gl(struct gfs2_glock *gl)
}
/**
- * gfs2_pte_inval - Sync and invalidate all PTEs associated with a glock
- * @gl: the glock
- *
- */
-
-static void gfs2_pte_inval(struct gfs2_glock *gl)
-{
- struct gfs2_inode *ip;
- struct inode *inode;
-
- ip = gl->gl_object;
- inode = &ip->i_inode;
- if (!ip || !S_ISREG(inode->i_mode))
- return;
-
- unmap_shared_mapping_range(inode->i_mapping, 0, 0);
- if (test_bit(GIF_SW_PAGED, &ip->i_flags))
- set_bit(GLF_DIRTY, &gl->gl_flags);
-
-}
-
-/**
- * meta_go_sync - sync out the metadata for this glock
+ * rgrp_go_sync - sync out the metadata for this glock
* @gl: the glock
*
* Called when demoting or unlocking an EX glock. We must flush
@@ -106,36 +84,42 @@ static void gfs2_pte_inval(struct gfs2_glock *gl)
* not return to caller to demote/unlock the glock until I/O is complete.
*/
-static void meta_go_sync(struct gfs2_glock *gl)
+static void rgrp_go_sync(struct gfs2_glock *gl)
{
- if (gl->gl_state != LM_ST_EXCLUSIVE)
+ struct address_space *metamapping = gl->gl_aspace->i_mapping;
+ int error;
+
+ if (!test_and_clear_bit(GLF_DIRTY, &gl->gl_flags))
return;
+ BUG_ON(gl->gl_state != LM_ST_EXCLUSIVE);
- if (test_and_clear_bit(GLF_DIRTY, &gl->gl_flags)) {
- gfs2_log_flush(gl->gl_sbd, gl);
- gfs2_meta_sync(gl);
- gfs2_ail_empty_gl(gl);
- }
+ gfs2_log_flush(gl->gl_sbd, gl);
+ filemap_fdatawrite(metamapping);
+ error = filemap_fdatawait(metamapping);
+ mapping_set_error(metamapping, error);
+ gfs2_ail_empty_gl(gl);
}
/**
- * meta_go_inval - invalidate the metadata for this glock
+ * rgrp_go_inval - invalidate the metadata for this glock
* @gl: the glock
* @flags:
*
+ * We never used LM_ST_DEFERRED with resource groups, so that we
+ * should always see the metadata flag set here.
+ *
*/
-static void meta_go_inval(struct gfs2_glock *gl, int flags)
+static void rgrp_go_inval(struct gfs2_glock *gl, int flags)
{
- if (!(flags & DIO_METADATA))
- return;
+ struct address_space *mapping = gl->gl_aspace->i_mapping;
- gfs2_meta_inval(gl);
- if (gl->gl_object == GFS2_I(gl->gl_sbd->sd_rindex))
- gl->gl_sbd->sd_rindex_uptodate = 0;
- else if (gl->gl_ops == &gfs2_rgrp_glops && gl->gl_object) {
- struct gfs2_rgrpd *rgd = (struct gfs2_rgrpd *)gl->gl_object;
+ BUG_ON(!(flags & DIO_METADATA));
+ gfs2_assert_withdraw(gl->gl_sbd, !atomic_read(&gl->gl_ail_count));
+ truncate_inode_pages(mapping, 0);
+ if (gl->gl_object) {
+ struct gfs2_rgrpd *rgd = (struct gfs2_rgrpd *)gl->gl_object;
rgd->rd_flags &= ~GFS2_RDF_UPTODATE;
}
}
@@ -152,48 +136,54 @@ static void inode_go_sync(struct gfs2_glock *gl)
struct address_space *metamapping = gl->gl_aspace->i_mapping;
int error;
- if (gl->gl_state != LM_ST_UNLOCKED)
- gfs2_pte_inval(gl);
- if (gl->gl_state != LM_ST_EXCLUSIVE)
- return;
-
if (ip && !S_ISREG(ip->i_inode.i_mode))
ip = NULL;
+ if (ip && test_and_clear_bit(GIF_SW_PAGED, &ip->i_flags))
+ unmap_shared_mapping_range(ip->i_inode.i_mapping, 0, 0);
