The watch_queue_set_size() allocation error paths return the ret value
set via the prior pipe_resize_ring() call, which will always be zero.
As a result, IOC_WATCH_QUEUE_SET_SIZE callers such as "keyctl watch"
fail to detect kernel wqueue->notes allocation failures and proceed to
KEYCTL_WATCH_KEY, with any notifications subsequently lost.
Fixes: 175295c8c9fa4 ("pipe: Add general notification queue support") Signed-off-by: David Disseldorp <ddiss@suse.de> Signed-off-by: Christian Brauner (Microsoft) <brauner@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
A recent commit moved enabling of runtime PM from adreno_gpu_init() to
adreno_load_gpu() (called on first open()), which means that unbind()
may now be called with runtime PM disabled in case the device was never
opened in between.
Make sure to only forcibly suspend and disable runtime PM at unbind() in
case runtime PM has been enabled to prevent a disable count imbalance.
This specifically avoids leaving runtime PM disabled when the device
is later opened after a successful bind:
msm_dpu ae01000.display-controller: [drm:adreno_load_gpu [msm]] *ERROR* Couldn't power up the GPU: -13
During kexec on ARM device, we notice that device_shutdown() only calls
pm_runtime_force_suspend() while shutting down the GPU. This means the GPU
kthread is still running and further, there maybe active submits.
This causes all kinds of issues during a kexec reboot:
Fix by calling adreno_system_suspend() in the device_shutdown() path.
[ Applied Rob Clark feedback on fixing adreno_unbind() similarly, also
tested as above. ]
Cc: Rob Clark <robdclark@chromium.org> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Ricardo Ribalda <ribalda@chromium.org> Cc: Ross Zwisler <zwisler@kernel.org> Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org> Reviewed-by: Ricardo Ribalda <ribalda@chromium.org> Reviewed-by: Rob Clark <robdclark@gmail.com>
Patchwork: https://patchwork.freedesktop.org/patch/517633/ Link: https://lore.kernel.org/r/20230109222547.1368644-1-joel@joelfernandes.org Signed-off-by: Rob Clark <robdclark@chromium.org>
Stable-dep-of: 9faeae71331e ("drm/msm/adreno: fix runtime PM imbalance at unbind") Signed-off-by: Sasha Levin <sashal@kernel.org>
Reviewed-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Stable-dep-of: 5d459c76613e ("drm/amdgpu/soc21: Add video cap query support for VCN_4_0_4") Signed-off-by: Sasha Levin <sashal@kernel.org>
As lockdep properly warns, we should not be locking i_rwsem while having
transactions started as the proper lock ordering used by all directory
handling operations is i_rwsem -> transaction start. Fix the lock
ordering by moving the locking of the directory earlier in
ext4_rename().
This patch fixes a return value check in kfd doorbell handling.
This function should return 0(error) only when the ida_simple_get
returns < 0(error), return > 0 is a success case.
Cc: Felix Kuehling <Felix.Kuehling@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Fixes: 17cd15143613 ("drm/amdkfd: Allocate doorbells only when needed") Acked-by: Christian Koenig <chriatian.koenig@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Shashank Sharma <shashank.sharma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
We're currently using stop_machine() to update ftrace & kprobes, which
means that the thread that takes text_mutex during may not be the same
as the thread that eventually patches the code. This isn't actually a
race because the lock is still held (preventing any other concurrent
accesses) and there is only one thread running during stop_machine(),
but it does trigger a lockdep failure.
This patch just elides the lockdep check during stop_machine.
When CONFIG_FRAME_POINTER is unset, the stack unwinding function
walk_stackframe randomly reads the stack and then, when KASAN is enabled,
it can lead to the following backtrace:
Let's revert commit 5cccc5fbb265 ("erofs: fix kvcalloc() misuse with
__GFP_NOFAIL") since kvmalloc() already supports __GFP_NOFAIL in commit 318e576ba783 ("mm: allow !GFP_KERNEL allocations for kvmalloc"). So
the original fix was wrong.
Actually there was some issue as [1] discussed, so before that mm fix
is landed, the warn could still happen but applying this commit first
will cause less.
The MT7530 switch from the MT7621 SoC has 2 ports which can be set up as
internal: port 5 and 6. Arınç reports that the GMAC1 attached to port 5
receives corrupted frames, unless port 6 (attached to GMAC0) has been
brought up by the driver. This is true regardless of whether port 5 is
used as a user port or as a CPU port (carrying DSA tags).
Offline debugging (blind for me) which began in the linked thread showed
experimentally that the configuration done by the driver for port 6
contains a step which is needed by port 5 as well - the write to
CORE_GSWPLL_GRP2 (note that I've no idea as to what it does, apart from
the comment "Set core clock into 500Mhz"). Prints put by Arınç show that
the reset value of CORE_GSWPLL_GRP2 is RG_GSWPLL_POSDIV_500M(1) |
RG_GSWPLL_FBKDIV_500M(40) (0x128), both on the MCM MT7530 from the
MT7621 SoC, as well as on the standalone MT7530 from MT7623NI Bananapi
BPI-R2. Apparently, port 5 on the standalone MT7530 can work under both
values of the register, while on the MT7621 SoC it cannot.
The call path that triggers the register write is:
mt753x_phylink_mac_config() for port 6
-> mt753x_pad_setup()
-> mt7530_pad_clk_setup()
so this fully explains the behavior noticed by Arınç, that bringing port
6 up is necessary.
The simplest fix for the problem is to extract the register writes which
are needed for both port 5 and 6 into a common mt7530_pll_setup()
function, which is called at mt7530_setup() time, immediately after
switch reset. We can argue that this mirrors the code layout introduced
in mt7531_setup() by commit 39e6fca1cede ("net: mt7531: only do PLL once
after the reset"), in that the PLL setup has the exact same positioning,
and further work to consolidate the separate setup() functions is not
hindered.
Testing confirms that:
- the slight reordering of writes to MT7530_P6ECR and to
CORE_GSWPLL_GRP1 / CORE_GSWPLL_GRP2 introduced by this change does not
appear to cause problems for the operation of port 6 on MT7621 and on
MT7623 (where port 5 also always worked)
- packets sent through port 5 are not corrupted anymore, regardless of
whether port 6 is enabled by phylink or not (or even present in the
device tree)
My algorithm for determining the Fixes: tag is as follows. Testing shows
that some logic from mt7530_pad_clk_setup() is needed even for port 5.
Prior to commit 6ae43c9d6729 ("net: dsa: mt7530: Convert to PHYLINK
API"), a call did exist for all phy_is_pseudo_fixed_link() ports - so
port 5 included. That commit replaced it with a temporary "Port 5 is not
supported!" comment, and the following commit 347b5aa9eb37 ("net: dsa:
mt7530: Add support for port 5") replaced that comment with a
configuration procedure in mt7530_setup_port5() which was insufficient
for port 5 to work. I'm laying the blame on the patch that claimed
support for port 5, although one would have also needed the change from
commit 1b0f892ff949 ("net: dsa: mt7530: setup core clock even in TRGMII
mode") for the write to be performed completely independently from port
6's configuration.
