| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
[ Upstream commit bfb4fb77f9a8ce33ce357224569eae5564eec573 ]
syszbot reports various ordering issues for lower instance locks and
team lock. Switch to using rtnl lock for protecting team device,
similar to bonding. Based on the patch by Tetsuo Handa.
Cc: Jiri Pirko <jiri@resnulli.us>
Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Reported-by: syzbot+705c61d60b091ef42c04@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=705c61d60b091ef42c04
Reported-by: syzbot+71fd22ae4b81631e22fd@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=71fd22ae4b81631e22fd
Fixes: 6b1d3c5f675c ("team: grab team lock during team_change_rx_flags")
Link: https://lkml.kernel.org/r/ZoZ2RH9BcahEB9Sb@nanopsycho.orion
Signed-off-by: Stanislav Fomichev <sdf@fomichev.me>
Link: https://patch.msgid.link/20250623153147.3413631-1-sdf@fomichev.me
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 95cbab48782bf62e4093837dc15ac6133902c12f ]
drm_panthor_gpu_info::shader_present is currently automatically offset
by 4 byte to meet Arm's 32-bit/64-bit field alignment rules, but those
constraints don't stand on 32-bit x86 and cause a mismatch when running
an x86 binary in a user emulated environment like FEX. It's also
generally agreed that uAPIs should explicitly pad their struct fields,
which we originally intended to do, but a mistake slipped through during
the submission process, leading drm_panthor_gpu_info::shader_present to
be misaligned.
This uAPI change doesn't break any of the existing users of panthor
which are either arm32 or arm64 where the 64-bit alignment of
u64 fields is already enforced a the compiler level.
Changes in v2:
- Rename the garbage field into pad0 and adjust the comment accordingly
- Add Liviu's A-b
Changes in v3:
- Add R-bs
Fixes: 0f25e493a246 ("drm/panthor: Add uAPI")
Acked-by: Liviu Dudau <liviu.dudau@arm.com>
Reviewed-by: Adrián Larumbe <adrian.larumbe@collabora.com>
Reviewed-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250606080932.4140010-2-boris.brezillon@collabora.com
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 12c409aa1ec2592280a2ddcc66ff8f3c7f7bb171 ]
Because pps_cdev_poll() returns unconditionally EPOLLIN,
a user space program that calls select/poll get always an immediate data
ready-to-read response. As a result the intended use to wait until next
data becomes ready does not work.
User space snippet:
struct pollfd pollfd = {
.fd = open("/dev/pps0", O_RDONLY),
.events = POLLIN|POLLERR,
.revents = 0 };
while(1) {
poll(&pollfd, 1, 2000/*ms*/); // returns immediate, but should wait
if(revents & EPOLLIN) { // always true
struct pps_fdata fdata;
memset(&fdata, 0, sizeof(memdata));
ioctl(PPS_FETCH, &fdata); // currently fetches data at max speed
}
}
Lets remember the last fetch event counter and compare this value
in pps_cdev_poll() with most recent event counter
and return 0 if they are equal.
Signed-off-by: Denis OSTERLAND-HEIM <denis.osterland@diehl.com>
Co-developed-by: Rodolfo Giometti <giometti@enneenne.com>
Signed-off-by: Rodolfo Giometti <giometti@enneenne.com>
Fixes: eae9d2ba0cfc ("LinuxPPS: core support")
Link: https://lore.kernel.org/all/f6bed779-6d59-4f0f-8a59-b6312bd83b4e@enneenne.com/
Acked-by: Rodolfo Giometti <giometti@enneenne.com>
Link: https://lore.kernel.org/r/c3c50ad1eb19ef553eca8a57c17f4c006413ab70.camel@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit ffaf1bf3737f706e4e9be876de4bc3c8fc578091 ]
The macro takes a parameter called "p" but references "fc" internally.
This happens to compile as long as callers pass a variable named fc,
but breaks otherwise. Rename the first parameter to “fc” to match the
usage and to be consistent with warnfc() / errorfc().
Fixes: a3ff937b33d9 ("prefix-handling analogues of errorf() and friends")
Signed-off-by: RubenKelevra <rubenkelevra@gmail.com>
Link: https://lore.kernel.org/20250617230927.1790401-1-rubenkelevra@gmail.com
Signed-off-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit ae1ae11fb277f1335d6bcd4935ba0ea985af3c32 ]
The move of the module sanity check to earlier skipped the audit logging
call in the case of failure and to a place where the previously used
context is unavailable.
Add an audit logging call for the module loading failure case and get
the module name when possible.
Link: https://issues.redhat.com/browse/RHEL-52839
Fixes: 02da2cbab452 ("module: move check_modinfo() early to early_mod_check()")
Signed-off-by: Richard Guy Briggs <rgb@redhat.com>
Reviewed-by: Petr Pavlu <petr.pavlu@suse.com>
Signed-off-by: Paul Moore <paul@paul-moore.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
commit 0dec7201788b9152f06321d0dab46eed93834cda upstream.
In file included from drivers/crypto/intel/qat/qat_common/adf_pm_dbgfs_utils.c:4:
include/linux/sprintf.h:11:54: error: unknown type name 'va_list'
11 | __printf(2, 0) int vsprintf(char *buf, const char *, va_list);
| ^~~~~~~
include/linux/sprintf.h:1:1: note: 'va_list' is defined in header '<stdarg.h>'; this is probably fixable by adding '#include <stdarg.h>'
Link: https://lkml.kernel.org/r/20250721173754.42865913@canb.auug.org.au
Fixes: 39ced19b9e60 ("lib/vsprintf: split out sprintf() and friends")
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Andriy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: Petr Mladek <pmladek@suse.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Cc: Sergey Senozhatsky <senozhatsky@chromium.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 95a16160ca1d75c66bf7a1c5e0bcaffb18e7c7fc upstream.
