| Age | Commit message (Collapse) | Author |
|
UBLK_U_IO_FETCH_IO_CMDS has the same order problem as the per-io
commands: ublk_batch_attach() links the fetch command into fcmd_head
and marks it cancelable only after dropping evts_lock. A cancel from
the control path (ublk_batch_cancel_queue()) can take it in between and
complete it, and the later mark puts a completed request on io_uring's
cancelable list.
Mark it before linking it. evts_lock orders the mark before the link,
which is where the cancel finds it.
Batch commands are not bounced to task work, so they can run from io-wq
without uring_lock, and io_uring's cancel walk can now find a fetch
command before it is linked. Two things make that safe:
- initialize fcmd->node: it came from kzalloc(), so list_empty() saw a
linked node and ublk_batch_cancel_cmd() would list_del_init() NULL
pointers
- on the -ENODEV path, complete the command with io_uring_cmd_done()
before freeing fcmd: that removes it from the cancelable list under
uring_lock, after which nothing can see fcmd
Also use data->cmd instead of fcmd->cmd after dropping evts_lock: once
fcmd is linked and not the active one, a control-path cancel may
complete and free it.
Fixes: a4d883755399 ("ublk: add UBLK_U_IO_FETCH_IO_CMDS for batch I/O processing")
Cc: stable@vger.kernel.org # v7.0+
Tested-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Link: https://patch.msgid.link/20261011063555.1142277-4-tom.leiming@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
FETCH, COMMIT_AND_FETCH and NEED_GET_DATA publish the command in io->cmd
before marking it cancelable. A cancel from the control path (STOP_DEV,
QUIESCE_DEV) can complete it in between:
issue path control-path cancel
io->cmd = C
take C, io_uring_cmd_done(C):
C not marked, nothing to unlink
ublk_prep_cancel(C)
the completed C is linked on the cancelable list
ring exit: the cancel walk hits it → GPF (KASAN, with a delay added)
Mark the command before ublk_fill_io_cmd() publishes it. The handlers
run with uring_lock held, as io_uring's cancel walk does, so marking
takes no lock and the walk can't see a marked command before it is
published. smp_wmb() orders the mark before the io->cmd store; the
cancel loads cmd with READ_ONCE(io->cmd) before reading cmd->flags.
A marked command has to be completed by io_uring_cmd_done(), so a failed
FETCH and the inline UBLK_IO_RES_OK of NEED_GET_DATA now do that. The CQE
is the same.
Fixes: 216c8f5ef0f2 ("ublk: replace monitor with cancelable uring_cmd")
Cc: stable@vger.kernel.org
Tested-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Link: https://patch.msgid.link/20261011063555.1142277-3-tom.leiming@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
A cancel completes a fetched io command and sets io->cmd to NULL, but
the io still counts as ready. Only ubq->canceling stops ublk_queue_rq()
from using the NULL io->cmd. Two paths clear that flag too early:
partial FETCH round recovery, two queues
------------------- --------------------
task A: FETCH tags 0-2 q0 ready: q0->canceling = false
task A exits: q0 task exits:
start_cancel(): mark queues start_cancel(): ub->canceling
tags 0-2: io->cmd = NULL is still set -> q0 not marked
task B: FETCH tag 3 q0: io->cmd = NULL
queue ready: canceling = false q1 ready
START_DEV END_USER_RECOVERY
read -> ublk_queue_cmd(NULL) read on q0 -> ublk_queue_cmd(NULL)
A FETCH round runs from the open of /dev/ublkcN until
ublk_reset_ch_dev() resets the queues. A canceled command can't be
fetched again in the same round. So the ready check only needs one
fact: did this round see a cancel? ub->canceling already records it:
ublk_start_cancel() sets it under cancel_mutex before any command is
taken. It is only cleared too early, when all queues are ready.
Fix:
- clear ub->canceling only in ublk_reset_ch_dev()
- when a queue gets ready, clear ubq->canceling only if ub->canceling
is not set, and check it under cancel_mutex
The check and the marking are ordered by cancel_mutex:
check first: queue cleared -> start_cancel() marks it again
-> command taken
cancel first: check sees ub->canceling -> queue stays canceling
While ub->canceling is set, no queue clears its flag. So the "all
queues marked" shortcut in ublk_start_cancel() is correct again.
UBLK_F_BATCH_IO follows the same rule. It does not dispatch through
io->cmd, so it never hit the oops, but a round which lost a fetch
command is treated the same way: it has to start over.
Fixes: 728cbac5fe21 ("ublk: move device reset into ublk_ch_release()")
Fixes: 3f3850785594 ("ublk: fix batch I/O recovery -ENODEV error")
Cc: stable@vger.kernel.org # v6.17+: needs cancel_mutex
Reported-by: Josef Bacik <josef@toxicpanda.com>
Closes: https://lore.kernel.org/linux-block/20260928-b4-ublk-cancel-stop-v1-0-4a4360232a46@toxicpanda.com/
Tested-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Link: https://patch.msgid.link/20261011063555.1142277-2-tom.leiming@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
The TP-LINK SM410U is a 2.5G copper module which is based on RealTek
RTL8221B, but doesn't seem to offer any way to access the PHY over I2C.
Add a quirk to disable in-band which isn't supported by the PHY.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/172a4fed74d35b1be51f4e8aba2222aa33845b08.1791679663.git.daniel@makrotopia.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The RX refill path should set the IC bit once every rx_coal_frames
descriptors and leave the rest to the RX watchdog (RIWT), as done by
commit d429b66e2ebc ("net: stmmac: Implement RX Coalesce Frames
setting").
Since commit 6fa9d691b91a ("net: stmmac: Prevent divide-by-zero") the
counter is increased by N+1 for each descriptor and then compared
against N, so it is always reset to 0. With a non-zero rx-frames, IC is
therefore set on every descriptor and every frame raises an interrupt.
Fix this by counting one descriptor at a time in a helper used by both
stmmac_rx_refill() and stmmac_rx_refill_zc(). Platforms without RIWT
(IC on every descriptor) are unaffected.
Cc: stable+noautosel@kernel.org # never worked
Fixes: 6fa9d691b91a ("net: stmmac: Prevent divide-by-zero")
Signed-off-by: Pengfei Zhang <zhangfeionline@gmail.com>
Link: https://patch.msgid.link/20261009111138.320089-1-zhangfeionline@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
dax_core_init() returns success when either the character device
number allocation or DAX bus registration fails, even though the
error paths have already released the initialized resources.
For modular DAX, this allows the module to remain loaded despite
failed initialization, leaving subsequent users and module removal
to access resources that have already been released.
Return the error code from the failed operation so that module
loading fails rather than reporting successful initialization
after the resources have been released.