+ if (!test_and_clear_bit(GLF_DIRTY, &gl->gl_flags))
+ return;
- if (test_bit(GLF_DIRTY, &gl->gl_flags)) {
- gfs2_log_flush(gl->gl_sbd, gl);
- filemap_fdatawrite(metamapping);
- if (ip) {
- struct address_space *mapping = ip->i_inode.i_mapping;
- filemap_fdatawrite(mapping);
- error = filemap_fdatawait(mapping);
- mapping_set_error(mapping, error);
- }
- error = filemap_fdatawait(metamapping);
- mapping_set_error(metamapping, error);
- clear_bit(GLF_DIRTY, &gl->gl_flags);
- gfs2_ail_empty_gl(gl);
+ BUG_ON(gl->gl_state != LM_ST_EXCLUSIVE);
+
+ gfs2_log_flush(gl->gl_sbd, gl);
+ filemap_fdatawrite(metamapping);
+ if (ip) {
+ struct address_space *mapping = ip->i_inode.i_mapping;
+ filemap_fdatawrite(mapping);
+ error = filemap_fdatawait(mapping);
+ mapping_set_error(mapping, error);
}
+ error = filemap_fdatawait(metamapping);
+ mapping_set_error(metamapping, error);
+ gfs2_ail_empty_gl(gl);
}
/**
* inode_go_inval - prepare a inode glock to be released
* @gl: the glock
* @flags:
+ *
+ * Normally we invlidate everything, but if we are moving into
+ * LM_ST_DEFERRED from LM_ST_SHARED or LM_ST_EXCLUSIVE then we
+ * can keep hold of the metadata, since it won't have changed.
*
*/
static void inode_go_inval(struct gfs2_glock *gl, int flags)
{
struct gfs2_inode *ip = gl->gl_object;
- int meta = (flags & DIO_METADATA);
- if (meta) {
- gfs2_meta_inval(gl);
+ gfs2_assert_withdraw(gl->gl_sbd, !atomic_read(&gl->gl_ail_count));
+
+ if (flags & DIO_METADATA) {
+ struct address_space *mapping = gl->gl_aspace->i_mapping;
+ truncate_inode_pages(mapping, 0);
if (ip)
set_bit(GIF_INVALID, &ip->i_flags);
}
+ if (ip == GFS2_I(gl->gl_sbd->sd_rindex))
+ gl->gl_sbd->sd_rindex_uptodate = 0;
if (ip && S_ISREG(ip->i_inode.i_mode))
truncate_inode_pages(ip->i_inode.i_mapping, 0);
}
@@ -395,7 +385,6 @@ static int trans_go_demote_ok(const struct gfs2_glock *gl)
}
const struct gfs2_glock_operations gfs2_meta_glops = {
- .go_xmote_th = meta_go_sync,
.go_type = LM_TYPE_META,
};
@@ -410,8 +399,8 @@ const struct gfs2_glock_operations gfs2_inode_glops = {
};
const struct gfs2_glock_operations gfs2_rgrp_glops = {
- .go_xmote_th = meta_go_sync,
- .go_inval = meta_go_inval,
+ .go_xmote_th = rgrp_go_sync,
+ .go_inval = rgrp_go_inval,
.go_demote_ok = rgrp_go_demote_ok,
.go_lock = rgrp_go_lock,
.go_unlock = rgrp_go_unlock,
diff --git a/fs/gfs2/meta_io.c b/fs/gfs2/meta_io.c
index 870d65a..8d6f132 100644
--- a/fs/gfs2/meta_io.c
+++ b/fs/gfs2/meta_io.c
@@ -89,27 +89,6 @@ void gfs2_aspace_put(struct inode *aspace)
}
/**
- * gfs2_meta_inval - Invalidate all buffers associated with a glock
- * @gl: the glock
- *
- */
-
-void gfs2_meta_inval(struct gfs2_glock *gl)
-{
- struct gfs2_sbd *sdp = gl->gl_sbd;
- struct inode *aspace = gl->gl_aspace;
- struct address_space *mapping = gl->gl_aspace->i_mapping;
-
- gfs2_assert_withdraw(sdp, !atomic_read(&gl->gl_ail_count));
-
- atomic_inc(&aspace->i_writecount);
- truncate_inode_pages(mapping, 0);