- Added a handler in DRM buddy manager to reset the cleared
flag for the blocks in the freelist.
- This is necessary because, upon resuming, the VRAM becomes
cluttered with BIOS data, yet the VRAM backend manager
believes that everything has been cleared.
v2:
- Add lock before accessing drm_buddy_clear_reset_blocks()(Matthew Auld)
- Force merge the two dirty blocks.(Matthew Auld)
- Add a new unit test case for this issue.(Matthew Auld)
- Having this function being able to flip the state either way would be
good. (Matthew Brost)
v3(Matthew Auld):
- Do merge step first to avoid the use of extra reset flag.
Signed-off-by: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Suggested-by: Christian König <christian.koenig@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Cc: stable@vger.kernel.org
Fixes: a68c7eaa7a8f ("drm/amdgpu: Enable clear page functionality")
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3812
Signed-off-by: Christian König <christian.koenig@amd.com>
Link: https://lore.kernel.org/r/20250716075125.240637-2-Arunpravin.PaneerSelvam@amd.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
[ Upstream commit 897e8601b9cff1d054cdd53047f568b0e1995726 ]
The s390x ISM device data sheet clearly states that only one
request-response sequence is allowable per ISM function at any point in
time. Unfortunately as of today the s390/ism driver in Linux does not
honor that requirement. This patch aims to rectify that.
This problem was discovered based on Aliaksei's bug report which states
that for certain workloads the ISM functions end up entering error state
(with PEC 2 as seen from the logs) after a while and as a consequence
connections handled by the respective function break, and for future
connection requests the ISM device is not considered -- given it is in a
dysfunctional state. During further debugging PEC 3A was observed as
well.
A kernel message like
[ 1211.244319] zpci: 061a:00:00.0: Event 0x2 reports an error for PCI function 0x61a
is a reliable indicator of the stated function entering error state
with PEC 2. Let me also point out that a kernel message like
[ 1211.244325] zpci: 061a:00:00.0: The ism driver bound to the device does not support error recovery
is a reliable indicator that the ISM function won't be auto-recovered
because the ISM driver currently lacks support for it.
On a technical level, without this synchronization, commands (inputs to
the FW) may be partially or fully overwritten (corrupted) by another CPU
trying to issue commands on the same function. There is hard evidence that
this can lead to DMB token values being used as DMB IOVAs, leading to
PEC 2 PCI events indicating invalid DMA. But this is only one of the
failure modes imaginable. In theory even completely losing one command
and executing another one twice and then trying to interpret the outputs
as if the command we intended to execute was actually executed and not
the other one is also possible. Frankly, I don't feel confident about
providing an exhaustive list of possible consequences.
Fixes: 684b89bc39ce ("s390/ism: add device driver for internal shared memory")
Reported-by: Aliaksei Makarau <Aliaksei.Makarau@ibm.com>
Tested-by: Mahanta Jambigi <mjambigi@linux.ibm.com>
Tested-by: Aliaksei Makarau <Aliaksei.Makarau@ibm.com>
Signed-off-by: Halil Pasic <pasic@linux.ibm.com>
Reviewed-by: Alexandra Winter <wintera@linux.ibm.com>
Signed-off-by: Alexandra Winter <wintera@linux.ibm.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250722161817.1298473-1-wintera@linux.ibm.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit c0f21029f123d1b15f8eddc8e3976bf0c8781c43 ]
Offload path is used for GRO with SW IPsec, and not just for HW
offload. So initialize it anyway.
Fixes: 585b64f5a620 ("xfrm: delay initialization of offload path till its actually requested")
Reported-by: Sabrina Dubroca <sd@queasysnail.net>
Closes: https://lore.kernel.org/all/aEGW_5HfPqU1rFjl@krikkit
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 2fd895842d49c23137ae48252dd211e5d6d8a3ed ]
When a call is released, rxrpc takes the spinlock and removes it from
->recvmsg_q in an effort to prevent racing recvmsg() invocations from
seeing the same call. Now, rxrpc_recvmsg() only takes the spinlock when
actually removing a call from the queue; it doesn't, however, take it in
the lead up to that when it checks to see if the queue is empty. It *does*
hold the socket lock, which prevents a recvmsg/recvmsg race - but this
doesn't prevent sendmsg from ending the call because sendmsg() drops the
socket lock and relies on the call->user_mutex.
Fix this by firstly removing the bit in rxrpc_release_call() that dequeues
the released call and, instead, rely on recvmsg() to simply discard
released calls (done in a preceding fix).
Secondly, rxrpc_notify_socket() is abandoned if the call is already marked
as released rather than trying to be clever by setting both pointers in
call->recvmsg_link to NULL to trick list_empty(). This isn't perfect and
can still race, resulting in a released call on the queue, but recvmsg()
will now clean that up.
Fixes: 17926a79320a ("[AF_RXRPC]: Provide secure RxRPC sockets for use by userspace and kernel both")
Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: Jeffrey Altman <jaltman@auristor.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: Junvyyang, Tencent Zhuque Lab <zhuque@tencent.com>
cc: LePremierHomme <kwqcheii@proton.me>
cc: Simon Horman <horms@kernel.org>
cc: linux-afs@lists.infradead.org
Link: https://patch.msgid.link/20250717074350.3767366-4-dhowells@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 962fb1f651c2cf2083e0c3ef53ba69e3b96d3fbc ]
If a call receives an event (such as incoming data), the call gets placed
on the socket's queue and a thread in recvmsg can be awakened to go and
process it. Once the thread has picked up the call off of the queue,
further events will cause it to be requeued, and once the socket lock is
dropped (recvmsg uses call->user_mutex to allow the socket to be used in
parallel), a second thread can come in and its recvmsg can pop the call off
the socket queue again.
