If nfs4_client is in courtesy state then there is no point to send
the callback. This causes nfsd4_shutdown_callback to hang since
cl_cb_inflight is not 0. This hang lasts about 15 minutes until TCP
notifies NFSD that the connection was dropped.
This patch modifies nfsd4_run_cb_work to skip the RPC call if
nfs4_client is in courtesy state.
Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Fixes: 66af25799940 ("NFSD: add courteous server support for thread with only delegation") Cc: stable@vger.kernel.org Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If getting acl_default fails, acl_access and acl_default will be released
simultaneously. However, acl_access will still retain a pointer pointing
to the released posix_acl, which will trigger a WARNING in
nfs3svc_release_getacl like this:
Clear acl_access/acl_default after posix_acl_release is called to prevent
UAF from being triggered.
Fixes: a257cdd0e217 ("[PATCH] NFSD: Add server support for NFSv3 ACLs.") Cc: stable@vger.kernel.org Link: https://lore.kernel.org/all/20241107014705.2509463-1-lilingfeng@huaweicloud.com/ Signed-off-by: Li Lingfeng <lilingfeng3@huawei.com> Reviewed-by: Rick Macklem <rmacklem@uoguelph.ca> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Currently the skb size after coalescing is only limited by the skb
layout (the skb must not carry frag_list). A single coalesced skb
covering several MSS can potentially fill completely the receive
buffer. In such a case, the snd win will zero until the receive buffer
will be empty again, affecting tput badly.
Fixes: 8268ed4c9d19 ("mptcp: introduce and use mptcp_try_coalesce()") Cc: stable@vger.kernel.org # please delay 2 weeks after 6.13-final release Signed-off-by: Paolo Abeni <pabeni@redhat.com> Reviewed-by: Mat Martineau <martineau@kernel.org> Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org> Link: https://patch.msgid.link/20241230-net-mptcp-rbuf-fixes-v1-3-8608af434ceb@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
==================================================================
BUG: KASAN: slab-out-of-bounds in ocfs2_match fs/ocfs2/dir.c:334
[inline]
BUG: KASAN: slab-out-of-bounds in ocfs2_search_dirblock+0x283/0x6e0
fs/ocfs2/dir.c:367
Read of size 1 at addr ffff88804d8b9982 by task syz-executor.2/14802
The two reports are all caused invalid negative i_size of dir inode. For
ocfs2, dir_inode can't be negative or zero.
Here add a check in which is called by ocfs2_check_dir_for_entry(). It
fixes the second report as ocfs2_check_dir_for_entry() must be called
before ocfs2_prepare_dir_for_insert(). Also set a up limit for dir with
OCFS2_INLINE_DATA_FL. The i_size can't be great than blocksize.
Declare ftrace_get_parent_ra_addr() as static to suppress clang
compiler warning that 'no previous prototype'. This function is
not intended to be called from other parts.
Fix follow error with clang-19:
arch/mips/kernel/ftrace.c:251:15: error: no previous prototype for function 'ftrace_get_parent_ra_addr' [-Werror,-Wmissing-prototypes]
251 | unsigned long ftrace_get_parent_ra_addr(unsigned long self_ra, unsigned long
| ^
arch/mips/kernel/ftrace.c:251:1: note: declare 'static' if the function is not intended to be used outside of this translation unit
251 | unsigned long ftrace_get_parent_ra_addr(unsigned long self_ra, unsigned long
| ^
| static
1 error generated.
Signed-off-by: WangYuli <wangyuli@uniontech.com> Acked-by: Masami Hiramatsu (Google) <mhiramat@kernel.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The ioctl and sysfs handlers unconditionally call the ->enable callback.
Not all drivers implement that callback, leading to NULL dereferences.
Example of affected drivers: ptp_s390.c, ptp_vclock.c and ptp_mock.c.
Instead use a dummy callback if no better was specified by the driver.
Fixes: d94ba80ebbea ("ptp: Added a brand new class driver for ptp clocks.") Cc: stable@vger.kernel.org Signed-off-by: Thomas Weißschuh <linux@weissschuh.net> Acked-by: Richard Cochran <richardcochran@gmail.com> Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com> Link: https://patch.msgid.link/20250123-ptp-enable-v1-1-b015834d3a47@weissschuh.net Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The NCSI state machine as it's currently implemented assumes that
transition to the next logical state is performed either explicitly by
calling `schedule_work(&ndp->work)` to re-queue itself or implicitly
after processing the predefined (ndp->pending_req_num) number of
replies. Thus to avoid the configuration FSM from advancing prematurely
and getting out of sync with the process it's essential to not skip
waiting for a reply.
This patch makes the code wait for reception of the Deselect Package
response for the last package probed before proceeding to channel
configuration.
Thanks go to Potin Lai and Cosmo Chou for the initial investigation and
testing.
Fixes: 8e13f70be05e ("net/ncsi: Probe single packages to avoid conflict") Cc: stable@vger.kernel.org Signed-off-by: Paul Fertser <fercerpav@gmail.com> Link: https://patch.msgid.link/20250116152900.8656-1-fercerpav@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
For registered buffers, fastrpc driver sends the buffer information
to remote subsystem. There is a problem with current implementation
where the page address is being sent with an offset leading to
improper buffer address on DSP. This is leads to functional failures
as DSP expects base address in page information and extracts offset
information from remote arguments. Mask the offset and pass the base
page address to DSP.
This issue is observed is a corner case when some buffer which is registered
with fastrpc framework is passed with some offset by user and then the DSP
implementation tried to read the data. As DSP expects base address and takes
care of offsetting with remote arguments, passing an offsetted address will
result in some unexpected data read in DSP.
All generic usecases usually pass the buffer as it is hence is problem is
not usually observed. If someone tries to pass offsetted buffer and then
tries to compare data at HLOS and DSP end, then the ambiguity will be observed.
The function do_otp_read() does not set the output parameter *retlen,
which is expected to contain the number of bytes actually read.
As a result, in onenand_otp_walk(), the tmp_retlen variable remains
uninitialized after calling do_otp_walk() and used to change
the values of the buf, len and retlen variables.
Found by Linux Verification Center (linuxtesting.org) with SVACE.
The "pipe" variable is a u8 which comes from the network. If it's more
than 127, then it results in memory corruption in the caller,
nci_hci_connect_gate().
