If komeda_pipeline_unbound_components() returns -EDEADLK,
it means that a deadlock happened in the locking context.
Currently, komeda is not dealing with the deadlock properly,producing the
following output when CONFIG_DEBUG_WW_MUTEX_SLOWPATH is enabled:
According to the call trace information,it can be located to be
WARN_ON(IS_ERR(c_st)) in the komeda_pipeline_unbound_components function;
Then follow the function.
komeda_pipeline_unbound_components
-> komeda_component_get_state_and_set_user
-> komeda_pipeline_get_state_and_set_crtc
-> komeda_pipeline_get_state
->drm_atomic_get_private_obj_state
-> drm_atomic_get_private_obj_state
-> drm_modeset_lock
ret = drm_modeset_lock(&obj->lock, state->acquire_ctx); if (ret)
return ERR_PTR(ret);
Here it return -EDEADLK.
deal with the deadlock as suggested by [1], using the
function drm_modeset_backoff().
[1] https://docs.kernel.org/gpu/drm-kms.html?highlight=kms#kms-locking
Therefore, handling this problem can be solved
by adding return -EDEADLK back to the drm_modeset_backoff processing flow
in the drm_mode_atomic_ioctl function.
syzbot reports a slab use-after-free in hci_conn_hash_flush [1].
After releasing an object using hci_conn_del_sysfs in the
hci_conn_cleanup function, releasing the same object again
using the hci_dev_put and hci_conn_put functions causes a double free.
Here's a simplified flow:
This patch drop the hci_dev_put and hci_conn_put function
call in hci_conn_cleanup function, because the object is
freed in hci_conn_del_sysfs function.
This patch also fixes the refcounting in hci_conn_add_sysfs() and
hci_conn_del_sysfs() to take into account device_add() failures.
This fixes CVE-2023-28464.
Link: https://syzkaller.appspot.com/bug?id=1bb51491ca5df96a5f724899d1dbb87afda61419 Signed-off-by: ZhengHan Wang <wzhmmmmm@gmail.com> Co-developed-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Signed-off-by: youwan Wang <wangyouwan@126.com> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
As talked about in commit d66d24ac300c ("ath10k: Keep track of which
interrupts fired, don't poll them"), if we access the copy engine
register at a bad time then ath10k can go boom. However, it's not
necessarily easy to know when it's safe to access them.
The ChromeOS test labs saw a crash that looked like this at
shutdown/reboot time (on a chromeos-5.15 kernel, but likely the
problem could also reproduce upstream):
Though there's no known way to reproduce the problem, it makes sense
that it would be the same issue where we're trying to access copy
engine registers when it's not allowed.
Let's fix this by changing how we "disable" the interrupts. Instead of
tweaking the copy engine registers we'll just use disable_irq() and
enable_irq(). Then we'll configure the interrupts once at power up
time.
When compiling with clang 16.0.6 and CONFIG_FORTIFY_SOURCE=y, I've
noticed the following (somewhat confusing due to absence of an actual
source code location):
In file included from drivers/net/wireless/ath/ath10k/debug.c:8:
In file included from ./include/linux/module.h:13:
In file included from ./include/linux/stat.h:19:
In file included from ./include/linux/time.h:60:
In file included from ./include/linux/time32.h:13:
In file included from ./include/linux/timex.h:67:
In file included from ./arch/x86/include/asm/timex.h:5:
In file included from ./arch/x86/include/asm/processor.h:23:
In file included from ./arch/x86/include/asm/msr.h:11:
In file included from ./arch/x86/include/asm/cpumask.h:5:
In file included from ./include/linux/cpumask.h:12:
In file included from ./include/linux/bitmap.h:11:
In file included from ./include/linux/string.h:254:
./include/linux/fortify-string.h:592:4: warning: call to '__read_overflow2_field'
declared with 'warning' attribute: detected read beyond size of field (2nd
parameter); maybe use struct_group()? [-Wattribute-warning]
__read_overflow2_field(q_size_field, size);
The compiler actually complains on 'ath10k_debug_get_et_strings()' where
fortification logic inteprets call to 'memcpy()' as an attempt to copy
the whole 'ath10k_gstrings_stats' array from it's first member and so
issues an overread warning. This warning may be silenced by passing
an address of the whole array and not the first member to 'memcpy()'.
Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru> Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20230829093652.234537-1-dmantipov@yandex.ru Signed-off-by: Sasha Levin <sashal@kernel.org>
When compiling with clang 16.0.6 and CONFIG_FORTIFY_SOURCE=y, I've
noticed the following (somewhat confusing due to absence of an actual
source code location):
In file included from drivers/net/wireless/ath/ath9k/debug.c:17:
In file included from ./include/linux/slab.h:16:
In file included from ./include/linux/gfp.h:7:
In file included from ./include/linux/mmzone.h:8:
In file included from ./include/linux/spinlock.h:56:
In file included from ./include/linux/preempt.h:79:
In file included from ./arch/x86/include/asm/preempt.h:9:
In file included from ./include/linux/thread_info.h:60:
In file included from ./arch/x86/include/asm/thread_info.h:53:
In file included from ./arch/x86/include/asm/cpufeature.h:5:
In file included from ./arch/x86/include/asm/processor.h:23:
In file included from ./arch/x86/include/asm/msr.h:11:
In file included from ./arch/x86/include/asm/cpumask.h:5:
In file included from ./include/linux/cpumask.h:12:
In file included from ./include/linux/bitmap.h:11:
In file included from ./include/linux/string.h:254:
./include/linux/fortify-string.h:592:4: warning: call to '__read_overflow2_field'
declared with 'warning' attribute: detected read beyond size of field (2nd
parameter); maybe use struct_group()? [-Wattribute-warning]
__read_overflow2_field(q_size_field, size);
In file included from drivers/net/wireless/ath/ath9k/htc_drv_debug.c:17:
In file included from drivers/net/wireless/ath/ath9k/htc.h:20:
In file included from ./include/linux/module.h:13:
In file included from ./include/linux/stat.h:19:
In file included from ./include/linux/time.h:60:
In file included from ./include/linux/time32.h:13:
In file included from ./include/linux/timex.h:67:
In file included from ./arch/x86/include/asm/timex.h:5:
In file included from ./arch/x86/include/asm/processor.h:23:
In file included from ./arch/x86/include/asm/msr.h:11:
In file included from ./arch/x86/include/asm/cpumask.h:5:
In file included from ./include/linux/cpumask.h:12:
In file included from ./include/linux/bitmap.h:11:
In file included from ./include/linux/string.h:254:
./include/linux/fortify-string.h:592:4: warning: call to '__read_overflow2_field'
declared with 'warning' attribute: detected read beyond size of field (2nd
parameter); maybe use struct_group()? [-Wattribute-warning]
__read_overflow2_field(q_size_field, size);
The compiler actually complains on 'ath9k_get_et_strings()' and
'ath9k_htc_get_et_strings()' due to the same reason: fortification logic
inteprets call to 'memcpy()' as an attempt to copy the whole array from
it's first member and so issues an overread warning. These warnings may
be silenced by passing an address of the whole array and not the first
member to 'memcpy()'.
Now that bpf_throw kfunc is the first such call instruction that has
noreturn semantics within the verifier, this also kicks in dead code
elimination in unprecedented ways. For one, any instruction following
a bpf_throw call will never be marked as seen. Moreover, if a callchain
ends up throwing, any instructions after the call instruction to the
eventually throwing subprog in callers will also never be marked as
seen.
