KVM/arm64 fixes for 6.3, part #3
- Ensure the guest PMU context is restored before the first KVM_RUN,
fixing an issue where EL0 event counting is broken after vCPU
save/restore
- Actually initialize ID_AA64PFR0_EL1.{CSV2,CSV3} based on the
sanitized, system-wide values for protected VMs
Pull drm fixes from Daniel Vetter:
"Mostly i915 fixes: dp mst for compression/dsc, perf ioctl uaf, ctx rpm
accounting, gt reset vs huc loading.
And a few individual driver fixes: ivpu dma fence&suspend, panfrost
mmap, nouveau color depth"
* tag 'drm-fixes-2023-04-06' of git://anongit.freedesktop.org/drm/drm:
accel/ivpu: Fix S3 system suspend when not idle
accel/ivpu: Add dma fence to command buffers only
drm/i915: Fix context runtime accounting
drm/i915: fix race condition UAF in i915_perf_add_config_ioctl
drm/i915: Use compressed bpp when calculating m/n value for DP MST DSC
drm/i915/huc: Cancel HuC delayed load timer on reset.
drm/i915/ttm: fix sparse warning
drm/panfrost: Fix the panfrost_mmu_map_fault_addr() error path
drm/nouveau/disp: Support more modes by checking with lower bpc
Pull sound fixes from Takashi Iwai:
"The majority of changes here are various fixes for Intel drivers,
and there is a change in ASoC PCM core for the format constraints.
In addition, a workaround for HD-audio HDMI regressions and usual
HD-audio quirks are found"
* tag 'sound-6.3-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
ALSA: hda/hdmi: Preserve the previous PCM device upon re-enablement
ALSA: hda/realtek: Add quirk for Clevo X370SNW
ALSA: hda/realtek: fix mute/micmute LEDs for a HP ProBook
ASoC: SOF: avoid a NULL dereference with unsupported widgets
ASoC: da7213.c: add missing pm_runtime_disable()
ASoC: hdac_hdmi: use set_stream() instead of set_tdm_slots()
ASoC: codecs: lpass: fix the order or clks turn off during suspend
ASoC: Intel: bytcr_rt5640: Add quirk for the Acer Iconia One 7 B1-750
ASoC: SOF: ipc4: Ensure DSP is in D0I0 during sof_ipc4_set_get_data()
ASoC: amd: yc: Add DMI entries to support Victus by HP Laptop 16-e1xxx (8A22)
ASoC: soc-pcm: fix hw->formats cleared by soc_pcm_hw_init() for dpcm
ASoC: Intel: soc-acpi: add table for Intel 'Rooks County' NUC M15
ASOC: Intel: sof_sdw: add quirk for Intel 'Rooks County' NUC M15
Pull x86 platform driver fixes from Hans de Goede:
- more think-lmi fixes
- one DMI quirk addition
* tag 'platform-drivers-x86-v6.3-5' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86:
platform/x86: thinkpad_acpi: Add missing T14s Gen1 type to s2idle quirk list
platform/x86: think-lmi: Clean up display of current_value on Thinkstation
platform/x86: think-lmi: Fix memory leaks when parsing ThinkStation WMI strings
platform/x86: think-lmi: Fix memory leak when showing current settings
Memory passed to kvfree_rcu() that is to be freed is tracked by a
per-CPU kfree_rcu_cpu structure, which in turn contains pointers
to kvfree_rcu_bulk_data structures that contain pointers to memory
that has not yet been handed to RCU, along with an kfree_rcu_cpu_work
structure that tracks the memory that has already been handed to RCU.
These structures track three categories of memory: (1) Memory for
kfree(), (2) Memory for kvfree(), and (3) Memory for both that arrived
during an OOM episode. The first two categories are tracked in a
cache-friendly manner involving a dynamically allocated page of pointers
(the aforementioned kvfree_rcu_bulk_data structures), while the third
uses a simple (but decidedly cache-unfriendly) linked list through the
rcu_head structures in each block of memory.
On a given CPU, these three categories are handled as a unit, with that
CPU's kfree_rcu_cpu_work structure having one pointer for each of the
three categories. Clearly, new memory for a given category cannot be
placed in the corresponding kfree_rcu_cpu_work structure until any old
memory has had its grace period elapse and thus has been removed. And
the kfree_rcu_monitor() function does in fact check for this.
