<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux-toradex.git/fs/smb/server/server.c, branch v6.14-rc6</title>
<subtitle>Linux kernel for Apalis and Colibri modules</subtitle>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/'/>
<entry>
<title>ksmbd: fix broken transfers when exceeding max simultaneous operations</title>
<updated>2024-12-16T04:20:03+00:00</updated>
<author>
<name>Marios Makassikis</name>
<email>mmakassikis@freebox.fr</email>
</author>
<published>2024-12-14T03:17:23+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=43fb7bce8866e793275c4f9f25af6a37745f3416'/>
<id>43fb7bce8866e793275c4f9f25af6a37745f3416</id>
<content type='text'>
Since commit 0a77d947f599 ("ksmbd: check outstanding simultaneous SMB
operations"), ksmbd enforces a maximum number of simultaneous operations
for a connection. The problem is that reaching the limit causes ksmbd to
close the socket, and the client has no indication that it should have
slowed down.

This behaviour can be reproduced by setting "smb2 max credits = 128" (or
lower), and transferring a large file (25GB).

smbclient fails as below:

  $ smbclient //192.168.1.254/testshare -U user%pass
  smb: \&gt; put file.bin
  cli_push returned NT_STATUS_USER_SESSION_DELETED
  putting file file.bin as \file.bin smb2cli_req_compound_submit:
  Insufficient credits. 0 available, 1 needed
  NT_STATUS_INTERNAL_ERROR closing remote file \file.bin
  smb: \&gt; smb2cli_req_compound_submit: Insufficient credits. 0 available,
  1 needed

Windows clients fail with 0x8007003b (with smaller files even).

Fix this by delaying reading from the socket until there's room to
allocate a request. This effectively applies backpressure on the client,
so the transfer completes, albeit at a slower rate.

Fixes: 0a77d947f599 ("ksmbd: check outstanding simultaneous SMB operations")
Signed-off-by: Marios Makassikis &lt;mmakassikis@freebox.fr&gt;
Signed-off-by: Namjae Jeon &lt;linkinjeon@kernel.org&gt;
Signed-off-by: Steve French &lt;stfrench@microsoft.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Since commit 0a77d947f599 ("ksmbd: check outstanding simultaneous SMB
operations"), ksmbd enforces a maximum number of simultaneous operations
for a connection. The problem is that reaching the limit causes ksmbd to
close the socket, and the client has no indication that it should have
slowed down.

This behaviour can be reproduced by setting "smb2 max credits = 128" (or
lower), and transferring a large file (25GB).

smbclient fails as below:

  $ smbclient //192.168.1.254/testshare -U user%pass
  smb: \&gt; put file.bin
  cli_push returned NT_STATUS_USER_SESSION_DELETED
  putting file file.bin as \file.bin smb2cli_req_compound_submit:
  Insufficient credits. 0 available, 1 needed
  NT_STATUS_INTERNAL_ERROR closing remote file \file.bin
  smb: \&gt; smb2cli_req_compound_submit: Insufficient credits. 0 available,
  1 needed

Windows clients fail with 0x8007003b (with smaller files even).

Fix this by delaying reading from the socket until there's room to
allocate a request. This effectively applies backpressure on the client,
so the transfer completes, albeit at a slower rate.

Fixes: 0a77d947f599 ("ksmbd: check outstanding simultaneous SMB operations")
Signed-off-by: Marios Makassikis &lt;mmakassikis@freebox.fr&gt;
Signed-off-by: Namjae Jeon &lt;linkinjeon@kernel.org&gt;
Signed-off-by: Steve French &lt;stfrench@microsoft.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>ksmbd: fix racy issue from session lookup and expire</title>
<updated>2024-12-10T23:48:06+00:00</updated>
<author>
<name>Namjae Jeon</name>
<email>linkinjeon@kernel.org</email>
</author>
<published>2024-12-05T12:38:47+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=b95629435b84b9ecc0c765995204a4d8a913ed52'/>
<id>b95629435b84b9ecc0c765995204a4d8a913ed52</id>
<content type='text'>
Increment the session reference count within the lock for lookup to avoid
racy issue with session expire.

