skb->len may be bigger than 65535 in UDP-based tunnels that have BIG TCP
enabled. If GSO aggregates packets that large, set the length in the UDP
header to 0, so that tcpdump can print such packets properly (treating
them as RFC 2675 jumbograms). Later in the pipeline, __udp_gso_segment
will set uh->len to the size of individual packets.
Signed-off-by: Alice Mikityanska <alice@isovalent.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20260710134242.216538-7-alice.kernel@fastmail.im
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
In the previous commit we started using uh->len = 0 as a marker of a GRO
packet bigger than 65536 bytes. Filter out malformed packets coming from
the wire with len=0 at udp_gro_receive to exclude them from GRO.
Note that a similar check was present in udp_gro_receive_segment, but
not in the UDP socket gro_receive flow. By adding an early check to
udp_gro_receive, the check in udp_gro_receive_segment can be dropped.
Signed-off-by: Alice Mikityanska <alice@isovalent.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20260710134242.216538-6-alice.kernel@fastmail.im
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Currently, gro_max_size and gro_ipv4_max_size can be set to values
bigger than 65536, and GRO will happily aggregate UDP to the configured
size (for example, with TCP traffic in VXLAN tunnels). However,
udp_gro_complete uses the 16-bit length field in the UDP header to store
the length of the aggregated packet. It leads to the packet truncation
later in udp_rcv.
Fix this by storing 0 to the UDP length field and by restoring the real
length from skb->len in udp_rcv. IP GRO already can store 0 to the IP
length field, and iph_totlen()/ipv6_payload_len() are capable of
restoring the real length, because the relevant packets (BIG TCP
tunneled in UDP tunnels) will have skb_is_gso_tcp == true.
Additionally, restrict handling uh->len=0 in udpv6_rcv to BIG TCP and
jumbograms only by using the udp_get_len helper.
Signed-off-by: Alice Mikityanska <alice@isovalent.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20260710134242.216538-5-alice.kernel@fastmail.im
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Wherever a GSO packet can occur, and its length is used to fill the UDP
header, use udp_set_len that assigns 0 if the length doesn't fit 16
bits, so that the packet can be properly parsed and segmented later,
instead of having truncated length.
Use udp_get_len in udp_validate_len to treat BIG TCP packets as valid.
Signed-off-by: Alice Mikityanska <alice@isovalent.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20260710134242.216538-4-alice.kernel@fastmail.im
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
skb_segment is called for partial GSO, when netif_needs_gso returns true
in validate_xmit_skb. Partial GSO is needed, for example, when
segmentation of tunneled traffic is offloaded to a NIC that only
supports inner checksum offload.
Currently, skb_segment clamps the segment length to 65534 bytes, because
gso_size == 65535 is a special value GSO_BY_FRAGS, and we don't want
to accidentally assign mss = 65535, as it would fall into the
GSO_BY_FRAGS check further in the function.
This implementation, however, artificially blocks len > 65534, which is
possible since the introduction of BIG TCP. To allow bigger lengths and
avoid resegmentation of BIG TCP packets, store the gso_by_frags flag in
the beginning and don't use a special value of mss for this purpose
after mss was modified.
Signed-off-by: Alice Mikityanska <alice@isovalent.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20260710134242.216538-3-alice.kernel@fastmail.im
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Since BIG TCP for UDP tunnels will start using len=0 in the UDP header
as an indicator of a GSO packet bigger than 65535 bytes, this commit
introduces the following getter and setters to use tree-wide, in order
to explicitly mark places where len=0 may be expected, and handle them
properly:
1. udp_set_len() sets uh->len to its real value if it's not bigger than
65535, and to 0 otherwise: to be used in GSO context with aggregated
packets.
2. udp_set_len_short() is to be used when the length is known to fit 16
bits. It WARNs when the caller tries to assign a bigger value if
CONFIG_DEBUG_NET=y.
3. udp_get_len_short() returns len in host byte order: to be used on the
RX side to deal with non-aggregated packets, or to access the raw value
of the len field.
4. udp_get_len() decodes uh->len set by udp_set_len(). It checks whether
the packet is GSO to guard from malformed packets.
At the moment udp_set_len() is not used, a following commit will start
using it after enabling len>65535 for GSO.
Raw uh->len (in network byte order) is still accessed in a few places
for checksum calculation purposes, and to decode len=0 in udpv6_rcv for
jumbograms. udp_rcv and udpv6_rcv will be addressed by the commit that
starts using udp_set_len() to set UDP len=0 for BIG TCP packets in UDP
tunnels.
