Commit graph

1,403,262 commits

Author SHA1 Message Date
Cristian Marussi
c23c03bf1f firmware: arm_scmi: Fix timeout checks on polling path
Polling mode transactions wait for a reply busy-looping without holding a
spinlock, but currently the timeout checks are based only on elapsed time:
as a result we could hit a false positive whenever our busy-looping thread
is pre-empted and scheduled out for a time greater than the polling
timeout.

Change the checks at the end of the busy-loop to make sure that the polling
wasn't indeed successful or an out-of-order reply caused the polling to be
forcibly terminated.

Fixes: 31d2f803c1 ("firmware: arm_scmi: Add sync_cmds_completed_on_ret transport flag")
Reported-by: Huangjie <huangjie1663@phytium.com.cn>
Closes: https://lore.kernel.org/arm-scmi/20250123083323.2363749-1-jackhuang021@gmail.com/
Signed-off-by: Cristian Marussi <cristian.marussi@arm.com>
Cc: stable@vger.kernel.org # 5.18.x
Message-Id: <20250310175800.1444293-1-cristian.marussi@arm.com>
Signed-off-by: Sudeep Holla <sudeep.holla@arm.com>
2025-04-08 10:31:59 +01:00
Cristian Marussi
9ca67840c0 firmware: arm_scmi: Balance device refcount when destroying devices
Using device_find_child() to lookup the proper SCMI device to destroy
causes an unbalance in device refcount, since device_find_child() calls an
implicit get_device(): this, in turns, inhibits the call of the provided
release methods upon devices destruction.

As a consequence, one of the structures that is not freed properly upon
destruction is the internal struct device_private dev->p populated by the
drivers subsystem core.

KMemleak detects this situation since loading/unloding some SCMI driver
causes related devices to be created/destroyed without calling any
device_release method.

unreferenced object 0xffff00000f583800 (size 512):
  comm "insmod", pid 227, jiffies 4294912190
  hex dump (first 32 bytes):
    00 00 00 00 ad 4e ad de ff ff ff ff 00 00 00 00  .....N..........
    ff ff ff ff ff ff ff ff 60 36 1d 8a 00 80 ff ff  ........`6......
  backtrace (crc 114e2eed):
    kmemleak_alloc+0xbc/0xd8
    __kmalloc_cache_noprof+0x2dc/0x398
    device_add+0x954/0x12d0
    device_register+0x28/0x40
    __scmi_device_create.part.0+0x1bc/0x380
    scmi_device_create+0x2d0/0x390
    scmi_create_protocol_devices+0x74/0xf8
    scmi_device_request_notifier+0x1f8/0x2a8
    notifier_call_chain+0x110/0x3b0
    blocking_notifier_call_chain+0x70/0xb0
    scmi_driver_register+0x350/0x7f0
    0xffff80000a3b3038
    do_one_initcall+0x12c/0x730
    do_init_module+0x1dc/0x640
    load_module+0x4b20/0x5b70
    init_module_from_file+0xec/0x158

$ ./scripts/faddr2line ./vmlinux device_add+0x954/0x12d0
device_add+0x954/0x12d0:
kmalloc_noprof at include/linux/slab.h:901
(inlined by) kzalloc_noprof at include/linux/slab.h:1037
(inlined by) device_private_init at drivers/base/core.c:3510
(inlined by) device_add at drivers/base/core.c:3561

Balance device refcount by issuing a put_device() on devices found via
device_find_child().

Reported-by: Alice Ryhl <aliceryhl@google.com>
Closes: https://lore.kernel.org/linux-arm-kernel/Z8nK3uFkspy61yjP@arm.com/T/#mc1f73a0ea5e41014fa145147b7b839fc988ada8f
CC: Sudeep Holla <sudeep.holla@arm.com>
CC: Catalin Marinas <catalin.marinas@arm.com>
Fixes: d4f9dddd21 ("firmware: arm_scmi: Add dynamic scmi devices creation")
Signed-off-by: Cristian Marussi <cristian.marussi@arm.com>
Tested-by: Alice Ryhl <aliceryhl@google.com>
Message-Id: <20250306185447.2039336-1-cristian.marussi@arm.com>
Signed-off-by: Sudeep Holla <sudeep.holla@arm.com>
2025-04-08 10:31:59 +01:00
Mark Brown
f4f20f7a50
ASoC: wcd938x: enable t14s audio headset
Merge series from srinivas.kandagatla@linaro.org:

On Lenovo ThinkPad T14s, the headset is connected via a HiFi Switch to
support CTIA and OMTP headsets. This switch is used to minimise pop and
click during headset type switching.

This patchset adds required bindings and changes to codec and dts to
tnable the regulator required to power this switch along with wiring up
gpio that control the headset switching.

