linux-stable.git/drivers/net/netconsole.c, branch linux-6.11.y

net: netconsole: fix wrong warning

2024-10-17T13:26:51+00:00

[ Upstream commit d94785bb46b6167382b1de3290eccc91fa98df53 ]

A warning is triggered when there is insufficient space in the buffer
for userdata. However, this is not an issue since userdata will be sent
in the next iteration.

Current warning message:

    ------------[ cut here ]------------
     WARNING: CPU: 13 PID: 3013042 at drivers/net/netconsole.c:1122 write_ext_msg+0x3b6/0x3d0
      ? write_ext_msg+0x3b6/0x3d0
      console_flush_all+0x1e9/0x330

The code incorrectly issues a warning when this_chunk is zero, which is
a valid scenario. The warning should only be triggered when this_chunk
is negative.

Fixes: 1ec9daf95093 ("net: netconsole: append userdata to fragmented netconsole messages")
Signed-off-by: Breno Leitao 
Reviewed-by: Simon Horman 
Link: https://patch.msgid.link/20241008094325.896208-1-leitao@debian.org
Signed-off-by: Jakub Kicinski 
Signed-off-by: Sasha Levin

Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

2024-07-15T20:19:17+00:00

Merge in late fixes to prepare for the 6.11 net-next PR.

Conflicts:
  93c3a96c301f ("net: pse-pd: Do not return EOPNOSUPP if config is null")
  4cddb0f15ea9 ("net: ethtool: pse-pd: Fix possible null-deref")
  30d7b6727724 ("net: ethtool: Add new power limit get and set features")
https://lore.kernel.org/20240715123204.623520bb@canb.auug.org.au/

Signed-off-by: Jakub Kicinski

net: netconsole: Disable target before netpoll cleanup

2024-07-14T14:38:44+00:00

Currently, netconsole cleans up the netpoll structure before disabling
the target. This approach can lead to race conditions, as message
senders (write_ext_msg() and write_msg()) check if the target is
enabled before using netpoll. The sender can validate that the target is
enabled, but, the netpoll might be de-allocated already, causing
undesired behaviours.

This patch reverses the order of operations:
1. Disable the target
2. Clean up the netpoll structure

This change eliminates the potential race condition, ensuring that
no messages are sent through a partially cleaned-up netpoll structure.

Fixes: 2382b15bcc39 ("netconsole: take care of NETDEV_UNREGISTER event")
Cc: stable@vger.kernel.org
Signed-off-by: Breno Leitao 
Reviewed-by: Eric Dumazet 
Link: https://patch.msgid.link/20240712143415.1141039-1-leitao@debian.org
Signed-off-by: Jakub Kicinski

net: netconsole: Eliminate redundant setting of enabled field

2024-07-12T01:32:48+00:00

When disabling a netconsole target, enabled_store() is called with
enabled=false. Currently, this results in updating the nt->enabled
field twice:

1. Inside the if/else block, with the target_list_lock spinlock held
2. Later, without the target_list_lock

This patch eliminates the redundancy by setting the field only once,
improving efficiency and reducing potential race conditions.

Signed-off-by: Breno Leitao 
Link: https://patch.msgid.link/20240709144403.544099-3-leitao@debian.org
Signed-off-by: Jakub Kicinski

net: netconsole: Remove unnecessary cast from bool

2024-07-12T01:32:48+00:00

The 'enabled' variable is already a bool, so casting it to its value
is redundant.

Remove the superfluous cast, improving code clarity without changing
functionality.

Signed-off-by: Breno Leitao 
Link: https://patch.msgid.link/20240709144403.544099-2-leitao@debian.org
Signed-off-by: Jakub Kicinski

netconsole: Do not shutdown dynamic configuration if cmdline is invalid

2024-05-30T00:19:50+00:00

If a user provides an invalid netconsole configuration during boot time
(e.g., specifying an invalid ethX interface), netconsole will be
entirely disabled. Consequently, the user won't be able to create new
entries in /sys/kernel/config/netconsole/ as that directory does not
exist.

