A reader can hang forever in __ceph_get_caps() when the client no
longer holds `FILE_RD`, but local cap state still says that the
capability is already wanted (via `mds_wanted`).
One way to trigger this is through MDS cap revocation. If another
client performs a conflicting operation, the MDS can revoke `FILE_RD`
from the reader; the next read then has to reacquire `FILE_RD`. If
the cap update that should request `FILE_RD` never reaches the MDS
after `cap->mds_wanted` was raised, the reader is left holding only
non-file caps while local `mds_wanted` still includes the file read
caps.
In that state, try_get_cap_refs() sees `need <= mds_wanted` and
returns 0, so __ceph_get_caps() just waits on `i_cap_wq`. If the cap
update that was supposed to request `FILE_RD never reaches the MDS
after `cap->mds_wanted was` raised, no further request is sent and the
waiter can sleep indefinitely until unrelated cap traffic happens to
wake it up.
The ordering issue is that `cap->mds_wanted` is updated in
__prep_cap() before the `CEPH_MSG_CLIENT_CAPS message` is actually
queued for send. That makes one field serve two different meanings at
once: what this client wants, and what the client believes the MDS
already knows it wants.
A proper fix would be to split those states and track whether a cap
update is actually in flight or has been observed by the MDS.
However, simply moving the `cap->mds_wanted assignment` later would
not be sufficient: queueing the message in the messenger does not
guarantee that the MDS processed that specific wanted set, and
reconnect or message loss can still invalidate that assumption.
Fixing that properly would require a larger rework of the cap state
machine.
To allow simpler backports to stable kernels, this patch implements a
simpler workaround:
- stop waiting forever in __ceph_get_caps(); after a bounded wait,
fall back to the renew path
- make ceph_renew_caps() issue a synchronous `OPEN` request whenever
the inode still does not actually hold the wanted caps, instead of
only calling ceph_check_caps()
The extra issued-vs-wanted check in ceph_renew_caps() is necessary
because the previous test only checked whether the inode still had any
real caps at all. That is not enough after revocation: the client can
still hold something like `pLs` and yet be missing `FILE_RD`
completely. In that case, falling back to ceph_check_caps() is not
sufficient, because it still trusts `cap->mds_wanted` and may resend
nothing. By requiring `(issued & wanted) == wanted` before taking the
asynchronous path, the code only uses ceph_check_caps() when the
`wanted caps` are already actually issued. Otherwise, it sends the
synchronous `OPEN` renew.
This preserves the existing asynchronous fast path when the wanted
caps are already issued, avoids changing cap-state semantics, and
fixes the hang by guaranteeing that a stalled waiter eventually
retries through a path that does not rely on the stale `mds_wanted`
state.
[ idryomov: move CEPH_GET_CAPS_WAIT_TIMEOUT from libceph.h to
mds_client.h, formatting ]
Cc: stable@vger.kernel.org
Fixes: 0a454bdd50 ("ceph: reorganize __send_cap for less spinlock abuse")
Signed-off-by: Max Kellermann <max.kellermann@ionos.com>
Reviewed-by: Alex Markuze <amarkuze@redhat.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
In a reply message to an RBD request, a positive result code indicates
a data payload, which is not allowed for writes. While
rbd_osd_req_callback() already resets a positive result code for writes
to zero, rbd_object_map_callback() does not. This allows a corrupted
reply to an object map update to trigger the rbd_assert(*result < 0) in
__rbd_obj_handle_request(). This happens, because
rbd_object_map_callback() calls rbd_obj_handle_request() ->
__rbd_obj_handle_request() and passes this positive result code. From
__rbd_obj_handle_request(), rbd_obj_advance_write() is called, which
leaves the positive result code unchanged and returns true. Therefore,
the if(done && *result) branch is executed in __rbd_obj_handle_request()
and the assertion triggers.
This patch fixes the issue by adjusting the logic in the
rbd_object_map_callback() path. A positive result code for an object map
update is now reset to zero (similar to rbd_osd_req_callback()), and the
message is subsequently handled the same way as if the result code was
zero from the beginning. Additionally, a WARN_ON_ONCE() is added for
this case.
Cc: stable@vger.kernel.org
Fixes: 22e8bd51bb ("rbd: support for object-map and fast-diff")
Signed-off-by: Raphael Zimmer <raphael.zimmer@tu-ilmenau.de>
Reviewed-by: Ilya Dryomov <idryomov@gmail.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
__decode_pg_temp() decodes an user-controlled length but only rejects
values large enough to overflow the allocation; it does not bound it to
CEPH_PG_MAX_SIZE. The helper backs both pg_temp and pg_upmap decoding, and
apply_upmap()/get_temp_osds() later copy the decoded list into the fixed-size
on-stack array struct ceph_osds.osds[CEPH_PG_MAX_SIZE]. A monitor that sends
an OSDMap with a pg_temp/pg_upmap entry longer than 32 thus causes a stack
out-of-bounds write.
An OSD set for a single PG can never exceed CEPH_PG_MAX_SIZE, so reject longer
entries at decode time. The bound is well below the old overflow threshold, so
it also covers the allocation-size overflow the previous check guarded against.
BUG: KASAN: stack-out-of-bounds in ceph_pg_to_up_acting_osds
Write of size 4 ... by task exploit
kasan_report (mm/kasan/report.c:595)
ceph_pg_to_up_acting_osds (net/ceph/osdmap.c:2617 net/ceph/osdmap.c:2833)
calc_target (net/ceph/osd_client.c:1638)
__submit_request (net/ceph/osd_client.c:2394)
ceph_osdc_start_request (net/ceph/osd_client.c:2490)
ceph_osdc_call (net/ceph/osd_client.c:5164)
rbd_dev_image_probe (drivers/block/rbd.c:6899)
do_rbd_add (drivers/block/rbd.c:7138)
...
kernel BUG at net/ceph/osdmap.c:2670!
[ idryomov: do the same in __decode_pg_upmap_items() ]
Cc: stable@vger.kernel.org
Fixes: a303bb0e58 ("libceph: introduce and switch to decode_pg_mapping()")
Reported-by: Weiming Shi <bestswngs@gmail.com>
Assisted-by: Claude:claude-opus-4-8
Signed-off-by: Xiang Mei <xmei5@asu.edu>
Reviewed-by: Alex Markuze <amarkuze@redhat.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
ceph_x_create_authorizer() caches au->buf->vec.iov_base and
au->buf->vec.iov_len in struct ceph_auth_handshake. These
cached values are then used by the messenger connect code when
sending the authorizer.
ceph_x_update_authorizer() can rebuild the authorizer when a newer
service ticket is available. If the rebuilt authorizer no longer
fits in the existing buffer, ceph_x_build_authorizer() drops its
reference to au->buf and allocates a new one. If this is the final
reference, ceph_buffer_put() frees the old ceph_buffer and its
vec.iov_base, but auth->authorizer_buf still points at that freed
memory.
A subsequent msgr1 reconnect can therefore queue the stale pointer
and trigger a KASAN slab-use-after-free in _copy_from_iter() while
tcp_sendmsg() copies the authorizer.
Refresh auth->authorizer_buf and auth->authorizer_buf_len after a
successful authorizer rebuild so the messenger sends the current
buffer.