Without this patchset, there will be lots of noise on headset and mic
will not we functional.
2025-04-08 10:25:20 +01:00
Mark Brown
8e5e0e7110
ASoC: fsl_sai: add several improvements
Merge series from Shengjiu Wang <shengjiu.wang@nxp.com>:

Add several improvements for the sai interface.
1.allow to set mclk rate with zero clk_id for master mode
2.add xlate_tdm_slot_mask() callback to avoid channel constrain
3.separate 'is_dsp_mode' for tx and rx
4.separate set_tdm_slot() for tx and rx
2025-04-08 10:25:16 +01:00
Mark Brown
94112d3d94
ASoC: codec: wcd93xx: Convert to GPIO descriptors
Merge series from "Peng Fan (OSS)" <peng.fan@oss.nxp.com>:

of_gpio.h is deprecated, so update driver to use gpiod API.

The current driver use value 0 to assert reset and 1 to deassert reset.
The DTSes in tree that use the codec are using GPIO_ACTIVE_LOW.
So it is safe to use devm_gpiod_get to get GPIO descriptors and
use gpiod_set_value to configure output with value 1 means raw value
0, value 0 means raw value 1.

Note:
I not have devices to test, just my best pratice to do the convertion.
2025-04-08 10:25:12 +01:00
Mark Brown
1f4db3cb1a
ASoC: Intel: avs: 16 channels support
Merge series from Cezary Rojewski <cezary.rojewski@intel.com>:

Relatively small delta-wise patchset which raises max channels supported
from 8 to 16.  The existing limitation is software-based, not hardware
based.  The hardware, as per HDAudio specification, section 1.2.2,
(relevant register at SDnFMT, section 3.3.41) supports the
configurations for years.  The avs-driver becomes the first consumer of
that configuration on the Linux kernel side.

Set starts off with update to string_helpers so that functionality added
with parse_int_array_user() can be utilized in kernel-kernel
interactions.

Follow up is rasing the cap on HDAudio-library side.  The format
selection procedure found in the library is good-to-go as is.

Everything that follows these two patches is avs-driver specific:
- raise channels_max for every DAI-driver template
- provide i2s_test module parameter for testing purposes.  When combined
  with I2S loopback card, allows to test 16ch on most Intel hardware post
  Broadwell era
- adjust TDM masks to reflect the 8 -> 16 channels change
2025-04-08 10:25:08 +01:00
Mark Brown
aa446b5d9a
ASoC: Intel: avs: Add support for FCL platform
Merge series from Cezary Rojewski <cezary.rojewski@intel.com>:

The patchset is fairly straightforward - add support for Automotive
platforms based on new DSP architecture, Frisco Lake (FCL), a
PantherLake (PTL)-based platform is an example of.  The cAVS
architecture which all Intel AudioDSP followed for years ends with
RaptorLake familty.  Like all the major updates, this one received new
name too - Audio Context Engine (ACE).

While the range of improvements and changes on the firmware/hardware
side is large, software survives this evolution without need of any
major refactoring.  Additional hardware changes brought with LunarLake
(LNL, ACE 2.0) call for update in PCM-area.  The GPDMAs previously
utilized for non-HDAudio transfer types are no longer there, everything
is running through HDAudio LINK on the Back-End side now.

In terms of code, the mtl.c file, provided with patch 05 'ASoC: Intel:
avs: PTL-based platforms support' hosts largest number of new handlers -
new IRQ and INT control and DSP-cores management.  Combined with lnl.c
and ptl.c which layer the architecture changes done over ACE
generations, provide support for PTL-based platforms e.g.: FCL.
The inheritance in summary:

	mtl.c <- lnl.c <- ptl.c

The functional update to HDAudio library is there to help avs-driver
read certain capabilities directly from the hardware.  Once the pointer
to LINK is obtained, there is no need to call AudioDSP firmware to get
the caps.
2025-04-08 10:25:04 +01:00
Mark Brown
25293be4a7
ASoC: Intel: avs: Update machine board card names
Merge series from Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>:

As discussed in user bug report [1] & [2], it was identified that HDA
card provides wrong FE device numbers to be used with UCM, at the same
time it was requested that we change card names to better names.

This patchset does that, by adding option for going back to old
behaviour in first patch. Using existing way of enumerating FEs from
topology in second patch. Then setting new names in all cards while
providing fallback to old name. Only exception is probe card, which is
used for debug purposes only, where we just directly rename card.

Do note that patch 2, will require updated topologies if topology
exposes more than 1 FE on card. Old topologies didn't assign id field
and it defaulted to 0, however when we use this field by setting
snd_soc_component_driver::use_dai_pcm_id field, we need topologies with
distinct values in FEs. Necessary changes are provided in avsdk and
avs-topology-xml repositories ([4] & [5]). linux-firmware update will
follow as soon as this changeset is merged.

