Driver Changes:
- Wedge from the timeout handler only after releasing the queue (Rodrigo)
- Fix a NULL pointer dereference (Francois)
- Remove redundant exec_queue_suspended (Lu)
- RTP / OA whitelist fixes (Ashutosh, Gustavo, Thomas)
- Return error on non-migratable faults requiring devmem (Matt Brost)
- Skip FORCE_WC and vm_bound check for external dma-bufs (Matt Auld)
- Hold notifier lock for write on inject test path (Shuicheng)
- Drop bogus static from finish in force_invalidate (Shuicheng)
- Fix double-free of managed BO in error path (Shuicheng)
- Don't attempt to process FAST_REQ or EVENT relays (Michal)
- Fix NPD in bo_meminfo (Matthew Auld)
- Prevent invalid cursor access for purged BOs (Matthew Auld)
- Fix offset alignment for MERT WHITELST_OA_MERT_MMIO_TRG (Ashutosh)
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Thomas Hellstrom <thomas.hellstrom@linux.intel.com>
Link: https://patch.msgid.link/akZ_UbrL94G4F2iA@fedora
In ATH11K_QMI_EVENT_FW_READY, ATH11K_FLAG_REGISTERED is set
unconditionally even when ath11k_core_qmi_firmware_ready() fails.
This leaves the driver in an inconsistent state where
initialization is considered complete although the firmware ready
handling did not finish successfully. During the subsequent SSR,
the driver enters the restart path based on this incorrect state
and dereferences uninitialized srng members, resulting in a NULL
pointer dereference.
Call trace:
ath11k_hal_srng_access_begin+0xc/0x60 [ath11k] (P)
ath11k_ce_cleanup_pipes+0x17c/0x180 [ath11k]
ath11k_core_restart+0x40/0x168 [ath11k]
Fix this by:
- skipping firmware_ready if ATH11K_FLAG_REGISTERED is already set
- setting ATH11K_FLAG_REGISTERED only when firmware_ready succeeds
- setting ATH11K_FLAG_QMI_FAIL and aborting the FW_READY handling
on error
Tested-on: WCN6750 hw1.0 AHB WLAN.MSL.2.0.c2-00204-QCAMSLSWPLZ-1
Fixes: 6fe62a8cec ("wifi: ath11k: Add cold boot calibration support on WCN6750")
Signed-off-by: Gaole Zhang <gaole.zhang@oss.qualcomm.com>
Reviewed-by: Baochen Qiang <baochen.qiang@oss.qualcomm.com>
Reviewed-by: Rameshkumar Sundaram <rameshkumar.sundaram@oss.qualcomm.com>
Link: https://patch.msgid.link/20260609090609.4041009-1-gaole.zhang@oss.qualcomm.com
Signed-off-by: Jeff Johnson <jeff.johnson@oss.qualcomm.com>
The aggr_reset_state() function uses timer_delete() (non-synchronous)
for the aggregation timer before proceeding to delete TID state and
before the structure is freed by callers like aggr_module_destroy().
If the timer callback (aggr_timeout) is executing when aggr_reset_state()
is called, the callback will continue to access aggr_conn fields like
rx_tid[] and stat[] which may be freed immediately after by
kfree(aggr_info->aggr_conn) in aggr_module_destroy().
Additionally, the timer callback can re-arm itself via mod_timer() while
aggr_reset_state() is running, creating a more complex race condition.
Use timer_delete_sync() instead to ensure any running timer callback
has completed before returning.