Cc: stable@vger.kernel.org
Reported-by: zdi-disclosures@trendmicro.com # ZDI-CAN-25737
Signed-off-by: Namjae Jeon &lt;linkinjeon@kernel.org&gt;
Signed-off-by: Steve French &lt;stfrench@microsoft.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Increment the session reference count within the lock for lookup to avoid
racy issue with session expire.

Cc: stable@vger.kernel.org
Reported-by: zdi-disclosures@trendmicro.com # ZDI-CAN-25737
Signed-off-by: Namjae Jeon &lt;linkinjeon@kernel.org&gt;
Signed-off-by: Steve French &lt;stfrench@microsoft.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>ksmbd: fix use-after-free in SMB request handling</title>
<updated>2024-11-26T00:58:27+00:00</updated>
<author>
<name>Yunseong Kim</name>
<email>yskelg@gmail.com</email>
</author>
<published>2024-11-25T07:45:55+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=9a8c5d89d327ff58e9b2517f8a6afb4181d32c6e'/>
<id>9a8c5d89d327ff58e9b2517f8a6afb4181d32c6e</id>
<content type='text'>
A race condition exists between SMB request handling in
`ksmbd_conn_handler_loop()` and the freeing of `ksmbd_conn` in the
workqueue handler `handle_ksmbd_work()`. This leads to a UAF.
- KASAN: slab-use-after-free Read in handle_ksmbd_work
- KASAN: slab-use-after-free in rtlock_slowlock_locked

This race condition arises as follows:
- `ksmbd_conn_handler_loop()` waits for `conn-&gt;r_count` to reach zero:
  `wait_event(conn-&gt;r_count_q, atomic_read(&amp;conn-&gt;r_count) == 0);`
- Meanwhile, `handle_ksmbd_work()` decrements `conn-&gt;r_count` using
  `atomic_dec_return(&amp;conn-&gt;r_count)`, and if it reaches zero, calls
  `ksmbd_conn_free()`, which frees `conn`.
- However, after `handle_ksmbd_work()` decrements `conn-&gt;r_count`,
  it may still access `conn-&gt;r_count_q` in the following line:
  `waitqueue_active(&amp;conn-&gt;r_count_q)` or `wake_up(&amp;conn-&gt;r_count_q)`
  This results in a UAF, as `conn` has already been freed.

The discovery of this UAF can be referenced in the following PR for
syzkaller's support for SMB requests.
Link: https://github.com/google/syzkaller/pull/5524

Fixes: ee426bfb9d09 ("ksmbd: add refcnt to ksmbd_conn struct")
Cc: linux-cifs@vger.kernel.org
Cc: stable@vger.kernel.org # v6.6.55+, v6.10.14+, v6.11.3+
Cc: syzkaller@googlegroups.com
Signed-off-by: Yunseong Kim &lt;yskelg@gmail.com&gt;
Acked-by: Namjae Jeon &lt;linkinjeon@kernel.org&gt;
Signed-off-by: Steve French &lt;stfrench@microsoft.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
A race condition exists between SMB request handling in
`ksmbd_conn_handler_loop()` and the freeing of `ksmbd_conn` in the
workqueue handler `handle_ksmbd_work()`. This leads to a UAF.
- KASAN: slab-use-after-free Read in handle_ksmbd_work
- KASAN: slab-use-after-free in rtlock_slowlock_locked

This race condition arises as follows:
- `ksmbd_conn_handler_loop()` waits for `conn-&gt;r_count` to reach zero:
  `wait_event(conn-&gt;r_count_q, atomic_read(&amp;conn-&gt;r_count) == 0);`
- Meanwhile, `handle_ksmbd_work()` decrements `conn-&gt;r_count` using
  `atomic_dec_return(&amp;conn-&gt;r_count)`, and if it reaches zero, calls
  `ksmbd_conn_free()`, which frees `conn`.
- However, after `handle_ksmbd_work()` decrements `conn-&gt;r_count`,
  it may still access `conn-&gt;r_count_q` in the following line:
  `waitqueue_active(&amp;conn-&gt;r_count_q)` or `wake_up(&amp;conn-&gt;r_count_q)`
  This results in a UAF, as `conn` has already been freed.