Cc: stable@vger.kernel.org Fixes: a1b0b9415817 ("NFC: nci: Create pipe on specific gate in nci_hci_connect_gate") Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org> Reviewed-by: Simon Horman <horms@kernel.org> Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Link: https://patch.msgid.link/bcf5453b-7204-4297-9c20-4d8c7dacf586@stanley.mountain Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Since nilfs_bmap_lookup_contig() in nilfs_fiemap() calculates its result
by being prepared to go through potentially maxblocks == INT_MAX blocks,
the value in n may experience an overflow caused by left shift of blkbits.
While it is extremely unlikely to occur, play it safe and cast right hand
expression to wider type to mitigate the issue.
Found by Linux Verification Center (linuxtesting.org) with static analysis
tool SVACE.
Mark did a conversion of ocfs2 to use folios and sent it to me as a
giant patch for review ;-)
So I've redone it as individual patches, and credited Mark for the patches
where his code is substantially the same. It's not a bad way to do it;
his patch had some bugs and my patches had some bugs. Hopefully all our
bugs were different from each other. And hopefully Mark likes all the
changes I made to his code!
This patch (of 23):
If we can't read the buffer, be sure to unlock the page before returning.
Link: https://lkml.kernel.org/r/20241205171653.3179945-1-willy@infradead.org Link: https://lkml.kernel.org/r/20241205171653.3179945-2-willy@infradead.org Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org> Reviewed-by: Joseph Qi <joseph.qi@linux.alibaba.com> Cc: Mark Fasheh <mark@fasheh.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Junxiao Bi <junxiao.bi@oracle.com> Cc: Changwei Ge <gechangwei@live.cn> Cc: Jun Piao <piaojun@huawei.com> Cc: Mark Tinguely <mark.tinguely@oracle.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 23aab037106d ("ocfs2: fix UBSAN warning in ocfs2_verify_volume()")
introduced a regression bug. The blksz_bits value is already converted to
CPU endian in the previous code; therefore, the code shouldn't use
le32_to_cpu() anymore.
Link: https://lkml.kernel.org/r/20250121112204.12834-1-heming.zhao@suse.com Fixes: 23aab037106d ("ocfs2: fix UBSAN warning in ocfs2_verify_volume()") Signed-off-by: Heming Zhao <heming.zhao@suse.com> Reviewed-by: Joseph Qi <joseph.qi@linux.alibaba.com> Cc: Mark Fasheh <mark@fasheh.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Junxiao Bi <junxiao.bi@oracle.com> Cc: Changwei Ge <gechangwei@live.cn> Cc: Jun Piao <piaojun@huawei.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
count and offset are passed from user space and not checked, only
offset is capped to 40 bits, which can be used to read/write out of
bounds of the device.
Fixes: 6e3f26456009 (“vfio/platform: read and write support for the device fd”) Cc: stable@vger.kernel.org Reported-by: Mostafa Saleh <smostafa@google.com> Reviewed-by: Eric Auger <eric.auger@redhat.com> Reviewed-by: Mostafa Saleh <smostafa@google.com> Tested-by: Mostafa Saleh <smostafa@google.com> Signed-off-by: Alex Williamson <alex.williamson@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When __nvmem_cell_entry_write() is called for an nvmem cell that does
not need bit shifting, it requires that the len parameter exactly
matches the nvmem cell size. However, when the nvmem cell has a nonzero
bit_offset, it was skipping this check.
Accepting values of len larger than the cell size results in
nvmem_cell_prepare_write_buffer() trying to write past the end of a heap
buffer that it allocates. Add a check to avoid that problem and instead
return -EINVAL when len doesn't match the number of bits expected by the
nvmem cell when bit_offset is nonzero.
This check uses cell->nbits in order to allow providing the smaller size
to cells that are shifted into another byte by bit_offset. For example,
a cell with nbits=8 and nonzero bit_offset would have bytes=2 but should
accept a 1-byte write here, although no current callers depend on this.
Fixes: 69aba7948cbe ("nvmem: Add a simple NVMEM framework for consumers") Cc: stable@vger.kernel.org Signed-off-by: Jennifer Berringer <jberring@redhat.com> Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org> Link: https://lore.kernel.org/r/20241230141901.263976-7-srinivas.kandagatla@linaro.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Let the nvmem core know what size the SDAM is, most notably this fixes
the size of /sys/bus/nvmem/devices/spmi_sdam*/nvmem being '0' and makes
user space work with that file.
If we encounter an error when registering alorithms with the crypto
framework, we just bail out and don't unregister the ones we
successfully registered in prior iterations of the loop.
Add code that goes back over the algos and unregisters them before
returning an error from qce_register_algs().
ctrl->handle will only be different than NULL for controls that have
mappings. This is because that assignment is only done inside
uvc_ctrl_set() for mapped controls.
Cc: stable@vger.kernel.org Fixes: e5225c820c05 ("media: uvcvideo: Send a control event when a Control Change interrupt arrives") Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com> Reviewed-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Ricardo Ribalda <ribalda@chromium.org> Link: https://lore.kernel.org/r/20241203-uvc-fix-async-v6-2-26c867231118@chromium.org Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The channel index is off by one unit if AS73211_SCAN_MASK_ALL is not
set (optimized path for color channel readings), and it must be shifted
instead of leaving an empty channel for the temperature when it is off.
Once the channel index is fixed, the uninitialized channel must be set
to zero to avoid pushing uninitialized data.
Add available_scan_masks for all channels and only-color channels to let
the IIO core demux and repack the enabled channels.
Cc: stable@vger.kernel.org Fixes: 403e5586b52e ("iio: light: as73211: New driver") Tested-by: Christian Eggers <ceggers@arri.de> Signed-off-by: Javier Carrasco <javier.carrasco.cruz@gmail.com> Link: https://patch.msgid.link/20241214-iio_memset_scan_holes-v4-1-260b395b8ed5@gmail.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
GCC 15 changed the default C standard version to C23, which should not
have impacted the kernel because it requests the gnu11 standard via
'-std=' in the main Makefile. However, the x86 compressed boot Makefile
uses its own set of KBUILD_CFLAGS without a '-std=' value (i.e., using
the default), resulting in errors from the kernel's definitions of bool,
true, and false in stddef.h, which are reserved keywords under C23.
./include/linux/stddef.h:11:9: error: expected identifier before ‘false’
11 | false = 0,
./include/linux/types.h:35:33: error: two or more data types in declaration specifiers
35 | typedef _Bool bool;
Set '-std=gnu11' in the x86 compressed boot Makefile to resolve the
error and consistently use the same C standard version for the entire
kernel.