In such a case, the first thread will be receiving stuff from the call and
the second thread will be blocked on call->user_mutex. The first thread
can, at this point, process both the event that it picked call for and the
event that the second thread picked the call for and may see the call
terminate - in which case the call will be "released", decoupling the call
from the user call ID assigned to it (RXRPC_USER_CALL_ID in the control
message).
The first thread will return okay, but then the second thread will wake up
holding the user_mutex and, if it sees that the call has been released by
the first thread, it will BUG thusly:
kernel BUG at net/rxrpc/recvmsg.c:474!
Fix this by just dequeuing the call and ignoring it if it is seen to be
already released. We can't tell userspace about it anyway as the user call
ID has become stale.
Fixes: 248f219cb8bc ("rxrpc: Rewrite the data and ack handling code")
Reported-by: Junvyyang, Tencent Zhuque Lab <zhuque@tencent.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: Jeffrey Altman <jaltman@auristor.com>
cc: LePremierHomme <kwqcheii@proton.me>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: Simon Horman <horms@kernel.org>
cc: linux-afs@lists.infradead.org
Link: https://patch.msgid.link/20250717074350.3767366-3-dhowells@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 2d72afb340657f03f7261e9243b44457a9228ac7 ]
A crash in conntrack was reported while trying to unlink the conntrack
entry from the hash bucket list:
[exception RIP: __nf_ct_delete_from_lists+172]
[..]
#7 [ff539b5a2b043aa0] nf_ct_delete at ffffffffc124d421 [nf_conntrack]
#8 [ff539b5a2b043ad0] nf_ct_gc_expired at ffffffffc124d999 [nf_conntrack]
#9 [ff539b5a2b043ae0] __nf_conntrack_find_get at ffffffffc124efbc [nf_conntrack]
[..]
The nf_conn struct is marked as allocated from slab but appears to be in
a partially initialised state:
ct hlist pointer is garbage; looks like the ct hash value
(hence crash).
ct->status is equal to IPS_CONFIRMED|IPS_DYING, which is expected
ct->timeout is 30000 (=30s), which is unexpected.
Everything else looks like normal udp conntrack entry. If we ignore
ct->status and pretend its 0, the entry matches those that are newly
allocated but not yet inserted into the hash:
- ct hlist pointers are overloaded and store/cache the raw tuple hash
- ct->timeout matches the relative time expected for a new udp flow
rather than the absolute 'jiffies' value.
If it were not for the presence of IPS_CONFIRMED,
__nf_conntrack_find_get() would have skipped the entry.
Theory is that we did hit following race:
cpu x cpu y cpu z
found entry E found entry E
E is expired <preemption>
nf_ct_delete()
return E to rcu slab
init_conntrack
E is re-inited,
ct->status set to 0
reply tuplehash hnnode.pprev
stores hash value.
cpu y found E right before it was deleted on cpu x.
E is now re-inited on cpu z. cpu y was preempted before
checking for expiry and/or confirm bit.
->refcnt set to 1
E now owned by skb
->timeout set to 30000
If cpu y were to resume now, it would observe E as
expired but would skip E due to missing CONFIRMED bit.
nf_conntrack_confirm gets called
sets: ct->status |= CONFIRMED
This is wrong: E is not yet added
to hashtable.
cpu y resumes, it observes E as expired but CONFIRMED:
<resumes>
nf_ct_expired()
-> yes (ct->timeout is 30s)
confirmed bit set.
cpu y will try to delete E from the hashtable:
nf_ct_delete() -> set DYING bit
__nf_ct_delete_from_lists
Even this scenario doesn't guarantee a crash:
cpu z still holds the table bucket lock(s) so y blocks:
wait for spinlock held by z
CONFIRMED is set but there is no
guarantee ct will be added to hash:
"chaintoolong" or "clash resolution"
logic both skip the insert step.
reply hnnode.pprev still stores the
hash value.
unlocks spinlock
return NF_DROP
<unblocks, then
crashes on hlist_nulls_del_rcu pprev>
In case CPU z does insert the entry into the hashtable, cpu y will unlink
E again right away but no crash occurs.
Without 'cpu y' race, 'garbage' hlist is of no consequence:
ct refcnt remains at 1, eventually skb will be free'd and E gets
destroyed via: nf_conntrack_put -> nf_conntrack_destroy -> nf_ct_destroy.
To resolve this, move the IPS_CONFIRMED assignment after the table
insertion but before the unlock.
Pablo points out that the confirm-bit-store could be reordered to happen
before hlist add resp. the timeout fixup, so switch to set_bit and
before_atomic memory barrier to prevent this.
It doesn't matter if other CPUs can observe a newly inserted entry right
before the CONFIRMED bit was set:
Such event cannot be distinguished from above "E is the old incarnation"
case: the entry will be skipped.
Also change nf_ct_should_gc() to first check the confirmed bit.
The gc sequence is:
1. Check if entry has expired, if not skip to next entry
2. Obtain a reference to the expired entry.
3. Call nf_ct_should_gc() to double-check step 1.
nf_ct_should_gc() is thus called only for entries that already failed an
expiry check. After this patch, once the confirmed bit check passes
ct->timeout has been altered to reflect the absolute 'best before' date
instead of a relative time. Step 3 will therefore not remove the entry.
Without this change to nf_ct_should_gc() we could still get this sequence:
1. Check if entry has expired.
2. Obtain a reference.
3. Call nf_ct_should_gc() to double-check step 1:
4 - entry is still observed as expired
5 - meanwhile, ct->timeout is corrected to absolute value on other CPU
and confirm bit gets set
6 - confirm bit is seen
7 - valid entry is removed again
First do check 6), then 4) so the gc expiry check always picks up either
confirmed bit unset (entry gets skipped) or expiry re-check failure for
re-inited conntrack objects.