[1]
Link: https://bugzilla.kernel.org/show_bug.cgi?id=219654
[2]
Link: https://github.com/alsa-project/alsa-ucm-conf/pull/499
[3]
Link: https://lore.kernel.org/linux-sound/20250127144445.2739017-1-amadeuszx.slawinski@linux.intel.com/
[4]
a879c8ae4b
[5]
9b94d52cbc
bdbc8d6ba9
2025-04-08 10:25:00 +01:00
Mark Brown
d4c29a3360
ASoC: codecs: lpass-wsa: fix VI capture setup.
Merge series from srinivas.kandagatla@linaro.org:

This two patches fixes below two issues with the VI setup.

1. Only one channel gets enabled on VI feedback patch instead of two
channels
2. recording rate is hardcoded to 8K instead dyamically setting it up.

Both of these issues are fixed in these patches.
2025-04-08 10:24:55 +01:00
Christian Brauner
4e5c53e038
Merge patch series "VFS: improve interface for lookup_one functions"
NeilBrown <neil@brown.name> says:

VFS has some functions with names containing "lookup_one_len" and others
without the "_len".  This difference has nothing to do with "len".  This
is an historical accident but can be confusing.

The functions without "_len" take a "mnt_idmap" pointer.  This is found
in the "vfsmount" and that is an important question when choosing which
to use: do you have a vfsmount, or are you "inside" the filesystem.  A
related question is "is permission checking relevant here?".

nfsd and cachefiles *do* have a vfsmount but *don't* use the non-_len
functions.  They pass nop_mnt_idmap and refuse to work on filesystems
which have any other idmap.

This series changes nfsd and cachefile to use the lookup_one family of
functions and to explictily pass &nop_mnt_idmap which is consistent with
all other vfs interfaces used where &nop_mnt_idmap is explicitly passed.

The remaining uses of the "_one" functions do not require permission
checks so these are renamed to be "_noperm" and the permission checking
is removed.

This series also changes these lookup function to take a qstr instead of
separate name and len.  In many cases this simplifies the call.

* patches from https://lore.kernel.org/r/20250319031545.2999807-2-neil@brown.name:
  VFS: change lookup_one_common and lookup_noperm_common to take a qstr
  Use try_lookup_noperm() instead of d_hash_and_lookup() outside of VFS
  VFS: rename lookup_one_len family to lookup_noperm and remove permission check
  cachefiles: Use lookup_one() rather than lookup_one_len()
  nfsd: Use lookup_one() rather than lookup_one_len()
  VFS: improve interface for lookup_one functions

Link: https://lore.kernel.org/r/20250319031545.2999807-2-neil@brown.name
Signed-off-by: Christian Brauner <brauner@kernel.org>
2025-04-08 11:24:42 +02:00
NeilBrown
390e34bc14
VFS: change lookup_one_common and lookup_noperm_common to take a qstr
These function already take a qstr of course, but they also currently
take a name/len was well and fill in the qstr.
Now they take a qstr that is already filled in, which is what all the
callers have.

Signed-off-by: NeilBrown <neilb@suse.de>
Link: https://lore.kernel.org/r/20250319031545.2999807-7-neil@brown.name
Signed-off-by: Christian Brauner <brauner@kernel.org>
2025-04-08 11:24:41 +02:00
NeilBrown
06c567403a
Use try_lookup_noperm() instead of d_hash_and_lookup() outside of VFS
try_lookup_noperm() and d_hash_and_lookup() are nearly identical.  The
former does some validation of the name where the latter doesn't.
Outside of the VFS that validation is likely valuable, and having only
one exported function for this task is certainly a good idea.

So make d_hash_and_lookup() local to VFS files and change all other
callers to try_lookup_noperm().  Note that the arguments are swapped.

Signed-off-by: NeilBrown <neilb@suse.de>
Link: https://lore.kernel.org/r/20250319031545.2999807-6-neil@brown.name
Signed-off-by: Christian Brauner <brauner@kernel.org>
2025-04-08 11:24:41 +02:00
NeilBrown
fa6fe07d15
VFS: rename lookup_one_len family to lookup_noperm and remove permission check
The lookup_one_len family of functions is (now) only used internally by
a filesystem on itself either
- in a context where permission checking is irrelevant such as by a
  virtual filesystem populating itself, or xfs accessing its ORPHANAGE
  or dquota accessing the quota file; or
- in a context where a permission check (MAY_EXEC on the parent) has just
  been performed such as a network filesystem finding in "silly-rename"
  file in the same directory.  This is also the context after the
  _parentat() functions where currently lookup_one_qstr_excl() is used.

So the permission check is pointless.

The name "one_len" is unhelpful in understanding the purpose of these
functions and should be changed.  Most of the callers pass the len as
"strlen()" so using a qstr and QSTR() can simplify the code.