-Wenum-enum-conversion was strengthened in clang-19 to warn for C, which
caused the kernel to move it to W=1 in commit 75b5ab134bb5 ("kbuild:
Move -Wenum-{compare-conditional,enum-conversion} into W=1") because
there were numerous instances that would break builds with -Werror.
Unfortunately, this is not a full solution, as more and more developers,
subsystems, and distributors are building with W=1 as well, so they
continue to see the numerous instances of this warning.
Since the move to W=1, there have not been many new instances that have
appeared through various build reports and the ones that have appeared
seem to be following similar existing patterns, suggesting that most
instances of this warning will not be real issues. The only alternatives
for silencing this warning are adding casts (which is generally seen as
an ugly practice) or refactoring the enums to macro defines or a unified
enum (which may be undesirable because of type safety in other parts of
the code).
Move the warning to W=2, where warnings that occur frequently but may be
relevant should reside.
In StorVSC, payload->range.len is used to indicate if this SCSI command
carries payload. This data is allocated as part of the private driver data
by the upper layer and may get passed to lower driver uninitialized.
For example, the SCSI error handling mid layer may send TEST_UNIT_READY or
REQUEST_SENSE while reusing the buffer from a failed command. The private
data section may have stale data from the previous command.
If the SCSI command doesn't carry payload, the driver may use this value as
is for communicating with host, resulting in possible corruption.
Fix this by always initializing this value.
Fixes: be0cf6ca301c ("scsi: storvsc: Set the tablesize based on the information given by the host") Cc: stable@kernel.org Tested-by: Roman Kisel <romank@linux.microsoft.com> Reviewed-by: Roman Kisel <romank@linux.microsoft.com> Reviewed-by: Michael Kelley <mhklinux@outlook.com> Signed-off-by: Long Li <longli@microsoft.com> Link: https://lore.kernel.org/r/1737601642-7759-1-git-send-email-longli@linuxonhyperv.com Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Positivo C6400 is equipped with ALC269VB, and it needs
ALC269VB_FIXUP_ASUS_ZENBOOK quirk to make its headset mic work.
Also must to limits the microphone boost.
dm-crypt uses tag_offset to index the integrity metadata for each crypt
sector. When the initial crypt_convert() returns BLK_STS_DEV_RESOURCE,
dm-crypt will try to continue the crypt/decrypt procedure in a kworker.
However, it resets tag_offset as zero instead of using the tag_offset
related with current sector. It may return unexpected data when using
random IV or return unexpected integrity related error.
Fix the problem by tracking tag_offset in per-IO convert_context.
Therefore, when the crypt/decrypt procedure continues in a kworker, it
could use the next tag_offset saved in convert_context.
Fixes: 8abec36d1274 ("dm crypt: do not wait for backlogged crypto request completion in softirq") Cc: stable@vger.kernel.org Signed-off-by: Hou Tao <houtao1@huawei.com> Signed-off-by: Mikulas Patocka <mpatocka@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The updates of io->sector are the leftovers when dm-crypt allocated
pages for partial write request. However, since commit cf2f1abfbd0db
("dm crypt: don't allocate pages for a partial request"), there is no
partial request anymore.
After the introduction of write request rb-tree, the updates of
io->sectors may interfere the insertion procedure, because ->sectors of
these write requests which have already been added in the rb-tree may be
changed during the insertion of new write request.
Fix it by removing these buggy updates of io->sectors. Considering these
updates only effect the write request rb-tree, the commit which
introduces the write request rb-tree is used as the fix tag.
The PE Reset State "0" returned by RTAS calls
"ibm_read_slot_reset_[state|state2]" indicates that the reset is
deactivated and the PE is in a state where MMIO and DMA are allowed.
However, the current implementation of "pseries_eeh_get_state()" does
not reflect this, causing drivers to incorrectly assume that MMIO and
DMA operations cannot be resumed.
The userspace drivers as a part of EEH recovery using VFIO ioctls fail
to detect when the recovery process is complete. The VFIO_EEH_PE_GET_STATE
ioctl does not report the expected EEH_PE_STATE_NORMAL state, preventing
userspace drivers from functioning properly on pseries systems.
The patch addresses this issue by updating 'pseries_eeh_get_state()'
to include "EEH_STATE_MMIO_ENABLED" and "EEH_STATE_DMA_ENABLED" in
the result mask for PE Reset State "0". This ensures correct state
reporting to the callers, aligning the behavior with the PAPR specification
and fixing the bug in EEH recovery for VFIO user workflows.
In the sh-sci driver, sci_ports[0] is used by earlycon. If the earlycon is
still active when sci_probe() is called and the new serial port is supposed
to map to sci_ports[0], return -EBUSY to prevent breaking the earlycon.
This situation should occurs in debug scenarios, and users should be
aware of the potential conflict.
The port_cfg object is used by serial_console_write(), which serves as
the write function for the earlycon device. Marking port_cfg as __initdata
causes it to be freed after kernel initialization, resulting in earlycon
becoming unavailable thereafter. Remove the __initdata macro from port_cfg
to resolve this issue.
The firmware used on MSM8916 exposes SOCINFO_VERSION(0, 8), which does not
have support for the serial_num field in the socinfo struct. There is an
existing check to avoid exposing the serial number in that case, but it's
not correct: When checking the item_size returned by SMEM, we need to make
sure the *end* of the serial_num is within bounds, instead of comparing
with the *start* offset. The serial_number currently exposed on MSM8916
devices is just an out of bounds read of whatever comes after the socinfo
struct in SMEM.
Fix this by changing offsetof() to offsetofend(), so that the size of the
field is also taken into account.
The duplicate kmalloc here is causing memory leak. The request
preparation in bot_send_write_request is also done in
usbg_prepare_w_request. Remove the duplicate work.
We submitted the command with TARGET_SCF_ACK_KREF, which requires
acknowledgment of command completion. If the command fails, make sure to
decrement the ref count.
Since brcmf_get_ip() can and actually will return NULL in this case the
call to brcmf_txfinalize() will result in a NULL pointer dereference inside
brcmf_txfinalize() when trying to update ifp->ndev->stats.tx_errors.
This will only happen if a flowring still has an skb.
Although the NULL pointer dereference has only been seen when trying to
update the tx statistic, all other uses of the ifp pointer have been
guarded as well with an early return if ifp is NULL.
RTL8821AE is stuck transmitting at the lowest rate allowed by the rate
mask. This is because the firmware doesn't know the device is connected
to a network.
Fix the macros SET_H2CCMD_MSRRPT_PARM_OPMODE and
SET_H2CCMD_MSRRPT_PARM_MACID_IND to work on the first byte of __cmd,
not the second. Now the firmware is correctly notified when the device
is connected to a network and it activates the rate control.