Fixes: 9567da0b408a ("device-dax: Introduce bus + driver model")
Cc: stable@vger.kernel.org
Signed-off-by: Jianyun Gao <jianyungao89@gmail.com>
Reviewed-by: Gregory Price (Meta) <gourry@gourry.net>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Link: https://patch.msgid.link/20261012151857.22534-1-jianyungao89@gmail.com
Signed-off-by: Alison Schofield <alison.schofield@intel.com>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth-next
Luiz Augusto von Dentz says:
====================
bluetooth-next pull request for net-next:
core:
- Move H:4 reassembly into the Bluetooth core
- Add generic support for vendor packets
- MGMT: Add management security level changed event
- MGMT: Allow connection rate request as peripheral
- L2CAP: Add locking annotations and fix locking of channel operations
- Enable context analysis for headers
drivers:
- btusb: Add support for Bulk Serialization Mode
- btusb: Build the driver from multiple source files
- btusb: Add support for Qualcomm multi-subsystem QCC2072
- btusb: Add support for Intel Otter Peak2 (OrP2)
- btusb: Add USB ID 13d3:3556 for RTL8821CE
- btusb: Add Realtek RTL8821CE device 13d3:3558
- btusb: Add device ID for Realtek RTL8761BUE (StarTech AV53C1)
- btusb: Add device ID for MediaTek MT7925 (13d3:3631)
- btusb: add ASUS 0b05:1825 and 0b05:1868 to QCA Rome quirks
- btintel: load unlocker.sfi to allow unsigned firmware
- btintel_pcie: Add support for Intel Draco
- btintel_pcie: Add PCI ID and CNVi ID for Titan Lake
- btintel_pcie: add MDBGC multi-buffer DBGC allocation
- btintel_pcie: unified decoder coredump format
- btmtk: Add MT7928 support
- btmtk: Improve BT firmware logging
- btbcm: Add Synaptics 4384 chip support
- btbcm: Add entry for BCM4356A3 UART bluetooth
* tag 'for-net-next-2026-10-14' of git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth-next: (120 commits)
Bluetooth: hci_sync: scan unfiltered for host-resolved RPAs
Bluetooth: Fix the LE Read Connection Interval response layout
Bluetooth: L2CAP: Remove dead configuration state code
Bluetooth: L2CAP: Only accept connection responses to outstanding requests
Bluetooth: L2CAP: Use default TxWindow for zero socket option
Bluetooth: L2CAP: Reject invalid TxWindow in configuration requests
Bluetooth: L2CAP: Reject signalling responses in invalid states
Bluetooth: L2CAP: Reject mode changes on connected channels
Bluetooth: hci_event: Fix LE link drop on remote features command status
Bluetooth: btusb: annotate Intel Otter Peak 2 USB ID
Bluetooth: btintel_pcie: Add PCI ID for Titan Lake
Bluetooth: btmtk: Add MT7928 support
Bluetooth: MSFT: Serialize handle map cleanup with event processing
Bluetooth: MGMT: Allow connection rate request as peripheral
Bluetooth: btintel: load unlocker.sfi to allow unsigned firmware
Bluetooth: btintel: Zero version data before parsing TLVs
Bluetooth: selftest: Pin module for selftest debugfs file operations
Bluetooth: L2CAP: Hold the listener while notifying child teardown
Bluetooth: hci_vhci: Pin module for debugfs file operations
Bluetooth: hci_sync: Protect IRK identity reads with RCU
...
====================
Link: https://patch.msgid.link/20261014184327.1648397-1-luiz.dentz@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
https://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless-next
Johannes Berg says:
====================
Lots of updates:
- mt76
- mt7625 NAN fixes
- mt7996 connection monitor fix
- offchannel/scan improvements
- MLO fixes
- rtw88
- RTL8723BS support
- SDIO TX performance improvements
- rtw89
- P2P_DEVICE support
- BT-coex updates
- mm81x
- monitor mode support
- rx improvements
- fix persistent NO_IR flag bug
- iwlwifi
- UHR sniffer support
- support for newer UEFI/BIOS
- cfg80211/mac80211
- latency improvements (e.g. some lockless ops)
- RTNL reduction & locking cleanups
* tag 'wireless-next-2026-10-15' of https://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless-next: (462 commits)
wifi: cfg80211: fix potential requested_regd leak
wifi: cfg80211: reject WEP/TKIP with multi-link
wifi: cfg80211: track SSID regardless of country data
wifi: mac80211: tests: fix stack variable bug
wifi: iwlwifi: mld: include VI packets in low latency detection
wifi: iwlwifi: trans: remove the unused write8 operation
wifi: iwlwifi: support SCD_QUEUE_CONFIG_CMD version 4
wifi: iwlwifi: mvm: uninline iwl_mvm_ftm_start versions
wifi: iwlwifi: mvm: fix beacon flags compiler warning
wifi: iwlwifi: mld: fix const in iwl_mld_mu_mimo_iface_iterator
wifi: iwlwifi: mld: avoid const in iwl_mld_chan_load_update_iter()
wifi: iwlwifi: pcie: correctly read WFPM_OTP_CFG1_ADDR
wifi: iwlwifi: mvm: remove dead sta check in rx_monitor_no_data
wifi: iwlwifi: mld: don't enable RSS if there is one queue.
wifi: iwlwifi: move context info APIs to a separate file
wifi: iwlwifi: fw: api: fix D3 sequence counter docs
wifi: iwlwifi: mvm: fix the VHT MCS sent to a v1 firmware
wifi: iwlwifi: bump maximum core version for BZ/SC/DR to 109
wifi: iwlwifi: dbg: validate the index before accessing an array
wifi: iwlwifi: don't mask ucode_ver in iwl_sar_geo_support()
...
====================
Link: https://patch.msgid.link/20261015105035.137870-3-johannes@sipsolutions.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
When nvdimm_clear_poison() fails, it returns a negative error code.
However, 'cleared' is a signed long while 'size' is size_t.
The comparison 'cleared < size' promotes 'cleared' to an unsigned long.
So the error condition can be bypassed and the function can
incorrectly return success if the subsequent flush succeeds.
Explicitly check for negative error codes before comparing with 'size'.
Fixes: 868f036fee4b ("libnvdimm: fix mishandled nvdimm_clear_poison() return value")
Assisted-by: LLM
Reported-by: Dan Carpenter <error27@gmail.com>
Closes: https://lore.kernel.org/all/90d3d353-28e9-4f6d-b141-a9b7157d5514@moroto.mountain/
Signed-off-by: Serhat Kumral <serhatkumral1@gmail.com>
Reviewed-by: Alison Schofield <alison.schofield@intel.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Link: https://patch.msgid.link/20261002165641.14479-1-serhatkumral1@gmail.com
Signed-off-by: Alison Schofield <alison.schofield@intel.com>
|
|
PCI core/ASPM service driver allows controlling ASPM state through
pci_disable_link_state() and pci_enable_link_state() API. It was
decided earlier (see the Link below), to not allow ASPM changes when OS
does not have control over it but only log a warning about the problem
(commit 2add0ec14c25 ("PCI/ASPM: Warn when driver asks to disable ASPM,
but we can't do it")). Similarly, if ASPM is not enabled through
config, ASPM cannot be disabled.
A number of drivers have added workarounds to force ASPM off with own
writes into the Link Control Register (some even with comments
explaining why PCI core does not disable it under some circumstances).
According to the comments, some drivers require ASPM to be off for
reliable operation.
Having custom ASPM handling in drivers is problematic because the state
kept in the ASPM service driver is not updated by the changes made
outside the link state management API.
As the first step to address this issue, make pci_disable_link_state()
unconditionally disable ASPM so the motivation for drivers to come
up with custom ASPM handling code is eliminated.
Make pci_disable_link_state() function comment to comply kerneldoc
formatting while changing the description.
[bhelgaas: To take advantage of the PCI core ASPM handling, drivers that
need to quirk ASPM should depend on CONFIG_PCIEASPM and have their custom
ASPM code removed.]
Link: https://lore.kernel.org/all/CANUX_P3F5YhbZX3WGU-j1AGpbXb_T9Bis2ErhvKkFMtDvzatVQ@mail.gmail.com/
Link: https://lore.kernel.org/all/20230511131441.45704-1-ilpo.jarvinen@linux.intel.com/
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
[bhelgaas: extract just the aspm.c change and update commit log]
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Link: https://patch.msgid.link/20230918131103.24119-4-ilpo.jarvinen@linux.intel.com
|
|
amd_spi_fifo_xfer() unconditionally decrements xfer->len when a tx_buf
is present. The SPI core does not reject zero-length transfers, so a
transfer submitted through the SPI_IOC_MESSAGE ioctl with len == 0 and
a tx_buf makes the u32 len wrap to 0xFFFFFFFF, and the following
per-byte loop writes far past the FIFO window.
