Age | Commit message (Collapse) | Author | Files | Lines |
|
The conversion to netfs in the 6.3 kernel caused a regression when
maximum write size is set by the server to an unexpected value which is
not a multiple of 4096 (similarly if the user overrides the maximum
write size by setting mount parm "wsize", but sets it to a value that
is not a multiple of 4096). When negotiated write size is not a
multiple of 4096 the netfs code can skip the end of the final
page when doing large sequential writes, causing data corruption.
This section of code is being rewritten/removed due to a large
netfs change, but until that point (ie for the 6.3 kernel until now)
we can not support non-standard maximum write sizes.
Add a warning if a user specifies a wsize on mount that is not
a multiple of 4096 (and round down), also add a change where we
round down the maximum write size if the server negotiates a value
that is not a multiple of 4096 (we also have to check to make sure that
we do not round it down to zero).
Reported-by: R. Diez" <rdiez-2006@rd10.de>
Fixes: d08089f649a0 ("cifs: Change the I/O paths to use an iterator rather than a page list")
Suggested-by: Ronnie Sahlberg <ronniesahlberg@gmail.com>
Acked-by: Ronnie Sahlberg <ronniesahlberg@gmail.com>
Tested-by: Matthew Ruffell <matthew.ruffell@canonical.com>
Reviewed-by: Shyam Prasad N <sprasad@microsoft.com>
Cc: stable@vger.kernel.org # v6.3+
Cc: David Howells <dhowells@redhat.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Based on our implementation of multichannel, it is entirely
possible that a server struct may not be found in any channel
of an SMB session.
In such cases, we should be prepared to move on and search for
the server struct in the next session.
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
When a tcon is marked for need_reconnect, the intention
is to have it reconnected.
This change adjusts tcon->status in cifs_tree_connect
when need_reconnect is set. Also, this change has a minor
correction in resetting need_reconnect on success. It makes
sure that it is done with tc_lock held.
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
We have several places in the code where we treat the
error -EAGAIN very differently. Some code retry for
arbitrary number of times.
Introducing this new mount option named "retrans", so
that all these handlers of -EAGAIN can retry a fixed
number of times. This applies only to soft mounts.
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Fixes no-op checkpatch errors and warnings.
Signed-off-by: Pierre Mariani <pierre.mariani@gmail.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
cifs_chan_is_iface_active checks the channels of a session to see
if the associated iface is active. This should always happen
with chan_lock held. However, these two callers of this function
were missing this locking.
This change makes sure the function calls are protected with
proper locking.
Fixes: b54034a73baf ("cifs: during reconnect, update interface if necessary")
Fixes: fa1d0508bdd4 ("cifs: account for primary channel in the interface list")
Cc: stable@vger.kernel.org
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Once the server disables multichannel for an active multichannel
session, on the following reconnect, the client would reduce
the number of channels to 1. However, it could be the case that
the tree connect was active on one of these disabled channels.
This results in an unrecoverable state.
This change fixes that by making sure that whenever a channel
is being terminated, the session and tcon are marked for
reconnect too. This could mean a few redundant tree connect
calls to the server, but considering that this is not a frequent
event, we should be okay.
