Convert the lock for lkbidr to an rwlock. Most idr lookups will use
the read lock.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
The conversion to rhashtable introduced a hash table lock per lockspace,
in place of per bucket locks. To make this more scalable, switch to
using a rwlock for hash table access. The common case fast path uses
it as a read lock.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Currently the scand kthread acts like a garbage collection for expired
rsbs on toss list, to clean them up after a certain timeout. It triggers
every couple of seconds and iterates over the toss list while holding
ls_rsbtbl_lock for the whole hash bucket iteration.
To reduce the amount of time holding ls_rsbtbl_lock, we now handle the
disposal of expired rsbs using a per-lockspace timer that expires for the
earliest tossed rsb on the lockspace toss queue. This toss queue is
ordered according to the rsb res_toss_time with the earliest tossed rsb
as the first entry. The toss timer will only trylock() necessary locks,
since it is low priority garbage collection, and will rearm the timer
if trylock() fails. If the timer function does not find any expired
rsb's, it rearms the timer with the next earliest expired rsb.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
In the past we had problems when an rsb had a reference counter greater
than one while in the toss state. An rsb in the toss state is not
actively used for locking, and should not have any other references
apart from the single ref keeping it on the rsb hash. Shift to freeing
rsb's directly rather than using kref_put to free them, since the ref
counting is not meant to be used in this state. Add warnings if ref
counting is seen while an rsb is in the toss state.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Replace our own hash table with the more advanced rhashtable
for keeping rsb structs.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
To prepare for using rhashtable, add two rsb lists for iterating
through rsb's in two uncommon cases where this is necesssary:
- when dumping rsb state from debugfs, now using seq_list.
- when looking at all rsb's during recovery.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
There are several places where lock processing can perform two hash table
lookups, first in the "keep" list, and if not found, in the "toss" list.
This patch introduces a new rsb state flag "RSB_TOSS" to represent the
difference between the state of being on keep vs toss list, so that the
two lists can be combined. This avoids cases of two lookups.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Prepare to replace our own hash table with rhashtable by replacing
the per-bucket locks in our own hash table with a single lock.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Increment the ls_count value while dlm_scand is processing a
lockspace so that release_lockspace()/remove_lockspace() will
wait for dlm_scand to finish.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Move dlm message processing from an ordered workqueue context to an
ordered softirq context. Handling dlm messages in softirq will allow
requests to be cleared more quickly and efficiently, and should avoid
longer queues of incomplete requests. Later patches are expected to
run completion/blocking callbacks directly from this message processing
context, further reducing context switches required to complete a request.
In the longer term, concurrent message processing could be implemented.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Use spin_lock_bh for all spinlocks involved in message processing,
in preparation for softirq message processing. DLM lock requests
from user space involve dlm processing in user context, in addition
to the standard kernel context, necessitating bh variants.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Remove an explicit schedule() call in the message processing path,
in preparation for softirq message processing.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Convert ls_recv_active rw_semaphore to an rwlock to avoid
sleeping, in preparation for softirq message processing.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
The end of the recovery process transitioned to normal message
processing by temporarily blocking the receiving context,
processing saved messages, then unblocking the receiving
context. To avoid blocking the receiving context, the old
wait_queue and mutex are replaced by a new rwlock and the new
RECV_MSG_BLOCKED flag. Received messages are added to the
list of saved messages, protected by the rwlock, until the
flag is cleared, which happens when all saved messages have
been processed.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Convert the rsb struct res_lock from a mutex to a spinlock
in preparation for processing messages in softirq context.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Convert the waiters mutex to a spinlock in prepration for
processing messages in softirq context.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
The waiters_mutex no longer needs to be used in the waiters recovery
functions dlm_recover_waiters_pre() and dlm_recover_waiters_pre().
During recovery, ordinary locking operations are paused, and the
recovery thread is the only context accessing the waiters list,
so the lock is not needed.