Signed-off-by: Alice Mikityanska <alice@isovalent.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Acked-by: Jason A. Donenfeld <Jason@zx2c4.com>
Link: https://patch.msgid.link/20260710134242.216538-2-alice.kernel@fastmail.im
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
huge_ptep_get() can be handed a virtual address pointing to the middle
of a contpmd/contpte mapped hugetlb folio (examples of callers are
pagemap_hugetlb_range, page_mapped_in_vma).
The arm64 helper rewalks the pgtables in find_num_contig to answer
whether the huge pte we have maps a contpmd or a contpte hugetlb folio,
and returns CONT_PMDS or CONT_PTES, so that it can collect a/d bits over
the contiguous ptes. We can falsely return CONT_PTES instead of
CONT_PMDS if the addr is not aligned. On systems where CONT_PTES !=
CONT_PMDS (meaning page size is 16K), we could collect excess A/D bit
state, meaning extra work for the kernel. Even worse, we may iterate
beyond the PTE table and dereference a garbage ptep pointer to access
physical memory we don't own. Since the ptep pointer is a linear map
address, we may run off the end of the linear map or into a hole,
dereference a VA not mapped into the kernel pgtables and cause kernel
panic.
Fix this by aligning the pmdp pointer down to a contpmd base before
checking equality with the passed huge pte pointer, to correctly answer
whether the huge pte is the base of a contpmd block.
Fixes: 29cb805196 ("arm64: hugetlb: Cleanup huge_pte size discovery mechanisms")
Cc: stable@vger.kernel.org
Acked-by: David Hildenbrand (Arm) <david@kernel.org>
Signed-off-by: Dev Jain <dev.jain@arm.com>
Acked-by: Muchun Song <muchun.song@linux.dev>
Signed-off-by: Will Deacon <will@kernel.org>
Commit 922fb2db02 ("dt-bindings: display: simple: Add Kyocera
tcg070wvlq panel") added the compatible string for the Kyocera
tcg070wvlq panel to the devicetree bindings documentation
(panel-simple.yaml), but the corresponding entry was not added
to the panel-simple driver.
Add the missing compatible entry to the driver.
Signed-off-by: Heiko Schocher <hs@nabladev.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patch.msgid.link/20260721041935.757240-1-hs@nabladev.com
This HP All-in-One desktop (model 27-ca1xxx) uses an ALC274 codec
and has no sound from either the internal speakers or the headphone
jack. The audio output requires a combination of a specific verb
setup and GPIO configuration to become functional. Additionally,
the power_save feature must be disabled; otherwise, audible clicks
or pops occur during power state transitions.
Link: https://bugzilla.kernel.org/show_bug.cgi?id=220694
Signed-off-by: Zhang Heng <zhangheng@kylinos.cn>
Link: https://patch.msgid.link/20260721113539.317561-1-zhangheng@kylinos.cn
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Setting up the interface when suspended/resuming fails on this card.
Adding a reset and delay quirk will eliminate this problem.
usb 1-1: New USB device found, idVendor=0124, idProduct=0c21
usb 1-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3
usb 1-1: Product: USB Headphone
usb 1-1: Manufacturer: Generic
usb 1-1: SerialNumber: 20210726905926
Signed-off-by: Lianqin Hu <hulianqin@vivo.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Link: https://patch.msgid.link/TYUPR06MB6217CBB68C8F868C076A4353D2C12@TYUPR06MB6217.apcprd06.prod.outlook.com
Convert atmel_ac97c_probe() to the managed APIs. Replace the open-coded
platform_get_resource() + ioremap() with devm_platform_ioremap_resource(),
which requests and maps the AC97C register window in one call. Switch the
clock to devm_clk_get_enabled(), the card to snd_devm_card_new(), and the
interrupt to devm_request_irq(). The now-unnecessary error-path cleanup
and the manual teardown in atmel_ac97c_remove() are dropped, since devm
handles them.
platform_get_irq() was already used; tighten its error check to irq < 0.
Both resource and IRQ lookups are equivalent for a platform-backed
device. The AC97C register window is owned solely by this driver, so the
new region request cannot conflict with another claimant, and it is
mapped exactly once (no double mapping).
No functional change; built for ARM (allmodconfig + SND_ATMEL_AC97C)
with LLVM=1 and sound/atmel/ac97c.o compiles cleanly.
Assisted-by: opencode:hy3-free
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Link: https://patch.msgid.link/20260721225411.815553-1-rosenp@gmail.com
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Merge the introduction of acpi_dev_is_video_device() along with some
following driver updates related to it (from Andy Shevchenko).