This patch renames these functions (include lookup_positive_unlocked()
which is part of the family despite the name) to have a name based on
"lookup_noperm".  They are changed to receive a 'struct qstr' instead
of separate name and len.  In a few cases the use of QSTR() results in a
new call to strlen().

try_lookup_noperm() takes a pointer to a qstr instead of the whole
qstr.  This is consistent with d_hash_and_lookup() (which is nearly
identical) and useful for lookup_noperm_unlocked().

The new lookup_noperm_common() doesn't take a qstr yet.  That will be
tidied up in a subsequent patch.

Signed-off-by: NeilBrown <neil@brown.name>
Link: https://lore.kernel.org/r/20250319031545.2999807-5-neil@brown.name
Signed-off-by: Christian Brauner <brauner@kernel.org>
2025-04-08 11:24:36 +02:00
Mateusz Guzik
e45960c279
fs: unconditionally use atime_needs_update() in pick_link()
Vast majority of the time the func returns false.

This avoids a branch to determine whether we are in RCU mode.

Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Link: https://lore.kernel.org/20250408073641.1799151-1-mjguzik@gmail.com
Signed-off-by: Christian Brauner <brauner@kernel.org>
2025-04-08 11:08:24 +02:00
Andy Shevchenko
b8c7a1ac88 gpiolib: of: Move Atmel HSMCI quirk up out of the regulator comment
The regulator comment in of_gpio_set_polarity_by_property()
made on top of a couple of the cases, while Atmel HSMCI quirk
is not related to that. Make it clear by moving Atmel HSMCI
quirk up out of the scope of the regulator comment.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20250402122058.1517393-3-andriy.shevchenko@linux.intel.com
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
2025-04-08 11:06:45 +02:00
Andy Shevchenko
2b9c536430 gpiolib: of: Fix the choice for Ingenic NAND quirk
The Ingenic NAND quirk has been added under CONFIG_LCD_HX8357 ifdeffery
which sounds quite wrong. Fix the choice for Ingenic NAND quirk
by wrapping it into own ifdeffery related to the respective driver.

Fixes: 3a7fd473bd ("mtd: rawnand: ingenic: move the GPIO quirk to gpiolib-of.c")
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20250402122058.1517393-2-andriy.shevchenko@linux.intel.com
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
2025-04-08 11:06:33 +02:00
Paolo Abeni
cd23e77e65 Merge branch 'net_sched-make-qlen_notify-idempotent'
Cong Wang says:

====================
net_sched: make ->qlen_notify() idempotent

Gerrard reported a vulnerability exists in fq_codel where manipulating
the MTU can cause codel_dequeue() to drop all packets. The parent qdisc's
sch->q.qlen is only updated via ->qlen_notify() if the fq_codel queue
remains non-empty after the drops. This discrepancy in qlen between
fq_codel and its parent can lead to a use-after-free condition.

Let's fix this by making all existing ->qlen_notify() idempotent so that
the sch->q.qlen check will be no longer necessary.

Patch 1~5 make all existing ->qlen_notify() idempotent to prepare for
patch 6 which removes the sch->q.qlen check. They are followed by 5
selftests for each type of Qdisc's we touch here.

All existing and new Qdisc selftests pass after this patchset.

Fixes: 4b549a2ef4 ("fq_codel: Fair Queue Codel AQM")
Fixes: 76e3cc126b ("codel: Controlled Delay AQM")
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
====================

Link: https://patch.msgid.link/20250403211033.166059-1-xiyou.wangcong@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2025-04-08 10:58:16 +02:00
Cong Wang
ce94507f5f selftests/tc-testing: Add a test case for FQ_CODEL with ETS parent
Add a test case for FQ_CODEL with ETS parent to verify packet drop
behavior when the queue becomes empty. This helps ensure proper
notification mechanisms between qdiscs.

Note this is best-effort, it is hard to play with those parameters
perfectly to always trigger ->qlen_notify().

Cc: Pedro Tammela <pctammela@mojatatu.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Reviewed-by: Victor Nogueira <victor@mojatatu.com>
Link: https://patch.msgid.link/20250403211636.166257-6-xiyou.wangcong@gmail.com
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2025-04-08 10:58:16 +02:00
Cong Wang
0d5c27ecb6 selftests/tc-testing: Add a test case for FQ_CODEL with DRR parent
Add a test case for FQ_CODEL with DRR parent to verify packet drop
behavior when the queue becomes empty. This helps ensure proper
notification mechanisms between qdiscs.

Note this is best-effort, it is hard to play with those parameters
perfectly to always trigger ->qlen_notify().

Cc: Pedro Tammela <pctammela@mojatatu.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Reviewed-by: Victor Nogueira <victor@mojatatu.com>
Link: https://patch.msgid.link/20250403211636.166257-5-xiyou.wangcong@gmail.com
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2025-04-08 10:58:16 +02:00
Cong Wang
72b05c1bf7 selftests/tc-testing: Add a test case for FQ_CODEL with HFSC parent
Add a test case for FQ_CODEL with HFSC parent to verify packet drop
behavior when the queue becomes empty. This helps ensure proper
notification mechanisms between qdiscs.

