* [PATCH 0/6] misc fixes for mds
@ 2013-07-17 8:28 Yan, Zheng
2013-07-17 8:28 ` [PATCH 1/6] mds: fix cap revoke confirmation Yan, Zheng
` (6 more replies)
0 siblings, 7 replies; 9+ messages in thread
From: Yan, Zheng @ 2013-07-17 8:28 UTC (permalink / raw)
To: ceph-devel; +Cc: sage, Yan, Zheng
From: "Yan, Zheng" <zheng.z.yan@intel.com>
these patches are also in:
git://github.com/ukernel/ceph.git wip-mds
Regards
Yan, Zheng
^ permalink raw reply [flat|nested] 9+ messages in thread
* [PATCH 1/6] mds: fix cap revoke confirmation
2013-07-17 8:28 [PATCH 0/6] misc fixes for mds Yan, Zheng
@ 2013-07-17 8:28 ` Yan, Zheng
2013-07-17 8:28 ` [PATCH 2/6] mds: revoke GSHARED cap when finishing xlock Yan, Zheng
` (5 subsequent siblings)
6 siblings, 0 replies; 9+ messages in thread
From: Yan, Zheng @ 2013-07-17 8:28 UTC (permalink / raw)
To: ceph-devel; +Cc: sage, Yan, Zheng
From: "Yan, Zheng" <zheng.z.yan@intel.com>
There are several issues in the Capability::confirm_receipt()
1. when receiving a client caps message with 'seq == last_sent',
it doesn't mean we finish revoking caps. The client can send
caps message that only flushes dirty metadata.
2. When receiving a client caps message with 'seq == N', we should
forget pending revocations whose seq numbers are less than N.
This is because, when revoking caps, we create a revoke_info
structure and set its seq number to 'last_sent', then increase
the 'last_sent'.
3. When client actively releases caps (by request), the code only
works for the 'seq == last_sent' case. If there are pending
revocations, we should update them as if the release message
is received before we revoke the corresponding caps.
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
---
src/mds/Capability.h | 44 ++++++++++++++++++++++++++++++++++----------
1 file changed, 34 insertions(+), 10 deletions(-)
diff --git a/src/mds/Capability.h b/src/mds/Capability.h
index fdecb90..d56bdc9 100644
--- a/src/mds/Capability.h
+++ b/src/mds/Capability.h
@@ -142,13 +142,11 @@ public:
_pending = c;
_issued |= c;
} else if (~_pending & c) {
- // adding bits only. remove obsolete revocations?
+ // note prior caps if there are pending revocations
+ if (!_revokes.empty())
+ _revokes.push_back(revoke_info(_pending, last_sent, last_issue));
_pending |= c;
_issued |= c;
- // drop old _revokes with no bits we don't have
- while (!_revokes.empty() &&
- (_revokes.back().before & ~_pending) == 0)
- _revokes.pop_back();
} else {
// no change.
assert(_pending == c);
@@ -169,16 +167,42 @@ public:
for (list<revoke_info>::iterator p = _revokes.begin(); p != _revokes.end(); ++p)
_issued |= p->before;
}
+ void _update_revokes(ceph_seq_t seq, unsigned caps) {
+ // can i forget any revocations?
+ while (!_revokes.empty() && _revokes.front().seq < seq)
+ _revokes.pop_front();
+
+ if (!_revokes.empty() && _revokes.front().seq == seq) {
+ list<revoke_info>::iterator p = _revokes.begin();
+ unsigned prev_pending = p->before;
+ p->before = caps;
+ // client actively released caps?
+ unsigned release = prev_pending & ~caps;
+ if (release) {
+ for (++p; p != _revokes.end(); ++p) {
+ // we issued new caps to client?
+ release &= prev_pending | ~(p->before);
+ if (release == 0)
+ break;
+ prev_pending = p->before;
+ p->before &= ~release;
+ }
+ if (release) {
+ // we issued new caps to client?
+ release &= prev_pending | ~_pending;
+ _pending &= ~release;
+ }
+ }
+ }
+ }
void confirm_receipt(ceph_seq_t seq, unsigned caps) {
if (seq == last_sent) {
- _pending = caps;
_revokes.clear();
_issued = caps;
+ // don't add bits
+ _pending &= caps;
} else {
- // can i forget any revocations?
