sched_ext generates warnings when cpu.weight / cpu.idle are set to
non-default values if the BPF scheduler doesn't implement weight support.
These warnings don't provide much value while adding constant annoyance. A
BPF scheduler may not implement any particular behavior and there's nothing
particularly special about missing cgroup weight support. Drop the warnings.
Signed-off-by: Tejun Heo <tj@kernel.org>
Replace kzalloc with kvzalloc for the exit_dump buffer allocation, which
can require large contiguous memory depending on the implementation.
This change prevents allocation failures by allowing the system to fall
back to vmalloc when contiguous memory allocation fails.
Since this buffer is only used for debugging purposes, physical memory
contiguity is not required, making vmalloc a suitable alternative.
Cc: stable@vger.kernel.org
Fixes: 07814a9439 ("sched_ext: Print debug dump after an error exit")
Suggested-by: Rik van Riel <riel@surriel.com>
Signed-off-by: Breno Leitao <leitao@debian.org>
Acked-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Commit d072acda48 ("rust: use custom FFI integer types") did not
update rust-analyzer to include the new crate.
To enable rust-analyzer support for these custom ffi types, add the
`ffi` crate as a dependency to the `bindings`, `uapi` and `kernel`
crates, which all directly depend on it.
Fixes: d072acda48 ("rust: use custom FFI integer types")
Signed-off-by: Lukas Fischer <kernel@o1oo11oo.de>
Reviewed-by: Tamir Duberstein <tamird@gmail.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20250404125150.85783-2-kernel@o1oo11oo.de
Signed-off-by: Miguel Ojeda <ojeda@kernel.org>
Some operations require checking, or ignoring, specific bits in an address
value. For example, this can be comparing address values to identify unique
structures.
Currently, the full address value is compared when filtering for duplicates.
This results in over counting and creation of extra records. This gives the
impression that more unique events occurred than did in reality.
Mask the address for physical rows on MI300.
[ bp: Simplify. ]
Fixes: 6f15e617cc ("RAS: Introduce a FRU memory poison manager")
Signed-off-by: Yazen Ghannam <yazen.ghannam@amd.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Cc: stable@vger.kernel.org
landlock_put_hierarchy() can be called when an error occurs in
landlock_merge_ruleset() due to insufficient memory. In this case, the
domain's audit details might not have been allocated yet, which would
cause landlock_free_hierarchy_details() to print a warning (but still
safely handle this case).
We could keep the WARN_ON_ONCE(!hierarchy) but it's not worth it for
this kind of function, so let's remove it entirely.
Cc: Paul Moore <paul@paul-moore.com>
Reported-by: syzbot+8bca99e91de7e060e4ea@syzkaller.appspotmail.com
Link: https://lore.kernel.org/r/20250331104709.897062-1-mic@digikod.net
Reviewed-by: Günther Noack <gnoack@google.com>
Signed-off-by: Mickaël Salaün <mic@digikod.net>
gpio irq which using three-cell scheme should always call
instance_match() function to find the correct irqdomain.
The select() function will be called with !DOMAIN_BUS_ANY,
so for specific gpio irq driver, it need to set bus token
explicitly, something like:
irq_domain_update_bus_token(girq->domain, DOMAIN_BUS_WIRED);
Signed-off-by: Yixun Lan <dlan@gentoo.org>
Link: https://lore.kernel.org/r/20250408-04-gpio-irq-threecell-v4-1-fd170d5e2d2b@gentoo.org
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
The FSD SoC contains two instances of Synopsys DWC QoS Ethernet IP, one in
FSYS0 block and other in PERIC block.
The ethernet interface is connected to a switch which is not managed by
Linux.
Adds device tree node for Ethernet in PERIC Block and enables the same for
FSD platform.
Signed-off-by: Pankaj Dubey <pankaj.dubey@samsung.com>
Signed-off-by: Jayati Sahu <jayati.sahu@samsung.com>
Signed-off-by: Swathi K S <swathi.ks@samsung.com>
Link: https://lore.kernel.org/r/20250307044904.59077-3-swathi.ks@samsung.com
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
The FSD SoC contains two instances of Synopsys DWC QoS Ethernet IP, one
in FSYS0 block and other in PERIC block.
The ethernet interface is managed by a switch which is not managed by
Linux.
Adds device tree node for Ethernet in FSYS0 Block and enables the same for
FSD platform.
Signed-off-by: Pankaj Dubey <pankaj.dubey@samsung.com>
Signed-off-by: Jayati Sahu <jayati.sahu@samsung.com>
Signed-off-by: Swathi K S <swathi.ks@samsung.com>
Link: https://lore.kernel.org/r/20250307044904.59077-2-swathi.ks@samsung.com
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Merge series from Patrice Chotard <patrice.chotard@foss.st.com>:
Make "resets" property mandatory.