This change cannot be backported to releases before 5.19. Without
commit 8a75a2c17410 ("netfilter: conntrack: remove unconfirmed list")
|= IPS_CONFIRMED line cannot be moved without further changes.
Cc: Razvan Cojocaru <rzvncj@gmail.com>
Link: https://lore.kernel.org/netfilter-devel/20250627142758.25664-1-fw@strlen.de/
Link: https://lore.kernel.org/netfilter-devel/4239da15-83ff-4ca4-939d-faef283471bb@gmail.com/
Fixes: 1397af5bfd7d ("netfilter: conntrack: remove the percpu dying list")
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 6851a0c228fc040dce8e4c393004209e7372e0a3 ]
The 'quirks' member already ran out of bits on some platforms some time
ago. Replace the integer member by a bitmap in order to have enough bits
in future. Replace raw bit operations by accessor macros.
Fixes: ff26b2dd6568 ("Bluetooth: Add quirk for broken READ_VOICE_SETTING")
Fixes: 127881334eaa ("Bluetooth: Add quirk for broken READ_PAGE_SCAN_TYPE")
Suggested-by: Pauli Virtanen <pav@iki.fi>
Tested-by: Ivan Pravdin <ipravdin.official@gmail.com>
Signed-off-by: Kiran K <kiran.k@intel.com>
Signed-off-by: Christian Eggers <ceggers@arri.de>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit cdee6a4416b2a57c89082929cc60e2275bb32a3a ]
Macro parameters should always be put into braces when accessing it.
Fixes: 4fc9857ab8c6 ("Bluetooth: hci_sync: Add check simultaneous roles support")
Signed-off-by: Christian Eggers <ceggers@arri.de>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit dfef8d87a031ac1a46dde3de804e0fcf3c3a6afd ]
The provided macro parameter is named 'dev' (rather than 'hdev', which
may be a variable on the stack where the macro is used).
Fixes: a9a830a676a9 ("Bluetooth: hci_event: Fix sending HCI_OP_READ_ENC_KEY_SIZE")
Fixes: 6126ffabba6b ("Bluetooth: Introduce HCI_CONN_FLAG_DEVICE_PRIVACY device flag")
Signed-off-by: Christian Eggers <ceggers@arri.de>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 444020f4bf06fb86805ee7e7ceec0375485fd94d ]
This reverts commit e3eac9f32ec0 ("wifi: cfg80211: Annotate struct
cfg80211_scan_request with __counted_by").
This really has been a completely failed experiment. There were
no actual bugs found, and yet at this point we already have four
"fixes" to it, with nothing to show for but code churn, and it
never even made the code any safer.
In all of the cases that ended up getting "fixed", the structure
is also internally inconsistent after the n_channels setting as
the channel list isn't actually filled yet. You cannot scan with
such a structure, that's just wrong. In mac80211, the struct is
also reused multiple times, so initializing it once is no good.
Some previous "fixes" (e.g. one in brcm80211) are also just setting
n_channels before accessing the array, under the assumption that the
code is correct and the array can be accessed, further showing that
the whole thing is just pointless when the allocation count and use
count are not separate.
If we really wanted to fix it, we'd need to separately track the
number of channels allocated and the number of channels currently
used, but given that no bugs were found despite the numerous syzbot
reports, that'd just be a waste of time.
Remove the __counted_by() annotation. We really should also remove
a number of the n_channels settings that are setting up a structure
that's inconsistent, but that can wait.
Reported-by: syzbot+e834e757bd9b3d3e1251@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=e834e757bd9b3d3e1251
Fixes: e3eac9f32ec0 ("wifi: cfg80211: Annotate struct cfg80211_scan_request with __counted_by")
Link: https://patch.msgid.link/20250714142130.9b0bbb7e1f07.I09112ccde72d445e11348fc2bef68942cb2ffc94@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit cf0233491b3a15933234a26efd9ecbc1c0764674 ]
If the PHY driver uses another PHY internally (e.g. in case of eUSB2,
repeaters are represented as PHYs), then it would trigger the following
lockdep splat because all PHYs use a single static lockdep key and thus
lockdep can not identify whether there is a dependency or not and
reports a false positive.
Make PHY subsystem use dynamic lockdep keys, assigning each driver a
separate key. This way lockdep can correctly identify dependency graph
between mutexes.