- while (!_revokes.empty() &&
- _revokes.front().seq <= seq)
- _revokes.pop_front();
+ _update_revokes(seq, caps);
_calc_issued();
}
//check_rdcaps_list();
--
1.8.1.4
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [PATCH 2/6] mds: revoke GSHARED cap when finishing xlock
2013-07-17 8:28 [PATCH 0/6] misc fixes for mds Yan, Zheng
2013-07-17 8:28 ` [PATCH 1/6] mds: fix cap revoke confirmation Yan, Zheng
@ 2013-07-17 8:28 ` Yan, Zheng
2013-07-17 8:28 ` [PATCH 3/6] mds: remove "type != CEPH_LOCK_DN" check in Locker::cancel_locking() Yan, Zheng
` (4 subsequent siblings)
6 siblings, 0 replies; 9+ messages in thread
From: Yan, Zheng @ 2013-07-17 8:28 UTC (permalink / raw)
To: ceph-devel; +Cc: sage, Yan, Zheng
From: "Yan, Zheng" <zheng.z.yan@intel.com>
If lock state is LOCK_XLOCKDONE, the xlocker can have GSHARED cap.
So when finishing xlock, we may need to revoke the GSHARED cap.
In most cases Locker::_finish_xlock() directly set lock state to
LOCK_LOCK or LOCK_EXCL, which hides the issue. If 'num_rdlock > 0'
or 'num_wrlock > 0' when finishing xlock, the issue reveals.
(lock get stuck in LOCK_XLOCKDONE forever)
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
---
src/mds/Locker.cc | 42 +++++++++++++++---------------------------
1 file changed, 15 insertions(+), 27 deletions(-)
diff --git a/src/mds/Locker.cc b/src/mds/Locker.cc
index 30e014a..afaac1c 100644
--- a/src/mds/Locker.cc
+++ b/src/mds/Locker.cc
@@ -547,8 +547,6 @@ void Locker::cancel_locking(Mutation *mut, set<CInode*> *pneed_issue)
bool need_issue = false;
if (lock->get_state() == LOCK_PREXLOCK)
_finish_xlock(lock, &need_issue);
- if (lock->is_stable())
- eval(lock, &need_issue);
if (need_issue)
pneed_issue->insert(static_cast<CInode *>(lock->get_parent()));
}
@@ -1461,16 +1459,21 @@ bool Locker::xlock_start(SimpleLock *lock, MDRequest *mut)
void Locker::_finish_xlock(SimpleLock *lock, bool *pneed_issue)
{
assert(!lock->is_stable());
- if (lock->get_type() != CEPH_LOCK_DN && (static_cast<CInode*>(lock->get_parent())->get_loner()) >= 0)
+ if (lock->get_num_rdlocks() == 0 &&
+ lock->get_num_wrlocks() == 0 &&
+ lock->get_num_client_lease() == 0 &&
+ lock->get_type() != CEPH_LOCK_DN &&
+ (static_cast<CInode*>(lock->get_parent())->get_loner()) >= 0) {
lock->set_state(LOCK_EXCL);
- else
- lock->set_state(LOCK_LOCK);
- if (lock->get_type() == CEPH_LOCK_DN && lock->get_parent()->is_replicated() &&
- !lock->is_waiter_for(SimpleLock::WAIT_WR))
- simple_sync(lock, pneed_issue);
- if (lock->get_cap_shift())
- *pneed_issue = true;
- lock->get_parent()->auth_unpin(lock);
+ lock->get_parent()->auth_unpin(lock);
+ lock->finish_waiters(SimpleLock::WAIT_STABLE|SimpleLock::WAIT_WR|SimpleLock::WAIT_RD);
+ if (lock->get_cap_shift())
+ *pneed_issue = true;
+ } else {
+ // the xlocker may have CEPH_CAP_GSHARED, need to revoke it
+ // if next state is LOCK_LOCK
+ eval_gather(lock, true, pneed_issue);
+ }
}
void Locker::xlock_finish(SimpleLock *lock, Mutation *mut, bool *pneed_issue)
@@ -1508,24 +1511,9 @@ void Locker::xlock_finish(SimpleLock *lock, Mutation *mut, bool *pneed_issue)
SimpleLock::WAIT_WR |
SimpleLock::WAIT_RD, 0);
} else {
- if (lock->get_num_xlocks() == 0 &&
- lock->get_num_rdlocks() == 0 &&
- lock->get_num_wrlocks() == 0 &&
- lock->get_num_client_lease() == 0) {
+ if (lock->get_num_xlocks() == 0)
_finish_xlock(lock, &do_issue);
- }
-
- // others waiting?