Cc: stable@vger.kernel.org
Fixes: 0bed9b5c52 ("libceph: add update_authorizer auth method")
Closes: https://lore.kernel.org/all/E378850E-106C-427B-A241-970EB2D054D7@gmail.com/
Signed-off-by: Shuangpeng Bai <shuangpeng.kernel@gmail.com>
Reviewed-by: Alex Markuze <amarkuze@redhat.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
The ceph_readdir() function allocates a ceph_mds_request via
ceph_mdsc_create_request() and stores it in dfi->last_readdir. In
the directory entry processing loop, if the entry's offset is less
than ctx->pos or if the inode pointer is unexpectedly NULL, the
function returns -EIO without releasing the reference held by
dfi->last_readdir, causing a refcount leak.
Fix this by adding ceph_mdsc_put_request(dfi->last_readdir) before
returning on these error paths. Also set dfi->last_readdir to NULL
for safety, matching the cleanup done at the normal exit.
Cc: stable@vger.kernel.org
Fixes: af9ffa6df7 ("ceph: add support to readdir for encrypted names")
Signed-off-by: WenTao Liang <vulab@iscas.ac.cn>
Reviewed-by: Viacheslav Dubeyko <slava@dubeyko.com>
Reviewed-by: Alex Markuze <amarkuze@redhat.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Localized read selection can walk a parent bucket whose name exists in
the CRUSH map while its type has no matching entry in type_names.
get_immediate_parent() then dereferences a NULL type_cn and passes an
invalid pointer into strcmp(), causing a null-ptr-deref.
Skip such malformed parent buckets unless both the bucket name and type
name metadata are present. This keeps malformed hierarchy data from
crashing locality lookup and safely falls back to "not local".
[ idryomov: add WARN_ON_ONCE ]
Cc: stable@vger.kernel.org
Fixes: 117d96a04f ("libceph: support for balanced and localized reads")
Reported-by: Yuan Tan <yuantan098@gmail.com>
Reported-by: Zhengchuan Liang <zcliangcn@gmail.com>
Reported-by: Xin Liu <bird@lzu.edu.cn>
Assisted-by: Codex:GPT-5.4
Signed-off-by: Zhao Zhang <zzhan461@ucr.edu>
Signed-off-by: Ren Wei <n05ec@lzu.edu.cn>
Reviewed-by: Viacheslav Dubeyko <slava@dubeyko.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
ceph_destroy_client() tears down the monitor client before removing
the per-client debugfs files. A concurrent read of the monmap debugfs
file can enter monmap_show() after ceph_monc_stop() has freed
monc->monmap, triggering a use-after-free.
Remove the debugfs files before stopping the OSD and monitor clients.
debugfs_remove() drains active handlers and prevents new accesses, so
the debugfs callbacks can no longer race the rest of client teardown.
Cc: stable@vger.kernel.org
Fixes: 76aa844d5b ("ceph: debugfs")
Reported-by: Yuan Tan <yuantan098@gmail.com>
Reported-by: Zhengchuan Liang <zcliangcn@gmail.com>
Reported-by: Xin Liu <bird@lzu.edu.cn>
Assisted-by: Codex:GPT-5.4
Signed-off-by: Douya Le <ldy3087146292@gmail.com>
Signed-off-by: Ren Wei <n05ec@lzu.edu.cn>
Reviewed-by: Viacheslav Dubeyko <slava@dubeyko.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
handle_get_version_reply() uses msg->front_alloc_len as the decode
boundary for MON_GET_VERSION_REPLY. That is the size of the reused
reply buffer, not the number of bytes actually received.
A truncated reply can therefore pass ceph_decode_need() and decode the
second u64 from stale tail bytes left in the buffer by an earlier
message, causing an uninitialized memory read.
Use msg->front.iov_len as the receive-side decode boundary, matching
other libceph reply handlers and limiting decoding to the bytes that
were actually read from the wire.
Cc: stable@vger.kernel.org
Fixes: 513a8243d6 ("libceph: mon_get_version request infrastructure")
Reported-by: Yuan Tan <yuantan098@gmail.com>
Reported-by: Zhengchuan Liang <zcliangcn@gmail.com>
Reported-by: Xin Liu <bird@lzu.edu.cn>
Assisted-by: Codex:GPT-5.4
Signed-off-by: Douya Le <ldy3087146292@gmail.com>
Signed-off-by: Ren Wei <n05ec@lzu.edu.cn>
Reviewed-by: Viacheslav Dubeyko <slava@dubeyko.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
write_folio_nounlock() increments fsc->writeback_count to track
in-flight writeback operations. On several error paths where the
function returns early (folio lookup failure, snapshot context
allocation failure, and writepages submission failure), the function
returns without calling atomic_long_dec_return() to decrement the
counter.
Each leaked increment keeps the counter above zero, which can prevent
the filesystem from cleanly unmounting or suspending writes.
Add atomic_long_dec_return() calls on all error paths that currently
return without decrementing the counter.
Cc: stable@vger.kernel.org
Fixes: d55207717d ("ceph: add encryption support to writepage and writepages")
Signed-off-by: Wentao Liang <vulab@iscas.ac.cn>
Reviewed-by: Viacheslav Dubeyko <Slava.Dubeyko@ibm.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
decode_lockers() in cls_lock_client.c contains two bare decode operations
that allow a malicious or compromised OSD to trigger slab-out-of-bounds
reads:
1. ceph_decode_32(p) at the num_lockers field has no preceding bounds
check. ceph_start_decoding() accepts struct_len=0 as valid -- the
internal ceph_decode_need(p, end, 0, bad) always passes -- so when an
OSD sends struct_len=0, ceph_start_decoding() returns success with
p == end. The immediately following bare ceph_decode_32(p) then reads
4 bytes past the validated buffer boundary. The garbage value is
passed directly to kzalloc_objs() as the locker count.
The sibling function decode_watchers() in osd_client.c already uses
ceph_decode_32_safe() after its own ceph_start_decoding() call.
decode_lockers() was the only site using the bare variant.
2. ceph_decode_8(p) after the decode_locker() loop has no preceding
bounds check. If an OSD crafts num_lockers such that the loop
advances p exactly to end, the subsequent bare ceph_decode_8(p) reads
one byte past the validated buffer boundary. The result is passed
directly into *type, which is used as a lock type discriminator by
callers, giving an OSD-controlled one-byte OOB read with direct
influence over the lock type field.
Fix both by replacing bare operations with their safe variants:
ceph_decode_32(p) -> ceph_decode_32_safe(p, end, *num_lockers,
err_inval)
ceph_decode_8(p) -> ceph_decode_8_safe(p, end, *type,
err_free_lockers)
The goto targets differ intentionally:
err_inval: is a new label returning -EINVAL directly. It is used for
the pre-allocation failure path where *lockers is not yet allocated
and must not be passed to ceph_free_lockers().
err_free_lockers: is the existing label. It is used for the
post-allocation failure path where *lockers is allocated and must
be freed.
ret is set to -EINVAL before ceph_decode_8_safe() so that
err_free_lockers returns the correct error code on bounds violation.
Without this, err_free_lockers would return a stale ret value (0 from
the successful decode_locker() loop), silently swallowing the error.
-EINVAL is correct for both failure paths. The data received from the
OSD is structurally malformed. -ENOMEM would misrepresent the failure
class to callers and to stable@ backporters triaging error paths.
Attacker model: a malicious or compromised OSD in a multi-tenant Ceph
deployment can trigger this against any kernel client that issues the
lock.get_info class method (e.g. during RBD exclusive lock acquisition).