============================================
WARNING: possible recursive locking detected
6.15.0-rc7-next-20250522-12896-g3932f283970c #3455 Not tainted
--------------------------------------------
kworker/u51:0/78 is trying to acquire lock:
ffff0008116554f0 (&phy->mutex){+.+.}-{4:4}, at: phy_init+0x4c/0x12c
but task is already holding lock:
ffff000813c10cf0 (&phy->mutex){+.+.}-{4:4}, at: phy_init+0x4c/0x12c
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(&phy->mutex);
lock(&phy->mutex);
*** DEADLOCK ***
May be due to missing lock nesting notation
4 locks held by kworker/u51:0/78:
#0: ffff000800010948 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_one_work+0x18c/0x5ec
#1: ffff80008036bdb0 (deferred_probe_work){+.+.}-{0:0}, at: process_one_work+0x1b4/0x5ec
#2: ffff0008094ac8f8 (&dev->mutex){....}-{4:4}, at: __device_attach+0x38/0x188
#3: ffff000813c10cf0 (&phy->mutex){+.+.}-{4:4}, at: phy_init+0x4c/0x12c
stack backtrace:
CPU: 0 UID: 0 PID: 78 Comm: kworker/u51:0 Not tainted 6.15.0-rc7-next-20250522-12896-g3932f283970c #3455 PREEMPT
Hardware name: Qualcomm CRD, BIOS 6.0.240904.BOOT.MXF.2.4-00528.1-HAMOA-1 09/ 4/2024
Workqueue: events_unbound deferred_probe_work_func
Call trace:
show_stack+0x18/0x24 (C)
dump_stack_lvl+0x90/0xd0
dump_stack+0x18/0x24
print_deadlock_bug+0x258/0x348
__lock_acquire+0x10fc/0x1f84
lock_acquire+0x1c8/0x338
__mutex_lock+0xb8/0x59c
mutex_lock_nested+0x24/0x30
phy_init+0x4c/0x12c
snps_eusb2_hsphy_init+0x54/0x1a0
phy_init+0xe0/0x12c
dwc3_core_init+0x450/0x10b4
dwc3_core_probe+0xce4/0x15fc
dwc3_probe+0x64/0xb0
platform_probe+0x68/0xc4
really_probe+0xbc/0x298
__driver_probe_device+0x78/0x12c
driver_probe_device+0x3c/0x160
__device_attach_driver+0xb8/0x138
bus_for_each_drv+0x84/0xe0
__device_attach+0x9c/0x188
device_initial_probe+0x14/0x20
bus_probe_device+0xac/0xb0
deferred_probe_work_func+0x8c/0xc8
process_one_work+0x208/0x5ec
worker_thread+0x1c0/0x368
kthread+0x14c/0x20c
ret_from_fork+0x10/0x20
Fixes: 3584f6392f09 ("phy: qcom: phy-qcom-snps-eusb2: Add support for eUSB2 repeater")
Fixes: e2463559ff1d ("phy: amlogic: Add Amlogic AXG PCIE PHY Driver")
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Reviewed-by: Abel Vesa <abel.vesa@linaro.org>
Reported-by: Johan Hovold <johan+linaro@kernel.org>
Link: https://lore.kernel.org/lkml/ZnpoAVGJMG4Zu-Jw@hovoldconsulting.com/
Reviewed-by: Johan Hovold <johan+linaro@kernel.org>
Tested-by: Johan Hovold <johan+linaro@kernel.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20250605-phy-subinit-v3-1-1e1e849e10cd@oss.qualcomm.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
commit 89635eae076cd8eaa5cb752f66538c9dc6c9fdc3 upstream.
When netfslib is issuing subrequests, the subrequests start processing
immediately and may complete before we reach the end of the issuing
function. At the end of the issuing function we set NETFS_RREQ_ALL_QUEUED
to indicate to the collector that we aren't going to issue any more subreqs
and that it can do the final notifications and cleanup.
Now, this isn't a problem if the request is synchronous
(NETFS_RREQ_OFFLOAD_COLLECTION is unset) as the result collection will be
done in-thread and we're guaranteed an opportunity to run the collector.
However, if the request is asynchronous, collection is primarily triggered
by the termination of subrequests queuing it on a workqueue. Now, a race
can occur here if the app thread sets ALL_QUEUED after the last subrequest
terminates.
This can happen most easily with the copy2cache code (as used by Ceph)
where, in the collection routine of a read request, an asynchronous write
request is spawned to copy data to the cache. Folios are added to the
write request as they're unlocked, but there may be a delay before
ALL_QUEUED is set as the write subrequests may complete before we get
there.
If all the write subreqs have finished by the ALL_QUEUED point, no further
events happen and the collection never happens, leaving the request
hanging.
Fix this by queuing the collector after setting ALL_QUEUED. This is a bit
heavy-handed and it may be sufficient to do it only if there are no extant
subreqs.
Also add a tracepoint to cross-reference both requests in a copy-to-request
operation and add a trace to the netfs_rreq tracepoint to indicate the
setting of ALL_QUEUED.
Fixes: e2d46f2ec332 ("netfs: Change the read result collector to only use one work item")
Reported-by: Max Kellermann <max.kellermann@ionos.com>
Link: https://lore.kernel.org/r/CAKPOu+8z_ijTLHdiCYGU_Uk7yYD=shxyGLwfe-L7AV3DhebS3w@mail.gmail.com/
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/20250711151005.2956810-3-dhowells@redhat.com
Reviewed-by: Paulo Alcantara (Red Hat) <pc@manguebit.org>
cc: Paulo Alcantara <pc@manguebit.org>
cc: Viacheslav Dubeyko <slava@dubeyko.com>
cc: Alex Markuze <amarkuze@redhat.com>
cc: Ilya Dryomov <idryomov@gmail.com>
cc: netfs@lists.linux.dev
cc: ceph-devel@vger.kernel.org
cc: linux-fsdevel@vger.kernel.org
cc: stable@vger.kernel.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
[ Upstream commit 18cdb3d982da8976b28d57691eb256ec5688fad2 ]
syzbot found a potential access to uninit-value in nf_flow_pppoe_proto()
Blamed commit forgot the Ethernet header.
BUG: KMSAN: uninit-value in nf_flow_offload_inet_hook+0x7e4/0x940 net/netfilter/nf_flow_table_inet.c:27
nf_flow_offload_inet_hook+0x7e4/0x940 net/netfilter/nf_flow_table_inet.c:27
nf_hook_entry_hookfn include/linux/netfilter.h:157 [inline]
nf_hook_slow+0xe1/0x3d0 net/netfilter/core.c:623
nf_hook_ingress include/linux/netfilter_netdev.h:34 [inline]
nf_ingress net/core/dev.c:5742 [inline]
__netif_receive_skb_core+0x4aff/0x70c0 net/core/dev.c:5837
__netif_receive_skb_one_core net/core/dev.c:5975 [inline]
__netif_receive_skb+0xcc/0xac0 net/core/dev.c:6090
netif_receive_skb_internal net/core/dev.c:6176 [inline]
netif_receive_skb+0x57/0x630 net/core/dev.c:6235
tun_rx_batched+0x1df/0x980 drivers/net/tun.c:1485
tun_get_user+0x4ee0/0x6b40 drivers/net/tun.c:1938
tun_chr_write_iter+0x3e9/0x5c0 drivers/net/tun.c:1984
new_sync_write fs/read_write.c:593 [inline]
vfs_write+0xb4b/0x1580 fs/read_write.c:686
ksys_write fs/read_write.c:738 [inline]
__do_sys_write fs/read_write.c:749 [inline]
Reported-by: syzbot+bf6ed459397e307c3ad2@syzkaller.appspotmail.com
Closes: https://lore.kernel.org/netdev/686bc073.a00a0220.c7b3.0086.GAE@google.com/T/#u
Fixes: 87b3593bed18 ("netfilter: flowtable: validate pppoe header")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Pablo Neira Ayuso <pablo@netfilter.org>
Link: https://patch.msgid.link/20250707124517.614489-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 4eb56b0761e75034dd35067a81da4c280c178262 ]
- trace_erofs_readpages => trace_erofs_readahead;
- Rename a redundant statement `nrpages = readahead_count(rac);`;
- Move the tracepoint to the beginning of z_erofs_readahead().
