Inspired by syzbot reports using a stack of multiple ipvlan devices.
Reduce stack size needed in ipvlan_process_v6_outbound() by moving
the flowi6 struct used for the route lookup in an non inlined
helper. ipvlan_route_v6_outbound() needs 120 bytes on the stack,
immediately reclaimed.
Also make sure ipvlan_process_v4_outbound() is not inlined.
We might also have to lower MAX_NEST_DEV, because only syzbot uses
setups with more than four stacked devices.
Fixes: 2ad7bf363841 ("ipvlan: Initial check-in of the IPVLAN driver.") Reported-by: syzbot <syzkaller@googlegroups.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Mahesh Bandewar <maheshb@google.com> Cc: Willem de Bruijn <willemb@google.com> Reviewed-by: Willem de Bruijn <willemb@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Let's move the SOCK_RCU_FREE part up a bit, before we are inserting
the socket into hashtables. Note, that the race is really harmless;
the bpf callers are handling this situation (where listener socket
doesn't have SOCK_RCU_FREE set) correctly, so the only
annoyance is a WARN_ONCE.
More details from Eric regarding SOCK_RCU_FREE timeline:
Commit 3b24d854cb35 ("tcp/dccp: do not touch listener sk_refcnt under
synflood") added SOCK_RCU_FREE. At that time, the precise location of
sock_set_flag(sk, SOCK_RCU_FREE) did not matter, because the thread calling
__inet_hash() owns a reference on sk. SOCK_RCU_FREE was only tested
at dismantle time.
Commit 6acc9b432e67 ("bpf: Add helper to retrieve socket in BPF")
started checking SOCK_RCU_FREE _after_ the lookup to infer whether
the refcount has been taken care of.
Fixes: 6acc9b432e67 ("bpf: Add helper to retrieve socket in BPF") Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Stanislav Fomichev <sdf@google.com> Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
The listen sk is currently stored in two hash tables,
listening_hash (hashed by port) and lhash2 (hashed by port and address).
After commit 0ee58dad5b06 ("net: tcp6: prefer listeners bound to an address")
and commit d9fbc7f6431f ("net: tcp: prefer listeners bound to an address"),
the TCP-SYN lookup fast path does not use listening_hash.
The commit 05c0b35709c5 ("tcp: seq_file: Replace listening_hash with lhash2")
also moved the seq_file (/proc/net/tcp) iteration usage from
listening_hash to lhash2.
There are still a few listening_hash usages left.
One of them is inet_reuseport_add_sock() which uses the listening_hash
to search a listen sk during the listen() system call. This turns
out to be very slow on use cases that listen on many different
VIPs at a popular port (e.g. 443). [ On top of the slowness in
adding to the tail in the IPv6 case ]. The latter patch has a
selftest to demonstrate this case.
This patch takes this chance to move all remaining listening_hash
usages to lhash2 and then retire listening_hash.
Since most changes need to be done together, it is hard to cut
the listening_hash to lhash2 switch into small patches. The
changes in this patch is highlighted here for the review
purpose.
1. Because of the listening_hash removal, lhash2 can use the
sk->sk_nulls_node instead of the icsk->icsk_listen_portaddr_node.
This will also keep the sk_unhashed() check to work as is
after stop adding sk to listening_hash.
The union is removed from inet_listen_hashbucket because
only nulls_head is needed.
2. icsk->icsk_listen_portaddr_node and its helpers are removed.
3. The current lhash2 users needs to iterate with sk_nulls_node
instead of icsk_listen_portaddr_node.
One case is in the inet[6]_lhash2_lookup().
Another case is the seq_file iterator in tcp_ipv4.c.
One thing to note is sk_nulls_next() is needed
because the old inet_lhash2_for_each_icsk_continue()
does a "next" first before iterating.
4. Move the remaining listening_hash usage to lhash2
inet_reuseport_add_sock() which this series is
trying to improve.
inet_diag.c and mptcp_diag.c are the final two
remaining use cases and is moved to lhash2 now also.
Signed-off-by: Martin KaFai Lau <kafai@fb.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Stable-dep-of: 871019b22d1b ("net: set SOCK_RCU_FREE before inserting socket into hashtable") Signed-off-by: Sasha Levin <sashal@kernel.org>
This patch folds lhash2 related functions into __inet_hash and
inet_unhash. This will make the removal of the listening_hash
in a latter patch easier to review.
First, this patch folds inet_hash2 into __inet_hash.
For unhash, the current call sequence is like
inet_unhash() => __inet_unhash() => inet_unhash2().
The specific testing cases in __inet_unhash() are mostly related
to TCP_LISTEN sk and its caller inet_unhash() already has
the TCP_LISTEN test, so this patch folds both __inet_unhash() and
inet_unhash2() into inet_unhash().
Note that all listening_hash users also have lhash2 initialized,
so the !h->lhash2 check is no longer needed.
Signed-off-by: Martin KaFai Lau <kafai@fb.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Stable-dep-of: 871019b22d1b ("net: set SOCK_RCU_FREE before inserting socket into hashtable") Signed-off-by: Sasha Levin <sashal@kernel.org>
After commit 0ee58dad5b06 ("net: tcp6: prefer listeners bound to an address")
and commit d9fbc7f6431f ("net: tcp: prefer listeners bound to an address"),
the count is no longer used. This patch removes it.