The tempting way to fix this would be to emit extra 'int3' instructions
which bump the jited_len of a program, and ensure that during runtime
when a program throws, we can discover its boundaries even if the call
instruction to bpf_throw (or to subprogs that always throw) is emitted
as the final instruction in the program.
An example of such a program would be this:
do_something():
...
r0 = 0
exit
foo():
r1 = 0
call bpf_throw
r0 = 0
exit
bar(cond):
if r1 != 0 goto pc+2
call do_something
exit
call foo
r0 = 0 // Never seen by verifier
exit //
main(ctx):
r1 = ...
call bar
r0 = 0
exit
Here, if we do end up throwing, the stacktrace would be the following:
bpf_throw
foo
bar
main
In bar, the final instruction emitted will be the call to foo, as such,
the return address will be the subsequent instruction (which the JIT
emits as int3 on x86). This will end up lying outside the jited_len of
the program, thus, when unwinding, we will fail to discover the return
address as belonging to any program and end up in a panic due to the
unreliable stack unwinding of BPF programs that we never expect.
To remedy this case, make bpf_prog_ksym_find treat IP == ksym.end as
part of the BPF program, so that is_bpf_text_address returns true when
such a case occurs, and we are able to unwind reliably when the final
instruction ends up being a call instruction.
This is based on alx driver commit 881d0327db37 ("net: alx: Work around
the DMA RX overflow issue").
The alx and atl1c drivers had RX overflow error which was why a custom
allocator was created to avoid certain addresses. The simpler workaround
then created for alx driver, but not for atl1c due to lack of tester.
Instead of using a custom allocator, check the allocated skb address and
use skb_reserve() to move away from problematic 0x...fc0 address.
When compiling with clang 16.0.6 and CONFIG_FORTIFY_SOURCE=y, I've
noticed the following (somewhat confusing due to absence of an actual
source code location):
In file included from drivers/net/wireless/virtual/mac80211_hwsim.c:18:
In file included from ./include/linux/slab.h:16:
In file included from ./include/linux/gfp.h:7:
In file included from ./include/linux/mmzone.h:8:
In file included from ./include/linux/spinlock.h:56:
In file included from ./include/linux/preempt.h:79:
In file included from ./arch/x86/include/asm/preempt.h:9:
In file included from ./include/linux/thread_info.h:60:
In file included from ./arch/x86/include/asm/thread_info.h:53:
In file included from ./arch/x86/include/asm/cpufeature.h:5:
In file included from ./arch/x86/include/asm/processor.h:23:
In file included from ./arch/x86/include/asm/msr.h:11:
In file included from ./arch/x86/include/asm/cpumask.h:5:
In file included from ./include/linux/cpumask.h:12:
In file included from ./include/linux/bitmap.h:11:
In file included from ./include/linux/string.h:254:
./include/linux/fortify-string.h:592:4: warning: call to '__read_overflow2_field'
declared with 'warning' attribute: detected read beyond size of field (2nd
parameter); maybe use struct_group()? [-Wattribute-warning]
__read_overflow2_field(q_size_field, size);
The compiler actually complains on 'mac80211_hwsim_get_et_strings()' where
fortification logic inteprets call to 'memcpy()' as an attempt to copy the
whole 'mac80211_hwsim_gstrings_stats' array from its first member and so
issues an overread warning. This warning may be silenced by passing
an address of the whole array and not the first member to 'memcpy()'.
Qi Zheng reported crashes in a production environment and provided a
simplified example as a reproducer:
| For example, if we use Qemu to start a two NUMA node kernel,
| one of the nodes has 2M memory (less than NODE_MIN_SIZE),
| and the other node has 2G, then we will encounter the
| following panic:
|
| BUG: kernel NULL pointer dereference, address: 0000000000000000
| <...>
| RIP: 0010:_raw_spin_lock_irqsave+0x22/0x40
| <...>
| Call Trace:
| <TASK>
| deactivate_slab()
| bootstrap()
| kmem_cache_init()
| start_kernel()
| secondary_startup_64_no_verify()
The crashes happen because of inconsistency between the nodemask that
has nodes with less than 4MB as memoryless, and the actual memory fed
into the core mm.
The commit:
9391a3f9c7f1 ("[PATCH] x86_64: Clear more state when ignoring empty node in SRAT parsing")
... that introduced minimal size of a NUMA node does not explain why
a node size cannot be less than 4MB and what boot failures this
restriction might fix.
Fixes have been submitted to the core MM code to tighten up the
memory topologies it accepts and to not crash on weird input:
mm: page_alloc: skip memoryless nodes entirely
mm: memory_hotplug: drop memoryless node from fallback lists
Andrew has accepted them into the -mm tree, but there are no
stable SHA1's yet.
This patch drops the limitation for minimal node size on x86:
- which works around the crash without the fixes to the core MM.
- makes x86 topologies less weird,
- removes an arbitrary and undocumented limitation on NUMA topologies.
[ mingo: Improved changelog clarity. ]
Reported-by: Qi Zheng <zhengqi.arch@bytedance.com> Tested-by: Mario Casquero <mcasquer@redhat.com> Signed-off-by: Mike Rapoport (IBM) <rppt@kernel.org> Signed-off-by: Ingo Molnar <mingo@kernel.org> Acked-by: David Hildenbrand <david@redhat.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: Rik van Riel <riel@surriel.com> Link: https://lore.kernel.org/r/ZS+2qqjEO5/867br@gmail.com Signed-off-by: Sasha Levin <sashal@kernel.org>
All callers of work_on_cpu() share the same lock class key for all the
functions queued. As a result the workqueue related locking scenario for
a function A may be spuriously accounted as an inversion against the
locking scenario of function B such as in the following model:
long A(void *arg)
{
mutex_lock(&mutex);
mutex_unlock(&mutex);
}
launchA and launchB running concurrently have no chance to deadlock.
However the above can be reported by lockdep as a possible locking
inversion because the works containing A() and B() are treated as
belonging to the same locking class.
The following shows an existing example of such a spurious lockdep splat:
======================================================
WARNING: possible circular locking dependency detected 6.6.0-rc1-00065-g934ebd6e5359 #35409 Not tainted
------------------------------------------------------
kworker/0:1/9 is trying to acquire lock: ffffffff9bc72f30 (cpu_hotplug_lock){++++}-{0:0}, at: _cpu_down+0x57/0x2b0
but task is already holding lock: ffff9e3bc0057e60 ((work_completion)(&wfc.work)){+.+.}-{0:0}, at: process_scheduled_works+0x216/0x500
which lock already depends on the new lock.
the existing dependency chain (in reverse order) is:
On SAM9 hardware two cascaded 16 bit timers are used to form a 32 bit
high resolution timer that is used as scheduler clock when the kernel
has been configured that way (CONFIG_ATMEL_CLOCKSOURCE_TCB).
The driver initially triggers a reset-to-zero of the two timers but this
reset is only performed on the next rising clock. For the first timer
this is ok - it will be in the next 60ns (16MHz clock). For the chained
second timer this will only happen after the first timer overflows, i.e.
after 2^16 clocks (~4ms with a 16MHz clock). So with other words the
scheduler clock resets to 0 after the first 2^16 clock cycles.