Apart from misconfiguration, another issue arises when ethX is loaded as
a module and the netconsole= line in the command line points to ethX,
resulting in an obvious failure. This renders netconsole unusable, as
/sys/kernel/config/netconsole/ will never appear. This is more annoying
since users reconfigure (or just toggle) the configuratin later (see
commit 5fbd6cdbe304b ("netconsole: Attach cmdline target to dynamic
target"))

Create /sys/kernel/config/netconsole/ even if the command line arguments
are invalid, so, users can create dynamic entries in netconsole.

Reported-by: Aijay Adams 
Signed-off-by: Breno Leitao 
Link: https://lore.kernel.org/r/20240528084225.3215853-1-leitao@debian.org
Signed-off-by: Jakub Kicinski

net: netconsole: Add continuation line prefix to userdata messages

2024-03-11T21:07:57+00:00

Add a space (' ') prefix to every userdata line to match docs for
dev-kmsg. To account for this extra character in each userdata entry,
reduce userdata entry names (directory name) from 54 characters to 53.

According to the dev-kmsg docs, a space is used for subsequent lines to
mark them as continuation lines.

> A line starting with ' ', is a continuation line, adding
> key/value pairs to the log message, which provide the machine
> readable context of the message, for reliable processing in
> userspace.

Testing for this patch::

 cd /sys/kernel/config/netconsole && mkdir cmdline0
 cd cmdline0
 mkdir userdata/test && echo "hello" > userdata/test/value
 mkdir userdata/test2 && echo "hello2" > userdata/test2/value
 echo "message" > /dev/kmsg

Outputs::

 6.8.0-rc5-virtme,12,493,231373579,-;message
  test=hello
  test2=hello2

And I confirmed all testing works as expected from the original patchset

Fixes: df03f830d099 ("net: netconsole: cache userdata formatted string in netconsole_target")
Signed-off-by: Matthew Wood 
Reviewed-by: Breno Leitao 
Link: https://lore.kernel.org/r/20240308002525.248672-1-thepacketgeek@gmail.com
Signed-off-by: Jakub Kicinski

net: netconsole: append userdata to fragmented netconsole messages

2024-02-09T10:23:46+00:00

Regardless of whether the original message body or formatted userdata
exceeds the MAX_PRINT_CHUNK, append userdata to the netconsole message
starting with the first chunk that has available space after writing the
body.

Co-developed-by: Breno Leitao 
Signed-off-by: Breno Leitao 
Signed-off-by: Matthew Wood 
Signed-off-by: David S. Miller

net: netconsole: append userdata to netconsole messages

2024-02-09T10:23:46+00:00

Append userdata to outgoing unfragmented (<1000 bytes) netconsole messages.
When sending messages the userdata string is already formatted and stored
in netconsole_target->userdata_complete.

Always write the outgoing message to buf, so userdata can be appended in
a standard fashion. This is a change from only using buf when the
release needs to be prepended to the message.

Co-developed-by: Breno Leitao 
Signed-off-by: Breno Leitao 
Signed-off-by: Matthew Wood 
Signed-off-by: David S. Miller

net: netconsole: cache userdata formatted string in netconsole_target

2024-02-09T10:23:45+00:00

Store a formatted string for userdata that will be appended to netconsole
messages. The string has a capacity of 4KB, as calculated by the userdatum
entry length of 256 bytes and a max of 16 userdata entries.

Update the stored netconsole_target->userdata_complete string with the new
formatted userdata values when a userdatum is created, edited, or
removed. Each userdata entry contains a trailing newline, which will be
formatted as such in netconsole messages::

    6.7.0-rc8-virtme,12,500,1646292204,-;test
    release=foo
    something=bar
    6.7.0-rc8-virtme,12,500,1646292204,-;another test
    release=foo
    something=bar

Enforcement of MAX_USERDATA_ITEMS is done in userdatum_make_item;
update_userdata will not check for this case but will skip any userdata
children over the limit of MAX_USERDATA_ITEMs.

If a userdata entry/dir is created but no value is provided, that entry
will be skipped. This is in part because update_userdata() can't be
called in userdatum_make_item() since the item will not have been added
to the userdata config_group children yet. To preserve the experience of
adding an empty userdata that doesn't show up in the netconsole
messages, purposefully skip empty userdata items even when
update_userdata() can be called.

Co-developed-by: Breno Leitao 
Signed-off-by: Breno Leitao 
Signed-off-by: Matthew Wood 
Signed-off-by: David S. Miller