Signed-off-by: Martin KaFai Lau <kafai@fb.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Stable-dep-of: 871019b22d1b ("net: set SOCK_RCU_FREE before inserting socket into hashtable") Signed-off-by: Sasha Levin <sashal@kernel.org>
Iterate over the tcp listen sockets and pick those that have mptcp ulp
info attached.
mptcp_diag_get_info() is modified to prefer msk->first for mptcp sockets
in listen state. This reports accurate number for recv and send queue
(pending / max connection backlog counters).
Sample output:
ss -Mil
State Recv-Q Send-Q Local Address:Port Peer Address:Port
LISTEN 0 20 127.0.0.1:12000 0.0.0.0:*
subflows_max:2
Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Stable-dep-of: 871019b22d1b ("net: set SOCK_RCU_FREE before inserting socket into hashtable") Signed-off-by: Sasha Levin <sashal@kernel.org>
Raw access to cb->arg[] is deprecated, use a context structure.
Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Stable-dep-of: 871019b22d1b ("net: set SOCK_RCU_FREE before inserting socket into hashtable") Signed-off-by: Sasha Levin <sashal@kernel.org>
Commit 0abd1557e21c added rcu_dereference() for dereferencing ip->i_gl
in gfs2_permission. This now causes lockdep to complain when
gfs2_permission is called in non-RCU context:
WARNING: suspicious RCU usage in gfs2_permission
Switch to rcu_dereference_check() and check for the MAY_NOT_BLOCK flag
to shut up lockdep when we know that dereferencing ip->i_gl is safe.
Fixes: 0abd1557e21c ("gfs2: fix an oops in gfs2_permission") Reported-by: syzbot+3e5130844b0c0e2b4948@syzkaller.appspotmail.com Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
RPC client pipefs dentries cleanup is in separated rpc_remove_pipedir()
workqueue,which takes care about pipefs superblock locking.
In some special scenarios, when kernel frees the pipefs sb of the
current client and immediately alloctes a new pipefs sb,
rpc_remove_pipedir function would misjudge the existence of pipefs
sb which is not the one it used to hold. As a result,
the rpc_remove_pipedir would clean the released freed pipefs dentries.
To fix this issue, rpc_remove_pipedir should check whether the
current pipefs sb is consistent with the original pipefs sb.
If the client is doing pnfs IO and Kerberos is configured and EXCHANGEID
successfully negotiated SP4_MACH_CRED and WRITE/COMMIT are on the
list of state protected operations, then we need to make sure to
choose the DS's rpc_client structure instead of the MDS's one.
Fixes: fb91fb0ee7b2 ("NFS: Move call to nfs4_state_protect_write() to nfs4_write_setup()") Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
This IS_ERR() check was deleted during in a cleanup because, at the time,
the rpcb_call_async() function could not return an error pointer. That
changed in commit 25cf32ad5dba ("SUNRPC: Handle allocation failure in
rpc_new_task()") and now it can return an error pointer. Put the check
back.
A related revert was done in commit 13bd90141804 ("Revert "SUNRPC:
Remove unreachable error condition"").
If connect() is returning ECONNRESET, it usually means that nothing is
listening on that port. If so, a rebind might be required in order to
obtain the new port on which the RPC service is listening.
Fixes: fd01b2597941 ("SUNRPC: ECONNREFUSED should cause a rebind.") Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
'meson' directory contains two separate drivers, so it should be added
to Makefile compilation hierarchy unconditionally, because otherwise the
meson-ao-cec-g12a won't be compiled if meson-ao-cec is not selected.
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com> Fixes: 4be5e8648b0c ("media: move CEC platform drivers to a separate directory") Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Sasha Levin <sashal@kernel.org>
The notifier is added to the global notifier list when registered. When
the module is removed, the struct csi2rx_priv in which the notifier is
embedded, is destroyed. As a result the notifier list has a reference to
a notifier that no longer exists. This causes invalid memory accesses
when the list is iterated over. Similar for when the probe fails.
Unregister and clean up the notifier to avoid this.
Fixes: 1fc3b37f34f6 ("media: v4l: cadence: Add Cadence MIPI-CSI2 RX driver") Signed-off-by: Pratyush Yadav <p.yadav@ti.com> Tested-by: Julien Massot <julien.massot@collabora.com> Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com> Reviewed-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com> Reviewed-by: Maxime Ripard <mripard@kernel.org> Signed-off-by: Jai Luthra <j-luthra@ti.com> Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Sasha Levin <sashal@kernel.org>
Except on x86, preempt_count is always accessed with READ_ONCE().
Repeated invocations in macros like irq_count() produce repeated loads.
These redundant instructions appear in various fast paths. In the one
shown below, for example, irq_count() is evaluated during kernel entry
if !tick_nohz_full_cpu(smp_processor_id()).
This patch doesn't prevent the pointless btst and beqs instructions
above, but it does eliminate 2 of the 3 pointless move instructions
here and elsewhere.
On x86, preempt_count is per-cpu data and the problem does not arise
presumably because the compiler is free to optimize more effectively.
This patch was tested on m68k and x86. I was expecting no changes
to object code for x86 and mostly that's what I saw. However, there
were a few places where code generation was perturbed for some reason.
The performance issue addressed here is minor on uniprocessor m68k. I
got a 0.01% improvement from this patch for a simple "find /sys -false"
benchmark. For architectures and workloads susceptible to cache line bounce
the improvement is expected to be larger. The only SMP architecture I have
is x86, and as x86 unaffected I have not done any further measurements.
Now that there are three different instances of doing the addition trick
to the preempt_count() and NMI_MASK, HARDIRQ_MASK and SOFTIRQ_OFFSET
macros, it deserves a helper function defined in the preempt.h header.