- lock->finish_waiters(SimpleLock::WAIT_STABLE |
- SimpleLock::WAIT_WR |
- SimpleLock::WAIT_RD, 0);
}
-
- // eval?
- if (!lock->is_stable())
- eval_gather(lock, false, &do_issue);
- else if (lock->get_parent()->is_auth())
- try_eval(lock, &do_issue);
if (do_issue) {
CInode *in = static_cast<CInode*>(lock->get_parent());
--
1.8.1.4
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [PATCH 3/6] mds: remove "type != CEPH_LOCK_DN" check in Locker::cancel_locking()
2013-07-17 8:28 [PATCH 0/6] misc fixes for mds Yan, Zheng
2013-07-17 8:28 ` [PATCH 1/6] mds: fix cap revoke confirmation Yan, Zheng
2013-07-17 8:28 ` [PATCH 2/6] mds: revoke GSHARED cap when finishing xlock Yan, Zheng
@ 2013-07-17 8:28 ` Yan, Zheng
2013-07-17 8:28 ` [PATCH 4/6] mds: handle "state == LOCK_LOCK_XLOCK" when cancelling xlock Yan, Zheng
` (3 subsequent siblings)
6 siblings, 0 replies; 9+ messages in thread
From: Yan, Zheng @ 2013-07-17 8:28 UTC (permalink / raw)
To: ceph-devel; +Cc: sage, Yan, Zheng
From: "Yan, Zheng" <zheng.z.yan@intel.com>
For acquiring/cancelling xlock, the lock state transitions for
dentry lock and other types of locks are the same. So I think
the "type != CEPH_LOCK_DN" check doesn't make sense.
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
---
src/mds/Locker.cc | 12 +++++-------
1 file changed, 5 insertions(+), 7 deletions(-)
diff --git a/src/mds/Locker.cc b/src/mds/Locker.cc
index afaac1c..aaf6590 100644
--- a/src/mds/Locker.cc
+++ b/src/mds/Locker.cc
@@ -543,13 +543,11 @@ void Locker::cancel_locking(Mutation *mut, set<CInode*> *pneed_issue)
dout(10) << "cancel_locking " << *lock << " on " << *mut << dendl;
if (lock->get_parent()->is_auth()) {
- if (lock->get_type() != CEPH_LOCK_DN) {
- bool need_issue = false;
- if (lock->get_state() == LOCK_PREXLOCK)
- _finish_xlock(lock, &need_issue);
- if (need_issue)
- pneed_issue->insert(static_cast<CInode *>(lock->get_parent()));
- }
+ bool need_issue = false;
+ if (lock->get_state() == LOCK_PREXLOCK)
+ _finish_xlock(lock, &need_issue);
+ if (need_issue)
+ pneed_issue->insert(static_cast<CInode *>(lock->get_parent()));
}
mut->finish_locking(lock);
}
--
1.8.1.4
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [PATCH 4/6] mds: handle "state == LOCK_LOCK_XLOCK" when cancelling xlock
2013-07-17 8:28 [PATCH 0/6] misc fixes for mds Yan, Zheng
` (2 preceding siblings ...)
2013-07-17 8:28 ` [PATCH 3/6] mds: remove "type != CEPH_LOCK_DN" check in Locker::cancel_locking() Yan, Zheng
@ 2013-07-17 8:28 ` Yan, Zheng
2013-07-17 8:28 ` [PATCH 5/6] mds: wake xlock waiter when xlock is done Yan, Zheng
` (2 subsequent siblings)
6 siblings, 0 replies; 9+ messages in thread
From: Yan, Zheng @ 2013-07-17 8:28 UTC (permalink / raw)
To: ceph-devel; +Cc: sage, Yan, Zheng
From: "Yan, Zheng" <zheng.z.yan@intel.com>
If we find lock state is LOCK_LOCK_XLOCK when cancelling xlock,
set lock state to LOCK_XLOCK_DONE and call Locker::eval_gather().
This makes sure the lock will eventually transit to a stable state.
