Eliminate kernel panics when getting invalid responses from controller.
Take controller offline instead of causing kernel panics.
Link: https://lore.kernel.org/r/159622929306.30579.16523318707596752828.stgit@brunhilda Reviewed-by: Scott Teel <scott.teel@microsemi.com> Reviewed-by: Scott Benesh <scott.benesh@microsemi.com> Reviewed-by: Prasad Munirathnam <Prasad.Munirathnam@microsemi.com> Reviewed-by: Martin Wilck <mwilck@suse.com> Signed-off-by: Kevin Barnett <kevin.barnett@microsemi.com> Signed-off-by: Don Brace <don.brace@microsemi.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Add topology filename override based on system DMI data matching,
typically to account for a different hardware layout.
In ACPI based systems, the tplg_filename is pre-defined in an ACPI
machine table. When a DMI quirk is detected, the
sof_pdata->tplg_filename is not set with the hard-coded ACPI value,
and instead is set with the DMI-specific filename.
syzbot is reporting that del_timer_sync() is called from
mwifiex_usb_cleanup_tx_aggr() from mwifiex_unregister_dev() without
checking timer_setup() from mwifiex_usb_tx_init() was called [1].
Ganapathi Bhat proposed a possibly cleaner fix, but it seems that
that fix was forgotten [2].
"grep -FrB1 'del_timer' drivers/ | grep -FA1 '.function)'" says that
currently there are 28 locations which call del_timer[_sync]() only if
that timer's function field was initialized (because timer_setup() sets
that timer's function field). Therefore, let's use same approach here.
The current code for bridge address events has two shortcomings in its
control sequence:
1. after disabling address events via PNSO, we don't flush the remaining
events from the event_wq. So if the feature is re-enabled fast
enough, stale events could leak over.
2. PNSO and the events' arrival via the READ ccw device are unordered.
So even if we flushed the workqueue, it's difficult to say whether
the READ device might produce more events onto the workqueue
afterwards.
Fix this by
1. explicitly fencing off the events when we no longer care, in the
READ device's event handler. This ensures that once we flush the
workqueue, it doesn't get additional address events.
2. Flush the workqueue after disabling the events & fencing them off.
As the code that triggers the flush will typically hold the sbp_lock,
we need to rework the worker code to avoid a deadlock here in case
of a 'notifications-stopped' event. In case of lock contention,
requeue such an event with a delay. We'll eventually aquire the lock,
or spot that the feature has been disabled and the event can thus be
discarded.
This leaves the theoretical race that a stale event could arrive
_after_ we re-enabled ourselves to receive events again. Such an event
would be impossible to distinguish from a 'good' event, nothing we can
do about it.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com> Reviewed-by: Alexandra Winter <wintera@linux.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
When a usrjquota or grpjquota mount option is used multiple times, we
will leak memory allocated for the file name. Make sure the last setting
is used and all the previous ones are properly freed.
Reported-by: syzbot+c9e294bbe0333a6b7640@syzkaller.appspotmail.com Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Sasha Levin <sashal@kernel.org>
A fb_ioctl() FBIOPUT_VSCREENINFO call with invalid xres setting
or yres setting in struct fb_var_screeninfo will result in a
KASAN: vmalloc-out-of-bounds failure in bitfill_aligned() as
the margins are being cleared. The margins are cleared in
chunks and if the xres setting or yres setting is a value of
zero upto the chunk size, the failure will occur.
Add a margin check to validate xres and yres settings.
While aborting the I/O, the firmware cleanup task timed out and driver
deleted the I/O from active command list. Some time later the firmware
sent the cleanup task response and driver again deleted the I/O from
active command list causing firmware to send completion for non-existent
I/O and list_del corruption of active command list.
Add fix to check if I/O is present before deleting it from the active
command list to ensure firmware sends valid I/O completion and protect
against list_del corruption.
Link: https://lore.kernel.org/r/20200908095657.26821-4-mrangankar@marvell.com Signed-off-by: Nilesh Javali <njavali@marvell.com> Signed-off-by: Manish Rangankar <mrangankar@marvell.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Protect active command list for non-I/O commands like login response,
logout response, text response, and recovery cleanup of active list to
avoid list corruption.
Link: https://lore.kernel.org/r/20200908095657.26821-5-mrangankar@marvell.com Signed-off-by: Nilesh Javali <njavali@marvell.com> Signed-off-by: Manish Rangankar <mrangankar@marvell.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
For short time cable pulls, the in-flight I/O to the firmware is never
cleaned up, resulting in the behaviour of stale I/O completion causing
list_del corruption and soft lockup of the system.
On link down event, mark all the connections for recovery, causing cleanup
of all the in-flight I/O immediately.
Link: https://lore.kernel.org/r/20200908095657.26821-7-mrangankar@marvell.com Signed-off-by: Nilesh Javali <njavali@marvell.com> Signed-off-by: Manish Rangankar <mrangankar@marvell.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The Acer One S1003 2-in-1 keyboard dock uses a Synaptics S910xx touchpad
which is connected to an ITE 8910 USB keyboard controller chip.
This keyboard has the same quirk for its rfkill / airplane mode hotkey as
other keyboards with ITE keyboard chips, it only sends a single release
event when pressed and released, it never sends a press event.
This commit adds this keyboards USB id to the hid-ite id-table, fixing
the rfkill key not working on this keyboard. Note that like for the
Acer Aspire Switch 10 (SW5-012) the id-table entry matches on the
HID_GROUP_GENERIC generic group so that hid-ite only binds to the
keyboard interface and the mouse/touchpad interface is left untouched
so that hid-multitouch can bind to it.
Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz> Signed-off-by: Sasha Levin <sashal@kernel.org>
When wlc_phy_txpwr_srom_read_lcnphy fails in wlc_phy_attach_lcnphy,
the allocated pi->u.pi_lcnphy is leaked, since struct brcms_phy will be
freed in the caller function.
Fix this by calling wlc_phy_detach_lcnphy in the error handler of
wlc_phy_txpwr_srom_read_lcnphy before returning.
Andrii reported that with latest clang, when building selftests, we have
error likes:
error: progs/test_sysctl_loop1.c:23:16: in function sysctl_tcp_mem i32 (%struct.bpf_sysctl*):
Looks like the BPF stack limit of 512 bytes is exceeded.
Please move large on stack variables into BPF per-cpu array map.
The error is triggered by the following LLVM patch:
https://reviews.llvm.org/D87134
For example, the following code is from test_sysctl_loop1.c:
static __always_inline int is_tcp_mem(struct bpf_sysctl *ctx)
{
volatile char tcp_mem_name[] = "net/ipv4/tcp_mem/very_very_very_very_long_pointless_string";
...
}
Without the above LLVM patch, the compiler did optimization to load the string
(59 bytes long) with 7 64bit loads, 1 8bit load and 1 16bit load,
occupying 64 byte stack size.
With the above LLVM patch, the compiler only uses 8bit loads, but subregister is 32bit.
So stack requirements become 4 * 59 = 236 bytes. Together with other stuff on
the stack, total stack size exceeds 512 bytes, hence compiler complains and quits.
To fix the issue, removing "volatile" key word or changing "volatile" to
"const"/"static const" does not work, the string is put in .rodata.str1.1 section,
which libbpf did not process it and errors out with
libbpf: elf: skipping unrecognized data section(6) .rodata.str1.1
libbpf: prog 'sysctl_tcp_mem': bad map relo against '.L__const.is_tcp_mem.tcp_mem_name'
in section '.rodata.str1.1'
Defining the string const as global variable can fix the issue as it puts the string constant
in '.rodata' section which is recognized by libbpf. In the future, when libbpf can process
'.rodata.str*.*' properly, the global definition can be changed back to local definition.
Defining tcp_mem_name as a global, however, triggered a verifier failure.
./test_progs -n 7/21
libbpf: load bpf program failed: Permission denied
libbpf: -- BEGIN DUMP LOG ---
libbpf:
invalid stack off=0 size=1
verification time 6975 usec
stack depth 160+64
processed 889 insns (limit 1000000) max_states_per_insn 4 total_states
14 peak_states 14 mark_read 10
libbpf: -- END LOG --
libbpf: failed to load program 'sysctl_tcp_mem'
libbpf: failed to load object 'test_sysctl_loop2.o'
test_bpf_verif_scale:FAIL:114
#7/21 test_sysctl_loop2.o:FAIL
This actually exposed a bpf program bug. In test_sysctl_loop{1,2}, we have code
like
const char tcp_mem_name[] = "<...long string...>";
...
char name[64];
...
for (i = 0; i < sizeof(tcp_mem_name); ++i)
if (name[i] != tcp_mem_name[i])
return 0;
In the above code, if sizeof(tcp_mem_name) > 64, name[i] access may be
out of bound. The sizeof(tcp_mem_name) is 59 for test_sysctl_loop1.c and
79 for test_sysctl_loop2.c.
Without promotion-to-global change, old compiler generates code where
the overflowed stack access is actually filled with valid value, so hiding
the bpf program bug. With promotion-to-global change, the code is different,
more specifically, the previous loading constants to stack is gone, and
"name" occupies stack[-64:0] and overflow access triggers a verifier error.
To fix the issue, adjust "name" buffer size properly.
Emit a warning when ->done or ->free are called on an already freed
srb. There is a hidden use-after-free bug in the driver which corrupts
the srb memory pool which originates from the cleanup callbacks.
An extensive search didn't bring any lights on the real problem. The
initial fix was to set both pointers to NULL and try to catch invalid
accesses. But instead the memory corruption was gone and the driver
didn't crash. Since not all calling places check for NULL pointer, add
explicitly default handlers. With this we workaround the memory
corruption and add a debug help.
Link: https://lore.kernel.org/r/20200908081516.8561-2-dwagner@suse.de Reviewed-by: Martin Wilck <mwilck@suse.com> Reviewed-by: Arun Easi <aeasi@marvell.com> Signed-off-by: Daniel Wagner <dwagner@suse.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
It is trivial to trigger a WARN_ON_ONCE(1) in iomap_dio_actor() by
unprivileged users which would taint the kernel, or worse - panic if
panic_on_warn or panic_on_taint is set. Hence, just convert it to
pr_warn_ratelimited() to let users know their workloads are racing.
Thank Dave Chinner for the initial analysis of the racing reproducers.
Signed-off-by: Qian Cai <cai@lca.pw> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Since l2cap_sock_teardown_cb doesn't acquire the channel lock before
setting the socket as zapped, it could potentially race with
l2cap_sock_release which frees the socket. Thus, wait until the cleanup
is complete before marking the socket as zapped.
This race was reproduced on a JBL GO speaker after the remote device
rejected L2CAP connection due to resource unavailability.