Thanks go to Arınç for describing the problem, for debugging and for
testing.
NDC caches contexts of frequently used queue's (Rx and Tx queues)
contexts. Due to a HW errata when NDC detects fault/poision while
accessing contexts it could go into an illegal state where a cache
line could get locked forever. To makesure all cache lines in NDC
are available for optimum performance upon fault/lockerror/posion
errors scan through all cache lines in NDC and clear the lock bit.
Fixes: d5ff4853f5b8 ("octeontx2-af: NPA AQ instruction enqueue support") Signed-off-by: Suman Ghosh <sumang@marvell.com> Signed-off-by: Sunil Kovvuri Goutham <sgoutham@marvell.com> Signed-off-by: Sai Krishna <saikrishnag@marvell.com> Reviewed-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Before determining whether the msg has unsupported options, it has been
prematurely terminated by the wrong status check.
For the application, the general usages of MSG_FASTOPEN likes
fd = socket(...)
/* rather than connect */
sendto(fd, data, len, MSG_FASTOPEN)
Hence, We need to check the flag before state check, because the sock
state here is always SMC_INIT when applications tries MSG_FASTOPEN.
Once we found unsupported options, fallback it to TCP.
Fixes: ecf28acb843b ("net/smc: handle sockopts forcing fallback") Signed-off-by: D. Wythe <alibuda@linux.alibaba.com> Signed-off-by: Simon Horman <simon.horman@corigine.com>
v2 -> v1: Optimize code style Reviewed-by: Tony Lu <tonylu@linux.alibaba.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
When creating the TLV to send to the FW for configuring DSCP mode PFC,the
PFCENABLE field was being masked with a 4 bit mask (0xF), but this is an 8
bit bitmask for enabled classes for PFC. This means that traffic classes
4-7 could not be enabled for PFC.
Remove the mask completely, as it is not necessary, as we are assigning 8
bits to an 8 bit field.
Fixes: c56146c4c21c ("ice: Add DSCP support") Signed-off-by: Dave Ertman <david.m.ertman@intel.com> Signed-off-by: Karen Ostrowska <karen.ostrowska@intel.com> Tested-by: Gurucharan G <gurucharanx.g@intel.com> (A Contingent worker at Intel) Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
[1210423.486503] WARNING: CPU: 8 PID: 1524732 at fs/ext4/ext4_jbd2.c:75 ext4_journal_check_start+0x68/0xb0
Reported-by: <flole@flole.de> Suggested-by: Jan Kara <jack@suse.cz> Link: https://bugzilla.kernel.org/show_bug.cgi?id=217123 Fixes: 459ceebfe186 ("nfsd: use vfs_iter_read/write") Reviewed-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
If disk_scan_partitions() is called with 'FMODE_EXCL',
blkdev_get_by_dev() will be called without 'FMODE_EXCL', however, follow
blkdev_put() is still called with 'FMODE_EXCL', which will cause
'bd_holders' counter to leak.
Fix the problem by using the right mode for blkdev_put().
Reported-by: syzbot+2bcc0d79e548c4f62a59@syzkaller.appspotmail.com Link: https://lore.kernel.org/lkml/f9649d501bc8c3444769418f6c26263555d9d3be.camel@linux.ibm.com/T/ Tested-by: Julian Ruess <julianr@linux.ibm.com> Fixes: 92ef14daac46 ("block: fix scan partition for exclusively open device again") Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@kernel.dk> Signed-off-by: Sasha Levin <sashal@kernel.org>
REGMAP is a hidden (not user visible) symbol. Users cannot set it
directly thru "make *config", so drivers should select it instead of
depending on it if they need it.
Consistently using "select" or "depends on" can also help reduce
Kconfig circular dependency issues.
Therefore, change the use of "depends on REGMAP" to "select REGMAP".
Fixes: 04776abc3efa ("platform/x86: mlx-platform: Add physical bus number auto detection") Signed-off-by: Randy Dunlap <rdunlap@infradead.org> Cc: Vadim Pasternak <vadimp@mellanox.com> Cc: Darren Hart <dvhart@infradead.org> Cc: Hans de Goede <hdegoede@redhat.com> Cc: Mark Gross <markgross@kernel.org> Cc: platform-driver-x86@vger.kernel.org Link: https://lore.kernel.org/r/20230226053953.4681-7-rdunlap@infradead.org Signed-off-by: Hans de Goede <hdegoede@redhat.com> Reviewed-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
REGMAP is a hidden (not user visible) symbol. Users cannot set it
directly thru "make *config", so drivers should select it instead of
depending on it if they need it.
Consistently using "select" or "depends on" can also help reduce
Kconfig circular dependency issues.
Therefore, change the use of "depends on REGMAP" to "select REGMAP".
For NVSW_SN2201, select REGMAP_I2C instead of depending on it.
Fixes: 627642557ce5 ("platform/mellanox: mlxreg-hotplug: Modify to use a regmap interface") Fixes: 3971fc1bddc1 ("platform/mellanox: Introduce support for Mellanox register access driver") Fixes: d9bc053dd186 ("platform/mellanox: mlxreg-lc: Add initial support for Nvidia line card devices") Fixes: d6a845788cec ("platform/mellanox: Add support for new SN2201 system") Signed-off-by: Randy Dunlap <rdunlap@infradead.org> Cc: Darren Hart <dvhart@infradead.org> Cc: Hans de Goede <hdegoede@redhat.com> Cc: Michael Shych <michaelsh@nvidia.com> Cc: Mark Gross <markgross@kernel.org> Cc: Vadim Pasternak <vadimp@nvidia.com> Cc: platform-driver-x86@vger.kernel.org Link: https://lore.kernel.org/r/20230226053953.4681-6-rdunlap@infradead.org Signed-off-by: Hans de Goede <hdegoede@redhat.com> Reviewed-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Customers using GKE 1.25 and 1.26 are facing conntrack issues
root caused to commit e62fa7b5eb99 ("netfilter: conntrack: make
max chain length random").
Even if we assume Uniform Hashing, a bucket often reachs 8 chained
items while the load factor of the hash table is smaller than 0.5
With a limit of 16, we reach load factors of 3.
With a limit of 32, we reach load factors of 11.
With a limit of 40, we reach load factors of 15.
With a limit of 50, we reach load factors of 24.
This patch changes MIN_CHAINLEN to 50, to minimize risks.
Ideally, we could in the future add a cushion based on expected
load factor (2 * nf_conntrack_max / nf_conntrack_buckets),
because some setups might expect unusual values.
