This driver runs also on Tegra SoCs without a Tegra20 APB DMA controller
(e.g. Tegra234).
Remove the Kconfig dependency on TEGRA20_APB_DMA; in addition, amend the
help text to reflect the fact that this driver works on SoCs different from
Tegra114.
Fixes: bb9667d8187b ("arm64: tegra: Add SPI device tree nodes for Tegra234") Signed-off-by: Francesco Lavra <flavra@baylibre.com> Link: https://patch.msgid.link/20251126095027.4102004-1-flavra@baylibre.com Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
According to [1], the C906 vector registers are 128 bits wide.
The 'thead,vlenb' property specifies the vector register length
in bytes, so its value must be set to 16.
As well as checking that the parent hasn't changed after getting the
lock we need to check that the dentry hasn't been unhashed.
Otherwise we might try to rename something that has been removed.
Reported-by: syzbot+bfc9a0ccf0de47d04e8c@syzkaller.appspotmail.com Fixes: d2c995581c7c ("ovl: Call ovl_create_temp() without lock held.") Signed-off-by: NeilBrown <neil@brown.name> Link: https://patch.msgid.link/176429295510.634289.1552337113663461690@noble.neil.brown.name Tested-by: syzbot+bfc9a0ccf0de47d04e8c@syzkaller.appspotmail.com Reviewed-by: Amir Goldstein <amir73il@gmail.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The allocation of a cell's anonymous key is done in a background thread
along with other cell setup such as doing a DNS upcall. In the reported
bug, this is triggered by afs_parse_source() parsing the device name given
to mount() and calling afs_lookup_cell() with the name of the cell.
The normal key lookup then tries to use the key description on the
anonymous authentication key as the reference for request_key() - but it
may not yet be set and so an oops can happen.
This has been made more likely to happen by the fix for dynamic lookup
failure.
Fix this by firstly allocating a reference name and attaching it to the
afs_cell record when the record is created. It can share the memory
allocation with the cell name (unfortunately it can't just overlap the cell
name by prepending it with "afs@" as the cell name already has a '.'
prepended for other purposes). This reference name is then passed to
request_key().
Secondly, the anon key is now allocated on demand at the point a key is
requested in afs_request_key() if it is not already allocated. A mutex is
used to prevent multiple allocation for a cell.
Thirdly, make afs_request_key_rcu() return NULL if the anonymous key isn't
yet allocated (if we need it) and then the caller can return -ECHILD to
drop out of RCU-mode and afs_request_key() can be called.
Note that the anonymous key is kind of necessary to make the key lookup
cache work as that doesn't currently cache a negative lookup, but it's
probably worth some investigation to see if NULL can be used instead.
Fixes: 330e2c514823 ("afs: Fix dynamic lookup to fail on cell lookup failure") Reported-by: syzbot+41c68824eefb67cdf00c@syzkaller.appspotmail.com Signed-off-by: David Howells <dhowells@redhat.com> Link: https://patch.msgid.link/800328.1764325145@warthog.procyon.org.uk
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
cc: linux-fsdevel@vger.kernel.org Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The "probe_setup_failed" label calls pm_runtime_disable(), but
pm_runtime_enable() was placed after a possible jump to this label.
When cqspi_setup_flash() fails, control jumps to the label without
pm_runtime_enable() being called, leading to unbalanced PM runtime
reference counting.
Move pm_runtime_enable() and associated calls above the first
possible branch to "probe_setup_failed" to ensure balanced
enable/disable calls across all error paths.
Fixes: 30dbc1c8d50f ("spi: cadence-qspi: defer runtime support on socfpga if reset bit is enabled") Signed-off-by: Anurag Dutta <a-dutta@ti.com> Link: https://patch.msgid.link/20251105161146.2019090-2-a-dutta@ti.com Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Fix runtime PM usage count underflow caused by calling
pm_runtime_put_autosuspend() twice with only one corresponding
pm_runtime_get_noresume() call. This triggers the warning:
"Runtime PM usage count underflow!"
Remove the duplicate put call to balance the runtime PM reference
counting.
Fixes: 30dbc1c8d50f ("spi: cadence-qspi: defer runtime support on socfpga if reset bit is enabled") Signed-off-by: Anurag Dutta <a-dutta@ti.com> Link: https://patch.msgid.link/20251105161146.2019090-3-a-dutta@ti.com Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The error status mask for a type 3/4 subspace is used for reading the
error status, and the bitwise inverse is used for clearing the error
with the intent being to preserve any of the non-error bits. However,
we were previously applying the mask to extract the status and then
applying the inverse to the result which ended up clearing all bits.
Instead, store the inverse mask in the preserve mask and then use that
on the original value read from the error status so that only the error
is cleared.
GCE can only fetch the command buffer address from a 32-bit register.
Some SoCs support a 35-bit command buffer address for GCE, which
requires a right shift of 3 bits before setting the address into
the 32-bit register. A comment has been added to the header of
cmdq_get_shift_pa() to explain this requirement.
To prevent the GCE command buffer address from being DMA mapped beyond
its supported bit range, the DMA bit mask for the device is set during
initialization.
Additionally, to ensure the correct shift is applied when setting or
reading the register that stores the GCE command buffer address,
new APIs, cmdq_convert_gce_addr() and cmdq_revert_gce_addr(), have
been introduced for consistent operations on this register.
The variable type for the command buffer address has been standardized
to dma_addr_t to prevent handling issues caused by type mismatches.
Fixes: 0858fde496f8 ("mailbox: cmdq: variablize address shift in platform") Signed-off-by: Jason-JH Lin <jason-jh.lin@mediatek.com> Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Signed-off-by: Jassi Brar <jassisinghbrar@gmail.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
devm_pm_runtime_enable() can fail due to memory allocation.
The current code ignores its return value, potentially causing
pm_runtime_resume_and_get() to operate on uninitialized runtime
PM state.
Check the return value of devm_pm_runtime_enable() and return on failure.
The calibrated timestamp is calculated from the nominal value using the
formula:
ts_gain[ns] ≈ ts_sensitivity - (ts_trim_coeff * val) / 1000.
The values of ts_sensitivity and ts_trim_coeff are not the same for all
devices, so it is necessary to differentiate them based on the part name.
For the correct values please consult the relevant AN.
Fixes: cb3b6b8e1bc0 ("iio: imu: st_lsm6dsx: add odr calibration feature") Signed-off-by: Mario Tesi <mario.tesi@st.com> Acked-by: Lorenzo Bianconi <lorenzo@kernel.org> Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
There are currently two situations that can trigger the PTP interrupt,
one is the PPS event, the other is the PEROUT event. However, the irq
handler fec_pps_interrupt() does not check the irq event type and
directly registers a PPS event into the system, but the event may be
a PEROUT event. This is incorrect because PEROUT is an output signal,
while PPS is the input of the kernel PPS system. Therefore, add a check
for the event type, if pps_enable is true, it means that the current
event is a PPS event, and then the PPS event is registered.
Fixes: 350749b909bf ("net: fec: Add support for periodic output signal of PPS") Signed-off-by: Wei Fang <wei.fang@nxp.com> Link: https://patch.msgid.link/20251125085210.1094306-5-wei.fang@nxp.com Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
In the current driver, PPS and PEROUT use the same channel to generate
the events, so they cannot be enabled at the same time. Otherwise, the
later configuration will overwrite the earlier configuration. Therefore,
when configuring PPS, the driver will check whether PEROUT is enabled.
Similarly, when configuring PEROUT, the driver will check whether PPS
is enabled.
Fixes: 350749b909bf ("net: fec: Add support for periodic output signal of PPS") Signed-off-by: Wei Fang <wei.fang@nxp.com> Link: https://patch.msgid.link/20251125085210.1094306-4-wei.fang@nxp.com Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
If the previously set PEROUT is already active, updating it will cause
the new PEROUT to start immediately instead of at the specified time.
This is because fep->reload_period is updated whithout check whether
the PEROUT is enabled, and the old PEROUT is not disabled. Therefore,
the pulse period will be updated immediately in the pulse interrupt
handler fec_pps_interrupt().
Currently, the driver does not support directly updating PEROUT and it
will make the logic be more complicated. To fix the current issue, add
a check before enabling the PEROUT, the driver will return an error if
PEROUT is enabled. If users wants to update a new PEROUT, they should
disable the old PEROUT first.
Fixes: 350749b909bf ("net: fec: Add support for periodic output signal of PPS") Signed-off-by: Wei Fang <wei.fang@nxp.com> Link: https://patch.msgid.link/20251125085210.1094306-3-wei.fang@nxp.com Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The PEROUT allows the user to set a specified future time to output the
periodic signal. If the future time is far from the current time, the FEC
driver will use hrtimer to configure PEROUT one second before the future
time. However, the hrtimer will not be canceled if the PEROUT is disabled
before the hrtimer expires. So the PEROUT will be configured when the
hrtimer expires, which is not as expected. Therefore, cancel the hrtimer
in fec_ptp_pps_disable() to fix this issue.