Here is a dmesg log with debug logs from a repro of this bug:
[ 3465.424086] Bluetooth: hci_core.c:hci_acldata_packet() hci0 len 16 handle 0x0003 flags 0x0002
[ 3465.424090] Bluetooth: hci_conn.c:hci_conn_enter_active_mode() hcon 00000000cfedd07d mode 0
[ 3465.424094] Bluetooth: l2cap_core.c:l2cap_recv_acldata() conn 000000007eae8952 len 16 flags 0x2
[ 3465.424098] Bluetooth: l2cap_core.c:l2cap_recv_frame() len 12, cid 0x0001
[ 3465.424102] Bluetooth: l2cap_core.c:l2cap_raw_recv() conn 000000007eae8952
[ 3465.424175] Bluetooth: l2cap_core.c:l2cap_sig_channel() code 0x03 len 8 id 0x0c
[ 3465.424180] Bluetooth: l2cap_core.c:l2cap_connect_create_rsp() dcid 0x0045 scid 0x0000 result 0x02 status 0x00
[ 3465.424189] Bluetooth: l2cap_core.c:l2cap_chan_put() chan 000000006acf9bff orig refcnt 4
[ 3465.424196] Bluetooth: l2cap_core.c:l2cap_chan_del() chan 000000006acf9bff, conn 000000007eae8952, err 111, state BT_CONNECT
[ 3465.424203] Bluetooth: l2cap_sock.c:l2cap_sock_teardown_cb() chan 000000006acf9bff state BT_CONNECT
[ 3465.424221] Bluetooth: l2cap_core.c:l2cap_chan_put() chan 000000006acf9bff orig refcnt 3
[ 3465.424226] Bluetooth: hci_core.h:hci_conn_drop() hcon 00000000cfedd07d orig refcnt 6
[ 3465.424234] BUG: spinlock bad magic on CPU#2, kworker/u17:0/159
[ 3465.425626] Bluetooth: hci_sock.c:hci_sock_sendmsg() sock 000000002bb0cb64 sk 00000000a7964053
[ 3465.430330] lock: 0xffffff804410aac0, .magic: 00000000, .owner: <none>/-1, .owner_cpu: 0
[ 3465.430332] Causing a watchdog bite!
I got a use-after-free report when doing some fuzz test:
If ttm_bo_init() fails, the "gbo" and "gbo->bo.base" will be
freed by ttm_buffer_object_destroy() in ttm_bo_init(). But
then drm_gem_vram_create() and drm_gem_vram_init() will free
"gbo" and "gbo->bo.base" again.
If ttm_bo_init() fails, the "gbo" will be freed by
ttm_buffer_object_destroy() in ttm_bo_init(). But then
drm_gem_vram_create() and drm_gem_vram_init() will free
"gbo" again.
Reported-by: Hulk Robot <hulkci@huawei.com> Reported-by: butt3rflyh4ck <butterflyhuangxx@gmail.com> Signed-off-by: Jia Yang <jiayang5@huawei.com> Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de> Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de> Link: https://patchwork.freedesktop.org/patch/msgid/20200714083238.28479-2-tzimmermann@suse.de Signed-off-by: Sasha Levin <sashal@kernel.org>
Some integrated OHCI controller hubs do not expose all ports of the hub
to pins on the SoC. In some cases the unconnected ports generate
spurious over-current events. For example the Broadcom 56060/Ranger 2 SoC
contains a nominally 3 port hub but only the first port is wired.
Default behaviour for ohci-platform driver is to use global over-current
protection mode (AKA "ganged"). This leads to the spurious over-current
events affecting all ports in the hub.
We now alter the default to use per-port over-current protection.
This patch results in the following configuration changes depending
on quirks:
- For quirk OHCI_QUIRK_SUPERIO no changes. These systems remain set up
for ganged power switching and no over-current protection.
- For quirk OHCI_QUIRK_AMD756 or OHCI_QUIRK_HUB_POWER power switching
remains at none, while over-current protection is now guaranteed to be
set to per-port rather than the previous behaviour where it was either
none or global over-current protection depending on the value at
function entry.
Suggested-by: Alan Stern <stern@rowland.harvard.edu> Acked-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Hamish Martin <hamish.martin@alliedtelesis.co.nz> Link: https://lore.kernel.org/r/20200910212512.16670-1-hamish.martin@alliedtelesis.co.nz Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
There's an overflow bug in the realtime allocator. If the rt volume is
large enough to handle a single allocation request that is larger than
the maximum bmap extent length and the rt bitmap ends exactly on a
bitmap block boundary, it's possible that the near allocator will try to
check the freeness of a range that extends past the end of the bitmap.
This fails with a corruption error and shuts down the fs.
Therefore, constrain maxlen so that the range scan cannot run off the
end of the rt bitmap.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Sasha Levin <sashal@kernel.org>
If dev_pm_opp_attach_genpd() is called multiple times (once for each CPU
sharing the table), then it would result in unwanted behavior like
memory leak, attaching the domain multiple times, etc.
Handle that by checking and returning earlier if the domains are already
attached. Now that dev_pm_opp_detach_genpd() can get called multiple
times as well, we need to protect that too.
Note that the virtual device pointers aren't returned in this case, as
they may become unavailable to some callers during the middle of the
operation.
unlock_new_inode() is only meant to be called after a new inode has
already been inserted into the hash table. But reiserfs_new_inode() can
call it even before it has inserted the inode, triggering the WARNING in
unlock_new_inode(). Fix this by only calling unlock_new_inode() if the
inode has the I_NEW flag set, indicating that it's in the table.
This addresses the syzbot report "WARNING in unlock_new_inode"
(https://syzkaller.appspot.com/bug?extid=187510916eb6a14598f7).