Except that the kfree_rcu_monitor() function checks these pointers one
at a time. This means that if the previous kfree_rcu() memory passed
to RCU had only category 1 and the current one has only category 2, the
kfree_rcu_monitor() function will send that current category-2 memory
along immediately. This can result in memory being freed too soon,
that is, out from under unsuspecting RCU readers.
To see this, consider the following sequence of events, in which:
o Task A on CPU 0 calls rcu_read_lock(), then uses "from_cset",
then is preempted.
o CPU 1 calls kfree_rcu(cset, rcu_head) in order to free "from_cset"
after a later grace period. Except that "from_cset" is freed
right after the previous grace period ended, so that "from_cset"
is immediately freed. Task A resumes and references "from_cset"'s
member, after which nothing good happens.
In full detail:
CPU 0 CPU 1
---------------------- ----------------------
count_memcg_event_mm()
|rcu_read_lock() <---
|mem_cgroup_from_task()
|// css_set_ptr is the "from_cset" mentioned on CPU 1
|css_set_ptr = rcu_dereference((task)->cgroups)
|// Hard irq comes, current task is scheduled out.
cgroup_attach_task()
|cgroup_migrate()
|cgroup_migrate_execute()
|css_set_move_task(task, from_cset, to_cset, true)
|cgroup_move_task(task, to_cset)
|rcu_assign_pointer(.., to_cset)
|...
|cgroup_migrate_finish()
|put_css_set_locked(from_cset)
|from_cset->refcount return 0
|kfree_rcu(cset, rcu_head) // free from_cset after new gp
|add_ptr_to_bulk_krc_lock()
|schedule_delayed_work(&krcp->monitor_work, ..)
kfree_rcu_monitor()
|krcp->bulk_head[0]'s work attached to krwp->bulk_head_free[]
|queue_rcu_work(system_wq, &krwp->rcu_work)
|if rwork->rcu.work is not in WORK_STRUCT_PENDING_BIT state,
|call_rcu(&rwork->rcu, rcu_work_rcufn) <--- request new gp
// There is a perious call_rcu(.., rcu_work_rcufn)
// gp end, rcu_work_rcufn() is called.
rcu_work_rcufn()
|__queue_work(.., rwork->wq, &rwork->work);
|kfree_rcu_work()
|krwp->bulk_head_free[0] bulk is freed before new gp end!!!
|The "from_cset" is freed before new gp end.
// the task resumes some time later.
|css_set_ptr->subsys[(subsys_id) <--- Caused kernel crash, because css_set_ptr is freed.
This commit therefore causes kfree_rcu_monitor() to refrain from moving
kfree_rcu() memory to the kfree_rcu_cpu_work structure until the RCU
grace period has completed for all three categories.
v2: Use helper function instead of inserted code block at kfree_rcu_monitor().
Fixes: 34c8817455 ("rcu: Support kfree_bulk() interface in kfree_rcu()")
Fixes: 5f3c8d6204 ("rcu/tree: Maintain separate array for vmalloc ptrs")
Reported-by: Mukesh Ojha <quic_mojha@quicinc.com>
Signed-off-by: Ziwei Dai <ziwei.dai@unisoc.com>
Reviewed-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
Tested-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Pull asm-generic fixes from Arnd Bergmann:
"These are minor fixes to address false-positive build warnings:
Some of the less common I/O accessors are missing __force casts and
cause sparse warnings for their implied byteswap, and a recent change
to __generic_cmpxchg_local() causes a warning about constant integer
truncation"
* tag 'asm-generic-fixes-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/arnd/asm-generic:
asm-generic: avoid __generic_cmpxchg_local warnings
asm-generic/io.h: suppress endianness warnings for relaxed accessors
asm-generic/io.h: suppress endianness warnings for readq() and writeq()
The current code path can lead to warnings because of uninitialized device,
which contains, as a consequence, uninitialized kobject. The uninitialized
device is passed to of_platform_populate, which will at some point, while
creating child device, try to get a reference on uninitialized parent,
resulting in the following warning:
kobject: '(null)' ((ptrval)): is not initialized, yet kobject_get() is
being called.
The warning is observed after migrating a kernel 5.10.x to 6.1.x.
Reverting commit 0d70af3c25 ("fpga: bridge: Use standard dev_release for
class driver") seems to remove the warning.