Fixes: 350749b909bf ("net: fec: Add support for periodic output signal of PPS") Signed-off-by: Wei Fang <wei.fang@nxp.com> Link: https://patch.msgid.link/20251125085210.1094306-2-wei.fang@nxp.com Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
On transmit, we are currently relying on skb->dev being set by
mctp_local_output() when we first set up the skb destination fields.
However, forwarded skbs do not use the local_output path, so will retain
their incoming netdev as their ->dev on tx. This does not work when
we're forwarding between interfaces.
Set skb->dev unconditionally in the transmit path, to allow for proper
forwarding.
We keep the skb->dev initialisation in mctp_local_output(), as we use it
for fragmentation.
Fixes: 269936db5eb3 ("net: mctp: separate routing database from routing operations") Suggested-by: Vince Chang <vince_chang@aspeedtech.com> Signed-off-by: Jeremy Kerr <jk@codeconstruct.com.au> Link: https://patch.msgid.link/20251125-dev-forward-v1-1-54ecffcd0616@codeconstruct.com.au Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The atlantic driver can receive packets with more than MAX_SKB_FRAGS (17)
fragments when handling large multi-descriptor packets. This causes an
out-of-bounds write in skb_add_rx_frag_netmem() leading to kernel panic.
The issue occurs because the driver doesn't check the total number of
fragments before calling skb_add_rx_frag(). When a packet requires more
than MAX_SKB_FRAGS fragments, the fragment index exceeds the array bounds.
Fix by assuming there will be an extra frag if buff->len > AQ_CFG_RX_HDR_SIZE,
then all fragments are accounted for. And reusing the existing check to
prevent the overflow earlier in the code path.
This crash occurred in production with an Aquantia AQC113 10G NIC.
Fix a potential counter roll-over issue in fbnic_mbx_alloc_rx_msgs()
when calculating descriptor slots. The issue occurs when head - tail
results in a large positive value (unsigned) and the compiler interprets
head - tail - 1 as a signed value.
Since FBNIC_IPC_MBX_DESC_LEN is a power of two, use a masking operation,
which is a common way of avoiding this problem when dealing with these
sort of ring space calculations.
When using the SGMII PCS as a fixed-link chip-to-chip connection, it is
easy to miss the fact that traffic passes only at 1G, since that's what
any normal such connection would use.
When using the SGMII PCS connected towards an on-board PHY or an SFP
module, it is immediately noticeable that when the link resolves to a
speed other than 1G, traffic from the MAC fails to pass: TX counters
increase, but nothing gets decoded by the other end, and no local RX
counters increase either.
Artificially lowering a fixed-link rate to speed = <100> makes us able
to see the same issue as in the case of having an SGMII PHY.
Some debugging shows that the XPCS configuration is A-OK, but that the
MAC Configuration Table entry for the port has the SPEED bits still set
to 1000Mbps, due to a special condition in the driver. Deleting that
condition, and letting the resolved link speed be programmed directly
into the MAC speed field, results in a functional link at all 3 speeds.
This piece of evidence, based on testing on both generations with SGMII
support (SJA1105S and SJA1110A) directly contradicts the statement from
the blamed commit that "the MAC is fixed at 1 Gbps and we need to
configure the PCS only (if even that)". Worse, that statement is not
backed by any documentation, and no one from NXP knows what it might
refer to.
I am unable to recall sufficient context regarding my testing from March
2020 to understand what led me to draw such a braindead and factually
incorrect conclusion. Yet, there is nothing of value regarding forcing
the MAC speed, either for SGMII or 2500Base-X (introduced at a later
stage), so remove all such logic.
Fixes: ffe10e679cec ("net: dsa: sja1105: Add support for the SGMII port") Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Link: https://patch.msgid.link/20251122111324.136761-1-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Prior to commit a25e7962db0d7 ("PCI/P2PDMA: Refactor the p2pdma mapping
helpers"), P2P segments were mapped using the pci_p2pdma_map_segment()
helper. This helper was responsible for populating sg->dma_address,
marking the bus address, and also setting sg_dma_len(sg).
The refactor[1] removed this helper and moved the mapping logic directly
into the callers. While iommu_dma_map_sg() was correctly updated to set
the length in the new flow, it was missed in dma_direct_map_sg().
Thus, in dma_direct_map_sg(), the PCI_P2PDMA_MAP_BUS_ADDR case sets the
dma_address and marks the segment, but immediately executes 'continue',
which causes the loop to skip the standard assignment logic at the end:
sg_dma_len(sg) = sg->length;
As a result, when CONFIG_NEED_SG_DMA_LENGTH is enabled, the dma_length
field remains uninitialized (zero) for P2P bus address mappings. This
breaks upper-layer drivers (for e.g. RDMA/IB) that rely on sg_dma_len()
to determine the transfer size.
Fix this by explicitly setting the DMA length in the
PCI_P2PDMA_MAP_BUS_ADDR case before continuing to the next scatterlist
entry.
Fixes: a25e7962db0d7 ("PCI/P2PDMA: Refactor the p2pdma mapping helpers") Reported-by: Jacob Moroni <jmoroni@google.com> Signed-off-by: Pranjal Shrivastava <praan@google.com>
[1]
https://lore.kernel.org/all/ac14a0e94355bf898de65d023ccf8a2ad22a3ece.1746424934.git.leon@kernel.org/
Since commit 30f241fcf52a ("xsk: Fix immature cq descriptor
production"), the descriptor number is stored in skb control block and
xsk_cq_submit_addr_locked() relies on it to put the umem addrs onto
pool's completion queue.
skb control block shouldn't be used for this purpose as after transmit
xsk doesn't have control over it and other subsystems could use it. This
leads to the following kernel panic due to a NULL pointer dereference.
Instead use the skb destructor_arg pointer along with pointer tagging.
As pointers are always aligned to 8B, use the bottom bit to indicate
whether this a single address or an allocated struct containing several
addresses.
Fixes: 30f241fcf52a ("xsk: Fix immature cq descriptor production") Closes: https://lore.kernel.org/netdev/0435b904-f44f-48f8-afb0-68868474bf1c@nop.hu/ Suggested-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Fernando Fernandez Mancera <fmancera@suse.de> Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com> Reviewed-by: Jason Xing <kerneljasonxing@gmail.com> Link: https://patch.msgid.link/20251124171409.3845-1-fmancera@suse.de Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
We are unnecessarily setting a bunch of skb fields per each processed
descriptor, which is redundant for fragmented frames.
Let us set these respective members for first fragment only. To address
both paths that we have within xsk_build_skb(), move assignments onto
xsk_set_destructor_arg() and rename it to xsk_skb_init_misc().
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com> Acked-by: Stanislav Fomichev <sdf@fomichev.me> Reviewed-by: Jason Xing <kerneljasonxing@gmail.com> Acked-by: Martin KaFai Lau <martin.lau@kernel.org> Link: https://patch.msgid.link/20250925160009.2474816-2-maciej.fijalkowski@intel.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Stable-dep-of: 0ebc27a4c67d ("xsk: avoid data corruption on cq descriptor number") Signed-off-by: Sasha Levin <sashal@kernel.org>
Currently, when skb is null, the driver prints an error and then
dereferences skb on the next line.
To fix this, let's add a 'break' after the error message to switch
to sxgbe_rx_refill(), which is similar to the approach taken by the
other drivers in this particular case, e.g. calxeda with xgmac_rx().
Found during a code review.
Fixes: 1edb9ca69e8a ("net: sxgbe: add basic framework for Samsung 10Gb ethernet driver") Signed-off-by: Alexey Kodanev <aleksei.kodanev@bell-sw.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20251121123834.97748-1-aleksei.kodanev@bell-sw.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Attempting to add a port device that is already up will expectedly fail,
but not before modifying the team device header_ops.
In the case of the syzbot reproducer the gre0 device is
already in state UP when it attempts to add it as a
port device of team0, this fails but before that
header_ops->create of team0 is changed from eth_header to ipgre_header
in the call to team_dev_type_check_change.
Later when we end up in ipgre_header() struct ip_tunnel* points to nonsense
as the private data of the device still holds a struct team.
Example sequence of iproute2 commands to reproduce the hang/BUG():
ip link add dev team0 type team
ip link add dev gre0 type gre
ip link set dev gre0 up
ip link set dev gre0 master team0
ip link set dev team0 up
ping -I team0 1.1.1.1
Move team_dev_type_check_change down where all other checks have passed
as it changes the dev type with no way to restore it in case
one of the checks that follow it fail.