Fixes: bdcd817079 ("Add ath6kl cleaned up driver")
Cc: stable@vger.kernel.org
Signed-off-by: Daniel Hodges <git@danielhodges.dev>
Reviewed-by: Vasanthakumar Thiagarajan <vasanthakumar.thiagarajan@oss.qualcomm.com>
Link: https://patch.msgid.link/20260206185207.30098-1-git@danielhodges.dev
Signed-off-by: Jeff Johnson <jeff.johnson@oss.qualcomm.com>
ath9k_hif_request_firmware() re-arms an asynchronous firmware load via
request_firmware_nowait(), passing hif_dev as the completion context, and
then still dereferences hif_dev:
dev_info(&hif_dev->udev->dev, "ath9k_htc: Firmware %s requested\n",
hif_dev->fw_name);
The re-armed callback ath9k_hif_usb_firmware_cb() runs on the "events"
workqueue and, when the firmware is missing, walks the retry chain into
ath9k_hif_usb_firmware_fail() -> complete_all(&hif_dev->fw_done). That
releases the wait_for_completion(&hif_dev->fw_done) in a concurrent
ath9k_hif_usb_disconnect(), which then kfree()s hif_dev. The trailing
dev_info() in the frame that re-armed the request can therefore read freed
memory (hif_dev->udev, the first field of struct hif_device_usb):
BUG: KASAN: slab-use-after-free in ath9k_hif_request_firmware
Read of size 8 ... by task kworker/...
ath9k_hif_request_firmware
ath9k_hif_usb_firmware_cb drivers/net/wireless/ath/ath9k/hif_usb.c:1247
request_firmware_work_func
Allocated by ...:
ath9k_hif_usb_probe drivers/net/wireless/ath/ath9k/hif_usb.c
Freed by ...:
ath9k_hif_usb_disconnect -> kfree drivers/net/wireless/ath/ath9k/hif_usb.c
The fw_done barrier only makes disconnect wait for the firmware chain to
*terminate*; it does not protect the outer ath9k_hif_request_firmware()
frame that re-armed the request and keeps touching hif_dev afterwards.
Drop the post-request dev_info(): it is the only use of hif_dev after the
async request is armed, and it is purely informational (the dev_err() on the
failure path runs only when request_firmware_nowait() did not arm a callback,
so hif_dev is still alive there).
This was first reported by syzbot as a single, non-reproduced crash that was
later auto-obsoleted, and was independently rediscovered by the reFuzz fuzzer,
which produced a C reproducer (USB-gadget connect/disconnect of an ath9k_htc
device whose firmware download fails). The vulnerable code is unchanged and
still present in v7.1-rc6, where the slab-use-after-free reproduces under KASAN
once the (sub-microsecond) race window is widened.
Fixes: e904cf6fe2 ("ath9k_htc: introduce support for different fw versions")
Reported-by: syzbot+50122cbc2874b1eb25b0@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=50122cbc2874b1eb25b0
Signed-off-by: Cheng Yongkang <teel4res@gmail.com>
Acked-by: Toke Høiland-Jørgensen <toke@toke.dk>
Link: https://patch.msgid.link/20260605153210.20471-1-1020691186@qq.com
Signed-off-by: Jeff Johnson <jeff.johnson@oss.qualcomm.com>
The commit cited below should not have been merged. It attemted to fix an
existing problem ansd thereby introduced new problems by keeping the
pi_state in state Q_REQUEUE_PI_IN_PROGRESS and leaking it.
Based on the commit description the intention was to handle the case
when task_blocks_on_rt_mutex() returns -EDEADLK and the following
remove_waiter() dereferences the NULL pointer in waiter->task.
That is already handled by Davidlohr in commit 40a25d59e8
("locking/rtmutex: Skip remove_waiter() when waiter is not enqueued") and
requires no further acting.
Revert the commit breaking the "waiter == owner" case again.
Fixes: 74e144274a ("futex/requeue: Prevent NULL pointer dereference in remove_waiter() on self-deadlock")
Reported-by: Michael Bommarito <michael.bommarito@gmail.com>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Cc: stable@vger.kernel.org
Link: https://patch.msgid.link/20260701131150.0Ijhq4Dw@linutronix.de
Closes: https://lore.kernel.org/all/20260629020049.2082397-1-michael.bommarito@gmail.com
The batadv_skb_set_priority() receives an SKB with the inner ethernet
header at position "offset". When it tries to extract the IPv4 and IPv6
header, it needs to skip the ethernet header to get access to the IP
header.
But for VLAN header, it performs the access with the struct vlan_ethhdr.