Link: https://lore.kernel.org/r/20200628070057.820213-1-ebiggers@kernel.org Reported-by: syzbot+187510916eb6a14598f7@syzkaller.appspotmail.com Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Sasha Levin <sashal@kernel.org>
Protect against potential stack overflow that might happen when bpf2bpf
calls get combined with tailcalls. Limit the caller's stack depth for
such case down to 256 so that the worst case scenario would result in 8k
stack size (32 which is tailcall limit * 256 = 8k).
The T820, G31 & G52 GPUs integrated by Amlogic in the respective GXM,
G12A/SM1 & G12B SoCs needs a quirk in the PWR registers at the GPU reset
time.
Since the Amlogic's integration of the GPU cores with the SoC is not
publicly documented we do not know what does these values, but they
permit having a fully functional GPU running with Panfrost.
Signed-off-by: Neil Armstrong <narmstrong@baylibre.com>
[Steven: Fix typo in commit log] Reviewed-by: Steven Price <steven.price@arm.com> Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com> Signed-off-by: Steven Price <steven.price@arm.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200916150147.25753-3-narmstrong@baylibre.com Signed-off-by: Sasha Levin <sashal@kernel.org>
This adds the required GPU quirks, including the quirk in the PWR
registers at the GPU reset time and the IOMMU quirk for shareability
issues observed on G52 in Amlogic G12B SoCs.
Calls to usb_kill_anchored_urbs() after usb_kill_urb() on multiprocessor
systems create a race condition in which usb_kill_anchored_urbs() deallocates
the URB before the completer callback is called in usb_kill_urb(), resulting
in a use-after-free.
To fix this, add proper lock protection to usb_kill_urb() calls that can
possibly run concurrently with usb_kill_anchored_urbs().
This patch implements error handling and propagates the error value of
flexcan_chip_stop(). This function will be called from flexcan_suspend()
in an upcoming patch in some SoCs which support LPSR mode.
Add a new function flexcan_chip_stop_disable_on_error() that tries to
disable the chip even in case of errors.
Signed-off-by: Joakim Zhang <qiangqing.zhang@nxp.com>
[mkl: introduce flexcan_chip_stop_disable_on_error() and use it in flexcan_close()] Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de> Link: https://lore.kernel.org/r/20200922144429.2613631-11-mkl@pengutronix.de Signed-off-by: Sasha Levin <sashal@kernel.org>
When shifting a boolean variable by more than 31 bits and putting the
result into a u64 variable, we need to cast the boolean into unsigned 64
bits to prevent possible overflow.
Reported-by: kernel test robot <lkp@intel.com> Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Handle broken union functional descriptors where the master-interface
doesn't exist or where its class is of neither Communication or Data
type (as required by the specification) by falling back to
"combined-interface" probing.
Note that this still allows for handling union descriptors with switched
interfaces.
This specifically makes the Whistler radio scanners TRX series devices
work with the driver without adding further quirks to the device-id
table.
Reported-by: Daniel Caujolle-Bert <f1rmb.daniel@gmail.com> Tested-by: Daniel Caujolle-Bert <f1rmb.daniel@gmail.com> Acked-by: Oliver Neukum <oneukum@suse.com> Signed-off-by: Johan Hovold <johan@kernel.org> Link: https://lore.kernel.org/r/20200921135951.24045-3-johan@kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The vht capability of MAX_MPDU_LENGTH is 11454 in rtw88; however, the rx
buffer size for each packet is 8192. When receiving packets that are
larger than rx buffer size, it will leads to rx buffer ring overflow.
When we fail to read inode, some data accessed in udf_evict_inode() may
be uninitialized. Move the accesses to !is_bad_inode() branch.
Reported-by: syzbot+91f02b28f9bb5f5f1341@syzkaller.appspotmail.com Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Sasha Levin <sashal@kernel.org>
Although UDF standard allows it, we don't support sparing table larger
than a single block. Check it during mount so that we don't try to
access memory beyond end of buffer.
Reported-by: syzbot+9991561e714f597095da@syzkaller.appspotmail.com Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Sasha Levin <sashal@kernel.org>
There are reports that 8822CE fails to work rtw88 with "failed to read DBI
register" error. Also I have a system with 8723DE which freezes the whole
system when the rtw88 is probing the device.
According to [1], platform firmware may not properly power manage the
device during shutdown. I did some expirements and putting the device to
D3 can workaround the issue.
So let's power cycle the device by putting the device to D3 at shutdown
to prevent the issue from happening.
BUG: KASAN: use-after-free in __lock_acquire+0x3fd4/0x4180
kernel/locking/lockdep.c:3831
Read of size 8 at addr ffff8880683b0018 by task syz-executor.0/3377
Since struct _mic_vring_info and vring are allocated together and follow
vring, if the vring_size() is not four bytes aligned, which will cause
the start address of struct _mic_vring_info is not four byte aligned.
For example, when vring entries is 128, the vring_size() will be 5126
bytes. The _mic_vring_info struct layout in ddr looks like:
0x90002400: 0000000000390000EE0100000000C0FF
Here 0x39 is the avail_idx member, and 0xC0FFEE01 is the magic member.
When EP use ioread32(magic) to reads the magic in RC's share memory, it
will cause kernel panic on ARM64 platform due to the cross-byte io read.
Here read magic in user space use le32toh(vr0->info->magic) will meet
the same issue.