The driver also never checks the cumulative FIFO offset against
AMD_SPI_FIFO_SIZE. The core splits each transfer down to the
controller's max transfer size but does not bound the total message
length, so a message of several transfers whose total exceeds the FIFO
overflows it with out-of-bounds MMIO writes (tx) and reads (rx).
Skip zero-length transfers and reject any transfer that would run past
the end of the FIFO window.
Tested with a user-space replica of the fifo_xfer length arithmetic
(ASAN build): a zero-length transfer with tx_buf makes len-- wrap to
0xffffffff, and skipping zero-length transfers avoids the wrap. No AMD
SPI hardware was available, so the FIFO bound check was verified by
inspection against AMD_SPI_FIFO_SIZE.
Fixes: bbb336f39efc ("spi: spi-amd: Add AMD SPI controller driver support")
Cc: stable@vger.kernel.org # 5.8+
Reviewed-by: Guanglei Zhu <zhugl3@xiaopeng.com>
Signed-off-by: Weibin Liu <liuwb@xiaopeng.com>
Link: https://patch.msgid.link/20261014-spi-amd-fifo-fix-v1-1-9fec9ab753e8@xiaopeng.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
The MSI-X Table lives in device MMIO space behind a BAR and the MSI Mask
register in Configuration Space, so neither is reachable while the Link is
down. While a Downstream Port has the Link contained by DPC it completes
these accesses with Unsupported Request, and reads return all ones.
If the upstream Root Port implements the RP Extensions for DPC, it
additionally reports that UR completion as an RP PIO error and triggers a
second containment event, this time at the Root Port, which contains every
device below it. So a contained Link on one Downstream Port turns into a
far wider outage that takes down unrelated devices.
pci_free_irq_vectors() is called from driver error_detected() and
prepare-for-reset callbacks, i.e. while the Link is contained and before
the reset and the pci_restore_state() that follows it, and it masks every
descriptor. Skip the programming when pci_channel_offline(), which also
covers surprise removal. The msix_ctrl and msi_mask caches are still
updated, so the restore paths replay the intended state once the Link is
back up, and report_slot_reset() clears the offline state before the driver
callback runs, so recovery is unaffected. pci_msix_write_tph_tag() flushes
its Vector Control update with an unconditional read, so return -EIO there
rather than issue it for a write that was skipped; the caller disables TPH
in response. error_state is only set once containment has occurred, so this
covers the case where the kernel knows the Link is down; it is not mutual
exclusion against a containment event that begins concurrently.
This does not attempt to make every Configuration Space access safe while
the Link is contained.
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260909164908.2562818-1-vidyas@nvidia.com
|
|
for-7.4/block
Pull NVMe updates from Keith:
"- Parallelize nvme-tcp I/O queue setup and use the caller's network
namespace
- New multipath sysfs attributes for path state, available
iopolicies, and fail on controller loss
- Multipath fixes for failfast/ANA path evaluation and discovery
subsystem iopolicy reporting
- Apple M4 ANS2 NVMe support
- Traffic based keep alive support for the target
- DH-HMAC-CHAP hardening for host and target: enforce bidirectional
auth, require auth before property set, fix state UAF, reject short
payloads, restrict secret visibility
- Target reservation preempt fixes
- Target passthru I/O use-after-free fix
- Fix out-of-bounds reads and memory leaks from device and host
supplied data: FDP config log, ZNS zone reports, file namespace
range errors
- Enforce partition checks for unprivileged passthrough
- nvme-rdma fix for bogus remote invalidation
- nvmet-fc fix for recursive locking hang on port unlink
- Revert DMA opt max_hw_sectors clamp due to performance regressions
- Misc cleanups"
* tag 'nvme-7.4-2026-10-15' of git://git.infradead.org/nvme: (28 commits)
Revert "nvme-pci: clamp max_hw_sectors based on DMA optimized limitation"
nvme: make the DH-HMAC-CHAP secret attributes readable by root only
nvmet: zns: don't leak stale memory in zone management receive
nvme: bound FDP config log reads to the received buffer
nvme: pass partition flags to passthrough checks
nvmet-auth: reject undersized AUTH_SEND payloads
nvmet-fc: don't drop the last assoc reference in nvmet_fc_delete_ctrl()
nvmet: require authentication before property set
nvmet: free the bvec on file namespace range errors
nvme-auth: fail when the controller does not authenticate itself
nvmet-auth: serialize access to DH-HMAC-CHAP state to fix UAF
nvmet: fix preempt of another registrant by the reservation holder
nvmet: fix use-after-free of the old registrant in nvmet_pr_preempt()
nvme-multipath: add fail_on_ctrl_loss sysfs attribute
nvme-multipath: fix path state evaluation for failfast and ANA
nvme-multipath: report "none" iopolicy for discovery subsystems
nvme-rdma: don't complete a request after a bogus remote invalidation
nvmet: fix use-after-free in passthru I/O hotpath
nvmet: introduce struct nvmet_passthru
nvmet: Add traffic based keep alive support to configfs
...
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc
Pull SoC fixes from Arnd Bergmann:
"A small number of last minute bug fixes came up since LPC.
Most of these are for devicetree files:
- Tegra264 needs fan descriptions to keep the system from overheating
- A cleanup on TI DRA7 caused a regression on PCI support and needs a
fixup
- The beaglev-fire board can no longer probe the mcp3564 sensor after
a an inconsistent update in 7.1
- A few fixes address compile time dtc warnings on tegra
The other three patches are all addressing minor problems
in tegra drivers and the sama5 defconfig"
* tag 'soc-fixes-7.3-3' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc:
ARM: configs: sama5: enable current Microchip KSZ DSA symbols
soc/tegra: pmc: Fix uninitialised clock rate
riscv: dts: microchip: beaglev-fire: remove double definition of gpio interrupts
ARM: dts: ti/omap: dra7: fix PCIe PHY clock divider definition
soc/tegra: fuse: Add missing newline to APBMISC error message
arm64: tegra: Add PCIe root ports on Tegra264
arm64: tegra: Reorder reg and reg-names to match bindings
arm64: tegra: Add PWM fan on Jetson AGX Thor DevKit
arm64: tegra: Add PWM controllers on Tegra264
|
|
tegra_spi_probe() initializes SPI_COMMAND1 with only SPI_M_S, leaving
CS_POL_INACTIVE clear for every chip select. This drives every native
active-low chip select low until tegra_spi_setup() runs for that device.
SPI children are registered and probed one at a time. A synchronous
probe of an earlier child can therefore transfer while a later child's
chip select is still asserted. On a Tegra234 system with active-low
devices on CS0 and CS1, this caused both devices to be selected when the
CS1 device probed first. Its initialization then intermittently failed.
A logic analyzer showed CS0 remained low throughout the CS1 transfer and
went high only after the CS1 probe failed.
Walk the controller's device tree children before SPI_COMMAND1 is
written and set CS_POL_INACTIVE for each valid chip select without
spi-cs-high. That way the line is already deasserted when the first
child transfers and does not change level when setup runs.
Disabled children are walked too. The core only registers available
children, so tegra_spi_setup() never runs for a disabled one and probe
is the only chance to park its chip select at the level its hardware
needs.
If an enabled child shares a chip select with disabled ones, its
polarity takes priority. If only disabled children use a chip select
and they disagree, warn and leave the line inactive high.
Fixes: f333a331adfa ("spi/tegra114: add spi driver")
Assisted-by: Claude:Opus-5
Signed-off-by: Abraham Zukor <abe@maticrobots.com>
Link: https://patch.msgid.link/20261010195503.623287-1-abe@maticrobots.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Both fsl_lpspi_dma_transfer() and fsl_lpspi_pio_transfer() only return
negative error values in case of error or 0 in case of success.
Directly return ret, there is no need to check for positive error values
and return them as 0.