This struct contains both both the ethernet header and the VLAN header. It
is therefore incorrect to skip over the whole vlan_ethhdr size to get
access to the vlan_ethhdr.
Cc: stable@vger.kernel.org
Fixes: c54f38c9aa ("batman-adv: set skb priority according to content")
Signed-off-by: Sven Eckelmann <sven@narfation.org>
scu_supply is enabled alongside scu and scu_x2 during normal SRC
operation, but rsnd_src_suspend() and rsnd_src_resume() only disable
and re-enable scu and scu_x2. The supply clock is left enabled across
a system suspend and its prepare/enable refcount becomes unbalanced
after a suspend/resume cycle.
Disable scu_supply in rsnd_src_suspend() and re-enable it in
rsnd_src_resume() so the SRC clocks are managed consistently across
system PM transitions.
Fixes: ef19ecf042 ("ASoC: rsnd: Add system suspend/resume support")
Signed-off-by: John Madieu <john.madieu.xa@bp.renesas.com>
Acked-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://patch.msgid.link/20260630175329.4145703-1-john.madieu.xa@bp.renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
siw_create_qp() currently calls siw_qp_add() before the queues, CQ
pointers, state, completion, and device list entry are ready. A QPN
lookup can therefore reach a QP that is still being constructed.
Move siw_qp_add() to the end of siw_create_qp(), after QP
initialization and before adding the QP to the siw device list.
Fixes: f29dd55b02 ("rdma/siw: queue pair methods")
Link: https://patch.msgid.link/r/20260630060040.966461-1-ruoyuw560@gmail.com
Suggested-by: Bernard Metzler <bernard.metzler@linux.dev>
Signed-off-by: Ruoyu Wang <ruoyuw560@gmail.com>
Acked-by: Bernard Metzler <bernard.metzler@linux.dev>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
When vm_insert_pages() fails, the error path calls vma->vm_ops->close(vma)
which internally calls drm_gem_vm_close() → drm_gem_object_put(),
releasing the GEM object reference acquired at the start of the function.
However, the close_vma label then falls through to put_obj, which calls
drm_gem_object_put() a second time on the same object.
If the first put releases the last reference, the object is freed and the
second put accesses freed memory, causing a use-after-free.
Fix by returning directly from close_vma instead of falling through to
put_obj, since the close handler already performs all necessary cleanup
including the object put.
Cc: stable@vger.kernel.org
Fixes: e486147c91 ("accel/amdxdna: Add BO import and export")
Signed-off-by: Wentao Liang <vulab@iscas.ac.cn>
Reviewed-by: Lizhi Hou <lizhi.hou@amd.com>
Signed-off-by: Lizhi Hou <lizhi.hou@amd.com>
Link: https://patch.msgid.link/20260625113239.49764-1-vulab@iscas.ac.cn
In hns_roce_cleanup_mhop_hem_table(), the expression:
obj = i * buf_chunk_size / table->obj_size;
is evaluated using 32-bit unsigned arithmetic because
'buf_chunk_size' is u32 and the usual arithmetic conversions convert
'i' to unsigned int. The result is assigned to a u64 variable, but the
multiplication may overflow before the assignment.
For sufficiently large HEM tables, this produces an incorrect object
index passed to hns_roce_table_mhop_put().
Cast 'i' to u64 before the multiplication so that the intermediate
calculation is performed with 64-bit arithmetic.
Found by Linux Verification Center (linuxtesting.org) with SVACE.
Fixes: a25d13cbe8 ("RDMA/hns: Add the interfaces to support multi hop addressing for the contexts in hip08")
Link: https://patch.msgid.link/r/20260627095951.51378-1-listdansp@mail.ru
Signed-off-by: Danila Chernetsov <listdansp@mail.ru>
Reviewed-by: Junxian Huang <huangjunxian6@hisilicon.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Previously, the user QP creation path would only attempt to
populate iwqp->iwpbl if the user-provided req.user_wqe_bufs
field was non-zero. The problem is that iwqp->iwpbl is
unconditionally dereferenced later on in irdma_setup_virt_qp.