This commit aggregates device_initialize() and device_add() into
device_register() but this new call is done AFTER of_platform_populate
Fixes: 0d70af3c25 ("fpga: bridge: Use standard dev_release for class driver")
Signed-off-by: Alexis Lothoré <alexis.lothore@bootlin.com>
Acked-by: Xu Yilun <yilun.xu@intel.com>
Link: https://lore.kernel.org/r/20230404133102.2837535-2-alexis.lothore@bootlin.com
Signed-off-by: Xu Yilun <yilun.xu@intel.com>
Smatch Warns:
sound/firewire/tascam/tascam-stream.c:493 snd_tscm_stream_start_duplex()
warn: missing unwind goto?
The direct return will cause the stream list of "&tscm->domain" unemptied
and the session in "tscm" unfinished if amdtp_domain_start() returns with
an error.
Fix this by changing the direct return to a goto which will empty the
stream list of "&tscm->domain" and finish the session in "tscm".
The snd_tscm_stream_start_duplex() function is called in the prepare
callback of PCM. According to "ALSA Kernel API Documentation", the prepare
callback of PCM will be called many times at each setup. So, if the
"&d->streams" list is not emptied, when the prepare callback is called
next time, snd_tscm_stream_start_duplex() will receive -EBUSY from
amdtp_domain_add_stream() that tries to add an existing stream to the
domain. The error handling code after the "error" label will be executed
in this case, and the "&d->streams" list will be emptied. So not emptying
the "&d->streams" list will not cause an issue. But it is more efficient
and readable to empty it on the first error by changing the direct return
to a goto statement.
The session in "tscm" has been begun before amdtp_domain_start(), so it
needs to be finished when amdtp_domain_start() fails.
Fixes: c281d46a51 ("ALSA: firewire-tascam: support AMDTP domain")
Signed-off-by: Xu Biang <xubiang@hust.edu.cn>
Reviewed-by: Dan Carpenter <error27@gmail.com>
Acked-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20230406132801.105108-1-xubiang@hust.edu.cn
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The BTRFS_FS_CSUM_IMPL_FAST flag is currently set whenever a non-generic
crc32c is detected, which is the incorrect check if the file system uses
a different checksumming algorithm. Refactor the code to only check
this if crc32c is actually used. Note that in an ideal world the
information if an algorithm is hardware accelerated or not should be
provided by the crypto API instead, but that's left for another day.
CC: stable@vger.kernel.org # 5.4.x: c8a5f8ca9a: btrfs: print checksum type and implementation at mount time
CC: stable@vger.kernel.org # 5.4.x
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Commit d7b9416fe5 ("btrfs: remove btrfs_end_io_wq") converted the read
and I/O handling from btrfs_workqueues to Linux workqueues, and as part
of that lost the code to apply the thread_pool= based max_active limit
on remount. Restore it.
Fixes: d7b9416fe5 ("btrfs: remove btrfs_end_io_wq")
CC: stable@vger.kernel.org # 6.0+
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Pull NVMe fix from Christoph:
"nvme fixes for Linux 6.3
- fix discard support without oncs (Keith Busch)"
* tag 'nvme-6.3-2023-04-06' of git://git.infradead.org/nvme:
nvme: fix discard support without oncs
block size is one very key setting for block layer, and bad block size
could panic kernel easily.
Make sure that block size is set correctly.
Meantime if ublk_validate_params() fails, clear ub->params so that disk
is prevented from being added.
Fixes: 71f28f3136 ("ublk_drv: add io_uring based userspace block driver")
Reported-and-tested-by: Breno Leitao <leitao@debian.org>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
For memory alloc that store user data from nla[NFTA_OBJ_USERDATA],
use GFP_KERNEL_ACCOUNT is more suitable.
Fixes: 33758c8914 ("memcg: enable accounting for nft objects")
Signed-off-by: Chen Aotian <chenaotian2@163.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Recent attempt to ensure PREROUTING hook is executed again when a
decrypted ipsec packet received on a bridge passes through the network
stack a second time broke the physdev match in INPUT hook.
We can't discard the nf_bridge info strct from sabotage_in hook, as
this is needed by the physdev match.
Keep the struct around and handle this with another conditional instead.