[ idryomov: trim changelog, formatting ]
Cc: stable@vger.kernel.org
Fixes: d4ed4a5305 ("libceph: support for lock.lock_info")
Signed-off-by: Pavitra Jha <jhapavitra98@gmail.com>
Reviewed-by: Viacheslav Dubeyko <Slava.Dubeyko@ibm.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
ceph_handle_caps() reads snap_trace_len from the wire-format
ceph_mds_caps header and uses it unconditionally to build a fake
end pointer (snaptrace + snaptrace_len) that is later handed to
ceph_update_snap_trace() in the CEPH_CAP_OP_IMPORT case:
snaptrace = h + 1;
snaptrace_len = le32_to_cpu(h->snap_trace_len);
p = snaptrace + snaptrace_len;
...
case CEPH_CAP_OP_IMPORT:
if (snaptrace_len) {
...
if (ceph_update_snap_trace(mdsc, snaptrace,
snaptrace + snaptrace_len,
false, &realm)) { ... }
ceph_update_snap_trace() then decodes a struct ceph_mds_snap_realm
from snaptrace using ceph_decode_need(&p, e, sizeof(*ri), bad)
with the attacker-supplied fake end e == snaptrace + snaptrace_len.
With snaptrace_len == 0xFFFFFFFF the bound check is trivially
satisfied, ri = p reads sizeof(struct ceph_mds_snap_realm) past
the legitimate msg->front buffer, and ri->num_snaps /
ri->num_prior_parent_snaps then drive further out-of-bounds
reads of the encoded snap arrays.
The eleven msg_version >= 2 .. msg_version >= 12 decoder blocks
above the op switch each catch this OOB through their
ceph_decode_*_safe() / ceph_decode_need() helpers, but they sit
behind a hdr.version-gated if, so a malicious or compromised
MDS that sets msg->hdr.version = 1 reaches the IMPORT path with
no version-gated decoder having validated snap_trace_len. The
shape has been present since ceph_handle_caps() was introduced.
Validate snap_trace_len against the message front buffer before
consuming it, using the canonical ceph_decode_need() / ceph_has_room()
helper. The helper bounds the length with subtraction (n <= end - p,
guarded by end >= p) rather than pointer addition, so it is wrap-safe
for the attacker-controlled u32 length on 32-bit builds where
p + snap_trace_len could overflow the address space. This matches the
rest of the ceph decode path (e.g. the pool_ns_len check a few lines
below), and the existing goto bad cleanup already covers this exit
path.
Cc: stable@vger.kernel.org
Fixes: a8599bd821 ("ceph: capability management")
Signed-off-by: Bryam Vargas <hexlabsecurity@proton.me>
Reviewed-by: Viacheslav Dubeyko <Slava.Dubeyko@ibm.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
A message of type CEPH_MSG_MON_MAP contains a monmap that is sent from a
monitor to the client. This monmap contains information about the
existing monitors in the cluster. Currently, a monmap indicating that
there are zero monitors in the cluster is treated as valid. However, it
is impossible to have zero monitors in the cluster and still receive a
valid monmap from a monitor. Therefore, such a monmap must be corrupted
and should be treated as invalid. Furthermore, a monmap with a monitor
count of zero can subsequently crash the client when attempting to open
a session with a monitor in __open_session(). This happens because the
"BUG_ON(monc->monmap->num_mon < 1)" assertion in pick_new_mon() is
triggered.
This patch extends a check in ceph_monmap_decode() to also reject
arriving mon_maps with num_mon == 0 rather than only with
num_mon > CEPH_MAX_MON.
[ idryomov: drop "log output for unusual values of num_mon" part ]
Cc: stable@vger.kernel.org
Signed-off-by: Raphael Zimmer <raphael.zimmer@tu-ilmenau.de>
Reviewed-by: Ilya Dryomov <idryomov@gmail.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
CRUSH bucket type 0 is reserved for devices. The mapper relies on
that invariant and uses type 0 to identify leaf devices.
If crush_decode() accepts a bucket with type 0, a malformed CRUSH map
can make the mapper treat a negative bucket ID as a device and pass it
to is_out(), which then indexes the OSD weight array with a negative
value.
Reject zero bucket types while decoding the CRUSH map so the invalid
state never reaches the mapper.
Cc: stable@vger.kernel.org
Fixes: f24e9980eb ("ceph: OSD client")
Reported-by: Yuan Tan <yuantan098@gmail.com>
Reported-by: Zhengchuan Liang <zcliangcn@gmail.com>
Reported-by: Xin Liu <bird@lzu.edu.cn>
Assisted-by: Codex:GPT-5.4
Signed-off-by: Douya Le <ldy3087146292@gmail.com>
Signed-off-by: Ren Wei <n05ec@lzu.edu.cn>
Reviewed-by: Ilya Dryomov <idryomov@gmail.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
If a message of type CEPH_MSG_OSD_MAP contains a (maliciously) corrupted
osdmap, out-of-bounds memory accesses may occur in
decode_new_up_state_weight(). This happens because the bounds check for
the new_state part is based on calculating its length depending on a len
value read from the incoming message. This calculation may overflow
leading to an incorrect bounds check. Subsequently, out-of-bounds reads
may occur when decoding this part.
This patch switches the multiplication to use check_mul_overflow() to
abort processing the osdmap if an overflow occurred. Therefore,
osdmaps/messages containing large values for len that result in a
multiplication overflow are treated as invalid.
[ idryomov: rename new_state_len -> new_state_item_size, formatting ]
Cc: stable@vger.kernel.org
Fixes: 930c532869 ("libceph: apply new_state before new_up_client on incrementals")
Signed-off-by: Raphael Zimmer <raphael.zimmer@tu-ilmenau.de>
Reviewed-by: Viacheslav Dubeyko <Slava.Dubeyko@ibm.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
- Call flush_cache_vmap() after populating new vmemmap pages, on all
architectures. This avoids spurious faults on RISC-V
microarchitectures that cache PTEs marked as non-present
- Disable LTO for the vDSO to prevent the compiler from eliding
functions that are used, but which don't appear to be
- Fix an issue with libgcc's unwinder and signal handlers by dropping
an unnecessary CFI landing pad instruction in __vdso_rt_sigreturn
(similar to what was done on ARM64)
- Avoid reading uninitialized memory under certain conditions in
hwprobe_get_cpus()
- Save some memory and I$ when CONFIG_DYNAMIC_FTRACE=n by avoiding
our four-byte function alignment requirement in that case
- Avoid clang warnings about null-pointer arithmetic in the I/O-port
accessor macros (inb, outb, etc.) by ifdeffing them out when
!CONFIG_HAS_IOPORT
- Make the build of the lazy TLB flushing code in the vmalloc path
depend on CONFIG_64BIT and CONFIG_MMU (since those platforms are the
only ones that use it)
-----BEGIN PGP SIGNATURE-----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=gKi/
-----END PGP SIGNATURE-----
Merge tag 'riscv-for-linus-7.2-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux
Pull RISC-V fixes from Paul Walmsley:
- Call flush_cache_vmap() after populating new vmemmap pages, on all
architectures. This avoids spurious faults on RISC-V
microarchitectures that cache PTEs marked as non-present
- Disable LTO for the vDSO to prevent the compiler from eliding
functions that are used, but which don't appear to be
- Fix an issue with libgcc's unwinder and signal handlers by dropping
an unnecessary CFI landing pad instruction in __vdso_rt_sigreturn
(similar to what was done on ARM64)
- Avoid reading uninitialized memory under certain conditions in
hwprobe_get_cpus()
- Save some memory and I$ when CONFIG_DYNAMIC_FTRACE=n by avoiding our
four-byte function alignment requirement in that case
- Avoid clang warnings about null-pointer arithmetic in the I/O-port
accessor macros (inb, outb, etc.) by ifdeffing them out when
!CONFIG_HAS_IOPORT
- Make the build of the lazy TLB flushing code in the vmalloc path
depend on CONFIG_64BIT and CONFIG_MMU (since those platforms are the
only ones that use it)
* tag 'riscv-for-linus-7.2-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:
riscv: hwprobe: Avoid uninitialized read in hwprobe_get_cpus()
arch/riscv: vdso: remove CFI landing pad from rt_sigreturn
riscv: vdso: Do not use LTO for the vDSO
riscv: io: avoid null-pointer arithmetic in PIO helpers
riscv: Gate FUNCTION_ALIGNMENT_4B on DYNAMIC_FTRACE
mm/sparse-vmemmap: flush_cache_vmap() after hotplugging vmemmap
riscv: mm: Make mark_new_valid_map() stuff depend on 64BIT && MMU
-----BEGIN PGP SIGNATURE-----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=qHz/
-----END PGP SIGNATURE-----
Merge tag 'block-7.2-20260717' of git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux
Pull block fixes from Jens Axboe:
- Fixes for the dio bounce buffer helpers: correct the alignment of
bounced dio read bios to avoid a double unpin, handle huge zero
folios in bio_free_folios(), and don't warn on the larger-order folio
attempts in the greedy allocation path.