Fixes: ee1d21794e55 ("cifs: handle when server stops supporting multichannel")
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Validate next header's offset in ->next_header() so that it isn't
smaller than MID_HEADER_SIZE(server) and then standard_receive3() or
->receive() ends up writing off the end of the buffer because
'pdu_length - MID_HEADER_SIZE(server)' wraps up to a huge length:
BUG: KASAN: slab-out-of-bounds in _copy_to_iter+0x4fc/0x840
Write of size 701 at addr ffff88800caf407f by task cifsd/1090
CPU: 0 PID: 1090 Comm: cifsd Not tainted 6.7.0-rc4 #5
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS
rel-1.16.2-3-gd478f380-rebuilt.opensuse.org 04/01/2014
Call Trace:
<TASK>
dump_stack_lvl+0x4a/0x80
print_report+0xcf/0x650
? srso_alias_return_thunk+0x5/0xfbef5
? srso_alias_return_thunk+0x5/0xfbef5
? __phys_addr+0x46/0x90
kasan_report+0xd8/0x110
? _copy_to_iter+0x4fc/0x840
? _copy_to_iter+0x4fc/0x840
kasan_check_range+0x105/0x1b0
__asan_memcpy+0x3c/0x60
_copy_to_iter+0x4fc/0x840
? srso_alias_return_thunk+0x5/0xfbef5
? hlock_class+0x32/0xc0
? srso_alias_return_thunk+0x5/0xfbef5
? __pfx__copy_to_iter+0x10/0x10
? srso_alias_return_thunk+0x5/0xfbef5
? lock_is_held_type+0x90/0x100
? srso_alias_return_thunk+0x5/0xfbef5
? __might_resched+0x278/0x360
? __pfx___might_resched+0x10/0x10
? srso_alias_return_thunk+0x5/0xfbef5
__skb_datagram_iter+0x2c2/0x460
? __pfx_simple_copy_to_iter+0x10/0x10
skb_copy_datagram_iter+0x6c/0x110
tcp_recvmsg_locked+0x9be/0xf40
? __pfx_tcp_recvmsg_locked+0x10/0x10
? mark_held_locks+0x5d/0x90
? srso_alias_return_thunk+0x5/0xfbef5
tcp_recvmsg+0xe2/0x310
? __pfx_tcp_recvmsg+0x10/0x10
? srso_alias_return_thunk+0x5/0xfbef5
? srso_alias_return_thunk+0x5/0xfbef5
? lock_acquire+0x14a/0x3a0
? srso_alias_return_thunk+0x5/0xfbef5
inet_recvmsg+0xd0/0x370
? __pfx_inet_recvmsg+0x10/0x10
? __pfx_lock_release+0x10/0x10
? do_raw_spin_trylock+0xd1/0x120
sock_recvmsg+0x10d/0x150
cifs_readv_from_socket+0x25a/0x490 [cifs]
? __pfx_cifs_readv_from_socket+0x10/0x10 [cifs]
? srso_alias_return_thunk+0x5/0xfbef5
cifs_read_from_socket+0xb5/0x100 [cifs]
? __pfx_cifs_read_from_socket+0x10/0x10 [cifs]
? __pfx_lock_release+0x10/0x10
? do_raw_spin_trylock+0xd1/0x120
? _raw_spin_unlock+0x23/0x40
? srso_alias_return_thunk+0x5/0xfbef5
? __smb2_find_mid+0x126/0x230 [cifs]
cifs_demultiplex_thread+0xd39/0x1270 [cifs]
? __pfx_cifs_demultiplex_thread+0x10/0x10 [cifs]
? __pfx_lock_release+0x10/0x10
? srso_alias_return_thunk+0x5/0xfbef5
? mark_held_locks+0x1a/0x90
? lockdep_hardirqs_on_prepare+0x136/0x210
? srso_alias_return_thunk+0x5/0xfbef5
? srso_alias_return_thunk+0x5/0xfbef5
? __kthread_parkme+0xce/0xf0
? __pfx_cifs_demultiplex_thread+0x10/0x10 [cifs]
kthread+0x18d/0x1d0
? kthread+0xdb/0x1d0
? __pfx_kthread+0x10/0x10
ret_from_fork+0x34/0x60
? __pfx_kthread+0x10/0x10
ret_from_fork_asm+0x1b/0x30
</TASK>
Fixes: 8ce79ec359ad ("cifs: update multiplex loop to handle compounded responses")
Cc: stable@vger.kernel.org
Reported-by: Robert Morris <rtm@csail.mit.edu>
Signed-off-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Reconnect worker currently assumes that the server struct
is alive and only takes reference on the server if it needs
to call smb2_reconnect.