Also make sure to preserve the origial mtu assignment:
- If port_dev is not the same type as dev, dev takes mtu from port_dev
- If port_dev is the same type as dev, port_dev takes mtu from dev
This is done by adding a conditional before the call to dev_set_mtu
to prevent it from assigning port_dev->mtu = dev->mtu and instead
letting team_dev_type_check_change assign dev->mtu = port_dev->mtu.
The conditional is needed because the patch moves the call to
team_dev_type_check_change past dev_set_mtu.
Testing:
- team device driver in-tree selftests
- Add/remove various devices as slaves of team device
- syzbot
Reported-by: syzbot+a2a3b519de727b0f7903@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=a2a3b519de727b0f7903 Fixes: 1d76efe1577b ("team: add support for non-ethernet devices") Signed-off-by: Nikola Z. Ivanov <zlatistiv@gmail.com> Reviewed-by: Jiri Pirko <jiri@nvidia.com> Link: https://patch.msgid.link/20251122002027.695151-1-zlatistiv@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The rate limiting validation condition currently checks the output
variable max_bw_value[i] instead of the input value
maxrate->tc_maxrate[i]. This causes the validation to compare an
uninitialized or stale value rather than the actual requested rate.
The condition should check the input rate to properly validate against
the upper limit:
} else if (maxrate->tc_maxrate[i] <= upper_limit_gbps) {
This aligns with the pattern used in the first branch, which correctly
checks maxrate->tc_maxrate[i] against upper_limit_mbps.
The current implementation can lead to unreliable validation behavior:
- For rates between 25.5 Gbps and 255 Gbps, if max_bw_value[i] is 0
from initialization, the GBPS path may be taken regardless of whether
the actual rate is within bounds
- When processing multiple TCs (i > 0), max_bw_value[i] contains the
value computed for the previous TC, affecting the validation logic
- The overflow check for rates exceeding 255 Gbps may not trigger
consistently depending on previous array values
This patch ensures the validation correctly examines the requested rate
value for proper bounds checking.
Fixes: 43b27d1bd88a ("net/mlx5e: Fix wraparound in rate limiting for values above 255 Gbps") Signed-off-by: Danielle Costantino <dcostantino@meta.com> Reviewed-by: Gal Pressman <gal@nvidia.com> Link: https://patch.msgid.link/20251124180043.2314428-1-dcostantino@meta.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
To initialize the taprio block in lan966x, it is required to configure
the register REVISIT_DLY. The purpose of this register is to set the
delay before revisit the next gate and the value of this register depends
on the system clock. The problem is that the we calculated wrong the value
of the system clock period in picoseconds. The actual system clock is
~165.617754MHZ and this correspond to a period of 6038 pico seconds and
not 15125 as currently set.
Fixes: e462b2717380b4 ("net: lan966x: Add offload support for taprio") Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20251121061411.810571-1-horatiu.vultur@microchip.com Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
gpy_update_interface() returns early in case the PHY is internal or
connected via USXGMII. In this case the gigabit master/slave property
as well as MDI/MDI-X status also won't be read which seems wrong.
Always read those properties by moving the logic to retrieve them to
gpy_read_status().
Fixes: fd8825cd8c6fc ("net: phy: mxl-gpy: Add PHY Auto/MDI/MDI-X set driver for GPY211 chips") Fixes: 311abcdddc00a ("net: phy: add support to get Master-Slave configuration") Suggested-by: "Russell King (Oracle)" <linux@armlinux.org.uk> Signed-off-by: Daniel Golle <daniel@makrotopia.org> Link: https://patch.msgid.link/71fccf3f56742116eb18cc070d2a9810479ea7f9.1763650701.git.daniel@makrotopia.org Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Commit 7a1bb49461b1 ("net: aquantia: fix potential IOMMU fault after
driver unbind") and commit ed4d81c4b3f2 ("net: aquantia: when cleaning
hw cache it should be toggled") fixed cache invalidation for ATL B0, but
ATL2 was left with only interrupt disabling. This allowed hardware to
write to cached descriptors after DMA memory was unmapped, triggering
SMMU faults. Once cache invalidation is applied to ATL2, the translation
fault can't be observed anymore.
Add shared aq_hw_invalidate_descriptor_cache() helper and use it in both
ATL B0 and ATL2 hw_stop() implementations for consistent behavior.
Fixes: e54dcf4bba3e ("net: atlantic: basic A2 init/deinit hw_ops") Tested-by: Carol Soto <csoto@nvidia.com> Signed-off-by: Kai-Heng Feng <kaihengf@nvidia.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20251120041537.62184-1-kaihengf@nvidia.com Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
This passes the address of the pointer "&punit_ipcdev" when the intent
was to pass the pointer itself "punit_ipcdev" (without the ampersand).
This means that the:
complete(&ipcdev->cmd_complete);
in intel_punit_ioc() will write to a wrong memory address corrupting it.
Fixes: fdca4f16f57d ("platform:x86: add Intel P-Unit mailbox IPC driver") Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org> Link: https://patch.msgid.link/aSCmoBipSQ_tlD-D@stanley.mountain Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
As the interface mode doesn't need to be updated on PHYs connected with
USXGMII and integrated PHYs, gpy_update_interface() should just return 0
in these cases rather than -EINVAL which has wrongly been introduced by
commit 7a495dde27ebc ("net: phy: mxl-gpy: Change gpy_update_interface()
function return type"), as this breaks support for those PHYs.
Fixes: 7a495dde27ebc ("net: phy: mxl-gpy: Change gpy_update_interface() function return type") Signed-off-by: Daniel Golle <daniel@makrotopia.org> Reviewed-by: Maxime Chevallier <maxime.chevallier@bootlin.com> Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Link: https://patch.msgid.link/f744f721a1fcc5e2e936428c62ff2c7d94d2a293.1763648168.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The sii902x driver was caching HDMI detection state in a sink_is_hdmi field
and checking it in mode_set() to determine whether to set HDMI or DVI
output mode. This approach had two problems:
1. With DRM_BRIDGE_ATTACH_NO_CONNECTOR (used by modern display drivers like
TIDSS), the bridge's get_modes() is never called. Instead, the
drm_bridge_connector helper calls the bridge's edid_read() and updates the
connector itself. This meant sink_is_hdmi was never populated, causing the
driver to default to DVI mode and breaking HDMI audio.
2. The mode_set() callback doesn't receive atomic state or connector
pointer, making it impossible to check connector->display_info.is_hdmi
directly at that point.
Fix this by moving the HDMI vs DVI decision from mode_set() to
atomic_enable(), where we can access the connector via
drm_atomic_get_new_connector_for_encoder(). This works for both connector
models:
- With DRM_BRIDGE_ATTACH_NO_CONNECTOR: Returns the drm_bridge_connector
created by the display driver, which has already been updated by the
helper's call to drm_edid_connector_update()
- Without DRM_BRIDGE_ATTACH_NO_CONNECTOR (legacy): Returns the connector
embedded in sii902x struct, which gets updated by the bridge's own
get_modes()
Fixes: 3de47e1309c2 ("drm/bridge: sii902x: use display info is_hdmi") Signed-off-by: Devarsh Thakkar <devarsht@ti.com> Reviewed-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com> Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org> Link: https://patch.msgid.link/20251030151635.3019864-1-devarsht@ti.com Signed-off-by: Sasha Levin <sashal@kernel.org>
As explain in commit fa349e396e48 ("veth: Fix race with AF_XDP exposing
old or uninitialized descriptors") for veth there is a chance after
napi_complete_done() that another CPU can manage start another NAPI
instance running veth_pool(). For NAPI this is correctly handled as the
napi_schedule_prep() check will prevent multiple instances from getting
scheduled, but for the remaining code in veth_pool() this can run
concurrent with the newly started NAPI instance.
The problem/race is that xdp_clear_return_frame_no_direct() isn't
designed to be nested.
Prior to commit 401cb7dae813 ("net: Reference bpf_redirect_info via
task_struct on PREEMPT_RT.") the temporary BPF net context
bpf_redirect_info was stored per CPU, where this wasn't an issue. Since
this commit the BPF context is stored in 'current' task_struct. When
running veth in threaded-NAPI mode, then the kthread becomes the storage
area. Now a race exists between two concurrent veth_pool() function calls
one exiting NAPI and one running new NAPI, both using the same BPF net
context.
Race is when another CPU gets within the xdp_set_return_frame_no_direct()
section before exiting veth_pool() calls the clear-function
xdp_clear_return_frame_no_direct().
