From 87124e581bfeaa5864662a435b6ee2a19e91b905 Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Mon, 23 Jul 2007 09:54:36 +0100 Subject: [GFS2] Fix two races relating to glock callbacks One of the races relates to referencing a variable while not holding its protecting spinlock. The patch simply moves the test inside the spin lock. The other races occurs when a demote to unlocked request occurs during the time a demote to shared request is already running. This of course only happens in the case that the lock was in the exclusive mode to start with. The patch adds a check to see if another demote request has occurred in the mean time and if it has, then it performs a second demote. Signed-off-by: Steven Whitehouse --- fs/gfs2/glock.c | 24 ++++++++++++++++++------ 1 file changed, 18 insertions(+), 6 deletions(-) (limited to 'fs/gfs2/glock.c') diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 3f0974e..6a3eeba 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -545,12 +545,14 @@ static int rq_demote(struct gfs2_glock *gl) return 0; } set_bit(GLF_LOCK, &gl->gl_flags); - spin_unlock(&gl->gl_spin); if (gl->gl_demote_state == LM_ST_UNLOCKED || - gl->gl_state != LM_ST_EXCLUSIVE) + gl->gl_state != LM_ST_EXCLUSIVE) { + spin_unlock(&gl->gl_spin); gfs2_glock_drop_th(gl); - else + } else { + spin_unlock(&gl->gl_spin); gfs2_glock_xmote_th(gl, NULL); + } spin_lock(&gl->gl_spin); return 0; @@ -760,10 +762,20 @@ static void xmote_bh(struct gfs2_glock *gl, unsigned int ret) if (!gh) { gl->gl_stamp = jiffies; - if (ret & LM_OUT_CANCELED) + if (ret & LM_OUT_CANCELED) { op_done = 0; - else + } else { + spin_lock(&gl->gl_spin); + if (gl->gl_state != gl->gl_demote_state) { + gl->gl_req_bh = NULL; + spin_unlock(&gl->gl_spin); + gfs2_glock_drop_th(gl); + gfs2_glock_put(gl); + return; + } gfs2_demote_wake(gl); + spin_unlock(&gl->gl_spin); + } } else { spin_lock(&gl->gl_spin); list_del_init(&gh->gh_list); @@ -817,7 +829,7 @@ out: * */ -void gfs2_glock_xmote_th(struct gfs2_glock *gl, struct gfs2_holder *gh) +static void gfs2_glock_xmote_th(struct gfs2_glock *gl, struct gfs2_holder *gh) { struct gfs2_sbd *sdp = gl->gl_sbd; int flags = gh ? gh->gh_flags : 0; -- cgit v1.1 From 26caee5bc643b318fa2e9bd4f66dace1755ec413 Mon Sep 17 00:00:00 2001 From: Josef Whiter Date: Mon, 23 Jul 2007 10:02:40 +0100 Subject: [GFS2] Fix calculation of demote state If a glock is in the exclusive state and a request for demote to deferred has been received, then further requests for demote to shared are being ignored. This patch fixes that by ensuring that we demote to unlocked in that case. Signed-off-by: Josef Whiter Signed-off-by: Steven Whitehouse --- fs/gfs2/glock.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'fs/gfs2/glock.c') diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 6a3eeba..6b6ae45 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -697,8 +697,9 @@ static void handle_callback(struct gfs2_glock *gl, unsigned int state, int remot } return; } - } else if (gl->gl_demote_state != LM_ST_UNLOCKED) { - gl->gl_demote_state = state; + } else if (gl->gl_demote_state != LM_ST_UNLOCKED && + gl->gl_demote_state != state) { + gl->gl_demote_state = LM_ST_UNLOCKED; } spin_unlock(&gl->gl_spin); } -- cgit v1.1 From aa0481e58a9a97a97035725a712920b5fe32f348 Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Sat, 21 Jul 2007 17:03:22 +0200 Subject: [GFS2] Clean up duplicate includes in fs/gfs2/ This patch cleans up duplicate includes in fs/gfs2/ Signed-off-by: Jesper Juhl Signed-off-by: Steven Whitehouse --- fs/gfs2/glock.