With the new ability to disable channels based on whether the
server has multichannel disabled, this becomes a problem when
we need to disable established channels. While disabling the
channels and deallocating the server, there could be reconnect
work that could not be cancelled (because it started).
This change forces the reconnect worker to unconditionally
take a reference on the server when it runs.
Also, this change now allows smb2_reconnect to know if it was
called by the reconnect worker. Based on this, the cifs_put_tcp_session
can decide whether it can cancel the reconnect work synchronously or not.
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
This reverts commit 19a4b9d6c372cab6a3b2c9a061a236136fe95274.
This earlier commit was making an assumption that each mod_delayed_work
called for the reconnect work would result in smb2_reconnect_server
being called twice. This assumption turns out to be untrue. So reverting
this change for now.
I will submit a follow-up patch to fix the actual problem in a different
way.
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
My last change in this area introduced a change which
accounted for primary channel in the interface ref count.
However, it did not reduce this ref count on deallocation
of the primary channel. i.e. during umount.
Fixing this leak here, by dropping this ref count for
primary channel while freeing up the session.
Fixes: fa1d0508bdd4 ("cifs: account for primary channel in the interface list")
Cc: stable@vger.kernel.org
Reported-by: Paulo Alcantara <pc@manguebit.com>
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
When a server stops supporting multichannel, we will
keep attempting reconnects to the secondary channels today.
Avoid this by freeing extra channels when negotiate
returns no multichannel support.
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
When the user mounts with multichannel option, but the
server does not support it, there can be a time in future
where it can be supported.
With this change, such a case is handled.
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
|
|
The delayed work for reconnect takes server struct
as a parameter. But it does so without holding a ref
to it. Normally, this may not show a problem as
the reconnect work is only cancelled on umount.
However, since we now plan to support scaling down of
channels, and the scale down can happen from reconnect
work itself, we need to fix it.
This change takes a reference on the server struct
before it is passed to the delayed work. And drops
the reference in the delayed work itself. Or if
the delayed work is successfully cancelled, by the
process that cancels it.
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
The only reason why cifs_sb gets passed today to cifs_try_adding_channels
is to pass the local_nls field for the new channels and binding session.
However, the ses struct already has local_nls field that is setup during
the first cifs_setup_session. So there is no need to pass cifs_sb.
This change removes cifs_sb from the arg list for this and the functions
that it calls and uses ses->local_nls instead.
Cc: stable@vger.kernel.org
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
Reviewed-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
So far, SMB multichannel could only scale up, but not
scale down the number of channels. In this series of
patch, we now allow the client to deal with the case
of multichannel disabled on the server when the share
is mounted. With that change, we now need the ability
to scale down the channels.
This change allows the client to deal with cases of
missing channels more gracefully.
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
During a session reconnect, it is possible that the
server moved to another physical server (happens in case
of Azure files). So at this time, force a query of server
interfaces again (in case of multichannel session), such
that the secondary channels connect to the right
IP addresses (possibly updated now).
Cc: stable@vger.kernel.org
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
We introduced a helper function to be used by non-cifsd threads to
mark the connection for reconnect. For multichannel, when only
a particular channel needs to be reconnected, this had a bug.
This change fixes that by marking that particular channel
for reconnect.
Fixes: dca65818c80c ("cifs: use a different reconnect helper for non-cifsd threads")
Cc: stable@vger.kernel.org
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
Reviewed-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
If @ses->chan_count <= 1, then for-loop body will not be executed so
no need to check it twice.
Reviewed-by: Shyam Prasad N <sprasad@microsoft.com>
Signed-off-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
We were passing 0 as the xid for the call to query
server interfaces. This is not great for debugging.
This change adds a real xid.
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
Reviewed-by: Bharath SM <bharathsm@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Honor 'nohandlecache' mount option by not starting laundromat thread
even when SMB server supports directory leases. Do not waste system
resources by having laundromat thread running with no directory
caching at all.
Fixes: 2da338ff752a ("smb3: do not start laundromat thread when dir leases disabled")
Signed-off-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.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 ensure that SMB
mounts do not see their mount address overwritten in such cases.