- Try a slab allocation in bio_alloc_bioset() before falling back to
the mempool, restoring the previous behavior for non-sleeping
allocations from a cache-enabled bioset.
- Serialize elevator changes for the same queue using the writer lock.
- Fix a race in blk_time_get_ns() where a task preempted between
setting PF_BLOCK_TS and the cached-timestamp reload could return 0.
- blk-cgroup fix for leaks and the online flag on a radix_tree_insert()
failure in blkg_create().
- Free the copied pages when blk_rq_map_kern() fails after
blk_rq_append_bio() rejects the bio.
- Remove manually added partitions on loop device detach, fixing dead
partition devices left behind and a subsequent LOOP_CONFIGURE -EBUSY
- Bound the AIX partition lvd scan to the sector that was actually
read.
- Show the block operation in error injection rules (Jackie)
* tag 'block-7.2-20260717' of git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux:
block: fix aligning of bounced dio read bios
block: handle huge zero folios in bio_free_folios
block: try slab allocation in bio_alloc_bioset() before mempool
block: show operation in error injection rules
block: serialize elevator changes for the same queue using a writer lock
block: free copied pages when blk_rq_map_kern() fails
block: do not warn when doing greedy allocation in folio_alloc_greedy()
partitions: aix: bound the lvd scan to one sector
blk-cgroup: fix leaks and online flag on radix_tree_insert failure
loop: remove manually added partitions on detach
block: fix race in blk_time_get_ns() returning 0
-----BEGIN PGP SIGNATURE-----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=Ue52
-----END PGP SIGNATURE-----
Merge tag 'io_uring-7.2-20260717' of git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux
Pull io_uring fixes from Jens Axboe:
- Fix a use-after-free in the bpf-ops struct_ops path, where the same
io_uring_bpf_ops map could be registered more than once.
- Fix the deferred iovec free for the provided-buffer grow path, which
could leave the caller with a dangling iovec and result in repeated
frees. Follow-up to the earlier fix in this series.
- Zero-check the unused addr3/pad2 SQE fields for unlinkat
* tag 'io_uring-7.2-20260717' of git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux:
io_uring/bpf-ops: reject re-registration of an already-bound ops
io_uring/fs: check unused sqe fields for unlinkat
io_uring/kbuf: free the replaced iovec after a successful grow
A couple of fairly routine driver fixes, nothing too remarkable.
-----BEGIN PGP SIGNATURE-----
iQEzBAABCgAdFiEEreZoqmdXGLWf4p/qJNaLcl1Uh9AFAmpcxP4ACgkQJNaLcl1U
h9Cnigf9FRw9pWxDNEhhABYvy7A9kcSTYWDoYEoljdsvYzGCWpbkS9VCmiSOcxlA
WYrVkrknUrjnd2twMyqWSs3MGePcncEchlOKOqTF8bA12dKSiRgFlldZ5MoRgYrf
GAbFzlu0Z3wZgGK2n8NwHwQ4ut0LcM4PYYKqQ+nABbURmfCA+oCnbE2QhCdRxHUB
MtHiA1fzoBnnCHsDkRnI5MTbkKWA2r2nqzQY7pzojtwrR4xghm1MUO16gD5V8VM9
w5ndpdLIer3FtHpwy1OqtuvXnZJEu8xWqQ/uAv+HfdHsRvckPB6hKt6GkSKVdut/
R4e8ZbIuSfEwcwy8VC4fRQjVw5BNVQ==
=b2Lc
-----END PGP SIGNATURE-----
Merge tag 'spi-fix-v7.2-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi
Pull spi fixes from Mark Brown:
"A couple of fairly routine driver fixes, nothing too remarkable"
* tag 'spi-fix-v7.2-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi:
spi: cadence-quadspi: Fix indirect write timeout when DMA read mode is enabled
spi: dw-dma: Wait for controller idle before completing Tx
One straightforward driver fix for some incorrectly described bitfields
in the ltc3676 driver.
