Matthieu Baerts says:
====================
mptcp: various small and unrelated improvements
Here are various unrelated patches:
- Patch 1: sched: remove unused structure.
- Patch 2: sched: split the validation part, a preparation for later.
- Patch 3: pm: clarify code, not to think there is a possible UaF.
Note: a previous version has already been sent individually to Netdev.
- Patch 4: subflow: simplify subflow_hmac_valid by passing subflow_req.
- Patch 5: mib: add counter for MPJoin rejected by the PM.
- Patch 6: selftests: validate this new MPJoinRejected counter.
- Patch 7: selftests: define nlh variable only where needed.
- Patch 8: selftests: show how to use IPPROTO_MPTCP with getaddrinfo.
Note: a previous version has already been sent individually to Netdev.
v1: https://lore.kernel.org/20250411-net-next-mptcp-sched-mib-sft-misc-v1-0-85ac8c6654c3@kernel.org
====================
Link: https://patch.msgid.link/20250413-net-next-mptcp-sched-mib-sft-misc-v2-0-0f83a4350150@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
mptcp_connect.c is a startup tutorial of MPTCP programming, however
there is a lack of ai_protocol(IPPROTO_MPTCP) usage. Add comment for
getaddrinfo MPTCP support.
This patch first uses IPPROTO_MPTCP to get addrinfo, and if glibc
version is too old, it falls back to using IPPROTO_TCP.
Co-developed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250413-net-next-mptcp-sched-mib-sft-misc-v2-8-0f83a4350150@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
It's strange that 'nlh' variable is set to NULL in get_mptcpinfo() and then
this NULL pointer is passed to recv_nlmsg(). In fact, this variable should
be defined in recv_nlmsg(), not get_mptcpinfo().
So this patch drops this useless 'nlh' parameter of recv_nlmsg() and define
'nlh' variable in recv_nlmsg().
Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250413-net-next-mptcp-sched-mib-sft-misc-v2-7-0f83a4350150@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The parent commit adds this new counter, incremented when receiving a
connection request, if the PM didn't allow the creation of new subflows.
Most of the time, it is then kept at 0, except when the PM limits cause
the receiver side to reject new MPJoin connections. This is the case in
the following tests:
- single subflow, limited by server
- multiple subflows, limited by server
- subflows limited by server w cookies
- userspace pm type rejects join
- userspace pm type prevents mp_prio
Simply set join_syn_rej=1 when checking the MPJoin counters for these
tests.
Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250413-net-next-mptcp-sched-mib-sft-misc-v2-6-0f83a4350150@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This counter is useful to understand why some paths are rejected, and
not created as expected.
It is incremented when receiving a connection request, if the PM didn't
allow the creation of new subflows.
Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250413-net-next-mptcp-sched-mib-sft-misc-v2-5-0f83a4350150@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
subflow_hmac_valid() needs to access the MPTCP socket and the subflow
request, but not the request sock that is passed in argument.
Instead, the subflow request can be directly passed to avoid getting it
via an additional cast.
Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250413-net-next-mptcp-sched-mib-sft-misc-v2-4-0f83a4350150@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Commit e4c28e3d5c ("mptcp: pm: move generic PM helpers to pm.c")
removed an unnecessary if-check, which resulted in returning a freed
pointer.
This still works due to the implicit boolean conversion when returning
the freed pointer from mptcp_remove_anno_list_by_saddr(), but it can be
confusing and potentially error-prone. To improve clarity, add a local
variable to explicitly return a boolean value instead.
Signed-off-by: Thorsten Blum <thorsten.blum@linux.dev>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250413-net-next-mptcp-sched-mib-sft-misc-v2-3-0f83a4350150@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
A new interface .validate has been added in struct bpf_struct_ops
recently. This patch prepares a future struct_ops support by
implementing it as a new helper mptcp_validate_scheduler() for struct
mptcp_sched_ops.
In this helper, check whether the required ops "get_subflow" of struct
mptcp_sched_ops has been implemented.
Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250413-net-next-mptcp-sched-mib-sft-misc-v2-2-0f83a4350150@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This is a follow-up of commit b68b106b0f ("mptcp: sched: reduce size
for unused data"), now removing the mptcp_sched_data structure.
Now is a good time to do that, because the previously mentioned WIP work
has been updated, no longer depending on this structure.