The discovery of this UAF can be referenced in the following PR for
syzkaller's support for SMB requests.
Link: https://github.com/google/syzkaller/pull/5524

Fixes: ee426bfb9d09 ("ksmbd: add refcnt to ksmbd_conn struct")
Cc: linux-cifs@vger.kernel.org
Cc: stable@vger.kernel.org # v6.6.55+, v6.10.14+, v6.11.3+
Cc: syzkaller@googlegroups.com
Signed-off-by: Yunseong Kim &lt;yskelg@gmail.com&gt;
Acked-by: Namjae Jeon &lt;linkinjeon@kernel.org&gt;
Signed-off-by: Steve French &lt;stfrench@microsoft.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>ksmbd: use __GFP_RETRY_MAYFAIL</title>
<updated>2024-11-26T00:58:02+00:00</updated>
<author>
<name>Namjae Jeon</name>
<email>linkinjeon@kernel.org</email>
</author>
<published>2024-11-22T07:33:25+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=0066f623bce8f98b69b752ee03d46a5047c281b8'/>
<id>0066f623bce8f98b69b752ee03d46a5047c281b8</id>
<content type='text'>
Prefer to report ENOMEM rather than incur the oom for allocations in
ksmbd. __GFP_NORETRY could not achieve that, It would fail the allocations
just too easily. __GFP_RETRY_MAYFAIL will keep retrying the allocation
until there is no more progress and fail the allocation instead go OOM
and let the caller to deal with it.

Signed-off-by: Namjae Jeon &lt;linkinjeon@kernel.org&gt;
Signed-off-by: Steve French &lt;stfrench@microsoft.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Prefer to report ENOMEM rather than incur the oom for allocations in
ksmbd. __GFP_NORETRY could not achieve that, It would fail the allocations
just too easily. __GFP_RETRY_MAYFAIL will keep retrying the allocation
until there is no more progress and fail the allocation instead go OOM
and let the caller to deal with it.

Signed-off-by: Namjae Jeon &lt;linkinjeon@kernel.org&gt;
Signed-off-by: Steve French &lt;stfrench@microsoft.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>ksmbd: check outstanding simultaneous SMB operations</title>
<updated>2024-11-05T00:26:38+00:00</updated>
<author>
<name>Namjae Jeon</name>
<email>linkinjeon@kernel.org</email>
</author>
<published>2024-11-04T04:43:06+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=0a77d947f599b1f39065015bec99390d0c0022ee'/>
<id>0a77d947f599b1f39065015bec99390d0c0022ee</id>
<content type='text'>
If Client send simultaneous SMB operations to ksmbd, It exhausts too much
memory through the "ksmbd_work_cache”. It will cause OOM issue.
ksmbd has a credit mechanism but it can't handle this problem. This patch
add the check if it exceeds max credits to prevent this problem by assuming
that one smb request consumes at least one credit.

Cc: stable@vger.kernel.org # v5.15+
Reported-by: Norbert Szetei &lt;norbert@doyensec.com&gt;
Tested-by: Norbert Szetei &lt;norbert@doyensec.com&gt;
Signed-off-by: Namjae Jeon &lt;linkinjeon@kernel.org&gt;
Signed-off-by: Steve French &lt;stfrench@microsoft.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
If Client send simultaneous SMB operations to ksmbd, It exhausts too much
memory through the "ksmbd_work_cache”. It will cause OOM issue.
ksmbd has a credit mechanism but it can't handle this problem. This patch
add the check if it exceeds max credits to prevent this problem by assuming
that one smb request consumes at least one credit.

Cc: stable@vger.kernel.org # v5.15+
Reported-by: Norbert Szetei &lt;norbert@doyensec.com&gt;
Tested-by: Norbert Szetei &lt;norbert@doyensec.com&gt;
Signed-off-by: Namjae Jeon &lt;linkinjeon@kernel.org&gt;
Signed-off-by: Steve French &lt;stfrench@microsoft.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>ksmbd: fix slab-use-after-free in smb3_preauth_hash_rsp</title>
<updated>2024-11-05T00:26:37+00:00</updated>
<author>
<name>Namjae Jeon</name>
<email>linkinjeon@kernel.org</email>
</author>
<published>2024-11-04T04:40:41+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=b8fc56fbca7482c1e5c0e3351c6ae78982e25ada'/>
<id>b8fc56fbca7482c1e5c0e3351c6ae78982e25ada</id>
<content type='text'>
ksmbd_user_session_put should be called under smb3_preauth_hash_rsp().
It will avoid freeing session before calling smb3_preauth_hash_rsp().