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
Link: https://patch.msgid.link/20261009-spi-fsl-lpspi-more-cleanups-v1-2-2d4549d5205e@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
spi_controller if possible
Instead of passing a spi_controller pointer to functions and using
spi_controller_get_devdata() to get the struct fsl_lpspi_data pointer,
directly pass a struct fsl_lpspi_data pointer.
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20261009-spi-fsl-lpspi-more-cleanups-v1-1-2d4549d5205e@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Since commit faa8e404ad8e ("spi: imx: support dynamic burst length for
ECSPI DMA mode"), the driver allocates its private DMA TX/RX buffers
with kzalloc(dma_len), where dma_len is only rounded up to 4 bytes, but
maps, syncs and unmaps them with DMA_CACHE_ALIGNED_LEN(dma_len), i.e.
rounded up to the cache line size. As a result, the DMA API operates on
memory beyond the end of the allocation.
This was observed on a custom i.MX8MP platform with an mcp251xfd CAN
controller under heavy RPMSG load, as a series of refcount_t warnings
triggered from rpmsg_recv_done() in virtio_rpmsg_bus:
refcount_t: addition on 0; use-after-free.
refcount_t: underflow; use-after-free.
refcount_t: saturated; leaking memory.
Leading finally to an "Unable to handle kernel paging request" oops.
Reproducing this problem took between 30 minutes and roughly 30 hours
under the same workload.
Hardware watchpoints on the refcount of a struct rpmsg_endpoint showed
it was overwritten from 2 to 1 by __pi_memcpy_generic() in the
irq/155-spi2.0 thread. Additionally it was found that a "normal"
rpmsg_recv_done() took 11µs +/-2µs from its kref_get to kref_put. In
the error case the same sequence took about 215µs +/-10µs. Based on
this, an LLM was used to identify the size mismatch fixed here.
Allocate the buffers with DMA_CACHE_ALIGNED_LEN(dma_len) so they cover
the full mapped length, and use the same length for
dma_sync_single_for_cpu() to stay consistent with map and unmap.
With this patch applied, the same workload ran for more than 200 hours
without the problem reproducing.
Assisted-by: LLM
Fixes: faa8e404ad8e ("spi: imx: support dynamic burst length for ECSPI DMA mode")
Cc: stable@vger.kernel.org
Signed-off-by: Richard Leitner <richard.leitner@linux.dev>
Link: https://patch.msgid.link/20261015-fix-spi-imx-dma-v1-1-e28259671392@linux.dev
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Commit 4d4c10f763d7 ("PCI: Explicitly put devices into D0 when
initializing") added pci_pm_power_up_and_verify_state() to pci_pm_init().
The helper ignores pci_power_up()'s return value indicating that BARs need
restoring. Its resume callers restore saved configuration afterwards;
pci_pm_init() does not.
D3hot-to-D0 may reset a device and lose its BARs (PCI PM 1.2, sec 5.4.1).
After enumeration powers it up, pci_enable_device() reads D0 and skips the
full power-up path. Previously, a device still in D3hot at enable time took
that path, including BAR restoration.
On AMD Venice SP7, two AMD xHCI controllers [1022:17d2] are in runtime
D3hot with No_Soft_Reset clear before kdump. Diagnostic reads on
v7.3-rc6-37 show the following for 0000:01:00.4 in the crash kernel:
PMCSR BAR0
before the platform power call 0003 f4100004
after the platform power call 0003 f4100004
after native D0 power-up 0000 00000004
The upper BAR DWORD stays zero. pci_power_up() returns 1, but the BAR
address remains zero while dev->resource[0] still contains 0xf4100000.
Without BAR restoration, the crash kernel hangs during boot. The last
console output includes:
pci 0000:01:00.4: enabling device (0000 -> 0002)
Restoring both controllers' BARs gives matching readbacks and allows kdump
to save a vmcore and reboot. An AMD Turin xHCI [1022:1557] shows the same
BAR loss; restoring its BAR also allows kdump to complete.
Read the initial power state first. Otherwise, pci_power_up() may request a
BAR restore based on a cached PCI_UNKNOWN state, even when the device is
already in D0.
For devices already in D0, only synchronize the platform state so
that ACPI calls such as _REG still run. For other states, use
pci_set_full_power_state() to power up the device and restore BARs
when needed.
Fixes: 4d4c10f763d7 ("PCI: Explicitly put devices into D0 when initializing")
Suggested-by: Mario Limonciello <mario.limonciello@amd.com>
Suggested-by: Rafael J. Wysocki <rafael@kernel.org>
Link: https://lore.kernel.org/r/CAJZ5v0ie2AvCbdv8-zkpYzHLcTwUPUj+=WWOQjDAPmSepSAQPg@mail.gmail.com/
Signed-off-by: Keng-Yu Lin <keng-yu.lin@hpe.com>
[bhelgaas: shorten comment]
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org>> ---
Reviewed-by: Desnes Nunes <desnesn@redhat.com>
Reviewed-by: Rafael J. Wysocki (Intel) <rafael@kernel.org>
Cc: stable@vger.kernel.org
Link: https://patch.msgid.link/20261015020811.1132314-1-keng-yu.lin@hpe.com
|
|
The devres conversion replaced manual pm_runtime_enable()/disable() with
devm_pm_runtime_set_active_enabled() and dropped the remove-time runtime
PM teardown. The managed release tears runtime PM down in the wrong
order: it calls pm_runtime_dont_use_autosuspend() before
pm_runtime_disable(), i.e. while runtime PM is still enabled, and devres
is LIFO so the devm_clk_get_enabled() release runs afterwards.
At remove(), pm_runtime_put_sync() leaves the device active with the
autosuspend timer armed. Clearing use_autosuspend then makes rpm_idle()
suspend immediately, and xiic_i2c_runtime_suspend() clk_disable()s the
clock. The later devm_clk_get_enabled() release clk_disable_unprepare()s
the already-disabled clock, so clk_core_disable() WARNs ("clkN already
disabled") on every teardown.
Drop the managed helper and restore the non-managed runtime PM setup and
teardown, so runtime PM is enabled once in probe and disabled once in
remove and the clock enable count stays balanced.
Order the remove teardown so runtime PM is disabled before the usage
count is dropped: pm_runtime_disable() followed by pm_runtime_put_noidle()
instead of pm_runtime_put_sync(). Dropping the count first with autosuspend
still enabled arms the autosuspend timer, and a preemption before
pm_runtime_disable() lets it expire, suspend the device and clk_disable()
the clock, which the devm release then disables again - the same WARN.
Restore the non-managed IRQ request as well. The probe error paths now
unwind runtime PM by hand via goto, and a devm-registered handler is only
freed later during devres unwind, so it would stay live across the manual
teardown (and across any future failing step added after it). Request the
IRQ with request_threaded_irq() and free it explicitly in the probe error
path and in remove. The probe error path also clears use_autosuspend, to
match the teardown in remove().
Fixes: 50c63491ff26 ("i2c: xiic: switch to devres managed APIs")
Signed-off-by: Abdurrahman Hussain <abdurrahman@nexthop.ai>
Cc: <stable@vger.kernel.org> # v7.1+
Signed-off-by: Andi Shyti <andi.shyti@kernel.org>
Link: https://patch.msgid.link/20261014-i2c-xiic-restore-runtime-pm-teardown-v8-1-40668c270069@nexthop.ai
|
|
syzbot reports a lockdep splat when I/O is dispatched to a ublk device
with UBLK_F_SHMEM_ZC enabled:
WARNING: suspicious RCU usage
lib/maple_tree.c:720 suspicious rcu_dereference_check() usage!
...