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'fs/gfs2/glock.c') diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 6b6ae45..d403fd7 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -25,8 +25,6 @@ #include #include #include -#include -#include #include "gfs2.h" #include "incore.h" -- cgit v1.1 From 7b08fc620109c2f66575e9ae884f45c37933ea18 Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Tue, 24 Jul 2007 13:53:36 +0100 Subject: [GFS2] Fix an oops in glock dumping This fixes an oops which was occurring during glock dumping due to the seq file code not taking a reference to the glock. Also this fixes a memory leak which occurred in certain cases, in turn preventing the filesystem from unmounting. Signed-off-by: Steven Whitehouse --- fs/gfs2/glock.c | 73 +++++++++++++++++++++++++++------------------------------ 1 file changed, 35 insertions(+), 38 deletions(-) (limited to 'fs/gfs2/glock.c') diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index d403fd7..e4bc8ae 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -46,7 +46,6 @@ struct glock_iter { int hash; /* hash bucket index */ struct gfs2_sbd *sdp; /* incore superblock */ struct gfs2_glock *gl; /* current glock struct */ - struct hlist_head *hb_list; /* current hash bucket ptr */ struct seq_file *seq; /* sequence file for debugfs */ char string[512]; /* scratch space */ }; @@ -1990,47 +1989,38 @@ int __init gfs2_glock_init(void) static int gfs2_glock_iter_next(struct glock_iter *gi) { + struct gfs2_glock *gl; + read_lock(gl_lock_addr(gi->hash)); - while (1) { - if (!gi->hb_list) { /* If we don't have a hash bucket yet */ - gi->hb_list = &gl_hash_table[gi->hash].hb_list; - if (hlist_empty(gi->hb_list)) { - read_unlock(gl_lock_addr(gi->hash)); - gi->hash++; - read_lock(gl_lock_addr(gi->hash)); - gi->hb_list = NULL; - if (gi->hash >= GFS2_GL_HASH_SIZE) { - read_unlock(gl_lock_addr(gi->hash)); - return 1; - } - else - continue; - } - if (!hlist_empty(gi->hb_list)) { - gi->gl = list_entry(gi->hb_list->first, - struct gfs2_glock, - gl_list); - } - } else { - if (gi->gl->gl_list.next == NULL) { - read_unlock(gl_lock_addr(gi->hash)); - gi->hash++; - read_lock(gl_lock_addr(gi->hash)); - gi->hb_list = NULL; - continue; - } - gi->gl = list_entry(gi->gl->gl_list.next, - struct gfs2_glock, gl_list); - } + gl = gi->gl; + if (gl) { + gi->gl = hlist_entry(gl->gl_list.next, struct gfs2_glock, + gl_list); if (gi->gl) - break; + gfs2_glock_hold(gi->gl); } read_unlock(gl_lock_addr(gi->hash)); + if (gl) + gfs2_glock_put(gl); + + while(gi->gl == NULL) { + gi->hash++; + if (gi->hash >= GFS2_GL_HASH_SIZE) + return 1; + read_lock(gl_lock_addr(gi->hash)); + gi->gl = hlist_entry(gl_hash_table[gi->hash].hb_list.first, + struct gfs2_glock, gl_list); + if (gi->gl) + gfs2_glock_hold(gi->gl); + read_unlock(gl_lock_addr(gi->hash)); + } return 0; } static void gfs2_glock_iter_free(struct glock_iter *gi) { + if (gi->gl) + gfs2_glock_put(gi->gl); kfree(gi); } @@ -2044,12 +2034,17 @@ static