* acpi-video:
platform/x86: thinkpad_acpi: Convert to use acpi_dev_is_video_device() helper
PCI/VGA: Convert to use acpi_dev_is_video_device() helper
i2c: acpi: Convert to use acpi_dev_is_video_device() helper
ACPI: video: Convert to use acpi_dev_is_video_device() helper
ACPI: scan: Convert to use acpi_dev_is_video_device() helper
ACPI: utils: Introduce acpi_dev_is_video_device() helper
The NPCM845 timer0 node references a single clock, but its
clock-names property is not described by the timer binding. Drop the
undocumented name so the DTS matches the binding.
Signed-off-by: Tomer Maimon <tmaimon77@gmail.com>
Link: https://patch.msgid.link/20260708165929.2233934-2-tmaimon77@gmail.com
Signed-off-by: Andrew Jeffery <andrew@codeconstruct.com.au>
* icc-misc
interconnect: qcom: add COMPILE_TEST
interconnect: qcom: simplify allocation
interconnect: debugfs-client: add NULL check for platform_device_alloc
interconnect: Fix use after free in icc_get() and of_icc_get_by_index()
Signed-off-by: Georgi Djakov <djakov@kernel.org>
Add Epoch Subsystem (EPSS) L3 scaling support on Qualcomm Shikra SoC.
EPSS hardware on Shikra is similar to other SoCs but supports only twelve
L3 frequency entries (LUT). Reading the LUT beyond supported frequencies
can expose incorrect frequencies.
Introduce new compatible to represent this constrained variant of EPSS.
* icc-shikra
dt-bindings: interconnect: qcom,osm-l3: Add EPSS L3 DT binding for Qualcomm Shikra SoC
interconnect: qcom: Add EPSS L3 scaling support for Shikra SoC
Link: https://patch.msgid.link/20260603-shikra_epss_l3-v3-0-3c2e0b796e78@oss.qualcomm.com
Signed-off-by: Georgi Djakov <djakov@kernel.org>
In of_icc_get_by_index() and icc_get(), if the dynamic allocation for
path->name fails via kasprintf(), the error handling path directly
calls kfree(path) to free the path object and returns an error.
However, prior to this point, path_find() calls path_init(), which
already links the path's requests into the req_list of the respective
interconnect nodes via hlist_add_head(). Directly invoking kfree(path)
leaves dangling pointers in the hlist. A subsequent call to icc_get()
or icc_set_bw() will traverse or modify these corrupted lists, triggering
a slab use afterfree.
KASAN report showing the vulnerability when reproducing via debugfs:
BUG: KASAN: slab-use-after-free in path_find+0x6f8/0xcfc
Write of size 8 at addr fff000000d43f748 by task sh/1
...
Call trace:
kasan_report+0xac/0xfc
path_find+0x6f8/0xcfc
icc_get+0x148/0x380
icc_get_set+0xf8/0x2d0
...
Freed by task 1:
kfree+0x1a0/0x4a4
icc_get+0x2cc/0x380
icc_get_set+0xf8/0x2d0
Fix this by replacing kfree(path) with the proper teardown function,
icc_put(path), which safely removes the requests from the req_list using
hlist_del() and drops the provider usage references before freeing the
memory.
Additionally, in icc_get(), ensure that the icc_lock mutex is released
prior to calling icc_put(path) to avoid a deadlock, as icc_put()
internally acquires the same lock.
Fixes: 3791163602 ("interconnect: Handle memory allocation errors")
Cc: stable@vger.kernel.org
Signed-off-by: Kuan-Wei Chiu <visitorckw@gmail.com>
Link: https://patch.msgid.link/20260416190840.1753468-1-visitorckw@gmail.com
Signed-off-by: Georgi Djakov <djakov@kernel.org>
scmi_protocol_device_unrequest() drops scmi_requested_devices_mtx while
notifying listeners but continues to retain the per-protocol list head.
When two SCMI drivers for the same protocol unregister concurrently, one
thread can remove the final request and free the list head while the other
is running its notifier. The latter then dereferences the freed list head
after reacquiring the mutex and can free it a second time.
Complete the list and IDR updates, including freeing an empty list head,
before dropping the mutex. Keep the blocking notifier outside the critical
section and retain only the detached request across the callback.
Fixes: d3cd7c525f ("firmware: arm_scmi: Refactor protocol device creation")
Reported-by: Sashiko <sashiko-bot@kernel.org>
Link: https://patch.msgid.link/20260722095250.2011630-1-sudeep.holla@kernel.org
Signed-off-by: Sudeep Holla <sudeep.holla@kernel.org>
Implement completion counters for the EFA device. Each completion
counter is backed by two EFA event counters, one for success
completions and one for error completions.