Update spi-stm32-ospi driver and dt-bindings accordingly.
A number of test suites call functions that expect the returned
drm_display_mode to be destroyed eventually.
However, none of the tests called drm_mode_destroy, which results in a
memory leak.
Since drm_mode_destroy takes two pointers as argument, we can't use a
kunit wrapper. Let's just create a helper every test suite can use.
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250408-drm-kunit-drm-display-mode-memleak-v1-1-996305a2e75a@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
Merge series from Bartosz Golaszewski <brgl@bgdev.pl>:
struct gpio_chip now has callbacks for setting line values that return
an integer, allowing to indicate failures. We're in the process of
converting all GPIO drivers to using the new API. This series converts
the only GPIO controller under drivers/regulator/ and - while at it -
refactors the code a bit.
SFDP of MX25L3255E is JESD216, which does not include the Quad
Enable bit Requirement in BFPT. As a result, during BFPT parsing,
the quad_enable method is not set to spi_nor_sr1_bit6_quad_enable.
Therefore, it is necessary to correct this setting by late_init.
In addition, MX25L3255E also supports 1-4-4 page program in 3-byte
address mode. However, since the 3-byte address 1-4-4 page program
is not defined in SFDP, it needs to be configured in late_init.
Signed-off-by: Cheng Ming Lin <chengminglin@mxic.com.tw>
Acked-by: Pratyush Yadav <pratyush@kernel.org>
Signed-off-by: Pratyush Yadav <pratyush@kernel.org>
[pratyush@kernel.org: move params declaration to top, and use it everywhere]
Link: https://lore.kernel.org/r/20250407075400.1113177-4-linchengming884@gmail.com
To clean up the code and reduce duplication, this entry is removed.
Signed-off-by: Cheng Ming Lin <chengminglin@mxic.com.tw>
Reviewed-by: Tudor Ambarus <tudor.ambarus@linaro.org>
Signed-off-by: Pratyush Yadav <pratyush@kernel.org>
Link: https://lore.kernel.org/r/20250407075400.1113177-3-linchengming884@gmail.com
Many flash devices share the same ID but have different part numbers.
To avoid confusion, the part number field is removed.
Additionally, since SFDP already provides size information and
functionality covered by no_sfdp_flags, these fields are also removed.
Furthermore, when 4-byte address instruction table is available,
the SPI_NOR_4B_OPCODES flag is no longer needed and is removed.
Signed-off-by: Cheng Ming Lin <chengminglin@mxic.com.tw>
Acked-by: Tudor Ambarus <tudor.ambarus@linaro.org>
Signed-off-by: Pratyush Yadav <pratyush@kernel.org>
Link: https://lore.kernel.org/r/20250407075400.1113177-2-linchengming884@gmail.com
The code for enabling and disabling the irq is so similar that it can
easily be handled by one function. Like in most other RTC drivers. Save
the duplicated code and one layer of indirection.
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Link: https://lore.kernel.org/r/20250320102218.10781-2-wsa+renesas@sang-engineering.com
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Commit 5d2b55e55e ("panel/boe-tv101wum-ll2: Use refcounted allocation
in place of devm_kzalloc()") switched from a kmalloc + drm_panel_init
call to a devm_drm_panel_alloc one.
However, the variable it was storing the allocated pointer in doesn't
exist, resulting in a compilation breakage.
Fixes: 5d2b55e55e ("panel/boe-tv101wum-ll2: Use refcounted allocation in place of devm_kzalloc()")
Signed-off-by: Maxime Ripard <mripard@kernel.org>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Tested-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250408122008.1676235-3-mripard@kernel.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Commit 77dcbce637 ("panel/th101mb31ig002-28a: Use refcounted
allocation in place of devm_kzalloc()") switched from a kmalloc +
drm_panel_init call to a devm_drm_panel_alloc one.
However, the variable it was storing the allocated pointer in doesn't
exist, resulting in a compilation breakage.
Fixes: 77dcbce637 ("panel/th101mb31ig002-28a: Use refcounted allocation in place of devm_kzalloc()")
Signed-off-by: Maxime Ripard <mripard@kernel.org>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Tested-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250408122008.1676235-2-mripard@kernel.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Commit 9d7d7c3c9a ("panel/auo-a030jtn01: Use refcounted allocation in
place of devm_kzalloc()") switched from a kmalloc + drm_panel_init call
to a devm_drm_panel_alloc one.
However, the variable it was storing the allocated pointer in doesn't
exist, resulting in a compilation breakage.