So add round_up(x,4) for vring_size, then the struct _mic_vring_info
will store in this way:
0x90002400: 000000000000000000000039C0FFEE01
Which will avoid kernel panic when read magic in struct _mic_vring_info.
trace-cmd report doesn't show events from target subsystem because
scsi_command_size() leaks through event format string:
[target:target_sequencer_start] function scsi_command_size not defined
[target:target_cmd_complete] function scsi_command_size not defined
Addition of scsi_command_size() to plugin_scsi.c in trace-cmd doesn't
help because an expression is used inside TP_printk(). trace-cmd event
parser doesn't understand minus sign inside [ ]:
Error: expected ']' but read '-'
Rather than duplicating kernel code in plugin_scsi.c, provide a dedicated
field for CONTROL byte.
Link: https://lore.kernel.org/r/20200929125957.83069-1-r.bolshakov@yadro.com Reviewed-by: Mike Christie <michael.christie@oracle.com> Signed-off-by: Roman Bolshakov <r.bolshakov@yadro.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
get_gendisk grabs a reference on the disk and file operation, so this
code will leak both of them while having absolutely no use for the
gendisk itself.
This effectively reverts commit 2df83fa4bce421f ("PM / Hibernate: Use
get_gendisk to verify partition if resume_file is integer format")
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
This is because pcpu_freelist_head.lock is accessed in both NMI and
non-NMI context. Fix this issue by using raw_spin_trylock() in NMI.
Since NMI interrupts non-NMI context, when NMI context tries to lock the
raw_spinlock, non-NMI context of the same CPU may already have locked a
lock and is blocked from unlocking the lock. For a system with N CPUs,
there could be N NMIs at the same time, and they may block N non-NMI
raw_spinlocks. This is tricky for pcpu_freelist_push(), where unlike
_pop(), failing _push() means leaking memory. This issue is more likely to
trigger in non-SMP system.
Fix this issue with an extra list, pcpu_freelist.extralist. The extralist
is primarily used to take _push() when raw_spin_trylock() failed on all
the per CPU lists. It should be empty most of the time. The following
table summarizes the behavior of pcpu_freelist in NMI and non-NMI:
non-NMI pop(): use _lock(); check per CPU lists first;
if all per CPU lists are empty, check extralist;
if extralist is empty, return NULL.
non-NMI push(): use _lock(); only push to per CPU lists.
NMI pop(): use _trylock(); check per CPU lists first;
if all per CPU lists are locked or empty, check extralist;
if extralist is locked or empty, return NULL.
NMI push(): use _trylock(); check per CPU lists first;
if all per CPU lists are locked; try push to extralist;
if extralist is also locked, keep trying on per CPU lists.
Reported-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Song Liu <songliubraving@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Martin KaFai Lau <kafai@fb.com> Link: https://lore.kernel.org/bpf/20201005165838.3735218-1-songliubraving@fb.com Signed-off-by: Sasha Levin <sashal@kernel.org>
As expected, when the device detect a MMIC error, it returns a specific
status. However, it also strip IV from the frame (don't ask me why).
So, with the current code, mac80211 detects a corrupted frame and it
drops it before it handle the MMIC error. The expected behavior would be
to detect MMIC error then to renegotiate the EAP session.
So, this patch correctly informs mac80211 that IV is not available. So,
mac80211 correctly takes into account the MMIC error.
GRE tunnel has its own header_ops, ipgre_header_ops, and sets it
conditionally. When it is set, it assumes the outer IP header is
already created before ipgre_xmit().
This is not true when we send packets through a raw packet socket,
where L2 headers are supposed to be constructed by user. Packet
socket calls dev_validate_header() to validate the header. But
GRE tunnel does not set dev->hard_header_len, so that check can
be simply bypassed, therefore uninit memory could be passed down
to ipgre_xmit(). Similar for dev->needed_headroom.
dev->hard_header_len is supposed to be the length of the header
created by dev->header_ops->create(), so it should be used whenever
header_ops is set, and dev->needed_headroom should be used when it
is not set.
Reported-and-tested-by: syzbot+4a2c52677a8a1aa283cb@syzkaller.appspotmail.com Cc: William Tu <u9012063@gmail.com> Acked-by: Willem de Bruijn <willemb@google.com> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Acked-by: Xie He <xie.he.0141@gmail.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Number of bytes allocated for mft record should be equal to the mft record
size stored in ntfs superblock as reported by syzbot, userspace might
trigger out-of-bounds read by dereferencing ctx->attr in ntfs_attr_find()
pm_runtime_get_sync() increments the runtime PM usage counter even
when it returns an error code. Thus a pairing decrement is needed on
the error handling path to keep the counter balanced. For other error
paths after this call, things are the same.
Fix this by adding pm_runtime_put_noidle() after 'err_runtime_disable'
label. But in this case, the error path after pm_runtime_put_sync()
will decrease PM usage counter twice. Thus add an extra
pm_runtime_get_noresume() in this path to balance PM counter.
The patch just remembers the "nodes" entry pointer in space as I think
it's created as subdirectory when parent "spaces" is created. In
function drop_space() we will lost the pointer reference to nds because
configfs_remove_default_groups(). However as this subdirectory is always
available when "spaces" exists it will just be freed when "spaces" will be
freed.
Signed-off-by: Alexander Aring <aahringo@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
There are few list handling issues while adding and deleting
node in the registered buf list in the driver.
1. list addition - buffer added into the list during buf_init
while not deleted during cleanup.
2. list deletion - In capture streamoff, the list was reinitialized.
As a result, if any node was present in the list, it would
lead to issue while cleaning up that node during buf_cleanup.