Fixes: e62fa7b5eb99 ("netfilter: conntrack: make max chain length random") Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
When the sd driver revalidates host-managed SMR disks, it calls
disk_set_zoned() which changes the zone_write_granularity attribute value
to the logical block size regardless of the device type. After that, the sd
driver overwrites the value in sd_zbc_read_zone() with the physical block
size, since ZBC/ZAC requires this for host-managed disks. Between the calls
to disk_set_zoned() and sd_zbc_read_zone(), there exists a window where the
attribute shows the logical block size as the zone_write_granularity value,
which is wrong for host-managed disks. The duration of the window is from
20ms to 200ms, depending on report zone command execution time.
To avoid the wrong zone_write_granularity value between disk_set_zoned()
and sd_zbc_read_zone(), modify the value not in sd_zbc_read_zone() but
just after disk_set_zoned() call.
Fixes: 172572cc4fbf ("block: introduce zone_write_granularity limit") Signed-off-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com> Link: https://lore.kernel.org/r/20230306063024.3376959-1-shinichiro.kawasaki@wdc.com Reviewed-by: Damien Le Moal <damien.lemoal@opensource.wdc.com> Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The firmware only supports Logical Disk IDs up to 240 and LD ID 255 (0xFF)
is reserved for deleted LDs. However, in some cases, firmware was assigning
LD ID 254 (0xFE) to deleted LDs and this was causing the driver to mark the
wrong disk as deleted. This in turn caused the wrong disk device to be
taken offline by the SCSI midlayer.
To address this issue, limit the LD ID range from 255 to 240. This ensures
the deleted LD ID is properly identified and removed by the driver without
accidently deleting any valid LDs.
Fixes: 4d5cbca6306a ("scsi: megaraid_sas: Early detection of VD deletion through RaidMap update") Reported-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Chandrakanth Patil <chandrakanth.patil@broadcom.com> Signed-off-by: Sumit Saxena <sumit.saxena@broadcom.com> Link: https://lore.kernel.org/r/20230302105342.34933-2-chandrakanth.patil@broadcom.com Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Adrien reports that incorrect data is transmitted when a single
page straddles multiple records. We would transmit the same
data in all iterations of the loop.
Fix data corruption issue with SerDes connected PHYs operating at 1.25
Gbps speed where we could previously observe about 30% packet loss while
the bad packet counter was increasing.
As almost all boards with MediaTek MT7622 or MT7986 use either the MT7531
switch IC operating at 3.125Gbps SerDes rate or single-port PHYs using
rate-adaptation to 2500Base-X mode, this issue only got exposed now when
we started trying to use SFP modules operating with 1.25 Gbps with the
BananaPi R3 board.
The fix is to set bit 12 which disables the RX FIFO clear function when
setting up MAC MCR, MediaTek SDK did the same change stating:
"If without this patch, kernel might receive invalid packets that are
corrupted by GMAC."[1]
Currently link up can't be detected in forced mode if polling
isn't used. Only link up interrupt source we have is aneg
complete which isn't applicable in forced mode. Therefore we
have to use energy-on as link up indicator.
Fixes: 4ffe361110db ("net: phy: smsc: skip ENERGYON interrupt if disabled") Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
The last invocation of btf_datasec_resolve should invoke btf_var_resolve
by means of env_stack_push, instead it returns EINVAL. The reason is that
env_stack_push is never executed for the second VAR.
if (!env_type_is_resolve_sink(env, var_type) &&
!env_type_is_resolved(env, var_type_id)) {
env_stack_set_next_member(env, i + 1);
return env_stack_push(env, var_type, var_type_id);
}
env_type_is_resolve_sink() changes its behaviour based on resolve_mode.
For RESOLVE_PTR, we can simplify the if condition to the following:
Since we're dealing with a VAR the clause evaluates to false. This is
not sufficient to trigger the bug however. The log output and EINVAL
are only generated if btf_type_id_size() fails.
Most types are sized, so for example a VAR referring to an INT is not a
problem. The bug is only triggered if a VAR points at a modifier. Since
we skipped btf_var_resolve that modifier was also never resolved, which
means that btf_resolved_type_id returns 0 aka VOID for the modifier.
This in turn causes btf_type_id_size to return NULL, triggering EINVAL.
To summarise, the following conditions are necessary:
- VAR pointing at PTR, STRUCT, UNION or ARRAY
- Followed by a VAR pointing at TYPEDEF, VOLATILE, CONST, RESTRICT or
TYPE_TAG
The fix is to reset resolve_mode to RESOLVE_TBD before attempting to
resolve a VAR from a DATASEC.
Fixes: 5cc8aa5eb834 ("bpf: kernel side support for BTF Var and DataSec") Signed-off-by: Lorenz Bauer <lmb@isovalent.com> Link: https://lore.kernel.org/r/20230306112138.155352-2-lmb@isovalent.com Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
It's always the case and e.g. xdp_convert_buff_to_frame() relies on
this.
IOW, the following:
for (u32 i = 0; i < 0xdead; i++) {
xdpf = xdp_convert_buff_to_frame(&xdp);
xdp_convert_frame_to_buff(xdpf, &xdp);
}
shouldn't ever modify @xdpf's contents or the pointer itself.
However, "live packet" code wrongly treats &xdp_frame as part of its
context placed *before* the data_hard_start. With such flow,
data_hard_start is sizeof(*xdpf) off to the right and no longer points
to the XDP frame.
Instead of replacing `sizeof(ctx)` with `offsetof(ctx, xdpf)` in several
places and praying that there are no more miscalcs left somewhere in the
code, unionize ::frm with ::data in a flex array, so that both starts
pointing to the actual data_hard_start and the XDP frame actually starts
being a part of it, i.e. a part of the headroom, not the context.
A nice side effect is that the maximum frame size for this mode gets
increased by 40 bytes, as xdp_buff::frame_sz includes everything from
data_hard_start (-> includes xdpf already) to the end of XDP/skb shared
info.
Also update %MAX_PKT_SIZE accordingly in the selftests code. Leave it
hardcoded for 64 bit && 4k pages, it can be made more flexible later on.
Minor: align `&head->data` with how `head->frm` is assigned for
consistency.
Minor #2: rename 'frm' to 'frame' in &xdp_page_head while at it for
clarity.
(was found while testing XDP traffic generator on ice, which calls
xdp_convert_frame_to_buff() for each XDP frame)
Fixes: c23b8c0ec3f7 ("bpf: Add "live packet" mode for XDP in BPF_PROG_RUN") Acked-by: Toke Høiland-Jørgensen <toke@redhat.com> Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com> Link: https://lore.kernel.org/r/20230215185440.4126672-1-aleksander.lobakin@intel.com Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
At btrfs_drop_extent_map_range() we are clearing the EXTENT_FLAG_LOGGING
bit on a 'flags' variable that was not initialized. This makes static
checkers complain about it, so initialize the 'flags' variable before
clearing the bit.
In practice this has no consequences, because EXTENT_FLAG_LOGGING should
not be set when btrfs_drop_extent_map_range() is called, as an fsync locks
the inode in exclusive mode, locks the inode's mmap semaphore in exclusive
mode too and it always flushes all delalloc.