Add the interrupt_context_level() helper and replace the three instances
that do that logic with it.
Instead of having branches that adds noise to the branch prediction, use
the addition logic to set the bit for the level of interrupt context that
the state is currently in. This copies the logic from perf's
get_recursion_context() function.
Currently we are setting the rate in the tx cmd for
mgmt frames (e.g. during connection establishment).
This was problematic when sending mgmt frames in eSR mode,
as we don't know what link this frame will be sent on
(This is decided by the FW), so we don't know what is the
lowest rate.
Fix this by not setting the rate in tx cmd and rely
on FW to choose the right one.
Set rate only for injected frames with fixed rate,
or when no sta is given.
Also set for important frames (EAPOL etc.) the High Priority flag.
Fixes: 055b22e770dd ("iwlwifi: mvm: Set Tx rate and flags when there is not station") Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20230913145231.6c7e59620ee0.I6eaed3ccdd6dd62b9e664facc484081fc5275843@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
These enums are passed to set/test_bit(). The set/test_bit() functions
take a bit number instead of a shifted value. Passing a shifted value
is a double shift bug like doing BIT(BIT(1)). The double shift bug
doesn't cause a problem here because we are only checking 0 and 1 but
if the value was 5 or above then it can lead to a buffer overflow.
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org> Reviewed-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Reviewed-by: Sam Protsenko <semen.protsenko@linaro.org> Signed-off-by: Thierry Reding <thierry.reding@gmail.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
When software 'pci unplug' using IGT is executed we got a sysfs directory
entry is NULL for differant ras blocks like hdp, umc, etc.
Before call 'sysfs_remove_file_from_group' and 'sysfs_remove_group'
check that 'sd' is not NULL.
The allocated memory for qdev->dumb_heads should be released
in qxl_destroy_monitors_object before qxl suspend.
otherwise,qxl_create_monitors_object will be called to
reallocate memory for qdev->dumb_heads after qxl resume,
it will cause memory leak.
We need to check for an active device as otherwise we get warnings
for some mcbsp instances for "Runtime PM usage count underflow!".
Reported-by: Andreas Kemnade <andreas@kemnade.info> Signed-off-by: Tony Lindgren <tony@atomide.com> Link: https://lore.kernel.org/r/20231030052340.13415-1-tony@atomide.com Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
When entering kdb/kgdb on a kernel panic, it was be observed that the
console isn't flushed before the `kdb` prompt came up. Specifically,
when using the buddy lockup detector on arm64 and running:
echo HARDLOCKUP > /sys/kernel/debug/provoke-crash/DIRECT
I could see:
[ 26.161099] lkdtm: Performing direct entry HARDLOCKUP
[ 32.499881] watchdog: Watchdog detected hard LOCKUP on cpu 6
[ 32.552865] Sending NMI from CPU 5 to CPUs 6:
[ 32.557359] NMI backtrace for cpu 6
... [backtrace for cpu 6] ...
[ 32.558353] NMI backtrace for cpu 5
... [backtrace for cpu 5] ...
[ 32.867471] Sending NMI from CPU 5 to CPUs 0-4,7:
[ 32.872321] NMI backtrace forP cpuANC: Hard LOCKUP
Entering kdb (current=..., pid 0) on processor 5 due to Keyboard Entry
[5]kdb>
As you can see, backtraces for the other CPUs start printing and get
interleaved with the kdb PANIC print.
Let's replicate the commands to flush the console in the kdb panic
entry point to avoid this.
[Why & How]
Check whether assigned timing generator is NULL or not before
accessing its funcs to prevent NULL dereference.
Reviewed-by: Jun Lei <jun.lei@amd.com> Acked-by: Hersen Wu <hersenxs.wu@amd.com> Signed-off-by: Wayne Lin <wayne.lin@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
imon driver probes two USB interfaces, and at the probe of the second
interface, the driver assumes blindly that the first interface got
bound with the same imon driver. It's usually true, but it's still
possible that the first interface is bound with another driver via a
malformed descriptor. Then it may lead to a memory corruption, as
spotted by syzkaller; imon driver accesses the data from drvdata as
struct imon_context object although it's a completely different one
that was assigned by another driver.
This patch adds a sanity check -- whether the first interface is
really bound with the imon driver or not -- for avoiding the problem
above at the probe time.
Reported-by: syzbot+59875ffef5cb9c9b29e9@syzkaller.appspotmail.com Closes: https://lore.kernel.org/all/000000000000a838aa0603cc74d6@google.com/ Tested-by: Ricardo B. Marliere <ricardo@marliere.net> Link: https://lore.kernel.org/r/20230922005152.163640-1-ricardo@marliere.net Signed-off-by: Takashi Iwai <tiwai@suse.de> Signed-off-by: Sean Young <sean@mess.org> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Sasha Levin <sashal@kernel.org>
Fix documentation for struct ccs_quirk, a device specific struct for
managing deviations from the standard. The flags field was drifted away
from where it should have been.
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com> Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Sasha Levin <sashal@kernel.org>
In RCU mode, we might race with gfs2_evict_inode(), which zeroes
->i_gl. Freeing of the object it points to is RCU-delayed, so
if we manage to fetch the pointer before it's been replaced with
NULL, we are fine. Check if we'd fetched NULL and treat that
as "bail out and tell the caller to get out of RCU mode".