Call Trace:
mas_root lib/maple_tree.c:720 [inline]
mas_start+0x3cc/0x640 lib/maple_tree.c:1250
mas_state_walk lib/maple_tree.c:3006 [inline]
mas_walk+0x1db/0x490 lib/maple_tree.c:4458
ublk_try_buf_match drivers/block/ublk_drv.c:5639 [inline]
ublk_init_iod+0x5d2/0x1120 drivers/block/ublk_drv.c:521
ublk_dispatch_req+0x97/0x9c0 drivers/block/ublk_drv.c:1776
tctx_task_work_run+0x73d/0xa80 io_uring/tw.c:96
ublk_try_buf_match() calls mas_walk() on the buffer tree without
holding the RCU read lock or the tree lock. Writers are already
excluded because ublk_lock_buf_tree() freezes the queue, so there is
no real race, but maple tree readers must still hold one of the two.
Take the RCU read lock around the walk.
Fixes: 4d4a512a1f87 ("ublk: add PFN-based buffer matching in I/O path")
Reported-by: syzbot+b48868c46e62c1c4cb2c@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=b48868c46e62c1c4cb2c
Assisted-by: LLM
Signed-off-by: Paolo Bertinetti <paolobertinetti200f4@proton.me>
Reviewed-by: Ming Lei <tom.leiming@gmail.com>
Link: https://patch.msgid.link/179010105967.8478.2344875677216734809@proton.me
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/iommu/linux
Pull IOMMU fixes from Joerg Roedel:
"AMD IOMMU:
- Revert an interrupt remapping check removal that caused GPU
regressions on some platforms
Generic page tables:
- Fix KUnit builds with distributed Clang ThinLTO by preserving
compiler flags, preventing x86-64 code generation from arm64
bitcode"
* tag 'iommu-fixes-v7.3-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/iommu/linux:
iommupt/kunit: Use if_changed_rule for the kunit
Revert "iommu/amd: Remove redundant check in irq_remapping_select()"
|
|
Previously, the IRDMAQPSQ_PBLADDR FRWR WQE field was being populated
using a mask that was right-shifted, whereas it should have been the
address being right-shifted.
This ended up working fine in many cases because the lowermost 12 bits
are always zero (the addresses are page aligned) and the uppermost 12
bits are usually zero too, but it could break for DMA addresses that
are wider than 52 bits.
Fix this by right-shifting the address instead of the mask, similar
to how it was done in the original i40iw driver.
Tested with NFS over RDMA with some added debug logs to ensure it
was hitting both the direct (single page) path as well as multi
page (< 512).
Fixes: 3f49d6842569 ("RDMA/irdma: Implement HW Admin Queue OPs")
Signed-off-by: Jacob Moroni <jmoroni@google.com>
Link: https://patch.msgid.link/20261015161756.3078300-1-jmoroni@google.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
|
|
If driver_f.qp_priv_init() fails, execution jumps to bail_rq_wq, which
bypasses bail_qpn. Because alloc_qpn() has already succeeded at this point,
the allocated QP number is left unreleased.
Jump to bail_qpn instead to ensure the QPN is properly freed via
rvt_free_qpn().
Fixes: 5190f052a3654 ("IB/hfi1: Allow the driver to initialize QP priv struct")
Signed-off-by: Serhat Kumral <serhatkumral1@gmail.com>
Link: https://patch.msgid.link/20261015102518.15780-1-serhatkumral1@gmail.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
|
|
of_overlay_remove_all() has no in-tree callers.
It is also subtly broken. It passes &ovcs->id to of_overlay_remove(),
which zeroes *ovcs_id once the changeset is reverted, before calling
free_overlay_changeset(). Through the aliased pointer that zeroes
ovcs->id itself, so free_overlay_changeset()'s "ovcs->id > 0" guard is
false and both idr_remove() and list_del() are skipped: the
overlay_changeset is kfree()d while still linked in ovcs_list and
registered in ovcs_idr.
Rather than fix a helper nobody calls, remove it, along with its
!CONFIG_OF_OVERLAY stub and the paragraph documenting it in
overlay-notes.rst and in the zh_CN translation.
ovcs_list itself stays: overlay_removal_is_ok() still walks it to check
that a changeset is topmost.
Reported-by: Sashiko AI <sashiko-bot@kernel.org>
Closes: https://lore.kernel.org/20260901014554.83A8D1F000E9@smtp.kernel.org
Suggested-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/all/20260928182658.GA185307-robh@kernel.org/
Signed-off-by: Abdurrahman Hussain <abdurrahman@nexthop.ai>
Link: https://patch.msgid.link/20261014-b4-of-overlay-remove-all-fix-v2-1-cf5ea2cc9dc3@nexthop.ai
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
|
|
bcm_qspi_mspi_exec_mem_op() builds a local command buffer cmd[6]
holding one opcode byte, up to four address bytes and one spare
byte, and sets t[0].len = op->addr.nbytes + op->dummy.nbytes + 1.
bcm_qspi_exec_mem_op() only validates op->addr.nbytes <= 4 and never
bounds op->dummy.nbytes, which spi_mem_check_op() also leaves
unrestricted. With addr.nbytes = 4 and dummy.nbytes >= 2 the length
becomes 7 or more, so write_to_hw() reads past cmd[] on the kernel
stack and pushes the leaked bytes into the hardware TXRAM.
Reject operations whose dummy phase does not fit the remaining byte
of cmd[] by requiring op->dummy.nbytes <= 1, so such transfers fall
back to the generic SPI path instead of overrunning the buffer.
Reviewed-by: Guanglei Zhu <zhugl3@xiaopeng.com>
Fixes: 5f195ee7d830c ("spi: bcm-qspi: Implement the spi_mem interface")
Cc: stable@vger.kernel.org # 4.18+
Signed-off-by: Weibin Liu <liuwb@xiaopeng.com>
Link: https://patch.msgid.link/20261015-bcm-qspi-dummy-bytes-v1-1-c319da3c2497@xiaopeng.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
* for-next/fw-rmi:
: RMM v2.0 base RMI support
firmware: arm_rmm: hotplug: Skip memory added to ZONE_MOVABLE
firmware: arm_rmm: Add wrappers for Realm related RMI commands
arm64: Block hibernate and kexec while RMM is active
PM: hibernate: Add an arch hook for preventing hibernation
firmware: arm_rmm: Ensure the RMM has GPT entries for memory
firmware: arm_rmm: Activate the RMM
firmware: arm_rmm: Add support for SRO
firmware: arm_rmm: Configure the RMM with the host's page size
firmware: arm_rmm: Check for RMI support at init
firmware: arm_rmm: Add SMC definitions for calling the RMM
arm64: mm: Handle Granule Protection Faults (GPFs)
|
|
'for-next/preemptible-this-cpu', 'for-next/cpu-errata', 'for-next/mpam', 'for-next/kselftest', 'for-next/efi', 'for-next/secondary-boot-prep', 'for-next/sysreg' and 'for-next/gcs', remote-tracking branch 'arm64/for-next/perf' into for-next/core
* arm64/for-next/perf: (22 commits)
drivers/perf: apple_m1: Add macOS 27 M2 Event
perf/marvell: cn10k_tad: Publish the OF module alias
perf: xgene: initialize lock before requesting IRQ
perf: marvell: avoid large stack variables
perf/dwc_pcie: Fix PCI device reference leak in probe
coresight: perf: Suppress ITRACE start records
perf: arm_spe: Suppress redundant ITRACE start records
perf/core: Export perf_event_itrace_started()
perf/cxl: Clear stale overflow status before using a counter
perf/cxl: Don't log through pmu.dev in the overflow interrupt handler
perf/cxl: Validate the hardware-reported counter width
perf/cxl: Unfreeze counters after handling an overflow interrupt
perf/cxl: Don't share the overflow interrupt, and keep it pinned
cxl/pci: Add the PMUs after configuring events
perf/cxl: Split the MSI vector out of info->irq
perf/cxl: Accept an overflow interrupt on MSI message number 0
perf/cxl: Fix the counter overflow delta fixup
perf/cxl: Clear stale event fields before reprogramming a counter
perf/cxl: Program the requested event group on configurable counters
drivers/perf: hisi: Add cycle event for new MN PMU
...