When aspect_ratio_crop_init() fails, curr_stream needs
to be freed just like what we've done in the following
error paths. However, current code is returning directly
and ends up leaking memory.
bFrameIndex and bFormatIndex can be negotiated by the camera during
probing, resulting in the camera choosing a different format than
expected. v4l2 can already accommodate such changes, but the code was
not updating the proper fields.
Without such a change, v4l2 would potentially interpret the payload
incorrectly, causing corrupted output. This was happening on the
Elgato HD60 S+, which currently always renegotiates to format 1.
As an aside, the Elgato firmware is buggy and should not be renegotating,
but it is still a valid thing for the camera to do. Both macOS and Windows
will properly probe and read uncorrupted images from this camera.
With this change, both qv4l2 and chromium can now read uncorrupted video
from the Elgato HD60 S+.
[Add blank lines, remove periods at the of messages]
If an exception needs to be handled while reading an MSR - which is in
most of the cases caused by a #GP on a non-existent MSR - then this
is most likely the incarnation of a BIOS or a hardware bug. Such bug
violates the architectural guarantee that MCA banks are present with all
MSRs belonging to them.
The proper fix belongs in the hardware/firmware - not in the kernel.
Handling an #MC exception which is raised while an NMI is being handled
would cause the nasty NMI nesting issue because of the shortcoming of
IRET of reenabling NMIs when executed. And the machine is in an #MC
context already so <Deity> be at its side.
Tracing MSR accesses while in #MC is another no-no due to tracing being
inherently a bad idea in atomic context:
vmlinux.o: warning: objtool: do_machine_check()+0x4a: call to mce_rdmsrl() leaves .noinstr.text section
so remove all that "additional" functionality from mce_rdmsrl() and
provide it with a special exception handler which panics the machine
when that MSR is not accessible.
The exception handler prints a human-readable message explaining what
the panic reason is but, what is more, it panics while in the #GP
handler and latter won't have executed an IRET, thus opening the NMI
nesting issue in the case when the #MC has happened while handling
an NMI. (#MC itself won't be reenabled until MCG_STATUS hasn't been
cleared).
Suggested-by: Andy Lutomirski <luto@kernel.org> Suggested-by: Peter Zijlstra <peterz@infradead.org>
[ Add missing prototypes for ex_handler_* ] Reported-by: kernel test robot <lkp@intel.com> Signed-off-by: Borislav Petkov <bp@suse.de> Reviewed-by: Tony Luck <tony.luck@intel.com> Link: https://lkml.kernel.org/r/20200906212130.GA28456@zn.tnic Signed-off-by: Sasha Levin <sashal@kernel.org>
Use a clock divider tuned to a 200MHz FSI bus frequency (the maximum). Use
of the previous divider at 200MHz results in corrupt data from endpoint
devices. Ideally the clock divider would be calculated from the FSI clock,
but that would require some significant work on the FSI driver. With FSI
frequencies slower than 200MHz, the SPI clock will simply run slower, but
safely.
Signed-off-by: Brad Bishop <bradleyb@fuzziesquirrel.com> Signed-off-by: Eddie James <eajames@linux.ibm.com> Signed-off-by: Joel Stanley <joel@jms.id.au> Link: https://lore.kernel.org/r/20200909222857.28653-3-eajames@linux.ibm.com Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
In bttv_probe if some functions such as pci_enable_device,
pci_set_dma_mask and request_mem_region fails the allocated
memory for btv should be released.
pm_runtime_get_sync() increments the runtime PM usage counter even
when it returns an error code. Thus a pairing decrement is needed on
the error handling path to keep the counter balanced.
pm_runtime_get_sync() increments the runtime PM usage counter even
when it returns an error code. Thus a pairing decrement is needed on
the error handling path to keep the counter balanced.
Signed-off-by: Dinghao Liu <dinghao.liu@zju.edu.cn> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
pm_runtime_get_sync() increments the runtime PM usage counter even
when it returns an error code. Thus a pairing decrement is needed on
the error handling path to keep the counter balanced.
Also, call pm_runtime_disable() when pm_runtime_get_sync() returns
an error code.
pm_runtime_get_sync() increments the runtime PM usage counter even
when it returns an error code. Thus a pairing decrement is needed on
the error handling path to keep the counter balanced.
pm_runtime_get_sync() increments the runtime PM usage counter even
when it returns an error code, causing incorrect ref count if
pm_runtime_put_noidle() is not called in error handling paths.
Thus call pm_runtime_put_noidle() if pm_runtime_get_sync() fails.
On calling pm_runtime_get_sync() the reference count of the device
is incremented. In case of failure, decrement the
reference count before returning the error.
On calling pm_runtime_get_sync() the reference count of the device
is incremented. In case of failure, decrement the
reference count before returning the error.
pm_runtime_get_sync() increments the runtime PM usage counter even
when it returns an error code, causing incorrect ref count if
pm_runtime_put_noidle() is not called in error handling paths.
Thus call pm_runtime_put_noidle() if pm_runtime_get_sync() fails.
delta_run_work() calls delta_get_sync() that increments
the reference counter. In case of failure, decrement the reference
count by calling delta_put_autosuspend().
The patrol scrubber in Skylake and Cascade Lake systems can be configured
to report uncorrected errors using a special signature in the machine
check bank and to signal using CMCI instead of machine check.
Update the severity calculation mechanism to allow specifying the model,
minimum stepping and range of machine check bank numbers.
Add a new rule to detect the special signature (on model 0x55, stepping
>=4 in any of the memory controller banks).