The hid-sensor-hub creates the individual device structs and transfers them
to the created mfd platform-devices via the platform_data in the mfd_cell.
Before e651a1da442a ("HID: hid-sensor-hub: Allow parallel synchronous reads")
the sensor-hub was managing access centrally, with one "completion" in the
hub's data structure, which needed to be finished on removal at the latest.
The mentioned commit then moved this central management to each hid sensor
device, resulting on a completion in each struct hid_sensor_hub_device.
The remove procedure was adapted to go through all sensor devices and
finish any pending "completion".
What this didn't take into account was, platform_device_add_data() that is
used by mfd_add{_hotplug}_devices() does a kmemdup on the submitted
platform-data. So the data the platform-device gets is a copy of the
original data, meaning that the device worked on a different completion
than what sensor_hub_remove() currently wants to access.
To fix that, use device_for_each_child() to go through each child-device
similar to how mfd_remove_devices() unregisters the devices later and
with that get the live platform_data to finalize the correct completion.
The comparison function cmpworker() violates the C standard's
requirements for qsort() comparison functions, which mandate symmetry
and transitivity:
Symmetry: If x < y, then y > x.
Transitivity: If x < y and y < z, then x < z.
In its current implementation, cmpworker() incorrectly returns 0 when
w1->tid < w2->tid, which breaks both symmetry and transitivity. This
violation causes undefined behavior, potentially leading to issues such
as memory corruption in glibc [1].
Fix the issue by returning -1 when w1->tid < w2->tid, ensuring
compliance with the C standard and preventing undefined behavior.
GCC 15 changed the default C standard version to C23, which should not
have impacted the kernel because it requests the gnu11 standard via
'-std=' in the main Makefile. However, the EFI libstub Makefile uses its
own set of KBUILD_CFLAGS for x86 without a '-std=' value (i.e., using
the default), resulting in errors from the kernel's definitions of bool,
true, and false in stddef.h, which are reserved keywords under C23.
./include/linux/stddef.h:11:9: error: expected identifier before ‘false’
11 | false = 0,
./include/linux/types.h:35:33: error: two or more data types in declaration specifiers
35 | typedef _Bool bool;
Set '-std=gnu11' in the x86 cflags to resolve the error and consistently
use the same C standard version for the entire kernel. All other
architectures reuse KBUILD_CFLAGS from the rest of the kernel, so this
issue is not visible for them.
blkcg_fill_root_iostats() iterates over @block_class's devices by
class_dev_iter_(init|next)(), but does not end iterating with
class_dev_iter_exit(), so causes the class's subsystem refcount leakage.
Fix by ending the iterating with class_dev_iter_exit().
Fixes: ef45fe470e1e ("blk-cgroup: show global disk stats in root cgroup io.stat") Reviewed-by: Michal Koutný <mkoutny@suse.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: stable@vger.kernel.org Acked-by: Tejun Heo <tj@kernel.org> Signed-off-by: Zijun Hu <quic_zijuhu@quicinc.com> Link: https://lore.kernel.org/r/20250105-class_fix-v6-2-3a2f1768d4d4@quicinc.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit c45ae598fc16 ("clk: qcom: support for alpha mode configuration")
added support for configuring alpha mode, but it seems that the feature
was never working in practice.
The value of the alpha_{en,mode}_mask members of the configuration gets
added to the value parameter passed to the regmap_update_bits() function,
however the same values are not getting applied to the bitmask. As the
result, the respective bits in the USER_CTL register are never modifed
which leads to improper configuration of several PLLs.
The following table shows the PLL configurations where the 'alpha_en_mask'
member is set and which are passed as a parameter for the
clk_alpha_pll_configure() function. In the table the 'expected rate' column
shows the rate the PLL should run at with the given configuration, and
the 'real rate' column shows the rate the PLL runs at actually. The real
rates has been verified on hardwareOn IPQ* platforms, on other platforms,
those are computed values only.
As it can be seen from the above, there are several PLLs which are
configured incorrectly.
Change the code to apply both 'alpha_en_mask' and 'alpha_mode_mask'
values to the bitmask in order to configure the alpha mode correctly.
Applying the 'alpha_en_mask' fixes the initial rate of the PLLs showed
in the table above. Since the 'alpha_mode_mask' is not used by any driver
currently, that part of the change causes no functional changes.
While testing the MMC nodes proposed in [1], it was noted that mmc0/1
would fail to initialize, with "mmc: fatal err update clk timeout" in
the kernel logs. A closer look at the clock definitions showed that the MMC
MPs had the "CLK_SET_RATE_NO_REPARENT" flag set. No reason was given for
adding this flag in the first place, and its original purpose is unknown,
but it doesn't seem to make sense and results in severe limitations to MMC
speeds. Thus, remove this flag from the 3 MMC MPs.
One of the possible ways to enable the input MTU auto-selection for L2CAP
connections is supposed to be through passing a special "0" value for it
as a socket option. Commit [1] added one of those into avdtp. However, it
simply wouldn't work because the kernel still treats the specified value
as invalid and denies the setting attempt. Recorded BlueZ logs include the
following:
Found by Linux Verification Center (linuxtesting.org).
Fixes: 4b6e228e297b ("Bluetooth: Auto tune if input MTU is set to 0") Cc: stable@vger.kernel.org Signed-off-by: Fedor Pchelkin <pchelkin@ispras.ru> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A NULL sock pointer is passed into l2cap_sock_alloc() when it is called
from l2cap_sock_new_connection_cb() and the error handling paths should
also be aware of it.
Seemingly a more elegant solution would be to swap bt_sock_alloc() and
l2cap_chan_create() calls since they are not interdependent to that moment
but then l2cap_chan_create() adds the soon to be deallocated and still
dummy-initialized channel to the global list accessible by many L2CAP
paths. The channel would be removed from the list in short period of time
but be a bit more straight-forward here and just check for NULL instead of
changing the order of function calls.
Found by Linux Verification Center (linuxtesting.org) with SVACE static
analysis tool.
Fixes: 7c4f78cdb8e7 ("Bluetooth: L2CAP: do not leave dangling sk pointer on error in l2cap_sock_create()") Cc: stable@vger.kernel.org Signed-off-by: Fedor Pchelkin <pchelkin@ispras.ru> Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We try to reuse the same vsie page when re-executing the vsie with a
given SCB address. The result is that we use the same shadow SCB --
residing in the vsie page -- and can avoid flushing the TLB when
re-running the vsie on a CPU.