* for-next/misc:
arm64: cpufeature: Enable BBML2 for NVIDIA Rigel
arm64: cputype: Add NVIDIA Rigel definitions
arm64: raise default ARCH_MMAP_RND_BITS for 48-bit and 52-bit VA
arm64/debug: mask debug exceptions when called from common code
arm64: clear_page[s] using memset
arm64/sve: Don't zero the SVE state buffer when the SVE state is live
arm64/cpufeature: Drop redundant define(ID_AA64MMFR0..._SUPPORTED_LPA2)
arm64: Use proper include variant
arm64: include: fix repeated word 'to' in comment
selftests: arm64: fix typo "hander" in comment
arm64: fix typos in comments
docs: arm64: Fix typo "initalised" in booting.rst
arm64: alternative: Remove 'cfg' parameter
arm64: mte-kasan: Use IS_ENABLED() rather than ALTERNATIVE() 'cfg' parameter
arm64/cpufeature: Change description for ARM64_HAS_LPA2
* for-next/be-gone:
drivers: Kconfig: Simplify dependencies on ARM64 && !CPU_BIG_ENDIAN
arm64: Kbuild: Remove vestigial big-endian support
gve: Add explicit ARM64 dependency to Kconfig
crypto: hisilicon/sec2: Remove check of CONFIG_CPU_LITTLE_ENDIAN
arm64: Remove all usage of __AARCH64EB__
arm64: Remove all usage of CONFIG_CPU_BIG_ENDIAN
KVM: arm64: Remove support for a big-endian hypervisor object
arm64: vdso32: Always build compat vDSO object as little-endian
arm64: assembler: Remove endianness helper macros
arm64: lib: Assume a little-endian kernel in optimised string routines
arm64: lib: Assume a little-endian kernel in custom library routines
arm64: crypto: Assume a little-endian kernel
arm64: bpf: Remove big-endian support from the JIT compiler
selftests/arm64: Remove big-endian support
* for-next/smccc-bus:
: Switch Arm SMCCC firmware services to an SMCCC bus
arm64: rsi: Correct type of realm_config::hash_algo
firmware: smccc: Pass only the required device probe arguments
firmware: arm_rmm: Add SPDX and sort the Kconfig and Makefile entries
coco: guest: arm64: Remove dummy CCA platform device
coco: guest: arm64: Add sysfs ABI for CCA Realm guests
firmware: smccc: arm-cca-guest: Bind the TSM provider to an SMCCC device
virt: coco: arm-cca-guest: Rename TSM report source file
arm64: realm: Move Realm memory encryption ops to RSI code
firmware: arm_rmm: Move RSI support out of arch/arm64
firmware: hwrng: arm_smccc_trng: Register as an SMCCC device
firmware: smccc: Add an Arm SMCCC bus
* for-next/preemptible-this-cpu:
: Preemptible this_cpu_*() operations
arm64: percpu: Remove _pcp_protect*() wrappers
arm64: percpu: Implement preemptible CMPXCHG128 ops
arm64: percpu: Implement preemptible CMPXCHG ops
arm64: percpu: Implement preemptible XCHG ops
arm64: percpu: Implement preemptible return RMW ops
arm64: percpu: Implement preemptible void RMW ops
arm64: percpu: Implement preemptible read/write ops
arm64: percpu: Add infrastructure for preemptible this_cpu_*() ops
arm64: entry: sdei: Make 'tsk' available
arm64: entry: sdei: Restore all clobberable GPRs
arm64: gpr-num: add __GPR_NUM() helper
arm64: gpr-num: Add wxN aliases for wN registers
arm64: percpu: Factor out percpu offset asm
arm64: ptrace: Always inline pt_regs_{read,write}_reg()
arm64: preempt: Treat should_resched() as unlikely
arm64: preempt: Simplify and optimize __preempt_count_dec_and_test()
arm64: cmpxchg128: LSE: Remove redundant operands
arm64: cmpxchg: LL/SC: Avoid redundant extension
* for-next/cpu-errata:
: Various CPU errata workarounds
arm64: errata: Add NXP i.MX8QM/8QP workaround for A53 cache coherency issue
arm64: cpufeature: Add ARM64_CPUCAP_SOC_ERRATUM for SoC-level errata
* for-next/mpam: (30 commits)
: MPAM updates
arm_mpam: make the mon_sel guard conditional only
arm_mpam: On saving mbwu state read the monitors before clearing CFG_MBWU_CTL
arm_mpam: detect and enable MPAM-Fb PCC support
arm_mpam: change MPAM-Fb error IRQ to use a threaded IRQ handler
arm_mpam: add MPAM-Fb MSC firmware access support
arm_mpam: prepare mon_sel locking for MPAM-Fb
arm_mpam: propagate MSC access errors in mpam_reset_class_locked()
arm_mpam: propagate MSC access errors for interrupt control
arm_mpam: propagate MSC access errors for mpam_reprogram_ris_partid()
arm_mpam: propagate MSC access errors for state saving function
arm_mpam: propagate MSC access errors for __ris_msmon_read()
arm_mpam: propagate MSC access errors for msmon helpers
arm_mpam: propagate MSC access errors for MBWU counters
arm_mpam: propagate MSC access errors for hw_probe functions
arm_mpam: let low level MSC accessors return an error
MAINTAINERS: Use linux-arm-kernel list for the MPAM driver
MAINTAINERS: Add mpam.rst to the MPAM DRIVER entry
arm_mpam: resctrl: Don't stop early when tearing down a class
arm_mpam: Switch to kvzmalloc_objs() for allocation of component cfg
arm_mpam: Don't loop forever if there is the maximum possible amount of PARTIDs
...
* for-next/kselftest:
: arm64 kselftest updates
kselftest/arm64: signal: copy each test binary only once
kselftest/arm64/mte: Use MTE safe memory accessors
kselftest/arm64/mte: Introduce MTE safe memory accessors
kselftest/arm64: Extend gcs-locking to cover cases where enable is unlocked
* for-next/efi:
: arm64 EFI updates/fixes
arm64/mm: Treat faults taken by EFI runtime firmware as kernel faults
arm64/efi: Clear PTE_WRITE when applying EFI_MEMORY_RO
* for-next/secondary-boot-prep:
: cpuinfo cleanup in preparation for parallel secondary bring-up
arm64: smp: Defer RCU registration during secondary CPU bringup
arm64: smp: Don't bother printing the I-cache policy for each CPU
arm64: smp: Defer update of secondary CPU capabilities
arm64: smp: Tidy up cpuinfo init and cpufeature updates
arm64: smp: Tidy up smp_prepare_cpus()
arm64: cpufeature: Store every ID register with its overrides applied
arm64: cpufeature: Read MPAMIDR_EL1 in __cpuinfo_store_cpu()
arm64: cpufeature: Add read_cpuid_with_overrides()
arm64: cpufeature: Check arm64_ftr_regs[] before the first store
* for-next/sysreg:
: sysreg updates
perf: arm_brbe: Remove the check for TME transaction states
arm64: sysreg: Mark fields the architecture dropped as RES0
arm64: sysreg: Fix the names of two misspelt fields
arm64: sysreg: Fix three incorrect enum values
* for-next/gcs:
: GCS updates/fixes
kselftest/arm64: Check for -EPERM not -EBUSY in the locking test
arm64: gcs: Return -EPERM not -EBUSY for prctl() locking failures
|
|
If a single PTE cannot be stored in the current bitmap that should be a
failure and reported back to userspace. Allowing a PTE to be split creates
a possibility that dirty memory will be reported as clean.
The gather must also cover the entire PTE that is being cleaned, not just
the part that fits within the bitmap.