The driver creates umem for counters from private descriptor ioctl
attributes using core utility.
Read operations are not implemented as the counter values are accessed
directly from userspace through the mapped memory.
Reviewed-by: Yonatan Nachum <ynachum@amazon.com>
Signed-off-by: Michael Margolin <mrgolin@amazon.com>
Link: https://patch.msgid.link/20260722083603.30334-7-mrgolin@amazon.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Track completion counter objects in the resource tracking database so
they are visible through the rdma netlink interface. The rdma tool
displays the comp_cntr count in the resource summary.
Add RDMA_RESTRACK_COMP_CNTR type, embed rdma_restrack_entry in
ib_comp_cntr, and add the res_to_dev mapping. Register the resource
on create and remove it on destroy.
Reviewed-by: Yonatan Nachum <ynachum@amazon.com>
Signed-off-by: Michael Margolin <mrgolin@amazon.com>
Link: https://patch.msgid.link/20260722083603.30334-5-mrgolin@amazon.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Add a dedicated query interface for completion counter capabilities
via UVERBS_METHOD_QUERY_COMP_CNTR_CAPS on the device object.
The query returns the maximum number of counters, maximum counter
value, and a bitmask of supported QP attach operations. Each field
is an optional ioctl attribute, allowing userspace to request only
the capabilities it needs.
Drivers implement the query_comp_cntr_caps operation to report
device-specific capabilities.
Reviewed-by: Yonatan Nachum <ynachum@amazon.com>
Signed-off-by: Michael Margolin <mrgolin@amazon.com>
Link: https://patch.msgid.link/20260722083603.30334-4-mrgolin@amazon.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Reject comp_cntr destroy while it is attached to any QP. Track
attachments using an xarray in ib_qp keyed by the attach op_mask.
Use op bitmask to reject overlapping attaches early.
Reviewed-by: Yonatan Nachum <ynachum@amazon.com>
Signed-off-by: Michael Margolin <mrgolin@amazon.com>
Link: https://patch.msgid.link/20260722083603.30334-3-mrgolin@amazon.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Add core infrastructure for Completion Counters, a light-weight
alternative to polling CQ for tracking operation completions.
Define the UVERBS_OBJECT_COMP_CNTR ioctl object with create, destroy,
modify and read methods for both success and error counters. Add a QP
attach method on the QP object to associate a completion counter with a
queue pair.
Add ib_comp_cntr struct, ib_comp_cntr_attach_attr, device ops, and
DECLARE_RDMA_OBJ_SIZE for driver object allocation.
Only userspace Completion Counters are supported at this stage.
Reviewed-by: Yonatan Nachum <ynachum@amazon.com>
Signed-off-by: Michael Margolin <mrgolin@amazon.com>
Link: https://patch.msgid.link/20260722083603.30334-2-mrgolin@amazon.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
When accessing a PD via the netlink path the only synchronization
mechanism for the said PD is rdma_restrack_get().
Currently, rdma_restrack_del() is invoked at the end of
ib_dealloc_pd_user(), which is too late, since by that point
vendor-specific resources associated with the PD might already be
freed. This can leave a short window where the PD remains accessible
through restrack, leading to a potential use-after-free.
Fix this by moving the rdma_restrack_begin_del() call to the start of
ib_dealloc_pd_user(), ensuring that the PD is removed from restrack
before its internal resources are released. This guarantees that no new
users hold references to a PD that is in the process of destruction.
In addition, this change preserves the intended inverted order
between create and destroy routines: resources are added to
restrack at the end of successful creation, and hence shall be removed
from the restrack first thing during the destruction flow, which keeps
the lifecycle management consistent and predictable.
Fixes: 91a7c58fce ("RDMA: Restore ability to fail on PD deallocate")
Signed-off-by: Patrisious Haddad <phaddad@nvidia.com>
Reviewed-by: Michael Guralnik <michaelgur@nvidia.com>
Signed-off-by: Edward Srouji <edwards@nvidia.com>
Link: https://patch.msgid.link/20260713-restrack-uaf-fix-resub-v2-8-bbe8bb270d51@nvidia.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
When accessing a dmah via the netlink path the only synchronization
mechanism for the said dmah is rdma_restrack_get().
Currently, rdma_restrack_del() is invoked at the end of
uverbs_free_dmah(), which is too late, since by that point
vendor-specific resources associated with the dmah might already be
freed. This can leave a short window where the dmah remains accessible
through restrack, leading to a potential use-after-free.
