Merge git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-2.6-nmw

* git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-2.6-nmw:
  GFS2: Fix recovery stuck bug (try #2)
  GFS2: Fix typo in stuffed file data copy handling
  Revert "GFS2: recovery stuck on transaction lock"
  GFS2: Make "try" lock not try quite so hard
  GFS2: remove dependency on __GFP_NOFAIL
  GFS2: Simplify gfs2_write_alloc_required
  GFS2: Wait for journal id on mount if not specified on mount command line
  GFS2: Use nobh_writepage

11 files changed, 169 insertions, 88 deletions

diff --git a/fs/gfs2/aops.c b/fs/gfs2/aops.c
index 9f8b52500d63..5e96cbd8a454 100644
--- a/fs/gfs2/aops.c
+++ b/fs/gfs2/aops.c
@@ -136,10 +136,7 @@ static int gfs2_writeback_writepage(struct page *page,
         if (ret <= 0)
                 return ret;
 
-        ret = mpage_writepage(page, gfs2_get_block_noalloc, wbc);
-        if (ret == -EAGAIN)
-                ret = block_write_full_page(page, gfs2_get_block_noalloc, wbc);
-        return ret;
+        return nobh_writepage(page, gfs2_get_block_noalloc, wbc);
 }
 
 /**
@@ -637,9 +634,7 @@ static int gfs2_write_begin(struct file *file, struct address_space *mapping,
                 }
         }
 
-        error = gfs2_write_alloc_required(ip, pos, len, &alloc_required);
-        if (error)
-                goto out_unlock;
+        alloc_required = gfs2_write_alloc_required(ip, pos, len);
 
         if (alloc_required || gfs2_is_jdata(ip))
                 gfs2_write_calc_reserv(ip, len, &data_blocks, &ind_blocks);
diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c
index 84da64b551b2..6f482809d1a3 100644
--- a/fs/gfs2/bmap.c
+++ b/fs/gfs2/bmap.c
@@ -1040,7 +1040,7 @@ static int trunc_start(struct gfs2_inode *ip, u64 size)
                 goto out;
 
         if (gfs2_is_stuffed(ip)) {
-                u64 dsize = size + sizeof(struct gfs2_inode);
+                u64 dsize = size + sizeof(struct gfs2_dinode);
                 ip->i_disksize = size;
                 ip->i_inode.i_mtime = ip->i_inode.i_ctime = CURRENT_TIME;
                 gfs2_trans_add_bh(ip->i_gl, dibh, 1);
@@ -1244,13 +1244,12 @@ int gfs2_file_dealloc(struct gfs2_inode *ip)
  * @ip: the file being written to
  * @offset: the offset to write to
  * @len: the number of bytes being written
- * @alloc_required: set to 1 if an alloc is required, 0 otherwise
  *
- * Returns: errno
+ * Returns: 1 if an alloc is required, 0 otherwise
  */
 
 int gfs2_write_alloc_required(struct gfs2_inode *ip, u64 offset,
-                              unsigned int len, int *alloc_required)
+                              unsigned int len)
 {
         struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
         struct buffer_head bh;
@@ -1258,26 +1257,23 @@ int gfs2_write_alloc_required(struct gfs2_inode *ip, u64 offset,
         u64 lblock, lblock_stop, size;
         u64 end_of_file;
 
-        *alloc_required = 0;
-
         if (!len)
                 return 0;
 
         if (gfs2_is_stuffed(ip)) {
                 if (offset + len >
                     sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode))
-                        *alloc_required = 1;
+                        return 1;
                 return 0;
         }
 