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
When lots of quota changes are made, there may be cases in which an
inode's quota information is increased and then decreased, such as when
blocks are added to a file, then deleted from it. If the timing is
right, function do_qc can add pending quota changes to a transaction,
then later, another call to do_qc can negate those changes, resulting
in a net gain of 0. The quota_change information is recorded in the qc
buffer (and qd element of the inode as well). The buffer is added to the
transaction by the first call to do_qc, but a subsequent call changes
the value from non-zero back to zero. At that point it's too late to
remove the buffer_head from the transaction. Later, when the quota sync
code is called, the zero-change qd element is discovered and flagged as
an assert warning. If the fs is mounted with errors=panic, the kernel
will panic.
This is usually seen when files are truncated and the quota changes are
negated by punch_hole/truncate which uses gfs2_quota_hold and
gfs2_quota_unhold rather than block allocations that use gfs2_quota_lock
and gfs2_quota_unlock which automatically do quota sync.
This patch solves the problem by adding a check to qd_check_sync such
that net-zero quota changes already added to the transaction are no
longer deemed necessary to be synced, and skipped.
In this case references are taken for the qd and the slot from do_qc
so those need to be put. The normal sequence of events for a normal
non-zero quota change is as follows:
In the net-zero change case, we add a check to qd_check_sync so it puts
the qd and slot references acquired in gfs2_quota_change and skip the
unneeded sync.
Signed-off-by: Bob Peterson <rpeterso@redhat.com> Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
drivers/media/test-drivers/vivid/vivid-rds-gen.c: In function 'vivid_rds_gen_fill':
drivers/media/test-drivers/vivid/vivid-rds-gen.c:147:56: warning: '.' directive output may be truncated writing 1 byte into a region of size between 0 and 3 [-Wformat-truncation=]
147 | snprintf(rds->psname, sizeof(rds->psname), "%6d.%1d",
| ^
drivers/media/test-drivers/vivid/vivid-rds-gen.c:147:52: note: directive argument in the range [0, 9]
147 | snprintf(rds->psname, sizeof(rds->psname), "%6d.%1d",
| ^~~~~~~~~
drivers/media/test-drivers/vivid/vivid-rds-gen.c:147:9: note: 'snprintf' output between 9 and 12 bytes into a destination of size 9
147 | snprintf(rds->psname, sizeof(rds->psname), "%6d.%1d",
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
148 | freq / 16, ((freq & 0xf) * 10) / 16);
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Syzkaller reported the following issue:
UBSAN: shift-out-of-bounds in drivers/media/usb/gspca/cpia1.c:1031:27
shift exponent 245 is too large for 32-bit type 'int'
When the value of the variable "sd->params.exposure.gain" exceeds the
number of bits in an integer, a shift-out-of-bounds error is reported. It
is triggered because the variable "currentexp" cannot be left-shifted by
more than the number of bits in an integer. In order to avoid invalid
range during left-shift, the conditional expression is added.
Reported-by: syzbot+e27f3dbdab04e43b9f73@syzkaller.appspotmail.com Closes: https://lore.kernel.org/all/20230818164522.12806-1-coolrrsh@gmail.com Link: https://syzkaller.appspot.com/bug?extid=e27f3dbdab04e43b9f73 Signed-off-by: Rajeshwar R Shinde <coolrrsh@gmail.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Sasha Levin <sashal@kernel.org>
The `i3c_master_bus_init` function may attach the I2C devices before the
I3C bus initialization. In this flow, the DAT `alloc_entry`` will be used
before the DAT `init`. Additionally, if the `i3c_master_bus_init` fails,
the DAT `cleanup` will execute before the device is detached, which will
execue DAT `free_entry` function. The above scenario can cause the driver
to use DAT_data when it is NULL.
The following codes have an implicit conversion from size_t to u32:
(u32)max_size = (size_t)virtio_max_dma_size(vdev);
This may lead overflow, Ex (size_t)4G -> (u32)0. Once
virtio_max_dma_size() has a larger size than U32_MAX, use U32_MAX
instead.
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Message-Id: <20230904061045.510460-1-pizhenwei@bytedance.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Make sure we don't OOPS in case clock-frequency is set to 0 in a DT. The
variable set here is later used as a divisor.
Signed-off-by: Axel Lin <axel.lin@ingics.com> Acked-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Wolfram Sang <wsa@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Do not loop over ring headers in hci_dma_irq_handler() that are not
allocated and enabled in hci_dma_init(). Otherwise out of bounds access
will occur from rings->headers[i] access when i >= number of allocated
ring headers.
W=1 warns about null argument to kprintf:
In file included from fs/9p/xattr.c:12:
In function ‘v9fs_xattr_get’,
inlined from ‘v9fs_listxattr’ at fs/9p/xattr.c:142:9:
include/net/9p/9p.h:55:2: error: ‘%s’ directive argument is null
[-Werror=format-overflow=]
55 | _p9_debug(level, __func__, fmt, ##__VA_ARGS__)
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Use an empty string instead of :
- this is ok 9p-wise because p9pdu_vwritef serializes a null string
and an empty string the same way (one '0' word for length)
- since this degrades the print statements, add new single quotes for
xattr's name delimter (Old: "file = (null)", new: "file = ''")
| BUG: KCSAN: data-race in p9_fd_create / p9_fd_create
|
| read-write to 0xffff888130fb3d48 of 4 bytes by task 15599 on cpu 0:
| p9_fd_open net/9p/trans_fd.c:842 [inline]
| p9_fd_create+0x210/0x250 net/9p/trans_fd.c:1092
| p9_client_create+0x595/0xa70 net/9p/client.c:1010
| v9fs_session_init+0xf9/0xd90 fs/9p/v9fs.c:410
| v9fs_mount+0x69/0x630 fs/9p/vfs_super.c:123
| legacy_get_tree+0x74/0xd0 fs/fs_context.c:611
| vfs_get_tree+0x51/0x190 fs/super.c:1519
| do_new_mount+0x203/0x660 fs/namespace.c:3335
| path_mount+0x496/0xb30 fs/namespace.c:3662
| do_mount fs/namespace.c:3675 [inline]
| __do_sys_mount fs/namespace.c:3884 [inline]
| [...]