Fix this by moving the rdma_restrack_begin_del() call to the start of
uverbs_free_dmah(), ensuring that the dmah is removed from restrack
before its internal resources are released. This guarantees that no new
users hold references to a dmah that is in the process of destruction.
In addition, this change preserves the intended inverted order
between create and destroy routines: resources are added to
restrack at the end of successful creation, and hence shall be removed
from the restrack first thing during the destruction flow, which keeps
the lifecycle management consistent and predictable.
Fixes: d83edab562 ("RDMA/core: Introduce a DMAH object and its alloc/free APIs")
Signed-off-by: Patrisious Haddad <phaddad@nvidia.com>
Reviewed-by: Michael Guralnik <michaelgur@nvidia.com>
Signed-off-by: Edward Srouji <edwards@nvidia.com>
Link: https://patch.msgid.link/20260713-restrack-uaf-fix-resub-v2-7-bbe8bb270d51@nvidia.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
When accessing a CQ via the netlink path the only synchronization
mechanism for the said CQ is rdma_restrack_get().
Currently, rdma_restrack_del() is invoked at the end of
ib_free_cq(), which is too late, since by that point
vendor-specific resources associated with the CQ might already be
freed. This can leave a short window where the CQ remains accessible
through restrack, leading to a potential use-after-free.
Fix this by moving the rdma_restrack_del() call to be before the freeing
of the vendor-specific resources ensuring that the CQ is removed from
restrack before its internal resources are released.
This guarantees that no new users hold references to a CQ that is in
the process of destruction.
Fixes: 43d781b9fa ("RDMA: Allow fail of destroy CQ")
Signed-off-by: Patrisious Haddad <phaddad@nvidia.com>
Reviewed-by: Michael Guralnik <michaelgur@nvidia.com>
Signed-off-by: Edward Srouji <edwards@nvidia.com>
Link: https://patch.msgid.link/20260713-restrack-uaf-fix-resub-v2-6-bbe8bb270d51@nvidia.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
When accessing a counter via the netlink path the only synchronization
mechanism for the said counter is rdma_restrack_get().
Currently, rdma_restrack_del() is invoked at the end of
counter_release(), which is too late, since by that point
vendor-specific resources associated with the counter might already be
freed. This can leave a short window where the counter remains
accessible through restrack, leading to a potential use-after-free.
Fix this by moving the rdma_restrack_del() call to be before the
freeing of the vendor-specific resources, ensuring that the counter is
removed from restrack before its internal resources are released.
This guarantees that no new users hold references to a counter that is
in the process of destruction.
Fixes: 99fa331dc8 ("RDMA/counter: Add "auto" configuration mode support")
Signed-off-by: Patrisious Haddad <phaddad@nvidia.com>
Reviewed-by: Michael Guralnik <michaelgur@nvidia.com>
Signed-off-by: Edward Srouji <edwards@nvidia.com>
Link: https://patch.msgid.link/20260713-restrack-uaf-fix-resub-v2-5-bbe8bb270d51@nvidia.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
When accessing a SRQ via the netlink path the only synchronization
mechanism for the said SRQ is rdma_restrack_get().
Currently, rdma_restrack_del() is invoked at the end of
ib_destroy_srq_user(), which is too late, since by that point
vendor-specific resources associated with the SRQ might already be
freed. This can leave a short window where the SRQ remains accessible
through restrack, leading to a potential use-after-free.
Fix this by moving the rdma_restrack_begin_del() call to the start of
ib_destroy_srq_user(), ensuring that the SRQ is removed from restrack
before its internal resources are released. This guarantees that no new
users hold references to a SRQ that is in the process of destruction.
In addition, this change preserves the intended inverted order
between create and destroy routines: resources are added to
restrack at the end of successful creation, and hence shall be removed
from the restrack first thing during the destruction flow, which keeps
the lifecycle management consistent and predictable.
Fixes: 48f8a70e89 ("RDMA/restrack: Add support to get resource tracking for SRQ")
Signed-off-by: Patrisious Haddad <phaddad@nvidia.com>
Reviewed-by: Michael Guralnik <michaelgur@nvidia.com>
Signed-off-by: Edward Srouji <edwards@nvidia.com>
Link: https://patch.msgid.link/20260713-restrack-uaf-fix-resub-v2-4-bbe8bb270d51@nvidia.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
When accessing a CQ via the netlink path the only synchronization
mechanism for the said CQ is rdma_restrack_get().