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250413-net-next-mptcp-sched-mib-sft-misc-v2-1-0f83a4350150@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
-----BEGIN PGP SIGNATURE-----
iQEzBAABCAAdFiEEq1nRK9aeMoq1VSgcnJ2qBz9kQNkFAmf+LBEACgkQnJ2qBz9k
QNkVywf/YPZpyP2ZOp6bvRX+dPGNUABM9aPWtlS5We+EENxUwxzK7sdkYxhmIuwj
3B/lXLVJ2H20mUqgfxIq18gjjB2oqjrSUdIG8ZzTKmE2WnpsZDRXqkLhlsXUbu0q
O3qWjP8PIuUVaW0H1rmt88EMbosE0a9++Dlebrpsr+ryP+V5RjZEwS22a4C377WZ
ddTdENHTAGdxrlTlQNabGa+bBYy8TgGDRvWg7xIMHQSTaPzTRLBM8oJZbtAkhO39
r+Aht/hh8PtRAKNJFljXi0+euX/ZDhBAKS72irs9yvrAKPw6++V30h7eVPuGkjIV
ulJgDYMf98WlivqvFJmdo9ZoeUKDSw==
=LCvT
-----END PGP SIGNATURE-----
Merge tag 'fs_for_v6.15-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs
Pull isofs fix from Jan Kara:
"Fix a case where isofs could be reading beyond end of the passed
file handle if its type was incorrectly set"
* tag 'fs_for_v6.15-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs:
isofs: Prevent the use of too small fid
This code is indented one extra tab. Delete the tab.
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/Z_4458uUI3LURa8M@stanley.mountain
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Replace the old direct use of platform_device APIs with newer KUnit APIs
and the faux bus.
The dummy codec driver device doesn't need to be a platform device.
It can be a faux bus device.
The dummy GPIO driver still must be a platform_device so that a
software_node can be added to it before it probes. But use the new
KUnit-managed APIs to create the platform_device and platform_driver.
These will cleanup automatically when a test completes or fails.
Also use KUnit resource cleanup to destroy the faux bus driver and the GPIO
software node instead of doing this "manually" in test exit() functions.
Signed-off-by: Richard Fitzgerald <rf@opensource.cirrus.com>
Link: https://patch.msgid.link/20250415105414.471039-1-rf@opensource.cirrus.com
Signed-off-by: Takashi Iwai <tiwai@suse.de>
If userspace reads "uevent" device attribute at the same time as another
threads unbinds the device from its driver, change to dev->driver from a
valid pointer to NULL may result in crash. Fix this by using READ_ONCE()
when fetching the pointer, and take bus' drivers klist lock to make sure
driver instance will not disappear while we access it.
Use WRITE_ONCE() when setting the driver pointer to ensure there is no
tearing.
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Reviewed-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Link: https://lore.kernel.org/r/20250311052417.1846985-3-dmitry.torokhov@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This reverts commit c0a40097f0.
Probing a device can take arbitrary long time. In the field we observed
that, for example, probing a bad micro-SD cards in an external USB card
reader (or maybe cards were good but cables were flaky) sometimes takes
longer than 2 minutes due to multiple retries at various levels of the
stack. We can not block uevent_show() method for that long because udev
is reading that attribute very often and that blocks udev and interferes
with booting of the system.
The change that introduced locking was concerned with dev_uevent()
racing with unbinding the driver. However we can handle it without
locking (which will be done in subsequent patch).
There was also claim that synchronization with probe() is needed to
properly load USB drivers, however this is a red herring: the change
adding the lock was introduced in May of last year and USB loading and
probing worked properly for many years before that.
Revert the harmful locking.
Cc: stable@vger.kernel.org
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Reviewed-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Link: https://lore.kernel.org/r/20250311052417.1846985-1-dmitry.torokhov@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Add device tree entries for the SPMI controller
Reviewed-by: Nick Chan <towinchenmi@gmail.com>
Signed-off-by: Sasha Finkelstein <fnkl.kernel@gmail.com>
Reviewed-by: Neal Gompa <neal@gompa.dev>
Link: https://lore.kernel.org/r/20250409-spmi-v4-3-eb81ecfd1f64@gmail.com
Signed-off-by: Sven Peter <sven@svenpeter.dev>
The original commit message and the wording "uncork" in the code comment
indicate that it is expected that the suppressed event instances are
automatically sent after unsuppressing.
This is not the case, instead they are discarded.
In effect this means that no "changed" events are emitted on the device
itself by default.
While each discovered partition does trigger a changed event on the
device, devices without partitions don't have any event emitted.
This makes udev miss the device creation and prompted workarounds in
userspace. See the linked util-linux/losetup bug.
Explicitly emit the events and drop the confusingly worded comments.
Link: https://github.com/util-linux/util-linux/issues/2434
Fixes: 498ef5c777 ("loop: suppress uevents while reconfiguring the device")
Cc: stable@vger.kernel.org
Signed-off-by: Thomas Weißschuh <thomas.weissschuh@linutronix.de>
Link: https://lore.kernel.org/r/20250415-loop-uevent-changed-v2-1-0c4e6a923b2a@linutronix.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Set cmd->iocb.ki_ioprio to the ioprio of loop device's request.