(LOCK_XLOCK_DONE's next state is stable)
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
---
src/mds/Locker.cc | 12 ++++++++++--
1 file changed, 10 insertions(+), 2 deletions(-)
diff --git a/src/mds/Locker.cc b/src/mds/Locker.cc
index aaf6590..47ea6fc 100644
--- a/src/mds/Locker.cc
+++ b/src/mds/Locker.cc
@@ -544,8 +544,13 @@ void Locker::cancel_locking(Mutation *mut, set<CInode*> *pneed_issue)
if (lock->get_parent()->is_auth()) {
bool need_issue = false;
- if (lock->get_state() == LOCK_PREXLOCK)
+ if (lock->get_state() == LOCK_PREXLOCK) {
_finish_xlock(lock, &need_issue);
+ } else if (lock->get_state() == LOCK_LOCK_XLOCK &&
+ lock->get_num_xlocks() == 0) {
+ lock->set_state(LOCK_XLOCKDONE);
+ eval_gather(lock, true, &need_issue);
+ }
if (need_issue)
pneed_issue->insert(static_cast<CInode *>(lock->get_parent()));
}
@@ -1509,8 +1514,11 @@ void Locker::xlock_finish(SimpleLock *lock, Mutation *mut, bool *pneed_issue)
SimpleLock::WAIT_WR |
SimpleLock::WAIT_RD, 0);
} else {
- if (lock->get_num_xlocks() == 0)
+ if (lock->get_num_xlocks() == 0) {
+ if (lock->get_state() == LOCK_LOCK_XLOCK)
+ lock->set_state(LOCK_XLOCKDONE);
_finish_xlock(lock, &do_issue);
+ }
}
if (do_issue) {
--
1.8.1.4
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [PATCH 5/6] mds: wake xlock waiter when xlock is done
2013-07-17 8:28 [PATCH 0/6] misc fixes for mds Yan, Zheng
` (3 preceding siblings ...)
2013-07-17 8:28 ` [PATCH 4/6] mds: handle "state == LOCK_LOCK_XLOCK" when cancelling xlock Yan, Zheng
@ 2013-07-17 8:28 ` Yan, Zheng
2013-07-17 8:28 ` [PATCH 6/6] mds: change LOCK_SCAN to unstable state Yan, Zheng
2013-07-23 5:34 ` [PATCH 0/6] misc fixes for mds Yan, Zheng
6 siblings, 0 replies; 9+ messages in thread
From: Yan, Zheng @ 2013-07-17 8:28 UTC (permalink / raw)
To: ceph-devel; +Cc: sage, Yan, Zheng
From: "Yan, Zheng" <zheng.z.yan@intel.com>
Once a lock is in LOCK_XLOCKDONE state, client already holds xlock
can acquire extra xlock. So wake up xlock waiter when we set lock
state to LOCK_XLOCKDONE.
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
---
src/mds/Locker.cc | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/src/mds/Locker.cc b/src/mds/Locker.cc
index 47ea6fc..a4b50db 100644
--- a/src/mds/Locker.cc
+++ b/src/mds/Locker.cc
@@ -477,6 +477,7 @@ void Locker::set_xlocks_done(Mutation *mut, bool skip_dentry)
continue;
dout(10) << "set_xlocks_done on " << **p << " " << *(*p)->get_parent() << dendl;
(*p)->set_xlock_done();
+ (*p)->finish_waiters(ScatterLock::WAIT_XLOCK);
}
}
@@ -1419,7 +1420,7 @@ bool Locker::xlock_start(SimpleLock *lock, MDRequest *mut)
}
}
- lock->add_waiter(SimpleLock::WAIT_WR|SimpleLock::WAIT_STABLE, new C_MDS_RetryRequest(mdcache, mut));
+ lock->add_waiter(SimpleLock::WAIT_XLOCK|SimpleLock::WAIT_STABLE, new C_MDS_RetryRequest(mdcache, mut));
nudge_log(lock);
return false;
} else {
--
1.8.1.4
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [PATCH 6/6] mds: change LOCK_SCAN to unstable state
2013-07-17 8:28 [PATCH 0/6] misc fixes for mds Yan, Zheng
` (4 preceding siblings ...)
2013-07-17 8:28 ` [PATCH 5/6] mds: wake xlock waiter when xlock is done Yan, Zheng
@ 2013-07-17 8:28 ` Yan, Zheng
2013-07-23 5:34 ` [PATCH 0/6] misc fixes for mds Yan, Zheng
6 siblings, 0 replies; 9+ messages in thread
From: Yan, Zheng @ 2013-07-17 8:28 UTC (permalink / raw)
To: ceph-devel; +Cc: sage, Yan, Zheng
From: "Yan, Zheng" <zheng.z.yan@intel.com>
commit 0071b8e75b (mds: stay in SCAN state in file_eval) makes
Locker::file_eval() ignore lock in LOCK_SCAN state. If there
is no request changes the lock state, the lock can be stuck in
LOCK_SCAN state forever. This can cause client read/write hang
because lock in LOCK_SCAN state does not allow Frw caps.