Cc: stable@vger.kernel.org # v5.15+
Reported-by: Norbert Szetei &lt;norbert@doyensec.com&gt;
Tested-by: Norbert Szetei &lt;norbert@doyensec.com&gt;
Signed-off-by: Namjae Jeon &lt;linkinjeon@kernel.org&gt;
Signed-off-by: Steve French &lt;stfrench@microsoft.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
ksmbd_user_session_put should be called under smb3_preauth_hash_rsp().
It will avoid freeing session before calling smb3_preauth_hash_rsp().

Cc: stable@vger.kernel.org # v5.15+
Reported-by: Norbert Szetei &lt;norbert@doyensec.com&gt;
Tested-by: Norbert Szetei &lt;norbert@doyensec.com&gt;
Signed-off-by: Namjae Jeon &lt;linkinjeon@kernel.org&gt;
Signed-off-by: Steve French &lt;stfrench@microsoft.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>ksmbd: fix user-after-free from session log off</title>
<updated>2024-10-10T02:23:17+00:00</updated>
<author>
<name>Namjae Jeon</name>
<email>linkinjeon@kernel.org</email>
</author>
<published>2024-10-08T13:42:57+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=7aa8804c0b67b3cb263a472d17f2cb50d7f1a930'/>
<id>7aa8804c0b67b3cb263a472d17f2cb50d7f1a930</id>
<content type='text'>
There is racy issue between smb2 session log off and smb2 session setup.
It will cause user-after-free from session log off.
This add session_lock when setting SMB2_SESSION_EXPIRED and referece
count to session struct not to free session while it is being used.

Cc: stable@vger.kernel.org # v5.15+
Reported-by: zdi-disclosures@trendmicro.com # ZDI-CAN-25282
Signed-off-by: Namjae Jeon &lt;linkinjeon@kernel.org&gt;
Signed-off-by: Steve French &lt;stfrench@microsoft.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
There is racy issue between smb2 session log off and smb2 session setup.
It will cause user-after-free from session log off.
This add session_lock when setting SMB2_SESSION_EXPIRED and referece
count to session struct not to free session while it is being used.

Cc: stable@vger.kernel.org # v5.15+
Reported-by: zdi-disclosures@trendmicro.com # ZDI-CAN-25282
Signed-off-by: Namjae Jeon &lt;linkinjeon@kernel.org&gt;
Signed-off-by: Steve French &lt;stfrench@microsoft.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>ksmbd: Correct typos in multiple comments across various files</title>
<updated>2024-09-26T02:33:22+00:00</updated>
<author>
<name>Shen Lichuan</name>
<email>shenlichuan@vivo.com</email>
</author>
<published>2024-09-25T07:43:23+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=9e676e571d39eb6189bf6d55a9c401ba2dd13410'/>
<id>9e676e571d39eb6189bf6d55a9c401ba2dd13410</id>
<content type='text'>
Fixed some confusing typos that were currently identified witch codespell,
the details are as follows:

-in the code comments:
fs/smb/common/smb2pdu.h:9: specfication ==&gt; specification
fs/smb/common/smb2pdu.h:494: usally ==&gt; usually
fs/smb/common/smb2pdu.h:1064: Attrubutes ==&gt; Attributes
fs/smb/server/connection.c:28: cleand ==&gt; cleaned
fs/smb/server/ksmbd_netlink.h:216: struture ==&gt; structure
fs/smb/server/oplock.c:799: conains ==&gt; contains
fs/smb/server/oplock.c:1487: containted ==&gt; contained
fs/smb/server/server.c:282: proccessing ==&gt; processing
fs/smb/server/smb_common.c:491: comforms ==&gt; conforms
fs/smb/server/xattr.h:102: ATTRIBUITE ==&gt; ATTRIBUTE

Signed-off-by: Shen Lichuan &lt;shenlichuan@vivo.com&gt;
Acked-by: Namjae Jeon &lt;linkinjeon@kernel.org&gt;
Signed-off-by: Steve French &lt;stfrench@microsoft.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Fixed some confusing typos that were currently identified witch codespell,
the details are as follows:

-in the code comments:
fs/smb/common/smb2pdu.h:9: specfication ==&gt; specification
fs/smb/common/smb2pdu.h:494: usally ==&gt; usually
fs/smb/common/smb2pdu.h:1064: Attrubutes ==&gt; Attributes
fs/smb/server/connection.c:28: cleand ==&gt; cleaned
fs/smb/server/ksmbd_netlink.h:216: struture ==&gt; structure
fs/smb/server/oplock.c:799: conains ==&gt; contains
fs/smb/server/oplock.c:1487: containted ==&gt; contained
fs/smb/server/server.c:282: proccessing ==&gt; processing
fs/smb/server/smb_common.c:491: comforms ==&gt; conforms
fs/smb/server/xattr.h:102: ATTRIBUITE ==&gt; ATTRIBUTE

Signed-off-by: Shen Lichuan &lt;shenlichuan@vivo.com&gt;
Acked-by: Namjae Jeon &lt;linkinjeon@kernel.org&gt;
Signed-off-by: Steve French &lt;stfrench@microsoft.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>smb: move SMB2 Status code to common header file</title>
<updated>2024-09-15T15:42:44+00:00</updated>
<author>
<name>ChenXiaoSong</name>
<email>chenxiaosong@kylinos.cn</email>
</author>
<published>2024-08-22T08:21:00+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=78181a5504a401e421e65d0257a33f904e0e7c29'/>
<id>78181a5504a401e421e65d0257a33f904e0e7c29</id>
<content type='text'>
There are only 4 different definitions between the client and server:

  - STATUS_SERVER_UNAVAILABLE: from client/smb2status.h
  - STATUS_FILE_NOT_AVAILABLE: from client/smb2status.h
  - STATUS_NO_PREAUTH_INTEGRITY_HASH_OVERLAP: from server/smbstatus.h
  - STATUS_INVALID_LOCK_RANGE: from server/smbstatus.h

Rename client/smb2status.h to common/smb2status.h, and merge the
2 different definitions of server to common header file.

Signed-off-by: ChenXiaoSong &lt;chenxiaosong@kylinos.cn&gt;
Acked-by: Namjae Jeon &lt;linkinjeon@kernel.org&gt;
Signed-off-by: Steve French &lt;stfrench@microsoft.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
There are only 4 different definitions between the client and server:

  - STATUS_SERVER_UNAVAILABLE: from client/smb2status.h
  - STATUS_FILE_NOT_AVAILABLE: from client/smb2status.h
  - STATUS_NO_PREAUTH_INTEGRITY_HASH_OVERLAP: from server/smbstatus.h
  - STATUS_INVALID_LOCK_RANGE: from server/smbstatus.h

Rename client/smb2status.h to common/smb2status.h, and merge the
2 different definitions of server to common header file.

Signed-off-by: ChenXiaoSong &lt;chenxiaosong@kylinos.cn&gt;
Acked-by: Namjae Jeon &lt;linkinjeon@kernel.org&gt;
Signed-off-by: Steve French &lt;stfrench@microsoft.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>ksmbd: add durable scavenger timer</title>
<updated>2024-07-15T02:45:40+00:00</updated>
<author>
<name>Namjae Jeon</name>
<email>linkinjeon@kernel.org</email>
</author>
<published>2024-07-14T23:29:39+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=d484d621d40f4a8b8959008802d79bef3609641b'/>
<id>d484d621d40f4a8b8959008802d79bef3609641b</id>
<content type='text'>
Launch ksmbd-durable-scavenger kernel thread to scan durable fps that
have not been reclaimed by a client within the configured time.

Signed-off-by: Namjae Jeon &lt;linkinjeon@kernel.org&gt;
Signed-off-by: Steve French &lt;stfrench@microsoft.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Launch ksmbd-durable-scavenger kernel thread to scan durable fps that
have not been reclaimed by a client within the configured time.

Signed-off-by: Namjae Jeon &lt;linkinjeon@kernel.org&gt;
Signed-off-by: Steve French &lt;stfrench@microsoft.com&gt;
</pre>
</div>
</content>
</entry>
</feed>