Fixes: 2b272bb558 ("netfilter: br_netfilter: disable sabotage_in hook after first suppression")
Reported-and-tested-by: Farid BENAMROUCHE <fariouche@yahoo.fr>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
It could have never worked, as snd_emu10k1_fx8010_playback_prepare() and
snd_emu10k1_fx8010_playback_hw_free() assume the emu10k1 offset for the
ETRAM, and the default DSP code includes no handler for it. It also
wouldn't make a lot of sense to make it work, as Audigy has an own, much
simpler, pass-through mechanism. So just skip creation of the device.
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20230405201220.2197938-1-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Due to two copy/pastos, closing the MIC or EFX capture device would
make a running ADC capture hang due to unsetting its interrupt handler.
In principle, this would have also allowed dereferencing dangling
pointers, but we're actually rather thorough at disabling and flushing
the ints.
While it may sound like one, this actually wasn't a hypothetical bug:
PortAudio will open a capture stream at startup (and close it right
away) even if not asked to. If the first device is busy, it will just
proceed with the next one ... thus killing a concurrent capture.
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20230405201220.2197923-1-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The BIOS botches this one completely - it says the 2nd S/PDIF output is
used, while in fact it's the 1st one. This is tested on DP45SG, but I'm
assuming it's valid for the other boards in the series as well.
Also add some comments regarding the pins.
FWIW, the codec is apparently still sold by Tempo Semiconductor, Inc.,
where one can download the documentation.
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20230405201220.2197826-2-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Since SQE memory is shared with userspace, we should only be reading it
once. We cannot read it multiple times, particularly when it's read once
for validation and then read again for the actual use.
ublk_ch_uring_cmd() is safe when called as a retry operation, as the
memory backing is stable at that point. But for normal issue, we want
to ensure that we only read ublksrv_io_cmd once. Wrap the function in
a helper that reads the value into an on-stack copy of the struct.
Cc: stable@vger.kernel.org # 6.0+
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When deleting the netns and recreating a new one while re-adding the
veth interface, there is a small window of time during which the old
veth interface has not yet been removed. This can cause the new addition
to fail. To resolve this issue, we can either wait for a short while to
ensure that the old veth interface is deleted, or we can specifically
remove the veth interface.
Before this patch:
# ./rps_default_mask.sh
empty rps_default_mask [ ok ]
changing rps_default_mask dont affect existing devices [ ok ]
changing rps_default_mask dont affect existing netns [ ok ]
changing rps_default_mask affect newly created devices [ ok ]
changing rps_default_mask don't affect newly child netns[II][ ok ]
rps_default_mask is 0 by default in child netns [ ok ]
RTNETLINK answers: File exists
changing rps_default_mask in child ns don't affect the main one[ ok ]
cat: /sys/class/net/vethC11an1/queues/rx-0/rps_cpus: No such file or directory
changing rps_default_mask in child ns affects new childns devices./rps_default_mask.sh: line 36: [: -eq: unary operator expected
[fail] expected 1 found
changing rps_default_mask in child ns don't affect existing devices[ ok ]
After this patch:
# ./rps_default_mask.sh
empty rps_default_mask [ ok ]
changing rps_default_mask dont affect existing devices [ ok ]
changing rps_default_mask dont affect existing netns [ ok ]
changing rps_default_mask affect newly created devices [ ok ]
changing rps_default_mask don't affect newly child netns[II][ ok ]
rps_default_mask is 0 by default in child netns [ ok ]
changing rps_default_mask in child ns don't affect the main one[ ok ]
changing rps_default_mask in child ns affects new childns devices[ ok ]
changing rps_default_mask in child ns don't affect existing devices[ ok ]
Fixes: 3a7d84eae0 ("self-tests: more rps self tests")
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Link: https://lore.kernel.org/r/20230404072411.879476-1-liuhangbin@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Not all fec MDIO bus drivers support C45 mode transactions. The older fec
hardware block in many ColdFire SoCs does not appear to support them, at
least according to most of the different ColdFire SoC reference manuals.
The bits used to generate C45 access on the iMX parts, in the OP field
of the MMFR register, are documented as generating non-compliant MII
frames (it is not documented as to exactly how they are non-compliant).
Commit 8d03ad1ab0 ("net: fec: Separate C22 and C45 transactions")
means the fec driver will always register c45 MDIO read and write
methods. During probe these will always be accessed now generating
non-compliant MII accesses on ColdFire based devices.
Add a quirk define, FEC_QUIRK_HAS_MDIO_C45, that can be used to
distinguish silicon that supports MDIO C45 framing or not. Add this to
all the existing iMX quirks, so they will be behave as they do now (*).