-----BEGIN PGP SIGNATURE-----
iQEzBAABCgAdFiEEreZoqmdXGLWf4p/qJNaLcl1Uh9AFAmpcxWoACgkQJNaLcl1U
h9Bj2Af+J+ibj5u+dpsT9GdFRkV3dkZtR7wAShFlbVYRI0dtEvrvAOFQ+/dh8whX
vRs8qvn8JtpXepUs2DsZlvB70CEdK5yYIuAQBJAj1QpolWLd069q0TK3fHAdqD6R
61KCpuhbQ8JTCYzL/BCmKhYTqr9A9vhC0hn6cFflqT3Ngunxvayal/pIiutC0S0W
AT57CKul41+DUPJ7XhKlCuuo2ZUL6wiYKkecRp8+hsMWllUaN9Jg+gsPKZQXwL18
ykKOsfYqQR+qinUNRqZTvzJUlMZNJCOpeTZnrVx1gIRv03UpCFnLSCQlDDRWqHkZ
CfDemVw6oZdyHY+0qNroKYENpbqWoA==
=SYeH
-----END PGP SIGNATURE-----
Merge tag 'regulator-fix-v7.2-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator
Pull regulator fix from Mark Brown:
"One straightforward driver fix for some incorrectly described
bitfields in the ltc3676 driver"
* tag 'regulator-fix-v7.2-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator:
regulator: ltc3676: Fix incorrect IRQSTAT bit offsets
- Reject too long acpi_rsdp= boot parameter values
(Thorsten Blum)
- Validate console=uart8250 baud rate to fix early boot hang
(Thorsten Blum)
- Remove dead Makefile rule (Ethan Nelson-Moore)
Signed-off-by: Ingo Molnar <mingo@kernel.org>
-----BEGIN PGP SIGNATURE-----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=upW0
-----END PGP SIGNATURE-----
Merge tag 'x86-urgent-2026-07-19' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fixes from Ingo Molnar:
- Reject too long acpi_rsdp= boot parameter values (Thorsten Blum)
- Validate console=uart8250 baud rate to fix early boot hang (Thorsten
Blum)
- Remove dead Makefile rule (Ethan Nelson-Moore)
* tag 'x86-urgent-2026-07-19' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/boot: Validate console=uart8250 baud rate to fix early boot hang
x86/boot: Reject too long acpi_rsdp= values
x86/cpu: Remove Makefile rule for removed UMC CPU support
- Fix checksum lib on machines without the vector facility where the
non-vector fallback made csum_partial() calculate the checksum from
address 0 instead of the provided buffer
- Fix cpum_cf perf event initialization missing speculation barrier for
user controlled event numbers used as generic event array indexes
-----BEGIN PGP SIGNATURE-----
iQEzBAABCgAdFiEE3QHqV+H2a8xAv27vjYWKoQLXFBgFAmpcDpgACgkQjYWKoQLX
FBizogf/fOTDdmXPx2xp5PFadlaoVnRmycslrPDog+lVp2tl+RXlAU6mKT27gcXT
2G/PCF9SjYQSQdGztMNPCh6Az1cAo+XvhADtElumRAjlybTYlk++aallv5+buHRS
Z2jEnurYVWQaQP/PpFDIUhh7Yf6ggrctXuRdYgyJv/5Siy4OQYQEpWQazQXyfOZI
506gQ2M4mD8dIJRgZygM7tQztR7K8IFDIUPzxMCiBKj9XQQBpZ0Q9SOm07nL6dtS
MHcXbyd80XbvVoYNCZHquktJ4ZOvru+sA65Hd94TSHOQRDxuuiDueo6VYPPqBX5o
/ENY4mFt0aD7m0lbTqMyxrUI2bDzLg==
=WPyD
-----END PGP SIGNATURE-----
Merge tag 's390-7.2-5' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux
Pull s390 fixes from Vasily Gorbik:
- Fix checksum lib on machines without the vector facility where the
non-vector fallback made csum_partial() calculate the checksum from
address 0 instead of the provided buffer
- Fix cpum_cf perf event initialization missing speculation barrier for
user controlled event numbers used as generic event array indexes
* tag 's390-7.2-5' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux:
s390/perf_cpum_cf: Add missing array_index_nospec() to __hw_perf_event_init()
s390/checksum: Fix csum_partial() without vector facility
The biggest core change is the reliable wake fix for scsi_schedule_eh
which is used by both libata and libsas which could otherwise cause
error handler hangs due to rare races. All other fixes are in drivers
(well except the export symbol removal) the next biggest being the
target PR-OUT transportid parsing fix.
Signed-off-by: James E.J. Bottomley <James.Bottomley@HansenPartnership.com>
-----BEGIN PGP SIGNATURE-----
iLgEABMIAGAWIQTnYEDbdso9F2cI+arnQslM7pishQUCalvAVRsUgAAAAAAEAA5t
YW51MiwyLjUrMS4xMiwyLDImHGphbWVzLmJvdHRvbWxleUBoYW5zZW5wYXJ0bmVy
c2hpcC5jb20ACgkQ50LJTO6YrIUI/QD+Kr/rBo60RkNJfsdC6vmHj0+pkkSpQNkM
XGTZXjPz+poBAL7ybOwWvQWYoJ2LLDg10F9dSb68DxE2FJAS3lzM0IGf
=dhtX
-----END PGP SIGNATURE-----
Merge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Pull SCSI fixes from James Bottomley:
"The biggest core change is the reliable wake fix for scsi_schedule_eh
which is used by both libata and libsas which could otherwise cause
error handler hangs due to rare races.
All other fixes are in drivers (well except the export symbol removal)
the next biggest being the target PR-OUT transportid parsing fix"
* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
scsi: hpsa: Fix DMA mapping leak on IOACCEL2 reset path
scsi: elx: efct: Fix refcount leak in efct_hw_io_abort()
scsi: elx: efct: Fix I/O leak on unsupported additional CDB
scsi: core: wake eh reliably when using scsi_schedule_eh
scsi: target: core: Fix iSCSI ISID use-after-free in REGISTER AND MOVE
scsi: target: Bound PR-OUT TransportID parsing to the received buffer
scsi: lpfc: Fix memory leak in lpfc_sli4_driver_resource_setup()
scsi: sg: Report request-table problems when any status is set
scsi: ufs: core: tracing: Do not dereference pointers in TP_printk()
scsi: bfa: Reduce kernel stack usage in bfa_fcs_lport_fdmi_build_portattr_block()
scsi: xen: scsiback: Free the command tag on the TMR submit-failure path
scsi: xen: scsiback: Free unsubmitted command instead of double-putting it
scsi: core: Remove export for scsi_device_from_queue()
imx: fix empty SMBus block reads in atomic and IRQ paths
spacemit: return IRQ_NONE for empty interrupt status
mlxbf: fix a use-after-free issue
mediatek: restore WRRD support without automatic restart
CREDITS: add Wolfram Sang
-----BEGIN PGP SIGNATURE-----
iHUEABYKAB0WIQScDfrjQa34uOld1VLaeAVmJtMtbgUCaltX3AAKCRDaeAVmJtMt
bqqgAPsFCe+3Y/71UNBWzOEI99zUlyFu7OH7Rt/015HVXsmelgEAjkQ0X5RX2Njg
AswoeT+OGvxxVxlNeNMMrxAM4s6caQU=
=yKh8
-----END PGP SIGNATURE-----
Merge tag 'i2c-fixes-7.2-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/andi.shyti/linux
Pull i2c fixes from Andi Shyti:
"A handful of small fixes for host controller drivers.
One patch also adds Wolfram Sang to CREDITS after more than a decade
of work on I2C"
* tag 'i2c-fixes-7.2-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/andi.shyti/linux:
i2c: mediatek: fix WRRD for SoCs without auto_restart option
i2c: mlxbf: Fix use-after-free in mlxbf_i2c_init_resource()
i2c: spacemit: fix spurious IRQ handling returning IRQ_HANDLED
i2c: imx: fix locked bus on SMBus block-read of 0 (IRQ)
i2c: imx: fix locked bus on SMBus block-read of 0 (atomic)
CREDITS: Add Wolfram Sang
-----BEGIN PGP SIGNATURE-----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=K5Ks
-----END PGP SIGNATURE-----
Merge tag 'v7.2-rc3-smb3-server-fixes' of git://git.samba.org/ksmbd
Pull smb server fixes from Steve French:
"ksmbd server fixes, mostly addressing malformed SMB request
handling and connection/session lifetime issues, including
two information-disclosure or memory-safety bugs in the SMB2
request/response paths.
- validate FILE_ALLOCATION_INFORMATION before block rounding to
prevent a client-controlled overflow from truncating a file.
- pin connections while asynchronous oplock and lease-break
notifications are pending.
- initialize compound SMB2 READ alignment padding, preventing
disclosure of uninitialized heap bytes.
- release the allocated alternate-stream xattr name after rename.
- size multichannel binding session-key buffers for the largest
permitted key, avoiding a stack buffer overflow.
- remove a disconnecting connection's channels from every session,
including channels whose binding state has since changed.
- serialize binding preauthentication-session lookup and update
against its teardown.