Note this is best-effort, it is hard to play with those parameters
perfectly to always trigger ->qlen_notify().

Cc: Pedro Tammela <pctammela@mojatatu.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Reviewed-by: Victor Nogueira <victor@mojatatu.com>
Link: https://patch.msgid.link/20250403211636.166257-4-xiyou.wangcong@gmail.com
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2025-04-08 10:58:15 +02:00
Cong Wang
4cb1837ac5 selftests/tc-testing: Add a test case for FQ_CODEL with QFQ parent
Add a test case for FQ_CODEL with QFQ parent to verify packet drop
behavior when the queue becomes empty. This helps ensure proper
notification mechanisms between qdiscs.

Note this is best-effort, it is hard to play with those parameters
perfectly to always trigger ->qlen_notify().

Cc: Pedro Tammela <pctammela@mojatatu.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Reviewed-by: Victor Nogueira <victor@mojatatu.com>
Link: https://patch.msgid.link/20250403211636.166257-3-xiyou.wangcong@gmail.com
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2025-04-08 10:58:15 +02:00
Cong Wang
cbe9588b12 selftests/tc-testing: Add a test case for FQ_CODEL with HTB parent
Add a test case for FQ_CODEL with HTB parent to verify packet drop
behavior when the queue becomes empty. This helps ensure proper
notification mechanisms between qdiscs.

Note this is best-effort, it is hard to play with those parameters
perfectly to always trigger ->qlen_notify().

Cc: Pedro Tammela <pctammela@mojatatu.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Reviewed-by: Victor Nogueira <victor@mojatatu.com>
Link: https://patch.msgid.link/20250403211636.166257-2-xiyou.wangcong@gmail.com
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2025-04-08 10:58:15 +02:00
Bartosz Golaszewski
9ed74dfa08 Stable tag for the GPIO tree to pull from
-----BEGIN PGP SIGNATURE-----
 
 iQJHBAABCgAxFiEEQp8+kY+LLUocC4bMphj1TA10mKEFAmf04k0THHRnbHhAbGlu
 dXRyb25peC5kZQAKCRCmGPVMDXSYocLBD/0bUwFH+WqgO+nHEiVulSgaYBZvyX8E
 fIVSQBCVmZdjmJSeuzyeaw+NIe76I4ISHZ9TH93Ki88f4siTkKhjl0M8W01sq4m6
 X37lGuZZJF1nqEAtU1qqZ4g+MyN/lLVHD8fe05D9fGimX7d1zvugE4xkSR8rIK6T
 nTLm90O5t8VbSwERCHhUK0ZuayvWUnLqZQBOU0CRK1ikyG0WXmYb8lTpU8PKM8tT
 xDfStiU2X3ERDqJflIJanz+o8fEcEm5LF45MbhQsXvgUQb0PRl4HHOnESKvhbGc/
 JQ83ocUXIQ7eeUGx8nZVmN6K/8NrSjCG8Dvwe6SLBgeIkdAHfyyi6igdfWaYG/h+
 LTsF9OODYDKcLGaf4wg+tK/Z46HC9aHnEvCDLO9A9h0vHHtXKdldkccmmFuPsKLF
 JN2FBNYcldItEjCHtRixJBdT1EpSy46VghJrnOD6tPvBzwqHYYSFbkrlWv1NF3Sw
 8hg5bYgemQ+8lYirsPWhAUCxgwb7f+IsNn/VwKufJ8ohr85O4EMOS+IDFtiKQJEQ
 A9KDRfzOwSUj0Fp5Y26s8rGSUHDcXqnkQZsJsHx8Bwe3SSndZtstTUgNhCkl9VXY
 aqxq+dKtvXhE4ZZZwMYa4BDoGqrWbJiGYbSidmTaDfc56LxOa4MVh1kwAo0WcEDY
 wsK6a7fRSsWYWw==
 =pJLt
 -----END PGP SIGNATURE-----

Merge tag 'irqdomain-04-08-25' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip into gpio/for-next

Stable tag for the GPIO tree to pull from
2025-04-08 10:58:07 +02:00
Cong Wang
342debc121 codel: remove sch->q.qlen check before qdisc_tree_reduce_backlog()
After making all ->qlen_notify() callbacks idempotent, now it is safe to
remove the check of qlen!=0 from both fq_codel_dequeue() and
codel_qdisc_dequeue().