Currently, rdma_restrack_del() is invoked at the end of
ib_destroy_cq_user(), which is too late, since by that point
vendor-specific resources associated with the CQ might already be
freed. This can leave a short window where the CQ remains accessible
through restrack, leading to a potential use-after-free.
Fix this by moving the rdma_restrack_begin_del() call to the start of
ib_destroy_cq_user(), ensuring that the CQ is removed from restrack
before its internal resources are released. This guarantees that no new
users hold references to a CQ that is in the process of destruction.
In addition, this change preserves the intended inverted order
between create and destroy routines: resources are added to
restrack at the end of successful creation, and hence shall be removed
from the restrack first thing during the destruction flow, which keeps
the lifecycle management consistent and predictable.
Fixes: 08f294a152 ("RDMA/core: Add resource tracking for create and destroy CQs")
Signed-off-by: Patrisious Haddad <phaddad@nvidia.com>
Reviewed-by: Michael Guralnik <michaelgur@nvidia.com>
Signed-off-by: Edward Srouji <edwards@nvidia.com>
Link: https://patch.msgid.link/20260713-restrack-uaf-fix-resub-v2-3-bbe8bb270d51@nvidia.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
When querying a QP via the netlink flow the only synchronization
mechanism for the said QP is rdma_restrack_get(), meanwhile during the
QP destroy path rdma_restrack_del() is called at the end of the
ib_destroy_qp_user() function which is too late, since by then the
vendor specific resources for said QP would already be destroyed, and
till the rdma_restrack_del() is called this QP can still be accessed,
which could cause the use after free below.
Fix this by moving the rdma_restrack_begin_del() to the start of the
ib_destroy_qp_user(), which in turn waits for all usages of the QP to be
done then removes it from the database to prevent access to it while it
is being destroyed.
RIP: 0010:ib_query_qp+0x15/0x50 [ib_core]
Code: 48 83 05 5d 8e b9 ff 01 eb b5 66 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 c7 46 40 00 00 00 00 48 c7 46 78 00 00 00 00 <48> 8b 07 48 8b 80 88 01 00 00 48 85 c0 74 1a 48 83 05 54 91 b9 ff
RSP: 0018:ff11000108a8f2f0 EFLAGS: 00010202
RAX: 0000000000000000 RBX: ff11000108a8f370 RCX: ff11000108a8f370
RDX: 0000000000000000 RSI: ff11000108a8f3d8 RDI: 0000000000000000
RBP: ff1100010de5a000 R08: 0000000000000e80 R09: 0000000000000004
R10: ff110001057a604c R11: 0000000000000000 R12: ff11000108a8f370
R13: ff110001090e8000 R14: 0000000000000000 R15: ff110001057a602c
FS: 00007f2ffd8db6c0(0000) GS:ff110008dc90b000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000000 CR3: 000000010b9a7004 CR4: 0000000000373eb0
Call Trace:
<TASK>
mlx5_ib_gsi_query_qp+0x21/0x50 [mlx5_ib]
mlx5_ib_query_qp+0x689/0x9d0 [mlx5_ib]
ib_query_qp+0x35/0x50 [ib_core]
fill_res_qp_entry_query.isra.0+0x47/0x280 [ib_core]
? __wake_up+0x40/0x50
? netlink_broadcast_filtered+0x15a/0x550
? kobject_uevent_env+0x562/0x710
? ep_poll_callback+0x242/0x270
? __nla_put+0xc/0x20
? nla_put+0x28/0x40
? nla_put_string+0x2e/0x40 [ib_core]
fill_res_qp_entry+0x138/0x190 [ib_core]
res_get_common_dumpit+0x4a5/0x800 [ib_core]
? fill_res_qp_entry_query.isra.0+0x280/0x280 [ib_core]
nldev_res_get_qp_dumpit+0x1e/0x30 [ib_core]
netlink_dump+0x16f/0x450
__netlink_dump_start+0x1ce/0x2e0
rdma_nl_rcv_msg+0x1d3/0x330 [ib_core]
? nldev_res_get_qp_raw_dumpit+0x30/0x30 [ib_core]
rdma_nl_rcv_skb.constprop.0.isra.0+0x108/0x180 [ib_core]
rdma_nl_rcv+0x12/0x20 [ib_core]
netlink_unicast+0x255/0x380
? __alloc_skb+0xfa/0x1e0
netlink_sendmsg+0x1f3/0x420
__sock_sendmsg+0x38/0x60
____sys_sendmsg+0x1e8/0x230
? copy_msghdr_from_user+0xea/0x170
___sys_sendmsg+0x7c/0xb0
? __futex_wait+0x95/0xf0
? __futex_wake_mark+0x40/0x40
? futex_wait+0x67/0x100
? futex_wake+0xac/0x1b0
__sys_sendmsg+0x5f/0xb0
do_syscall_64+0x55/0xb90
entry_SYSCALL_64_after_hwframe+0x4b/0x53
Fixes: 514aee660d ("RDMA: Globally allocate and release QP memory")
Signed-off-by: Patrisious Haddad <phaddad@nvidia.com>
Reviewed-by: Michael Guralnik <michaelgur@nvidia.com>
Signed-off-by: Edward Srouji <edwards@nvidia.com>
Link: https://patch.msgid.link/20260713-restrack-uaf-fix-resub-v2-2-bbe8bb270d51@nvidia.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Add rdma_restrack_abort_del(), rdma_restrack_begin_del() and
rdma_restrack_commit_del() functions to allow deleting a resource from
the xarray to effectively prevent future access to it and wait for all
current users to finish while preserving its index in the xarray to
allow to re-insert it if needed with guaranteed success.