Also add a comment about why we clear EXTENT_FLAG_LOGGING on a copy of the
flags of the split extent map.
Reported-by: Dan Carpenter <error27@gmail.com> Link: https://lore.kernel.org/linux-btrfs/Y%2FyipSVozUDEZKow@kili/ Fixes: 6d5f4f2e44e4 ("btrfs: drop extent map range more efficiently") Signed-off-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
When booting with an initial ramdisk on platforms where physical memory
does not start at address zero (e.g. on Amiga):
initrd: 0ef0602c - 0f800000
Zone ranges:
DMA [mem 0x0000000008000000-0x000000f7ffffffff]
Normal empty
Movable zone start for each node
Early memory node ranges
node 0: [mem 0x0000000008000000-0x000000000f7fffff]
Initmem setup node 0 [mem 0x0000000008000000-0x000000000f7fffff]
Unable to handle kernel access at virtual address (ptrval)
Oops: 00000000
Modules linked in:
PC: [<00201d3c>] memcmp+0x28/0x56
As phys_to_virt() relies on m68k_memoffset and module_fixup(), it must
not be called before paging_init(). Hence postpone the phys_to_virt
handling for the initial ramdisk until after calling paging_init().
While at it, reduce #ifdef clutter by using IS_ENABLED() instead.
The xtables packet traverser performs an unconditional local_bh_disable(),
but the nf_tables evaluation loop does not.
Functions that are called from either xtables or nftables must assume
that they can be called in process context.
inet_twsk_deschedule_put() assumes that no softirq interrupt can occur.
If tproxy is used from nf_tables its possible that we'll deadlock
trying to aquire a lock already held in process context.
Add a small helper that takes care of this and use it.
It seems that change was unintentional, we have userspace code that
needs the mark while listening for events like REPLY, DESTROY, etc.
Also include 0-marks in requested dumps, as they were before that fix.
Fixes: c7296b10ea6b ("netfilter: ctnetlink: fix compilation warning after data race fixes in ct mark") Signed-off-by: Ivan Delalande <colona@arista.com> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The driver needs to keep track of all the possible concurrent TPA (GRO/LRO)
completions on the aggregation ring. On P5 chips, the maximum number
of concurrent TPA is 256 and the amount of memory we allocate is order-5
on systems using 4K pages. Memory allocation failure has been reported:
Instead of allocating this big chunk of memory and dividing it up for the
concurrent TPA instances, allocate each small chunk separately for each
TPA instance. This will reduce it to order-0 allocations.
Fixes: 3ad87a400c64 ("bnxt_en: Expand bnxt_tpa_info struct to support 57500 chips.") Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com> Reviewed-by: Damodharam Ammepalli <damodharam.ammepalli@broadcom.com> Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com> Signed-off-by: Michael Chan <michael.chan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
The locking in phy_probe() and phy_remove() does very little to prevent
any races with e.g. phy_attach_direct(), but instead causes lockdep ABBA
warnings. Remove it.
======================================================
WARNING: possible circular locking dependency detected
6.2.0-dirty #1108 Tainted: G W E
------------------------------------------------------
ip/415 is trying to acquire lock: ffff5c268f81ef50 (&dev->lock){+.+.}-{3:3}, at: phy_attach_direct+0x17c/0x3a0 [libphy]
but task is already holding lock: ffffaef6496cb518 (rtnl_mutex){+.+.}-{3:3}, at: rtnetlink_rcv_msg+0x154/0x560
which lock already depends on the new lock.
the existing dependency chain (in reverse order) is:
Fixes: d692bc994712 ("net: phy: add core phylib sfp support") Reported-by: Marc Zyngier <maz@kernel.org> Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
When MAC is not support PMT, driver will check PHY's WoL capability
and set device wakeup capability in stmmac_init_phy(). We can enable
the WoL through ethtool, the driver would enable the device wake up
flag. Now the device_may_wakeup() return true.
But if there is a way which enable the PHY's WoL capability derectly,
like in BIOS. The driver would not know the enable thing and would not
set the device wake up flag. The phy_suspend may failed like this:
[ 32.409063] PM: dpm_run_callback(): mdio_bus_phy_suspend+0x0/0x50 returns -16
[ 32.409065] PM: Device stmmac-1:00 failed to suspend: error -16
[ 32.409067] PM: Some devices failed to suspend, or early wake event detected
Add to set the device wakeup enable flag according to the get_wol
function result in PHY can fix the error in this scene.
v2: add a Fixes tag.
Fixes: cd554d33c29a ("net: stmmac: Support WOL with phy") Signed-off-by: Rongguang Wei <weirongguang@kylinos.cn> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
DMA2 and DMA3 planes on msm8998 should use corresponding DMA2 and DMA3
clocks rather than CURSOR0/1 clocks (which are used for the CURSOR
planes). Correct corresponding SSPP declarations.
In this case, "copied" always is 0, the infinite loop occurs.
According to the Linux system call man page, 0 should be returned in this
case. Therefore, in tcp_bpf_recvmsg_parser(), if the length is 0, directly
return. Also modify several other functions with the same problem.
Fixes: 698fcd555908 ("udp: Implement udp_bpf_recvmsg() for sockmap") Fixes: efd2512dd28c ("af_unix: Implement unix_dgram_bpf_recvmsg()") Fixes: 653d74294f4d ("bpf, sockmap: Fix race in ingress receive verdict with redirect to self") Fixes: 38506f4bbc9d ("bpf, sockmap: convert to generic sk_msg interface") Signed-off-by: Liu Jian <liujian56@huawei.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: John Fastabend <john.fastabend@gmail.com> Cc: Jakub Sitnicki <jakub@cloudflare.com> Link: https://lore.kernel.org/bpf/20230303080946.1146638-1-liujian56@huawei.com Signed-off-by: Sasha Levin <sashal@kernel.org>
ice_get_module_eeprom() is broken since commit 175ddf7bd78f ("ice:
Reimplement module reads used by ethtool") In this refactor,
ice_get_module_eeprom() reads the eeprom in blocks of size 8.
But the condition that should protect the buffer overflow
ignores the last block. The last block always contains zeros.
Bug uncovered by ethtool upstream commit 9538f384b535
("netlink: eeprom: Defer page requests to individual parsers")
After this commit, ethtool reads a block with length = 1;
to read the SFF-8024 identifier value.
Fixes: 175ddf7bd78f ("ice: Reimplement module reads used by ethtool") Signed-off-by: Petr Oros <poros@redhat.com> Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Tested-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
syzbot reported use-after-free in cfusbl_device_notify() [1]. This
causes a stack trace like below:
BUG: KASAN: use-after-free in cfusbl_device_notify+0x7c9/0x870 net/caif/caif_usb.c:138
Read of size 8 at addr ffff88807ac4e6f0 by task kworker/u4:6/1214
When unregistering a net device, unregister_netdevice_many_notify()
sets the device's reg_state to NETREG_UNREGISTERING, calls notifiers
with NETDEV_UNREGISTER, and adds the device to the todo list.