Fixes: 401cb7dae8130 ("net: Reference bpf_redirect_info via task_struct on PREEMPT_RT.") Signed-off-by: Jesper Dangaard Brouer <hawk@kernel.org> Link: https://patch.msgid.link/176356963888.337072.4805242001928705046.stgit@firesoul Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The change eed467b517e8 ("Bluetooth: fix passkey uninitialized when used")
introduced a goto that bypasses the creation of temporary mackey and ltk
which are later used by the likes of DHKey Check step.
Later ffee202a78c2 ("Bluetooth: Always request for user confirmation for
Just Works (LE SC)") which means confirm_hint is always set in case
JUST_WORKS so the branch checking for an existing LTK becomes pointless
as confirm_hint will always be set, so this just merge both cases of
malicious or legitimate devices to be confirmed before continuing with the
pairing procedure.
Link: https://github.com/bluez/bluez/issues/1622 Fixes: eed467b517e8 ("Bluetooth: fix passkey uninitialized when used") Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The hdev lock/lookup/unlock/use pattern in the packet RX path doesn't
ensure hci_conn* is not concurrently modified/deleted. This locking
appears to be leftover from before conn_hash started using RCU
commit bf4c63252490b ("Bluetooth: convert conn hash to RCU")
and not clear if it had purpose since then.
Currently, there are code paths that delete hci_conn* from elsewhere
than the ordered hdev->workqueue where the RX work runs in. E.g.
commit 5af1f84ed13a ("Bluetooth: hci_sync: Fix UAF on hci_abort_conn_sync")
introduced some of these, and there probably were a few others before
it. It's better to do the locking so that even if these run
concurrently no UAF is possible.
Move the lookup of hci_conn and associated socket-specific conn to
protocol recv handlers, and do them within a single critical section
to cover hci_conn* usage and lookup.
syzkaller has reported a crash that appears to be this issue:
There is a potential race condition between sock bind and socket write
iter. bind may free the same cmd via mgmt_pending before write iter sends
the cmd, just as syzbot reported in UAF[1].
Here we use hci_dev_lock to synchronize the two, thereby avoiding the
UAF mentioned in [1].
[1]
syzbot reported:
BUG: KASAN: slab-use-after-free in mgmt_pending_remove+0x3b/0x210 net/bluetooth/mgmt_util.c:316
Read of size 8 at addr ffff888077164818 by task syz.0.17/5989
Call Trace:
mgmt_pending_remove+0x3b/0x210 net/bluetooth/mgmt_util.c:316
set_link_security+0x5c2/0x710 net/bluetooth/mgmt.c:1918
hci_mgmt_cmd+0x9c9/0xef0 net/bluetooth/hci_sock.c:1719
hci_sock_sendmsg+0x6ca/0xef0 net/bluetooth/hci_sock.c:1839
sock_sendmsg_nosec net/socket.c:727 [inline]
__sock_sendmsg+0x21c/0x270 net/socket.c:742
sock_write_iter+0x279/0x360 net/socket.c:1195
Fixes: 6fe26f694c82 ("Bluetooth: MGMT: Protect mgmt_pending list with its own lock") Reported-by: syzbot+9aa47cd4633a3cf92a80@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=9aa47cd4633a3cf92a80 Tested-by: syzbot+9aa47cd4633a3cf92a80@syzkaller.appspotmail.com Signed-off-by: Edward Adam Davis <eadavis@qq.com> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
HCI_OP_NOP means no command was actually sent so there is no point in
triggering cmd_timer which may cause a hdev->reset in the process since
it is assumed that the controller is stuck processing a command.
Fixes: e2d471b7806b ("Bluetooth: ISO: Fix not using SID from adv report") Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
When performing reset tests and encountering abnormal card drop issues
that lead to a kernel crash, it is necessary to perform a null check
before releasing resources to avoid attempting to release a null pointer.
Fixes: ceac1cb0259d ("Bluetooth: btusb: mediatek: add ISO data transmission functions") Signed-off-by: Chris Lu <chris.lu@mediatek.com> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The URB received in gs_usb_receive_bulk_callback() contains a struct
gs_host_frame. The length of the data after the header depends on the
gs_host_frame hf::flags and the active device features (e.g. time
stamping).
Introduce a new function gs_usb_get_minimum_length() and check that we have
at least received the required amount of data before accessing it. Only
copy the data to that skb that has actually been received.
The driver expects to receive a struct gs_host_frame in
gs_usb_receive_bulk_callback().
Use struct_group to describe the header of the struct gs_host_frame and
check that we have at least received the header before accessing any
members of it.
To resubmit the URB, do not dereference the pointer chain
"dev->parent->hf_size_rx" but use "parent->hf_size_rx" instead. Since
"urb->context" contains "parent", it is always defined, while "dev" is not
defined if the URB it too short.
The driver lacks the cleanup of failed transfers of URBs. This reduces the
number of available URBs per error by 1. This leads to reduced performance
and ultimately to a complete stop of the transmission.
If the sending of a bulk URB fails do proper cleanup:
- increase netdev stats
- mark the echo_sbk as free
- free the driver's context and do accounting
- wake the send queue
The `kvaser_usb_leaf_wait_cmd()` and `kvaser_usb_leaf_read_bulk_callback`
functions contain logic to zero-length commands. These commands are used
to align data to the USB endpoint's wMaxPacketSize boundary.
The driver attempts to skip these placeholders by aligning the buffer
position `pos` to the next packet boundary using `round_up()` function.
However, if zero-length command is found exactly on a packet boundary
(i.e., `pos` is a multiple of wMaxPacketSize, including 0), `round_up`
function will return the unchanged value of `pos`. This prevents `pos`
to be increased, causing an infinite loop in the parsing logic.
This patch fixes this in the function by using `pos + 1` instead.
This ensures that even if `pos` is on a boundary, the calculation is
based on `pos + 1`, forcing `round_up()` to always return the next
aligned boundary.
Fixes: 7259124eac7d ("can: kvaser_usb: Split driver into kvaser_usb_core.c and kvaser_usb_leaf.c") Signed-off-by: Seungjin Bae <eeodqql09@gmail.com> Reviewed-by: Jimmy Assarsson <extja@kvaser.com> Tested-by: Jimmy Assarsson <extja@kvaser.com> Link: https://patch.msgid.link/20251023162709.348240-1-eeodqql09@gmail.com Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de> Signed-off-by: Sasha Levin <sashal@kernel.org>
This software node change doesn't actually fix any current issues
with the kernel, it is an improvement to the lookup process rather
than fixing a live bug. It also causes a couple of regressions with
shipping laptops, which relied on the label based lookup.
There is a fix for the regressions in mainline, the first 5 patches
of [1]. However, those patches are fairly substantial changes and
given the patch causing the regression doesn't actually fix a bug
it seems better to just revert it in stable.
When scheduling the deferred balance callbacks, check SCX_RQ_BAL_CB_PENDING
instead of SCX_RQ_BAL_PENDING. This way schedule_deferred() properly tests
whether there is already a pending request for queue_balance_callback() to
be invoked at the end of .balance().
Fixes: a8ad873113d3 ("sched_ext: defer queue_balance_callback() until after ops.dispatch") Signed-off-by: Emil Tsalapatis <emil@etsalapatis.com> Signed-off-by: Tejun Heo <tj@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If we load a BPF scheduler while another scheduler is already running,
alloc_kick_pseqs() would be called again, overwriting the previously
allocated arrays.
Fix by moving the alloc_kick_pseqs() call after the scx_enable_state()
check, ensuring that the arrays are only allocated when a scheduler can
actually be loaded.
Fixes: 14c1da3895a11 ("sched_ext: Allocate scx_kick_cpus_pnt_seqs lazily using kvzalloc()") Signed-off-by: Andrea Righi <arighi@nvidia.com> Signed-off-by: Tejun Heo <tj@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Some ICL/TGL platforms with combo PHY ports suffer from signal integrity
issues at HBR3. While certain systems include a Parade PS8461 mux to
mitigate this, its presence cannot be reliably detected. Furthermore,
broken or missing VBT entries make it unsafe to rely on VBT for enforcing
link rate limits.
To address this introduce a device specific quirk to cap the eDP link rate
to HBR2 (540000 kHz). This will override any higher advertised rates from
the sink or DPCD for specific devices.
Currently, the quirk is added for Dell XPS 13 7390 2-in-1 which is reported
in gitlab issue #5969 [1].
v2: Align the quirk with the intended quirk name and refactor the
condition to use min(). (Jani)
v3: Use condition `rate > 540000`. Drop extra parentheses. (Ville)
Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/5969 Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com> Link: https://lore.kernel.org/r/20250710052041.1238567-3-ankit.k.nautiyal@intel.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This reverts commit 584cf613c24a4250d9be4819efc841aa2624d5b6.