|
| read-write to 0xffff888130fb3d48 of 4 bytes by task 15563 on cpu 1:
| p9_fd_open net/9p/trans_fd.c:842 [inline]
| p9_fd_create+0x210/0x250 net/9p/trans_fd.c:1092
| p9_client_create+0x595/0xa70 net/9p/client.c:1010
| v9fs_session_init+0xf9/0xd90 fs/9p/v9fs.c:410
| v9fs_mount+0x69/0x630 fs/9p/vfs_super.c:123
| legacy_get_tree+0x74/0xd0 fs/fs_context.c:611
| vfs_get_tree+0x51/0x190 fs/super.c:1519
| do_new_mount+0x203/0x660 fs/namespace.c:3335
| path_mount+0x496/0xb30 fs/namespace.c:3662
| do_mount fs/namespace.c:3675 [inline]
| __do_sys_mount fs/namespace.c:3884 [inline]
| [...]
|
| value changed: 0x00008002 -> 0x00008802
Within p9_fd_open(), O_NONBLOCK is added to f_flags of the read and
write files. This may happen concurrently if e.g. mounting process
modifies the fd in another thread.
Mark the plain read-modify-writes as intentional data-races, with the
assumption that the result of executing the accesses concurrently will
always result in the same result despite the accesses themselves not
being atomic.
Previously, gadget assignment to the net device occurred exclusively
during the initial binding attempt.
Nevertheless, the gadget pointer could change during bind/unbind
cycles due to various conditions, including the unloading/loading
of the UDC device driver or the detachment/reconnection of an
OTG-capable USB hub device.
This patch relocates the gether_set_gadget() function out from
ncm_opts->bound condition check, ensuring that the correct gadget
is assigned during each bind request.
The provided logs demonstrate the consistency of ncm_opts throughout
the power cycle, while the gadget may change.
* OTG hub connected during boot up and assignment of gadget and
ncm_opts pointer
[ 2.366301] usb 2-1.5: New USB device found, idVendor=2996, idProduct=0105
[ 2.366304] usb 2-1.5: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[ 2.366306] usb 2-1.5: Product: H2H Bridge
[ 2.366308] usb 2-1.5: Manufacturer: Aptiv
[ 2.366309] usb 2-1.5: SerialNumber: 13FEB2021
[ 2.427989] usb 2-1.5: New USB device found, VID=2996, PID=0105
[ 2.428959] dabridge 2-1.5:1.0: dabridge 2-4 total endpoints=5, 0000000093a8d681
[ 2.429710] dabridge 2-1.5:1.0: P(0105) D(22.06.22) F(17.3.16) H(1.1) high-speed
[ 2.429714] dabridge 2-1.5:1.0: Hub 2-2 P(0151) V(06.87)
[ 2.429956] dabridge 2-1.5:1.0: All downstream ports in host mode
* OTG capable hub disconnecte or assume driver unload.
[ 97.203114] usb 2-1: USB disconnect, device number 2
[ 97.203118] usb 2-1.1: USB disconnect, device number 3
[ 97.209217] usb 2-1.5: USB disconnect, device number 4
[ 97.230990] dabr_udc deleted
* Reconnect the OTG hub or load driver assaign new gadget pointer.
[ 111.534035] usb 2-1.1: New USB device found, idVendor=2996, idProduct=0120, bcdDevice= 6.87
[ 111.534038] usb 2-1.1: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[ 111.534040] usb 2-1.1: Product: Vendor
[ 111.534041] usb 2-1.1: Manufacturer: Aptiv
[ 111.534042] usb 2-1.1: SerialNumber: Superior
[ 111.535175] usb 2-1.1: New USB device found, VID=2996, PID=0120
[ 111.610995] usb 2-1.5: new high-speed USB device number 8 using xhci-hcd
[ 111.630052] usb 2-1.5: New USB device found, idVendor=2996, idProduct=0105, bcdDevice=21.02
[ 111.630055] usb 2-1.5: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[ 111.630057] usb 2-1.5: Product: H2H Bridge
[ 111.630058] usb 2-1.5: Manufacturer: Aptiv
[ 111.630059] usb 2-1.5: SerialNumber: 13FEB2021
[ 111.687464] usb 2-1.5: New USB device found, VID=2996, PID=0105
[ 111.690375] dabridge 2-1.5:1.0: dabridge 2-8 total endpoints=5, 000000000d87c961
[ 111.691172] dabridge 2-1.5:1.0: P(0105) D(22.06.22) F(17.3.16) H(1.1) high-speed
[ 111.691176] dabridge 2-1.5:1.0: Hub 2-6 P(0151) V(06.87)
[ 111.691646] dabridge 2-1.5:1.0: All downstream ports in host mode
[ 111.692298] gadget 00000000dc72f7a9 --------> new gadget ptr on connect
* NCM opts and associated gadget pointer during second ncm_bind
[ 113.271786] NCM opts 00000000aa304ac9 -----> same opts ptr used during first bind
[ 113.271788] NCM gadget 00000000dc72f7a9 ----> however new gaget ptr, that will not set
in net_device due to ncm_opts->bound = true
After repairing a corrupted file system with exfatprogs' fsck.exfat,
zero-size directories may result. It is also possible to create
zero-size directories in other exFAT implementation, such as Paragon
ufsd dirver.