Later on, devices in the todo list are processed by netdev_run_todo().
netdev_run_todo() waits devices' reference count become 1 while
rebdoadcasting NETDEV_UNREGISTER notification.
When cfusbl_device_notify() is called with NETDEV_UNREGISTER multiple
times, the parent device might be freed. This could cause UAF.
Processing NETDEV_UNREGISTER multiple times also causes inbalance of
reference count for the module.
This patch fixes the issue by accepting only first NETDEV_UNREGISTER
notification.
Move the LAN7800 internal phy (phy ID 0x0007c132) specific register
accesses to the phy driver (microchip.c).
Fix the error reported by Enguerrand de Ribaucourt in December 2022,
"Some operations during the cable switch workaround modify the register
LAN88XX_INT_MASK of the PHY. However, this register is specific to the
LAN8835 PHY. For instance, if a DP8322I PHY is connected to the LAN7801,
that register (0x19), corresponds to the LED and MAC address
configuration, resulting in unapropriate behavior."
I did not test with the DP8322I PHY, but I tested with an EVB-LAN7800
with the internal PHY.
Fixes: 09e8bf6570a4 ("lan78xx: workaround of forced 100 Full/Half duplex mode error") Signed-off-by: Yuiko Oshino <yuiko.oshino@microchip.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Link: https://lore.kernel.org/r/20230301154307.30438-1-yuiko.oshino@microchip.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
When creating counters with initial delay configured, the enable_on_exec
field is not set. So we need to enable the counters later. The problem
is, when a workload is specified the target__none() is true. So we also
need to check stat_config.initial_delay.
In this change, we add a new field 'initial_delay' for struct target
which could be shared by other subcommands. And define
target__enable_on_exec() which returns whether enable_on_exec should be
set on normal cases.
Fixes: 98b344effc11cb1b ("perf stat: Do not delay the workload with --delay") Signed-off-by: Changbin Du <changbin.du@huawei.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Hui Wang <hw.huiwang@huawei.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: https://lore.kernel.org/r/20230302031146.2801588-2-changbin.du@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Florian reported a regression and sent a patch with the following
changelog:
<quote>
There is a noticeable tcp performance regression (loopback or cross-netns),
seen with iperf3 -Z (sendfile mode) when generic retpolines are needed.
With SK_RECLAIM_THRESHOLD checks gone number of calls to enter/leave
memory pressure happen much more often. For TCP indirect calls are
used.
We can't remove the if-set-return short-circuit check in
tcp_enter_memory_pressure because there are callers other than
sk_enter_memory_pressure. Doing a check in the sk wrapper too
reduces the indirect calls enough to recover some performance.
"iperf3 -c $peer -t 60 -Z -f g", connected via veth in another netns.
</quote>
It seems we forgot to upstream this indirect call mitigation we
had for years, lets do this instead.
[edumazet] - It seems we forgot to upstream this indirect call
mitigation we had for years, let's do this instead.
- Changed to INDIRECT_CALL_INET_1() to avoid bots reports.
ctx->crypto_send.info is not protected by lock_sock in
do_tls_getsockopt_conf(). A race condition between do_tls_getsockopt_conf()
and error paths of do_tls_setsockopt_conf() may lead to a use-after-free
or null-deref.
More discussion: https://lore.kernel.org/all/Y/ht6gQL+u6fj3dG@hog/
The test_local_dnat_portonly() function initiates the client-side as
soon as it sets the listening side to the background. This could lead to
a race condition where the server may not be ready to listen. To ensure
that the server-side is up and running before initiating the
client-side, a delay is introduced to the test_local_dnat_portonly()
function.
Before the fix:
# ./nft_nat.sh
PASS: netns routing/connectivity: ns0-rthlYrBU can reach ns1-rthlYrBU and ns2-rthlYrBU
PASS: ping to ns1-rthlYrBU was ip NATted to ns2-rthlYrBU
PASS: ping to ns1-rthlYrBU OK after ip nat output chain flush
PASS: ipv6 ping to ns1-rthlYrBU was ip6 NATted to ns2-rthlYrBU
2023/02/27 04:11:03 socat[6055] E connect(5, AF=2 10.0.1.99:2000, 16): Connection refused
ERROR: inet port rewrite
After the fix:
# ./nft_nat.sh
PASS: netns routing/connectivity: ns0-9sPJV6JJ can reach ns1-9sPJV6JJ and ns2-9sPJV6JJ
PASS: ping to ns1-9sPJV6JJ was ip NATted to ns2-9sPJV6JJ
PASS: ping to ns1-9sPJV6JJ OK after ip nat output chain flush
PASS: ipv6 ping to ns1-9sPJV6JJ was ip6 NATted to ns2-9sPJV6JJ
PASS: inet port rewrite without l3 address
Fixes: a00bd5b5d626 ("netfilter: nat: really support inet nat without l3 address") Signed-off-by: Hangbin Liu <liuhangbin@gmail.com> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
When the police was removed from the port, then it was trying to
remove the police from the police id and not from the actual
police index.
The police id represents the id of the police and police index
represents the position in HW where the police is situated.
The port police id can be any number while the port police index
is a number based on the port chip port.
Fix this by deleting the police from HW that is situated at the
police index and not police id.
Fixes: 15b8f3a8fe54 ("net: lan966x: Add port police support using tc-matchall") Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com> Reviewed-by: Simon Horman <simon.horman@corigine.com> Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Fixes: 688bdaed3ad7 ("ila: Add generic ILA translation facility") Reported-by: syzbot <syzkaller@googlegroups.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
Gaurav reports that TLS Rx is broken with async crypto
accelerators. The commit under fixes missed updating
the retval byte counting logic when updating how records
are stored. Even tho both before and after the change
'decrypted' was updated inside the main loop, it was
completely overwritten when processing the async
completions. Now that the rx_list only holds
non-zero-copy records we need to add, not overwrite.
It looks like U-Boot fails to start the kernel properly when the
compatible string of the board isn't fsl,T1040RDB, so stop overriding it
from the rev-a.dts.
Fixes: bd04d92fe8a1 ("powerpc: dts: t1040rdb: fix ports names for Seville Ethernet switch") Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
While bringing hardware up we should perform a full reset including the
switch bit (BGMAC_BCMA_IOCTL_SW_RESET aka SICF_SWRST). It's what
specification says and what reference driver does.
This seems to be critical for the BCM5358. Without this hardware doesn't
get initialized properly and doesn't seem to transmit or receive any
packets.
Originally bgmac was calling bgmac_chip_reset() before setting
"has_robosw" property which resulted in expected behaviour. That has
changed as a side effect of adding platform device support which
regressed BCM5358 support.
The rptr_addr is set in the preempt_init_ring(), which is called from
a5xx_gpu_init(). It uses shadowptr() to set the address, however the
shadow_iova is not yet initialized at that time. Move the rptr_addr
setting to the a5xx_preempt_hw_init() which is called after setting the
shadow_iova, getting the correct value for the address.