So, when we allocate a fresh vsie page, or when we reuse a vsie page for
a different SCB address -- reusing the shadow SCB in different context --
we set ihcpu=0xffff to trigger the flush.
However, after we looked up the SCB address in the radix tree, but before
we grabbed the vsie page by raising the refcount to 2, someone could reuse
the vsie page for a different SCB address, adjusting page->index and the
radix tree. In that case, we would be reusing the vsie page with a
wrong page->index.
Another corner case is that we might set the SCB address for a vsie
page, but fail the insertion into the radix tree. Whoever would reuse
that page would remove the corresponding radix tree entry -- which might
now be a valid entry pointing at another page, resulting in the wrong
vsie page getting removed from the radix tree.
Let's handle such races better, by validating that the SCB address of a
vsie page didn't change after we grabbed it (not reuse for a different
SCB; the alternative would be performing another tree lookup), and by
setting the SCB address to invalid until the insertion in the tree
succeeded (SCB addresses are aligned to 512, so ULONG_MAX is invalid).
These scenarios are rare, the effects a bit unclear, and these issues were
only found by code inspection. Let's CC stable to be safe.
Fixes: a3508fbe9dc6 ("KVM: s390: vsie: initial support for nested virtualization") Cc: stable@vger.kernel.org Signed-off-by: David Hildenbrand <david@redhat.com> Reviewed-by: Claudio Imbrenda <imbrenda@linux.ibm.com> Reviewed-by: Christoph Schlameuss <schlameuss@linux.ibm.com> Tested-by: Christoph Schlameuss <schlameuss@linux.ibm.com>
Message-ID: <20250107154344.1003072-2-david@redhat.com> Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Explicitly verify the target vCPU is fully online _prior_ to clamping the
index in kvm_get_vcpu(). If the index is "bad", the nospec clamping will
generate '0', i.e. KVM will return vCPU0 instead of NULL.
In practice, the bug is unlikely to cause problems, as it will only come
into play if userspace or the guest is buggy or misbehaving, e.g. KVM may
send interrupts to vCPU0 instead of dropping them on the floor.
However, returning vCPU0 when it shouldn't exist per online_vcpus is
problematic now that KVM uses an xarray for the vCPUs array, as KVM needs
to insert into the xarray before publishing the vCPU to userspace (see
commit c5b077549136 ("KVM: Convert the kvm->vcpus array to a xarray")),
i.e. before vCPU creation is guaranteed to succeed.
As a result, incorrectly providing access to vCPU0 will trigger a
use-after-free if vCPU0 is dereferenced and kvm_vm_ioctl_create_vcpu()
bails out of vCPU creation due to an error and frees vCPU0. Commit afb2acb2e3a3 ("KVM: Fix vcpu_array[0] races") papered over that issue, but
in doing so introduced an unsolvable teardown conundrum. Preventing
accesses to vCPU0 before it's fully online will allow reverting commit afb2acb2e3a3, without re-introducing the vcpu_array[0] UAF race.
Fixes: 1d487e9bf8ba ("KVM: fix spectrev1 gadgets") Cc: stable@vger.kernel.org Cc: Will Deacon <will@kernel.org> Cc: Michal Luczaj <mhal@rbox.co> Reviewed-by: Pankaj Gupta <pankaj.gupta@amd.com> Acked-by: Will Deacon <will@kernel.org> Link: https://lore.kernel.org/r/20241009150455.1057573-2-seanjc@google.com Signed-off-by: Sean Christopherson <seanjc@google.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
During mass manufacturing, we noticed the mmc_rx_crc_error counter,
as reported by "ethtool -S eth0 | grep mmc_rx_crc_error", to increase
above zero during nuttcp speedtests. Most of the time, this did not
affect the achieved speed, but it prompted this investigation.
Cycling through the rx_delay range on six boards (see table below) of
various ages shows that there is a large good region from 0x12 to 0x35
where we see zero crc errors on all tested boards.
The old rx_delay value (0x10) seems to have always been on the edge for
the KSZ9031RNX that is usually placed on Puma.
Choose "rx_delay = 0x23" to put us smack in the middle of the good
region. This works fine as well with the KSZ9131RNX PHY that was used
for a small number of boards during the COVID chip shortages.
Board S/N PHY rx_delay good region
--------- --- --------------------
Puma TT0069903 KSZ9031RNX 0x11 0x35
Puma TT0157733 KSZ9031RNX 0x11 0x35
Puma TT0681551 KSZ9031RNX 0x12 0x37
Puma TT0681156 KSZ9031RNX 0x10 0x38
Puma 17496030079 KSZ9031RNX 0x10 0x37 (Puma v1.2 from 2017)
Puma TT0681720 KSZ9131RNX 0x02 0x39 (alternative PHY used in very few boards)
Intersection of good regions = 0x12 0x35
Middle of good region = 0x23
Most of these sizes and counts are capped at 256MB so the math doesn't
result in an integer overflow. The "relocs" count needs to be checked
as well. Otherwise on 32bit systems the calculation of "full_data"
could be wrong.
Including m68k's <asm/raw_io.h> in vga.h on nommu platforms results
in conflicting defines with io_no.h for various I/O macros from the
__raw_read and __raw_write families. An example error is
In file included from arch/m68k/include/asm/vga.h:12,
from include/video/vga.h:22,
from include/linux/vgaarb.h:34,
from drivers/video/aperture.c:12:
>> arch/m68k/include/asm/raw_io.h:39: warning: "__raw_readb" redefined
39 | #define __raw_readb in_8
|
In file included from arch/m68k/include/asm/io.h:6,
from include/linux/io.h:13,
from include/linux/irq.h:20,
from include/asm-generic/hardirq.h:17,
from ./arch/m68k/include/generated/asm/hardirq.h:1,
from include/linux/hardirq.h:11,
from include/linux/interrupt.h:11,
from include/linux/trace_recursion.h:5,
from include/linux/ftrace.h:10,
from include/linux/kprobes.h:28,
from include/linux/kgdb.h:19,
from include/linux/fb.h:6,
from drivers/video/aperture.c:5:
arch/m68k/include/asm/io_no.h:16: note: this is the location of the previous definition
16 | #define __raw_readb(addr) \
|
Include <asm/io.h>, which avoids raw_io.h on nommu platforms.
Also change the defined values of some of the read/write symbols in
vga.h to __raw_read/__raw_write as the raw_in/raw_out symbols are not
generally available.