This is a preparatory change for subsequent patches in the series
which will use these functions to fix the cleanup flow.
Signed-off-by: Patrisious Haddad <phaddad@nvidia.com>
Reviewed-by: Michael Guralnik <michaelgur@nvidia.com>
Signed-off-by: Edward Srouji <edwards@nvidia.com>
Link: https://patch.msgid.link/20260713-restrack-uaf-fix-resub-v2-1-bbe8bb270d51@nvidia.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
platform_device_alloc() may return NULL when memory allocation fails.
Add proper error handling to prevent a potential NULL pointer
dereference when calling platform_device_add() with a NULL pointer.
Signed-off-by: Li Jun <lijun01@kylinos.cn>
Link: https://patch.msgid.link/20260602103120.2966458-1-lijun01@kylinos.cn
Signed-off-by: Georgi Djakov <djakov@kernel.org>
mana_table_store_ud_qp() publishes a QP at its send-queue id before
inserting the receive-queue id, dropping the XArray lock between the two
xa_insert_irq() calls. A concurrent completion handler can look up the QP
and take a transient reference. When the second insertion fails, the
rollback erased only the send-queue entry and returned, leaving both the
initial table reference and the transient reference outstanding while RDMA
core frees the QP, causing a use-after-free.
Drain the reference as normal destruction does: drop the initial reference
and wait for qp->free, releasing the QP only after every concurrent lookup
returns its reference.
Fixes: 8001e9257e ("RDMA/mana_ib: extend mana QP table")
Link: https://patch.msgid.link/20260721-if-mana-table-store-qp-qids-partiall-v1-1-8fb3d2d2b559@nvidia.com
Reviewed-by: Konstantin Taranov <kotaranov@microsoft.com>
Reviewed-by: Long Li <longli@microsoft.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
The interrupt-status packet reports transmit credits returned by the
device. edge_interrupt_callback() adds the 16-bit value to txCredits
without checking maxTxCredits.
edge_write() uses txCredits minus the software FIFO count as the amount
of data that fits. Since the FIFO is allocated with maxTxCredits bytes,
txCredits exceeding maxTxCredits can cause OOB write in ring buffer.
Cap accumulated credits at maxTxCredits. Conforming devices should never
hit the cap.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Cc: stable@vger.kernel.org
Assisted-by: Codex:GPT-5
Signed-off-by: Sunho Park <shpark061104@gmail.com>
Signed-off-by: Johan Hovold <johan@kernel.org>
Use a flexible array member to reduce allocation by 1.
Add __counted_by for extra runtime analysis. Move counting variable
assignment after allocation before any array access.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Link: https://patch.msgid.link/20260609222454.37352-1-rosenp@gmail.com
Signed-off-by: Georgi Djakov <djakov@kernel.org>
Update the CQE structure to include hardware timestamp.
When firmware supports RDMA completion timestamps,
the hardware populates the timestamp field.
Co-developed-by: Allen Hubbe <allen.hubbe@amd.com>
Signed-off-by: Allen Hubbe <allen.hubbe@amd.com>
Signed-off-by: Abhijit Gangurde <abhijit.gangurde@amd.com>
Link: https://patch.msgid.link/20260610154216.712374-5-abhijit.gangurde@amd.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Enable user space applications to access the PHC state page when
firmware RDMA completion timestamp is supported.
This mapping allows user space to convert RDMA completion timestamps
to system wall time without kernel transitions, minimizing latency
overhead. Applications can directly read the PHC state through mmap,
enabling efficient timestamp correlation for precision timing
applications.