(*) it seems that some iMX parts may not support C45 transactions either.
The iMX25 and iMX50 Reference Manuals contain similar wording to
the ColdFire Reference Manuals on this.
Fixes: 8d03ad1ab0 ("net: fec: Separate C22 and C45 transactions")
Signed-off-by: Greg Ungerer <gerg@linux-m68k.org>
Reviewed-by: Wei Fang <wei.fang@nxp.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20230404052207.3064861-1-gerg@linux-m68k.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Since commit 75c2f26da0 ("PCI: imx6: Add i.MX PCIe EP mode support")
the i.MX6 PCI driver is no longer selected by default. The
existing PCI_IMX6 was made a hidden option, selected by new options
PCI_IMX6_HOST (for the existing support) and PCI_IMX6_EP (for the
endpoint mode), but there has been no corresponding update to
imx_v6_v7_defconfig so the PCI_IMX6 ends up getting disabled.
Switch imx_v6_v7_defconfig to PCI_IMX6_HOST to preserve the existing
functionality.
This is based on the same fix done in commit 0cd5780eb6 ("arm64:
defconfig: Fix unintentional disablement of PCI on i.MX").
Fixes: 75c2f26da0 ("PCI: imx6: Add i.MX PCIe EP mode support")
Reported-by: Mattias Barthel <mattiasbarthel@gmail.com>
Signed-off-by: Fabio Estevam <festevam@denx.de>
Signed-off-by: Shawn Guo <shawnguo@kernel.org>
There is a concurrency bug that may cause the wrong value to be loaded
when a CPU is modifying the maple tree.
CPU1:
mtree_insert_range()
mas_insert()
mas_store_root()
...
mas_root_expand()
...
rcu_assign_pointer(mas->tree->ma_root, mte_mk_root(mas->node));
ma_set_meta(node, maple_leaf_64, 0, slot); <---IP
CPU2:
mtree_load()
mtree_lookup_walk()
ma_data_end();
When CPU1 is about to execute the instruction pointed to by IP, the
ma_data_end() executed by CPU2 may return the wrong end position, which
will cause the value loaded by mtree_load() to be wrong.
An example of triggering the bug:
Add mdelay(100) between rcu_assign_pointer() and ma_set_meta() in
mas_root_expand().
static DEFINE_MTREE(tree);
int work(void *p) {
unsigned long val;
for (int i = 0 ; i< 30; ++i) {
val = (unsigned long)mtree_load(&tree, 8);
mdelay(5);
pr_info("%lu",val);
}
return 0;
}
mt_init_flags(&tree, MT_FLAGS_USE_RCU);
mtree_insert(&tree, 0, (void*)12345, GFP_KERNEL);
run_thread(work)
mtree_insert(&tree, 1, (void*)56789, GFP_KERNEL);
In RCU mode, mtree_load() should always return the value before or after
the data structure is modified, and in this example mtree_load(&tree, 8)
may return 56789 which is not expected, it should always return NULL. Fix
it by put ma_set_meta() before rcu_assign_pointer().
Link: https://lkml.kernel.org/r/20230314124203.91572-4-zhangpeng.00@bytedance.com
Fixes: 54a611b605 ("Maple Tree: add new data structure")
Signed-off-by: Peng Zhang <zhangpeng.00@bytedance.com>
Reviewed-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
The si->lock must be held when deleting the si from the available list.
Otherwise, another thread can re-add the si to the available list, which
can lead to memory corruption. The only place we have found where this
happens is in the swapoff path. This case can be described as below:
core 0 core 1
swapoff
del_from_avail_list(si) waiting
try lock si->lock acquire swap_avail_lock
and re-add si into
swap_avail_head
acquire si->lock but missing si already being added again, and continuing
to clear SWP_WRITEOK, etc.
It can be easily found that a massive warning messages can be triggered
inside get_swap_pages() by some special cases, for example, we call
madvise(MADV_PAGEOUT) on blocks of touched memory concurrently, meanwhile,
run much swapon-swapoff operations (e.g. stress-ng-swap).
However, in the worst case, panic can be caused by the above scene. In
swapoff(), the memory used by si could be kept in swap_info[] after
turning off a swap. This means memory corruption will not be caused
immediately until allocated and reset for a new swap in the swapon path.