Commit 584cf613c24a ("drm/i915/dp: Reject HBR3 when sink doesn't support
TPS4") introduced a blanket rejection of HBR3 link rate when the sink does
not support TPS4.
While this was intended to address instability observed on certain eDP
panels [1], there seem to be edp panels that do not follow the
specification. These eDP panels do not advertise TPS4 support, but require
HBR3 to operate at their fixed native resolution [2].
As a result, the change causes blank screens on such panels. Apparently,
Windows driver does not enforce this restriction, and the issue is not seen
there.
Therefore, revert the commit to restore functionality for such panels,
and align behaviour with Windows driver.
Jari Ruusu [Sat, 22 Nov 2025 07:28:00 +0000 (07:28 +0000)]
tty/vt: fix up incorrect backport to stable releases
Below is a patch for 6.12.58+ and 6.17.8+ stable branches only.
Upstream does not need this.
Signed-off-by: Jari Ruusu <jariruusu@protonmail.com> Fixes: da7e8b382396 ("tty/vt: Add missing return value for VT_RESIZE in vt_ioctl()") Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[why]
1. With allow_0_dtb_clk enabled, the time required to latch DTBCLK to 600 MHz
depends on the SMU. If DTBCLK is not latched to 600 MHz before set_mode completes,
gating DTBCLK causes the DP2 sink to lose its clock source.
2. The existing DTBCLK gating sequence ungates DTBCLK based on both pix_clk and ref_dtbclk,
but gates DTBCLK when either pix_clk or ref_dtbclk is zero.
pix_clk can be zero outside the set_mode sequence before DTBCLK is properly latched,
which can lead to DTBCLK being gated by mistake.
[how]
Consider both pixel_clk and ref_dtbclk when determining when it is safe to gate DTBCLK;
this is more accurate.
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4701 Fixes: 5949e7c4890c ("drm/amd/display: Enable Dynamic DTBCLK Switch") Reviewed-by: Charlene Liu <charlene.liu@amd.com> Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by: Fangzhi Zuo <Jerry.Zuo@amd.com> Signed-off-by: Roman Li <roman.li@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit d04eb0c402780ca037b62a6aecf23b863545ebca) Cc: stable@vger.kernel.org Signed-off-by: Sasha Levin <sashal@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fix inverted WARN_ON_ONCE condition that prevented normal address
removal counter updates. The current code only executes decrement
logic when the counter is already 0 (abnormal state), while
normal removals (counter > 0) are ignored.
CPU: 1 UID: 0 PID: 1300113 Comm: xfs_scrub Not tainted 6.18.0-rc4-djwx #rc4 PREEMPT(lazy) 3d744dd94e92690f00a04398d2bd8631dcef1954
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.0-4.module+el8.8.0+21164+ed375313 04/01/2014
==================================================================
On further analysis, I realized that the second parameter to min() is
not correct. xfs_ifork::if_bytes is the size of the xfs_ifork::if_data
buffer. if_bytes can be smaller than the data fork size because:
(a) the forkoff code tries to keep the data area as large as possible
(b) for symbolic links, if_bytes is the ondisk file size + 1
(c) forkoff is always a multiple of 8.
Case in point: for a single-byte symlink target, forkoff will be
8 but the buffer will only be 2 bytes long.
In other words, the logic here is wrong and we walk off the end of the
incore buffer. Fix that.
Cc: stable@vger.kernel.org # v6.10 Fixes: 2651923d8d8db0 ("xfs: online repair of symbolic links") Signed-off-by: Darrick J. Wong <djwong@kernel.org> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Carlos Maiolino <cem@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The changes modernizes the code by aligning it with current kernel best
practices. It improves code clarity and consistency, as strncpy is deprecated
as explained in Documentation/process/deprecated.rst. This change does
not alter the functionality or introduce any behavioral changes.
Suggested-by: Dave Chinner <david@fromorbit.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Carlos Maiolino <cmaiolino@redhat.com> Signed-off-by: Marcelo Moreira <marcelomoreira1905@gmail.com> Reviewed-by: Darrick J. Wong <djwong@kernel.org> Signed-off-by: Carlos Maiolino <cem@kernel.org>
Stable-dep-of: 678e1cc2f482 ("xfs: fix out of bounds memory read error in symlink repair") Signed-off-by: Sasha Levin <sashal@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Disable Panel Replay on MST links until it's properly implemented. For
instance the required VSC SDP is not programmed on MST and FEC is not
enabled if Panel Replay is enabled.
Currently we are ignoriong drm_dp_dpcd_read return values when reading PSR
and Panel Replay capability DPCD register. Rework intel_psr_dpcd a bit to
take care of checking the return value.
smb: client: fix incomplete backport in cfids_invalidation_worker()
The previous commit bdb596ceb4b7 ("smb: client: fix potential UAF in
smb2_close_cached_fid()") was an incomplete backport and missed one
kref_put() call in cfids_invalidation_worker() that should have been
converted to close_cached_dir().
On PF passthrough environment, after hibernate and then resume, coralgemm
will cause gpu page fault.
Mode1 reset happens during hibernate, but partition mode is not restored
on resume, register mmCP_HYP_XCP_CTL and mmCP_PSP_XCP_CTL is not right
after resume. When CP access the MQD BO, wrong stride size is used,
this will cause out of bound access on the MQD BO, resulting page fault.
The fix is to ensure gfx_v9_4_3_switch_compute_partition() is called
when resume from a hibernation.
KFD resume is called separately during a reset recovery or resume from
suspend sequence. Hence it's not required to be called as part of
partition switch.
Signed-off-by: Samuel Zhang <guoqing.zhang@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 5d1b32cfe4a676fe552416cb5ae847b215463a1a) Signed-off-by: Sasha Levin <sashal@kernel.org>
If we are logging a new name make sure our inode has the runtime flag
BTRFS_INODE_COPY_EVERYTHING set so that at btrfs_log_inode() we will find
new inode refs/extrefs in the subvolume tree and copy them into the log
tree.
We are currently doing it when adding a new link but we are missing it
when renaming.
An example where this makes a new name not persisted:
1) create symlink with name foo in directory A
2) fsync directory A, which persists the symlink
3) rename the symlink from foo to bar
4) fsync directory A to persist the new symlink name
Step 4 isn't working correctly as it's not logging the new name and also
leaving the old inode ref in the log tree, so after a power failure the
symlink still has the old name of "foo". This is because when we first
fsync directoy A we log the symlink's inode (as it's a new entry) and at
btrfs_log_inode() we set the log mode to LOG_INODE_ALL and then because
we are using that mode and the inode has the runtime flag
BTRFS_INODE_NEEDS_FULL_SYNC set, we clear that flag as well as the flag
BTRFS_INODE_COPY_EVERYTHING. That means the next time we log the inode,
during the rename through the call to btrfs_log_new_name() (calling
btrfs_log_inode_parent() and then btrfs_log_inode()), we will not search
the subvolume tree for new refs/extrefs and jump directory to the
'log_extents' label.
Fix this by making sure we set BTRFS_INODE_COPY_EVERYTHING on an inode
when we are about to log a new name. A test case for fstests will follow
soon.
Reported-by: Vyacheslav Kovalevsky <slava.kovalevskiy.2014@gmail.com> Link: https://lore.kernel.org/linux-btrfs/ac949c74-90c2-4b9a-b7fd-1ffc5c3175c7@gmail.com/ Reviewed-by: Boris Burkov <boris@bur.io> Reviewed-by: Qu Wenruo <wqu@suse.com> Signed-off-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The help message incorrectly listed '-t' as the short option for
--threads, but the actual getopt_long configuration uses '-e'.
This mismatch can confuse users and lead to incorrect command-line
usage. This patch updates the usage string to correctly show:
"-e, --threads NRTHR"
to match the implementation.
Note: checkpatch.pl reports a false-positive spelling warning on
'Run', which is intentional.
Make knav_dma_open_channel consistently return NULL on error instead
of ERR_PTR. Currently the header include/linux/soc/ti/knav_dma.h
returns NULL when the driver is disabled, but the driver
implementation does not even return NULL or ERR_PTR on failure,
causing inconsistency in the users. This results in a crash in
netcp_free_navigator_resources as followed (trimmed):
Unhandled fault: alignment exception (0x221) at 0xfffffff2
[fffffff2] *pgd=80000800207003, *pmd=82ffda003, *pte=00000000
Internal error: : 221 [#1] SMP ARM
Modules linked in:
CPU: 0 UID: 0 PID: 1 Comm: swapper/0 Not tainted 6.17.0-rc7 #1 NONE
Hardware name: Keystone
PC is at knav_dma_close_channel+0x30/0x19c
LR is at netcp_free_navigator_resources+0x2c/0x28c
[... TRIM...]