Signed-off-by: Gao Xiang <hsiangkao@linux.alibaba.com>
Reviewed-by: Hongbo Li <lihongbo22@huawei.com>
Link: https://lore.kernel.org/r/20250514120820.2739288-1-hsiangkao@linux.alibaba.com
Signed-off-by: Gao Xiang <hsiangkao@linux.alibaba.com>
Stable-dep-of: d53238b614e0 ("erofs: fix to add missing tracepoint in erofs_readahead()")
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 4cdf1bdd45ac78a088773722f009883af30ad318 ]
If THP is disabled and when a block device with logical block size >
page size is present, the following null ptr deref panic happens during
boot:
[ [13.2 mK AOSAN: null-ptr-deref in range [0x0000000000000000-0x0000000000K0 0 0[07]
[ 13.017749] RIP: 0010:create_empty_buffers+0x3b/0x380
<snip>
[ 13.025448] Call Trace:
[ 13.025692] <TASK>
[ 13.025895] block_read_full_folio+0x610/0x780
[ 13.026379] ? __pfx_blkdev_get_block+0x10/0x10
[ 13.027008] ? __folio_batch_add_and_move+0x1fa/0x2b0
[ 13.027548] ? __pfx_blkdev_read_folio+0x10/0x10
[ 13.028080] filemap_read_folio+0x9b/0x200
[ 13.028526] ? __pfx_filemap_read_folio+0x10/0x10
[ 13.029030] ? __filemap_get_folio+0x43/0x620
[ 13.029497] do_read_cache_folio+0x155/0x3b0
[ 13.029962] ? __pfx_blkdev_read_folio+0x10/0x10
[ 13.030381] read_part_sector+0xb7/0x2a0
[ 13.030805] read_lba+0x174/0x2c0
<snip>
[ 13.045348] nvme_scan_ns+0x684/0x850 [nvme_core]
[ 13.045858] ? __pfx_nvme_scan_ns+0x10/0x10 [nvme_core]
[ 13.046414] ? _raw_spin_unlock+0x15/0x40
[ 13.046843] ? __switch_to+0x523/0x10a0
[ 13.047253] ? kvm_clock_get_cycles+0x14/0x30
[ 13.047742] ? __pfx_nvme_scan_ns_async+0x10/0x10 [nvme_core]
[ 13.048353] async_run_entry_fn+0x96/0x4f0
[ 13.048787] process_one_work+0x667/0x10a0
[ 13.049219] worker_thread+0x63c/0xf60
As large folio support depends on THP, only allow bs > ps block devices
if THP is enabled.
Fixes: 47dd67532303 ("block/bdev: lift block size restrictions to 64k")
Signed-off-by: Pankaj Raghav <p.raghav@samsung.com>
Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Link: https://lore.kernel.org/r/20250704092134.289491-1-p.raghav@samsung.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit c5fd399a24c8e2865524361f7dc4d4a6899be4f4 ]
mac80211 identifies a short beacon by the presence of the next
TBTT field, however the standard actually doesn't explicitly state that
the next TBTT can't be in a long beacon or even that it is required in
a short beacon - and as a result this validation does not work for all
vendor implementations.
The standard explicitly states that an S1G long beacon shall contain
the S1G beacon compatibility element as the first element in a beacon
transmitted at a TBTT that is not a TSBTT (Target Short Beacon
Transmission Time) as per IEEE80211-2024 11.1.3.10.1. This is validated
by 9.3.4.3 Table 9-76 which states that the S1G beacon compatibility
element is only allowed in the full set and is not allowed in the
minimum set of elements permitted for use within short beacons.
Correctly identify short beacons by the lack of an S1G beacon
compatibility element as the first element in an S1G beacon frame.
Fixes: 9eaffe5078ca ("cfg80211: convert S1G beacon to scan results")
Signed-off-by: Simon Wadsworth <simon@morsemicro.com>
Signed-off-by: Lachlan Hodges <lachlan.hodges@morsemicro.com>
Link: https://patch.msgid.link/20250701075541.162619-1-lachlan.hodges@morsemicro.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
commit 82241a83cd15aaaf28200a40ad1a8b480012edaf upstream.
On some large machines with a high number of CPUs running a 64K pagesize
kernel, we found that the 'RES' field is always 0 displayed by the top
command for some processes, which will cause a lot of confusion for users.
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
875525 root 20 0 12480 0 0 R 0.3 0.0 0:00.08 top
1 root 20 0 172800 0 0 S 0.0 0.0 0:04.52 systemd
The main reason is that the batch size of the percpu counter is quite
large on these machines, caching a significant percpu value, since
converting mm's rss stats into percpu_counter by commit f1a7941243c1 ("mm:
convert mm's rss stats into percpu_counter"). Intuitively, the batch
number should be optimized, but on some paths, performance may take
precedence over statistical accuracy. Therefore, introducing a new
interface to add the percpu statistical count and display it to users,
which can remove the confusion. In addition, this change is not expected
to be on a performance-critical path, so the modification should be
acceptable.