While there was a check for iwqp->iwpbl != NULL, this check
would only occur if req.user_wqe_bufs was non-zero. The end
result is that a user could send a zero user_wqe_bufs value
and trigger a null ptr deref.
Fix this by unconditionally calling irdma_get_pbl and bailing
if it fails, similar to the CQ and SRQ paths.
Fixes: b48c24c2d7 ("RDMA/irdma: Implement device supported verb APIs")
Link: https://patch.msgid.link/r/20260617164013.280790-1-jmoroni@google.com
Signed-off-by: Jacob Moroni <jmoroni@google.com>
Reviewed-by: David Hu <xuehaohu@google.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Pull networking fixes from Paolo Abeni:
"Including fixes from netfilter and batman-adv.
Current release - new code bugs:
- netfilter: cthelper: cap to maximum number of expectation per master
Previous releases - regressions:
- netpoll: fix a use-after-free on shutdown path
- tcp: restore RCU grace period in tcp_ao_destroy_sock
- ipv6: fix NULL deref in fib6_walk_continiue() on multi-batch dump
- batman-adv: dat: ensure accessible eth_hdr proto field
- eth:
- virtio_net: disable cb when NAPI is busy-polled
- lan743x: Initialize eth_syslock spinlock before use
Previous releases - always broken:
- netfilter:
- nft_set_pipapo: don't leak bad clone into future transaction
- sched:
- sch_teql: Introduce slaves_lock to avoid race condition and UAF
- replace direct dequeue call with peek and qdisc_dequeue_peeked
- sctp: add INIT verification after cookie unpacking
- tipc: fix out-of-bounds read in broadcast Gap ACK blocks
- seg6: validate SRH length before reading fixed fields
- eth:
- mlx5e: fix use-after-free of metadata_dst on RX SC delete
- enetc: check the number of BDs needed for xdp_frame
- fbnic: don't cache shinfo across skb realloc"
* tag 'net-7.2-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (58 commits)
net/mlx5: HWS, fix matcher leak on resize target setup failure
net/sched: hhf: clear heavy-hitter state on reset
net/sched: dualpi2: clear stale classification on filter miss
net/sched: act_bpf: use rcu_dereference_bh() to read the filter
selftests: drv-net: tso: don't touch dangerous feature bits
cxgb4: Fix decode strings dump for T6 adapters
virtio_net: disable cb when NAPI is busy-polled
sctp: fix addr_wq_timer race in sctp_free_addr_wq()
selftests: net: bump default cmd() timeout to 20 seconds
bridge: stp: Fix a potential use-after-free when deleting a bridge
net/sched: sch_teql: Introduce slaves_lock to avoid race condition and UAF
net: gianfar: dispose irq mappings on probe failure and device removal
net: lan743x: Initialize eth_syslock spinlock before use
net: libwx: fix VMDQ mask for 1-queue mode
net: airoha: fix max receive size configuration
fsl/fman: Free init resources on KeyGen failure in fman_init()
netfilter: nftables: restrict checkum update offset
netfilter: nftables: restrict linklayer and network header writes
netfilter: nfnetlink_queue: restrict writes to network header
netfilter: nft_fib: reject fib expression on the netdev egress hook
...
Pull MFD fix from Lee Jones:
- Add MFD mailing list to MAINTAINERS
* tag 'mfd-fixes-7.2' of git://git.kernel.org/pub/scm/linux/kernel/git/lee/mfd:
MAINTAINERS: Add a mailing list entry to MFD
When data buffers are reused, e.g. for interval sample records, the
first record determines the data length, and the size of the buffer for
user copy. Current monwriter code does not check if the data length was
changed for subsequent records, which also would never happen for valid
user programs.
However, a malicious user could change the data length, resulting in out
of bounds user copy to the kernel buffer, and memory corruption. By
default, the monwriter misc device is created with root-only permissions,
so practical impact is typically low.
Fix this by checking for changed data length and rejecting such records.