As described in the specification, the lower directory size limits
is 0 bytes.
Without this commit, sub-directories and files cannot be created
under a zero-size directory, and it cannot be removed.
Signed-off-by: Yuezhang Mo <Yuezhang.Mo@sony.com> Reviewed-by: Andy Wu <Andy.Wu@sony.com> Reviewed-by: Aoyama Wataru <wataru.aoyama@sony.com> Signed-off-by: Namjae Jeon <linkinjeon@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Due to a hardware issue in A and B steppings of Intel IPU E2000, it expects
wrong endianness in ATS invalidation message body. This problem can lead to
outdated translations being returned as valid and finally cause system
instability.
To prevent such issues, add quirk_intel_e2000_no_ats() to disable ATS for
vulnerable IPU E2000 devices.
Use FIELD_GET() to extract PCIe Negotiated and Maximum Link Width fields
instead of custom masking and shifting.
Link: https://lore.kernel.org/r/20230919125648.1920-7-ilpo.jarvinen@linux.intel.com Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
[bhelgaas: drop duplicate include of <linux/bitfield.h>] Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
fc_lport_ptp_setup() did not check the return value of fc_rport_create()
which can return NULL and would cause a NULL pointer dereference. Address
this issue by checking return value of fc_rport_create() and log error
message on fc_rport_create() failed.
In get_esi() PCI errors are checked inside line-split "if" conditions (in
addition to the file not following the coding style). To make the code in
get_esi() more readable, fix the coding style and use the usual error
handling pattern with a separate variable.
In addition, initialization of 'error' variable at declaration is not
needed.
Use FIELD_GET() to extract PCIe Negotiated Link Width field instead of
custom masking and shifting.
Similarly, change custom code that misleadingly used
PCI_EXP_LNKSTA_NLW_SHIFT to prepare value for PCI_EXP_LNKCAP write
to use FIELD_PREP() with correct field define (PCI_EXP_LNKCAP_MLW).
While AudioDSP drivers assign streams exclusively of HOST or LINK type,
nothing blocks a user to attempt to assign a COUPLED stream. As
supplied substream instance may be a stub, what is the case when
code-loading, such scenario ends with null-ptr-deref.
Place IRQ handlers such as gic_handle_irq() in the irqentry section even
if FUNCTION_GRAPH_TRACER is not enabled. Without this, the stack
depot's filter_irq_stacks() does not correctly filter out IRQ stacks in
those configurations, which hampers deduplication and eventually leads
to "Stack depot reached limit capacity" splats with KASAN.
A similar fix was done for arm64 in commit f6794950f0e5ba37e3bbed
("arm64: set __exception_irq_entry with __irq_entry as a default").
Built-in firmware of cptkbd handles scrolling by itself (when middle
button is pressed) but with issues: it does not support horizontal and
hi-res scrolling and upon middle button release it sends middle button
click even if there was a scrolling event. Commit 3cb5ff0220e3 ("HID:
lenovo: Hide middle-button press until release") workarounds last
issue but it's impossible to workaround scrolling-related issues
without firmware modification.
Likely, Dennis Schneider has reverse engineered the firmware and
provided an instruction on how to patch it [1]. However,
aforementioned workaround prevents userspace (libinput) from knowing
exact moment when middle button has been pressed down and performing
"On-Button scrolling". This commit detects correctly-behaving patched
firmware if cursor movement events has been received during middle
button being pressed and stops applying workaround for this device.
Currently there is not check against the agno of the iag while
allocating new inodes to avoid fragmentation problem. Added the check
which is required.
Currently while searching for dmtree_t for sufficient free blocks there
is an array out of bounds while getting element in tp->dm_stree. To add
the required check for out of bound we first need to determine the type
of dmtree. Thus added an extra parameter to dbFindLeaf so that the type
of tree can be determined and the required check can be applied.
Both db_maxag and db_agpref are used as the index of the
db_agfree array, but there is currently no validity check for
db_maxag and db_agpref, which can lead to errors.
The following is related bug reported by Syzbot:
UBSAN: array-index-out-of-bounds in fs/jfs/jfs_dmap.c:639:20
index 7936 is out of range for type 'atomic_t[128]'
Add checking that the values of db_maxag and db_agpref are valid
indexes for the db_agfree array.
Reported-by: syzbot+38e876a8aa44b7115c76@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=38e876a8aa44b7115c76 Signed-off-by: Juntong Deng <juntong.deng@outlook.com> Signed-off-by: Dave Kleikamp <dave.kleikamp@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
In practice the driver should never send more commands than are allocated
to a queue's event pool. In the unlikely event that this happens, the code
asserts a BUG_ON, and in the case that the kernel is not configured to
crash on panic returns a junk event pointer from the empty event list
causing things to spiral from there. This BUG_ON is a historical artifact
of the ibmvfc driver first being upstreamed, and it is well known now that
the use of BUG_ON is bad practice except in the most unrecoverable
scenario. There is nothing about this scenario that prevents the driver
from recovering and carrying on.
Remove the BUG_ON in question from ibmvfc_get_event() and return a NULL
pointer in the case of an empty event pool. Update all call sites to
ibmvfc_get_event() to check for a NULL pointer and perfrom the appropriate
failure or recovery action.
