instead of removing '..' in a given path, call
kern_path with LOOKUP_BENEATH flag to prevent
the out of share access.

ran various test on this:
smb2-cat-async smb://127.0.0.1/homes/../out_of_share
smb2-cat-async smb://127.0.0.1/homes/foo/../../out_of_share
smbclient //127.0.0.1/homes -c "mkdir ../foo2"
smbclient //127.0.0.1/homes -c "rename bar ../bar"

Cc: Ronnie Sahlberg <ronniesahlberg@gmail.com>
Cc: Ralph Boehme <slow@samba.org>
Tested-by: NSteve French <smfrench@gmail.com>
Tested-by: NNamjae Jeon <linkinjeon@kernel.org>
Acked-by: NNamjae Jeon <linkinjeon@kernel.org>
Signed-off-by: NHyunchul Lee <hyc.lee@gmail.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

Use  LOOKUP_NO_SYMLINKS flags for default lookup to prohibit the middle of
symlink component lookup and remove follow symlinks parameter support.
We re-implement it as reparse point later.

Test result:
smbclient -Ulinkinjeon%1234 //172.30.1.42/share -c
"get hacked/passwd passwd"
NT_STATUS_OBJECT_NAME_NOT_FOUND opening remote file \hacked\passwd

Cc: Ralph Böhme <slow@samba.org>
Cc: Steve French <smfrench@gmail.com>
Acked-by: NRonnie Sahlberg <lsahlber@redhat.com>
Signed-off-by: NNamjae Jeon <linkinjeon@kernel.org>
Signed-off-by: NSteve French <stfrench@microsoft.com>

When second smb2 pdu has invalid protocol id, ksmbd doesn't detect it
and allow to process smb2 request. This patch add the check it in
ksmbd_verify_smb_message() and don't use protocol id of smb2 request as
protocol id of response.
Reviewed-by: NRonnie Sahlberg <ronniesahlberg@gmail.com>
Reviewed-by: NRalph Böhme <slow@samba.org>
Reported-by: NRonnie Sahlberg <lsahlber@redhat.com>
Signed-off-by: NNamjae Jeon <linkinjeon@kernel.org>
Signed-off-by: NSteve French <stfrench@microsoft.com>

Windows client expect to get default stream name(::DATA) in
FILE_STREAM_INFORMATION response even if there is no stream data in file.
This patch fix update failure when writing ppt or doc files.
Signed-off-by: NNamjae Jeon <linkinjeon@kernel.org>
Reviewed-By: NTom Talpey <tom@talpey.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

Add validation to check whether req->InputBufferLength is smaller than
smb2_ea_info_req structure size.

Cc: Ronnie Sahlberg <ronniesahlberg@gmail.com>
Cc: Ralph Böhme <slow@samba.org>
Cc: Steve French <smfrench@gmail.com>
Signed-off-by: NNamjae Jeon <linkinjeon@kernel.org>
Signed-off-by: NSteve French <stfrench@microsoft.com>

Because of .., files outside the share directory
could be accessed. To prevent this, normalize
the given path and remove all . and ..
components.

In addition to the usual large set of regression tests (smbtorture
and xfstests), ran various tests on this to specifically check
path name validation including libsmb2 tests to verify path
normalization:

 ./examples/smb2-ls-async smb://172.30.1.15/homes2/../
 ./examples/smb2-ls-async smb://172.30.1.15/homes2/foo/../
 ./examples/smb2-ls-async smb://172.30.1.15/homes2/foo/../../
 ./examples/smb2-ls-async smb://172.30.1.15/homes2/foo/../
 ./examples/smb2-ls-async smb://172.30.1.15/homes2/foo/..bar/
 ./examples/smb2-ls-async smb://172.30.1.15/homes2/foo/bar../
 ./examples/smb2-ls-async smb://172.30.1.15/homes2/foo/bar..
 ./examples/smb2-ls-async smb://172.30.1.15/homes2/foo/bar../../../../Signed-off-by: NHyunchul Lee <hyc.lee@gmail.com>
Signed-off-by: NNamjae Jeon <linkinjeon@kernel.org>
Signed-off-by: NSteve French <stfrench@microsoft.com>