Access to the waiters list from debugfs functions is avoided by
taking the top level recovery lock in the debugfs dump function.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Add a new struct to save the current position in the rsb masters_list
while sending the rsb names to other nodes. The rsb names are sent in
multiple chunks, and for each new chunk, the new "dlm_dir_dump" struct
saves the last position in the masters_list. The new struct is also
used to save more information to sanity check the recovery process.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Move the rsb root_list from the lockspace to a stack variable since
it is now only used by the ls_recover() function.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Add a new "masters_list" for master rsb structs, with a new
rwlock. The new list is created and used during the recovery
process to send the master rsb names to new nodes. With this
change, the current "root_list" can be used without locking.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Move dlm_create_root_list() and dlm_release_root_list() to
recover.c and declare them static because they are only used
there.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Replace GFP_NOFS with GFP_ATOMIC. Also stop using idr_preload which
uses a non-bh spin_lock. This is further preparation for softirq
message processing.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Remove the context parameter for message allocations and
always use GFP_ATOMIC. This prepares for softirq message
processing.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Use the new KMEM_CACHE() macro instead of direct kmem_cache_create
to simplify the creation of SLAB caches.
Signed-off-by: Kunwu Chan <chentao@kylinos.cn>
Acked-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Get rid of the unnecessary refcounting on callback structs.
Copy interesting callback info into the lkb struct rather
than maintaining pointers to callback structs from the lkb.
This goes back to the way things were done prior to
commit 61bed0baa4 ("fs: dlm: use a non-static queue for callbacks").
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
This patch fixes the following issue:
node 1 is dir
node 2 is master
node 3 is other
1->2: unlock
2: put final lkb, rsb moved to toss
2->1: unlock_reply
1: queue lkb callback with EUNLOCK
2->1: remove
1: receive_remove ignored (rsb on keep because of queued lkb callback)
1: complete lkb callback, put_lkb, move rsb to toss
3->1: lookup
1->3: lookup_reply master=2
3->2: request
2->3: request_reply EBADR
In summary:
An unexpected lkb reference causes the rsb to remain on the wrong list.
The rsb being on the wrong list causes receive_remove to be ignored.
An ignored receive_remove causes inconsistent dir and master state.
This sequence requires an unusually long delay in delivering the unlock
callback, because the remove message from 2->1 usually happens after
some seconds. So, it's not known exactly how frequently this sequence
occurs in pratice. It's possible that the same end result could also
have another unknown cause.
The solution for this issue is to further separate callback state
from the lkb, so that an lkb reference (and from that, an rsb ref)
are not held while a callback remains queued. Then, within the
unlock_reply, the lkb will be freed and the rsb moved to the toss
list. So, the receive_remove will not be ignored.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
This patch combines the failure and default cases for enqueue and
dequeue a callback to the lkb callback queue that should end in both
cases as it should never happen.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Save lkb callback info when queueing the callback so that the
lkb struct is not needed in the callback workqueue processing.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Remove the ability to dump pending lkb callbacks from debugfs.
The prepares for separating lkb structs from callbacks.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Stop using lkb structs in the callback tracepoints so that lkb
references are not needed. This prepares for separating lkb
structs from callbacks.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Use the new KMEM_CACHE() macro instead of direct kmem_cache_create
to simplify the creation of SLAB caches.
Signed-off-by: Kunwu Chan <chentao@kylinos.cn>
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
This patch fixes the copy lvb decision for user space lock requests.
Checking dlm_lvb_operations is done earlier, where granted/requested
lock modes are available to use in the matrix.
The decision had been moved to the wrong location, where granted mode
and requested mode where the same, which causes the dlm_lvb_operations
matix to produce the wrong copy decision. For PW or EX requests, the
caller could get invalid lvb data.
Fixes: 61bed0baa4 ("fs: dlm: use a non-static queue for callbacks")
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
- Fix mistaken variable assignment that caused a refcounting problem.
- Revert a recent change that began using atomic counters where they
were not needed (for lkb wait_count.)
- Add comments around forced state reset for waiting lock operations
during recovery.