Reported-by: Gerrard Tai <gerrard.tai@starlabs.sg>
Fixes: 4b549a2ef4 ("fq_codel: Fair Queue Codel AQM")
Fixes: 76e3cc126b ("codel: Controlled Delay AQM")
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250403211636.166257-1-xiyou.wangcong@gmail.com
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2025-04-08 10:57:56 +02:00
Cong Wang
a7a15f39c6 sch_ets: make est_qlen_notify() idempotent
est_qlen_notify() deletes its class from its active list with
list_del() when qlen is 0, therefore, it is not idempotent and
not friendly to its callers, like fq_codel_dequeue().

Let's make it idempotent to ease qdisc_tree_reduce_backlog() callers'
life. Also change other list_del()'s to list_del_init() just to be
extra safe.

Reported-by: Gerrard Tai <gerrard.tai@starlabs.sg>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Link: https://patch.msgid.link/20250403211033.166059-6-xiyou.wangcong@gmail.com
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2025-04-08 10:57:49 +02:00
Cong Wang
55f9eca4bf sch_qfq: make qfq_qlen_notify() idempotent
qfq_qlen_notify() always deletes its class from its active list
with list_del_init() _and_ calls qfq_deactivate_agg() when the whole list
becomes empty.

To make it idempotent, just skip everything when it is not in the active
list.

Also change other list_del()'s to list_del_init() just to be extra safe.

Reported-by: Gerrard Tai <gerrard.tai@starlabs.sg>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250403211033.166059-5-xiyou.wangcong@gmail.com
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2025-04-08 10:57:49 +02:00
Cong Wang
51eb3b6554 sch_hfsc: make hfsc_qlen_notify() idempotent
hfsc_qlen_notify() is not idempotent either and not friendly
to its callers, like fq_codel_dequeue(). Let's make it idempotent
to ease qdisc_tree_reduce_backlog() callers' life:

1. update_vf() decreases cl->cl_nactive, so we can check whether it is
non-zero before calling it.

2. eltree_remove() always removes RB node cl->el_node, but we can use
   RB_EMPTY_NODE() + RB_CLEAR_NODE() to make it safe.

Reported-by: Gerrard Tai <gerrard.tai@starlabs.sg>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250403211033.166059-4-xiyou.wangcong@gmail.com
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2025-04-08 10:57:49 +02:00
Cong Wang
df008598b3 sch_drr: make drr_qlen_notify() idempotent
drr_qlen_notify() always deletes the DRR class from its active list
with list_del(), therefore, it is not idempotent and not friendly
to its callers, like fq_codel_dequeue().

Let's make it idempotent to ease qdisc_tree_reduce_backlog() callers'
life. Also change other list_del()'s to list_del_init() just to be
extra safe.

Reported-by: Gerrard Tai <gerrard.tai@starlabs.sg>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250403211033.166059-3-xiyou.wangcong@gmail.com
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2025-04-08 10:57:49 +02:00
Cong Wang
5ba8b837b5 sch_htb: make htb_qlen_notify() idempotent
htb_qlen_notify() always deactivates the HTB class and in fact could
trigger a warning if it is already deactivated. Therefore, it is not
idempotent and not friendly to its callers, like fq_codel_dequeue().

Let's make it idempotent to ease qdisc_tree_reduce_backlog() callers'
life.

Reported-by: Gerrard Tai <gerrard.tai@starlabs.sg>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250403211033.166059-2-xiyou.wangcong@gmail.com
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2025-04-08 10:57:49 +02:00
Krzysztof Kozlowski
c5672e310a gpio: zynq: Fix wakeup source leaks on device unbind
Device can be unbound, so driver must also release memory for the wakeup
source.

Cc: stable@vger.kernel.org
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20250406202245.53854-2-krzysztof.kozlowski@linaro.org
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
2025-04-08 10:56:38 +02:00
Krzysztof Kozlowski
da47605e43 gpio: mpc8xxx: Fix wakeup source leaks on device unbind
Device can be unbound, so driver must also release memory for the wakeup
source.

Cc: stable@vger.kernel.org
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20250406202245.53854-1-krzysztof.kozlowski@linaro.org
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
2025-04-08 10:56:25 +02:00
Bartosz Golaszewski
2de1cf175c gpio: TODO: track the removal of regulator-related workarounds
The GPIOD_FLAGS_BIT_NONEXCLUSIVE flag and devm_gpiod_unhinge() function
should be replaced with a better solution. The pwrseq subsystem is a good
candidate. GPIOs themselves should remain a unique resource. Add a task
for tracking the removal of these deprecated symbols.

Acked-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20250401-gpio-todo-remove-nonexclusive-v2-4-7c1380797b0d@linaro.org
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
2025-04-08 10:41:59 +02:00
Bartosz Golaszewski
3af64f175b MAINTAINERS: add more keywords for the GPIO subsystem entry
Add GPIOD_FLAGS_BIT_NONEXCLUSIVE and devm_gpiod_unhinge as keywords to
the GPIO entry so that we get notified if anybody tries to use these
deprecated symbols. We'll drop them from here once we remove them from
the kernel.