Call trace:
knav_dma_close_channel from netcp_free_navigator_resources+0x2c/0x28c
netcp_free_navigator_resources from netcp_ndo_open+0x430/0x46c
netcp_ndo_open from __dev_open+0x114/0x29c
__dev_open from __dev_change_flags+0x190/0x208
__dev_change_flags from netif_change_flags+0x1c/0x58
netif_change_flags from dev_change_flags+0x38/0xa0
dev_change_flags from ip_auto_config+0x2c4/0x11f0
ip_auto_config from do_one_initcall+0x58/0x200
do_one_initcall from kernel_init_freeable+0x1cc/0x238
kernel_init_freeable from kernel_init+0x1c/0x12c
kernel_init from ret_from_fork+0x14/0x38
[... TRIM...]
Standardize the error handling by making the function return NULL on
all error conditions. The API is used in just the netcp_core.c so the
impact is limited.
Note, this change, in effect reverts commit 5b6cb43b4d62 ("net:
ethernet: ti: netcp_core: return error while dma channel open issue"),
but provides a less error prone implementation.
Suggested-by: Simon Horman <horms@kernel.org> Suggested-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Nishanth Menon <nm@ti.com> Reviewed-by: Jacob Keller <jacob.e.keller@intel.com> Link: https://patch.msgid.link/20251103162811.3730055-1-nm@ti.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
According to UFS specifications, the power-off sequence for a UFS device
includes:
- Sending an SSU command with Power_Condition=3 and await a response.
- Asserting RST_N low.
- Turning off REF_CLK.
- Turning off VCC.
- Turning off VCCQ/VCCQ2.
As part of ufs shutdown, after the SSU command completion, asserting
hardware reset (HWRST) triggers the device firmware to wake up and
execute its reset routine. This routine initializes hardware blocks and
takes a few milliseconds to complete. During this time, the ICCQ draws a
large current.
This large ICCQ current may cause issues for the regulator which is
supplying power to UFS, because the turn off request from UFS driver to
the regulator framework will be immediately followed by low power
mode(LPM) request by regulator framework. This is done by framework
because UFS which is the only client is requesting for disable. So if
the rail is still in the process of shutting down while ICCQ exceeds LPM
current thresholds, and LPM mode is activated in hardware during this
state, it may trigger an overcurrent protection (OCP) fault in the
regulator.
To prevent this, a 10ms delay is added after asserting HWRST. This
allows the reset operation to complete while power rails remain active
and in high-power mode.
Currently there is no way for Host to query whether the reset is
completed or not and hence this the delay is based on experiments with
Qualcomm UFS controllers across multiple UFS vendors.
Signed-off-by: Nitin Rawat <nitin.rawat@oss.qualcomm.com> Reviewed-by: Manivannan Sadhasivam <mani@kernel.org> Link: https://patch.msgid.link/20251012173828.9880-1-nitin.rawat@oss.qualcomm.com Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Since 8b3a087f7f65 ("ALSA: usb-audio: Unify virtual type units type to
UAC3 values") usb-audio is using UAC3_CLOCK_SOURCE instead of
bDescriptorSubtype, later refactored with e0ccdef9265 ("ALSA: usb-audio:
Clean up check_input_term()") into parse_term_uac2_clock_source().
This breaks the clock source selection for at least my
1397:0003 BEHRINGER International GmbH FCA610 Pro.
Fix by using UAC2_CLOCK_SOURCE in parse_term_uac2_clock_source().
If user provides a large value (such as 0x80) for parameter
prefetch_mem_region_instance in vm_bind ioctl, it will cause
BIT(prefetch_region) overflow as below:
"
------------[ cut here ]------------
UBSAN: shift-out-of-bounds in drivers/gpu/drm/xe/xe_vm.c:3414:7
shift exponent 128 is too large for 64-bit type 'long unsigned int'
CPU: 8 UID: 0 PID: 53120 Comm: xe_exec_system_ Tainted: G W 6.18.0-rc1-lgci-xe-kernel+ #200 PREEMPT(voluntary)
Tainted: [W]=WARN
Hardware name: ASUS System Product Name/PRIME Z790-P WIFI, BIOS 0812 02/24/2023
Call Trace:
<TASK>
dump_stack_lvl+0xa0/0xc0
dump_stack+0x10/0x20
ubsan_epilogue+0x9/0x40
__ubsan_handle_shift_out_of_bounds+0x10e/0x170
? mutex_unlock+0x12/0x20
xe_vm_bind_ioctl.cold+0x20/0x3c [xe]
...
"
Fix it by validating prefetch_region before the BIT() usage.
v2: Add Closes and Cc stable kernels. (Matt)
Reported-by: Koen Koning <koen.koning@intel.com> Reported-by: Peter Senna Tschudin <peter.senna@linux.intel.com> Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs") Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/6478 Cc: <stable@vger.kernel.org> # v6.8+ Reviewed-by: Matthew Auld <matthew.auld@intel.com> Signed-off-by: Shuicheng Lin <shuicheng.lin@intel.com> Signed-off-by: Matthew Auld <matthew.auld@intel.com> Link: https://patch.msgid.link/20251112181005.2120521-2-shuicheng.lin@intel.com
(cherry picked from commit 8f565bdd14eec5611cc041dba4650e42ccdf71d9) Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
(cherry picked from commit d52dea485cd3c98cfeeb474cf66cf95df2ab142f) Signed-off-by: Shuicheng Lin <shuicheng.lin@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Fix bug where make nconfig doesn't initialize the default locale, which
causes ncurses menu borders to be displayed incorrectly (lqqqqk) in
UTF-8 terminals that don't support VT100 ACS by default, such as PuTTY.
Fix bug where make menuconfig doesn't initialize the default locale, which
causes ncurses menu borders to be displayed incorrectly (lqqqqk) in
UTF-8 terminals that don't support VT100 ACS by default, such as PuTTY.
When a netdev issues a RX async resync request for a TLS connection,
the TLS module handles it by logging record headers and attempting to
match them to the tcp_sn provided by the device. If a match is found,
the TLS module approves the tcp_sn for resynchronization.
While waiting for a device response, the TLS module also increments
rcd_delta each time a new TLS record is received, tracking the distance
from the original resync request.
However, if the device response is delayed or fails (e.g due to
unstable connection and device getting out of tracking, hardware
errors, resource exhaustion etc.), the TLS module keeps logging and
incrementing, which can lead to a WARN() when rcd_delta exceeds the
threshold.
To address this, introduce tls_offload_rx_resync_async_request_cancel()
to explicitly cancel resync requests when a device response failure is
detected. Call this helper also as a final safeguard when rcd_delta
crosses its threshold, as reaching this point implies that earlier
cancellation did not occur.
Update tls_offload_rx_resync_async_request_start() and
tls_offload_rx_resync_async_request_end() to get a struct
tls_offload_resync_async parameter directly, rather than
extracting it from struct sock.
This change aligns the function signatures with the upcoming
tls_offload_rx_resync_async_request_cancel() helper, which
will be introduced in a subsequent patch.
In bareudp.sh, this script uses /bin/sh and it will load another lib.sh
BASH script at the very beginning.
But on some operating systems like Ubuntu, /bin/sh is actually pointed to
DASH, thus it will try to run BASH commands with DASH and consequently
leads to syntax issues:
# ./bareudp.sh: 4: ./lib.sh: Bad substitution
# ./bareudp.sh: 5: ./lib.sh: source: not found
# ./bareudp.sh: 24: ./lib.sh: Syntax error: "(" unexpected
Fix this by explicitly using BASH for bareudp.sh. This fixes test
execution failures on systems where /bin/sh is not BASH.
In very rare cases, DFS mounts could end up with SMB sessions without
any IPC connections. These mounts are only possible when having
unexpired cached DFS referrals, hence not requiring any IPC
connections during the mount process.
Try to establish those missing IPC connections when refreshing DFS
referrals. If the server is still rejecting it, then simply ignore
and leave expired cached DFS referral for any potential DFS failovers.