It looks like that the scheduler does not like this and behaves wrongly
over its lifetime, e.g. some tasks are scheduled with a long delay. Why
that is and if there are additional requirements for this behaviour has
not been further analysed.
There is a simple fix for resetting the second timer as well when the
first timer is reset and this is to set the ATMEL_TC_ASWTRG_SET bit in
the Channel Mode register (CMR) of the first timer. This will also rise
the TIOA line (clock input of the second timer) when a software trigger
respective SYNC is issued.
Signed-off-by: Ronald Wahl <ronald.wahl@raritan.com> Acked-by: Alexandre Belloni <alexandre.belloni@bootlin.com> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org> Link: https://lore.kernel.org/r/20231007161803.31342-1-rwahl@gmx.de Signed-off-by: Sasha Levin <sashal@kernel.org>
'rb->aux_pages' allocated by kcalloc() is a pointer array which is used to
maintains AUX trace pages. The allocated page for this array is physically
contiguous (and virtually contiguous) with an order of 0..MAX_ORDER. If the
size of pointer array crosses the limitation set by MAX_ORDER, it reveals a
WARNING.
So bail out early with -ENOMEM if the request AUX area is out of bound,
e.g.:
#perf record -C 0 -m ,4G -e arm_spe_0// -- sleep 1
failed to mmap with 12 (Cannot allocate memory)
In some cases running with the test-ww_mutex code, I was seeing
odd behavior where sometimes it seemed flush_workqueue was
returning before all the work threads were finished.
Often this would cause strange crashes as the mutexes would be
freed while they were being used.
Looking at the code, there is a lifetime problem as the
controlling thread that spawns the work allocates the
"struct stress" structures that are passed to the workqueue
threads. Then when the workqueue threads are finished,
they free the stress struct that was passed to them.
Unfortunately the workqueue work_struct node is in the stress
struct. Which means the work_struct is freed before the work
thread returns and while flush_workqueue is waiting.
It seems like a better idea to have the controlling thread
both allocate and free the stress structures, so that we can
be sure we don't corrupt the workqueue by freeing the structure
prematurely.
So this patch reworks the test to do so, and with this change
I no longer see the early flush_workqueue returns.
In the tree search v2 ioctl we use the type size_t, which is an unsigned
long, to track the buffer size in the local variable 'buf_size'. An
unsigned long is 32 bits wide on a 32 bits architecture. The buffer size
defined in struct btrfs_ioctl_search_args_v2 is a u64, so when we later
try to copy the local variable 'buf_size' to the argument struct, when
the search returns -EOVERFLOW, we copy only 32 bits which will be a
problem on big endian systems.
Fix this by using a u64 type for the buffer sizes, not only at
btrfs_ioctl_tree_search_v2(), but also everywhere down the call chain
so that we can use the u64 at btrfs_ioctl_tree_search_v2().
Fixes: cc68a8a5a433 ("btrfs: new ioctl TREE_SEARCH_V2") Reported-by: Dan Carpenter <dan.carpenter@linaro.org> Link: https://lore.kernel.org/linux-btrfs/ce6f4bd6-9453-4ffe-ba00-cee35495e10f@moroto.mountain/ Signed-off-by: Filipe Manana <fdmanana@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The commit above made quirks with an OEMID fail to be applied, as they
were checking card->cid.oemid for the full 16 bits defined in MMC_FIXUP
macros but the field would only contain the bottom 8 bits.
eMMC v5.1A might have bogus values in OEMID's higher bits so another fix
will be made, but it has been decided to revert this until that is ready.
The HDMI hotplug callback to the hdmi-codec is currently registered when
jack is set.
The hotplug not only serves to report the ASoC jack state but also to get
the ELD. It should be registered when the component probes instead, so it
does not depend on the card driver registering a jack for the HDMI to
properly report the ELD.
Fixes: 25ce4f2b3593 ("ASoC: hdmi-codec: Get ELD in before reporting plugged event") Signed-off-by: Jerome Brunet <jbrunet@baylibre.com> Link: https://lore.kernel.org/r/20231106104013.704356-1-jbrunet@baylibre.com Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
If DRM_IOCTL_SYNCOBJ_TIMELINE_WAIT is invoked with the
DRM_SYNCOBJ_WAIT_FLAGS_WAIT_AVAILABLE flag set but no fence has yet been
submitted for the given timeline point the call will fail immediately
with EINVAL. This does not match the intended behavior where the call
should wait until the fence has been submitted (or the timeout expires).
The following small example program illustrates the issue. It should
wait for 5 seconds and then print ETIME, but instead it terminates right
away after printing EINVAL.
The ipv6 redirect target was derived from the ipv4 one, i.e. its
identical to a 'dnat' with the first (primary) address assigned to the
network interface. The code has been moved around to make it usable
from nf_tables too, but its still the same as it was back when this
was added in 2012.
IPv6, however, has different types of addresses, if the 'wrong' address
comes first the redirection does not work.
In Daniels case, the addresses are:
inet6 ::ffff:192 ...
inet6 2a01: ...
... so the function attempts to redirect to the mapped address.
Add more checks before the address is deemed correct:
1. If the packets' daddr is scoped, search for a scoped address too
2. skip tentative addresses
3. skip mapped addresses
Use the first address that appears to match our needs.
`nf_nat_redirect_ipv4` takes a `struct nf_nat_ipv4_multi_range_compat`,
but converts it internally to a `struct nf_nat_range2`. Change the
function to take the latter, factor out the code now shared with
`nf_nat_redirect_ipv6`, move the conversion to the xt_REDIRECT module,
and update the ipv4 range initialization in the nft_redir module.
Replace a bare hex constant for 127.0.0.1 with a macro.
/* Can't setup nat info for confirmed ct. */
if (nf_ct_is_confirmed(ct))
return NF_ACCEPT;
This means that `ct` cannot be null or the kernel will crash, and
implies that `ctinfo` is `IP_CT_NEW` or `IP_CT_RELATED`.
nft_redir has separate ipv4 and ipv6 call-backs which share much of
their code, and an inet one switch containing a switch that calls one of
the others based on the family of the packet. Merge the ipv4 and ipv6
ones into the inet one in order to get rid of the duplicate code.
Const-qualify the `priv` pointer since we don't need to write through
it.
Assign `priv->flags` to the range instead of OR-ing it in.
Set the `NF_NAT_RANGE_PROTO_SPECIFIED` flag once during init, rather
than on every eval.
Signed-off-by: Jeremy Sowden <jeremy@azazel.net> Signed-off-by: Florian Westphal <fw@strlen.de>
Stable-dep-of: 80abbe8a8263 ("netfilter: nat: fix ipv6 nat redirect with mapped and scoped addresses") Signed-off-by: Sasha Levin <sashal@kernel.org>
Add the code to handle an invalid state when both bits S_RX_EVENT
(indicating a transaction) and S_START_BUSY (indicating the end
of transaction - transition of START_BUSY from 1 to 0) are set in
the interrupt status register during a slave read.
Signed-off-by: Roman Bacik <roman.bacik@broadcom.com> Fixes: 1ca1b4516088 ("i2c: iproc: handle Master aborted error") Acked-by: Ray Jui <ray.jui@broadcom.com> Signed-off-by: Wolfram Sang <wsa@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
So far we ignore the setting of IFF_MULTICAST. Fix this and clear bit
AcceptMulticast if IFF_MULTICAST isn't set.