Co-developed-by: Allen Hubbe <allen.hubbe@amd.com>
Signed-off-by: Allen Hubbe <allen.hubbe@amd.com>
Signed-off-by: Abhijit Gangurde <abhijit.gangurde@amd.com>
Link: https://patch.msgid.link/20260610154216.712374-4-abhijit.gangurde@amd.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Add a page associated with the PHC that can be mapped to user space,
allowing applications to access hardware timestamp information.
In order to synchronize between kernel and user space, a sequence
number is incremented at the beginning and end of each update.
An odd number means the data is being updated while an even number
means the update is complete. To guarantee that the data structure
was accessed atomically, user space will:
repeat:
seq1 = <read sequence>
goto <repeat> if odd
<read PHC state>
seq2 = <read sequence>
if seq1 != seq2 goto repeat
This mechanism acts as a guard against reading invalid state during
concurrent updates.
Co-developed-by: Allen Hubbe <allen.hubbe@amd.com>
Signed-off-by: Allen Hubbe <allen.hubbe@amd.com>
Signed-off-by: Abhijit Gangurde <abhijit.gangurde@amd.com>
Link: https://patch.msgid.link/20260610154216.712374-3-abhijit.gangurde@amd.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Currently, the driver only registers the PTP Hardware Clock (PHC) if
Ethernet hardware timestamping is supported. Update the registration
logic to register the PHC if the device supports either Ethernet
hardware timestamping or RDMA completion timestamping.
Co-developed-by: Allen Hubbe <allen.hubbe@amd.com>
Signed-off-by: Allen Hubbe <allen.hubbe@amd.com>
Signed-off-by: Abhijit Gangurde <abhijit.gangurde@amd.com>
Link: https://patch.msgid.link/20260610154216.712374-2-abhijit.gangurde@amd.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Each CEQ interrupt handler only schedules eqc->tasklet. The tasklet calls
erdma_ceq_completion_handler(), which reads the DMA-coherent EQ ring
through get_next_valid_eqe() and updates eq->dbrec through notify_eq().
erdma_ceqs_uninit() frees each CEQ IRQ and then destroys its EQ.
free_irq() prevents another hard IRQ and waits for an in-flight handler,
but it does not drain a tasklet that the handler already scheduled. The
tasklet can therefore access eq->qbuf or eq->dbrec after
erdma_eq_destroy() frees them.
Clearing ceq_cb->ready does not synchronize with a tasklet that already
passed the check at the start of erdma_ceq_completion_handler().
Kill the tasklet after free_irq(), when no handler can schedule it again,
and before erdma_ceq_uninit_one() releases the EQ buffers.
Fixes: f2a0a630b9 ("RDMA/erdma: Add event queue implementation")
Co-developed-by: Ijae Kim <ae878000@gmail.com>
Signed-off-by: Ijae Kim <ae878000@gmail.com>
Signed-off-by: Myeonghun Pak <mhun512@gmail.com>
Link: https://patch.msgid.link/20260721082545.47395-1-mhun512@gmail.com
Acked-by: Cheng Xu <chengyou@linux.alibaba.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
The cid tables are visible to the cid kfuncs while being modified: the
first enable publishes the global pointers before filling them,
ops.init_cids() overrides rewrite them in place, and re-enables rebuild
them in place. A racing TRACING or SYSCALL program can read unfilled
entries, including uninitialized memory in the kmalloc'd tables, or torn
topo updates.
Tie the tables' lifetimes to the root sched instead: each root enable
builds a fresh set privately and publishes the per-table __rcu globals once
the layout is final, and root disable unpublishes and RCU-frees the set. A
non-NULL global is now always a fully built table which stays valid for the
reader's RCU read section, and lookups stay two loads. Kfuncs treat NULL as
no-mapping, also after the scheduler exits instead of reporting the stale
last mapping.
The cid kfuncs are available whether the root scheduler is cid-form or
cpu-form, the latter to allow gradual migration to cids. Every root
therefore builds and publishes a default mapping.
Every reader must either be gated on scheduler liveness or NULL-check
inside an RCU read section. Fix the two kfuncs that were neither:
scx_bpf_this_cid() read the table with no RCU or preemption protection and
scx_bpf_task_cid() relied on KF_RCU, which doesn't put a sleepable program
in an RCU read section. The hotplug callbacks are instead serialized by
retiring the tables inside the cpus_read_lock() section that clears
scx_root.
v2: Document why every root builds the tables (desc + cid.c comment).
Reported-by: Andrea Righi <arighi@nvidia.com>
Closes: https://lore.kernel.org/r/al3tLtPZZkFjMveK@gpd4
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