struct glock_iter *gfs2_glock_iter_init(struct gfs2_sbd *sdp) gi->sdp = sdp; gi->hash = 0; - gi->gl = NULL; - gi->hb_list = NULL; gi->seq = NULL; memset(gi->string, 0, sizeof(gi->string)); - if (gfs2_glock_iter_next(gi)) { + read_lock(gl_lock_addr(gi->hash)); + gi->gl = hlist_entry(gl_hash_table[gi->hash].hb_list.first, + struct gfs2_glock, gl_list); + if (gi->gl) + gfs2_glock_hold(gi->gl); + read_unlock(gl_lock_addr(gi->hash)); + + if (!gi->gl && gfs2_glock_iter_next(gi)) { gfs2_glock_iter_free(gi); return NULL; } @@ -2066,7 +2061,7 @@ static void *gfs2_glock_seq_start(struct seq_file *file, loff_t *pos) if (!gi) return NULL; - while (n--) { + while(n--) { if (gfs2_glock_iter_next(gi)) { gfs2_glock_iter_free(gi); return NULL; @@ -2093,7 +2088,9 @@ static void *gfs2_glock_seq_next(struct seq_file *file, void *iter_ptr, static void gfs2_glock_seq_stop(struct seq_file *file, void *iter_ptr) { - /* nothing for now */ + struct glock_iter *gi = iter_ptr; + if (gi) + gfs2_glock_iter_free(gi); } static int gfs2_glock_seq_show(struct seq_file *file, void *iter_ptr) -- cgit v1.1 From 4ef290025ccde7c52ba219cf733a4295acd5401f Mon Sep 17 00:00:00 2001 From: Denis Cheng Date: Tue, 31 Jul 2007 18:31:11 +0800 Subject: [GFS2] mark struct *_operations const these struct *_operations are all method tables, thus should be const. Signed-off-by: Denis Cheng Signed-off-by: Steven Whitehouse --- fs/gfs2/glock.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/gfs2/glock.c') diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index e4bc8ae..0054b7d 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -2103,7 +2103,7 @@ static int gfs2_glock_seq_show(struct seq_file *file, void *iter_ptr) return 0; } -static struct seq_operations gfs2_glock_seq_ops = { +static const struct seq_operations gfs2_glock_seq_ops = { .start = gfs2_glock_seq_start, .next = gfs2_glock_seq_next, .stop = gfs2_glock_seq_stop, -- cgit v1.1 From 8fbbfd214c853102b614f4705c1904ed14f5a808 Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Wed, 1 Aug 2007 13:57:10 +0100 Subject: [GFS2] Reduce number of gfs2_scand processes to one We only need a single gfs2_scand process rather than the one per filesystem which we had previously. As a result the parameter determining the frequency of gfs2_scand runs becomes a module parameter rather than a mount parameter as it was before. Signed-off-by: Steven Whitehouse --- fs/gfs2/glock.c | 63 +++++++++++++++++++++++++++++++++++++++++++-------------- 1 file changed, 48 insertions(+), 15 deletions(-) (limited to 'fs/gfs2/glock.c') diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 0054b7d..559937c7 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -25,6 +25,8 @@ #include #include #include +#include +#include #include "gfs2.h" #include "incore.h" @@ -58,6 +60,8 @@ static void gfs2_glock_xmote_th(struct gfs2_glock *gl, struct gfs2_holder *gh); static void gfs2_glock_drop_th(struct gfs2_glock *gl); static DECLARE_RWSEM(gfs2_umount_flush_sem); static struct dentry *gfs2_root; +static struct task_struct *scand_process; +static unsigned int scand_secs = 5; #define GFS2_GL_HASH_SHIFT 15 #define GFS2_GL_HASH_SIZE (1 << GFS2_GL_HASH_SHIFT) @@ -1627,7 +1631,7 @@ static int examine_bucket(glock_examiner examiner, struct gfs2_sbd *sdp, goto out; gl = list_entry(head->first, struct gfs2_glock, gl_list); while(1) { - if (gl->gl_sbd == sdp) { + if (!sdp || gl->gl_sbd == sdp) { gfs2_glock_hold(gl); read_unlock(gl_lock_addr(hash)); if (prev) @@ -1645,6 +1649,7 @@ out: read_unlock(gl_lock_addr(hash)); if (prev) gfs2_glock_put(prev); + cond_resched(); return has_entries; } @@ -1673,20 +1678,6 @@ out_schedule: } /** - * gfs2_scand_internal - Look for glocks and inodes to toss from memory - * @sdp: the filesystem - * - */ - -void gfs2_scand_internal(struct gfs2_sbd *sdp) -{ - unsigned int x; - - for (x = 0; x < GFS2_GL_HASH_SIZE; x++) - examine_bucket(scan_glock, sdp, x); -} - -/** * clear_glock - look at a glock and see if we can free it from glock cache * @gl: the glock to look at * @@ -1973,6 +1964,35 @@ static int gfs2_dump_lockstate(struct gfs2_sbd *sdp) return error; } +/** + * gfs2_scand - Look for cached glocks and inodes to toss from memory + * @sdp: Pointer to GFS2 superblock + * + * One of these daemons runs, finding candidates to add to sd_reclaim_list. + * See gfs2_glockd() + */ + +static int gfs2_scand(void *data) +{ + unsigned x; + unsigned delay; + + while (!kthread_should_stop()) { + for (x = 0; x < GFS2_GL_HASH_SIZE; x++) + examine_bucket(scan_glock, NULL, x); + if (freezing(current)) + refrigerator(); + delay = scand_secs; + if (delay < 1) + delay = 1; + schedule_timeout_interruptible(delay * HZ); + } + + return 0; +} + + + int __init gfs2_glock_init(void) { unsigned i; @@ -1984,9 +2004,22 @@ int __init gfs2_glock_init(void) rwlock_init(&gl_hash_locks[i]); } #endif + + scand_process = kthread_run(gfs2_scand, NULL, "gfs2_scand"); + if (IS_ERR(scand_process)) + return PTR_ERR(scand_process); + return 0; } +void gfs2_glock_exit(void) +{ + kthread_stop(scand_process); +} + +module_param(scand_secs, uint, S_IRUGO|S_IWUSR); +MODULE_PARM_DESC(scand_secs, "The number of seconds between scand runs"); + static int gfs2_glock_iter_next(struct glock_iter *gi) { struct gfs2_glock *gl; -- cgit v1.1 From a947e0335699a1d387c3826e5b8eff9e0afe505e Mon Sep 17 00:00:00 2001 From: Abhijith Das Date: Tue, 21 Aug 2007 09:57:29 -0500 Subject: [GFS2] Wendy's dump lockname in hex & fix glock dump With this patch, gfs2 glockdump through the debugfs filesystem will only dump glocks for the specified filesystem instead of all glocks. Also, to aid debugging, the glock number is dumped in hex instead of decimal. Signed-off-by: Steven Whitehouse Signed-off-by: S. Wendy Cheng Signed-off-by: Abhijith Das --- fs/gfs2/glock.c | 26 +++++++++++++------------- 1 file changed, 13 insertions(+), 13 deletions(-) (limited to 'fs/gfs2/glock.c') diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 559937c7..3d94918 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -1844,7 +1844,7 @@ static int dump_glock(struct glock_iter *gi, struct gfs2_glock *gl) spin_lock(&gl->gl_spin); - print_dbg(gi, "Glock 0x%p (%u, %llu)\n", gl, gl->gl_name.ln_type, + print_dbg(gi, "Glock 0x%p (%u, 0x%llx)\n", gl, gl->gl_name.ln_type, (unsigned long long)gl->gl_name.ln_number); print_dbg(gi, " gl_flags ="); for (x = 0; x < 32; x++) { @@ -2024,20 +2024,21 @@ static int gfs2_glock_iter_next(struct glock_iter *gi) { struct gfs2_glock *gl; +restart: read_lock(gl_lock_addr(gi->hash)); gl = gi->gl; if (gl) { - gi->gl = hlist_entry(gl->gl_list.next, struct gfs2_glock, - gl_list); + gi->gl = hlist_entry(gl->gl_list.next, + struct gfs2_glock, gl_list); if (gi->gl) gfs2_glock_hold(gi->gl); } read_unlock(gl_lock_addr(gi->hash)); if (gl) gfs2_glock_put(gl); - - while(gi->gl == NULL) { + if (gl && gi->gl == NULL) gi->hash++; + while(gi->gl == NULL) { if (gi->hash >= GFS2_GL_HASH_SIZE) return 1; read_lock(gl_lock_addr(gi->hash)); @@ -2046,7 +2047,12 @@ static int gfs2_glock_iter_next(struct glock_iter *gi) if (gi->gl) gfs2_glock_hold(gi->gl); read_unlock(gl_lock_addr(gi->hash)); + gi->hash++; } + + if (gi->sdp != gi->gl->gl_sbd) + goto restart; + return 0; } @@ -2068,16 +2074,10 @@ static struct glock_iter *gfs2_glock_iter_init(struct gfs2_sbd *sdp) gi->sdp = sdp; gi->hash = 0; gi->seq = NULL; + gi->gl = NULL; memset(gi->string, 0, sizeof(gi->string)); - read_lock(gl_lock_addr(gi->hash)); - gi->gl = hlist_entry(gl_hash_table[gi->hash].hb_list.first, - struct gfs2_glock, gl_list); - if (gi->gl) - gfs2_glock_hold(gi->gl); - read_unlock(gl_lock_addr(gi->hash)); - - if (!gi->gl && gfs2_glock_iter_next(gi)) { + if (gfs2_glock_iter_next(gi)) { gfs2_glock_iter_free(gi); return NULL; } -- cgit v1.1 From c4f68a130fc1795e4a75ec5bdaf9e85d86c22419 Mon Sep 17 00:00:00 2001 From: Benjamin Marzinski Date: Thu, 23 Aug 2007 13:19:05 -0500 Subject: [GFS2] delay glock demote for a minimum hold time When a lot of IO, with some distributed mmap IO, is run on a GFS2 filesystem in a cluster, it will deadlock. The reason is that do_no_page() will repeatedly call gfs2_sharewrite_nopage(), because each node keeps giving up the glock too early, and is forced to call unmap_mapping_range(). This bumps the mapping->truncate_count sequence count, forcing do_no_page() to retry. This patch institutes a minimum glock hold time a tenth a second. This insures that even in heavy contention cases, the node has enough time to get some useful work done before it gives up the glock. A second issue is that when gfs2_glock_dq() is called from within a page fault to demote a lock, and the associated page needs to be written out, it will try to acqire a lock on it, but it has already been locked at a higher level. This patch puts makes gfs2_glock_dq() use the work queue as well, to avoid this issue. This is the same patch as Steve Whitehouse originally proposed to fix this issue, execpt that gfs2_glock_dq() now grabs a reference to the glock before it queues up the work on it. Signed-off-by: Benjamin E. Marzinski Signed-off-by: Steven Whitehouse --- fs/gfs2/glock.c | 75 +++++++++++++++++++++++++++++++++++++++++++++------------ 1 file changed, 59 insertions(+), 16 deletions(-) (limited to 'fs/gfs2/glock.c') diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 3d94918..931368a 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -27,6 +27,8 @@ #include #include #include +#include +#include #include "gfs2.h" #include "incore.h" @@ -58,10 +60,13 @@ static int gfs2_dump_lockstate(struct gfs2_sbd *sdp); static int dump_glock(struct glock_iter *gi, struct gfs2_glock *gl); static void gfs2_glock_xmote_th(struct gfs2_glock *gl, struct gfs2_holder *gh); static void gfs2_glock_drop_th(struct gfs2_glock *gl); +static void run_queue(struct gfs2_glock *gl); + static