Addresses-Coverity reported Uninitialized variables warninig :

/fs/ksmbd/smb2pdu.c: 5525 in set_file_basic_info()
5519                    if (!rc) {
5520                            inode->i_ctime = ctime;
5521                            mark_inode_dirty(inode);
5522                    }
5523                    inode_unlock(inode);
5524            }
>>>     CID 1506805:  Uninitialized variables  (UNINIT)
>>>     Using uninitialized value "rc".
5525            return rc;
5526     }
5527
5528     static int set_file_allocation_info(struct ksmbd_work *work,
5529                                 struct ksmbd_file *fp, char *buf)
5530     {

Addresses-Coverity: ("Uninitialized variable")
Signed-off-by: NNamjae Jeon <linkinjeon@kernel.org>
Signed-off-by: NSteve French <stfrench@microsoft.com>

This log happens on servers with a network bridge since
the bridge does not have a specified link speed.
This is not a real error so change the error log to debug instead.
Signed-off-by: NPer Forlin <perfn@axis.com>
Signed-off-by: NNamjae Jeon <linkinjeon@kernel.org>
Signed-off-by: NSteve French <stfrench@microsoft.com>

Permission checking and copying over ownership information is the task
of the underlying filesystem not ksmbd. The order is also wrong here.
This modifies the inode before notify_change(). If notify_change() fails
this will have changed ownership nonetheless. All of this is unnecessary
though since the underlying filesystem's ->setattr handler will do all
this (if required) by itself.

Cc: Steve French <stfrench@microsoft.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Namjae Jeon <namjae.jeon@samsung.com>
Cc: Hyunchul Lee <hyc.lee@gmail.com>
Cc: Sergey Senozhatsky <senozhatsky@chromium.org>
Cc: linux-cifs@vger.kernel.org
Signed-off-by: NChristian Brauner <christian.brauner@ubuntu.com>
Signed-off-by: NNamjae Jeon <linkinjeon@kernel.org>
Signed-off-by: NSteve French <stfrench@microsoft.com>

It seems the error was accidently ignored until now. Make sure it is
surfaced.

Cc: Steve French <stfrench@microsoft.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Namjae Jeon <namjae.jeon@samsung.com>
Cc: Hyunchul Lee <hyc.lee@gmail.com>
Cc: Sergey Senozhatsky <senozhatsky@chromium.org>
Cc: linux-cifs@vger.kernel.org
Signed-off-by: NChristian Brauner <christian.brauner@ubuntu.com>
Signed-off-by: NNamjae Jeon <linkinjeon@kernel.org>
Signed-off-by: NSteve French <stfrench@microsoft.com>

When creating new filesystem objects ksmbd translates between k*ids and
s*ids. For this it often uses struct smb_fattr and stashes the k*ids in
cf_uid and cf_gid. Let cf_uid and cf_gid always contain the final
information taking any potential idmapped mounts into account. When
finally translation cf_*id into s*ids translate them into the user
namespace of ksmbd since that is the relevant user namespace here.

Cc: Steve French <stfrench@microsoft.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Namjae Jeon <namjae.jeon@samsung.com>
Cc: Hyunchul Lee <hyc.lee@gmail.com>
Cc: Sergey Senozhatsky <senozhatsky@chromium.org>
Cc: linux-cifs@vger.kernel.org
Signed-off-by: NChristian Brauner <christian.brauner@ubuntu.com>
Signed-off-by: NNamjae Jeon <linkinjeon@kernel.org>
Signed-off-by: NSteve French <stfrench@microsoft.com>

When transferring ownership information to the
client the k*ids are translated into raw *ids before they are sent over
the wire. The function currently erroneously translates the k*ids
according to the mount's idmapping. Instead, reporting the owning *ids
to userspace the underlying k*ids need to be mapped up in the caller's
user namespace. This is how stat() works.
The caller in this instance is ksmbd itself and ksmbd always runs in the
initial user namespace. Translate according to that.

The idmapping of the mount is already taken into account by the lower
filesystem and so kstat->*id will contain the mapped k*ids.