-        *alloc_required = 1;
         shift = sdp->sd_sb.sb_bsize_shift;
         BUG_ON(gfs2_is_dir(ip));
         end_of_file = (ip->i_disksize + sdp->sd_sb.sb_bsize - 1) >> shift;
         lblock = offset >> shift;
         lblock_stop = (offset + len + sdp->sd_sb.sb_bsize - 1) >> shift;
         if (lblock_stop > end_of_file)
-                return 0;
+                return 1;
 
         size = (lblock_stop - lblock) << shift;
         do {
@@ -1285,12 +1281,11 @@ int gfs2_write_alloc_required(struct gfs2_inode *ip, u64 offset,
                 bh.b_size = size;
                 gfs2_block_map(&ip->i_inode, lblock, &bh, 0);
                 if (!buffer_mapped(&bh))
-                        return 0;
+                        return 1;
                 size -= bh.b_size;
                 lblock += (bh.b_size >> ip->i_inode.i_blkbits);
         } while(size > 0);
 
-        *alloc_required = 0;
         return 0;
 }
 
diff --git a/fs/gfs2/bmap.h b/fs/gfs2/bmap.h
index c983177e05ac..a20a5213135a 100644
--- a/fs/gfs2/bmap.h
+++ b/fs/gfs2/bmap.h
@@ -52,6 +52,6 @@ int gfs2_truncatei(struct gfs2_inode *ip, u64 size);
 int gfs2_truncatei_resume(struct gfs2_inode *ip);
 int gfs2_file_dealloc(struct gfs2_inode *ip);
 int gfs2_write_alloc_required(struct gfs2_inode *ip, u64 offset,
-                              unsigned int len, int *alloc_required);
+                              unsigned int len);
 
 #endif /* __BMAP_DOT_H__ */
diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c
index 6b48d7c268b2..b9dd88a78dd4 100644
--- a/fs/gfs2/dir.c
+++ b/fs/gfs2/dir.c
@@ -955,7 +955,12 @@ static int dir_split_leaf(struct inode *inode, const struct qstr *name)
         /* Change the pointers.
            Don't bother distinguishing stuffed from non-stuffed.
            This code is complicated enough already. */
-        lp = kmalloc(half_len * sizeof(__be64), GFP_NOFS | __GFP_NOFAIL);
+        lp = kmalloc(half_len * sizeof(__be64), GFP_NOFS);
+        if (!lp) {
+                error = -ENOMEM;
+                goto fail_brelse;
+        }
+
         /*  Change the pointers  */
         for (x = 0; x < half_len; x++)
                 lp[x] = cpu_to_be64(bn);
@@ -1063,7 +1068,9 @@ static int dir_double_exhash(struct gfs2_inode *dip)
 
         /*  Allocate both the "from" and "to" buffers in one big chunk  */
 
-        buf = kcalloc(3, sdp->sd_hash_bsize, GFP_NOFS | __GFP_NOFAIL);
+        buf = kcalloc(3, sdp->sd_hash_bsize, GFP_NOFS);
+        if (!buf)
+                return -ENOMEM;
 
         for (block = dip->i_disksize >> sdp->sd_hash_bsize_shift; block--;) {
                 error = gfs2_dir_read_data(dip, (char *)buf,
diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c
index ed9a94f0ef15..4edd662c8232 100644
--- a/fs/gfs2/file.c
+++ b/fs/gfs2/file.c
@@ -351,7 +351,6 @@ static int gfs2_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
         unsigned long last_index;
         u64 pos = page->index << PAGE_CACHE_SHIFT;
         unsigned int data_blocks, ind_blocks, rblocks;
-        int alloc_required = 0;
         struct gfs2_holder gh;
         struct gfs2_alloc *al;
         int ret;
@@ -364,8 +363,7 @@ static int gfs2_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
         set_bit(GLF_DIRTY, &ip->i_gl->gl_flags);
         set_bit(GIF_SW_PAGED, &ip->i_flags);
 