Note: Based on the implementations I've seen it doesn't seem to be 100% clear
what a driver is supposed to do if IFF_ALLMULTI is set but IFF_MULTICAST
is not. This patch is based on the understanding that IFF_MULTICAST has
precedence.
If one of the underlying disks of raid or dm is set to read-only, then
each io will generate new log, which will cause message storm. This
environment is indeed problematic, however we can't make sure our
naive custormer won't do this, hence use pr_warn_ratelimited() to
prevent message storm in this case.
Signed-off-by: Yu Kuai <yukuai3@huawei.com> Fixes: 57e95e4670d1 ("block: fix and cleanup bio_check_ro") Signed-off-by: Ye Bin <yebin10@huawei.com> Link: https://lore.kernel.org/r/20231107111247.2157820-1-yukuai1@huaweicloud.com Signed-off-by: Jens Axboe <axboe@kernel.dk> Signed-off-by: Sasha Levin <sashal@kernel.org>
Note that we always hold a reference to sock when attempting
to submit close_work. Therefore, if we have successfully
canceled close_work from pending, we MUST release that reference
to avoid potential leaks.
Fixes: 42bfba9eaa33 ("net/smc: immediate termination for SMCD link groups") Signed-off-by: D. Wythe <alibuda@linux.alibaba.com> Reviewed-by: Dust Li <dust.li@linux.alibaba.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
This patch re-fix the issues mentioned by commit 22a825c541d7
("net/smc: fix NULL sndbuf_desc in smc_cdc_tx_handler()").
Blocking sending message do solve the issues though, but it also
prevents the peer to receive the final message. Besides, in logic,
whether the sndbuf_desc is NULL or not have no impact on the processing
of cdc message sending.
Hence that, this patch allows the cdc message sending but to check the
sndbuf_desc with care in smc_cdc_tx_handler().
Fixes: 22a825c541d7 ("net/smc: fix NULL sndbuf_desc in smc_cdc_tx_handler()") Signed-off-by: D. Wythe <alibuda@linux.alibaba.com> Reviewed-by: Dust Li <dust.li@linux.alibaba.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Dues to __set_bit is not atomic, the DEAD or DONE might be lost.
if the DEAD flag lost, the state SMC_CLOSED will be never be reached
in smc_close_passive_work:
if (sock_flag(sk, SOCK_DEAD) &&
smc_close_sent_any_close(conn)) {
sk->sk_state = SMC_CLOSED;
} else {
/* just shutdown, but not yet closed locally */
sk->sk_state = SMC_APPFINCLOSEWAIT;
}
Replace sock_set_flags or __set_bit to set_bit will fix this problem.
Since set_bit is atomic.
Fixes: b38d732477e4 ("smc: socket closing and linkgroup cleanup") Signed-off-by: D. Wythe <alibuda@linux.alibaba.com> Reviewed-by: Dust Li <dust.li@linux.alibaba.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
In the PMTU test, when all previous tests are skipped and the new test
passes, the exit code is set to 0. However, the current check mistakenly
treats this as an assignment, causing the check to pass every time.
Consequently, regardless of how many tests have failed, if the latest test
passes, the PMTU test will report a pass.
Fixes: 2a9d3716b810 ("selftests: pmtu.sh: improve the test result processing") Signed-off-by: Hangbin Liu <liuhangbin@gmail.com> Acked-by: Po-Hsu Lin <po-hsu.lin@canonical.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
From XGMAC Core 3.20 and later, each Flexible PPS has individual PPSEN bit
to select Fixed mode or Flexible mode. The PPSEN must be set, or it stays
in Fixed PPS mode by default.
XGMAC Core prior 3.20, only PPSEN0(bit 4) is writable. PPSEN{1,2,3} are
read-only reserved, and they are already in Flexible mode by default, our
new code always set PPSEN{1,2,3} do not make things worse ;-)
Fixes: 95eaf3cd0a90 ("net: stmmac: dwxgmac: Add Flexible PPS support") Reviewed-by: Serge Semin <fancer.lancer@gmail.com> Reviewed-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Furong Xu <0x1207@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
after normal termination @entry is left with the value NULL
This is not correct in the case where UINT_MAX has an entry in the idr.
In that case @entry will be non-NULL after termination.
No current code depends on the documentation being correct, but to
save future code we should fix it.
Also fix idr_for_each_entry_continue_ul(). While this is not documented
as leaving @entry as NULL, the mellanox driver appears to depend on
it doing so. So make that explicit in the documentation as well as in
the code.
Fixes: e33d2b74d805 ("idr: fix overflow case for idr_for_each_entry_ul()") Cc: Matthew Wilcox <willy@infradead.org> Cc: Chris Mi <chrism@mellanox.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
RTL8168H and RTL8107E ethernet adapters erroneously filter unicast
eapol packets unless allmulti is enabled. These devices correspond to
RTL_GIGA_MAC_VER_46 and VER_48. Add an exception for VER_46 and VER_48
in the same way that VER_35 has an exception.
Fixes: 6e1d0b898818 ("r8169:add support for RTL8168H and RTL8107E") Signed-off-by: Patrick Thompson <ptf@google.com> Reviewed-by: Jacob Keller <jacob.e.keller@intel.com> Reviewed-by: Heiner Kallweit <hkallweit1@gmail.com> Link: https://lore.kernel.org/r/20231030205031.177855-1-ptf@google.com Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Initially, commit 4237c75c0a35 ("[MLSXFRM]: Auto-labeling of child
sockets") introduced security_inet_conn_request() in some functions
where reqsk is allocated. The hook is added just after the allocation,
so reqsk's IPv6 remote address was not initialised then.
However, SELinux/Smack started to read it in netlbl_req_setattr()
after commit e1adea927080 ("calipso: Allow request sockets to be
relabelled by the lsm.").
Commit 284904aa7946 ("lsm: Relocate the IPv4 security_inet_conn_request()
hooks") fixed that kind of issue only in TCPv4 because IPv6 labeling was
not supported at that time. Finally, the same issue was introduced again
in IPv6.
Let's apply the same fix on DCCPv6 and TCPv6.
Fixes: e1adea927080 ("calipso: Allow request sockets to be relabelled by the lsm.") Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com> Acked-by: Paul Moore <paul@paul-moore.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Initially, commit 4237c75c0a35 ("[MLSXFRM]: Auto-labeling of child
sockets") introduced security_inet_conn_request() in some functions
where reqsk is allocated. The hook is added just after the allocation,
so reqsk's IPv4 remote address was not initialised then.
However, SELinux/Smack started to read it in netlbl_req_setattr()
after the cited commits.
This bug was partially fixed by commit 284904aa7946 ("lsm: Relocate
the IPv4 security_inet_conn_request() hooks").
This patch fixes the last bug in DCCPv4.