DECLARE_RWSEM(gfs2_umount_flush_sem); static struct dentry *gfs2_root; static struct task_struct *scand_process; static unsigned int scand_secs = 5; +static struct workqueue_struct *glock_workqueue; #define GFS2_GL_HASH_SHIFT 15 #define GFS2_GL_HASH_SIZE (1 << GFS2_GL_HASH_SHIFT) @@ -277,6 +282,18 @@ static struct gfs2_glock *gfs2_glock_find(const struct gfs2_sbd *sdp, return gl; } +static void glock_work_func(struct work_struct *work) +{ + struct gfs2_glock *gl = container_of(work, struct gfs2_glock, gl_work.work); + + spin_lock(&gl->gl_spin); + if (test_and_clear_bit(GLF_PENDING_DEMOTE, &gl->gl_flags)) + set_bit(GLF_DEMOTE, &gl->gl_flags); + run_queue(gl); + spin_unlock(&gl->gl_spin); + gfs2_glock_put(gl); +} + /** * gfs2_glock_get() - Get a glock, or create one if one doesn't exist * @sdp: The GFS2 superblock @@ -316,6 +333,7 @@ int gfs2_glock_get(struct gfs2_sbd *sdp, u64 number, gl->gl_name = name; atomic_set(&gl->gl_ref, 1); gl->gl_state = LM_ST_UNLOCKED; + gl->gl_demote_state = LM_ST_EXCLUSIVE; gl->gl_hash = hash; gl->gl_owner_pid = 0; gl->gl_ip = 0; @@ -324,10 +342,12 @@ int gfs2_glock_get(struct gfs2_sbd *sdp, u64 number, gl->gl_req_bh = NULL; gl->gl_vn = 0; gl->gl_stamp = jiffies; + gl->gl_tchange = jiffies; gl->gl_object = NULL; gl->gl_sbd = sdp; gl->gl_aspace = NULL; lops_init_le(&gl->gl_le, &gfs2_glock_lops); + INIT_DELAYED_WORK(&gl->gl_work, glock_work_func); /* If this glock protects actual on-disk data or metadata blocks, create a VFS inode to manage the pages/buffers holding them. */ @@ -441,6 +461,8 @@ static void wait_on_holder(struct gfs2_holder *gh) static void gfs2_demote_wake(struct gfs2_glock *gl) { + BUG_ON(!spin_is_locked(&gl->gl_spin)); + gl->gl_demote_state = LM_ST_EXCLUSIVE; clear_bit(GLF_DEMOTE, &gl->gl_flags); smp_mb__after_clear_bit(); wake_up_bit(&gl->gl_flags, GLF_DEMOTE); @@ -682,10 +704,14 @@ static void gfs2_glmutex_unlock(struct gfs2_glock *gl) * practise: LM_ST_SHARED and LM_ST_UNLOCKED */ -static void handle_callback(struct gfs2_glock *gl, unsigned int state, int remote) +static void handle_callback(struct gfs2_glock *gl, unsigned int state, + int remote, unsigned long delay) { + int bit = delay ? GLF_PENDING_DEMOTE : GLF_DEMOTE; + spin_lock(&gl->gl_spin); - if (test_and_set_bit(GLF_DEMOTE, &gl->gl_flags) == 0) { + set_bit(bit, &gl->gl_flags); + if (gl->gl_demote_state == LM_ST_EXCLUSIVE) { gl->gl_demote_state = state; gl->gl_demote_time = jiffies; if (remote && gl->gl_ops->go_type == LM_TYPE_IOPEN && @@ -727,6 +753,7 @@ static void state_change(struct gfs2_glock *gl, unsigned int new_state) } gl->gl_state = new_state; + gl->gl_tchange = jiffies; } /** @@ -813,7 +840,6 @@ out: gl->gl_req_gh = NULL; gl->gl_req_bh = NULL; clear_bit(GLF_LOCK, &gl->gl_flags); - run_queue(gl); spin_unlock(&gl->gl_spin); } @@ -885,7 +911,6 @@ static void drop_bh(struct gfs2_glock *gl, unsigned int ret) gfs2_assert_warn(sdp, !ret); state_change(gl, LM_ST_UNLOCKED); - gfs2_demote_wake(gl); if (glops->go_inval) glops->go_inval(gl, DIO_METADATA); @@ -898,10 +923,10 @@ static void drop_bh(struct gfs2_glock *gl, unsigned int ret) } spin_lock(&gl->gl_spin); + gfs2_demote_wake(gl); gl->gl_req_gh = NULL; gl->gl_req_bh = NULL; clear_bit(GLF_LOCK, &gl->gl_flags); - run_queue(gl); spin_unlock(&gl->gl_spin); gfs2_glock_put(gl); @@ -1209,9 +1234,10 @@ void gfs2_glock_dq(struct gfs2_holder *gh) { struct gfs2_glock *gl = gh->gh_gl; const struct gfs2_glock_operations *glops = gl->gl_ops; + unsigned delay = 0; if (gh->gh_flags & GL_NOCACHE) - handle_callback(gl, LM_ST_UNLOCKED, 0); + handle_callback(gl, LM_ST_UNLOCKED, 0, 0); gfs2_glmutex_lock(gl); @@ -1229,8 +1255,14 @@ void gfs2_glock_dq(struct gfs2_holder *gh) } clear_bit(GLF_LOCK, &gl->gl_flags); - run_queue(gl); spin_unlock(&gl->gl_spin); + + gfs2_glock_hold(gl); + if (test_bit(GLF_PENDING_DEMOTE, &gl->gl_flags) && + !test_bit(GLF_DEMOTE, &gl->gl_flags)) + delay = gl->gl_ops->go_min_hold_time; + if (queue_delayed_work(glock_workqueue, &gl->gl_work, delay) == 0) + gfs2_glock_put(gl); } void gfs2_glock_dq_wait(struct gfs2_holder *gh) @@ -1457,18 +1489,21 @@ 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; - handle_callback(gl, state, 1); - - spin_lock(&gl->gl_spin); - run_queue(gl); - spin_unlock(&gl->gl_spin); + holdtime = gl->gl_tchange + gl->gl_ops->go_min_hold_time; + if (time_before(now, holdtime)) + delay = holdtime - now; - gfs2_glock_put(gl); + handle_callback(gl, state, 1, delay); + if (queue_delayed_work(glock_workqueue, &gl->gl_work, delay) == 0) + gfs2_glock_put(gl); } /** @@ -1509,7 +1544,8 @@ void gfs2_glock_cb(void *cb_data, unsigned int type, void *data) return; if (!gfs2_assert_warn(sdp, gl->gl_req_bh)) gl->gl_req_bh(gl, async->lc_ret); - gfs2_glock_put(gl); + if (queue_delayed_work(glock_workqueue, &gl->gl_work, 0) == 0) + gfs2_glock_put(gl); up_read(&gfs2_umount_flush_sem); return; } @@ -1602,7 +1638,7 @@ void gfs2_reclaim_glock(struct gfs2_sbd *sdp) if (gfs2_glmutex_trylock(gl)) { if (list_empty(&gl->gl_holders) && gl->gl_state != LM_ST_UNLOCKED && demote_ok(gl)) - handle_callback(gl, LM_ST_UNLOCKED, 0); + handle_callback(gl, LM_ST_UNLOCKED, 0, 0); gfs2_glmutex_unlock(gl); } @@ -1702,7 +1738,7 @@ static void clear_glock(struct gfs2_glock *gl) if (gfs2_glmutex_trylock(gl)) { if (list_empty(&gl->gl_holders) && gl->gl_state != LM_ST_UNLOCKED) - handle_callback(gl, LM_ST_UNLOCKED, 0); + handle_callback(gl, LM_ST_UNLOCKED, 0, 0); gfs2_glmutex_unlock(gl); } } @@ -2009,11 +2045,18 @@ int __init gfs2_glock_init(void) if (IS_ERR(scand_process)) return PTR_ERR(scand_process); + glock_workqueue = create_workqueue("glock_workqueue"); + if (IS_ERR(glock_workqueue)) { + kthread_stop(scand_process); + return PTR_ERR(glock_workqueue); + } + return 0; } void gfs2_glock_exit(void) { + destroy_workqueue(glock_workqueue); kthread_stop(scand_process); } -- cgit v1.1 From b4c20166dcfca106f0f416bfce200099ed76ab18 Mon Sep 17 00:00:00 2001 From: Abhijith Das Date: Thu, 13 Sep 2007 23:35:27 -0500 Subject: [GFS2] flocks from same process trip kernel BUG at fs/gfs2/glock.c:1118! This patch adds a new flag to the gfs2_holder structure GL_FLOCK. It is set on holders of glocks representing flocks. This flag is checked in add_to_queue() and a process is permitted to queue more than one holder onto a glock if it is set. This solves the issue of a process not being able to do multiple flocks on the same file. Through a single descriptor, a process can now promote and demote flocks. Through multiple descriptors a process can now queue multiple flocks on the same file. There's still