-        ret = gfs2_write_alloc_required(ip, pos, PAGE_CACHE_SIZE, &alloc_required);
-        if (ret || !alloc_required)
+        if (!gfs2_write_alloc_required(ip, pos, PAGE_CACHE_SIZE))
                 goto out_unlock;
         ret = -ENOMEM;
         al = gfs2_alloc_get(ip);
diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c
index 0898f3ec8212..9adf8f924e08 100644
--- a/fs/gfs2/glock.c
+++ b/fs/gfs2/glock.c
@@ -328,6 +328,30 @@ static void gfs2_holder_wake(struct gfs2_holder *gh)
 }
 
 /**
+ * do_error - Something unexpected has happened during a lock request
+ *
+ */
+
+static inline void do_error(struct gfs2_glock *gl, const int ret)
+{
+        struct gfs2_holder *gh, *tmp;
+
+        list_for_each_entry_safe(gh, tmp, &gl->gl_holders, gh_list) {
+                if (test_bit(HIF_HOLDER, &gh->gh_iflags))
+                        continue;
+                if (ret & LM_OUT_ERROR)
+                        gh->gh_error = -EIO;
+                else if (gh->gh_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB))
+                        gh->gh_error = GLR_TRYFAILED;
+                else
+                        continue;
+                list_del_init(&gh->gh_list);
+                trace_gfs2_glock_queue(gh, 0);
+                gfs2_holder_wake(gh);
+        }
+}
+
+/**
  * do_promote - promote as many requests as possible on the current queue
  * @gl: The glock
  * 
@@ -375,36 +399,13 @@ restart:
                 }
                 if (gh->gh_list.prev == &gl->gl_holders)
                         return 1;
+                do_error(gl, 0);
                 break;
         }
         return 0;
 }
 
 /**
- * do_error - Something unexpected has happened during a lock request
- *
- */
-
-static inline void do_error(struct gfs2_glock *gl, const int ret)
-{
-        struct gfs2_holder *gh, *tmp;
-
-        list_for_each_entry_safe(gh, tmp, &gl->gl_holders, gh_list) {
-                if (test_bit(HIF_HOLDER, &gh->gh_iflags))
-                        continue;
-                if (ret & LM_OUT_ERROR)
-                        gh->gh_error = -EIO;
-                else if (gh->gh_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB))
-                        gh->gh_error = GLR_TRYFAILED;
-                else
-                        continue;
-                list_del_init(&gh->gh_list);
-                trace_gfs2_glock_queue(gh, 0);
-                gfs2_holder_wake(gh);
-        }
-}
-
-/**
  * find_first_waiter - find the first gh that's waiting for the glock
  * @gl: the glock
  */
@@ -706,18 +707,8 @@ static void glock_work_func(struct work_struct *work)
 {
         unsigned long delay = 0;
         struct gfs2_glock *gl = container_of(work, struct gfs2_glock, gl_work.work);
-        struct gfs2_holder *gh;
         int drop_ref = 0;
 
-        if (unlikely(test_bit(GLF_FROZEN, &gl->gl_flags))) {
-                spin_lock(&gl->gl_spin);
-                gh = find_first_waiter(gl);
-                if (gh && (gh->gh_flags & LM_FLAG_NOEXP) &&
-                    test_and_clear_bit(GLF_FROZEN, &gl->gl_flags))
-                        set_bit(GLF_REPLY_PENDING, &gl->gl_flags);
-                spin_unlock(&gl->gl_spin);
-        }
-
         if (test_and_clear_bit(GLF_REPLY_PENDING, &gl->gl_flags)) {
                 finish_xmote(gl, gl->gl_reply);
                 drop_ref = 1;
@@ -1072,6 +1063,9 @@ int gfs2_glock_nq(struct gfs2_holder *gh)
 
         spin_lock(&gl->gl_spin);
         add_to_queue(gh);
+        if ((LM_FLAG_NOEXP & gh->gh_flags) &&
+            test_and_clear_bit(GLF_FROZEN, &gl->gl_flags))
+                set_bit(GLF_REPLY_PENDING, &gl->gl_flags);
         run_queue(gl, 1);
         spin_unlock(&gl->gl_spin);
 