- check that every compound request element contains StructureSize2
before reading it"
* tag 'v7.2-rc3-smb3-server-fixes' of git://git.samba.org/ksmbd:
ksmbd: validate compound request size before reading StructureSize2
ksmbd: lock the binding preauth session in smb3_preauth_hash_rsp
ksmbd: remove stale channels from all sessions on teardown
ksmbd: fix stack buffer overflow in multichannel session-key copy
ksmbd: fix memory leak of xattr_stream_name in smb2_rename()
ksmbd: zero the smb2_read alignment tail to avoid an infoleak
ksmbd: pin conn during async oplock break notification
ksmbd: fix integer overflow in set_file_allocation_info()
- Interrupt initialization and handling Fixes for the Designware
ahci_dwc driver (Rosen)
- Avoid possible infinite loop when scanning completion in the
Designware ahci_dwc driver (Rosen)
-----BEGIN PGP SIGNATURE-----
iHUEABYKAB0WIQSRPv8tYSvhwAzJdzjdoc3SxdoYdgUCalrPMQAKCRDdoc3SxdoY
dnQ3AP9NaxQu5xoACQtFtNPb8CyY35CRclAKZsKk+/clCNAVwwEAuUo2rgw/eAfP
w98ZEvN1HMRqW4dbatv2d3zNw2L8Iw8=
=9+gW
-----END PGP SIGNATURE-----
Merge tag 'ata-7.2-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/libata/linux
Pull ata fixes from Damien Le Moal:
- Interrupt initialization and handling fixes for the Designware
ahci_dwc driver (Rosen)
- Avoid possible infinite loop when scanning completion in the
Designware ahci_dwc driver (Rosen)
* tag 'ata-7.2-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/libata/linux:
ata: sata_dwc_460ex: fix infinite loop in NCQ tag completion bit-scanning
ata: sata_dwc_460ex: fix clear_interrupt_bit() clearing all pending interrupts
ata: sata_dwc_460ex: use platform_get_irq()
ata: sata_dwc_460ex: enable SATA interrupts only after IRQ handler is registered
Now that proper fixes have been found, let's revert this workaround.
This reverts commit a1fc7bf667.
Tested-by: Mario Limonciello (AMD) <superm1@kernel.org>
Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org>
Signed-off-by: Leo Li <sunpeng.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit f64a9be5653689ff43e148cd8a6483077488c8e5)
Cc: stable@vger.kernel.org # 8382cd234981: drm/amd/display: consolidate DCN vblank/flip handling onto vupdate_no_lock
Cc: stable@vger.kernel.org # 48ab86360af1: drm/amd/display: check GRPH_FLIP status before sending event
Cc: stable@vger.kernel.org
[Why]
After unifying DCN interrupt sources under VUPDATE_NO_LOCK, we have two
remaining issues to clean up:
1. On DCN, flip completion is now delivered from VUPDATE_NO_LOCK
(dm_crtc_high_irq_handler) instead of GRPH_PFLIP. But VUPDATE_NO_LOCK
fires every frame, regardless of whether a flip has latched.
2. There is a window during commit where a flip is armed (pflip_status =
SUBMITTED) but not yet programmed into HW. If the VUPDATE_NO_LOCK
fires in that window, its handler would deliver a flip event to
userspace before HW has latched to it. If userspace then renders to
what it believes is now the back buffer (but HW is still latched to
it!), it will cause display corruption. This issue seemed to have
been introduced by:
commit 1159898a88 ("drm/amd/display: Handle commit plane with no FB.")
Enabling replay or psr extended the duration of this window, and
hence made corruption more likely to be observed.
[How]
* Move acrtc->event/pflip_status arming to after
update_planes_and_stream_adapter() has programmed the flip into HW.
This closes the window where pflip_status is SUBMITTED but the flip is
not yet programmed.
* Add dc_get_flip_pending_on_otg(), which reads the HUBP flip-pending
status straight from HW for the pipe(s) bound to an OTG instance. It
is keyed only by otg_inst and does not take or mutate a
dc_plane_state, so it is safe to call from the OTG interrupt handler
without racing a concurrent commit that may be modifying plane state.
* Optimistically query for flip-pending after programming, in the event
that HW latched to the new fb between programming start and arming
event. If it latched, send the vblank event immediately, rather than
wait for the next vblank IRQ.
* In the VUPDATE_NO_LOCK handler, only deliver flip completion once
dc_get_flip_pending_on_otg() reports the flip is no longer pending.
Otherwise leave the flip armed and retry on the next vupdate.
* For DCE, maintain the existing behavior of arming flips before
programming, and relying on GRPH_FLIP to fire at HW latch.
v2:
* Drop flip_programmed completion object, instead move
event/pflip_status arming after programming.
* For DCN, optimistically query for flip pending immediately after
programming, and if it latched, send event right away.
v3:
* Fix event timestamps on optimistic flip latch detection, where it's
possible for it to run *before* the vupdate IRQ updates the timestamp.
* Add more docstrings for DCN vblank handling.
* Clean up if conditions in dm_arm_vblank_event().
* Code style cleanup on braces surrounding multi-line statements.
Fixes: 9b47278cec ("drm/amd/display: temp w/a for dGPU to enter idle optimizations")
Link: https://gitlab.freedesktop.org/drm/amd/-/work_items/3787
Link: https://gitlab.freedesktop.org/drm/amd/-/work_items/4141
Assisted-by: Copilot:claude-opus-4.8
Tested-by: Mario Limonciello (AMD) <superm1@kernel.org>
Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org>
Signed-off-by: Leo Li <sunpeng.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit f64a9be5653689ff43e148cd8a6483077488c8e5)
Cc: stable@vger.kernel.org # 8382cd234981: drm/amd/display: consolidate DCN vblank/flip handling onto vupdate_no_lock
Cc: stable@vger.kernel.org
[Why]
On DCN, vblank events were delivered from VSTARTUP/VUPDATE
(dm_crtc_high_irq/dm_vupdate_high_irq) and pageflip completion from
GRPH_PFLIP (dm_pflip_high_irq). These signals can be masked by hardware
by a few things:
* DPG - DCN can Dynamically Power Gate parts of the display pipe when a
self-refresh capable eDP is connected. DPG is engaged when there's
enough static frames (detected through drm_vblank_off). Once gated,
even though the OTG (output timing generator) is still enabled,
VSTARTUP and GRPH_FLIP are masked.
* GSL - Driver can use the Global Sync Lock to block HW from latching
onto double-buffered registers during programming, to prevent HW from
latching onto a partially programmed state. This will mask VSTARTUP,
GRPH_FLIP, and VUPDATE. See dcn20_pipe_control_lock().
* MALL - A DCN accessible cache introduced in DCN32+ DGPUs that can
store fb data to allow for longer DRAM sleep. When scanning out from
MALL, VSTARTUP is masked.
When masked, events are never delivered, which can show up as flip_done
timeouts in the wild.
However, there is an interrupt source on DCN that is never masked:
VUPDATE_NO_LOCK. It's simply an unmasked variant of VUPDATE, which fires
while the OTG is active, at the exact point hardware latches
double-buffered registers. It is therefore the natural single signal for
delivering both vblank and flip-completion events on DCN, and the
correct point to timestamp both VRR and non-VRR vblanks.
DCE's interrupt sources are different, it does not have an unmaskable
VUPDATE_NO_LOCK. The only unmaskable DCE interrupt is VLINE0, but it can
only be programmed as a vline offset from vsync_start, making it
unsuitable for VRR. Thus, we keep DCE untouched and use the existing mix
of interrupt sources.
[How]
For DCN1 and newer only:
* Factor the body of dm_crtc_high_irq() into dm_crtc_high_irq_handler()
and drive it from dm_vupdate_high_irq() (VUPDATE_NO_LOCK). DCE keeps
using dm_crtc_high_irq() (VSTARTUP) and dm_pflip_high_irq()
(GRPH_PFLIP) unchanged.