A panic message caused: (with CONFIG_PLIST_DEBUG enabled)
------------[ cut here ]------------
top: 00000000e58a3003, n: 0000000013e75cda, p: 000000008cd4451a
prev: 0000000035b1e58a, n: 000000008cd4451a, p: 000000002150ee8d
next: 000000008cd4451a, n: 000000008cd4451a, p: 000000008cd4451a
WARNING: CPU: 21 PID: 1843 at lib/plist.c:60 plist_check_prev_next_node+0x50/0x70
Modules linked in: rfkill(E) crct10dif_ce(E)...
CPU: 21 PID: 1843 Comm: stress-ng Kdump: ... 5.10.134+
Hardware name: Alibaba Cloud ECS, BIOS 0.0.0 02/06/2015
pstate: 60400005 (nZCv daif +PAN -UAO -TCO BTYPE=--)
pc : plist_check_prev_next_node+0x50/0x70
lr : plist_check_prev_next_node+0x50/0x70
sp : ffff0018009d3c30
x29: ffff0018009d3c40 x28: ffff800011b32a98
x27: 0000000000000000 x26: ffff001803908000
x25: ffff8000128ea088 x24: ffff800011b32a48
x23: 0000000000000028 x22: ffff001800875c00
x21: ffff800010f9e520 x20: ffff001800875c00
x19: ffff001800fdc6e0 x18: 0000000000000030
x17: 0000000000000000 x16: 0000000000000000
x15: 0736076307640766 x14: 0730073007380731
x13: 0736076307640766 x12: 0730073007380731
x11: 000000000004058d x10: 0000000085a85b76
x9 : ffff8000101436e4 x8 : ffff800011c8ce08
x7 : 0000000000000000 x6 : 0000000000000001
x5 : ffff0017df9ed338 x4 : 0000000000000001
x3 : ffff8017ce62a000 x2 : ffff0017df9ed340
x1 : 0000000000000000 x0 : 0000000000000000
Call trace:
plist_check_prev_next_node+0x50/0x70
plist_check_head+0x80/0xf0
plist_add+0x28/0x140
add_to_avail_list+0x9c/0xf0
_enable_swap_info+0x78/0xb4
__do_sys_swapon+0x918/0xa10
__arm64_sys_swapon+0x20/0x30
el0_svc_common+0x8c/0x220
do_el0_svc+0x2c/0x90
el0_svc+0x1c/0x30
el0_sync_handler+0xa8/0xb0
el0_sync+0x148/0x180
irq event stamp: 2082270
Now, si->lock locked before calling 'del_from_avail_list()' to make sure
other thread see the si had been deleted and SWP_WRITEOK cleared together,
will not reinsert again.
This problem exists in versions after stable 5.10.y.
Link: https://lkml.kernel.org/r/20230404154716.23058-1-rongwei.wang@linux.alibaba.com
Fixes: a2468cc9bf ("swap: choose swap device according to numa node")
Tested-by: Yongchen Yin <wb-yyc939293@alibaba-inc.com>
Signed-off-by: Rongwei Wang <rongwei.wang@linux.alibaba.com>
Cc: Bagas Sanjaya <bagasdotme@gmail.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Aaron Lu <aaron.lu@intel.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
The current nilfs2 sysfs support has issues with the timing of creation
and deletion of sysfs entries, potentially leading to null pointer
dereferences, use-after-free, and lockdep warnings.
Some of the sysfs attributes for nilfs2 per-filesystem instance refer to
metadata file "cpfile", "sufile", or "dat", but
nilfs_sysfs_create_device_group that creates those attributes is executed
before the inodes for these metadata files are loaded, and
nilfs_sysfs_delete_device_group which deletes these sysfs entries is
called after releasing their metadata file inodes.
Therefore, access to some of these sysfs attributes may occur outside of
the lifetime of these metadata files, resulting in inode NULL pointer
dereferences or use-after-free.
In addition, the call to nilfs_sysfs_create_device_group() is made during
the locking period of the semaphore "ns_sem" of nilfs object, so the
shrinker call caused by the memory allocation for the sysfs entries, may
derive lock dependencies "ns_sem" -> (shrinker) -> "locks acquired in
nilfs_evict_inode()".