I struggle to explain dividing an ARRAY_SIZE() by the size of an element
once again. As the latter equals to 2, only the half of EEPROM was ever
written. Drop the unexplainable division and write full ARRAY_SIZE().
Cc: stable@vger.kernel.org Fixes: 7a8685accb95 ("leds: lp8860: Introduce TI lp8860 4 channel LED driver") Signed-off-by: Alexander Sverdlin <alexander.sverdlin@siemens.com> Link: https://lore.kernel.org/r/20241114101402.2562878-1-alexander.sverdlin@siemens.com Signed-off-by: Lee Jones <lee@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The driver generates following warning when regulator support isn't
enabled in the kernel. Fix it.
drivers/cpufreq/s3c64xx-cpufreq.c: In function 's3c64xx_cpufreq_set_target':
>> drivers/cpufreq/s3c64xx-cpufreq.c:55:22: warning: variable 'old_freq' set but not used [-Wunused-but-set-variable]
55 | unsigned int old_freq, new_freq;
| ^~~~~~~~
>> drivers/cpufreq/s3c64xx-cpufreq.c:54:30: warning: variable 'dvfs' set but not used [-Wunused-but-set-variable]
54 | struct s3c64xx_dvfs *dvfs;
| ^~~~
Reported-by: kernel test robot <lkp@intel.com> Closes: https://lore.kernel.org/oe-kbuild-all/202501191803.CtfT7b2o-lkp@intel.com/ Cc: 5.4+ <stable@vger.kernel.org> # v5.4+ Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Link: https://patch.msgid.link/236b227e929e5adc04d1e9e7af6845a46c8e9432.1737525916.git.viresh.kumar@linaro.org Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
qdisc_tree_reduce_backlog() notifies parent qdisc only if child
qdisc becomes empty, therefore we need to reduce the backlog of the
child qdisc before calling it. Otherwise it would miss the opportunity
to call cops->qlen_notify(), in the case of DRR, it resulted in UAF
since DRR uses ->qlen_notify() to maintain its active list.
Fixes: f8d4bc455047 ("net/sched: netem: account for backlog updates from child qdisc") Cc: Martin Ottens <martin.ottens@fau.de> Reported-by: Mingi Cho <mincho@theori.io> Signed-off-by: Cong Wang <cong.wang@bytedance.com> Link: https://patch.msgid.link/20250204005841.223511-4-xiyou.wangcong@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
xen_hypercall_hvm(), which is used when running as a Xen PVH guest at
most only once during early boot, is clobbering %rbx. Depending on
whether the caller relies on %rbx to be preserved across the call or
not, this clobbering might result in an early crash of the system.
This can be avoided by using an already saved register instead of %rbx.
Fixes: b4845bb63838 ("x86/xen: add central hypercall functions") Signed-off-by: Juergen Gross <jgross@suse.com> Reviewed-by: Jan Beulich <jbeulich@suse.com> Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com> Signed-off-by: Juergen Gross <jgross@suse.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Firmware deinitialization performs MMIO accesses which are not
necessary if the device has already been removed. In some cases,
these accesses happen via readx_poll_timeout_atomic which ends up
timing out, resulting in a warning at hw_atl2_utils_fw.c:112:
The GPIO drivers with latch interrupt support (typically types starting
with PCAL) have interrupt status registers to determine which particular
inputs have caused an interrupt. Unfortunately there is no atomic
operation to read these registers and clear the interrupt. Clearing the
interrupt is done by reading the input registers.
The code was reading the interrupt status registers, and then reading
the input registers. If an input changed between these two events it was
lost.
The solution in this patch is to revert to the non-latch version of
code, i.e. remembering the previous input status, and looking for the
changes. This system results in no more I2C transfers, so is no slower.
The latch property of the device still means interrupts will still be
noticed if the input changes back to its initial state.
Fixes: 44896beae605 ("gpio: pca953x: add PCAL9535 interrupt support for Galileo Gen2") Signed-off-by: Mark Tomlinson <mark.tomlinson@alliedtelesis.co.nz> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Link: https://lore.kernel.org/r/20240606033102.2271916-1-mark.tomlinson@alliedtelesis.co.nz Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Commit 4094871db1d6 ("udp: only do GSO if # of segs > 1") avoided GSO
for small packets. But the kernel currently dismisses GSO requests only
after checking MTU/PMTU on gso_size. This means any packets, regardless
of their payload sizes, could be dropped when PMTU becomes smaller than
requested gso_size. We encountered this issue in production and it
caused a reliability problem that new QUIC connection cannot be
established before PMTU cache expired, while non GSO sockets still
worked fine at the same time.
Ideally, do not check any GSO related constraints when payload size is
smaller than requested gso_size, and return EMSGSIZE instead of EINVAL
on MTU/PMTU check failure to be more specific on the error cause.
Fixes: 4094871db1d6 ("udp: only do GSO if # of segs > 1") Signed-off-by: Yan Zhai <yan@cloudflare.com> Suggested-by: Willem de Bruijn <willemdebruijn.kernel@gmail.com> Reviewed-by: Willem de Bruijn <willemb@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Disable PCIe AER on the tg3 device on system reboot on a limited
list of Dell PowerEdge systems. This prevents a fatal PCIe AER event
on the tg3 device during the ACPI _PTS (prepare to sleep) method for
S5 on those systems. The _PTS is invoked by acpi_enter_sleep_state_prep()
as part of the kernel's reboot sequence as a result of commit 38f34dba806a ("PM: ACPI: reboot: Reinstate S5 for reboot").
There was an earlier fix for this problem by commit 2ca1c94ce0b6
("tg3: Disable tg3 device on system reboot to avoid triggering AER").
But it was discovered that this earlier fix caused a reboot hang
when some Dell PowerEdge servers were booted via ipxe. To address
this reboot hang, the earlier fix was essentially reverted by commit 9fc3bc764334 ("tg3: power down device only on SYSTEM_POWER_OFF").
This re-exposed the tg3 PCIe AER on reboot problem.
This fix is not an ideal solution because the root cause of the AER
is in system firmware. Instead, it's a targeted work-around in the
tg3 driver.
Note also that the PCIe AER must be disabled on the tg3 device even
if the system is configured to use "firmware first" error handling.
V3:
- Fix sparse warning on improper comparison of pdev->current_state
- Adhere to netdev comment style
Fixes: 9fc3bc764334 ("tg3: power down device only on SYSTEM_POWER_OFF") Signed-off-by: Lenny Szubowicz <lszubowi@redhat.com> Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com> Reviewed-by: Simon Horman <horms@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
If the hotplug detect of a display is low for longer than one second
(configurable through drm_dp_cec_unregister_delay), then the CEC adapter
is unregistered since we assume the display was disconnected. If the
HPD went low for less than one second, then we check if the properties
of the CEC adapter have changed, since that indicates that we actually
switch to new hardware and we have to unregister the old CEC device and
register a new one.