Fixes: da8cc66288a3 ("drm/msm: Allow a5xx to mark the RPTR shadow as privileged") Suggested-by: Rob Clark <robdclark@gmail.com> Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/522640/ Link: https://lore.kernel.org/r/20230214020956.164473-5-dmitry.baryshkov@linaro.org Signed-off-by: Rob Clark <robdclark@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Quoting Yassine: ring->memptrs->rptr is never updated and stays 0, so
the comparison always evaluates to false and get_next_ring always
returns ring 0 thinking it isn't empty.
Fix this by calling get_rptr() instead of reading rptr directly.
Reported-by: Yassine Oudjana <y.oudjana@protonmail.com> Fixes: 2ae99dd8e304 ("drm/msm: Implement preemption for A5XX targets") Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/522642/ Link: https://lore.kernel.org/r/20230214020956.164473-4-dmitry.baryshkov@linaro.org Signed-off-by: Rob Clark <robdclark@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Rather than writing CP_PREEMPT_ENABLE_GLOBAL twice, follow the vendor
kernel and set CP_PREEMPT_ENABLE_LOCAL register instead. a5xx_submit()
will override it during submission, but let's get the sequence correct.
Fixes: 2ae99dd8e304 ("drm/msm: Implement preemption for A5XX targets") Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/522638/ Link: https://lore.kernel.org/r/20230214020956.164473-2-dmitry.baryshkov@linaro.org Signed-off-by: Rob Clark <robdclark@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Do not map STATUS_OBJECT_NAME_INVALID to -EREMOTE under non-DFS
shares, or 'nodfs' mounts or CONFIG_CIFS_DFS_UPCALL=n builds.
Otherwise, in the slow path, get a referral to figure out whether it
is an actual DFS link.
This could be simply reproduced under a non-DFS share by running the
following
When we are renaming a directory to a different directory, we need to
update '..' entry in the moved directory. However nothing prevents moved
directory from being modified and even converted from the inline format
to the normal format. When such race happens the rename code gets
confused and we crash. Fix the problem by locking the moved directory.
CC: stable@vger.kernel.org Fixes: 4f33f5947495 ("ext4: let ext4_rename handle inline dir") Signed-off-by: Jan Kara <jack@suse.cz> Link: https://lore.kernel.org/r/20230126112221.11866-1-jack@suse.cz Signed-off-by: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Sasha Levin <sashal@kernel.org>
Remove the /proc/scsi/${proc_name} directory earlier to fix a race
condition between unloading and reloading kernel modules. This fixes a bug
introduced in 2009 by commit 1f857a84ea63 ("[SCSI] fix /proc memory leak in
the SCSI core").
As explained in commit 97484fe6a145 ("block: Do not reread partition table
on exclusively open device"), reread partition on the device that is
exclusively opened by someone else is problematic.
This patch will make sure partition scan will only be proceed if current
thread open the device exclusively, or the device is not opened
exclusively, and in the later case, other scanners and exclusive openers
will be blocked temporarily until partition scan is done.
Fixes: 45703ec20f31 ("block: remove the bd_openers checks in blk_drop_partitions") Cc: <stable@vger.kernel.org> Suggested-by: Jan Kara <jack@suse.cz> Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Link: https://lore.kernel.org/r/20230217022200.3092987-3-yukuai1@huaweicloud.com Signed-off-by: Jens Axboe <axboe@kernel.dk> Signed-off-by: Sasha Levin <sashal@kernel.org>
This patch can't fix the problem in a corner case that device can be
opened exclusively after the checking and before blkdev_get_by_dev().
We'll use a new solution to fix the problem in the next patch, and
the new solution doesn't need to change apis.
We now have some eDP+DSI dual panel systems floating around
where the DSI panel is the secondary LFP and thus needs to
consult "panel type 2" in VBT in order to locate all the
other panel type dependant stuff correctly.
To that end we need to pass in the devdata to
intel_bios_init_panel_late(), otherwise it'll just assume
we want the primary panel type. So let's try to just populate
the vbt.ports[] stuff and encoder->devdata for icl+ DSI
panels as well.
We can't do this on older platforms as there we risk a DSI
port aliasing with a HDMI/DP port, which is a totally legal
thing as the DSI ports live in their own little parallel
universe.
Lots of ADL machines out there with bogus VBTs that declare
two eDP child devices. In order for those to work we need to
figure out which power sequencer to use before we try the EDID
read. So let's do the panel VBT init early if we can, falling
back to the post-EDID init otherwise.
The post-EDID init panel_type=0xff approach of assuming the
power sequencer should already be enabled doesn't really work
with multiple eDP panels, and currently we just end up using
the same power sequencer for both eDP ports, which at least
confuses the wakeref tracking, and potentially also causes us
to toggle the VDD for the panel when we should not.
Introduce a place where we can initialize connector->panel
after it's been allocated. We already have a intel_panel_init()
so had to get creative with the name and came up with
intel_panel_init_alloc().
The flash decriptor contains the number of flash components that we use
to figure out how many flash chips there are connected. Therefore we
need to read it first before deciding how many chip selects the
controller has.
Reported-by: Marcin Witkowski <marcin.witkowski@intel.com> Fixes: 8cacd7e61303 ("spi: intel: Add support for second flash chip") Cc: stable@vger.kernel.org Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Link: https://lore.kernel.org/r/20230215110040.42186-1-mika.westerberg@linux.intel.com Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
state_lock, the spinlock type is meant to protect race against concurrent
MHI state transitions. In mhi_ep_set_m0_state(), while the state_lock is
being held, the channels are resumed in mhi_ep_resume_channels() if the
previous state was M3. This causes sleeping in atomic bug, since
mhi_ep_resume_channels() use mutex internally.
Since the state_lock is supposed to be held throughout the state change,
it is not ideal to drop the lock before calling mhi_ep_resume_channels().
So to fix this issue, let's change the type of state_lock to mutex. This
would also allow holding the lock throughout all state transitions thereby
avoiding any potential race.
Cc: <stable@vger.kernel.org> # 5.19 Fixes: 9ac68db5eafd ("bus: mhi: ep: Add support for suspending and resuming channels") Reported-by: Dan Carpenter <error27@gmail.com> Reviewed-by: Jeffrey Hugo <quic_jhugo@quicinc.com> Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
During graceful shutdown scenario, host will issue MHI RESET to the
endpoint device before initiating shutdown. In that case, it makes sense
to completely power down the MHI stack as sooner or later the access to
MMIO registers will be prohibited. Also, the stack needs to be powered
up in the case of SYS_ERR to recover the device.
The condition determining whether the preallocation can be used had
an off-by-one error so we didn't discard preallocation when new
allocation was just following it. This can then confuse code in
inode_getblk().
CC: stable@vger.kernel.org Fixes: 189b02ac597c ("udf: Discard preallocation before extending file with a hole") Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Sasha Levin <sashal@kernel.org>
This patch sets the stop tx flag before we commit the dlm message.