Fix both issues by computing the full IOVA range of every PTE and using
that to update the bitmap and the gather. The bitmap code now rejects
ranges that don't fit in it, and the gather always combined ranges.
The selftest quietly relied on a bug where the CONT was split, that is why
the half page size for the mock domain saw alternating dirties. Increase
the dirty test page size to PAGE_SIZE so it still works as intended.
Cc: stable@vger.kernel.org
Fixes: 4a00f9434891 ("iommupt: Add read_and_clear_dirty op")
Reported-by: Samarth Narang <samarth.colleges@gmail.com>
Closes: https://lore.kernel.org/all/20261013165814.309255
Link: https://patch.msgid.link/r/0-v1-5bb926ff73ec+576-iommufd_dirty_jgg@nvidia.com
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
Generally userspace is going to step the dirty bitmap through the IOVA
space. However if the bitmap ends in the middle of a large PTE this is
problematic. The first step will clear the dirty bit and report dirty to
the end of its bitmap while the second step will see the clear'd dirty and
report clean for the start of the bitmap. If it happens this ends up
losing dirties.
Further this function is missing bounds checking so if this does happen
ther is a good chance it will write memory beyond what the userspace was
expecting.
Return an error code if this illegal condition happens.
Fixes: 58ccf0190d19 ("vfio: Add an IOVA bitmap support")
Cc: stable@vger.kernel.org
Assisted-by: LLM
Signed-off-by: Andrea Parri <parri.andrea@gmail.com>
Link: https://patch.msgid.link/20260925161142.189813-2-parri.andrea@gmail.com
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
SDMA can randomly hang on Polaris 10. Hard to reproduce,
but has been occurring sporadically in the Mesa CI
on the RADV Polaris 10 job even when Mesa isn't using
the SDMA engine.
When Mesa actually uses the SDMA on Polaris, the hang
becomes more frequent and the reproduction rate goes up
significantly in the CI system.
We don't know why the hangs occur, but it seems
that inserting a single NOP packet before each
POLL_REGMEM and FENCE packet works around it.
At the moment it's unclear which GPUs are affected,
so let's apply this to SDMA 3.x for now.
Note that this issue is not documented or acknowledged by AMD.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit c343a27ed90823d08de125f7c447d96d056b75d9)
Cc: stable@vger.kernel.org
|
|
amdgpu_gem_object_close() detaches the user queue eviction fence from
the BO before it checks bo_va->ref_count, so closing any handle of a BO
removes the fence, even when the file still has another handle to the
BO and the bo_va with its GPU mappings stays. The BO can then be evicted
without suspending the user queues of the file, which keep accessing it
through mappings that are only marked for update.
Seen on navi31 with a SDMA user queue: a BO opened twice in the same
file (GEM_FLINK + GEM_OPEN) loses its eviction fence when the second
handle is closed (traced attach/detach of the fence):
attach bo=ffff8c78a6c84c00
GEM_CLOSE handle=7
detach bo=ffff8c78a6c84c00
while handle 6 and the mapping are still in use.
Only detach the fence when the last reference to the bo_va goes away,
matching amdgpu_gem_object_open(), which attaches it only when the
bo_va is created. With this the fence stays until the file is closed.
Fixes: 31f7efcdca4d ("drm/amdgpu: enable eviction fence")
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 8387982f03d4026b7dc11b0b5e3118298eee4d6a)
Cc: stable@vger.kernel.org
|
|
allocate_usb4_bandwidth_for_stream() and allocate_usb4_bandwidth() both
use ASSERT(stream->sink) before dereferencing stream->sink->link. ASSERT
is compiled out in production builds, so a NULL stream->sink causes a
NULL pointer dereference.
link_set_dpms_off() already guards two earlier accesses to stream->sink
with explicit NULL checks, confirming that stream->sink can legitimately
be NULL. The subsequent call to deallocate_usb4_bandwidth() had no
equivalent guard.
Replace the ASSERT() calls with proper NULL checks that return false
early, and add a matching stream->sink guard at the call site in
link_set_dpms_off() consistent with the surrounding pattern.
v2: Keep the existing ASSERT() calls for diagnostics (to report
unexpected NULL sink), and add explicit NULL checks that return false
before dereferencing stream->sink. Also add a matching stream->sink
guard at the deallocation call site in link_set_dpms_off(). (Alex)
Fixes: 8fa813b7fd1e ("drm/amd/display: Remove sink usage from DPMS")
Reported-by: Dan Carpenter <error27@gmail.com>
Cc: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Cc: Dominik Kaszewski <dominik.kaszewski@amd.com>
Cc: George Zhang <george.zhang@amd.com>
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Cc: Roman Li <Roman.Li@amd.com>
Cc: Ivan Lipski <ivan.lipski@amd.com>
Cc: Dan Wheeler <daniel.wheeler@amd.com>
Cc: Alex Hung <alex.hung@amd.com>
Cc: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Reviewed-by: Roman Li <roman.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 98029a0ce61feb52c563f65349dd9b25f672e355)
Cc: stable@vger.kernel.org
|
|
This was supposed to loop over the individual evicted list, as the
kernel and always_valid evicted lists were already handled in
amdgpu_vm_validate.
Without this, BOs that were created and evicted before a userq was made
would never end up in an eviction fence rearm.
Also add the missing ref to the BO before drm_exec locks and refs it.
Fixes: 4cdbba5a16aa ("drm/amdgpu: restructure VM state machine v4")
Signed-off-by: Benjamin Cheng <benjamin.cheng@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 18decfc9e8cc0fa70b6de9944cc27a37729edfbd)
Cc: stable@vger.kernel.org
|
|
The uapi documents drm_amdgpu_gem_list_handles_entry.alignment as
"physical start_addr alignment in bytes", and GEM_OP
GET_GEM_CREATE_INFO returns the same value in bytes, but
GEM_LIST_HANDLES returns tbo.page_alignment, which is in pages. For a BO
created with a 64 KiB alignment on navi31:
GET_GEM_CREATE_INFO: alignment=65536
GEM_LIST_HANDLES: alignment=16
so a user such as CRIU that recreates BOs from this list uses the wrong
alignment. Convert it to bytes.
Fixes: f9db1fc52ceb ("drm/amdgpu: Add ioctl to get all gem handles for a process")
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit aa4493ae949c13e79aba0731394dd7a7e431342b)
Cc: stable@vger.kernel.org
|
|
amdgpu_cs_parser_bos() allocates an HMM range for every userptr entry of
the BO list and looks up the user pages. Those ranges are only freed when
amdgpu_cs_parser_bos() itself fails or when amdgpu_cs_submit() checks
them under the notifier lock. If amdgpu_cs_patch_jobs(),
amdgpu_cs_vm_handling(), amdgpu_cs_sync_rings() or the early part of
amdgpu_cs_submit() fails, nothing frees them: neither
amdgpu_cs_parser_fini() nor amdgpu_bo_list_free() touch e->range.
Each such CS then leaks the range, its hmm_pfns array (8 bytes per page
of the userptr BO) and the BO reference the range holds. An IB on a ring
that patches the CS in place (e.g. VCN) with an unmapped va_start fails
in amdgpu_cs_patch_ibs() and takes this path.
Free the ranges on those error paths. Ranges amdgpu_cs_submit() already
dropped are NULL, so freeing them again is a no-op.
Fixes: fec8fdb54e8f ("drm/amdgpu: fix userptr HMM range handling v2")
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit c52a9542ed7cfd527feb938bd0accada2ac8d267)
Cc: stable@vger.kernel.org
|
|
amdgpu_debugfs_gpr_read() drops its runtime PM reference right after
reading the wave registers. When copying them to userspace fails, it
jumps to the err label, which drops the reference a second time. A read
into a bad buffer on Navi31 either steals a reference somebody else
holds (power/runtime_usage goes from 1 to 0) or underflows the count:
amdgpu 0000:03:00.0: Runtime PM usage count underflow!