Since nilfs2 may acquire "ns_sem" deep in the call stack holding other
locks via its error handler __nilfs_error(), this causes lockdep to report
circular locking. This is a false positive and no circular locking
actually occurs as no inodes exist yet when
nilfs_sysfs_create_device_group() is called. Fortunately, the lockdep
warnings can be resolved by simply moving the call to
nilfs_sysfs_create_device_group() out of "ns_sem".
This fixes these sysfs issues by revising where the device's sysfs
interface is created/deleted and keeping its lifetime within the lifetime
of the metadata files above.
Link: https://lkml.kernel.org/r/20230330205515.6167-1-konishi.ryusuke@gmail.com
Fixes: dd70edbde2 ("nilfs2: integrate sysfs support into driver")
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@gmail.com>
Reported-by: syzbot+979fa7f9c0d086fdc282@syzkaller.appspotmail.com
Link: https://lkml.kernel.org/r/0000000000003414b505f7885f7e@google.com
Reported-by: syzbot+5b7d542076d9bddc3c6a@syzkaller.appspotmail.com
Link: https://lkml.kernel.org/r/0000000000006ac86605f5f44eb9@google.com
Cc: Viacheslav Dubeyko <slava@dubeyko.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Device exclusive page table entries are used to prevent CPU access to a
page whilst it is being accessed from a device. Typically this is used to
implement atomic operations when the underlying bus does not support
atomic access. When a CPU thread encounters a device exclusive entry it
locks the page and restores the original entry after calling mmu notifiers
to signal drivers that exclusive access is no longer available.
The device exclusive entry holds a reference to the page making it safe to
access the struct page whilst the entry is present. However the fault
handling code does not hold the PTL when taking the page lock. This means
if there are multiple threads faulting concurrently on the device
exclusive entry one will remove the entry whilst others will wait on the
page lock without holding a reference.
This can lead to threads locking or waiting on a folio with a zero
refcount. Whilst mmap_lock prevents the pages getting freed via munmap()
they may still be freed by a migration. This leads to warnings such as
PAGE_FLAGS_CHECK_AT_FREE due to the page being locked when the refcount
drops to zero.
Fix this by trying to take a reference on the folio before locking it.
The code already checks the PTE under the PTL and aborts if the entry is
no longer there. It is also possible the folio has been unmapped, freed
and re-allocated allowing a reference to be taken on an unrelated folio.
This case is also detected by the PTE check and the folio is unlocked
without further changes.
Link: https://lkml.kernel.org/r/20230330012519.804116-1-apopple@nvidia.com
Fixes: b756a3b5e7 ("mm: device exclusive memory access")
Signed-off-by: Alistair Popple <apopple@nvidia.com>
Reviewed-by: Ralph Campbell <rcampbell@nvidia.com>
Reviewed-by: John Hubbard <jhubbard@nvidia.com>
Acked-by: David Hildenbrand <david@redhat.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
The finalization of nilfs_segctor_thread() can race with
nilfs_segctor_kill_thread() which terminates that thread, potentially
causing a use-after-free BUG as KASAN detected.
At the end of nilfs_segctor_thread(), it assigns NULL to "sc_task" member
of "struct nilfs_sc_info" to indicate the thread has finished, and then
notifies nilfs_segctor_kill_thread() of this using waitqueue
"sc_wait_task" on the struct nilfs_sc_info.
However, here, immediately after the NULL assignment to "sc_task", it is
possible that nilfs_segctor_kill_thread() will detect it and return to
continue the deallocation, freeing the nilfs_sc_info structure before the
thread does the notification.
This fixes the issue by protecting the NULL assignment to "sc_task" and
its notification, with spinlock "sc_state_lock" of the struct
nilfs_sc_info. Since nilfs_segctor_kill_thread() does a final check to
see if "sc_task" is NULL with "sc_state_lock" locked, this can eliminate
the race.
Link: https://lkml.kernel.org/r/20230327175318.8060-1-konishi.ryusuke@gmail.com
Reported-by: syzbot+b08ebcc22f8f3e6be43a@syzkaller.appspotmail.com
Link: https://lkml.kernel.org/r/00000000000000660d05f7dfa877@google.com
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
This patch fixes an issue that a hugetlb uffd-wr-protected mapping can be
writable even with uffd-wp bit set. It only happens with hugetlb private
mappings, when someone firstly wr-protects a missing pte (which will
install a pte marker), then a write to the same page without any prior
access to the page.