Link: https://lore.kernel.org/netdev/9944248dba1bce861375fcce9de663934d933ba9.camel@redhat.com/
Cc: <stable@vger.kernel.org> # 6.0+
Signed-off-by: Jordan Rife <jrife@google.com>
Acked-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
disabled
When no directory lease support, or for IPC shares where directories
can not be opened, do not start an unneeded laundromat thread for
that mount (it wastes resources).
Fixes: d14de8067e3f ("cifs: Add a laundromat thread for cached directories")
Reviewed-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Acked-by: Tom Talpey <tom@talpey.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Allow adjusting the maximum number of cached directories per share
(defaults to 16) via mount parm "max_cached_dirs"
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Since older dialects such as CIFS do not support multichannel
the macro CIFS_SERVER_IS_CHAN can be confusing (it requires SMB 3
or later) so shorten its name to "SERVER_IS_CHAN"
Suggested-by: Tom Talpey <tom@talpey.com>
Acked-by: Shyam Prasad N <sprasad@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
The ChannelSequence field in the SMB3 header is supposed to be
increased after reconnect to allow the server to distinguish
requests from before and after the reconnect. We had always
been setting it to zero. There are cases where incrementing
ChannelSequence on requests after network reconnects can reduce
the chance of data corruptions.
See MS-SMB2 3.2.4.1 and 3.2.7.1
Signed-off-by: Steve French <stfrench@microsoft.com>
Cc: stable@vger.kernel.org # 5.16+
|
|
Clang warns about exceeded stack frame size
fs/smb/client/connect.c:1109:1: warning: stack frame size (1048)
exceeds limit (1024) in 'cifs_demultiplex_thread'
[-Wframe-larger-than]
It turns out that clean_demultiplex_info() got inlined into
cifs_demultiplex_thread(), so mark it as noinline_for_stack to save
some stack space.
Signed-off-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Add new helper which declares and initialises target list of a DFS
referral rather having to do both separately.
No functional changes.
Signed-off-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
We need to specify charset, like "iocharset=utf-8", in mount options for
Chinese path if the nls_default don't support it, such as iso8859-1, the
default value for CONFIG_NLS_DEFAULT.
But now in reconnection the nls_default is used, instead of the one we
specified and used in mount, and this can lead to mount failure.
Signed-off-by: Winston Wen <wentao@uniontech.com>
Reviewed-by: Paulo Alcantara <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
When the number of responses with status of STATUS_IO_TIMEOUT
exceeds a specified threshold (NUM_STATUS_IO_TIMEOUT), we reconnect
the connection. But we do not return the mid, or the credits
returned for the mid, or reduce the number of in-flight requests.
This bug could result in the server->in_flight count to go bad,
and also cause a leak in the mids.
This change moves the check to a few lines below where the
response is decrypted, even of the response is read from the
transform header. This way, the code for returning the mids
can be reused.
Also, the cifs_reconnect was reconnecting just the transport
connection before. In case of multi-channel, this may not be
what we want to do after several timeouts. Changed that to
reconnect the session and the tree too.
Also renamed NUM_STATUS_IO_TIMEOUT to a more appropriate name
MAX_STATUS_IO_TIMEOUT.
Fixes: 8e670f77c4a5 ("Handle STATUS_IO_TIMEOUT gracefully")
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Currently, is_network_name_deleted and it's implementations
do not return anything if the network name did get deleted.
So the function doesn't fully achieve what it advertizes.
Changed the function to return a bool instead. It will now
return true if the error returned is STATUS_NETWORK_NAME_DELETED
and the share (tree id) was found to be connected. It returns
false otherwise.
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
Acked-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
We switch session state to SES_EXITING without cifs_tcp_ses_lock now,
it may lead to potential use-after-free issue.