@@ -1329,6 +1323,36 @@ void gfs2_glock_cb(struct gfs2_glock *gl, unsigned int state)
 }
 
 /**
+ * gfs2_should_freeze - Figure out if glock should be frozen
+ * @gl: The glock in question
+ *
+ * Glocks are not frozen if (a) the result of the dlm operation is
+ * an error, (b) the locking operation was an unlock operation or
+ * (c) if there is a "noexp" flagged request anywhere in the queue
+ *
+ * Returns: 1 if freezing should occur, 0 otherwise
+ */
+
+static int gfs2_should_freeze(const struct gfs2_glock *gl)
+{
+        const struct gfs2_holder *gh;
+
+        if (gl->gl_reply & ~LM_OUT_ST_MASK)
+                return 0;
+        if (gl->gl_target == LM_ST_UNLOCKED)
+                return 0;
+
+        list_for_each_entry(gh, &gl->gl_holders, gh_list) {
+                if (test_bit(HIF_HOLDER, &gh->gh_iflags))
+                        continue;
+                if (LM_FLAG_NOEXP & gh->gh_flags)
+                        return 0;
+        }
+
+        return 1;
+}
+
+/**
  * gfs2_glock_complete - Callback used by locking
  * @gl: Pointer to the glock
  * @ret: The return value from the dlm
@@ -1338,18 +1362,17 @@ 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;
+
         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))
+                if (gfs2_should_freeze(gl)) {
                         set_bit(GLF_FROZEN, &gl->gl_flags);
-                spin_unlock(&gl->gl_spin);
-                if (test_bit(GLF_FROZEN, &gl->gl_flags))
+                        spin_unlock(&gl->gl_spin);
                         return;
+                }
+                spin_unlock(&gl->gl_spin);
         }
         set_bit(GLF_REPLY_PENDING, &gl->gl_flags);
         gfs2_glock_hold(gl);
diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h
index b5d7363b22da..8fcbce48a128 100644
--- a/fs/gfs2/incore.h
+++ b/fs/gfs2/incore.h
@@ -460,6 +460,7 @@ enum {
         SDF_NOBARRIERS          = 3,
         SDF_NORECOVERY          = 4,
         SDF_DEMOTE              = 5,
+        SDF_NOJOURNALID         = 6,
 };
 
 #define GFS2_FSNAME_LEN         256
diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index 3593b3a7290e..45a4a36195d8 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -76,7 +76,7 @@ static struct gfs2_sbd *init_sbd(struct super_block *sb)
 
         sb->s_fs_info = sdp;
         sdp->sd_vfs = sb;
-
+        set_bit(SDF_NOJOURNALID, &sdp->sd_flags);
         gfs2_tune_init(&sdp->sd_tune);
 
         init_waitqueue_head(&sdp->sd_glock_wait);
@@ -1050,7 +1050,8 @@ static int gfs2_lm_mount(struct gfs2_sbd *sdp, int silent)
                         ret = match_int(&tmp[0], &option);
                         if (ret || option < 0) 
                                 goto hostdata_error;
-                        ls->ls_jid = option;
+                        if (test_and_clear_bit(SDF_NOJOURNALID, &sdp->sd_flags))
+                                ls->ls_jid = option;
                         break;
                 case Opt_id:
                         /* Obsolete, but left for backward compat purposes */
@@ -1102,6 +1103,24 @@ void gfs2_lm_unmount(struct gfs2_sbd *sdp)
                 lm->lm_unmount(sdp);
 }
 