Fixes: 9d7d7c3c9a ("panel/auo-a030jtn01: Use refcounted allocation in place of devm_kzalloc()")
Signed-off-by: Maxime Ripard <mripard@kernel.org>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Tested-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250408122008.1676235-1-mripard@kernel.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
It is good practice to clear running interrupts before removing the
driver. This is not really a bugfix because on current systems RuntimePM
will disable the module clock, so interrupts won't be initiated. The
dependency on that behaviour is subtle, though. Better be self-contained
and clean up when removing.
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Link: https://lore.kernel.org/r/20250312100105.36767-2-wsa+renesas@sang-engineering.com
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Use the BIT macro, use curly braces for else-blocks, don't split strings
over multiple lines, annotate the lock, update copyright.
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Link: https://lore.kernel.org/r/20250227134256.9167-18-wsa+renesas@sang-engineering.com
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
No need to store the resource for the registers in the per-device struct
because we only need it during probe. Remove some unneeded unlikely()
while here and correct the type of 'regsize'.
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Link: https://lore.kernel.org/r/20250227134256.9167-17-wsa+renesas@sang-engineering.com
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
The wrapper to enable interrupts is so thin that we can use it directly.
Also gets rid of an 'inline' which doesn't make sense here.
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Link: https://lore.kernel.org/r/20250227134256.9167-16-wsa+renesas@sang-engineering.com
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
We only need the alarm_irq handler. That means we can remove everything
related to periodic_irq and carry_irq. Also, the shared handler can go
since we only we need the alarm interrupt in any case.
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Link: https://lore.kernel.org/r/20250227134256.9167-15-wsa+renesas@sang-engineering.com
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Because periodic interrupts are emulated by the RTC core, the PIE
handling code can simply go away now. And with it the custom proc-file.
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Link: https://lore.kernel.org/r/20250227134256.9167-14-wsa+renesas@sang-engineering.com
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
With old, custom UIE handling removed, we can now disable the carry
interrupt in probe() and leave it like this. No further handling is
required.
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Link: https://lore.kernel.org/r/20250227134256.9167-13-wsa+renesas@sang-engineering.com
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Since commit e428c6a277 ("RTC: Clean out UIE icotl implementations"),
the flag for UIE cannot be set anymore. Because UIE is now handled via
regular alarms and a timerqueue by the RTC core, the UIE handling code
can simply go away now.
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Link: https://lore.kernel.org/r/20250227134256.9167-12-wsa+renesas@sang-engineering.com
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
The DT bindings for this driver define the interrupts in the order as
they are numbered in the interrupt controller. The old platform_data,
however, listed them in a different order. So, for DT based platforms,
they are mixed up. Assign them specifically for DT, so we can keep the
bindings stable. After the fix, 'rtctest' passes again on the Renesas
Genmai board (RZ-A1 / R7S72100).
Fixes: dab5aec64b ("rtc: sh: add support for rza series")
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Link: https://lore.kernel.org/r/20250227134256.9167-11-wsa+renesas@sang-engineering.com
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
There's a consistent pattern where the .cleanup_data() callback is
called when .prepare_data() fails, when it should really be called to
clean after a successful .prepare_data() as per the documentation.
Rewrite the error-handling paths to make sure we don't cleanup
un-prepared data.
Fixes: c781ff12a2 ("ethtool: Allow network drivers to dump arbitrary EEPROM data")
Reviewed-by: Kory Maincent <kory.maincent@bootlin.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Michal Kubecek <mkubecek@suse.cz>
Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Link: https://patch.msgid.link/20250407130511.75621-1-maxime.chevallier@bootlin.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
The tfilter_notify() and tfilter_del_notify() functions assume that
NLMSG_GOODSIZE is always enough to dump the filter chain. This is not
always the case, which can lead to silent notify failures (because the
return code of tfilter_notify() is not always checked). In particular,
this can lead to NLM_F_ECHO not being honoured even though an action
succeeds, which forces userspace to create workarounds[0].
Fix this by increasing the message size if dumping the filter chain into
the allocated skb fails. Use the size of the incoming skb as a size hint
if set, so we can start at a larger value when appropriate.
To trigger this, run the following commands:
# ip link add type veth
# tc qdisc replace dev veth0 root handle 1: fq_codel
# tc -echo filter add dev veth0 parent 1: u32 match u32 0 0 $(for i in $(seq 32); do echo action pedit munge ip dport set 22; done)
Before this fix, tc just returns:
Not a filter(cmd 2)
After the fix, we get the correct echo:
added filter dev veth0 parent 1: protocol all pref 49152 u32 chain 0 fh 800::800 order 2048 key ht 800 bkt 0 terminal flowid not_in_hw
match 00000000/00000000 at 0
action order 1: pedit action pass keys 1
index 1 ref 1 bind 1
key #0 at 20: val 00000016 mask ffff0000
[repeated 32 times]
[0] 106ef21860
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Reported-by: Frode Nordahl <frode.nordahl@canonical.com>
Closes: https://bugs.launchpad.net/ubuntu/+source/openvswitch/+bug/2018500
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Link: https://patch.msgid.link/20250407105542.16601-1-toke@redhat.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
WX_RXD_IPV6EX was incorrectly defined in Rx ring descriptor. In fact, this
field stores the 802.1ad ID from which the packet was received. The wrong
definition caused the statistics rx_csum_offload_errors to fail to grow
when receiving the 802.1ad packet with incorrect checksum.