Fixes: 389fb800ac8b ("netlabel: Label incoming TCP connections correctly in SELinux") Fixes: 07feee8f812f ("netlabel: Cleanup the Smack/NetLabel code to fix incoming TCP connections") Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com> Acked-by: Paul Moore <paul@paul-moore.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
When there are concurrent uref release and bpf timer init operations,
the following sequence diagram is possible. It will break the guarantee
provided by bpf_timer: bpf_timer will still be alive after userspace
application releases or unpins the map. It also will lead to kmemleak
for old kernel version which doesn't release bpf_timer when map is
released.
close(map_fd)
// put last uref
bpf_map_put_uref()
atomic_dec_and_test(map->usercnt)
array_map_free_timers()
bpf_timer_cancel_and_free()
// just return
read timer->timer is NULL
t = bpf_map_kmalloc_node()
timer->timer = t
unlock timer->lock
Fix the problem by checking map->usercnt after timer->timer is assigned,
so when there are concurrent uref release and bpf timer init, either
bpf_timer_cancel_and_free() from uref release reads a no-NULL timer
or the newly-added atomic64_read() returns a zero usercnt.
Because atomic_dec_and_test(map->usercnt) and READ_ONCE(timer->timer)
in bpf_timer_cancel_and_free() are not protected by a lock, so add
a memory barrier to guarantee the order between map->usercnt and
timer->timer. Also use WRITE_ONCE(timer->timer, x) to match the lockless
read of timer->timer in bpf_timer_cancel_and_free().
TIPC bearer-related names including link names must be null-terminated
strings. If a link name which is not null-terminated is passed through
netlink, strstr() and similar functions can cause buffer overrun. This
causes the above issue.
This patch changes the nla_policy for bearer-related names from NLA_STRING
to NLA_NUL_STRING. This resolves the issue by ensuring that only
null-terminated strings are accepted as bearer-related names.
syzbot reported similar uninit-value issue related to bearer names [2]. The
root cause of this issue is that a non-null-terminated bearer name was
passed. This patch also resolved this issue.
Fixes: 7be57fc69184 ("tipc: add link get/dump to new netlink api") Fixes: 0655f6a8635b ("tipc: add bearer disable/enable to new netlink api") Reported-and-tested-by: syzbot+5138ca807af9d2b42574@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=5138ca807af9d2b42574 [1] Reported-and-tested-by: syzbot+9425c47dccbcb4c17d51@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=9425c47dccbcb4c17d51 [2] Signed-off-by: Shigeru Yoshida <syoshida@redhat.com> Reviewed-by: Jiri Pirko <jiri@nvidia.com> Link: https://lore.kernel.org/r/20231030075540.3784537-1-syoshida@redhat.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The prp_fill_rct() function can fail. In that situation, it frees the
skb and returns NULL. Meanwhile on the success path, it returns the
original skb. So it's straight forward to fix bug by using the returned
value.
Fixes: 451d8123f897 ("net: prp: add packet handling support") Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org> Acked-by: Paolo Abeni <pabeni@redhat.com> Link: https://lore.kernel.org/r/57af1f28-7f57-4a96-bcd3-b7a0f2340845@moroto.mountain Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
LLC reads the mac header with eth_hdr without verifying that the skb
has an Ethernet header.
Syzbot was able to enter llc_rcv on a tun device. Tun can insert
packets without mac len and with user configurable skb->protocol
(passing a tun_pi header when not configuring IFF_NO_PI).
Add a mac_len test before all three eth_hdr(skb) calls under net/llc.
There are further uses in include/net/llc_pdu.h. All these are
protected by a test skb->protocol == ETH_P_802_2. Which does not
protect against this tun scenario.
But the mac_len test added in this patch in llc_fixup_skb will
indirectly protect those too. That is called from llc_rcv before any
other LLC code.
It is tempting to just add a blanket mac_len check in llc_rcv, but
not sure whether that could break valid LLC paths that do not assume
an Ethernet header. 802.2 LLC may be used on top of non-802.3
protocols in principle. The below referenced commit shows that used
to, on top of Token Ring.
At least one of the three eth_hdr uses goes back to before the start
of git history. But the one that syzbot exercises is introduced in
this commit. That commit is old enough (2008), that effectively all
stable kernels should receive this.
Fixes: f83f1768f833 ("[LLC]: skb allocation size for responses") Reported-by: syzbot+a8c7be6dee0de1b669cc@syzkaller.appspotmail.com Signed-off-by: Willem de Bruijn <willemb@google.com> Link: https://lore.kernel.org/r/20231025234251.3796495-1-willemdebruijn.kernel@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The put_device() calls rmi_release_function() which frees "fn" so the
dereference on the next line "fn->num_of_irqs" is a use after free.
Move the put_device() to the end to fix this.
Fixes: 24d28e4f1271 ("Input: synaptics-rmi4 - convert irq distribution to irq_domain") Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org> Link: https://lore.kernel.org/r/706efd36-7561-42f3-adfa-dd1d0bd4f5a1@moroto.mountain Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The suspend/resume functions currently utilize
clk_disable()/clk_enable() respectively which may be no-ops with certain
clock providers such as SCMI. Fix this to use clk_disable_unprepare()
and clk_prepare_enable() respectively as we should.
Instead of using one allocation per capture channel, use a single one. Also
store it in driver data instead of chip data.
This has several advantages:
- driver data isn't cleared when pwm_put() is called
- Reduces memory fragmentation
Also register the pwm chip only after the per capture channel data is
initialized as the capture callback relies on this initialization and it
might be called even before pwmchip_add() returns.
It would be still better to have struct sti_pwm_compat_data and the
per-channel data struct sti_cpt_ddata in a single memory chunk, but that's
not easily possible because the number of capture channels isn't known yet
when the driver data struct is allocated.
Currently, noinc writes are cached as if they were standard incrementing
writes, overwriting unrelated register values in the cache. Instead, we
want to cache the last value written to the register, as is done in the
accelerated noinc handler (regmap_noinc_readwrite).
Add check for the return value of kstrdup() and return the error
if it fails in order to avoid NULL pointer dereference.
Moreover, use kfree() in the later error handling in order to avoid
memory leak.
Fixes: c2f78f0cb294 ("media: vidtv: psi: add a Network Information Table (NIT)") Signed-off-by: Jiasheng Jiang <jiasheng@iscas.ac.cn> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Sasha Levin <sashal@kernel.org>
s3c_camif_register_video_node() works with video_device structure stored
as a field of camif_vp, so it should not be kfreed.
But there is video_device_release() on error path that do it.
Found by Linux Verification Center (linuxtesting.org) with SVACE.
Fixes: babde1c243b2 ("[media] V4L: Add driver for S3C24XX/S3C64XX SoC series camera interface") Signed-off-by: Katya Orlova <e.orlova@ispras.ru> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Sasha Levin <sashal@kernel.org>
There may be some a race condition between timer function
bttv_irq_timeout and bttv_remove. The timer is setup in
probe and there is no timer_delete operation in remove
function. When it hit kfree btv, the function might still be
invoked, which will cause use after free bug.
This bug is found by static analysis, it may be false positive.
Fix it by adding del_timer_sync invoking to the remove function.
This is odd to have a of_node_put() just after a for_each_child_of_node()
or a for_each_endpoint_of_node() loop. It should already be called
during the last iteration.
Afer commit 1fa5ae857bb1 ("driver core: get rid of struct device's
bus_id string array"), the name of device is allocated dynamically.
Therefore, it needs to be freed, which is done by the driver core for
us once all references to the device are gone. Therefore, move the
dev_set_name() call immediately before the call device_register(), which
either succeeds (then the freeing will be done upon subsequent remvoal),
or puts the reference in the error call. Also, it is not unusual that the
return value of dev_set_name is not checked.