Signed-off-by: Tyrel Datwyler <tyreld@linux.ibm.com> Link: https://lore.kernel.org/r/20230921225435.3537728-2-tyreld@linux.ibm.com Reviewed-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
If init debugfs failed during device registration due to memory allocation
failure, debugfs_remove_recursive() is called, after which debugfs_dir is
not set to NULL. debugfs_remove_recursive() will be called again during
device removal. As a result, illegal pointer is accessed.
Use FIELD_GET() to extract PCIe Negotiated Link Width field instead of
custom masking and shifting, and remove extract_width() which only
wraps that FIELD_GET().
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Link: https://lore.kernel.org/r/20230919125648.1920-2-ilpo.jarvinen@linux.intel.com Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Reviewed-by: Dean Luick <dean.luick@cornelisnetworks.com> Signed-off-by: Leon Romanovsky <leon@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
We found a hungtask bug in test_aead_vec_cfg as follows:
INFO: task cryptomgr_test:391009 blocked for more than 120 seconds.
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Call trace:
__switch_to+0x98/0xe0
__schedule+0x6c4/0xf40
schedule+0xd8/0x1b4
schedule_timeout+0x474/0x560
wait_for_common+0x368/0x4e0
wait_for_completion+0x20/0x30
wait_for_completion+0x20/0x30
test_aead_vec_cfg+0xab4/0xd50
test_aead+0x144/0x1f0
alg_test_aead+0xd8/0x1e0
alg_test+0x634/0x890
cryptomgr_test+0x40/0x70
kthread+0x1e0/0x220
ret_from_fork+0x10/0x18
Kernel panic - not syncing: hung_task: blocked tasks
For padata_do_parallel, when the return err is 0 or -EBUSY, it will call
wait_for_completion(&wait->completion) in test_aead_vec_cfg. In normal
case, aead_request_complete() will be called in pcrypt_aead_serial and the
return err is 0 for padata_do_parallel. But, when pinst->flags is
PADATA_RESET, the return err is -EBUSY for padata_do_parallel, and it
won't call aead_request_complete(). Therefore, test_aead_vec_cfg will
hung at wait_for_completion(&wait->completion), which will cause
hungtask.
The problem comes as following:
(padata_do_parallel) |
rcu_read_lock_bh(); |
err = -EINVAL; | (padata_replace)
| pinst->flags |= PADATA_RESET;
err = -EBUSY |
if (pinst->flags & PADATA_RESET) |
rcu_read_unlock_bh() |
return err
In order to resolve the problem, we replace the return err -EBUSY with
-EAGAIN, which means parallel_data is changing, and the caller should call
it again.
v3:
remove retry and just change the return err.
v2:
introduce padata_try_do_parallel() in pcrypt_aead_encrypt and
pcrypt_aead_decrypt to solve the hungtask.
Signed-off-by: Lu Jialin <lujialin4@huawei.com> Signed-off-by: Guo Zihua <guozihua@huawei.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: Sasha Levin <sashal@kernel.org>
Add members to struct snd_soc_card to store the PCI subsystem ID (SSID)
of the soundcard.
The PCI specification provides two registers to store a vendor-specific
SSID that can be read by drivers to uniquely identify a particular
"soundcard". This is defined in the PCI specification to distinguish
products that use the same silicon (and therefore have the same silicon
ID) so that product-specific differences can be applied.
PCI only defines 0xFFFF as an invalid value. 0x0000 is not defined as
invalid. So the usual pattern of zero-filling the struct and then
assuming a zero value unset will not work. A flag is included to
indicate when the SSID information has been filled in.
Unlike DMI information, which has a free-format entirely up to the vendor,
the PCI SSID has a strictly defined format and a registry of vendor IDs.
It is usual in Windows drivers that the SSID is used as the sole identifier
of the specific end-product and the Windows driver contains tables mapping
that to information about the hardware setup, rather than using ACPI
properties.
This SSID is important information for ASoC components that need to apply
hardware-specific configuration on PCI-based systems.
As the SSID is a generic part of the PCI specification and is treated as
identifying the "soundcard", it is reasonable to include this information
in struct snd_soc_card, instead of components inventing their own custom
ways to pass this information around.
Signed-off-by: Richard Fitzgerald <rf@opensource.cirrus.com> Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com> Link: https://lore.kernel.org/r/20230912163207.3498161-2-rf@opensource.cirrus.com Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Wrap the usb controllers in an intermediate simple-bus and use it to
constrain the dma address size of these usb controllers to the 40b
that they generate toward the interconnect. This is required because
the SoC uses 48b address sizes and this mismatch would lead to smmu
context faults [1] because the usb generates 40b addresses while the
smmu page tables are populated with 48b wide addresses.
[1]
xhci-hcd xhci-hcd.0.auto: xHCI Host Controller
xhci-hcd xhci-hcd.0.auto: new USB bus registered, assigned bus number 1
xhci-hcd xhci-hcd.0.auto: hcc params 0x0220f66d hci version 0x100 quirks 0x0000000002000010
xhci-hcd xhci-hcd.0.auto: irq 108, io mem 0x03100000
xhci-hcd xhci-hcd.0.auto: xHCI Host Controller
xhci-hcd xhci-hcd.0.auto: new USB bus registered, assigned bus number 2
xhci-hcd xhci-hcd.0.auto: Host supports USB 3.0 SuperSpeed
arm-smmu 5000000.iommu: Unhandled context fault: fsr=0x402, iova=0xffffffb000, fsynr=0x0, cbfrsynra=0xc01, cb=3
In certain types of chips, such as VEGA20, reading the amdgpu_regs_smc file could result in an abnormal null pointer access when the smc_rreg pointer is NULL. Below are the steps to reproduce this issue and the corresponding exception log:
Switching to a different reset sequence, enabling IOVCC before enabling
VCC.