+static int gfs2_journalid_wait(void *word)
+{
+        if (signal_pending(current))
+                return -EINTR;
+        schedule();
+        return 0;
+}
+
+static int wait_on_journal(struct gfs2_sbd *sdp)
+{
+        if (sdp->sd_args.ar_spectator)
+                return 0;
+        if (sdp->sd_lockstruct.ls_ops->lm_mount == NULL)
+                return 0;
+
+        return wait_on_bit(&sdp->sd_flags, SDF_NOJOURNALID, gfs2_journalid_wait, TASK_INTERRUPTIBLE);
+}
+
 void gfs2_online_uevent(struct gfs2_sbd *sdp)
 {
         struct super_block *sb = sdp->sd_vfs;
@@ -1194,6 +1213,10 @@ static int fill_super(struct super_block *sb, struct gfs2_args *args, int silent
         if (error)
                 goto fail_locking;
 
+        error = wait_on_journal(sdp);
+        if (error)
+                goto fail_sb;
+
         error = init_inodes(sdp, DO);
         if (error)
                 goto fail_sb;
diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c
index 8f02d3db8f42..8bb643cb2658 100644
--- a/fs/gfs2/quota.c
+++ b/fs/gfs2/quota.c
@@ -787,15 +787,9 @@ static int do_sync(unsigned int num_qd, struct gfs2_quota_data **qda)
                 goto out;
 
         for (x = 0; x < num_qd; x++) {
-                int alloc_required;
-
                 offset = qd2offset(qda[x]);
-                error = gfs2_write_alloc_required(ip, offset,
-                                                  sizeof(struct gfs2_quota),
-                                                  &alloc_required);
-                if (error)
-                        goto out_gunlock;
-                if (alloc_required)
+                if (gfs2_write_alloc_required(ip, offset,
+                                              sizeof(struct gfs2_quota)))
                         nalloc++;
         }
 
@@ -1584,10 +1578,7 @@ static int gfs2_set_dqblk(struct super_block *sb, int type, qid_t id,
                 goto out_i;
 
         offset = qd2offset(qd);
-        error = gfs2_write_alloc_required(ip, offset, sizeof(struct gfs2_quota),
-                                          &alloc_required);
-        if (error)
-                goto out_i;
+        alloc_required = gfs2_write_alloc_required(ip, offset, sizeof(struct gfs2_quota));
         if (alloc_required) {
                 al = gfs2_alloc_get(ip);
                 if (al == NULL)
diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c
index 4d1aad38f1b1..4140811a921c 100644
--- a/fs/gfs2/super.c
+++ b/fs/gfs2/super.c
@@ -342,8 +342,6 @@ int gfs2_jdesc_check(struct gfs2_jdesc *jd)
 {
         struct gfs2_inode *ip = GFS2_I(jd->jd_inode);
         struct gfs2_sbd *sdp = GFS2_SB(jd->jd_inode);
-        int ar;
-        int error;
 
         if (ip->i_disksize < (8 << 20) || ip->i_disksize > (1 << 30) ||
             (ip->i_disksize & (sdp->sd_sb.sb_bsize - 1))) {
@@ -352,13 +350,12 @@ int gfs2_jdesc_check(struct gfs2_jdesc *jd)
         }
         jd->jd_blocks = ip->i_disksize >> sdp->sd_sb.sb_bsize_shift;
 
-        error = gfs2_write_alloc_required(ip, 0, ip->i_disksize, &ar);
-        if (!error && ar) {
+        if (gfs2_write_alloc_required(ip, 0, ip->i_disksize)) {
                 gfs2_consist_inode(ip);
-                error = -EIO;
+                return -EIO;
         }
 
-        return error;
+        return 0;
 }
 
 /**
diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c
index 37f5393e68e6..d019d0d55e00 100644
--- a/fs/gfs2/sys.c
+++ b/fs/gfs2/sys.c
@@ -325,6 +325,30 @@ static ssize_t lkfirst_show(struct gfs2_sbd *sdp, char *buf)
         return sprintf(buf, "%d\n", ls->ls_first);
 }
 