Reported-by: Jay Shin <jaeshin@redhat.com> Signed-off-by: Paulo Alcantara (Red Hat) <pc@manguebit.org> Cc: David Howells <dhowells@redhat.com> Cc: linux-cifs@vger.kernel.org Signed-off-by: Steve French <stfrench@microsoft.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Fix warning caused from declaration under a case label. The proper way
is to declare variable at the beginning of the function. The warning
came from running clang using LLVM=1; and is as follows:
-test_cachestat.c:260:3: warning: label followed by a declaration is a C23 extension [-Wc23-extensions]
260 | char *map = mmap(NULL, filesize, PROT_READ | PROT_WRITE,
|
Link: https://lore.kernel.org/r/20250929115405.25695-2-sidharthseela@gmail.com Signed-off-by: Sidharth Seela <sidharthseela@gmail.com> Reviewed-by: SeongJae Park <sj@kernel.org> Reviewed-by: wang lian <lianux.mm@gmail.com> Reviewed-by: Dev Jain <dev.jain@arm.com> Acked-by: Shuah Khan <skhan@linuxfoundation.org> Acked-by: Nhat Pham <nphamcs@gmail.com> Signed-off-by: Shuah Khan <skhan@linuxfoundation.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
WildcatLake (WCL) is a variant of PantherLake (PTL) and shares the same
uncore PMU features with PTL. Therefore, directly reuse Pantherlake's
uncore PMU enabling code for WildcatLake.
The quirk for Victus 15-fa1xxx wasn't working on Victus 15-fa1031nt due to a different board id. This patch enables the existing quirk for the board id 8BC8.
Tested on HP Victus 15-fa1031nt (MB 8C2D). The LED behaviour works as intended.
Fix this regression by making scsi_host_busy() check whether the SCSI
host tag set has already been initialized. tag_set->ops is set by
scsi_mq_setup_tags() just before blk_mq_alloc_tag_set() is called. This
fix is based on the assumption that scsi_host_busy() and
scsi_mq_setup_tags() calls are serialized. This is the case in the UFS
driver.
Reported-by: Sebastian Reichel <sebastian.reichel@collabora.com> Closes: https://lore.kernel.org/linux-block/pnezafputodmqlpumwfbn644ohjybouveehcjhz2hmhtcf2rka@sdhoiivync4y/ Cc: Ming Lei <ming.lei@redhat.com> Cc: Jens Axboe <axboe@kernel.dk> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Reviewed-by: Ming Lei <ming.lei@redhat.com> Tested-by: Sebastian Reichel <sebastian.reichel@collabora.com> Link: https://patch.msgid.link/20251007214800.1678255-1-bvanassche@acm.org Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Since msm_vma_job_free() is called directly from the ioctl, this looks
like an error path cleanup issue. Which I think results from
prealloc_cleanup() called without a preceding successful
prealloc_allocate() call. So handle that case better.
Reported-by: Connor Abbott <cwabbott0@gmail.com> Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/678677/
Message-ID: <20251006153542.419998-1-robin.clark@oss.qualcomm.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The sched_ext code calls queue_balance_callback() during enqueue_task()
to defer operations that drop multiple locks until we can unpin them.
The call assumes that the rq lock is held until the callbacks are
invoked, and the pending callbacks will not be visible to any other
threads. This is enforced by a WARN_ON_ONCE() in rq_pin_lock().
However, balance_one() may actually drop the lock during a BPF dispatch
call. Another thread may win the race to get the rq lock and see the
pending callback. To avoid this, sched_ext must only queue the callback
after the dispatch calls have completed.
On systems with >4096 CPUs, scx_kick_cpus_pnt_seqs allocation fails during
boot because it exceeds the 32,768 byte percpu allocator limit.
Restructure to use DEFINE_PER_CPU() for the per-CPU pointers, with each CPU
pointing to its own kvzalloc'd array. Move allocation from boot time to
scx_enable() and free in scx_disable(), so the O(nr_cpu_ids^2) memory is only
consumed when sched_ext is active.
Use RCU to guard against racing with free. Arrays are freed via call_rcu()
and kick_cpus_irq_workfn() uses rcu_dereference_bh() with a NULL check.
While at it, rename to scx_kick_pseqs for brevity and update comments to
clarify these are pick_task sequence numbers.
v2: RCU protect scx_kick_seqs to manage kick_cpus_irq_workfn() racing
against disable as per Andrea.
v3: Fix bugs notcied by Andrea.
Reported-by: Phil Auld <pauld@redhat.com> Link: http://lkml.kernel.org/r/20251007133523.GA93086@pauld.westford.csb Cc: Andrea Righi <arighi@nvidia.com> Reviewed-by: Emil Tsalapatis <emil@etsalapatis.com> Reviewed-by: Phil Auld <pauld@redhat.com> Reviewed-by: Andrea Righi <arighi@nvidia.com> Signed-off-by: Tejun Heo <tj@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Commit 69896119dc9d ("MIPS: vdso: Switch to generic storage
implementation") switches to a generic vdso storage, which increases
the number of data pages from 1 to 4. But there is only one page
reserved, which causes segementation faults depending where the VDSO
area is randomized to. To fix this use the same size of reservation
and allocation of the VDSO data pages.
Fixes: 69896119dc9d ("MIPS: vdso: Switch to generic storage implementation") Reviewed-by: Thomas Weißschuh <thomas.weissschuh@linutronix.de> Reviewed-by: Huacai Chen <chenhuacai@loongson.cn> Reviewed-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Signed-off-by: Sasha Levin <sashal@kernel.org>
tk_aux_sysfs_init() returns immediately on error during the auxiliary clock
initialization loop without cleaning up previously allocated kobjects and
sysfs groups.
If kobject_create_and_add() or sysfs_create_group() fails during loop
iteration, the parent kobjects (tko and auxo) and any previously created
child kobjects are leaked.
Fix this by adding proper error handling with goto labels to ensure all
allocated resources are cleaned up on failure. kobject_put() on the
parent kobjects will handle cleanup of their children.
Fixes: 7b95663a3d96 ("timekeeping: Provide interface to control auxiliary clocks") Signed-off-by: Malaya Kumar Rout <mrout@redhat.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Link: https://patch.msgid.link/20251120150213.246777-1-mrout@redhat.com Signed-off-by: Sasha Levin <sashal@kernel.org>
During connect(), acting on a signal/timeout by disconnecting an already
established socket leads to several issues:
1. connect() invoking vsock_transport_cancel_pkt() ->
virtio_transport_purge_skbs() may race with sendmsg() invoking
virtio_transport_get_credit(). This results in a permanently elevated
`vvs->bytes_unsent`. Which, in turn, confuses the SOCK_LINGER handling.
2. connect() resetting a connected socket's state may race with socket
being placed in a sockmap. A disconnected socket remaining in a sockmap
breaks sockmap's assumptions. And gives rise to WARNs.
3. connect() transitioning SS_CONNECTED -> SS_UNCONNECTED allows for a
transport change/drop after TCP_ESTABLISHED. Which poses a problem for
any simultaneous sendmsg() or connect() and may result in a
use-after-free/null-ptr-deref.
Do not disconnect socket on signal/timeout. Keep the logic for unconnected
sockets: they don't linger, can't be placed in a sockmap, are rejected by
sendmsg().
Currently cpu-clock event always returns 0 count, e.g.,
perf stat -e cpu-clock -- sleep 1
Performance counter stats for 'sleep 1':
0 cpu-clock # 0.000 CPUs utilized
1.002308394 seconds time elapsed
The root cause is the commit 'bc4394e5e79c ("perf: Fix the throttle
error of some clock events")' adds PERF_EF_UPDATE flag check before
calling cpu_clock_event_update() to update the count, however the
PERF_EF_UPDATE flag is never set when the cpu-clock event is stopped in
counting mode (pmu->dev() -> cpu_clock_event_del() ->
cpu_clock_event_stop()). This leads to the cpu-clock event count is
never updated.
To fix this issue, force to set PERF_EF_UPDATE flag for cpu-clock event
just like what task-clock does.
Fixes: bc4394e5e79c ("perf: Fix the throttle error of some clock events") Signed-off-by: Dapeng Mi <dapeng1.mi@linux.intel.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Ian Rogers <irogers@google.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Link: https://patch.msgid.link/20251112080526.3971392-1-dapeng1.mi@linux.intel.com Signed-off-by: Sasha Levin <sashal@kernel.org>
Add proper cleanup of ctx->source and fc->source to the
cifs_parse_mount_err error handler. This ensures that memory allocated
for the source strings is correctly freed on all error paths, matching
the cleanup already performed in the success path by
smb3_cleanup_fs_context_contents().
Pointers are also set to NULL after freeing to prevent potential
double-free issues.
This change fixes a memory leak originally detected by syzbot. The
leak occurred when processing Opt_source mount options if an error
happened after ctx->source and fc->source were successfully
allocated but before the function completed.
The specific leak sequence was:
1. ctx->source = smb3_fs_context_fullpath(ctx, '/') allocates memory
2. fc->source = kstrdup(ctx->source, GFP_KERNEL) allocates more memory
3. A subsequent error jumps to cifs_parse_mount_err
4. The old error handler freed passwords but not the source strings,
causing the memory to leak.