- atomic_dec(&aspace->i_writecount);
-
- gfs2_assert_withdraw(sdp, !mapping->nrpages);
-}
-
-/**
* gfs2_meta_sync - Sync all buffers associated with a glock
* @gl: The glock
*
diff --git a/fs/gfs2/meta_io.h b/fs/gfs2/meta_io.h
index b1a5f36..de270c2 100644
--- a/fs/gfs2/meta_io.h
+++ b/fs/gfs2/meta_io.h
@@ -40,7 +40,6 @@ static inline void gfs2_buffer_copy_tail(struct buffer_head *to_bh,
struct inode *gfs2_aspace_get(struct gfs2_sbd *sdp);
void gfs2_aspace_put(struct inode *aspace);
-void gfs2_meta_inval(struct gfs2_glock *gl);
void gfs2_meta_sync(struct gfs2_glock *gl);
struct buffer_head *gfs2_meta_new(struct gfs2_glock *gl, u64 blkno);
diff --git a/fs/gfs2/ops_file.c b/fs/gfs2/ops_file.c
index 99d726f..48ec3d5 100644
--- a/fs/gfs2/ops_file.c
+++ b/fs/gfs2/ops_file.c
@@ -355,7 +355,6 @@ static int gfs2_page_mkwrite(struct vm_area_struct *vma, struct page *page)
if (ret)
goto out;
- set_bit(GIF_SW_PAGED, &ip->i_flags);
ret = gfs2_write_alloc_required(ip, pos, PAGE_CACHE_SIZE, &alloc_required);
if (ret || !alloc_required)
goto out_unlock;
@@ -396,6 +395,8 @@ static int gfs2_page_mkwrite(struct vm_area_struct *vma, struct page *page)
goto out_unlock_page;
}
ret = gfs2_allocate_page_backing(page);
+ if (!ret)
+ set_bit(GIF_SW_PAGED, &ip->i_flags);
out_unlock_page:
unlock_page(page);
diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index 804ca72..51883b3 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -296,15 +296,15 @@ static int gfs2_read_super(struct gfs2_sbd *sdp, sector_t sector)
__free_page(page);
return 0;
}
+
/**
* gfs2_read_sb - Read super block
* @sdp: The GFS2 superblock
- * @gl: the glock for the superblock (assumed to be held)
* @silent: Don't print message if mount fails
*
*/
-static int gfs2_read_sb(struct gfs2_sbd *sdp, struct gfs2_glock *gl, int silent)
+static int gfs2_read_sb(struct gfs2_sbd *sdp, int silent)
{
u32 hash_blocks, ind_blocks, leaf_blocks;
u32 tmp_blocks;
@@ -524,7 +524,7 @@ static int init_sb(struct gfs2_sbd *sdp, int silent)
return ret;
}
- ret = gfs2_read_sb(sdp, sb_gh.gh_gl, silent);
+ ret = gfs2_read_sb(sdp, silent);
if (ret) {
fs_err(sdp, "can't read superblock: %d\n", ret);
goto out;
--
1.6.0.3
^ permalink raw reply related [flat|nested] 34+ messages in thread
* [Cluster-devel] GFS2: Pre-pull patch posting
@ 2009-06-10 8:30 Steven Whitehouse
0 siblings, 0 replies; 34+ messages in thread
From: Steven Whitehouse @ 2009-06-10 8:30 UTC (permalink / raw)
To: cluster-devel.redhat.com
Hi,
As the merge window is more or less upon us, here is the content
of the GFS2 -nmw tree. There is nothing too startling this time
as the focus has very much been bug fixes and clean up.
We have a new mount option, commit= which does exactly the same
thing as the ext3 equivalent option. We have a long term plan to
make all the tunable parameters available as mount options and
thus to be able to eventually drop the sysfs interface to these
parameters.