This flag will report about unexpected transmissions after we
send the DLM_FIN message out, which should be the last message sent.
When we commit the dlm fin message, it could be that we already
got an ack back and the CLOSED state change already happened.
We should not set this flag when we are in CLOSED state. To avoid this
race we simply set the tx flag before the state change can be in
progress by moving it before dlm_midcomms_commit_mhandle().
Cc: stable@vger.kernel.org Fixes: 9c99c76781ff ("fs: dlm: add reliable connection if reconnect") Signed-off-by: Alexander Aring <aahringo@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
If we release a midcomms node structure, there should be nothing left
inside the dlm midcomms send queue. However, sometimes this is not true
because I believe some DLM_FIN message was not acked... if we run
into a shutdown timeout, then we should be sure there is no pending send
dlm message inside this queue when releasing midcomms node structure.
Cc: stable@vger.kernel.org Fixes: 9c99c76781ff ("fs: dlm: add reliable connection if reconnect") Signed-off-by: Alexander Aring <aahringo@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
To not get the console spammed about WARN_ON() of invalid states in the
dlm midcomms hot path handling we switch to WARN_ON_ONCE() to get it
only once that there might be an issue with the midcomms state handling.
Signed-off-by: Alexander Aring <aahringo@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com>
Stable-dep-of: 557856c7140b ("fs: dlm: be sure to call dlm_send_queue_flush()") Signed-off-by: Sasha Levin <sashal@kernel.org>
[ 151.834900] The buggy address belongs to the object at ffff88811a980c30
which belongs to the cache dlm_mhandle of size 88
[ 151.836894] The buggy address is located 48 bytes inside of
88-byte region [ffff88811a980c30, ffff88811a980c88)
[ 151.845822] Memory state around the buggy address:
[ 151.846602] ffff88811a980b00: fb fb fb fb fc fc fc fc fa fb fb fb fb fb fb fb
[ 151.847761] ffff88811a980b80: fb fb fb fc fc fc fc fa fb fb fb fb fb fb fb fb
[ 151.848921] >ffff88811a980c00: fb fb fc fc fc fc fa fb fb fb fb fb fb fb fb fb
[ 151.850076] ^
[ 151.851085] ffff88811a980c80: fb fc fc fc fc fa fb fb fb fb fb fb fb fb fb fb
[ 151.852269] ffff88811a980d00: fc fc fc fc fa fb fb fb fb fb fb fb fb fb fb fc
[ 151.853428] ==================================================================
[ 151.855618] Disabling lock debugging due to kernel taint
It is accessing a mhandle in dlm_midcomms_commit_mhandle() and the mhandle
was freed by a call_rcu() call in dlm_process_incoming_buffer(),
dlm_mhandle_delete(). It looks like it was freed because an ack of
this message was received. There is a short race between committing the
dlm message to be transmitted and getting an ack back. If the ack is
faster than returning from dlm_midcomms_commit_msg_3_2(), then we run
into a use-after free because we still need to reference the mhandle when
calling srcu_read_unlock().
To avoid that, we don't allow that mhandle to be freed between
dlm_midcomms_commit_msg_3_2() and srcu_read_unlock() by using rcu read
lock. We can do that because mhandle is protected by rcu handling.
Cc: stable@vger.kernel.org Fixes: 9c99c76781ff ("fs: dlm: add reliable connection if reconnect") Signed-off-by: Alexander Aring <aahringo@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
This patch adds tracepoints for send and recv cases of dlm messages and
dlm rcom messages. In case of send and dlm message we add the dlm rsb
resource name this dlm messages belongs to. This has the advantage to
follow dlm messages on a per lock basis. In case of recv message the
resource name can be extracted by follow the send message sequence
number.
The dlm message DLM_MSG_PURGE doesn't belong to a lock request and will
not set the resource name in a dlm_message trace. The same for all rcom
messages.
There is additional handling required for this debugging functionality
which is tried to be small as possible. Also the midcomms layer gets
aware of lock resource names, for now this is required to make a
connection between sequence number and lock resource names. It is for
debugging purpose only.
Signed-off-by: Alexander Aring <aahringo@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com>
Stable-dep-of: caf2799e0b84 ("fs: dlm: fix use after free in midcomms commit") Signed-off-by: Sasha Levin <sashal@kernel.org>
To allow more than just dereferencing the inner header we directly point
to the inner dlm packet which allows us to dereference the header, rcom
or message structure.
Signed-off-by: Alexander Aring <aahringo@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com>
Stable-dep-of: caf2799e0b84 ("fs: dlm: fix use after free in midcomms commit") Signed-off-by: Sasha Levin <sashal@kernel.org>
This patch removes the send repeat remove handling. This handling is
there to repeatingly DLM_MSG_REMOVE messages in cases the dlm stack
thinks it was not received at the first time. In cases of message drops
this functionality is necessary, but since the DLM midcomms layer
guarantees there are no messages drops between cluster nodes this
feature became not strict necessary anymore. Due message
delays/processing it could be that two send_repeat_remove() are sent out
while the other should be still on it's way. We remove the repeat remove
handling because we are sure that the message cannot be dropped due
communication errors.
Signed-off-by: Alexander Aring <aahringo@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com>
Stable-dep-of: caf2799e0b84 ("fs: dlm: fix use after free in midcomms commit") Signed-off-by: Sasha Levin <sashal@kernel.org>
The scand kthread can send dlm messages out, especially dlm remove
messages to free memory for unused rsb on other nodes. To send out dlm
messages, midcomms must be initialized. This patch moves the midcomms
start before scand is started.
Cc: stable@vger.kernel.org Fixes: ae5bef69e6aa ("[DLM] The core of the DLM for GFS2/CLVM") Signed-off-by: Alexander Aring <aahringo@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
This patch introduces leftovers of init, start, stop and exit
functionality. The dlm application layer should always call the midcomms
layer which getting aware of such event and redirect it to the lowcomms
layer. Some functionality which is currently handled inside the start
functionality of midcomms and lowcomms should be handled in the init
functionality as it only need to be initialized once when dlm is loaded.
Signed-off-by: Alexander Aring <aahringo@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com>
Stable-dep-of: 5282c6801f3b ("fs: dlm: start midcomms before scand") Signed-off-by: Sasha Levin <sashal@kernel.org>
This patch will fix a small issue when printing out that
dlm_midcomms_start() failed to start and it was printing out that the
dlm subcomponent lowcomms was failed but lowcomms is behind the midcomms
layer.
Signed-off-by: Alexander Aring <aahringo@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com>
Stable-dep-of: 5282c6801f3b ("fs: dlm: start midcomms before scand") Signed-off-by: Sasha Levin <sashal@kernel.org>
Call kvm_init() only after _all_ setup is complete, as kvm_init() exposes
/dev/kvm to userspace and thus allows userspace to create VMs (and call
other ioctls). E.g. KVM will encounter a NULL pointer when attempting to
add a vCPU to the per-CPU loaded_vmcss_on_cpu list if userspace is able to
create a VM before vmx_init() configures said list.