Consider the following execution processes:
Thread 1:
__cifs_put_smb_ses()
spin_lock(&cifs_tcp_ses_lock)
if (--ses->ses_count > 0)
spin_unlock(&cifs_tcp_ses_lock)
return
spin_unlock(&cifs_tcp_ses_lock)
---> **GAP**
spin_lock(&ses->ses_lock)
if (ses->ses_status == SES_GOOD)
ses->ses_status = SES_EXITING
spin_unlock(&ses->ses_lock)
Thread 2:
cifs_find_smb_ses()
spin_lock(&cifs_tcp_ses_lock)
list_for_each_entry(ses, ...)
spin_lock(&ses->ses_lock)
if (ses->ses_status == SES_EXITING)
spin_unlock(&ses->ses_lock)
continue
...
spin_unlock(&ses->ses_lock)
if (ret)
cifs_smb_ses_inc_refcount(ret)
spin_unlock(&cifs_tcp_ses_lock)
If thread 1 is preempted in the gap and thread 2 start executing, thread 2
will get the session, and soon thread 1 will switch the session state to
SES_EXITING and start releasing it, even though thread 1 had increased the
session's refcount and still uses it.
So switch session state under cifs_tcp_ses_lock to eliminate this gap.
Signed-off-by: Winston Wen <wentao@uniontech.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
When having two DFS root mounts that are connected to same namespace,
same mount options but different prefix paths, we can't really use the
shared @server->origin_fullpath when chasing DFS links in them.
Move the origin_fullpath field to cifs_tcon structure so when having
shared DFS root mounts with different prefix paths, and we need to
chase any DFS links, dfs_get_automount_devname() will pick up the
correct full path out of the @tcon that will be used for the new
mount.
Before patch
mount.cifs //dom/dfs/dir /mnt/1 -o ...
mount.cifs //dom/dfs /mnt/2 -o ...
# shared server, ses, tcon
# server: origin_fullpath=//dom/dfs/dir
# @server->origin_fullpath + '/dir/link1'
$ ls /mnt/2/dir/link1
ls: cannot open directory '/mnt/2/dir/link1': No such file or directory
After patch
mount.cifs //dom/dfs/dir /mnt/1 -o ...
mount.cifs //dom/dfs /mnt/2 -o ...
# shared server & ses
# tcon_1: origin_fullpath=//dom/dfs/dir
# tcon_2: origin_fullpath=//dom/dfs
# @tcon_2->origin_fullpath + '/dir/link1'
$ ls /mnt/2/dir/link1
dir0 dir1 dir10 dir3 dir5 dir6 dir7 dir9 target2_file.txt tsub
Fixes: 8e3554150d6c ("cifs: fix sharing of DFS connections")
Signed-off-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
This fixes the following warning reported by kernel test robot
fs/smb/client/connect.c:2974 generic_ip_connect() error: we
previously assumed 'socket' could be null (see line 2962)
Link: https://lore.kernel.org/all/202306170124.CtQqzf0I-lkp@intel.com/
Signed-off-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Fix potential dereference of ERR_PTR @tlink as reported by kernel test
robot
fs/smb/client/connect.c:2775 cifs_match_super() error: 'tlink'
dereferencing possible ERR_PTR()
Link: https://lore.kernel.org/all/202306170124.CtQqzf0I-lkp@intel.com/
Signed-off-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
iface_cmp used to simply do a memcmp of the two
provided struct sockaddrs. The comparison needs to do more
based on the address family. Similar logic was already
present in cifs_match_ipaddr. Doing something similar now.
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
The ordering of status checks at the beginning of
cifs_tree_connect is wrong. As a result, a tcon
which is good may stay marked as needing reconnect
infinitely.
Fixes: 2f0e4f034220 ("cifs: check only tcon status on tcon related functions")
Cc: stable@vger.kernel.org # 6.3
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Move CIFS/SMB3 related client and server files (cifs.ko and ksmbd.ko
and helper modules) to new fs/smb subdirectory:
fs/cifs --> fs/smb/client
fs/ksmbd --> fs/smb/server
fs/smbfs_common --> fs/smb/common
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|