Acked-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20250401-gpio-todo-remove-nonexclusive-v2-3-7c1380797b0d@linaro.org
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
2025-04-08 10:41:58 +02:00
Bartosz Golaszewski
686e54ea31 gpio: deprecate devm_gpiod_unhinge()
This function was introduced as a workaround for an issue with resource
ownership in the regulator subsystem. Rather than passing the ownership
of a GPIO, we should make the regulator core be able to deal with
resources it didn't request. Deprecate this function so that we don't
get more users in the tree.

Link: https://lore.kernel.org/r/20250401-gpio-todo-remove-nonexclusive-v2-2-7c1380797b0d@linaro.org
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
2025-04-08 10:41:58 +02:00
Bartosz Golaszewski
6deb8435f6 gpio: deprecate the GPIOD_FLAGS_BIT_NONEXCLUSIVE flag
The non-exclusive GPIO request flag looks like a functional feature but
is in fact a workaround for a corner-case that got out of hand. It should
be removed so deprecate it officially so that nobody uses it anymore.

Acked-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20250401-gpio-todo-remove-nonexclusive-v2-1-7c1380797b0d@linaro.org
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
2025-04-08 10:41:58 +02:00
Christian Brauner
c9b380a017
Merge patch series "fs: sort out cosmetic differences between stat funcs and add predicts"
Predict fastpaths in stat and during fdput().

* patches from https://lore.kernel.org/20250406235806.1637000-1-mjguzik@gmail.com:
  fs: predict not having to do anything in fdput()
  fs: sort out cosmetic differences between stat funcs and add predicts

Link: https://lore.kernel.org/20250406235806.1637000-1-mjguzik@gmail.com
Signed-off-by: Christian Brauner <brauner@kernel.org>
2025-04-08 10:28:10 +02:00
Mateusz Guzik
5f3e0b4a1f
fs: predict not having to do anything in fdput()
This matches the annotation in fdget().

Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Link: https://lore.kernel.org/20250406235806.1637000-2-mjguzik@gmail.com
Signed-off-by: Christian Brauner <brauner@kernel.org>
2025-04-08 10:28:07 +02:00
Mateusz Guzik
eaec2cd167
fs: sort out cosmetic differences between stat funcs and add predicts
This is a nop, but I did verify asm improves.

Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Link: https://lore.kernel.org/20250406235806.1637000-1-mjguzik@gmail.com
Signed-off-by: Christian Brauner <brauner@kernel.org>
2025-04-08 10:28:07 +02:00
Lad Prabhakar
b6f2c6bd4e clk: renesas: r9a09g057: Add clock and reset entries for GE3D
Add PLLGPU along with the necessary clock and reset entries for GE3D.

Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/20250309211402.80886-6-prabhakar.mahadev-lad.rj@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
2025-04-08 10:16:09 +02:00
Lad Prabhakar
360387a8f1 clk: renesas: rzv2h: Rename PLL field macros for consistency
Rename PLL field extraction macros to include the associated register name
(`CPG_PLL_CLK1` or `CPG_PLL_CLK2`) to maintain consistency with other PLL
register macros. Update all corresponding macro references accordingly.

Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/20250309211402.80886-5-prabhakar.mahadev-lad.rj@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
2025-04-08 10:16:09 +02:00
Lad Prabhakar
fea942bc15 clk: renesas: rzv2h: Add support for enabling PLLs
Some RZ/V2H(P) SoC variants do not have a GPU, resulting in PLLGPU being
disabled by default in TF-A. Add support for enabling PLL clocks in the
RZ/V2H(P) CPG driver to manage this.

Introduce `is_enabled` and `enable` callbacks to handle PLL state
transitions. With the `enable` callback, PLLGPU will be turned ON only
when the GPU node is enabled; otherwise, it will remain off. Define new
macros for PLL standby and monitor registers to facilitate this process.

Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/20250309211402.80886-4-prabhakar.mahadev-lad.rj@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
2025-04-08 10:16:09 +02:00
Lad Prabhakar
18510fd7bf clk: renesas: rzv2h: Remove unused type field from struct pll_clk
Remove the redundant `type` field from `struct pll_clk`, as it is not used
in the PLL clock handling logic.

Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/20250309211402.80886-3-prabhakar.mahadev-lad.rj@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
2025-04-08 10:16:09 +02:00
Lad Prabhakar
20fc4ea6d7 clk: renesas: rzv2h: Refactor PLL configuration handling
Refactor PLL handling by introducing a `struct pll` to encapsulate PLL
configuration parameters, ensuring consistency with the existing dynamic
divider structure.

Introduce the `PLL_PACK()` macro to simplify PLL structure initialization
and update the `DEF_PLL()` macro to use the new `pll` structure. Modify
relevant clock register functions to utilize the structured PLL data
instead of raw configuration values.