Unfortunately, the test for changed properties was written poorly, and
after a new CEC capability was added to the CEC core code the test always
returned true (i.e. the properties had changed).
As a result the CEC device was unregistered and re-registered for every
HPD toggle. If the CEC remote controller integration was also enabled
(CONFIG_MEDIA_CEC_RC was set), then the corresponding input device was
also unregistered and re-registered. As a result the input device in
/sys would keep incrementing its number, e.g.:
Since short HPD toggles are common, the number could over time get into
the thousands.
While not a serious issue (i.e. nothing crashes), it is not intended
to work that way.
This patch changes the test so that it only checks for the single CEC
capability that can actually change, and it ignores any other
capabilities, so this is now safe as well if new caps are added in
the future.
With the changed test the bit under #ifndef CONFIG_MEDIA_CEC_RC can be
dropped as well, so that's a nice cleanup.
When the set feature attempts fails with any NVME status code set in
nvme_set_queue_count, the function still report success. Though the
numbers of queues set to 0. This is done to support controllers in
degraded state (the admin queue is still up and running but no IO
queues).
Though there is an exception. When nvme_set_features reports an host
path error, nvme_set_queue_count should propagate this error as the
connectivity is lost, which means also the admin queue is not working
anymore.
Fixes: 9a0be7abb62f ("nvme: refactor set_queue_count") Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Daniel Wagner <wagi@kernel.org> Signed-off-by: Keith Busch <kbusch@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
If a command is queued to the final usable TRB of a ring segment, the
enqueue pointer is advanced to the subsequent link TRB and no further.
If the command is later aborted, when the abort completion is handled
the dequeue pointer is advanced to the first TRB of the next segment.
If no further commands are queued, xhci_handle_stopped_cmd_ring() sees
the ring pointers unequal and assumes that there is a pending command,
so it calls xhci_mod_cmd_timer() which crashes if cur_cmd was NULL.
Don't attempt timer setup if cur_cmd is NULL. The subsequent doorbell
ring likely is unnecessary too, but it's harmless. Leave it alone.
This is probably Bug 219532, but no confirmation has been received.
The issue has been independently reproduced and confirmed fixed using
a USB MCU programmed to NAK the Status stage of SET_ADDRESS forever.
Everything continued working normally after several prevented crashes.
- The HCD address_device callback now accepts a user-defined timeout value
in milliseconds, providing better control over command execution times.
- The default timeout value for the address_device command has been set
to 5000 ms, aligning with the USB 3.2 specification. However, this
timeout can be adjusted as needed.
- The xhci_setup_device function has been updated to accept the timeout
value, allowing it to specify the maximum wait time for the command
operation to complete.
- The hub driver has also been updated to accommodate the newly added
timeout parameter during the SET_ADDRESS request.
Syzkaller reports [1] encountering a common issue of utilizing a wrong
usb endpoint type during URB submitting stage. This, in turn, triggers
a warning shown below.
For now, enable simple endpoint checking (specifically, bulk and
interrupt eps, testing control one is not essential) to mitigate
the issue with a view to do other related cosmetic changes later,
if they are necessary.
Per the "Processor Specification Update" documentations referred by
the intel-microcode-20240312 release note, this microcode release has
fixed the issue for all affected models.
So don't disable PCID if the microcode is new enough. The precise
minimum microcode revision fixing the issue was provided by Pawan
Intel.
The F11 key on the new Lenovo Thinkpad T14 Gen 5, T16 Gen 3, and P14s
Gen 5 laptops includes a symbol showing a smartphone and a laptop
chained together. According to the user manual, it starts the Microsoft
Phone Link software used to connect to Android/iOS devices and relay
messages/calls or sync data.
As there are no suitable keycodes for this action, introduce a new one.
On a 32bit system the "keylen + sizeof(struct tipc_aead_key)" math could
have an integer wrapping issue. It doesn't matter because the "keylen"
is checked on the next line, but just to make life easier for static
analysis tools, let's re-order these conditions and avoid the integer
overflow.
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org> Reviewed-by: Simon Horman <horms@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Qualcomm regulator supports two power supply modes: HPM and LPM.
Currently, the sdhci-msm.c driver does not set the load to adjust
the current for eMMC and SD. If the regulator dont't set correct
load in LPM state, it will lead to the inability to properly
initialize eMMC and SD.
Set the correct regulator current for eMMC and SD to ensure that the
device can work normally even when the regulator is in LPM.
The GHES driver overrides the panic= setting by force-rebooting the
system after a fatal hw error has been reported. The intent being that
such an error would be reported earlier.
However, this is not optimal when a hard-to-debug issue requires long
time to reproduce and when that happens, the box will get rebooted after
30 seconds and thus destroy the whole hw context of when the error
happened.
So rip out the default GHES panic timeout and honor the global one.
In the panic disabled (panic=0) case, the error will still be logged to
dmesg for later inspection and if panic after a hw error is really
required, then that can be controlled the usual way - use panic= on the
cmdline or set it in the kernel .config's CONFIG_PANIC_TIMEOUT.
Reported-by: Feng Tang <feng.tang@linux.alibaba.com> Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de> Reviewed-by: Feng Tang <feng.tang@linux.alibaba.com> Reviewed-by: Ira Weiny <ira.weiny@intel.com> Link: https://patch.msgid.link/20250113125224.GFZ4UMiNtWIJvgpveU@fat_crate.local Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
When a 400KHz freq is used on this model of ELAN touchpad in Linux,
excessive smoothing (similar to when the touchpad's firmware detects
a noisy signal) is sometimes applied. As some devices' (e.g, Lenovo
V15 G4) ACPI tables specify a 400KHz frequency for this device and
some I2C busses (e.g, Designware I2C) default to a 400KHz freq,
force the speed to 100KHz as a workaround.
For future investigation: This problem may be related to the default
HCNT/LCNT values given by some busses' drivers, because they are not
specified in the aforementioned devices' ACPI tables, and because
the device works without issues on Windows at what is expected to be
a 400KHz frequency. The root cause of the issue is not known.
Signed-off-by: Randolph Ha <rha051117@gmail.com> Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The overflow_work is using system wq to do overflow checks and updates
for PHC device timecounter, which might be overhelmed by other tasks.