+static ssize_t lkfirst_store(struct gfs2_sbd *sdp, const char *buf, size_t len)
+{
+        unsigned first;
+        int rv;
+
+        rv = sscanf(buf, "%u", &first);
+        if (rv != 1 || first > 1)
+                return -EINVAL;
+        spin_lock(&sdp->sd_jindex_spin);
+        rv = -EBUSY;
+        if (test_bit(SDF_NOJOURNALID, &sdp->sd_flags) == 0)
+                goto out;
+        rv = -EINVAL;
+        if (sdp->sd_args.ar_spectator)
+                goto out;
+        if (sdp->sd_lockstruct.ls_ops->lm_mount == NULL)
+                goto out;
+        sdp->sd_lockstruct.ls_first = first;
+        rv = 0;
+out:
+        spin_unlock(&sdp->sd_jindex_spin);
+        return rv ? rv : len;
+}
+
 static ssize_t first_done_show(struct gfs2_sbd *sdp, char *buf)
 {
         struct lm_lockstruct *ls = &sdp->sd_lockstruct;
@@ -377,14 +401,41 @@ static ssize_t jid_show(struct gfs2_sbd *sdp, char *buf)
         return sprintf(buf, "%u\n", sdp->sd_lockstruct.ls_jid);
 }
 
+static ssize_t jid_store(struct gfs2_sbd *sdp, const char *buf, size_t len)
+{
+        unsigned jid;
+        int rv;
+
+        rv = sscanf(buf, "%u", &jid);
+        if (rv != 1)
+                return -EINVAL;
+
+        spin_lock(&sdp->sd_jindex_spin);
+        rv = -EINVAL;
+        if (sdp->sd_args.ar_spectator)
+                goto out;
+        if (sdp->sd_lockstruct.ls_ops->lm_mount == NULL)
+                goto out;
+        rv = -EBUSY;
+        if (test_and_clear_bit(SDF_NOJOURNALID, &sdp->sd_flags) == 0)
+                goto out;
+        sdp->sd_lockstruct.ls_jid = jid;
+        smp_mb__after_clear_bit();
+        wake_up_bit(&sdp->sd_flags, SDF_NOJOURNALID);
+        rv = 0;
+out:
+        spin_unlock(&sdp->sd_jindex_spin);
+        return rv ? rv : len;
+}
+
 #define GDLM_ATTR(_name,_mode,_show,_store) \
 static struct gfs2_attr gdlm_attr_##_name = __ATTR(_name,_mode,_show,_store)
 
 GDLM_ATTR(proto_name,           0444, proto_name_show,          NULL);
 GDLM_ATTR(block,                0644, block_show,               block_store);
 GDLM_ATTR(withdraw,             0644, withdraw_show,            withdraw_store);
-GDLM_ATTR(jid,                  0444, jid_show,                 NULL);
-GDLM_ATTR(first,                0444, lkfirst_show,             NULL);
+GDLM_ATTR(jid,                  0644, jid_show,                 jid_store);
+GDLM_ATTR(first,                0644, lkfirst_show,             lkfirst_store);
 GDLM_ATTR(first_done,           0444, first_done_show,          NULL);
 GDLM_ATTR(recover,              0600, NULL,                     recover_store);
 GDLM_ATTR(recover_done,         0444, recover_done_show,        NULL);
@@ -564,7 +615,7 @@ static int gfs2_uevent(struct kset *kset, struct kobject *kobj,
 
         add_uevent_var(env, "LOCKTABLE=%s", sdp->sd_table_name);
         add_uevent_var(env, "LOCKPROTO=%s", sdp->sd_proto_name);
-        if (!sdp->sd_args.ar_spectator)
+        if (!test_bit(SDF_NOJOURNALID, &sdp->sd_flags))
                 add_uevent_var(env, "JOURNALID=%u", sdp->sd_lockstruct.ls_jid);
         if (gfs2_uuid_valid(uuid))
                 add_uevent_var(env, "UUID=%pUB", uuid);