Fixes: 1fa5ae857bb1 ("driver core: get rid of struct device's bus_id string array") Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
[linux@dominikbrodowski.net: simplification, commit message modified] Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
As the comment of device_register() says, it should use put_device()
to give up the reference in the error path. Then, insofar resources
will be freed in pcmcia_release_dev(), the error path is no longer
needed. In particular, this means that the (previously missing) dropping
of the reference to &p_dev->function_config->ref is now handled by
pcmcia_release_dev().
If device_register() returns error in pccardd(), it leads two issues:
1. The socket_released has never been completed, it will block
pcmcia_unregister_socket(), because of waiting for completion
of socket_released.
2. The device name allocated by dev_set_name() is leaked.
Fix this two issues by calling put_device() when device_register() fails.
socket_released can be completed in pcmcia_release_socket(), the name can
be freed in kobject_cleanup().
The current implementation passes PIN_IO_INTA_OUT (2) as a mask and
PIN_IO_INTAPM (GENMASK(1, 0)) as a value.
Swap the variables to assign mask and value the right way.
This error was first introduced with the alarm support. For better or
worse it worked as expected because 0x02 was applied as a mask to 0x03,
resulting 0x02 anyway. This will of course not work for any other value.
This crash is due to the clearing out the cxl_memdev's driver context
(@cxlds) before the subsystem is done with it. This is ultimately due to
the region(s), that this memdev is a member, being torn down and expecting
to be able to de-reference @cxlds, like here:
static int cxl_region_decode_reset(struct cxl_region *cxlr, int count)
...
if (cxlds->rcd)
goto endpoint_reset;
...
Fix it by keeping the driver context valid until memdev-device
unregistration, and subsequently the entire stack of related
dependencies, unwinds.
Fixes: 9cc238c7a526 ("cxl/pci: Introduce cdevm_file_operations") Reported-by: Ira Weiny <ira.weiny@intel.com> Reviewed-by: Davidlohr Bueso <dave@stgolabs.net> Reviewed-by: Dave Jiang <dave.jiang@intel.com> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Reviewed-by: Ira Weiny <ira.weiny@intel.com> Tested-by: Ira Weiny <ira.weiny@intel.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When p9pdu_readf() is called with "s?d" attribute, it allocates a pointer
that will store a string. But when p9pdu_readf() fails while handling "d"
then this pointer will not be freed in p9_check_errors().
Fixes: 51a87c552dfd ("9p: rework client code to use new protocol support functions") Reviewed-by: Christian Schoenebeck <linux_oss@crudebyte.com> Signed-off-by: Hangyu Hua <hbh25y@gmail.com>
Message-ID: <20231027030302.11927-1-hbh25y@gmail.com> Signed-off-by: Dominique Martinet <asmadeus@codewreck.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
If this driver enables the xHC clocks while resuming from sleep, it calls
clk_prepare_enable() without checking for errors and blithely goes on to
read/write the xHC's registers -- which, with the xHC not being clocked,
at least on ARM32 usually causes an imprecise external abort exceptions
which cause kernel oops. Currently, the chips for which the driver does
the clock dance on suspend/resume seem to be the Broadcom STB SoCs, based
on ARM32 CPUs, as it seems...
Found by Linux Verification Center (linuxtesting.org) with the Svace static
analysis tool.
The AMD USB host controller (1022:43f7) isn't going into PCI D3 by default
without anything connected. This is because the policy that was introduced
by commit a611bf473d1f ("xhci-pci: Set runtime PM as default policy on all
xHC 1.2 or later devices") only covered 1.2 or later.
The 1.1 specification also has the same requirement as the 1.2
specification for D3 support. So expand the runtime PM as default policy
to all AMD 1.1 devices as well.
Fixes: a611bf473d1f ("xhci-pci: Set runtime PM as default policy on all xHC 1.2 or later devices") Link: https://composter.com.ua/documents/xHCI_Specification_for_USB.pdf Co-developed-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Basavaraj Natikar <Basavaraj.Natikar@amd.com> Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com> Link: https://lore.kernel.org/r/20231019102924.2797346-15-mathias.nyman@linux.intel.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The macro __SPIN_LOCK_INITIALIZER() is implementation specific. Users
that desire to initialize a spinlock in a struct must use
__SPIN_LOCK_UNLOCKED().
Use __SPIN_LOCK_UNLOCKED() for the spinlock_t in imc_global_refc.
Fixes: 76d588dddc459 ("powerpc/imc-pmu: Fix use of mutex in IRQs disabled section") Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://msgid.link/20230309134831.Nz12nqsU@linutronix.de Signed-off-by: Sasha Levin <sashal@kernel.org>
Sparse reports a size mismatch in the endian swap. The Opal
implementation[1] passes the value as a __be64, and the receiving
variable out_qsize is a u64, so the use of be32_to_cpu() appears to be
an error.
When MODULE_DEVICE_TABLE(tee, ) is built on a host with a different
endianness from the target architecture, it results in an incorrect
MODULE_ALIAS().
For example, see a case where drivers/char/hw_random/optee-rng.c
is built as a module for ARM little-endian.
If you build it on a little-endian host, you will get the correct
MODULE_ALIAS:
The same problem also occurs when you enable CONFIG_CPU_BIG_ENDIAN,
and build it on a little-endian host.
This issue has been unnoticed because the ARM kernel is configured for
little-endian by default, and most likely built on a little-endian host
(cross-build on x86 or native-build on ARM).
The uuid field must not be reversed because uuid_t is an array of __u8.
Fixes: 0fc1db9d1059 ("tee: add bus driver framework for TEE based devices") Signed-off-by: Masahiro Yamada <masahiroy@kernel.org> Reviewed-by: Sumit Garg <sumit.garg@linaro.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
If pxad_alloc_desc() fails on the first dma_pool_alloc() call, then
sw_desc->nb_desc is zero.
In such a case pxad_free_desc() is called and it will BUG_ON().
Remove this erroneous BUG_ON().
It is also useless, because if "sw_desc->nb_desc == 0", then, on the first
iteration of the for loop, i is -1 and the loop will not be executed.
(both i and sw_desc->nb_desc are 'int')
If a hub is disconnected that has device(s) that's attached to the usbip layer
the disconnect function might fail because it tries to release the port
on an already disconnected hub.
Fixes: 6080cd0e9239 ("staging: usbip: claim ports used by shared devices") Signed-off-by: Jonas Blixt <jonas.blixt@actia.se> Acked-by: Shuah Khan <skhan@linuxfoundation.org> Link: https://lore.kernel.org/r/20230615092810.1215490-1-jonas.blixt@actia.se Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The iio_generic_buffer can return garbage values when the total size of
scan data is not a multiple of the largest element in the scan. This can be
demonstrated by reading a scan, consisting, for example of one 4-byte and
one 2-byte element, where the 4-byte element is first in the buffer.
The IIO generic buffer code does not take into account the last two
padding bytes that are needed to ensure that the 4-byte data for next
scan is correctly aligned.
Add the padding bytes required to align the next sample with the scan size.
It is not allowed to call kfree_skb() from hardware interrupt
context or with hardware interrupts being disabled.
So replace kfree_skb() with dev_kfree_skb_irq() under
spin_lock_irqsave(). Compile tested only.
Zero is not a valid IRQ for in-kernel code and the irq_of_parse_and_map()
function returns zero on error. So this check for valid IRQs should only
accept values > 0.