But there is dedicated kthread in PTP subsystem designed for such
things. This patch changes the work queue to proper align with PTP
subsystem and to avoid overloading system work queue.
The adjfine() function acts the same way as overflow check worker,
we can postpone ptp aux worker till the next overflow period after
adjfine() was called.
syzbot is reporting too large allocation warning at tomoyo_write_control(),
for one can write a very very long line without new line character. To fix
this warning, I use __GFP_NOWARN rather than checking for KMALLOC_MAX_SIZE,
for practically a valid line should be always shorter than 32KB where the
"too small to fail" memory-allocation rule applies.
One might try to write a valid line that is longer than 32KB, but such
request will likely fail with -ENOMEM. Therefore, I feel that separately
returning -EINVAL when a line is longer than KMALLOC_MAX_SIZE is redundant.
There is no need to distinguish over-32KB and over-KMALLOC_MAX_SIZE.
Reported-by: syzbot+7536f77535e5210a5c76@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=7536f77535e5210a5c76 Reported-by: Leo Stone <leocstone@gmail.com> Closes: https://lkml.kernel.org/r/20241216021459.178759-2-leocstone@gmail.com Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by: Sasha Levin <sashal@kernel.org>
In 'wlc_phy_iqcal_gainparams_nphy()', add gain range check to WARN()
instead of possible out-of-bounds 'tbl_iqcal_gainparams_nphy' access.
Compile tested only.
Found by Linux Verification Center (linuxtesting.org) with SVACE.
Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru> Acked-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Kalle Valo <kvalo@kernel.org> Link: https://patch.msgid.link/20241210070441.836362-1-dmantipov@yandex.ru Signed-off-by: Sasha Levin <sashal@kernel.org>
The card-quirk was added to limit the clock-rate for a card with UHS-mode
support, although let's respect the quirk for non-UHS mode too, to make the
behaviour consistent.
Currently tun checks the group permission even if the user have matched.
Besides going against the usual permission semantic, this has a
very interesting implication: if the tun group is not among the
supplementary groups of the tun user, then effectively no one can
access the tun device. CAP_SYS_ADMIN still can, but its the same as
not setting the tun ownership.
This patch relaxes the group checking so that either the user match
or the group match is enough. This avoids the situation when no one
can access the device even though the ownership is properly set.
Also I simplified the logic by removing the redundant inversions:
tun_not_capable() --> !tun_capable()
Signed-off-by: Stas Sergeev <stsp2@yandex.ru> Reviewed-by: Willem de Bruijn <willemb@google.com> Acked-by: Jason Wang <jasowang@redhat.com> Link: https://patch.msgid.link/20241205073614.294773-1-stsp2@yandex.ru Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
syzbot attempts to write a buffer with a large size to a sysfs entry
with writes handled by handle_policy_update(), triggering a warning
in kmalloc.
Check the size specified for write buffers before allocating.
Reported-by: syzbot+4eb7a741b3216020043a@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=4eb7a741b3216020043a Signed-off-by: Leo Stone <leocstone@gmail.com>
[PM: subject tweak] Signed-off-by: Paul Moore <paul@paul-moore.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Shifting 1 << 31 on a 32-bit int causes signed integer overflow, which
leads to undefined behavior. To prevent this, cast 1 to u32 before
performing the shift, ensuring well-defined behavior.
This change explicitly avoids any potential overflow by ensuring that
the shift occurs on an unsigned 32-bit integer.
Lockdep has a set of configs used to determine the size of the static
arrays that it uses. However, the upper limit that was initially setup
for these configs is too high (30 bit shift). This equates to several
GiB of static memory for individual symbols. Using such high values
leads to linker errors:
$ make defconfig
$ ./scripts/config -e PROVE_LOCKING --set-val LOCKDEP_BITS 30
$ make olddefconfig all
[...]
ld: kernel image bigger than KERNEL_IMAGE_SIZE
ld: section .bss VMA wraps around address space
Adjust the upper limits to the maximum values that avoid these issues.
The need for anything more, likely points to a problem elsewhere. Note
that LOCKDEP_CHAINS_BITS was intentionally left out as its upper limit
had a different symptom and has already been fixed [1].
Reported-by: J. R. Okajima <hooanon05g@gmail.com> Closes: https://lore.kernel.org/all/30795.1620913191@jrobl/ [1] Cc: Peter Zijlstra <peterz@infradead.org> Cc: Boqun Feng <boqun.feng@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Waiman Long <longman@redhat.com> Cc: Will Deacon <will@kernel.org> Acked-by: Waiman Long <longman@redhat.com> Signed-off-by: Carlos Llamas <cmllamas@google.com> Signed-off-by: Boqun Feng <boqun.feng@gmail.com> Link: https://lore.kernel.org/r/20241024183631.643450-2-cmllamas@google.com Signed-off-by: Sasha Levin <sashal@kernel.org>
When steal time exceeds the measured delta when updating clock_task, we
currently try to catch up the excess in future updates.
However, this results in inaccurate run times for the future things using
clock_task, in some situations, as they end up getting additional steal
time that did not actually happen.
This is because there is a window between reading the elapsed time in
update_rq_clock() and sampling the steal time in update_rq_clock_task().
If the VCPU gets preempted between those two points, any additional
steal time is accounted to the outgoing task even though the calculated
delta did not actually contain any of that "stolen" time.
When this race happens, we can end up with steal time that exceeds the
calculated delta, and the previous code would try to catch up that excess
steal time in future clock updates, which is given to the next,
incoming task, even though it did not actually have any time stolen.
This behavior is particularly bad when steal time can be very long,
which we've seen when trying to extend steal time to contain the duration
that the host was suspended [0]. When this happens, clock_task stays
frozen, during which the running task stays running for the whole
duration, since its run time doesn't increase.
However the race can happen even under normal operation.
Ideally we would read the elapsed cpu time and the steal time atomically,
to prevent this race from happening in the first place, but doing so
is non-trivial.
Since the time between those two points isn't otherwise accounted anywhere,
neither to the outgoing task nor the incoming task (because the "end of
outgoing task" and "start of incoming task" timestamps are the same),
I would argue that the right thing to do is to simply drop any excess steal
time, in order to prevent these issues.
This BUG_ON is meant to catch backref cache problems, but these can
arise from either bugs in the backref cache or corruption in the extent
tree. Fix it to be a proper error.
Reviewed-by: Boris Burkov <boris@bur.io> Signed-off-by: Josef Bacik <josef@toxicpanda.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
projects / thirdparty / kernel / stable.git / log