[ bp: Rewrite it.
aegl: Productize it. ]
Suggested-by: Youquan Song <youquan.song@intel.com> Signed-off-by: Borislav Petkov <bp@suse.de> Co-developed-by: Tony Luck <tony.luck@intel.com> Signed-off-by: Tony Luck <tony.luck@intel.com> Signed-off-by: Borislav Petkov <bp@suse.de> Link: https://lkml.kernel.org/r/20200930021313.31810-2-tony.luck@intel.com Signed-off-by: Sasha Levin <sashal@kernel.org>
The CRn accessor functions use __force_order as a dummy operand to
prevent the compiler from reordering CRn reads/writes with respect to
each other.
The fact that the asm is volatile should be enough to prevent this:
volatile asm statements should be executed in program order. However GCC
4.9.x and 5.x have a bug that might result in reordering. This was fixed
in 8.1, 7.3 and 6.5. Versions prior to these, including 5.x and 4.9.x,
may reorder volatile asm statements with respect to each other.
There are some issues with __force_order as implemented:
- It is used only as an input operand for the write functions, and hence
doesn't do anything additional to prevent reordering writes.
- It allows memory accesses to be cached/reordered across write
functions, but CRn writes affect the semantics of memory accesses, so
this could be dangerous.
- __force_order is not actually defined in the kernel proper, but the
LLVM toolchain can in some cases require a definition: LLVM (as well
as GCC 4.9) requires it for PIE code, which is why the compressed
kernel has a definition, but also the clang integrated assembler may
consider the address of __force_order to be significant, resulting in
a reference that requires a definition.
Fix this by:
- Using a memory clobber for the write functions to additionally prevent
caching/reordering memory accesses across CRn writes.
- Using a dummy input operand with an arbitrary constant address for the
read functions, instead of a global variable. This will prevent reads
from being reordered across writes, while allowing memory loads to be
cached/reordered across CRn reads, which should be safe.
Signed-off-by: Pavel Machek (CIP) <pavel@denx.de> Acked-by: John Allen <john.allen@amd.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: Sasha Levin <sashal@kernel.org>
Printing "Bad RIP value" if copy_code() fails can be misleading for
userspace pointers, since copy_code() can fail if the instruction
pointer is valid but the code is paged out. This is because copy_code()
calls copy_from_user_nmi() for userspace pointers, which disables page
fault handling.
This is reproducible in OOM situations, where it's plausible that the
code may be reclaimed in the time between entry into the kernel and when
this message is printed. This leaves a misleading log in dmesg that
suggests instruction pointer corruption has occurred, which may alarm
users.
Change the message to state the error condition more precisely.
syzbot is reporting unkillable task [1], for the caller is failing to
handle a corrupted filesystem image which attempts to access beyond
the end of the device. While we need to fix the caller, flooding the
console with handle_bad_sector() message is unlikely useful.
Commit 21653a4181ff ("i2c: core: Call i2c_acpi_install_space_handler()
before i2c_acpi_register_devices()")'s intention was to only move the
acpi_install_address_space_handler() call to the point before where
the ACPI declared i2c-children of the adapter where instantiated by
i2c_acpi_register_devices().
But i2c_acpi_install_space_handler() had a call to
acpi_walk_dep_device_list() hidden (that is I missed it) at the end
of it, so as an unwanted side-effect now acpi_walk_dep_device_list()
was also being called before i2c_acpi_register_devices().
Move the acpi_walk_dep_device_list() call to the end of
i2c_acpi_register_devices(), so that it is once again called *after*
the i2c_client-s hanging of the adapter have been created.
This fixes the Microsoft Surface Go 2 hanging at boot.
Fixes: 21653a4181ff ("i2c: core: Call i2c_acpi_install_space_handler() before i2c_acpi_register_devices()") Link: https://bugzilla.kernel.org/show_bug.cgi?id=209627 Reported-by: Rainer Finke <rainer@finke.cc> Reported-by: Kieran Bingham <kieran.bingham@ideasonboard.com> Suggested-by: Maximilian Luz <luzmaximilian@gmail.com> Tested-by: Kieran Bingham <kieran.bingham@ideasonboard.com> Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Wolfram Sang <wsa@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
perf_event.h has macros that define the field offsets in the
data_src bitmask in perf records. The SNOOPX and REMOTE offsets
were both 37. These are distinct fields, and the bitfield layout
in perf_mem_data_src confirms that SNOOPX should be at offset 38.
Fixes: 52839e653b5629bd ("perf tools: Add support for printing new mem_info encodings") Signed-off-by: Al Grant <al.grant@foss.arm.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Andi Kleen <ak@linux.intel.com> Link: https://lkml.kernel.org/r/4ac9f5cc-4388-b34a-9999-418a4099415d@foss.arm.com Signed-off-by: Sasha Levin <sashal@kernel.org>
765cc3a4b224e ("sched/core: Optimize sched_feat() for !CONFIG_SCHED_DEBUG builds")
made sched features static for !CONFIG_SCHED_DEBUG configurations, but
overlooked the CONFIG_SCHED_DEBUG=y and !CONFIG_JUMP_LABEL cases.
For the latter echoing changes to /sys/kernel/debug/sched_features has
the nasty effect of effectively changing what sched_features reports,
but without actually changing the scheduler behaviour (since different
translation units get different sysctl_sched_features).
Fix CONFIG_SCHED_DEBUG=y and !CONFIG_JUMP_LABEL configurations by properly
restructuring ifdefs.