Switch to from_k*id_munged() which ensures that the overflow*id is
returned instead of the (*id_t)-1 when the k*id can't be translated.

Cc: Steve French <stfrench@microsoft.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Namjae Jeon <namjae.jeon@samsung.com>
Cc: Hyunchul Lee <hyc.lee@gmail.com>
Cc: Sergey Senozhatsky <senozhatsky@chromium.org>
Cc: linux-cifs@vger.kernel.org
Signed-off-by: NChristian Brauner <christian.brauner@ubuntu.com>
Signed-off-by: NNamjae Jeon <linkinjeon@kernel.org>
Signed-off-by: NSteve French <stfrench@microsoft.com>

It's great that the new in-kernel ksmbd server will support idmapped
mounts out of the box! However, lookup is currently broken. Lookup
helpers such as lookup_one_len() call inode_permission() internally to
ensure that the caller is privileged over the inode of the base dentry
they are trying to lookup under. So the permission checking here is
currently wrong.

Linux v5.15 will gain a new lookup helper lookup_one() that does take
idmappings into account. I've added it as part of my patch series to
make btrfs support idmapped mounts. The new helper is in linux-next as
part of David's (Sterba) btrfs for-next branch as commit
c972214c133b ("namei: add mapping aware lookup helper").

I've said it before during one of my first reviews: I would very much
recommend adding fstests to [1]. It already seems to have very
rudimentary cifs support. There is a completely generic idmapped mount
testsuite that supports idmapped mounts.

[1]: https://git.kernel.org/pub/scm/fs/xfs/xfsprogs-dev.git/
Cc: Colin Ian King <colin.king@canonical.com>
Cc: Steve French <stfrench@microsoft.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Namjae Jeon <namjae.jeon@samsung.com>
Cc: Hyunchul Lee <hyc.lee@gmail.com>
Cc: Sergey Senozhatsky <senozhatsky@chromium.org>
Cc: David Sterba <dsterba@suse.com>
Cc: linux-cifs@vger.kernel.org
Signed-off-by: NChristian Brauner <christian.brauner@ubuntu.com>
Signed-off-by: NNamjae Jeon <linkinjeon@kernel.org>
Signed-off-by: NSteve French <stfrench@microsoft.com>

When commanding chmod and chown on cifs&ksmbd, ksmbd allows it without file
permissions check. There is code to check it in settattr_prepare.
Instead of setting the inode directly, update the mode and uid/gid
through notify_change.
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

ksmbd is forcing to turn on FS_POSIX_ACL in Kconfig to use vfs acl
functions(posix_acl_alloc, get_acl, set_posix_acl). OpenWRT and other
platform doesn't use acl and this config is disable by default in
kernel. This patch use IS_ENABLED() to know acl config is enable and use
acl function if it is enable.
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

Use proper errno instead of -1 in smb2_get_ksmbd_tcon().
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

Update the comment for smb2_get_ksmbd_tcon().
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

Change data type of function that return only 0 or 1 to boolean.
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

If smb2_get_name() then name is an error pointer.  In the clean up
code, we try to kfree() it and that will lead to an Oops.  Set it to
NULL instead.
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

ksmbd create socket with IPv6 to listen both IPv4 and IPv6 connection
from client. Server should send IP addresses of NICs through network
interface info response. If Client connection is IPv4, Server should
fill IPv4 address in response buffer. But ss_family is always PF_INET6
on IPv6 socket. So This patch add ipv6_addr_v4mapped check to know
client connection is IPv4.
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

Dan report a warning that is missing error code in smb2_lock
from static checker. This patch add error code to avoid static checker
warning.
Reported-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

Windows client disconnect connection by wrong signed SMB2 session
setup response on SMB3 multichannel mode.
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

ksmbd does not support RSS mode stably.
RSS mode enabling will be set later.
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

Both Windows 10's SMB server and samba return STATUS_OBJECT_PATH_NOT_FOUND
when trying to access a nonexistent path.