This refactoring improves code readability, maintainability, and
alignment with the existing clock configuration approach.

Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/20250309211402.80886-2-prabhakar.mahadev-lad.rj@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
2025-04-08 10:16:09 +02:00
Thomas Zimmermann
1260ed7779 Merge drm/drm-fixes into drm-misc-fixes
Backmerging to get updates from v6.15-rc1.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
2025-04-08 10:15:47 +02:00
Tung Nguyen
69ae94725f tipc: fix memory leak in tipc_link_xmit
In case the backlog transmit queue for system-importance messages is overloaded,
tipc_link_xmit() returns -ENOBUFS but the skb list is not purged. This leads to
memory leak and failure when a skb is allocated.

This commit fixes this issue by purging the skb list before tipc_link_xmit()
returns.

Fixes: 365ad353c2 ("tipc: reduce risk of user starvation during link congestion")
Signed-off-by: Tung Nguyen <tung.quang.nguyen@est.tech>
Link: https://patch.msgid.link/20250403092431.514063-1-tung.quang.nguyen@est.tech
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2025-04-08 10:15:33 +02:00
Herbert Xu
af7e23c616 crypto: ccp - Silence may-be-uninitialized warning in sev_ioctl_do_pdh_export
The recent reordering of code in sev_ioctl_do_pdh_export triggered
a false-positive may-be-uninitialized warning from gcc:

In file included from ../include/linux/sched/task.h:13,
                 from ../include/linux/sched/signal.h:9,
                 from ../include/linux/rcuwait.h:6,
                 from ../include/linux/percpu-rwsem.h:7,
                 from ../include/linux/fs.h:34,
                 from ../include/linux/compat.h:17,
                 from ../arch/x86/include/asm/ia32.h:7,
                 from ../arch/x86/include/asm/elf.h:10,
                 from ../include/linux/elf.h:6,
                 from ../include/linux/module.h:19,
                 from ../drivers/crypto/ccp/sev-dev.c:11:
In function ‘copy_to_user’,
    inlined from ‘sev_ioctl_do_pdh_export’ at ../drivers/crypto/ccp/sev-dev.c:2036:7,
    inlined from ‘sev_ioctl’ at ../drivers/crypto/ccp/sev-dev.c:2249:9:
../include/linux/uaccess.h:225:16: warning: ‘input_cert_chain_address’ may be used uninitialized [-Wmaybe-uninitialized]
  225 |         return _copy_to_user(to, from, n);
      |                ^~~~~~~~~~~~~~~~~~~~~~~~~~
../drivers/crypto/ccp/sev-dev.c: In function ‘sev_ioctl’:
../drivers/crypto/ccp/sev-dev.c:1961:22: note: ‘input_cert_chain_address’ was declared here
 1961 |         void __user *input_cert_chain_address;
      |                      ^~~~~~~~~~~~~~~~~~~~~~~~

Silence it by moving the initialisation of the variables in question
prior to the NULL check.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Acked-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2025-04-08 15:54:38 +08:00
Ovidiu Panait
bfc68ebefb crypto: sun8i-ce-hash - use API helpers to setup fallback request
Rather than setting up the fallback request by hand, use
ahash_request_set_callback() and ahash_request_set_crypt() API helpers
to properly setup the new request.

Signed-off-by: Ovidiu Panait <ovidiu.panait.oss@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2025-04-08 15:54:38 +08:00
Ovidiu Panait
593c76e1aa crypto: sun8i-ce-hash - add IS_ENABLED() checks to debugfs stats
Add IS_ENABLED(CONFIG_CRYPTO_DEV_SUN8I_CE_DEBUG) checks before the
fallback counter updates to make sure the code is not included when
debugfs statistics support is not enabled.

Signed-off-by: Ovidiu Panait <ovidiu.panait.oss@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2025-04-08 15:54:38 +08:00
Ovidiu Panait
a8632253f3 crypto: sun8i-ce-hash - drop CONFIG_CRYPTO_DEV_SUN8I_CE_DEBUG ifdefs
'struct sun8i_ce_alg_template' is always defined, even with
CONFIG_CRYPTO_DEV_SUN8I_CE_DEBUG disabled, so the ifdef guards
are not needed.

Make sure the statistics have IS_ENABLED() checks instead.

Signed-off-by: Ovidiu Panait <ovidiu.panait.oss@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2025-04-08 15:54:38 +08:00
Ovidiu Panait
b745ab78a8 crypto: sun8i-ce-hash - factor out debugfs fallback statistics
In order to reduce code duplication, factor out to a separate function
the codepath that increments debugfs fallback stats.

Signed-off-by: Ovidiu Panait <ovidiu.panait.oss@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2025-04-08 15:54:38 +08:00