Another long term plan is to get rid of the files named
ops_somethingorother and to either merge them into other
files, or rename them to not have the ops_ prefix. This
patch series makes a start on that, and does all the easy
ones. As a result some functions with only one caller are
moved to the same file as their caller and made static.
The docs are also updated to reflect the fact that the
lock_dlm interface module no longer exists and that interface is
now built into GFS2.
Steve.
^ permalink raw reply [flat|nested] 34+ messages in thread
* [Cluster-devel] GFS2: Pre-pull patch posting
@ 2009-09-10 11:27 Steven Whitehouse
0 siblings, 0 replies; 34+ messages in thread
From: Steven Whitehouse @ 2009-09-10 11:27 UTC (permalink / raw)
To: cluster-devel.redhat.com
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.
^ permalink raw reply [flat|nested] 34+ messages in thread
* [Cluster-devel] GFS2: Pre-pull patch posting
@ 2010-03-01 15:08 Steven Whitehouse
0 siblings, 0 replies; 34+ messages in thread
From: Steven Whitehouse @ 2010-03-01 15:08 UTC (permalink / raw)
To: cluster-devel.redhat.com
Hi,
Not so many patches for GFS2 this merge window. The bulk of the changes are
aimed at reducing overheads when caching large numbers of inodes and
the consequent simplification of the umount code,
Steve.
^ permalink raw reply [flat|nested] 34+ messages in thread
* [Cluster-devel] GFS2: Pre-pull patch posting
@ 2010-03-11 17:21 Steven Whitehouse
0 siblings, 0 replies; 34+ messages in thread
From: Steven Whitehouse @ 2010-03-11 17:21 UTC (permalink / raw)
To: cluster-devel.redhat.com
Here are three small (but important!) fixes to GFS2.
Steve.
^ permalink raw reply [flat|nested] 34+ messages in thread
* [Cluster-devel] GFS2: Pre-pull patch posting
@ 2010-05-17 12:40 Steven Whitehouse
0 siblings, 0 replies; 34+ messages in thread
From: Steven Whitehouse @ 2010-05-17 12:40 UTC (permalink / raw)
To: cluster-devel.redhat.com
Hi,
Nothing very exciting this time.... mostly minor bug fixes and
a docs update. The gfs2_logd patch has been hanging around for
a long time and is now finally integrated. It is the first step
towards a longer term goal of improving performance in that
area,
Steve.
^ permalink raw reply [flat|nested] 34+ messages in thread
* [Cluster-devel] GFS2: Pre-pull patch posting
@ 2010-08-02 9:27 Steven Whitehouse
0 siblings, 0 replies; 34+ messages in thread
From: Steven Whitehouse @ 2010-08-02 9:27 UTC (permalink / raw)
To: cluster-devel.redhat.com
Hi,
Here is the current content of the GFS2 -nmw git tree. Mostly its
just clean up and bug fixes this time. There is one exception which
is the "wait for journal id" patch which is a new feature aimed
at (eventually) allowing us to simplify the userland support which
GFS2 requires,
Steve.
^ permalink raw reply [flat|nested] 34+ messages in thread
* [Cluster-devel] GFS2: Pre-pull patch posting
@ 2010-10-18 14:15 Steven Whitehouse
0 siblings, 0 replies; 34+ messages in thread
From: Steven Whitehouse @ 2010-10-18 14:15 UTC (permalink / raw)
To: cluster-devel.redhat.com
Hi,
I know the merge window isn't open yet, but at this stage I'm going to
hold off on any larger patches until the following merge window so this
patch set isn't likely to change much, hence kicking it out a bit early
for review.
There are a few interesting points to note in this patch set:
o GFS2 is updated to use the new truncate sequence
o Support for fallocate is added
o Clean up of some unused/obsolete mount options
I'm currently working on a patch to allow the glock hash table
to use RCU. That is currently a work-in-progress and that will
hopefully be ready for the succeeding merge window.
Steve.