In addition, the 'mm->rss_stat' is updated by using add_mm_counter() and
dec/inc_mm_counter(), which are all wrappers around
percpu_counter_add_batch(). In percpu_counter_add_batch(), there is
percpu batch caching to avoid 'fbc->lock' contention. This patch changes
task_mem() and task_statm() to get the accurate mm counters under the
'fbc->lock', but this should not exacerbate kernel 'mm->rss_stat' lock
contention due to the percpu batch caching of the mm counters. The
following test also confirm the theoretical analysis.
I run the stress-ng that stresses anon page faults in 32 threads on my 32
cores machine, while simultaneously running a script that starts 32
threads to busy-loop pread each stress-ng thread's /proc/pid/status
interface. From the following data, I did not observe any obvious impact
of this patch on the stress-ng tests.
w/o patch:
stress-ng: info: [6848] 4,399,219,085,152 CPU Cycles 67.327 B/sec
stress-ng: info: [6848] 1,616,524,844,832 Instructions 24.740 B/sec (0.367 instr. per cycle)
stress-ng: info: [6848] 39,529,792 Page Faults Total 0.605 M/sec
stress-ng: info: [6848] 39,529,792 Page Faults Minor 0.605 M/sec
w/patch:
stress-ng: info: [2485] 4,462,440,381,856 CPU Cycles 68.382 B/sec
stress-ng: info: [2485] 1,615,101,503,296 Instructions 24.750 B/sec (0.362 instr. per cycle)
stress-ng: info: [2485] 39,439,232 Page Faults Total 0.604 M/sec
stress-ng: info: [2485] 39,439,232 Page Faults Minor 0.604 M/sec
On comparing a very simple app which just allocates & touches some
memory against v6.1 (which doesn't have f1a7941243c1) and latest Linus
tree (4c06e63b9203) I can see that on latest Linus tree the values for
VmRSS, RssAnon and RssFile from /proc/self/status are all zeroes while
they do report values on v6.1 and a Linus tree with this patch.
Link: https://lkml.kernel.org/r/f4586b17f66f97c174f7fd1f8647374fdb53de1c.1749119050.git.baolin.wang@linux.alibaba.com
Fixes: f1a7941243c1 ("mm: convert mm's rss stats into percpu_counter")
Signed-off-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Reviewed-by: Aboorva Devarajan <aboorvad@linux.ibm.com>
Tested-by: Aboorva Devarajan <aboorvad@linux.ibm.com>
Tested-by Donet Tom <donettom@linux.ibm.com>
Acked-by: Shakeel Butt <shakeel.butt@linux.dev>
Acked-by: SeongJae Park <sj@kernel.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Cc: David Hildenbrand <david@redhat.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit f6bfc9afc7510cb5e6fbe0a17c507917b0120280 upstream.
Acquire GEM handles in drm_framebuffer_init() and release them in
the corresponding drm_framebuffer_cleanup(). Ties the handle's
lifetime to the framebuffer. Not all GEM buffer objects have GEM
handles. If not set, no refcounting takes place. This is the case
for some fbdev emulation. This is not a problem as these GEM objects
do not use dma-bufs and drivers will not release them while fbdev
emulation is running. Framebuffer flags keep a bit per color plane
of which the framebuffer holds a GEM handle reference.
As all drivers use drm_framebuffer_init(), they will now all hold
dma-buf references as fixed in commit 5307dce878d4 ("drm/gem: Acquire
references on GEM handles for framebuffers").
In the GEM framebuffer helpers, restore the original ref counting
on buffer objects. As the helpers for handle refcounting are now
no longer called from outside the DRM core, unexport the symbols.
v3:
- don't mix internal flags with mode flags (Christian)
v2:
- track framebuffer handle refs by flag
- drop gma500 cleanup (Christian)
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Fixes: 5307dce878d4 ("drm/gem: Acquire references on GEM handles for framebuffers")
Reported-by: Bert Karwatzki <spasswolf@web.de>
Closes: https://lore.kernel.org/dri-devel/20250703115915.3096-1-spasswolf@web.de/
Tested-by: Bert Karwatzki <spasswolf@web.de>
Tested-by: Mario Limonciello <superm1@kernel.org>
Tested-by: Borislav Petkov (AMD) <bp@alien8.de>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Anusha Srivatsa <asrivats@redhat.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Maxime Ripard <mripard@kernel.org>
Cc: Sumit Semwal <sumit.semwal@linaro.org>
Cc: "Christian König" <christian.koenig@amd.com>
Cc: linux-media@vger.kernel.org
Cc: dri-devel@lists.freedesktop.org
Cc: linaro-mm-sig@lists.linaro.org
Cc: <stable@vger.kernel.org>
Reviewed-by: Christian König <christian.koenig@amd.com>
Link: https://lore.kernel.org/r/20250707131224.249496-1-tzimmermann@suse.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit bd46cece51a36ef088f22ef0416ac13b0a46d5b0 upstream.
Object creation is a careful dance where we must guarantee that the
object is fully constructed before it is visible to other threads, and
GEM buffer objects are no difference.
Final publishing happens by calling drm_gem_handle_create(). After
that the only allowed thing to do is call drm_gem_object_put() because
a concurrent call to the GEM_CLOSE ioctl with a correctly guessed id
(which is trivial since we have a linear allocator) can already tear
down the object again.
Luckily most drivers get this right, the very few exceptions I've
pinged the relevant maintainers for. Unfortunately we also need
drm_gem_handle_create() when creating additional handles for an
already existing object (e.g. GETFB ioctl or the various bo import
ioctl), and hence we cannot have a drm_gem_handle_create_and_put() as
the only exported function to stop these issues from happening.