The fix is change LOCK_SCAN to a unstable state. Thank to the
CInode::STATE_RECOVERING check in Locker::eval_gather(), the
lock stays in the SCAN state while file is being recovering.
The lock will transit to a stable state once the recovery
finishes.
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
---
src/mds/Locker.cc | 18 +++++-------------
src/mds/MDCache.cc | 1 +
src/mds/locks.c | 3 +--
src/mds/locks.h | 1 -
4 files changed, 7 insertions(+), 16 deletions(-)
diff --git a/src/mds/Locker.cc b/src/mds/Locker.cc
index a4b50db..eef48f9 100644
--- a/src/mds/Locker.cc
+++ b/src/mds/Locker.cc
@@ -3435,7 +3435,6 @@ bool Locker::simple_sync(SimpleLock *lock, bool *need_issue)
switch (lock->get_state()) {
case LOCK_MIX: lock->set_state(LOCK_MIX_SYNC); break;
- case LOCK_SCAN:
case LOCK_LOCK: lock->set_state(LOCK_LOCK_SYNC); break;
case LOCK_XSYN: lock->set_state(LOCK_XSYN_SYNC); break;
case LOCK_EXCL: lock->set_state(LOCK_EXCL_SYNC); break;
@@ -3512,7 +3511,6 @@ void Locker::simple_excl(SimpleLock *lock, bool *need_issue)
in = static_cast<CInode *>(lock->get_parent());
switch (lock->get_state()) {
- case LOCK_SCAN:
case LOCK_LOCK: lock->set_state(LOCK_LOCK_EXCL); break;
case LOCK_SYNC: lock->set_state(LOCK_SYNC_EXCL); break;
case LOCK_XSYN: lock->set_state(LOCK_XSYN_EXCL); break;
@@ -3571,7 +3569,6 @@ void Locker::simple_lock(SimpleLock *lock, bool *need_issue)
int old_state = lock->get_state();
switch (lock->get_state()) {
- case LOCK_SCAN: lock->set_state(LOCK_SCAN_LOCK); break;
case LOCK_SYNC: lock->set_state(LOCK_SYNC_LOCK); break;
case LOCK_XSYN:
file_excl(static_cast<ScatterLock*>(lock), need_issue);
@@ -4157,10 +4154,6 @@ void Locker::file_eval(ScatterLock *lock, bool *need_issue)
if (lock->get_parent()->is_freezing_or_frozen())
return;
- // wait for scan
- if (lock->get_state() == LOCK_SCAN)
- return;
-
// excl -> *?
if (lock->get_state() == LOCK_EXCL) {
dout(20) << " is excl" << dendl;
@@ -4347,7 +4340,6 @@ void Locker::file_excl(ScatterLock *lock, bool *need_issue)
switch (lock->get_state()) {
case LOCK_SYNC: lock->set_state(LOCK_SYNC_EXCL); break;
case LOCK_MIX: lock->set_state(LOCK_MIX_EXCL); break;
- case LOCK_SCAN:
case LOCK_LOCK: lock->set_state(LOCK_LOCK_EXCL); break;
case LOCK_XSYN: lock->set_state(LOCK_XSYN_EXCL); break;
default: assert(0);
@@ -4458,12 +4450,12 @@ void Locker::file_recover(ScatterLock *lock)
issue_caps(in);
gather++;
}
- if (gather) {
- lock->get_parent()->auth_pin(lock);
- } else {
- lock->set_state(LOCK_SCAN);
+
+ lock->set_state(LOCK_SCAN);
+ if (gather)
+ in->state_set(CInode::STATE_NEEDSRECOVER);
+ else
mds->mdcache->queue_file_recover(in);
- }
}
diff --git a/src/mds/MDCache.cc b/src/mds/MDCache.cc
index e592dde..51efce8 100644
--- a/src/mds/MDCache.cc
+++ b/src/mds/MDCache.cc
@@ -5720,6 +5720,7 @@ void MDCache::identify_files_to_recover(vector<CInode*>& recover_q, vector<CInod
}
if (recover) {
+ in->auth_pin(&in->filelock);
in->filelock.set_state(LOCK_PRE_SCAN);
recover_q.push_back(in);