Fixes: ef4f3c19f9 ("net: wangxun: libwx add rx offload functions")
Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
Reviewed-by: Michal Kubiak <michal.kubiak@intel.com>
Link: https://patch.msgid.link/20250407103322.273241-1-jiawenwu@trustnetic.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Use SPDX-License-Identifier accross all the files of the xgbe driver to
ensure compliance with Linux kernel standards, thus removing the
boiler-plate template license text.
Signed-off-by: Raju Rangoju <Raju.Rangoju@amd.com>
Acked-by: Shyam Sundar S K <Shyam-sundar.S-k@amd.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250407102913.3063691-1-Raju.Rangoju@amd.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
struct gpio_chip now has callbacks for setting line values that return
an integer, allowing to indicate failures. Convert the driver to using
them.
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Link: https://patch.msgid.link/20250408-gpiochip-set-rv-spi-v1-1-597660766654@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
When running as a PVH dom0 the ACPI tables exposed to Linux are (mostly)
the native ones, thus exposing the C and P states, that can lead to
attachment of CPU idle and frequency drivers. However the entity in
control of the CPU C and P states is Xen, as dom0 doesn't have a full view
of the system load, neither has all CPUs assigned and identity pinned.
Like it's done for classic PV guests, prevent Linux from using idle or
frequency state drivers when running as a PVH dom0.
On an AMD EPYC 7543P system without this fix a Linux PVH dom0 will keep the
host CPUs spinning at 100% even when dom0 is completely idle, as it's
attempting to use the acpi_idle driver.
Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>
Reviewed-by: Jason Andryuk <jason.andryuk@amd.com>
Signed-off-by: Juergen Gross <jgross@suse.com>
Message-ID: <20250407101842.67228-1-roger.pau@citrix.com>
For persistent memories, add rxe_odp_flush_pmem_iova() so that ODP specific
steps are executed. Otherwise, no additional consideration is required.
Signed-off-by: Daisuke Matsuda <matsuda-daisuke@fujitsu.com>
Link: https://patch.msgid.link/20250324075649.3313968-2-matsuda-daisuke@fujitsu.com
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Remove the double negation and simplify the if condition.
No functional changes intended.
Signed-off-by: Thorsten Blum <thorsten.blum@linux.dev>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Link: https://patch.msgid.link/20250407091442.743478-1-thorsten.blum@linux.dev
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
struct gpio_chip now has callbacks for setting line values that return
an integer, allowing to indicate failures. Convert the driver to using
them.
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Link: https://patch.msgid.link/20250408-gpiochip-set-rv-regulator-v1-4-a18b6154b31a@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
Use mutex lock guards from linux/cleanup.h to simplify the driver code.
Note that ret must be initialized in order to avoid a build warning as
scoped_guard() is implemented as a for loop.
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Link: https://patch.msgid.link/20250408-gpiochip-set-rv-regulator-v1-3-a18b6154b31a@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
The __get_unaligned_cpu32 function is deprecated. So, replace it with
the more generic get_unaligned and just cast the input parameter.
Signed-off-by: Julian Vetter <julian@outer-limits.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250407083306.1553921-1-julian@outer-limits.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
If the destination buffer has a fixed length, strscpy_pad()
automatically determines its size using sizeof() when the argument is
omitted. This makes the explicit sizeof() calls unnecessary - remove
them.
No functional changes intended.
Signed-off-by: Thorsten Blum <thorsten.blum@linux.dev>
Link: https://patch.msgid.link/20250407082607.741919-2-thorsten.blum@linux.dev
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
The current code configures the Physical Function (PF) root node at TL1
and the Virtual Function (VF) root node at TL2.
This ensure at any given point of time PF traffic gets more priority.
PF root node
TL1
/ \
TL2 TL2 VF root node
/ \
TL3 TL3
/ \
TL4 TL4
/ \
SMQ SMQ
Due to a bug in the current code, the TL2 parent queue index on the
VF interface is not being configured, leading to 'SMQ Flush' errors
Fixes: 5e6808b4c6 ("octeontx2-pf: Add support for HTB offload")
Signed-off-by: Hariprasad Kelam <hkelam@marvell.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250407070341.2765426-1-hkelam@marvell.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>