Cc: stable@vger.kernel.org
Signed-off-by: Gerald Schaefer <gerald.schaefer@linux.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Every platform driver can be forced to match a device that doesn't match
its list of device IDs because of device_match_driver_override(), so
platform drivers that rely on the existence of a device's ACPI companion
object need to verify its presence.
Accordingly, add a requisite ACPI_COMPANION() check against NULL to the
ni903x_wdt watchdog driver.
Fixes: d37ec2fbab ("watchdog: ni903x_wdt: Convert to a platform driver")
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/2280455.irdbgypaU6@rafael.j.wysocki
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
The s32g_wdt driver uses two incorrect constants in the options field
of its watchdog_info struct. This bit mask should contain WDIOF_*
constants, but the driver uses two WDIOC_* ioctl constants (in addition
to correct WDIOF_* constants). This causes many incorrect bits to be
set in the bit mask. The functionality indicated by these ioctl
constants is supported by all drivers using the watchdog framework, so
this patch simply removes them.
Fixes: bd3f54ec55 ("watchdog: Add the Watchdog Timer for the NXP S32 platform")
Cc: stable@vger.kernel.org # 6.18+
Signed-off-by: Ethan Nelson-Moore <enelsonmoore@gmail.com>
Acked-by: Daniel Lezcano <daniel.lezcano@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20260505024409.60301-1-enelsonmoore@gmail.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
When a QP/CQ/SRQ is created, a two step process is used
where the buffer is allocated in userspace and explicitly
registered with the normal reg_mr mechanism prior to creating
the actual QP/CQ/SRQ object.
These special registrations are indicated via an ABI field
so the driver knows that they do not have a valid mkey and
to skip the actual CQP command submission.
Since these are real MR objects from the core's perspective,
it is possible for a user application to invoke rereg_mr on them
and cause a real CQP op to be emitted with the zero-initialized
mkey value of 0.
Fix this by preventing rereg_mr on these special regions.
Fixes: 5ac388db27 ("RDMA/irdma: Add support to re-register a memory region")
Signed-off-by: Jacob Moroni <jmoroni@google.com>
Reviewed-by: David Hu <xuehaohu@google.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
As documented in commit 8bd67ebb50 ("net: bridge: xmit: make sure we have
at least eth header len bytes"), it is possible by for a local user with
eBPF TC hook access to attach a tc filter which truncates the packet and
redirects to an batadv interface. But the code assumes that at least
ETH_HLEN bytes are available and thus might read outside of the available
buffer.
The batadv_interface_tx() must therefore always check itself if enough data
is available for the ethernet header and don't rely on min_header_len.
Cc: stable@vger.kernel.org
Fixes: c6c8fea297 ("net: Add batman-adv meshing protocol")
Reported-by: Sashiko <sashiko-bot@kernel.org>
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Add coverage for entry uprobes installed on CALL instructions while user
shadow stack is enabled. The test puts an entry uprobe on a helper whose
first instruction is a relative CALL, then verifies that the call/return
sequence completes without SIGSEGV.
This catches regressions where x86 uprobe CALL emulation updates the
regular user stack but leaves the CET shadow stack stale.
Signed-off-by: David Windsor <dwindsor@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/b957039191118c5eba97d01d80c494b859f115a6.1782777969.git.dwindsor@gmail.com
perf_event_remove_on_exec() removes events by calling
perf_event_exit_event(). For top-level events, this removes the event from
the context with DETACH_EXIT only.
This can leave inconsistent group state when a removed event is a group
leader and the group contains siblings without remove_on_exec. If the group
was active, the surviving siblings can remain active and attached to the
removed leader's sibling list, but are no longer represented by a valid
group leader on the PMU context active lists.
A later close of the removed leader uses DETACH_GROUP and can promote the
still-active siblings from this stale group state. The next schedule-in can
then add an already-linked active_list entry again, corrupting the PMU
context active list.
With DEBUG_LIST enabled, this is caught as a list_add double-add in
merge_sched_in().
Fix this by detaching group relationships when remove_on_exec removes an
event. This preserves the existing task-exit and revoke behavior, while
ensuring surviving siblings are ungrouped before the removed event leaves
the context.