Now unfortunately the implementation of drm_gem_handle_create() isn't
living up to standards: It does correctly finishe object
initialization at the global level, and hence is safe against a
concurrent tear down. But it also sets up the file-private aspects of
the handle, and that part goes wrong: We fully register the object in
the drm_file.object_idr before calling drm_vma_node_allow() or
obj->funcs->open, which opens up races against concurrent removal of
that handle in drm_gem_handle_delete().
Fix this with the usual two-stage approach of first reserving the
handle id, and then only registering the object after we've completed
the file-private setup.
Jacek reported this with a testcase of concurrently calling GEM_CLOSE
on a freshly-created object (which also destroys the object), but it
should be possible to hit this with just additional handles created
through import or GETFB without completed destroying the underlying
object with the concurrent GEM_CLOSE ioctl calls.
Note that the close-side of this race was fixed in f6cd7daecff5 ("drm:
Release driver references to handle before making it available
again"), which means a cool 9 years have passed until someone noticed
that we need to make this symmetry or there's still gaps left :-/
Without the 2-stage close approach we'd still have a race, therefore
that's an integral part of this bugfix.
More importantly, this means we can have NULL pointers behind
allocated id in our drm_file.object_idr. We need to check for that
now:
- drm_gem_handle_delete() checks for ERR_OR_NULL already
- drm_gem.c:object_lookup() also chekcs for NULL
- drm_gem_release() should never be called if there's another thread
still existing that could call into an IOCTL that creates a new
handle, so cannot race. For paranoia I added a NULL check to
drm_gem_object_release_handle() though.
- most drivers (etnaviv, i915, msm) are find because they use
idr_find(), which maps both ENOENT and NULL to NULL.
- drivers using idr_for_each_entry() should also be fine, because
idr_get_next does filter out NULL entries and continues the
iteration.
- The same holds for drm_show_memory_stats().
v2: Use drm_WARN_ON (Thomas)
Reported-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Tested-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: stable@vger.kernel.org
Cc: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Maxime Ripard <mripard@kernel.org>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: David Airlie <airlied@gmail.com>
Cc: Simona Vetter <simona@ffwll.ch>
Signed-off-by: Simona Vetter <simona.vetter@intel.com>
Signed-off-by: Simona Vetter <simona.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20250707151814.603897-1-simona.vetter@ffwll.ch
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 8af39ec5cf2be522c8eb43a3d8005ed59e4daaee upstream.
A small race exists between spsc_queue_push and the run-job worker, in
which spsc_queue_push may return not-first while the run-job worker has
already idled due to the job count being zero. If this race occurs, job
scheduling stops, leading to hangs while waiting on the job’s DMA
fences.
Seal this race by incrementing the job count before appending to the
SPSC queue.
This race was observed on a drm-tip 6.16-rc1 build with the Xe driver in
an SVM test case.
Fixes: 1b1f42d8fde4 ("drm: move amd_gpu_scheduler into common location")
Fixes: 27105db6c63a ("drm/amdgpu: Add SPSC queue to scheduler.")
Cc: stable@vger.kernel.org
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://lore.kernel.org/r/20250613212013.719312-1-matthew.brost@intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit fc582cd26e888b0652bc1494f252329453fd3b23 upstream.
syzbot reports that defer/local task_work adding via msg_ring can hit
a request that has been freed:
CPU: 1 UID: 0 PID: 19356 Comm: iou-wrk-19354 Not tainted 6.16.0-rc4-syzkaller-00108-g17bbde2e1716 #0 PREEMPT(full)
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 05/07/2025
Call Trace:
<TASK>
dump_stack_lvl+0x189/0x250 lib/dump_stack.c:120
print_address_description mm/kasan/report.c:408 [inline]
print_report+0xd2/0x2b0 mm/kasan/report.c:521
kasan_report+0x118/0x150 mm/kasan/report.c:634
io_req_local_work_add io_uring/io_uring.c:1184 [inline]
__io_req_task_work_add+0x589/0x950 io_uring/io_uring.c:1252
io_msg_remote_post io_uring/msg_ring.c:103 [inline]
io_msg_data_remote io_uring/msg_ring.c:133 [inline]
__io_msg_ring_data+0x820/0xaa0 io_uring/msg_ring.c:151
io_msg_ring_data io_uring/msg_ring.c:173 [inline]
io_msg_ring+0x134/0xa00 io_uring/msg_ring.c:314
__io_issue_sqe+0x17e/0x4b0 io_uring/io_uring.c:1739
io_issue_sqe+0x165/0xfd0 io_uring/io_uring.c:1762
io_wq_submit_work+0x6e9/0xb90 io_uring/io_uring.c:1874
io_worker_handle_work+0x7cd/0x1180 io_uring/io-wq.c:642
io_wq_worker+0x42f/0xeb0 io_uring/io-wq.c:696
ret_from_fork+0x3fc/0x770 arch/x86/kernel/process.c:148
ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:245
</TASK>
which is supposed to be safe with how requests are allocated. But msg
ring requests alloc and free on their own, and hence must defer freeing
to a sane time.
Add an rcu_head and use kfree_rcu() in both spots where requests are
freed. Only the one in io_msg_tw_complete() is strictly required as it
has been visible on the other ring, but use it consistently in the other
spot as well.
This should not cause any other issues outside of KASAN rightfully
complaining about it.
Link: https://lore.kernel.org/io-uring/686cd2ea.a00a0220.338033.0007.GAE@google.com/
Reported-by: syzbot+54cbbfb4db9145d26fc2@syzkaller.appspotmail.com
Cc: stable@vger.kernel.org
Fixes: 0617bb500bfa ("io_uring/msg_ring: improve handling of target CQE posting")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