The purpose is to inherit the original request ioprio in the aio
flow.
Signed-off-by: Yunlong Xing <yunlong.xing@unisoc.com>
Signed-off-by: Zhiguo Niu <zhiguo.niu@unisoc.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20250414030159.501180-1-yunlong.xing@unisoc.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Remove .owner field if calls are used which set it automatically.
Generated by: scripts/coccinelle/api/platform_no_drv_owner.cocci.
Signed-off-by: Chen Ni <nichen@iscas.ac.cn>
Link: https://lore.kernel.org/r/20250407092632.2952200-1-nichen@iscas.ac.cn
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Function dev_err() is redundant because platform_get_irq()
already prints an error.
Signed-off-by: Chen Ni <nichen@iscas.ac.cn>
Reviewed-by: Dominik Karol Piątkowski <dominik.karol.piatkowski@protonmail.com>
Link: https://lore.kernel.org/r/20250407031110.2382308-1-nichen@iscas.ac.cn
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Using Linux code style to replace typedef local_ppoll_mode_ioctl_t with
type short.
Adhering to Linux code style.
Reported by checkpatch.pl
WARNING: do not add new typedefs
Signed-off-by: Michael Rubin <matchstick@neverthere.org>
Link: https://lore.kernel.org/r/20250409055903.321438-22-matchstick@neverthere.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Using Linux code style to replace typedef autospoll_ioctl_t with type short.
Adhering to Linux code style.
Reported by checkpatch.pl
WARNING: do not add new typedefs
Signed-off-by: Michael Rubin <matchstick@neverthere.org>
Link: https://lore.kernel.org/r/20250409055903.321438-21-matchstick@neverthere.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Using Linux code style to replace typedef t1_delay_ioctl_t with type
unsigned int.
Adhering to Linux code style.
Reported by checkpatch.pl
WARNING: do not add new typedefs
Signed-off-by: Michael Rubin <matchstick@neverthere.org>
Link: https://lore.kernel.org/r/20250409055903.321438-20-matchstick@neverthere.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Using Linux code style to replace typedef rsc_ioctl_t with type int.
Adhering to Linux code style.
Reported by checkpatch.pl
WARNING: do not add new typedefs
Signed-off-by: Michael Rubin <matchstick@neverthere.org>
Link: https://lore.kernel.org/r/20250409055903.321438-19-matchstick@neverthere.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Using Linux code style to replace typedef event_ioctl_t with type short.
Adhering to Linux code style.
Reported by checkpatch.pl
WARNING: do not add new typedefs
Signed-off-by: Michael Rubin <matchstick@neverthere.org>
Link: https://lore.kernel.org/r/20250409055903.321438-18-matchstick@neverthere.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Using Linux code style for 'struct gpib_request_service2' to remove typedef.
Adhering to Linux code style.
Reported by checkpatch.pl
In general, a pointer, or a struct that has elements that can reasonably be
directly accessed should never be a typedef.
Signed-off-by: Michael Rubin <matchstick@neverthere.org>
Link: https://lore.kernel.org/r/20250409055903.321438-17-matchstick@neverthere.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Using Linux code style for 'struct gpib_select_device_path_ioctl' to
remove typedef.
Adhering to Linux code style.
Reported by checkpatch.pl
In general, a pointer, or a struct that has elements that can reasonably be
directly accessed should never be a typedef.
Signed-off-by: Michael Rubin <matchstick@neverthere.org>
Link: https://lore.kernel.org/r/20250409055903.321438-16-matchstick@neverthere.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Using Linux code style for 'struct gpib_sad_ioctl' to remove typedef.
Adhering to Linux code style.
Reported by checkpatch.pl
In general, a pointer, or a struct that has elements that can reasonably be
directly accessed should never be a typedef.
Signed-off-by: Michael Rubin <matchstick@neverthere.org>
Link: https://lore.kernel.org/r/20250409055903.321438-15-matchstick@neverthere.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Using Linux code style for 'struct gpib_pad_ioctl' to remove typedef.
Adhering to Linux code style.
Reported by checkpatch.pl
In general, a pointer, or a struct that has elements that can reasonably be
directly accessed should never be a typedef.
Signed-off-by: Michael Rubin <matchstick@neverthere.org>
Link: https://lore.kernel.org/r/20250409055903.321438-14-matchstick@neverthere.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Using Linux code style for 'struct gpib_ppoll_config_ioctl' to remove typedef.
Adhering to Linux code style.
Reported by checkpatch.pl
In general, a pointer, or a struct that has elements that can reasonably be
directly accessed should never be a typedef.
Signed-off-by: Michael Rubin <matchstick@neverthere.org>
Link: https://lore.kernel.org/r/20250409055903.321438-13-matchstick@neverthere.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>