Move the guts of kvm_arch_init() to a new helper, kvm_x86_vendor_init(),
so that VMX can do _all_ arch and vendor initialization before calling
kvm_init(). Calling kvm_init() must be the _very_ last step during init,
as kvm_init() exposes /dev/kvm to userspace, i.e. allows creating VMs.
No functional change intended.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20221130230934.1014142-14-seanjc@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Stable-dep-of: 3c3b35a745e3 ("KVM: VMX: Do _all_ initialization before exposing /dev/kvm to userspace") Signed-off-by: Sasha Levin <sashal@kernel.org>
Don't disable the eVMCS static key on module exit, kvm_intel.ko owns the
key so there can't possibly be users after the kvm_intel.ko is unloaded,
at least not without much bigger issues.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20221130230934.1014142-12-seanjc@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Stable-dep-of: 3c3b35a745e3 ("KVM: VMX: Do _all_ initialization before exposing /dev/kvm to userspace") Signed-off-by: Sasha Levin <sashal@kernel.org>
Reset the eVMCS controls in the per-CPU VP assist page during hardware
disabling instead of waiting until kvm-intel's module exit. The controls
are activated if and only if KVM creates a VM, i.e. don't need to be
reset if hardware is never enabled.
Doing the reset during hardware disabling will naturally fix a potential
NULL pointer deref bug once KVM disables CPU hotplug while enabling and
disabling hardware (which is necessary to fix a variety of bugs). If the
kernel is running as the root partition, the VP assist page is unmapped
during CPU hot unplug, and so KVM's clearing of the eVMCS controls needs
to occur with CPU hot(un)plug disabled, otherwise KVM could attempt to
write to a CPU's VP assist page after it's unmapped.
Reported-by: Vitaly Kuznetsov <vkuznets@redhat.com> Signed-off-by: Sean Christopherson <seanjc@google.com> Reviewed-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20221130230934.1014142-11-seanjc@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Stable-dep-of: 3c3b35a745e3 ("KVM: VMX: Do _all_ initialization before exposing /dev/kvm to userspace") Signed-off-by: Sasha Levin <sashal@kernel.org>
cb_context should be freed on the error path in nfc_se_io as stated by
commit 51e42304a409 ("nfc: fix memory leak of se_io context in
nfc_genl_se_io").
Make the error path in nfc_se_io unwind everything in reverse order, i.e.
free the cb_context after unlocking the device.
Suggested-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Signed-off-by: Fedor Pchelkin <pchelkin@ispras.ru> Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Link: https://lore.kernel.org/r/20230306212650.230322-1-pchelkin@ispras.ru Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The default maximum data buffer size for this interface is UHID_DATA_MAX
(4k). When data buffers are being processed, ensure this value is used
when ensuring the sanity, rather than a value between the user provided
value and HID_MAX_BUFFER_SIZE (16k).
Signed-off-by: Lee Jones <lee@kernel.org> Signed-off-by: Jiri Kosina <jkosina@suse.cz> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Presently, when a report is processed, its proposed size, provided by
the user of the API (as Report Size * Report Count) is compared against
the subsystem default HID_MAX_BUFFER_SIZE (16k). However, some
low-level HID drivers allocate a reduced amount of memory to their
buffers (e.g. UHID only allocates UHID_DATA_MAX (4k) buffers), rending
this check inadequate in some cases.
In these circumstances, if the received report ends up being smaller
than the proposed report size, the remainder of the buffer is zeroed.
That is, the space between sizeof(csize) (size of the current report)
and the rsize (size proposed i.e. Report Size * Report Count), which can
be handled up to HID_MAX_BUFFER_SIZE (16k). Meaning that memset()
shoots straight past the end of the buffer boundary and starts zeroing
out in-use values, often resulting in calamity.
This patch introduces a new variable into 'struct hid_ll_driver' where
individual low-level drivers can over-ride the default maximum value of
HID_MAX_BUFFER_SIZE (16k) with something more sympathetic to the
interface.
Signed-off-by: Lee Jones <lee@kernel.org> Signed-off-by: Jiri Kosina <jkosina@suse.cz> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If the boot loader inode has never been used before, the
EXT4_IOC_SWAP_BOOT inode will initialize it, including setting the
i_size to 0. However, if the "never before used" boot loader has a
non-zero i_size, then i_disksize will be non-zero, and the
inconsistency between i_size and i_disksize can trigger a kernel
warning:
Above issue happens as follows:
ext4_iget
ext4_find_inline_data_nolock ->i_inline_off=164 i_inline_size=60
ext4_try_add_inline_entry
__ext4_mark_inode_dirty
ext4_expand_extra_isize_ea ->i_extra_isize=32 s_want_extra_isize=44
ext4_xattr_shift_entries
->after shift i_inline_off is incorrect, actually is change to 176
ext4_try_add_inline_entry
ext4_update_inline_dir
get_max_inline_xattr_value_size
if (EXT4_I(inode)->i_inline_off)
entry = (struct ext4_xattr_entry *)((void *)raw_inode +
EXT4_I(inode)->i_inline_off);
free += EXT4_XATTR_SIZE(le32_to_cpu(entry->e_value_size));
->As entry is incorrect, then 'free' may be negative
ext4_update_inline_data
value = kzalloc(len, GFP_NOFS);
-> len is unsigned int, maybe very large, then trigger warning when
'kzalloc()'
To resolve the above issue we need to update 'i_inline_off' after
'ext4_xattr_shift_entries()'. We do not need to set
EXT4_STATE_MAY_INLINE_DATA flag here, since ext4_mark_inode_dirty()
already sets this flag if needed. Setting EXT4_STATE_MAY_INLINE_DATA
when it is needed may trigger a BUG_ON in ext4_writepages().
Reported-by: syzbot+d30838395804afc2fa6f@syzkaller.appspotmail.com Cc: stable@kernel.org Signed-off-by: Ye Bin <yebin10@huawei.com> Reviewed-by: Jan Kara <jack@suse.cz> Link: https://lore.kernel.org/r/20230307015253.2232062-3-yebin@huaweicloud.com Signed-off-by: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The only caller of ext4_find_inline_data_nolock() that needs setting of
EXT4_STATE_MAY_INLINE_DATA flag is ext4_iget_extra_inode(). In
ext4_write_inline_data_end() we just need to update inode->i_inline_off.
Since we are going to add one more caller that does not need to set
EXT4_STATE_MAY_INLINE_DATA, just move setting of EXT4_STATE_MAY_INLINE_DATA
out to ext4_iget_extra_inode().
Signed-off-by: Ye Bin <yebin10@huawei.com> Cc: stable@kernel.org Reviewed-by: Jan Kara <jack@suse.cz> Link: https://lore.kernel.org/r/20230307015253.2232062-2-yebin@huaweicloud.com Signed-off-by: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
projects / kernel.git / log