Fixes: 2b6b3b742019 ("ARM/dmaengine: edma: Merge the two drivers under drivers/dma/") Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org> Acked-by: Peter Ujfalusi <peter.ujfalusi@gmail.com> Link: https://lore.kernel.org/r/f15cb6a7-8449-4f79-98b6-34072f04edbc@moroto.mountain Signed-off-by: Vinod Koul <vkoul@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The USB host on Tegra3 works with 32-bit alignment. Previous code tried
to align the buffer, but it did align the wrapper struct instead, so
the buffer was at a constant offset of 8 bytes (two pointers) from
expected alignment. Since kmalloc() guarantees at least 8-byte
alignment already, the alignment-extending is removed.
Tegra USB controllers seem to issue DMA in full 32-bit words only and thus
may overwrite unevenly-sized buffers. One such occurrence is detected by
SLUB when receiving a reply to a 1-byte buffer (below). Fix this by
allocating a bounce buffer also for buffers with sizes not a multiple of 4.
=============================================================================
BUG kmalloc-64 (Tainted: G B ): kmalloc Redzone overwritten
-----------------------------------------------------------------------------
Redzone 8555ccc0: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc ................
Redzone 8555ccd0: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc ................
Redzone 8555cce0: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc ................
Redzone 8555ccf0: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc ................
Object 8555cd00: 01 00 00 00 cc cc cc cc cc cc cc cc cc cc cc cc ................
Object 8555cd10: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc ................
Object 8555cd20: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc ................
Object 8555cd30: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc ................
Redzone 8555cd40: cc cc cc cc ....
Padding 8555cd74: 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a ZZZZZZZZZZZZ
CPU: 3 PID: 41 Comm: kworker/3:1 Tainted: G B 6.6.0-rc1mq-00118-g59786f827ea1 #1115
Hardware name: NVIDIA Tegra SoC (Flattened Device Tree)
Workqueue: usb_hub_wq hub_event
[<8010ca28>] (unwind_backtrace) from [<801090a5>] (show_stack+0x11/0x14)
[<801090a5>] (show_stack) from [<805da2fb>] (dump_stack_lvl+0x4d/0x7c)
[<805da2fb>] (dump_stack_lvl) from [<8026464f>] (check_bytes_and_report+0xb3/0xe4)
[<8026464f>] (check_bytes_and_report) from [<802648e1>] (check_object+0x261/0x290)
[<802648e1>] (check_object) from [<802671b1>] (free_to_partial_list+0x105/0x3f8)
[<802671b1>] (free_to_partial_list) from [<80268613>] (__kmem_cache_free+0x103/0x128)
[<80268613>] (__kmem_cache_free) from [<80425a67>] (usb_get_status+0x73/0xac)
[<80425a67>] (usb_get_status) from [<80421b31>] (hub_probe+0x5e9/0xcec)
[<80421b31>] (hub_probe) from [<80428bbb>] (usb_probe_interface+0xbf/0x21c)
[<80428bbb>] (usb_probe_interface) from [<803ee13d>] (really_probe+0xa5/0x2c4)
[<803ee13d>] (really_probe) from [<803ee3d1>] (__driver_probe_device+0x75/0x174)
[<803ee3d1>] (__driver_probe_device) from [<803ee501>] (driver_probe_device+0x31/0x94)
usb 1-1: device descriptor read/8, error -71
In _dwc2_hcd_urb_enqueue(), "urb->hcpriv = NULL" is executed without
holding the lock "hsotg->lock". In _dwc2_hcd_urb_dequeue():
spin_lock_irqsave(&hsotg->lock, flags);
...
if (!urb->hcpriv) {
dev_dbg(hsotg->dev, "## urb->hcpriv is NULL ##\n");
goto out;
}
rc = dwc2_hcd_urb_dequeue(hsotg, urb->hcpriv); // Use urb->hcpriv
...
out:
spin_unlock_irqrestore(&hsotg->lock, flags);
When _dwc2_hcd_urb_enqueue() and _dwc2_hcd_urb_dequeue() are
concurrently executed, the NULL check of "urb->hcpriv" can be executed
before "urb->hcpriv = NULL". After urb->hcpriv is NULL, it can be used
in the function call to dwc2_hcd_urb_dequeue(), which can cause a NULL
pointer dereference.
This possible bug is found by an experimental static analysis tool
developed by myself. This tool analyzes the locking APIs to extract
function pairs that can be concurrently executed, and then analyzes the
instructions in the paired functions to identify possible concurrency
bugs including data races and atomicity violations. The above possible
bug is reported, when my tool analyzes the source code of Linux 6.5.
To fix this possible bug, "urb->hcpriv = NULL" should be executed with
holding the lock "hsotg->lock". After using this patch, my tool never
reports the possible bug, with the kernelconfiguration allyesconfig for
x86_64. Because I have no associated hardware, I cannot test the patch
in runtime testing, and just verify it according to the code logic.
idxd sub-drivers belong to bus dsa_bus_type. Thus, dsa_bus_type must be
registered in dsa bus init before idxd drivers can be registered.
But the order is wrong when both idxd and idxd_bus are builtin drivers.
In this case, idxd driver is compiled and linked before idxd_bus driver.
Since the initcall order is determined by the link order, idxd sub-drivers
are registered in idxd initcall before dsa_bus_type is registered
in idxd_bus initcall. idxd initcall fails:
[ 21.562803] calling idxd_init_module+0x0/0x110 @ 1
[ 21.570761] Driver 'idxd' was unable to register with bus_type 'dsa' because the bus was not initialized.
[ 21.586475] initcall idxd_init_module+0x0/0x110 returned -22 after 15717 usecs
[ 21.597178] calling dsa_bus_init+0x0/0x20 @ 1
To fix the issue, compile and link idxd_bus driver before idxd driver
to ensure the right registration order.
Fixes: d9e5481fca74 ("dmaengine: dsa: move dsa_bus_type out of idxd driver to standalone") Reported-by: Michael Prinke <michael.prinke@intel.com> Signed-off-by: Fenghua Yu <fenghua.yu@intel.com> Reviewed-by: Dave Jiang <dave.jiang@intel.com> Reviewed-by: Lijun Pan <lijun.pan@intel.com> Tested-by: Lijun Pan <lijun.pan@intel.com> Link: https://lore.kernel.org/r/20230924162232.1409454-1-fenghua.yu@intel.com Signed-off-by: Vinod Koul <vkoul@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
There is a pid leakage:
------------------------------
unreferenced object 0xffff88810c181940 (size 224):
comm "sshd", pid 8191, jiffies 4294946950 (age 524.570s)
hex dump (first 32 bytes):
01 00 00 00 00 00 00 00 00 00 00 00 ad 4e ad de .............N..
ff ff ff ff 6b 6b 6b 6b ff ff ff ff ff ff ff ff ....kkkk........
backtrace:
[<ffffffff814774e6>] kmem_cache_alloc+0x5c6/0x9b0
[<ffffffff81177342>] alloc_pid+0x72/0x570
[<ffffffff81140ac4>] copy_process+0x1374/0x2470
[<ffffffff81141d77>] kernel_clone+0xb7/0x900
[<ffffffff81142645>] __se_sys_clone+0x85/0xb0
[<ffffffff8114269b>] __x64_sys_clone+0x2b/0x30
[<ffffffff83965a72>] do_syscall_64+0x32/0x80
[<ffffffff83a00085>] entry_SYSCALL_64_after_hwframe+0x61/0xc6
It turns out that there is a race condition between disassociate_ctty() and
tty_signal_session_leader(), which caused this leakage.