-----BEGIN PGP SIGNATURE-----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=6RCz
-----END PGP SIGNATURE-----
Merge tag 'dlm-6.9' of git://git.kernel.org/pub/scm/linux/kernel/git/teigland/linux-dlm
Pull dlm updates from David Teigland:
- Fix mistaken variable assignment that caused a refcounting problem
- Revert a recent change that began using atomic counters where they
were not needed (for lkb wait_count)
- Add comments around forced state reset for waiting lock operations
during recovery
* tag 'dlm-6.9' of git://git.kernel.org/pub/scm/linux/kernel/git/teigland/linux-dlm:
dlm: add comments about forced waiters reset
dlm: revert atomic_t lkb_wait_count
dlm: fix user space lkb refcounting
When a lock is waiting for a reply for a remote operation, and recovery
interrupts this "waiters" state, the remote operation is voided by the
recovery, and no reply will be processed. The lkb waiters state for the
remote operation is forcibly reset/cleared, so that the lock operation
can be restarted after recovery. Improve the comments describing this.
Signed-off-by: David Teigland <teigland@redhat.com>
Revert "fs: dlm: handle lkb wait count as atomic_t"
This reverts commit 75a7d60134.
This counter does not need to be atomic. As the comment in
the reverted commit mentions, the counter is protected by
the rsb lock.
Signed-off-by: David Teigland <teigland@redhat.com>
This patch fixes to check on the right return value if it was the last
callback. The rv variable got overwritten by the return of
copy_result_to_user(). Fixing it by introducing a second variable for
the return value and don't let rv being overwritten.
Cc: stable@vger.kernel.org
Fixes: 61bed0baa4 ("fs: dlm: use a non-static queue for callbacks")
Reported-by: Valentin Vidić <vvidic@valentin-vidic.from.hr>
Closes: https://lore.kernel.org/gfs2/Ze4qSvzGJDt5yxC3@valentin-vidic.from.hr
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Most of the existing APIs have remained the same, but subsystems that
access file_lock fields directly need to reach into struct
file_lock_core now.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
Link: https://lore.kernel.org/r/20240131-flsplit-v3-37-c6129007ee8d@kernel.org
Reviewed-by: NeilBrown <neilb@suse.de>
Signed-off-by: Christian Brauner <brauner@kernel.org>
In a future patch, we're going to split file leases into their own
structure. Since a lot of the underlying machinery uses the same fields
move those into a new file_lock_core, and embed that inside struct
file_lock.
For now, add some macros to ensure that we can continue to build while
the conversion is in progress.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
Link: https://lore.kernel.org/r/20240131-flsplit-v3-17-c6129007ee8d@kernel.org
Reviewed-by: NeilBrown <neilb@suse.de>
Signed-off-by: Christian Brauner <brauner@kernel.org>
Convert to using the new file locking helper functions. Also, in later
patches we're going to introduce some temporary macros with names that
clash with the variable name in dlm_posix_unlock. Rename it.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
Link: https://lore.kernel.org/r/20240131-flsplit-v3-8-c6129007ee8d@kernel.org
Reviewed-by: NeilBrown <neilb@suse.de>
Signed-off-by: Christian Brauner <brauner@kernel.org>
Over the time the dlm debugfs format string has been changed but the
header wasn't updated. This patch changes the first line dump header and
their meaning to reflect the current formats.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
This patch fixes to set the type 4 format ops in case of table_open4().
It got accidentially changed by commit 541adb0d4d ("fs: dlm: debugfs
for queued callbacks") and since them toss debug dumps the same format
as format 5 that are the queued ast callbacks for lkbs.
Fixes: 541adb0d4d ("fs: dlm: debugfs for queued callbacks")
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
This patch uses the FL_SLEEP flag in struct file_lock to determine if
the lock request is a blocking or non-blocking request. Before dlm was
using IS_SETLKW() was being used which is not usable for lock requests
coming from lockd when EXPORT_OP_SAFE_ASYNC_LOCK inside the export flags
is set.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
This patch is changing the fl_owner value in case of an nfs lock request
to not be the pid of lockd. Instead this patch changes it to be the
owner value that nfs is giving us.
Currently there exists proved problems with this behaviour. One nfsd
server was created to export a gfs2 filesystem mount. Two nfs clients
doing a nfs mount of this export. Those two clients should conflict each
other operating on the same nfs file.
