diff options
author | Benjamin Herrenschmidt <benh@kernel.crashing.org> | 2011-05-20 15:36:52 +1000 |
---|---|---|
committer | Benjamin Herrenschmidt <benh@kernel.crashing.org> | 2011-05-20 15:36:52 +1000 |
commit | 880102e78547c1db158a17e36cf0cdd98e7ad710 (patch) | |
tree | 3fff9cc54c44dafe275cfabefb96c589e08d971d /fs/ocfs2/cluster/heartbeat.c | |
parent | 3d07f0e83d4323d2cd45cc583f7cf1957aca3cac (diff) | |
parent | 39ab05c8e0b519ff0a04a869f065746e6e8c3d95 (diff) |
Merge remote branch 'origin/master' into merge
Manual merge of arch/powerpc/kernel/smp.c and add missing scheduler_ipi()
call to arch/powerpc/platforms/cell/interrupt.c
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Diffstat (limited to 'fs/ocfs2/cluster/heartbeat.c')
-rw-r--r-- | fs/ocfs2/cluster/heartbeat.c | 61 |
1 files changed, 43 insertions, 18 deletions
diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c index 643720209a98..9a3e6bbff27b 100644 --- a/fs/ocfs2/cluster/heartbeat.c +++ b/fs/ocfs2/cluster/heartbeat.c @@ -539,25 +539,41 @@ static int o2hb_verify_crc(struct o2hb_region *reg, /* We want to make sure that nobody is heartbeating on top of us -- * this will help detect an invalid configuration. */ -static int o2hb_check_last_timestamp(struct o2hb_region *reg) +static void o2hb_check_last_timestamp(struct o2hb_region *reg) { - int node_num, ret; struct o2hb_disk_slot *slot; struct o2hb_disk_heartbeat_block *hb_block; + char *errstr; - node_num = o2nm_this_node(); - - ret = 1; - slot = ®->hr_slots[node_num]; + slot = ®->hr_slots[o2nm_this_node()]; /* Don't check on our 1st timestamp */ - if (slot->ds_last_time) { - hb_block = slot->ds_raw_block; + if (!slot->ds_last_time) + return; - if (le64_to_cpu(hb_block->hb_seq) != slot->ds_last_time) - ret = 0; - } + hb_block = slot->ds_raw_block; + if (le64_to_cpu(hb_block->hb_seq) == slot->ds_last_time && + le64_to_cpu(hb_block->hb_generation) == slot->ds_last_generation && + hb_block->hb_node == slot->ds_node_num) + return; - return ret; +#define ERRSTR1 "Another node is heartbeating on device" +#define ERRSTR2 "Heartbeat generation mismatch on device" +#define ERRSTR3 "Heartbeat sequence mismatch on device" + + if (hb_block->hb_node != slot->ds_node_num) + errstr = ERRSTR1; + else if (le64_to_cpu(hb_block->hb_generation) != + slot->ds_last_generation) + errstr = ERRSTR2; + else + errstr = ERRSTR3; + + mlog(ML_ERROR, "%s (%s): expected(%u:0x%llx, 0x%llx), " + "ondisk(%u:0x%llx, 0x%llx)\n", errstr, reg->hr_dev_name, + slot->ds_node_num, (unsigned long long)slot->ds_last_generation, + (unsigned long long)slot->ds_last_time, hb_block->hb_node, + (unsigned long long)le64_to_cpu(hb_block->hb_generation), + (unsigned long long)le64_to_cpu(hb_block->hb_seq)); } static inline void o2hb_prepare_block(struct o2hb_region *reg, @@ -983,9 +999,7 @@ static int o2hb_do_disk_heartbeat(struct o2hb_region *reg) /* With an up to date view of the slots, we can check that no * other node has been improperly configured to heartbeat in * our slot. */ - if (!o2hb_check_last_timestamp(reg)) - mlog(ML_ERROR, "Device \"%s\": another node is heartbeating " - "in our slot!\n", reg->hr_dev_name); + o2hb_check_last_timestamp(reg); /* fill in the proper info for our next heartbeat */ o2hb_prepare_block(reg, reg->hr_generation); @@ -999,8 +1013,8 @@ static int o2hb_do_disk_heartbeat(struct o2hb_region *reg) } i = -1; - while((i = find_next_bit(configured_nodes, O2NM_MAX_NODES, i + 1)) < O2NM_MAX_NODES) { - + while((i = find_next_bit(configured_nodes, + O2NM_MAX_NODES, i + 1)) < O2NM_MAX_NODES) { change |= o2hb_check_slot(reg, ®->hr_slots[i]); } @@ -1690,6 +1704,7 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg, struct file *filp = NULL; struct inode *inode = NULL; ssize_t ret = -EINVAL; + int live_threshold; if (reg->hr_bdev) goto out; @@ -1766,8 +1781,18 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg, * A node is considered live after it has beat LIVE_THRESHOLD * times. We're not steady until we've given them a chance * _after_ our first read. + * The default threshold is bare minimum so as to limit the delay + * during mounts. For global heartbeat, the threshold doubled for the + * first region. */ - atomic_set(®->hr_steady_iterations, O2HB_LIVE_THRESHOLD + 1); + live_threshold = O2HB_LIVE_THRESHOLD; + if (o2hb_global_heartbeat_active()) { + spin_lock(&o2hb_live_lock); + if (o2hb_pop_count(&o2hb_region_bitmap, O2NM_MAX_REGIONS) == 1) + live_threshold <<= 1; + spin_unlock(&o2hb_live_lock); + } + atomic_set(®->hr_steady_iterations, live_threshold + 1); hb_task = kthread_run(o2hb_thread, reg, "o2hb-%s", reg->hr_item.ci_name); |