The registers are already in the kernel buffer at that point, so finish
the hardware access first and copy the buffer with a single
copy_to_user() instead of the put_user() loop with its own error path.
Fixes: 9eee152aab56 ("drm/amdgpu/debugfs: fix ref count leak when pm_runtime_get_sync fails")
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 31035bbd74f7c3e10c163db875b8e67f8b50ee6a)
Cc: stable@vger.kernel.org
|
|
When amdgpu_hmm_range_alloc() fails amdgpu_gem_userptr_ioctl() returns
-ENOMEM right away and skips the cleanup every other error path goes
through. The GEM object, the bo->parent reference on the VM root and the
MMU notifier registration are all left behind.
Use release_object here, not user_pages_done, range is still NULL at
this point.
Fixes: e095b55155ef ("drm/amdgpu: use user provided hmm_range buffer in amdgpu_ttm_tt_get_user_pages")
Signed-off-by: Perry Yuan <perry.yuan@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit bbe739e407a542073ec1a7872b14843566b4919e)
Cc: stable@vger.kernel.org
|
|
add zero fan stop temperature support on smu 14.0.2
Closes: https://gitlab.freedesktop.org/drm/amd/-/work_items/5905
Signed-off-by: Kenneth Feng <kenneth.feng@amd.com>
Reviewed-by: Kevin Wang <kevin.wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit f990c94a2633d0ead477eb1b174e254a6d292e66)
Cc: stable@vger.kernel.org
|
|
Pair mutex_init() with mutex_destroy() on the GART window allocation
failure path and in amdgpu_ttm_buffer_entity_fini(). This is a no-op
without CONFIG_DEBUG_MUTEXES, but it keeps lockdep from seeing a live
mutex after the entity is freed or re-initialized.
Signed-off-by: Perry Yuan <perry.yuan@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 03ac40aeed034a0a0d5824ad26f8fbe1f6463698)
Cc: stable@vger.kernel.org
|
|
amdgpu_ttm_buffer_entity_init() checks num_gart_windows after
drm_sched_entity_init() has already attached the entity to a
scheduler run-queue. Returning -EINVAL from that check leaks the
entity.
Move the ARRAY_SIZE() check before any resources are acquired.
Fixes: 2c3725572561 ("drm/amdgpu: statically assign gart windows to ttm entities")
Signed-off-by: Perry Yuan <perry.yuan@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 4000d1cdd1eacd3422267d6c8b4f1c9333b0317e)
Cc: stable@vger.kernel.org
|
|
The PASS_MASK is for a different register not for UVD_FW_START.
It was a mistake to check this flag. It should be enough to check
that the register is non-zero.
Fixes: 889a2cfd889c ("drm/amdgpu/uvd3.1: Don't validate the firmware when already validated")
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Link: https://patch.msgid.link/20261006211607.75407-1-timur.kristof@gmail.com
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 9dee462c83259ca769aa1bf82228adee9831685c)
Cc: stable@vger.kernel.org
|
|
In roccat_release(), access the device using file-specific
reader data rather than global array, thus preventing the
case when original device was disconnected and a new device
has connected using the same slot indexed by minor number.
Reported-by: Sashiko <sashiko-bot@kernel.org>
Link: https://sashiko.dev/#/patchset/20260902094551.200587-1-dmantipov@yandex.ru?part=2
Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru>
Signed-off-by: Benjamin Tissoires <bentiss@kernel.org>
|
|
Cross-merge networking fixes after downstream PR (net-7.3-rc8).
No conflicts. Adjacent changes:
drivers/net/ethernet/microchip/lan865x/lan865x.c
2a3f6d2a9ac3 ("microchip: lan865x: move setting of netops to fix crash")
15358c2c3dca ("net: ethernet: oa_tc6: return ERR_PTR from oa_tc6_init()")
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The existing kerneldoc documentation for URB completion callbacks is
incorrect about the context in which the callbacks occur; it is not
always in_interrupt. State explicitly that a callback may occur in
any context with no guarantees, and likewise for the status of
CPU-local interrupts.
Documentation update only; no code changes.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Link: https://patch.msgid.link/de582feb-a0aa-4fc8-9818-890014de32a9@rowland.harvard.edu
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The raw event callback runs before HID core validates and zero-pads the
report. It passes the buffer to a helper which inspects data[0] and, for a
special report, reads the complete struct savu_mouse_report_special. A
truncated report can therefore cause an out-of-bounds read.
Require enough data for the special report before calling the helper.
Commit 47669bec44fe ("HID: asus: refactor the two workqueues and init
sequence") added the same kind of raw-event length validation to hid-asus.
Fixes: 6a2a6390cf09 ("HID: roccat: add support for Roccat Savu")
Cc: stable@vger.kernel.org
Signed-off-by: Jiale Yao <yaojiale02@163.com>
Signed-off-by: Benjamin Tissoires <bentiss@kernel.org>
|
|
The raw event callback runs before HID core validates and zero-pads the
report. It reads the report number without checking the received length
and passes matching data to roccat_report_event(), which copies a complete
struct ryos_report_special. A truncated report can therefore cause an
out-of-bounds read and expose adjacent data through the Roccat character
device.
Require a complete special report before inspecting or forwarding it.
Commit 47669bec44fe ("HID: asus: refactor the two workqueues and init
sequence") added the same kind of raw-event length validation to hid-asus.
Fixes: 6f3a19360545 ("HID: roccat: add support for Ryos MK keyboards")
Cc: stable@vger.kernel.org
Signed-off-by: Jiale Yao <yaojiale02@163.com>
Signed-off-by: Benjamin Tissoires <bentiss@kernel.org>
|
|
The raw event callback runs before HID core validates and zero-pads the
report. It passes the buffer to helpers which inspect data[0] and read a
complete struct pyra_mouse_event_button for button reports. A truncated
report can therefore cause an out-of-bounds read.
Require enough data for the button report before calling either helper.
Commit 47669bec44fe ("HID: asus: refactor the two workqueues and init
sequence") added the same kind of raw-event length validation to hid-asus.
Fixes: cb7cf3da0daa ("HID: roccat: add driver for Roccat Pyra mouse")
Cc: stable@vger.kernel.org
Signed-off-by: Jiale Yao <yaojiale02@163.com>
Signed-off-by: Benjamin Tissoires <bentiss@kernel.org>
|
|
The raw event callback runs before HID core validates and zero-pads the
report. It passes the buffer to helpers which inspect data[0] and read a
complete struct kovaplus_mouse_report_button for button reports. A
truncated report can therefore cause an out-of-bounds read.
Require enough data for the button report before calling either helper.
Commit 47669bec44fe ("HID: asus: refactor the two workqueues and init
sequence") added the same kind of raw-event length validation to hid-asus.
Fixes: 0e70f97f257e ("HID: roccat: Add support for Kova[+] mouse")
Cc: stable@vger.kernel.org
Signed-off-by: Jiale Yao <yaojiale02@163.com>
Signed-off-by: Benjamin Tissoires <bentiss@kernel.org>
|
|
The raw event callback runs before HID core validates and zero-pads the
report. It reads the report number without checking the received length
and passes matching data to roccat_report_event(), which copies a complete
struct konepure_mouse_report_button. A truncated report can therefore
cause an out-of-bounds read and expose adjacent data through the Roccat
character device.
Require a complete button report before inspecting or forwarding it.
Commit 47669bec44fe ("HID: asus: refactor the two workqueues and init
sequence") added the same kind of raw-event length validation to hid-asus.
Fixes: 8936aa31cd5f ("HID: roccat: add support for Roccat Kone Pure gaming mouse")
Cc: stable@vger.kernel.org
Signed-off-by: Jiale Yao <yaojiale02@163.com>
Signed-off-by: Benjamin Tissoires <bentiss@kernel.org>
|