* Stop registering VSTARTUP (crtc_irq) and GRPH_PFLIP (pageflip_irq) on
DCN, and stop enabling them in amdgpu_dm_crtc_set_vblank() /
manage_dm_interrupts(). Enable VUPDATE whenever vblank is enabled on
DCN (previously only in VRR mode). The secure-display vline0 interrupt
is left untouched.
* VUPDATE_NO_LOCK does not early-fire on an immediate (tearing / async)
flip, since HW latches the new address right away. Deliver the flip
completion event immediately after programming such flips in
amdgpu_dm_commit_planes(), and clear pflip_status so the next vupdate
handler does not double-send.
v2: Do not gate VUPDATE_NO_LOCK on DCN in dm_handle_vrr_transition()
Also toggle VUPDATE_NO_LOCK on DCN in dm_gpureset_toggle_interrupts()
Re-cook vblank event count and timestamp for immediate flips
Fixes: 9b47278cec ("drm/amd/display: temp w/a for dGPU to enter idle optimizations")
Link: https://gitlab.freedesktop.org/drm/amd/-/work_items/3787
Link: https://gitlab.freedesktop.org/drm/amd/-/work_items/4141
Assisted-by: Copilot:claude-opus-4.8
Co-developed-by: Matthew Schwartz <matthew.schwartz@linux.dev>
Signed-off-by: Matthew Schwartz <matthew.schwartz@linux.dev>
Tested-by: Mario Limonciello (AMD) <superm1@kernel.org>
Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org>
Signed-off-by: Leo Li <sunpeng.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit c87e6635d2db02c88ae8d09529362da672d34770)
Cc: stable@vger.kernel.org
Some AMD APU multi-function devices expose an integrated USB xHCI
controller. In some circumstances (such as larger VRAM), the PM core
can resume can fail when the xHCI controller is resuming in parallel
with the GPU/display function.
On affected systems, the xHCI controller can complete pci_pm_resume
and start resuming USB devices while the GPU is still in its much
longer resume path. This race condition leads to USB device resume
failures followed by:
xhci_hcd ...: xHCI host not responding to stop endpoint command
xhci_hcd ...: HC died; cleaning up
Create a device link from any xHCI controller sharing the same PCIe
root port as the APU display function. The link uses DL_FLAG_STATELESS
and DL_FLAG_PM_RUNTIME to ensure the GPU completes its resume before
the xHCI controller begins resuming USB devices.
This device link is done specifically in amdgpu so that if the
platform firmware has been modified such that this issue doesn't happen
the version can be detected and the workaround skipped.
Suggested-by: Aaron Ma <aaron.ma@canonical.com>
Reported-by: mrh@frame.work
Closes: https://bugzilla.kernel.org/show_bug.cgi?id=221073
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Tested-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Tested-by: Alexander F <superveridical@gmail.com>
Tested-by: Francis DB <francisdb@gmail.com>
Link: https://patch.msgid.link/20260713195313.1739762-1-mario.limonciello@amd.com
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 07c93d7eeb0d990bc1b8e3b1eafa464bc9feee97)
Cc: stable@vger.kernel.org
DCN42B enables DML2 and DML21 by default and defines
dcn42b_prepare_mcache_programming(), but the resource function table only
wires the callback when CONFIG_DRM_AMD_DC_DML21 is defined.
There is no in-tree Kconfig symbol named DRM_AMD_DC_DML21, so the
preprocessor always removes the callback entry. Sibling DCN42 and DCN401
resource tables wire their prepare_mcache_programming callbacks
unconditionally, and the core DC code already checks whether the callback
pointer is present before calling it.
Remove the stale guard so DCN42B exposes the callback relation that its
source and DML21 build world already provide.
This is an RFC patch draft from static conditional callback legality
auditing. It needs AMD display maintainer review before submission as a
final fix.
Signed-off-by: Pengpeng Hou <pengpeng@iscas.ac.cn>
Reviewed-by: George Zhang <george.zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 85453fb4ff726e1ddb9984ee83dca260903c5353)
In dm_update_crtc_state(), the skip_modeset path releases new_stream
via dc_stream_release() but does not set the pointer to NULL.
If a later error (e.g., color management failure) triggers the fail
label, the error path calls dc_stream_release() again on the same
dangling pointer, causing a double release and potential use-after-free.
Fix this by setting new_stream to NULL after the initial release.
Fixes: 9b690ef3c7 ("drm/amd/display: Avoid full modeset when not required")
Signed-off-by: WenTao Liang <vulab@iscas.ac.cn>
Reviewed-by: George Zhang <george.zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 99f3af19073b3ddbfd96e789124cce12c4277b28)
Cc: stable@vger.kernel.org
The Lenovo Legion 5 15ARH05 (Renoir) ships a BOE 0x08DF eDP panel that
advertises AUX/DPCD backlight control, so amdgpu's automatic detection
(amdgpu_backlight == -1) selects AUX. On this panel the AUX backlight
path has no effect: brightness writes are accepted but the panel level
never changes, the display is stuck at a fixed brightness and
max_brightness is reported as a bogus 511000. As a result neither the
desktop brightness slider nor the brightness hotkeys do anything.
Forcing PWM backlight (amdgpu.backlight=0) restores working control:
max_brightness becomes 65535 and the level tracks writes. This has long
been applied by users as a manual kernel-parameter workaround.
Extend the generic panel backlight quirk with a force_pwm flag, add an
entry for the Legion 5 15ARH05 / BOE 0x08DF panel, and have amdgpu
disable AUX backlight (use PWM) when the quirk matches and the user
lets the driver auto-select the backlight type.
Signed-off-by: Alessandro Rinaldi <ale@alerinaldi.it>
Tested-by: Alessandro Rinaldi <ale@alerinaldi.it>
Reviewed-by: George Zhang <george.zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 81b39f43e7e53589491e2eef6bad5389626b4b9c)
Cc: stable@vger.kernel.org
The change referenced by the Fixes tag releases the HIQ SDMA MQD trunk
buffer when device_queue_manager_init() fails after it has been
allocated.
However, the same failure path can also be reached after
init_mqd_managers() has succeeded. At that point dqm->mqd_mgrs[] contains
per-type MQD manager objects owned by the device queue manager. The
normal teardown path frees those objects from uninitialize(), but the
initialization error path only frees dqm itself.
Free the MQD managers from the initialization error path as well. This is
safe for earlier failures because dqm is zeroed when allocated and
init_mqd_managers() clears the entries it rolls back internally.
Fixes: b7cccc8286 ("drm/amdkfd: fix a memory leak in device_queue_manager_init()")
Signed-off-by: Guangshuo Li <lgs201920130244@gmail.com>
Reviewed-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 1fff2e07b6670bc5b8f7344a8708c136259cb176)
Cc: stable@vger.kernel.org
Allow using multiple SDMA schedulers only on GPUs where
we are allowed to do concurrent VM flushes.
This consideration is necessary because all GART windows
are mapped in VMID 0 (the kernel VMID) so each buffer
entity would flush VMID 0 concurrently.
Practically this means that we can't use multiple SDMA
engines for TTM on GFX6-8 and Navi 1x.
Fixes: 01c836788b ("drm/amdgpu: pass all the sdma scheds to amdgpu_mman")
Fixes: e4029f7a94 ("drm/amdgpu: only use working sdma schedulers for ttm")
Cc: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit a8171229bc836607fbc225d323ebc4d14489cfbb)
There were two mistakes in the previous implementation:
The check for AutomaticDCTransition should be inverted.
