From ddc09c8ddac8d0f170ba8caa8128801f358dccff Mon Sep 17 00:00:00 2001 From: Kurt Hackel Date: Fri, 5 Jan 2007 15:00:17 -0800 Subject: [PATCH] ocfs2_dlm: Fixes race between migrate and dirty dlmthread was removing lockres' from the dirty list and resetting the dirty flag before shuffling the list. This patch retains the dirty state flag until the lists are shuffled. Signed-off-by: Kurt Hackel Signed-off-by: Sunil Mushran Signed-off-by: Mark Fasheh --- fs/ocfs2/dlm/dlmcommon.h | 1 + fs/ocfs2/dlm/dlmmaster.c | 16 +++++++++++++++- fs/ocfs2/dlm/dlmthread.c | 30 +++++++++++++++++------------- 3 files changed, 33 insertions(+), 14 deletions(-) diff --git a/fs/ocfs2/dlm/dlmcommon.h b/fs/ocfs2/dlm/dlmcommon.h index 04048bb1a1bd..e95ecb2aaf14 100644 --- a/fs/ocfs2/dlm/dlmcommon.h +++ b/fs/ocfs2/dlm/dlmcommon.h @@ -223,6 +223,7 @@ static inline void __dlm_set_joining_node(struct dlm_ctxt *dlm, #define DLM_LOCK_RES_IN_PROGRESS 0x00000010 #define DLM_LOCK_RES_MIGRATING 0x00000020 #define DLM_LOCK_RES_DROPPING_REF 0x00000040 +#define DLM_LOCK_RES_BLOCK_DIRTY 0x00001000 /* max milliseconds to wait to sync up a network failure with a node death */ #define DLM_NODE_DEATH_WAIT_MAX (5 * 1000) diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c index 251c48028ea3..a65a87726d6a 100644 --- a/fs/ocfs2/dlm/dlmmaster.c +++ b/fs/ocfs2/dlm/dlmmaster.c @@ -2707,8 +2707,15 @@ static int dlm_mark_lockres_migrating(struct dlm_ctxt *dlm, __dlm_lockres_reserve_ast(res); spin_unlock(&res->spinlock); - /* now flush all the pending asts.. hang out for a bit */ + /* now flush all the pending asts */ dlm_kick_thread(dlm, res); + /* before waiting on DIRTY, block processes which may + * try to dirty the lockres before MIGRATING is set */ + spin_lock(&res->spinlock); + BUG_ON(res->state & DLM_LOCK_RES_BLOCK_DIRTY); + res->state |= DLM_LOCK_RES_BLOCK_DIRTY; + spin_unlock(&res->spinlock); + /* now wait on any pending asts and the DIRTY state */ wait_event(dlm->ast_wq, !dlm_lockres_is_dirty(dlm, res)); dlm_lockres_release_ast(dlm, res); @@ -2734,6 +2741,13 @@ again: mlog(0, "trying again...\n"); goto again; } + /* now that we are sure the MIGRATING state is there, drop + * the unneded state which blocked threads trying to DIRTY */ + spin_lock(&res->spinlock); + BUG_ON(!(res->state & DLM_LOCK_RES_BLOCK_DIRTY)); + BUG_ON(!(res->state & DLM_LOCK_RES_MIGRATING)); + res->state &= ~DLM_LOCK_RES_BLOCK_DIRTY; + spin_unlock(&res->spinlock); /* did the target go down or die? */ spin_lock(&dlm->spinlock); diff --git a/fs/ocfs2/dlm/dlmthread.c b/fs/ocfs2/dlm/dlmthread.c index baa99979904c..3b94e4dec351 100644 --- a/fs/ocfs2/dlm/dlmthread.c +++ b/fs/ocfs2/dlm/dlmthread.c @@ -95,7 +95,7 @@ int __dlm_lockres_has_locks(struct dlm_lock_resource *res) int __dlm_lockres_unused(struct dlm_lock_resource *res) { if (!__dlm_lockres_has_locks(res) && - list_empty(&res->dirty)) { + (list_empty(&res->dirty) && !(res->state & DLM_LOCK_RES_DIRTY))) { /* try not to scan the bitmap unless the first two * conditions are already true */ int bit = find_next_bit(res->refmap, O2NM_MAX_NODES, 0); @@ -455,12 +455,17 @@ void __dlm_dirty_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res) assert_spin_locked(&res->spinlock); /* don't shuffle secondary queues */ - if ((res->owner == dlm->node_num) && - !(res->state & DLM_LOCK_RES_DIRTY)) { - /* ref for dirty_list */ - dlm_lockres_get(res); - list_add_tail(&res->dirty, &dlm->dirty_list); - res->state |= DLM_LOCK_RES_DIRTY; + if ((res->owner == dlm->node_num)) { + if (res->state & (DLM_LOCK_RES_MIGRATING | + DLM_LOCK_RES_BLOCK_DIRTY)) + return; + + if (list_empty(&res->dirty)) { + /* ref for dirty_list */ + dlm_lockres_get(res); + list_add_tail(&res->dirty, &dlm->dirty_list); + res->state |= DLM_LOCK_RES_DIRTY; + } } } @@ -639,7 +644,7 @@ static int dlm_thread(void *data) dlm_lockres_get(res); spin_lock(&res->spinlock); - res->state &= ~DLM_LOCK_RES_DIRTY; + /* We clear the DLM_LOCK_RES_DIRTY state once we shuffle lists below */ list_del_init(&res->dirty); spin_unlock(&res->spinlock); spin_unlock(&dlm->spinlock); @@ -663,10 +668,11 @@ static int dlm_thread(void *data) /* it is now ok to move lockreses in these states * to the dirty list, assuming that they will only be * dirty for a short while. */ + BUG_ON(res->state & DLM_LOCK_RES_MIGRATING); if (res->state & (DLM_LOCK_RES_IN_PROGRESS | - DLM_LOCK_RES_MIGRATING | DLM_LOCK_RES_RECOVERING)) { /* move it to the tail and keep going */ + res->state &= ~DLM_LOCK_RES_DIRTY; spin_unlock(&res->spinlock); mlog(0, "delaying list shuffling for in-" "progress lockres %.*s, state=%d\n", @@ -687,6 +693,7 @@ static int dlm_thread(void *data) /* called while holding lockres lock */ dlm_shuffle_lists(dlm, res); + res->state &= ~DLM_LOCK_RES_DIRTY; spin_unlock(&res->spinlock); dlm_lockres_calc_usage(dlm, res); @@ -697,11 +704,8 @@ in_progress: /* if the lock was in-progress, stick * it on the back of the list */ if (delay) { - /* ref for dirty_list */ - dlm_lockres_get(res); spin_lock(&res->spinlock); - list_add_tail(&res->dirty, &dlm->dirty_list); - res->state |= DLM_LOCK_RES_DIRTY; + __dlm_dirty_lockres(dlm, res); spin_unlock(&res->spinlock); } dlm_lockres_put(res); -- 2.30.2