^ permalink raw reply [flat|nested] 34+ messages in thread
* [Cluster-devel] GFS2: Pre-pull patch posting
@ 2013-01-03 11:50 Steven Whitehouse
0 siblings, 0 replies; 34+ messages in thread
From: Steven Whitehouse @ 2013-01-03 11:50 UTC (permalink / raw)
To: cluster-devel.redhat.com
Hi,
Here are four small bug fixes for GFS2. There is no common theme here
really, just a few items that were fixed recently. The first fixes
lock name generation when the glock number is 0. The second fixes a
race allocating reservation structures and the final two fix a performance
issue by making small changes in the allocation code,
Steve.
^ permalink raw reply [flat|nested] 34+ messages in thread
* [Cluster-devel] GFS2: Pre-pull patch posting
@ 2013-04-05 9:57 Steven Whitehouse
0 siblings, 0 replies; 34+ messages in thread
From: Steven Whitehouse @ 2013-04-05 9:57 UTC (permalink / raw)
To: cluster-devel.redhat.com
Hi,
Here are a few GFS2 fixes which are pending. There are two patches
which fix up a couple of minor issues in the DLM interface code,
a missing error path in gfs2_rs_alloc(), two patches which fix problems
during "withdraw" and a fix for discards/FITRIM when using 4k sector
sized devices,
Steve.
^ permalink raw reply [flat|nested] 34+ messages in thread
* [Cluster-devel] GFS2: Pre-pull patch posting
@ 2015-02-10 10:36 Steven Whitehouse
0 siblings, 0 replies; 34+ messages in thread
From: Steven Whitehouse @ 2015-02-10 10:36 UTC (permalink / raw)
To: cluster-devel.redhat.com
This time we have mostly clean ups. There is a bug fix for a NULL dereference
relating to ACLs, and another which improves (but does not fix entirely) an
allocation fall-back code path. The other three patches are small clean ups.
Steve.
^ permalink raw reply [flat|nested] 34+ messages in thread
end of thread, other threads:[~2015-02-10 10:36 UTC | newest]
Thread overview: 34+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2009-03-18 12:23 [Cluster-devel] [GFS2] Pre-pull patch posting swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 01/18] GFS2: Fix remount argument parsing swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 02/18] GFS2: Bring back lvb-related stuff to lock_nolock to support quotas swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 03/18] GFS2: change gfs2_quota_scan into a shrinker swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 04/18] GFS2: Remove "double" locking in quota swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 05/18] GFS2: Merge lock_dlm module into GFS2 swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 06/18] GFS2: Remove unused field from glock swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 07/18] GFS2: Fix error path ref counting for root inode swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 08/18] GFS2: Fix deadlock on journal flush swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 09/18] GFS2: Support generation of discard requests swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 10/18] GFS2: Expose UUID via sysfs/uevent swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 11/18] GFS2: Add a "demote a glock" interface to sysfs swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 12/18] GFS2: Fix alignment issue and tidy gfs2_bitfit swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 13/18] GFS2: Support quota/noquota mount arguments swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 14/18] GFS2: fix sparse warnings: constant is so big it is swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 15/18] GFS2: fix sparse warning: Should it be static? swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 16/18] GFS2: Pagecache usage optimization on GFS2 swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 17/18] GFS2: Fix locking bug in failed shared to exclusive conversion swhiteho
2009-03-18 12:23 ` [Cluster-devel] [PATCH 18/18] GFS2: Clean up of glops.c swhiteho
-- strict thread matches above, loose matches on Subject: below --
2015-02-10 10:36 [Cluster-devel] GFS2: Pre-pull patch posting Steven Whitehouse
2013-04-05 9:57 Steven Whitehouse
2013-01-03 11:50 Steven Whitehouse
2010-10-18 14:15 Steven Whitehouse
2010-08-02 9:27 Steven Whitehouse
2010-05-17 12:40 Steven Whitehouse
2010-03-11 17:21 Steven Whitehouse
2010-03-01 15:08 Steven Whitehouse
2009-09-10 11:27 Steven Whitehouse
2009-06-10 8:30 Steven Whitehouse
2008-12-17 11:29 swhiteho
2008-09-26 12:00 Steven Whitehouse
2008-07-11 10:11 [Cluster-devel] [GFS2] " swhiteho
2008-04-17 8:37 swhiteho
2008-01-21 9:21 swhiteho
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).