Fixes: 2e498d0a74 ("perf: Add support for event removal on exec")
Signed-off-by: Taeyang Lee <0wn@theori.io>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/ai65GgZcC0LAlWLG@Taeyangs-MacBook-Pro.local
The cited commit hardcoded the hardware address comparison len to ETH_ALEN.
This breaks IPoIB, which uses 20-byte addresses. By truncating the
memcmp, the CMA may incorrectly assume the target address is
unchanged and fails to abort the stalled connection.
Fix this by replacing ETH_ALEN with the dynamic neigh->dev->addr_len
to correctly evaluate the full address regardless of the link layer.
Fixes: 925d046e7e ("RDMA/core: Add a netevent notifier to cma")
Signed-off-by: Or Gerlitz <ogerlitz@ddn.com>
Link: https://patch.msgid.link/20260617-fix-cma-ipoib-v1-1-03f869344304@ddn.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Commit 93d7356e4b ("arm64: sysreg: Describe ID_AA64DFR2_EL1 fields")
encodes the FEAT_BWE2 value of the BWE field as '0b0002'. Binary
literals only accept the digits 0 and 1, so the intended value is 2,
i.e. 0b0010.
The macro generated by gen-sysreg.awk currently expands to
#define ID_AA64DFR2_EL1_BWE_FEAT_BWE2 UL(0b0002)
is not legal C and would fail to compile if any in-tree code referenced
it. At present no caller uses this enum value, so the kernel still
builds cleanly, but the bug is latent.
Fix the typo by using the correct binary literal 0b0010.
Cc: Bin Guo <guobin@linux.alibaba.com>
Fixes: 93d7356e4b ("arm64: sysreg: Describe ID_AA64DFR2_EL1 fields")
Signed-off-by: Jia He <justin.he@arm.com>
Reviewed-by: Mark Brown <broonie@kernel.org>
Reviewed-by: Oliver Upton <oupton@kernel.org>
Acked-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Will Deacon <will@kernel.org>
During a page table walk for binding, xe_pt_stage_bind() explicitly
skips initializing the xe_res_cursor for purged BOs, treating them
similarly to NULL VMAs by only setting the cursor size.
However, xe_pt_hugepte_possible() and xe_pt_scan_64K() did not check
if the BO was purged before attempting to walk the cursor using
xe_res_dma() and xe_res_next(). Because the cursor was left
uninitialized for purged BOs, this falls through and triggers
warnings like:
WARNING: drivers/gpu/drm/xe/xe_res_cursor.h:274 at xe_res_next
Fix this by explicitly checking if the BO is purged in both
xe_pt_hugepte_possible() and xe_pt_scan_64K(), returning early just
as we do for NULL VMAs, avoiding the invalid cursor accesses entirely.
As a precaution, also zero-initialize the cursor in xe_pt_stage_bind()
to ensure we don't pass garbage data into the page table walkers
if we ever hit a similar edge case in the future.
Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/work_items/8418
Fixes: ad9843aac9 ("drm/xe/madvise: Implement purgeable buffer object support")
Assisted-by: Copilot:gemini-3.1-pro-preview
Reported-by: Matthew Schwartz <matthew.schwartz@linux.dev>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Arvind Yadav <arvind.yadav@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Tested-by: Matthew Schwartz <matthew.schwartz@linux.dev>
Link: https://patch.msgid.link/20260625152054.450125-8-matthew.auld@intel.com
(cherry picked from commit 4c7b9c6ece32440e5a435a92076d049450cd2d2e)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
When a buffer object is purged, its ttm.resource is set to NULL via the
TTM pipeline gutting flow. However, the BO remains in the client's
object list until userspace explicitly closes the GEM handle. If memory
stats are queried during this time, accessing bo->ttm.resource->mem_type
will result in a NULL pointer dereference.
Fix this by safely skipping purged BOs in bo_meminfo, as they no longer
consume any memory.