This issue was not addressed by commit e8c73eb7db0a ("cifs: client:
fix memory leak in smb3_fs_context_parse_param"), which only fixed
leaks from repeated fsconfig() calls but not this error path.
Patch updated with minor change suggested by kernel test robot
Reported-by: syzbot+87be6809ed9bf6d718e3@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=87be6809ed9bf6d718e3 Fixes: 24e0a1eff9e2 ("cifs: switch to new mount api") Reviewed-by: David Howells <dhowells@redhat.com> Signed-off-by: Shaurya Rane <ssrane_b23@ee.vjti.ac.in> Signed-off-by: Steve French <stfrench@microsoft.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The kernel UAPI headers already contain fixed-width integer types, there
is no need to rely on the libc types. There may not be a libc available
or the libc may not provides the <stdint.h>, like for example on nolibc.
This also aligns the header with the rest of the LoongArch UAPI headers.
Fixes: 803b0fc5c3f2 ("LoongArch: Add process management") Signed-off-by: Thomas Weißschuh <linux@weissschuh.net> Signed-off-by: Huacai Chen <chenhuacai@loongson.cn> Signed-off-by: Sasha Levin <sashal@kernel.org>
In commit 0345691b24c0 ("tick/rcu: Stop allowing RCU_SOFTIRQ in idle") the
new function report_idle_softirq() was created by breaking code out of the
existing can_stop_idle_tick() for kernels v5.18 and newer.
In doing so, the code essentially went from this form:
if (A) {
static int ratelimit;
if (ratelimit < 10 && !C && A&D) {
pr_warn("NOHZ tick-stop error: ...");
ratelimit++;
}
return false;
}
to a new function:
static bool report_idle_softirq(void)
{
static int ratelimit;
if (likely(!A))
return false;
if (ratelimit < 10)
return false;
...
pr_warn("NOHZ tick-stop error: local softirq work is pending, handler #%02x!!!\n",
pending);
ratelimit++;
return true;
}
commit a7e282c77785 ("tick/rcu: Fix bogus ratelimit condition") realized
ratelimit was essentially set to zero instead of ten, and hence *no*
softirq pending messages would ever be issued, but "fixed" it as:
- if (ratelimit < 10)
+ if (ratelimit >= 10)
return false;
However, this fix introduced another issue:
When ratelimit is greater than or equal 10, even if A is true, it will
directly return false. While ratelimit in the original code was only used
to control printing and will not affect the return value.
Restore the original logic and restrict ratelimit to control the printk and
not the return value.
Fixes: 0345691b24c0 ("tick/rcu: Stop allowing RCU_SOFTIRQ in idle") Fixes: a7e282c77785 ("tick/rcu: Fix bogus ratelimit condition") Signed-off-by: Wen Yang <wen.yang@linux.dev> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Link: https://patch.msgid.link/20251119174525.29470-1-wen.yang@linux.dev Signed-off-by: Sasha Levin <sashal@kernel.org>
Pause, Asym_Pause and Autoneg bits are not set when pl->supported is
initialized, so these link modes will not work for the fixed-link. This
leads to a TCP performance degradation issue observed on the i.MX943
platform.
The switch CPU port of i.MX943 is connected to an ENETC MAC, this link
is a fixed link and the link speed is 2.5Gbps. And one of the switch
user ports is the RGMII interface, and its link speed is 1Gbps. If the
flow-control of the fixed link is not enabled, we can easily observe
the iperf performance of TCP packets is very low. Because the inbound
rate on the CPU port is greater than the outbound rate on the user port,
the switch is prone to congestion, leading to the loss of some TCP
packets and requiring multiple retransmissions.
Solving this problem should be as simple as setting the Asym_Pause and
Pause bits. The reason why the Autoneg bit needs to be set, Russell
has gave a very good explanation in the thread [1], see below.
"As the advertising and lp_advertising bitmasks have to be non-empty,
and the swphy reports aneg capable, aneg complete, and AN enabled, then
for consistency with that state, Autoneg should be set. This is how it
was prior to the blamed commit."
With the final call to fput() on a file descriptor, the release action
may be deferred and scheduled on a work queue. The reference count of
that descriptor is still zero and it must not be used. It's possible
that a GPIO change, we want to notify the user-space about, happens
AFTER the reference count on the file descriptor associated with the
character device went down to zero but BEFORE the .release() callback
was called from the workqueue and so BEFORE we unregistered from the
notifier.
Using the regular get_file() routine in this situation triggers the
following warning:
struct file::f_count incremented from zero; use-after-free condition present!
So use the get_file_active() variant that will return NULL on file
descriptors that have been or are being released.
Fixes: 40b7c49950bd ("gpio: cdev: put emitting the line state events on a workqueue") Reported-by: Alexander Sverdlin <alexander.sverdlin@siemens.com> Closes: https://lore.kernel.org/all/5d605f7fc99456804911403102a4fe999a14cc85.camel@siemens.com/ Tested-by: Alexander Sverdlin <alexander.sverdlin@siemens.com> Link: https://lore.kernel.org/r/20251117-gpio-cdev-get-file-v1-1-28a16b5985b8@linaro.org Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Miao Wang reported a bug of SO_PEEK_OFF on AF_UNIX SOCK_STREAM
socket.
The unexpected behaviour is triggered when the peek offset is
larger than the recv queue and the thread is unblocked by new
data.
Let's assume a socket which has "aaaa" in the recv queue and
the peek offset is 4.
First, unix_stream_read_generic() reads the offset 4 and skips
the skb(s) of "aaaa" with the code below:
skip = max(sk_peek_offset(sk, flags), 0); /* @skip is 4. */
do {
...
while (skip >= unix_skb_len(skb)) {
skip -= unix_skb_len(skb);
...
skb = skb_peek_next(skb, &sk->sk_receive_queue);
if (!skb)
goto again; /* @skip is 0. */
}
The thread jumps to the 'again' label and goes to sleep since
new data has not arrived yet.
Later, new data "bbbb" unblocks the thread, and the thread jumps
to the 'redo:' label to restart the entire process from the first
skb in the recv queue.
do {
...
redo:
...
last = skb = skb_peek(&sk->sk_receive_queue);
...
again:
if (skb == NULL) {
...
timeo = unix_stream_data_wait(sk, timeo, last,
last_len, freezable);
...
goto redo; /* @skip is 0 !! */
However, the peek offset is not reset in the path.
If the buffer size is 8, recv() will return "aaaabbbb" without
skipping any data, and the final offset will be 12 (the original
offset 4 + peeked skbs' length 8).
After sleeping in unix_stream_read_generic(), we have to fetch the
peek offset again.
Let's move the redo label before mutex_lock(&u->iolock).
Fixes: 9f389e35674f ("af_unix: return data from multiple SKBs on recv() with MSG_PEEK flag") Reported-by: Miao Wang <shankerwangmiao@gmail.com> Closes: https://lore.kernel.org/netdev/3B969F90-F51F-4B9D-AB1A-994D9A54D460@gmail.com/ Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com> Link: https://patch.msgid.link/20251117174740.3684604-2-kuniyu@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The mlx5_irq_alloc() function can inadvertently free the entire rmap
and end up in a crash[1] when the other threads tries to access this,
when request_irq() fails due to exhausted IRQ vectors. This commit
modifies the cleanup to remove only the specific IRQ mapping that was
just added.
This prevents removal of other valid mappings and ensures precise
cleanup of the failed IRQ allocation's associated glue object.
Note: This error is observed when both fwctl and rds configs are enabled.
[1]
mlx5_core 0000:05:00.0: Successfully registered panic handler for port 1
mlx5_core 0000:05:00.0: mlx5_irq_alloc:293:(pid 66740): Failed to
request irq. err = -28
infiniband mlx5_0: mlx5_ib_test_wc:290:(pid 66740): Error -28 while
trying to test write-combining support
mlx5_core 0000:05:00.0: Successfully unregistered panic handler for port 1
mlx5_core 0000:06:00.0: Successfully registered panic handler for port 1
mlx5_core 0000:06:00.0: mlx5_irq_alloc:293:(pid 66740): Failed to
request irq. err = -28
infiniband mlx5_0: mlx5_ib_test_wc:290:(pid 66740): Error -28 while
trying to test write-combining support
mlx5_core 0000:06:00.0: Successfully unregistered panic handler for port 1
mlx5_core 0000:03:00.0: mlx5_irq_alloc:293:(pid 28895): Failed to
request irq. err = -28
mlx5_core 0000:05:00.0: mlx5_irq_alloc:293:(pid 28895): Failed to
request irq. err = -28
general protection fault, probably for non-canonical address
0xe277a58fde16f291: 0000 [#1] SMP NOPTI
projects / thirdparty / kernel / stable.git / log