We recently learned that the kernel should send
PPSMC_MSG_RunningOnAC when the flag is set, and not the
other way around.
The clocks also need to be recomputed, because the code in
the smu7_apply_state_adjust_rules() function selects
different limits on AC and DC.
Fixes: 96da0d8661 ("drm/amd/pm/smu7: Notify SMU7 of DC->AC switch")
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 516f8fc30a1b56af03f39e93c18707d13419fb1f)
Cc: stable@vger.kernel.org
AMD Ryzen Pinnacle Ridge (Zen+, family 0x17 model 0x08) CPUs have
PCI controllers that don't support PCIe dynamic speed switching,
causing system freezes during GPU initialization when enabled.
Disable dynamic speed switching when this CPU is detected.
Assisted-by: Claude:sonnet
Fixes: 466a7d1153 ("drm/amd: Use the first non-dGPU PCI device for BW limits")
Closes: https://gitlab.freedesktop.org/drm/amd/-/work_items/5436
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Link: https://patch.msgid.link/20260709031520.841611-1-mario.limonciello@amd.com
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 9ceb4e034a327a04155f32f1cd1a5031dfa5fe02)
Cc: stable@vger.kernel.org
We need the fence to reemit the gds switch or spm update
after a queue reset.
Fixes: a17ef94121 ("drm/amdgpu: rework ring reset backup and reemit v9")
Cc: timur.kristof@gmail.com
Cc: christian.koenig@amd.com
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit bc639a9eadc75822f7f15a4315c198a4b5513bd2)
Cc: stable@vger.kernel.org
There were two mistakes in the previous implementation:
The check for ATOM_PP_PLATFORM_CAP_HARDWAREDC should be
inverted. We recently learned that the kernel should send
PPSMC_MSG_RunningOnAC when the flag is set, and not the
other way around.
The clocks also need to be recomputed, because the code in
the si_apply_state_adjust_rules() function selects different
limits on AC and DC.
Fixes: 2d071f6457 ("drm/amd/pm/si: Notify the SMC when switching to AC")
Tested-by: Jeremy Klarenbeek <jeremy.klarenbeek99@gmail.com>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 358dd0a9ce66d898fa934887385327547d599d88)
Cc: stable@vger.kernel.org
When DPM is turned off with the amdgpu.dpm=0 module parameter,
the thermal work queue isn't initialized so we shouldn't
schedule any work on it.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit bd018d36171a695952c6d391471c279c9e05c8b2)
On DCE8-class ASICs (e.g. Bonaire), the resource pool contains digital
DIG stream encoders plus one analog DAC encoder. When assigning a stream
encoder for a second DisplayPort MST stream, if the preferred digital
encoder is already acquired, dce100_find_first_free_match_stream_enc_for_link()
falls back to the first free pool entry. That entry may be the analog
encoder, whose funcs table lacks DP hooks such as dp_set_stream_attribute.
The subsequent atomic commit then dereferences NULL function pointers in
link_set_dpms_on() and crashes.
Skip encoders without dp_set_stream_attribute when the stream uses a DP
signal (including MST). Use dc_is_dp_signal(stream->signal) for the MST
fallback path instead of checking only the link connector signal.
Tested on:
- GPU: AMD Radeon R7 260X (Bonaire / DCE8)
- Board: Supermicro C9X299-PG300
- Setup: DP MST daisy chain, hotplug second monitor or have it connected on boot
- Kernel: 7.1.3 (issue observed since 6.19)
- Result: kernel oops without patch; dual monitors stable with patch
Signed-off-by: Andriy Korud <a.korud@gmail.com>
Closes: https://gitlab.freedesktop.org/drm/amd/-/work_items/5162
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 28ec64943e3ee4d9b8d30cea61e380f1429953a8)
Cc: stable@vger.kernel.org
Always set native cursor mode when the CRTC is disabled,
to make sure it doesn't cause atomic commits to fail when
they are trying to disable the CRTC.
Fixes: 41af6215cd ("drm/amd/display: Reject cursor plane on DCE when scaled differently than primary")
Closes: https://gitlab.freedesktop.org/drm/amd/-/work_items/5432
Cc: Leo Li <sunpeng.li@amd.com>
Cc: Michel Dänzer <michel.daenzer@mailbox.org>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Tested-by: Viktor Jägersküpper <viktor_jaegerskuepper@freenet.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 2f79f0130f828cf26fe2dcf45291821616af7b47)
Cc: stable@vger.kernel.org
The old radeon driver has a documented workaround in ci_dpm.c
which claims that Bonaire 0x6658 with old memory controller
firmware is unstable with MCLK DPM, so as a precaution I
disabled MCLK DPM on this ASIC in amdgpu.
Note that the old MC firmware is not actually used with
amdgpu, but in theory it's possible that the VBIOS sets
up the ASIC with an old MC firmware that is already running
when amdgpu initializes (in which case amdgpu doesn't
load its own firmware).
What I expected to happen is that the GPU would simply use
its maximum memory clock, and indeed this is what seemed
to happen according to amdgpu_pm_info which reads the
current MCLK value from the SMU.
However, some users reported a huge perf regression
and upon a closer look it seems that the GPU seems to
not actually use the highest MCLK value, despite the SMU
reporting that it does.
Let's not disable MCLK DPM on Bonaire 0x6658 (R7 260X).
Keep MCLK DPM disabled on R9 M380 in the 2015 iMac
because that still hangs if we enable it.
Fixes: 9851f29cb0 ("drm/amd/pm/ci: Disable MCLK DPM on problematic CI ASICs")
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit d34acad064ee7d82bd18f5d87592c422d4d323ac)
Cc: stable@vger.kernel.org
When compiling the AMDGPU display driver for 32-bit architectures,
the linker reports undefined reference to `__udivdi3` in functions
get_dp_dto_frequency_100hz() and dcn401_get_dp_dto_frequency_100hz().
This is because the code uses 64-bit division (/) on 32-bit systems,
which GCC cannot handle directly and instead tries to call the missing
__udivdi3 helper function.
Replace the raw division with div_u64(), the kernel's standard 64-bit
division helper, to avoid the link error.
Signed-off-by: Linlin Yang <yanglinlin@kylinos.cn>
Reported-by: k2ci <kernel-bot@kylinos.cn>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 0421fc6ab3a8514e99156ff3c2cee13ee9af3fa7)
Cc: stable@vger.kernel.org
Currently the contents of IBs are abruptly cut off and don't
show the full contents. This patch makes sure to reserve
space for those contents too so they may be printed.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 4e2c0821509fed754e8c31d5053d152fbb3484a5)
Cc: stable@vger.kernel.org
These are in the dmesg logs but are missing from devcoredumps.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit fed7aa36d79802c3e02acd05aeae8b0a877e47c2)
Cc: stable@vger.kernel.org
amdgpu_acpi_vfct_bios() fetches the VFCT table with acpi_get_table()
but never releases it. acpi_get_table() takes a reference on the
table (incrementing its validation_count and mapping it on the 0->1
transition); without a paired acpi_put_table() the mapping is leaked
on every call, whether or not a matching VBIOS image is found.
Route all exit paths after the table is acquired through a common
acpi_put_table(). The VBIOS image is copied out with kmemdup() before
the table is released, so it remains valid for the caller.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Link: https://patch.msgid.link/20260708193518.702584-3-mario.limonciello@amd.com
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit ca5988682b4cba4cd125a0fa99b2de1239164ae4)
Cc: stable@vger.kernel.org