User is getting NPD on device resume, and possible theory is that in
bo_move(), if we need to evict something to SYSTEM to save the CCS state,
but the BO is marked as dontneed, this won't trigger a move but will
nuke the pages, leaving us with a NULL bo resource. And the meminfo()
doesn't look ready to handle a NULL resource.
v2 (Sashiko):
- There could potentially be other cases where we might end up with a
NULL resource, so make this a general NULL check for now.
Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/work_items/8419
Fixes: ad9843aac9 ("drm/xe/madvise: Implement purgeable buffer object support")
Assisted-by: Copilot:gemini-3.1-pro-preview
Reported-by: Matthew Schwartz <matthew.schwartz@linux.dev>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Arvind Yadav <arvind.yadav@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Tested-by: Matthew Schwartz <matthew.schwartz@linux.dev>
Link: https://patch.msgid.link/20260625152054.450125-6-matthew.auld@intel.com
(cherry picked from commit c9a8e7daa0afe3161111e27fd92176e608c7f186)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
The error path in hw_engine_init() explicitly frees a BO allocated
with xe_managed_bo_create_pin_map() via xe_bo_unpin_map_no_vm().
Since the managed BO already has a devm cleanup action registered,
this causes a double-free when devm unwinds during probe failure.
Remove the explicit free and let devm handle it, consistent with
all other xe_managed_bo_create_pin_map() callers.
Fixes: 0e1a47fcab ("drm/xe: Add a helper for DRM device-lifetime BO create")
Assisted-by: Claude:claude-opus-4.6
Reviewed-by: Zongyao Bai <zongyao.bai@intel.com>
Link: https://patch.msgid.link/20260626210631.3887291-1-shuicheng.lin@intel.com
Signed-off-by: Shuicheng Lin <shuicheng.lin@intel.com>
(cherry picked from commit e459a3bdeb117be496d7f229e2ea1f6c9fe4080b)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
The local "finish" pointer in xe_vma_userptr_force_invalidate() is
unconditionally written before each read, so the static storage class
serves no purpose. Worse, it makes the variable a process-wide shared
slot: the function's per-VM asserts do not exclude concurrent callers
on different VMs, so two such callers can race on the slot and take
the wrong if (finish) branch.
The function is gated by CONFIG_DRM_XE_USERPTR_INVAL_INJECT
(developer/test option, default n), so production builds are
unaffected.
Drop the static.
Fixes: 18c4e53695 ("drm/xe/userptr: Convert invalidation to two-pass MMU notifier")
Assisted-by: Claude:claude-opus-4.7
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Zongyao Bai <zongyao.bai@intel.com>
Link: https://patch.msgid.link/20260625224452.3243231-1-shuicheng.lin@intel.com
Signed-off-by: Shuicheng Lin <shuicheng.lin@intel.com>
(cherry picked from commit ed382e3b07fae51a09d7290485bff0592f6b168b)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
When CONFIG_DRM_XE_USERPTR_INVAL_INJECT=y, xe_pt_svm_userptr_pre_commit()
runs vma_check_userptr() with the svm notifier_lock taken for read. The
test injection causes vma_check_userptr() to call
xe_vma_userptr_force_invalidate(), which feeds into
xe_vma_userptr_do_inval() with drm_gpusvm_ctx.in_notifier=true. That
flag tells drm_gpusvm_unmap_pages() the caller already holds
notifier_lock for write and only asserts the mode. Because the caller
actually holds it for read, the assertion fires:
WARNING: drivers/gpu/drm/drm_gpusvm.c:1669 at \
drm_gpusvm_unmap_pages+0xd4/0x130 [drm_gpusvm_helper]
Call Trace:
xe_vma_userptr_do_inval+0x40d/0xfd0 [xe]
xe_vma_userptr_invalidate_pass1+0x3e6/0x8d0 [xe]
xe_vma_userptr_force_invalidate+0xde/0x290 [xe]
vma_check_userptr.constprop.0+0x1c6/0x220 [xe]
xe_pt_svm_userptr_pre_commit+0x6a3/0xc60 [xe]
...
xe_vm_bind_ioctl+0x3a0a/0x4480 [xe]
Acquire notifier_lock for write in pre-commit when the inject Kconfig
is enabled, via new helpers xe_pt_svm_userptr_notifier_lock()/_unlock().