the problem of a process deadlocking itself (because gfs2 blocking locks are not interruptible) by queueing incompatible deadlock. Signed-off-by: Abhijith Das Signed-off-by: Steven Whitehouse --- fs/gfs2/glock.c | 43 +++++++++++++++++++++++++------------------ 1 file changed, 25 insertions(+), 18 deletions(-) (limited to 'fs/gfs2/glock.c') diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 931368a..d631cad 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -1106,24 +1106,31 @@ static void add_to_queue(struct gfs2_holder *gh) if (test_and_set_bit(HIF_WAIT, &gh->gh_iflags)) BUG(); - existing = find_holder_by_owner(&gl->gl_holders, gh->gh_owner_pid); - if (existing) { - print_symbol(KERN_WARNING "original: %s\n", existing->gh_ip); - printk(KERN_INFO "pid : %d\n", existing->gh_owner_pid); - printk(KERN_INFO "lock type : %d lock state : %d\n", - existing->gh_gl->gl_name.ln_type, existing->gh_gl->gl_state); - print_symbol(KERN_WARNING "new: %s\n", gh->gh_ip); - printk(KERN_INFO "pid : %d\n", gh->gh_owner_pid); - printk(KERN_INFO "lock type : %d lock state : %d\n", - gl->gl_name.ln_type, gl->gl_state); - BUG(); - } - - existing = find_holder_by_owner(&gl->gl_waiters3, gh->gh_owner_pid); - if (existing) { - print_symbol(KERN_WARNING "original: %s\n", existing->gh_ip); - print_symbol(KERN_WARNING "new: %s\n", gh->gh_ip); - BUG(); + if (!(gh->gh_flags & GL_FLOCK)) { + existing = find_holder_by_owner(&gl->gl_holders, + gh->gh_owner_pid); + if (existing) { + print_symbol(KERN_WARNING "original: %s\n", + existing->gh_ip); + printk(KERN_INFO "pid : %d\n", existing->gh_owner_pid); + printk(KERN_INFO "lock type : %d lock state : %d\n", + existing->gh_gl->gl_name.ln_type, + existing->gh_gl->gl_state); + print_symbol(KERN_WARNING "new: %s\n", gh->gh_ip); + printk(KERN_INFO "pid : %d\n", gh->gh_owner_pid); + printk(KERN_INFO "lock type : %d lock state : %d\n", + gl->gl_name.ln_type, gl->gl_state); + BUG(); + } + + existing = find_holder_by_owner(&gl->gl_waiters3, + gh->gh_owner_pid); + if (existing) { + print_symbol(KERN_WARNING "original: %s\n", + existing->gh_ip); + print_symbol(KERN_WARNING "new: %s\n", gh->gh_ip); + BUG(); + } } if (gh->gh_flags & LM_FLAG_PRIORITY) -- cgit v1.1 From 49e61f2ef6f7d1d0296e3e30d366b28e0ca595c2 Mon Sep 17 00:00:00 2001 From: Wendy Cheng Date: Thu, 13 Sep 2007 17:52:42 -0400 Subject: [GFS2] Move inode deletion out of blocking_cb Move inode deletion code out of blocking_cb handle_callback route to avoid racy conditions that end up blocking lock_dlm1 thread. Fix bugzilla 286821. Signed-off-by: Wendy Cheng Signed-off-by: Steven Whitehouse --- fs/gfs2/glock.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) (limited to 'fs/gfs2/glock.c') diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index d631cad..a37efe4 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -716,12 +716,8 @@ static void handle_callback(struct gfs2_glock *gl, unsigned int state, gl->gl_demote_time = jiffies; if (remote && gl->gl_ops->go_type == LM_TYPE_IOPEN && gl->gl_object) { - struct inode *inode = igrab(gl->gl_object); + gfs2_glock_schedule_for_reclaim(gl); spin_unlock(&gl->gl_spin); - if (inode) { - d_prune_aliases(inode); - iput(inode); - } return; } } else if (gl->gl_demote_state != LM_ST_UNLOCKED && -- cgit v1.1