There also needs to be a delay after enabling the supplies and before
deasserting the reset. The datasheet specifies 1ms after the supplies
reach the required voltage. Use 10-20ms to also give the power supplies
some time to reach the required voltage, too.
This fixes intermittent panel initialization failures and screen
corruption during resume from sleep on panel xingbangda,xbd599 (e.g.
used in PinePhone).
Signed-off-by: Ondrej Jirman <megi@xff.cz> Signed-off-by: Frank Oltmanns <frank@oltmanns.dev> Reported-by: Samuel Holland <samuel@sholland.org> Reviewed-by: Guido Günther <agx@sigxcpu.org> Tested-by: Guido Günther <agx@sigxcpu.org> Signed-off-by: Guido Günther <agx@sigxcpu.org> Link: https://patchwork.freedesktop.org/patch/msgid/20230211171748.36692-2-frank@oltmanns.dev Signed-off-by: Sasha Levin <sashal@kernel.org>
In amdgpu_vkms_conn_get_modes(), the return value of drm_cvt_mode()
is assigned to mode, which will lead to a NULL pointer dereference
on failure of drm_cvt_mode(). Add a check to avoid null pointer
dereference.
Signed-off-by: Ma Ke <make_ruc2021@163.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
In tpg110_get_modes(), the return value of drm_mode_duplicate() is
assigned to mode, which will lead to a NULL pointer dereference on
failure of drm_mode_duplicate(). Add a check to avoid npd.
In versatile_panel_get_modes(), the return value of drm_mode_duplicate()
is assigned to mode, which will lead to a NULL pointer dereference
on failure of drm_mode_duplicate(). Add a check to avoid npd.
For pptable structs that use flexible array sizes, use flexible arrays.
Suggested-by: Felix Held <felix.held@amd.com> Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2874 Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The DP CTS test for EDID last block checksum expects the checksum for
the last block, invalid or not. Skip the validity check.
For the most part (*), the EDIDs returned by drm_get_edid() will be
valid anyway, and there's the CTS workaround to get the checksum for
completely invalid EDIDs. See commit 7948fe12d47a ("drm/msm/dp: return
correct edid checksum after corrupted edid checksum read").
This lets us remove one user of drm_edid_block_valid() with hopes the
function can be removed altogether in the future.
(*) drm_get_edid() ignores checksum errors on CTA extensions.
Cc: Abhinav Kumar <quic_abhinavk@quicinc.com> Cc: Dmitry Baryshkov <dmitry.baryshkov@linaro.org> Cc: Kuogee Hsieh <khsieh@codeaurora.org> Cc: Marijn Suijten <marijn.suijten@somainline.org> Cc: Rob Clark <robdclark@gmail.com> Cc: Sean Paul <sean@poorly.run> Cc: Stephen Boyd <swboyd@chromium.org> Cc: linux-arm-msm@vger.kernel.org Cc: freedreno@lists.freedesktop.org Signed-off-by: Jani Nikula <jani.nikula@intel.com> Reviewed-by: Stephen Boyd <swboyd@chromium.org> Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com> Reviewed-by: Kuogee Hsieh <quic_khsieh@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/555361/ Link: https://lore.kernel.org/r/20230901142034.580802-1-jani.nikula@intel.com Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Currently, user array duplications are sometimes done without an
overflow check. Sometimes the checks are done manually; sometimes the
array size is calculated with array_size() and sometimes by calculating
n * size directly in code.
Introduce wrappers for arrays for memdup_user() and vmemdup_user() to
provide a standardized and safe way for duplicating user arrays.
This is both for new code as well as replacing usage of (v)memdup_user()
in existing code that uses, e.g., n * size to calculate array sizes.
Suggested-by: David Airlie <airlied@redhat.com> Signed-off-by: Philipp Stanner <pstanner@redhat.com> Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com> Reviewed-by: Kees Cook <keescook@chromium.org> Reviewed-by: Zack Rusin <zackr@vmware.com> Signed-off-by: Dave Airlie <airlied@redhat.com> Link: https://patchwork.freedesktop.org/patch/msgid/20230920123612.16914-3-pstanner@redhat.com Signed-off-by: Sasha Levin <sashal@kernel.org>
[why]
Clip size increase will increase viewport, which could cause us to
switch to MPC combine.
If we skip full update, we are not able to change to MPC combine in
fast update. This will cause corruption showing on the video plane.
[how]
treat clip size increase of a surface larger than 5k as a full update.
Reviewed-by: Jun Lei <jun.lei@amd.com> Acked-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by: Wenjing Liu <wenjing.liu@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
prange->svm_bo unref can happen in both mmu callback and a callback after
migrate to system ram. Both are async call in different tasks. Sync svm_bo
unref operation to avoid random "use-after-free".
Signed-off-by: Xiaogang Chen <xiaogang.chen@amd.com> Reviewed-by: Philip Yang <Philip.Yang@amd.com> Reviewed-by: Jesse Zhang <Jesse.Zhang@amd.com> Tested-by: Jesse Zhang <Jesse.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
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.
projects / thirdparty / kernel / stable.git / log