diff --git a/src/mds/locks.c b/src/mds/locks.c
index 9031087..37e3f5e 100644
--- a/src/mds/locks.c
+++ b/src/mds/locks.c
@@ -122,8 +122,7 @@ const struct sm_state_t filelock[LOCK_MAX] = {
[LOCK_EXCL_XSYN] = { LOCK_XSYN, false, LOCK_LOCK, 0, 0, XCL, 0, 0, 0, 0, 0,CEPH_CAP_GCACHE|CEPH_CAP_GBUFFER,0,0 },
[LOCK_PRE_SCAN] = { LOCK_SCAN, false, LOCK_LOCK, 0, 0, 0, 0, 0, 0, 0, 0,0,0,0 },
- [LOCK_SCAN] = { 0, false, LOCK_LOCK, 0, 0, 0, 0, 0, 0, 0, 0,0,0,0 },
- [LOCK_SCAN_LOCK] = { LOCK_LOCK, false, LOCK_LOCK, 0, 0, 0, 0, 0, 0, 0, 0,0,0,0 },
+ [LOCK_SCAN] = { LOCK_LOCK, false, LOCK_LOCK, 0, 0, 0, 0, 0, 0, 0, 0,0,0,0 },
};
const struct sm_t sm_filelock = {
diff --git a/src/mds/locks.h b/src/mds/locks.h
index 2adcbf2..d1585ce 100644
--- a/src/mds/locks.h
+++ b/src/mds/locks.h
@@ -86,7 +86,6 @@ enum {
LOCK_PRE_SCAN,
LOCK_SCAN,
- LOCK_SCAN_LOCK,
LOCK_SNAP_SYNC,
--
1.8.1.4
^ permalink raw reply related [flat|nested] 9+ messages in thread
* Re: [PATCH 0/6] misc fixes for mds
2013-07-17 8:28 [PATCH 0/6] misc fixes for mds Yan, Zheng
` (5 preceding siblings ...)
2013-07-17 8:28 ` [PATCH 6/6] mds: change LOCK_SCAN to unstable state Yan, Zheng
@ 2013-07-23 5:34 ` Yan, Zheng
6 siblings, 0 replies; 9+ messages in thread
From: Yan, Zheng @ 2013-07-23 5:34 UTC (permalink / raw)
To: sage; +Cc: ceph-devel
On 07/17/2013 04:28 PM, Yan, Zheng wrote:
> From: "Yan, Zheng" <zheng.z.yan@intel.com>
>
> these patches are also in:
> git://github.com/ukernel/ceph.git wip-mds
>
I found bug in these patches, please ignore them.
^ permalink raw reply [flat|nested] 9+ messages in thread
* [PATCH 0/6] misc fixes for mds
@ 2013-08-05 6:10 Yan, Zheng
0 siblings, 0 replies; 9+ messages in thread
From: Yan, Zheng @ 2013-08-05 6:10 UTC (permalink / raw)
To: ceph-devel; +Cc: sage, Yan, Zheng
From: "Yan, Zheng" <zheng.z.yan@intel.com>
These patches are also in:
git://github.com/ukernel/ceph.git wip-mds
These patches together with my kernel patches have been tested
for about two weeks. They passed several 48+ hours stress testsï¼
no request hang was found.
Regards
Yan, Zheng
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 9+ messages in thread
end of thread, other threads:[~2013-08-05 6:10 UTC | newest]
Thread overview: 9+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2013-07-17 8:28 [PATCH 0/6] misc fixes for mds Yan, Zheng
2013-07-17 8:28 ` [PATCH 1/6] mds: fix cap revoke confirmation Yan, Zheng
2013-07-17 8:28 ` [PATCH 2/6] mds: revoke GSHARED cap when finishing xlock Yan, Zheng
2013-07-17 8:28 ` [PATCH 3/6] mds: remove "type != CEPH_LOCK_DN" check in Locker::cancel_locking() Yan, Zheng
2013-07-17 8:28 ` [PATCH 4/6] mds: handle "state == LOCK_LOCK_XLOCK" when cancelling xlock Yan, Zheng
2013-07-17 8:28 ` [PATCH 5/6] mds: wake xlock waiter when xlock is done Yan, Zheng
2013-07-17 8:28 ` [PATCH 6/6] mds: change LOCK_SCAN to unstable state Yan, Zheng
2013-07-23 5:34 ` [PATCH 0/6] misc fixes for mds Yan, Zheng
2013-08-05 6:10 Yan, Zheng
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.