Fixes: 765cc3a4b224e ("sched/core: Optimize sched_feat() for !CONFIG_SCHED_DEBUG builds") Co-developed-by: Daniel Bristot de Oliveira <bristot@redhat.com> Signed-off-by: Daniel Bristot de Oliveira <bristot@redhat.com> Signed-off-by: Juri Lelli <juri.lelli@redhat.com> Signed-off-by: Ingo Molnar <mingo@kernel.org> Acked-by: Patrick Bellasi <patrick.bellasi@matbug.net> Reviewed-by: Valentin Schneider <valentin.schneider@arm.com> Link: https://lore.kernel.org/r/20201013053114.160628-1-juri.lelli@redhat.com Signed-off-by: Sasha Levin <sashal@kernel.org>
The related system resources were not released when pci_set_dma_mask(),
pci_set_consistent_dma_mask(), or pci_iomap() return error in the
amd_ntb_init_pci() function. Add pci_release_regions() to fix it.
Fixes: a1b3695820aa ("NTB: Add support for AMD PCI-Express Non-Transparent Bridge") Signed-off-by: Kaige Li <likaige@loongson.cn> Signed-off-by: Jon Mason <jdmason@kudzu.us> Signed-off-by: Sasha Levin <sashal@kernel.org>
During shutdown the IOAPIC trigger mode is reset to edge triggered
while the vfio-pci INTx is still registered with a resampler.
This allows us to get into an infinite loop:
Commit 8be8f932e3db ("kvm: ioapic: Restrict lazy EOI update to
edge-triggered interrupts", 2020-05-04) acknowledges that this recursion
loop exists and tries to avoid it at the call to ioapic_lazy_update_eoi,
but at this point the scenario is already set, we have an edge interrupt
with resampler on the same gsi.
Fortunately, the only user of irq ack notifiers (in addition to resamplefd)
is i8254 timer interrupt reinjection. These are edge-triggered, so in
principle they would need the call to kvm_ioapic_update_eoi_one from
ioapic_lazy_update_eoi, but they already disable AVIC(*), so they don't
need the lazy EOI behavior. Therefore, remove the call to
kvm_ioapic_update_eoi_one from ioapic_lazy_update_eoi.
This fixes CVE-2020-27152. Note that this issue cannot happen with
SR-IOV assigned devices because virtual functions do not have INTx,
only MSI.
Fixes: f458d039db7e ("kvm: ioapic: Lazy update IOAPIC EOI") Suggested-by: Paolo Bonzini <pbonzini@redhat.com> Tested-by: Alex Williamson <alex.williamson@redhat.com> Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
When connecting a controller with a zero kato value using the following
command line
nvme connect -t tcp -n NQN -a ADDR -s PORT --keep-alive-tmo=0
the warning below can be reproduced:
WARNING: CPU: 1 PID: 241 at kernel/workqueue.c:1627 __queue_delayed_work+0x6d/0x90
with trace:
mod_delayed_work_on+0x59/0x90
nvmet_update_cc+0xee/0x100 [nvmet]
nvmet_execute_prop_set+0x72/0x80 [nvmet]
nvmet_tcp_try_recv_pdu+0x2f7/0x770 [nvmet_tcp]
nvmet_tcp_io_work+0x63f/0xb2d [nvmet_tcp]
...
This is caused by queuing up an uninitialized work. Althrough the
keep-alive timer is disabled during allocating the controller (fixed in 0d3b6a8d213a), ka_work still has a chance to run (called by
nvmet_start_ctrl).
Fixes: 0d3b6a8d213a ("nvmet: Disable keep-alive timer when kato is cleared to 0h") Signed-off-by: zhenwei pi <pizhenwei@bytedance.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Sasha Levin <sashal@kernel.org>
When an UE or memory error exception is encountered the MCE handler
tries to find the pfn using addr_to_pfn() which takes effective
address as an argument, later pfn is used to poison the page where
memory error occurred, recent rework in this area made addr_to_pfn
to run in real mode, which can be fatal as it may try to access
memory outside RMO region.
Have two helper functions to separate things to be done in real mode
and virtual mode without changing any functionality. This also fixes
the following error as the use of addr_to_pfn is now moved to virtual
mode.
Without this change following kernel crash is seen on hitting UE.
Every dump reported by OPAL is exported to userspace through a sysfs
interface and notified using kobject_uevent(). The userspace daemon
(opal_errd) then reads the dump and acknowledges that the dump is
saved safely to disk. Once acknowledged the kernel removes the
respective sysfs file entry causing respective resources to be
released including kobject.
However it's possible the userspace daemon may already be scanning
dump entries when a new sysfs dump entry is created by the kernel.
User daemon may read this new entry and ack it even before kernel can
notify userspace about it through kobject_uevent() call. If that
happens then we have a potential race between
dump_ack_store->kobject_put() and kobject_uevent which can lead to
use-after-free of a kernfs object resulting in a kernel crash.
This patch fixes this race by protecting the sysfs file
creation/notification by holding a reference count on kobject until we
safely send kobject_uevent().
The function create_dump_obj() returns the dump object which if used
by caller function will end up in use-after-free problem again.
However, the return value of create_dump_obj() function isn't being
used today and there is no need as well. Hence change it to return
void to make this fix complete.
There is an off-by-one array check that can lead to a out-of-bounds
write to devices->info[i]. Fix this by checking by using >= rather
than > for the size check. Also replace hard-coded array size limit
with ARRAY_SIZE on the array.
Addresses-Coverity: ("Out-of-bounds write") Fixes: cd9e9808d18f ("lightnvm: Support for Open-Channel SSDs") Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: Jens Axboe <axboe@kernel.dk> Signed-off-by: Sasha Levin <sashal@kernel.org>
projects / thirdparty / kernel / stable.git / log