Userfaultfd-wp trap for hugetlb was implemented in hugetlb_fault() before
reaching hugetlb_wp() to avoid taking more locks that userfault won't
need. However there's one CoW optimization path that can trigger
hugetlb_wp() inside hugetlb_no_page(), which will bypass the trap.
This patch skips hugetlb_wp() for CoW and retries the fault if uffd-wp bit
is detected. The new path will only trigger in the CoW optimization path
because generic hugetlb_fault() (e.g. when a present pte was
wr-protected) will resolve the uffd-wp bit already. Also make sure
anonymous UNSHARE won't be affected and can still be resolved, IOW only
skip CoW not CoR.
This patch will be needed for v5.19+ hence copy stable.
[peterx@redhat.com: v2]
Link: https://lkml.kernel.org/r/ZBzOqwF2wrHgBVZb@x1n
[peterx@redhat.com: v3]
Link: https://lkml.kernel.org/r/20230324142620.2344140-1-peterx@redhat.com
Link: https://lkml.kernel.org/r/20230321191840.1897940-1-peterx@redhat.com
Fixes: 166f3ecc0d ("mm/hugetlb: hook page faults for uffd write protection")
Signed-off-by: Peter Xu <peterx@redhat.com>
Reported-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
Tested-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
Acked-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Axel Rasmussen <axelrasmussen@google.com>
Cc: Mike Rapoport <rppt@linux.vnet.ibm.com>
Cc: Nadav Amit <nadav.amit@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Use the maple tree in RCU mode for VMA tracking.
The maple tree tracks the stack and is able to update the pivot
(lower/upper boundary) in-place to allow the page fault handler to write
to the tree while holding just the mmap read lock. This is safe as the
writes to the stack have a guard VMA which ensures there will always be a
NULL in the direction of the growth and thus will only update a pivot.
It is possible, but not recommended, to have VMAs that grow up/down
without guard VMAs. syzbot has constructed a testcase which sets up a VMA
to grow and consume the empty space. Overwriting the entire NULL entry
causes the tree to be altered in a way that is not safe for concurrent
readers; the readers may see a node being rewritten or one that does not
match the maple state they are using.
Enabling RCU mode allows the concurrent readers to see a stable node and
will return the expected result.
[Liam.Howlett@Oracle.com: we don't need to free the nodes with RCU[
Link: https://lore.kernel.org/linux-mm/000000000000b0a65805f663ace6@google.com/
Link: https://lkml.kernel.org/r/20230227173632.3292573-9-surenb@google.com
Fixes: d4af56c5c7 ("mm: start tracking VMAs with maple tree")
Signed-off-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Reported-by: syzbot+8d95422d3537159ca390@syzkaller.appspotmail.com
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Dereferencing RCU objects within the RCU callback without the RCU check
has caused lockdep to complain. Fix the RCU dereferencing by using the
RCU callback lock to ensure the operation is safe.
Also stop creating a new lock to use for dereferencing during destruction
of the tree or subtree. Instead, pass through a pointer to the tree that
has the lock that is held for RCU dereferencing checking. It also does
not make sense to use the maple state in the freeing scenario as the tree
walk is a special case where the tree no longer has the normal encodings
and parent pointers.
Link: https://lkml.kernel.org/r/20230227173632.3292573-8-surenb@google.com
Fixes: 54a611b605 ("Maple Tree: add new data structure")
Signed-off-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Reported-by: Suren Baghdasaryan <surenb@google.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
During the development of the maple tree, the strategy of freeing multiple
nodes changed and, in the process, the pivots were reused to store
pointers to dead nodes. To ensure the readers see accurate pivots, the
writers need to mark the nodes as dead and call smp_wmb() to ensure any
readers can identify the node as dead before using the pivot values.
There were two places where the old method of marking the node as dead
without smp_wmb() were being used, which resulted in RCU readers seeing
the wrong pivot value before seeing the node was dead. Fix this race
condition by using mte_set_node_dead() which has the smp_wmb() call to
ensure the race is closed.
Add a WARN_ON() to the ma_free_rcu() call to ensure all nodes being freed
are marked as dead to ensure there are no other call paths besides the two
updated paths.
This is necessary for the RCU mode of the maple tree.
Link: https://lkml.kernel.org/r/20230227173632.3292573-6-surenb@google.com
Fixes: 54a611b605 ("Maple Tree: add new data structure")
Signed-off-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