Rename xe_svm_assert_held_read() to
xe_svm_assert_held_read_or_inject_write() so it asserts the correct
mode under each build configuration. Production builds
(CONFIG_DRM_XE_USERPTR_INVAL_INJECT=n) keep the existing read-mode
behavior bit-for-bit.
Fixes: 9e97874148 ("drm/xe/userptr: replace xe_hmm with gpusvm")
Assisted-by: Claude:claude-opus-4.7
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Zongyao Bai <zongyao.bai@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patch.msgid.link/20260625215615.3016892-1-shuicheng.lin@intel.com
Signed-off-by: Shuicheng Lin <shuicheng.lin@intel.com>
(cherry picked from commit 80ccbd97ffee8ad2e73167d826fe7be548364365)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Currently, xe_display_bo_framebuffer_init() unconditionally attempts to
apply XE_BO_FLAG_FORCE_WC to the buffer and rejects the FB creation with
-EINVAL if the BO is already VM_BINDed.
However, for imported dma-bufs (ttm_bo_type_sg), this check doesn't seem
to make much sense since CPU caching policy is entirely controlled by
the exporter. Plus there is no place to set this flag, in the first
place. Also this is not rejected if not yet vm_binded, but that seems
arbitrary since setting or not setting FORCE_WC should a noop either
way, at this stage, and whether it is currently VM_BINDed makes no
difference.
Currently if we run an app and offload rendering to an external dGPU,
like NV or another xe device, the dma-buf passed back to the compositor
(igpu) will be an actual external import from xe pov, and it will be
missing FORCE_WC, and if the compositor side did a VM_BIND before
turning into it into an fb the whole thing gets rejected.
So it looks like we either need to reject outright, no matter what, or
this usecase is valid and we need to loosen the restriction for sg
buffers. Proposing here to loosen the restriction.
Assisted-by: Gemini:gemini-3.1-pro-preview
Link: https://gitlab.freedesktop.org/drm/xe/kernel/-/work_items/7919
Fixes: 44e694958b ("drm/xe/display: Implement display support")
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Maarten Lankhorst <dev@lankhorst.se>
Cc: <stable@vger.kernel.org> # v6.12+
Reviewed-by: Maarten Lankhorst <dev@lankhorst.se>
Link: https://patch.msgid.link/20260612170501.550816-2-matthew.auld@intel.com
(cherry picked from commit 3e493f88c84088ccd7b53cdd23ac5c875c9a60dd)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Whitelist OA registers on stream open and de-whitelist on stream
close/release. Whitelisting is only done when 'stream->sample' is
true. 'stream->sample' is only true when (a) xe_observation_paranoid is set
to false by system admin, or (b) the process is perfmon_capable(). This
therefore enforces the OA register whitelisting security requirements.
Fixes: 828a8eaf37 ("drm/xe/oa: Add MMIO trigger support")
Cc: stable@vger.kernel.org # v6.12+
Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Link: https://patch.msgid.link/20260615224227.34880-9-ashutosh.dixit@intel.com
(cherry picked from commit f8e6874f46f19a6a2a0f24a81689f90641bb402a)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Dave Airlie
Gaole Zhang
Daniel Hodges
Cheng Yongkang
Yang Xiuwei
Sebastian Andrzej Siewior
Sven Eckelmann
John Madieu
Ruoyu Wang
Wentao Liang
Danila Chernetsov
Chenguang Zhao
Joshua Crofts
Jacob Moroni
Linus Torvalds
Gerald Schaefer
David Howells
Manuel Ebner
Rafael J. Wysocki
Ethan Nelson-Moore
Yousef Alhouseen
Jiri Olsa
David Windsor
Taeyang Lee
Or Gerlitz
Wang Yan
Jia He
Ashutosh Dixit
Matthew Auld
Michal Wajdeczko
Shuicheng Lin
Matthew Brost