linux-stable.git/fs/dlm, branch v5.16

fs: dlm: avoid comms shutdown delay in release_lockspace

2021-09-01T16:29:14+00:00

When dlm_release_lockspace does active shutdown on connections to
other nodes, the active shutdown will wait for any exisitng passive
shutdowns to be resolved.  But, the sequence of operations during
dlm_release_lockspace can prevent the normal resolution of passive
shutdowns (processed normally by way of lockspace recovery.)
This disruption of passive shutdown handling can cause the active
shutdown to wait for a full timeout period, delaying the completion
of dlm_release_lockspace.

To fix this, make dlm_release_lockspace resolve existing passive
shutdowns (by calling dlm_clear_members earlier), before it does
active shutdowns.  The active shutdowns will not find any passive
shutdowns to wait for, and will not be delayed.

Reported-by: Chris Mackowski 
Signed-off-by: Alexander Aring 
Signed-off-by: David Teigland

fs: dlm: fix return -EINTR on recovery stopped

2021-08-19T16:33:03+00:00

This patch will return -EINTR instead of 1 if recovery is stopped. In
case of ping_members() the return value will be checked if the error is
-EINTR for signaling another recovery was triggered and the whole
recovery process will come to a clean end to process the next one.
Returning 1 will abort the recovery process and can leave the recovery
in a broken state.

It was reported with the following kernel log message attached and a gfs2
mount stopped working:

"dlm: bobvirt1: dlm_recover_members error 1"

whereas 1 was returned because of a conversion of "dlm_recovery_stopped()"
to an errno was missing which this patch will introduce. While on it all
other possible missing errno conversions at other places were added as
they are done as in other places.

It might be worth to check the error case at this recovery level,
because some of the functionality also returns -ENOBUFS and check why
recovery ends in a broken state. However this will fix the issue if
another recovery was triggered at some points of recovery handling.

Reported-by: Bob Peterson 
Signed-off-by: Alexander Aring 
Signed-off-by: David Teigland

fs: dlm: implement delayed ack handling

2021-08-19T16:33:03+00:00

This patch changes that we don't ack each message. Lowcomms will take
care about to send an ack back after a bulk of messages was processed.
Currently it's only when the whole receive buffer was processed, there
might better positions to send an ack back but only the lowcomms
implementation know when there are more data to receive. This patch has
also disadvantages that we might retransmit more on errors, however this
is a very rare case.

Tested with make_panic on gfs2 with three nodes by running:

trace-cmd record -p function -l 'dlm_send_ack' sleep 100

and

trace-cmd report | wc -l

Before patch:
- 20548
- 21376
- 21398

After patch:
- 18338
- 20679
- 19949

Signed-off-by: Alexander Aring 
Signed-off-by: David Teigland

fs: dlm: move receive loop into receive handler

2021-07-19T16:56:36+00:00

This patch moves the kernel_recvmsg() loop call into the
receive_from_sock() function instead of doing the loop outside the
function and abort the loop over it's return value.

Signed-off-by: Alexander Aring 
Signed-off-by: David Teigland

fs: dlm: fix multiple empty writequeue alloc

2021-07-19T16:56:35+00:00

This patch will add a mutex that a connection can allocate a writequeue
entry buffer only at a sleepable context at one time. If multiple caller
waits at the writequeue spinlock and the spinlock gets release it could
be that multiple new writequeue page buffers were allocated instead of
allocate one writequeue page buffer and other waiters will use remaining
buffer of it. It will only be the case for sleepable context which is
the common case. In non-sleepable contexts like retransmission we just
don't care about such behaviour.

Signed-off-by: Alexander Aring 
Signed-off-by: David Teigland

fs: dlm: generic connect func

2021-07-19T16:56:34+00:00

This patch adds a generic connect function for TCP and SCTP. If the
connect functionality differs from each other additional callbacks in
dlm_proto_ops were added. The sockopts callback handling will guarantee
that sockets created by connect() will use the same options as sockets
created by accept().

Signed-off-by: Alexander Aring 
Signed-off-by: David Teigland

fs: dlm: auto load sctp module

2021-07-19T16:56:26+00:00

This patch adds a "for now" better handling of missing SCTP support in
the kernel and try to load the sctp module if SCTP is set.

Signed-off-by: Alexander Aring 
Signed-off-by: David Teigland

fs: dlm: introduce generic listen

2021-07-19T16:53:43+00:00

This patch combines each transport layer listen functionality into one
listen function. Per transport layer differences are provided by
additional callbacks in dlm_proto_ops.

This patch drops silently sock_set_keepalive() for listen tcp sockets
only. This socket option is not set at connecting sockets, I also don't
see the sense of set keepalive for sockets which are created by accept()
only.

Signed-off-by: Alexander Aring 
Signed-off-by: David Teigland

fs: dlm: move to static proto ops

2021-07-19T16:53:43+00:00

This patch moves the per transport socket callbacks to a static const
array. We can support only one transport socket for the init namespace
which will be determinted by reading the dlm config at lowcomms_start().

Signed-off-by: Alexander Aring 
Signed-off-by: David Teigland

fs: dlm: introduce con_next_wq helper

2021-07-19T16:53:43+00:00

This patch introduce a function to determine if something is ready to
being send in the writequeue. It's not just that the writequeue is not
empty additional the first entry need to have a valid length field.

Signed-off-by: Alexander Aring 
Signed-off-by: David Teigland