The pid memleak is triggered by the following race:
task[sshd] task[bash]
----------------------- -----------------------
disassociate_ctty();
spin_lock_irq(¤t->sighand->siglock);
put_pid(current->signal->tty_old_pgrp);
current->signal->tty_old_pgrp = NULL;
tty = tty_kref_get(current->signal->tty);
spin_unlock_irq(¤t->sighand->siglock);
tty_vhangup();
tty_lock(tty);
...
tty_signal_session_leader();
spin_lock_irq(&p->sighand->siglock);
...
if (tty->ctrl.pgrp) //tty->ctrl.pgrp is not NULL
p->signal->tty_old_pgrp = get_pid(tty->ctrl.pgrp); //An extra get
spin_unlock_irq(&p->sighand->siglock);
...
tty_unlock(tty);
if (tty) {
tty_lock(tty);
...
put_pid(tty->ctrl.pgrp);
tty->ctrl.pgrp = NULL; //It's too late
...
tty_unlock(tty);
}
The issue is believed to be introduced by commit c8bcd9c5be24 ("tty:
Fix ->session locking") who moves the unlock of siglock in
disassociate_ctty() above "if (tty)", making a small window allowing
tty_signal_session_leader() to kick in. It can be easily reproduced by
adding a delay before "if (tty)" and at the entrance of
tty_signal_session_leader().
To fix this issue, we move "put_pid(current->signal->tty_old_pgrp)" after
"tty->ctrl.pgrp = NULL".
With below script, redundant compress extension will be parsed and added
by parse_options(), because parse_options() doesn't check whether the
extension is existed or not, fix it.
1. mount -t f2fs -o compress_extension=so /dev/vdb /mnt/f2fs
2. mount -t f2fs -o remount,compress_extension=so /mnt/f2fs
3. mount|grep f2fs
/dev/vdb on /mnt/f2fs type f2fs (...,compress_extension=so,compress_extension=so,...)
Fixes: 4c8ff7095bef ("f2fs: support data compression") Fixes: 151b1982be5d ("f2fs: compress: add nocompress extensions support") Signed-off-by: Chao Yu <chao@kernel.org> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
In f2fs_read_multi_pages(), once f2fs_decompress_cluster() was called if
we hit cached page in compress_inode's cache, dic may be released, it needs
break the loop rather than continuing it, in order to avoid accessing
invalid dic pointer.
In order to teach the compiler that 'trig->name' will never be truncated,
we need to tell it that 'cpu' is not negative.
When building with W=1, this fixes the following warnings:
drivers/leds/trigger/ledtrig-cpu.c: In function ‘ledtrig_cpu_init’:
drivers/leds/trigger/ledtrig-cpu.c:155:56: error: ‘%d’ directive output may be truncated writing between 1 and 11 bytes into a region of size 5 [-Werror=format-truncation=]
155 | snprintf(trig->name, MAX_NAME_LEN, "cpu%d", cpu);
| ^~
drivers/leds/trigger/ledtrig-cpu.c:155:52: note: directive argument in the range [-2147483648, 7]
155 | snprintf(trig->name, MAX_NAME_LEN, "cpu%d", cpu);
| ^~~~~~~
drivers/leds/trigger/ledtrig-cpu.c:155:17: note: ‘snprintf’ output between 5 and 15 bytes into a destination of size 8
155 | snprintf(trig->name, MAX_NAME_LEN, "cpu%d", cpu);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Disabling a PWM (i.e. calling pwm_apply_state with .enabled = false)
gives no guarantees what the PWM output does. It might freeze where it
currently is, or go in a High-Z state or drive the active or inactive
state, it might even continue to toggle.
To ensure that the LED gets really disabled, don't disable the PWM even
when .duty_cycle is zero.
This fixes disabling a leds-pwm LED on i.MX28. The PWM on this SoC is
one of those that freezes its output on disable, so if you disable an
LED that is full on, it stays on. If you disable a LED with half
brightness it goes off in 50% of the cases and full on in the other 50%.
The leds-turris-omnia driver uses three function for I2C access:
- i2c_smbus_write_byte_data() and i2c_smbus_read_byte_data(), which
cause an emulated SMBUS transfer,
- i2c_master_send(), which causes an ordinary I2C transfer.
The Turris Omnia MCU LED controller is not semantically SMBUS, it
operates as a simple I2C bus. It does not implement any of the SMBUS
specific features, like PEC, or procedure calls, or anything. Moreover
the I2C controller driver also does not implement SMBUS, and so the
emulated SMBUS procedure from drivers/i2c/i2c-core-smbus.c is used for
the SMBUS calls, which gives an unnecessary overhead.
When I first wrote the driver, I was unaware of these facts, and I
simply used the first function that worked.
Drop the I2C SMBUS calls and instead use simple I2C transfers.
Fixes: 089381b27abe ("leds: initial support for Turris Omnia LEDs") Signed-off-by: Marek Behún <kabel@kernel.org> Link: https://lore.kernel.org/r/20230918161104.20860-2-kabel@kernel.org Signed-off-by: Lee Jones <lee@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Do not lock driver mutex in the global LED panel brightness sysfs
accessors brightness_show() and brightness_store().
The mutex locking is unnecessary here. The I2C transfers are guarded by
I2C core locking mechanism, and the LED commands itself do not interfere
with other commands.
Fixes: 089381b27abe ("leds: initial support for Turris Omnia LEDs") Signed-off-by: Marek Behún <kabel@kernel.org> Reviewed-by: Lee Jones <lee@kernel.org> Link: https://lore.kernel.org/r/20230802160748.11208-2-kabel@kernel.org Signed-off-by: Lee Jones <lee@kernel.org>
Stable-dep-of: 6de283b96b31 ("leds: turris-omnia: Do not use SMBUS calls") Signed-off-by: Sasha Levin <sashal@kernel.org>
Commit 9e86b2ad4c11 changed the channel used for HPDET detection
(headphones vs lineout detection) from being hardcoded to
ARIZONA_ACCDET_MODE_HPL (HP left channel) to it being configurable
through arizona_pdata.hpdet_channel the DT/OF parsing added for
filling arizona_pdata on devicetree platforms ensures that
arizona_pdata.hpdet_channel gets set to ARIZONA_ACCDET_MODE_HPL
when not specified in the devicetree-node.
But on ACPI platforms where arizona_pdata is filled by
arizona_spi_acpi_probe() arizona_pdata.hpdet_channel was not
getting set, causing it to default to 0 aka ARIZONA_ACCDET_MODE_MIC.
This causes headphones to get misdetected as line-out on some models.
Fix this by setting hpdet_channel = ARIZONA_ACCDET_MODE_HPL.
Fixes: e933836744a2 ("mfd: arizona: Add support for ACPI enumeration of WM5102 connected over SPI") Signed-off-by: Hans de Goede <hdegoede@redhat.com> Link: https://lore.kernel.org/r/20231014205414.59415-1-hdegoede@redhat.com Signed-off-by: Lee Jones <lee@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
projects / thirdparty / kernel / stable.git / log