A small test program was written:
int main(int argc, const char *argv[])
{
struct flock fl = {
.l_type = F_WRLCK,
.l_whence = SEEK_SET,
.l_start = 1L,
.l_len = 1L,
};
int fd;
fd = open("filename", O_RDWR | O_CREAT, 0700);
printf("try to lock...\n");
fcntl(fd, F_SETLKW, &fl);
printf("locked!\n");
getc(stdin);
return 0;
}
Running on both clients at the same time and don't interrupting by
pressing any key. It will show that both clients are able to acquire the
lock which shouldn't be the case. The issue is here that the fl_owner
value is the same and the lock context of both clients should be
separated.
This patch lets lockd define how to deal with lock contexts and chose
hopefully the right fl_owner value. A test after this patch was made and
the locks conflicts each other which should be the case.
Acked-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Recent changes to kernel_connect() and kernel_bind() ensure that
callers are insulated from changes to the address parameter made by BPF
SOCK_ADDR hooks. This patch wraps direct calls to ops->connect() and
ops->bind() with kernel_connect() and kernel_bind() to protect callers
in such cases.
Link: https://lore.kernel.org/netdev/9944248dba1bce861375fcce9de663934d933ba9.camel@redhat.com/
Fixes: d74bad4e74 ("bpf: Hooks for sys_connect")
Fixes: 4fbac77d2d ("bpf: Hooks for sys_bind")
Cc: stable@vger.kernel.org
Signed-off-by: Jordan Rife <jrife@google.com>
Signed-off-by: David Teigland <teigland@redhat.com>
If there is a burst of message the receive worker will filling up the
processing queue but where are too slow to process dlm messages. This
patch will slow down the receiver worker to keep the buffer on the
socket layer to tell the sender to backoff. This is done by a threshold
to get the next buffers from the socket after all messages were
processed done by a flush_workqueue(). This however only occurs when we
have a message burst when we e.g. create 1 million locks. If we put more
and more new messages to process in the processqueue we will soon run out
of memory.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
In case of an final DLM message we can't should not send an ack out
after the final message. This patch moves the ack message before the
messages will be transmitted. If it's the final message and the
receiving node turns into DLM_CLOSED state another ack messages will
being received and turning the receiving node into DLM_ESTABLISHED
again.
Fixes: 1696c75f18 ("fs: dlm: add send ack threshold and append acks to msgs")
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
In case we running in a force shutdown in either midcomms or lowcomms
implementation we will make sure we reset all per midcomms node
information.
Fixes: 63e711b081 ("fs: dlm: create midcomms nodes when configure")
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
The idea of commit 63e711b081 ("fs: dlm: create midcomms nodes when
configure") is to set the midcomms node lifetime when a node joins or
leaves the cluster. Currently we can hit the following warning:
[10844.611495] ------------[ cut here ]------------
[10844.615913] WARNING: CPU: 4 PID: 84304 at fs/dlm/midcomms.c:1263
dlm_midcomms_remove_member+0x13f/0x180 [dlm]
or running in a state where we hit a midcomms node usage count in a
negative value:
[ 260.830782] node 2 users dec count -1
The first warning happens when the a specific node does not exists and
it was probably removed but dlm_midcomms_close() which is called when a
node leaves the cluster. The second kernel log message is probably in a
case when dlm_midcomms_addr() is called when a joined the cluster but
due fencing a node leaved the cluster without getting removed from the
lockspace. If the node joins the cluster and it was removed from the
cluster due fencing the first call is to remove the node from lockspaces
triggered by the user space. In both cases if the node wasn't found or
the user count is zero, we should ignore any additional midcomms handling
of dlm_midcomms_remove_member().
Fixes: 63e711b081 ("fs: dlm: create midcomms nodes when configure")
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
This patch will lookup existing nodes instead of always creating them
when dlm_midcomms_addr() is called. The idea is here to create midcomms
nodes when user space getting informed that nodes joins the cluster. This
is the case when dlm_midcomms_addr() is called, however it can be called
multiple times by user space to add several address configurations to one
node e.g. when using SCTP. Those multiple times need to be filtered out
and we doing that by looking up if the node exists before. Due configfs
entry it is safe that this function gets only called once at a time.
Fixes: 63e711b081 ("fs: dlm: create midcomms nodes when configure")
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
There is no need to clear the buffer used to build the file name.
snprintf() already guarantees that it is NULL terminated and such a
(useless) precaution was not done for the first string (i.e
ls_debug_rsb_dentry)
So, save a few LoC.
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