This fixes Windows 10 File History tool. The latter relies on the server
returning STATUS_OBJECT_PATH_NOT_FOUND to figure out what part of the
target path needs to be created. Returning STATUS_OBJECT_NAME_INVALID
will make it stop and display an error to the user.
Signed-off-by: NMarios Makassikis <mmakassikis@freebox.fr>
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

Support for faster packet signing (using GMAC instead of CMAC) can
now be negotiated to some newer servers, including Windows.
See MS-SMB2 section 2.2.3.17.

This patch adds support for sending the new negotiate context with two
supported signing algorithms(AES-CMAC, HMAC-SHA256).
If client add support for AES_GMAC, Server will be supported later
depend on it.
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

This patch add negotiate context verification code to check bounds.
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

When user insert wrong password, ksmbd return STATUS_INVALID_PARAMETER
error status to client. It will make user confusing whether it is not
password problem. This patch change error status to
STATUS_LOGON_FAILURE. and return STATUS_INSUFFICIENT_RESOURCES if memory
allocation failed on session setup.
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

Use smb2_compression_ctx instead of smb2_encryption_neg_context.
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

set RDMA capability for
FSCTL_QUERY_NETWORK_INTERFACE_INFO.
Signed-off-by: NHyunchul Lee <hyc.lee@gmail.com>
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

CID 1502845 (#1 of 1): Unused value (UNUSED_VALUE)
value_overwrite: Overwriting previous write to err with value from
vfs_lock_file(filp, 0U, rlock, NULL).
6880                err = vfs_lock_file(filp, 0, rlock, NULL);
6881                if (err)
6882                        pr_err("rollback unlock fail : %d\n", err);
Reported-by: NCoverity Scan <scan-admin@coverity.com>
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

Coverity Scan seems to report false alarm.

*** CID 1505930:    (USE_AFTER_FREE)
/fs/ksmbd/smb2pdu.c: 2527 in smb2_open()
>>> CID 1505930:    (USE_AFTER_FREE)
>>> Passing freed pointer "context" as an argument to
>>> "check_context_err".

This patch remove unneeded check_context_err to make coverity scan
happy.
Reported-by: NCoverity Scan <scan-admin@coverity.com>
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

For code cleanup, handle error cases first in
smb2_create_sd_buffers().
Reported-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NHyunchul Lee <hyc.lee@gmail.com>
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

instead of -ENOENT, make smb2_find_context_vals
return NULL if the given context cannot be found.
Reported-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NHyunchul Lee <hyc.lee@gmail.com>
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

the wait can be canceled by SMB2_CANCEL, SMB2_CLOSE,
SMB2_LOGOFF, disconnection or shutdown, we don't have
to use wait_event_interruptible.

And this remove the warning from Coverity:

CID 1502834 (#1 of 1): Unused value (UNUSED_VALUE)
returned_value: Assigning value from ksmbd_vfs_posix_lock_wait(flock)
to err here, but that stored value is overwritten before it can be used.
Signed-off-by: NHyunchul Lee <hyc.lee@gmail.com>
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

Append ksmbd_lock into the connection's
lock list and the ksmbd_file's lock list.
And when a file is closed, detach ksmbd_lock
from these lists and free it.
Signed-off-by: NHyunchul Lee <hyc.lee@gmail.com>
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

These lines are intended one tab too far.
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

This patch change data type of volatile/persistent id to u64 to make
issue from idr_find and idr_remove(). !HAS_FILE_ID check will protect
integer overflow issue from idr_find and idr_remove().
Reported-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

Add missing kfree(conv_name) on error path.
Reported-by: NCoverity Scan <scan-admin@coverity.com>
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

netdev can never be NULL in for_each_netdev loop.
This patch remove unneeded NULL check.
Reported-by: NCoverity Scan <scan-admin@coverity.com>
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>

There is a error handling case that passes control to label out_err
without pointer sess being assigned a value. The unassigned pointer
may be any garbage value and so the test of rc < 0 && sess maybe
true leading to sess being passed to the call to ksmbd_session_destroy.
Fix this by setting sess to NULL in this corner case.

Addresses-Coverity: ("Uninitialized pointer read")
Signed-off-by: NColin Ian King <colin.king@canonical.com>
Signed-off-by: NNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: NSteve French <stfrench@microsoft.com>