cpu.uclamp.min is a protection as described in cgroup-v2 Resource
Distribution Model
Documentation/admin-guide/cgroup-v2.rst
which means we try our best to preserve the minimum performance point of
tasks in this group. See full description of cpu.uclamp.min in the
cgroup-v2.rst.
But the current implementation makes it a limit, which is not what was
intended.
With this change the cgroup and per-task behaviors are the same, as
expected.
Additionally, we remove the confusing relationship between cgroup and
!user_defined flag.
We don't want for example RT tasks that are boosted by default to max to
change their boost value when they attach to a cgroup. If a cgroup wants
to limit the max performance point of tasks attached to it, then
cpu.uclamp.max must be set accordingly.
Or if they want to set different boost value based on cgroup, then
sysctl_sched_util_clamp_min_rt_default must be used to NOT boost to max
and set the right cpu.uclamp.min for each group to let the RT tasks
obtain the desired boost value when attached to that group.
As it stands the dependency on !user_defined flag adds an extra layer of
complexity that is not required now cpu.uclamp.min behaves properly as
a protection.
The propagation model of effective cpu.uclamp.min in child cgroups as
implemented by cpu_util_update_eff() is still correct. The parent
protection sets an upper limit of what the child cgroups will
effectively get.
Fixes: 3eac870a3247 (sched/uclamp: Use TG's clamps to restrict TASK's clamps) Signed-off-by: Qais Yousef <qais.yousef@arm.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lkml.kernel.org/r/20210510145032.1934078-2-qais.yousef@arm.com Signed-off-by: Sasha Levin <sashal@kernel.org>
The use of an enum named 'RST' conflicts with a #define macro
named 'RST' in arch/mips/include/asm/mach-rc32434/rb.h.
The MIPS use of RST was there first (AFAICT), so change the
media/i2c/ uses of RST to be named 'RSET'.
'git grep -w RSET' does not report any naming conflicts with the
new name.
This fixes multiple build errors:
arch/mips/include/asm/mach-rc32434/rb.h:15:14: error: expected identifier before '(' token
15 | #define RST (1 << 15)
| ^
drivers/media/i2c/s5c73m3/s5c73m3.h:356:2: note: in expansion of macro 'RST'
356 | RST,
| ^~~
../arch/mips/include/asm/mach-rc32434/rb.h:15:14: error: expected identifier before '(' token
15 | #define RST (1 << 15)
| ^
../drivers/media/i2c/s5k6aa.c:180:2: note: in expansion of macro 'RST'
180 | RST,
| ^~~
../arch/mips/include/asm/mach-rc32434/rb.h:15:14: error: expected identifier before '(' token
15 | #define RST (1 << 15)
| ^
../drivers/media/i2c/s5k5baf.c:238:2: note: in expansion of macro 'RST'
238 | RST,
| ^~~
and some others that I have trimmed.
Fixes: cac47f1822fc ("[media] V4L: Add S5C73M3 camera driver") Fixes: 8b99312b7214 ("[media] Add v4l2 subdev driver for S5K4ECGX sensor") Fixes: 7d459937dc09 ("[media] Add driver for Samsung S5K5BAF camera sensor") Fixes: bfa8dd3a0524 ("[media] v4l: Add v4l2 subdev driver for S5K6AAFX sensor") Signed-off-by: Randy Dunlap <rdunlap@infradead.org> Reported-by: kernel test robot <lkp@intel.com> Cc: Shawn Guo <shawnguo@kernel.org> Cc: Sascha Hauer <s.hauer@pengutronix.de> Cc: Pengutronix Kernel Team <kernel@pengutronix.de> Cc: Fabio Estevam <festevam@gmail.com> Cc: NXP Linux Team <linux-imx@nxp.com> Cc: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers) Cc: Andrzej Hajda <a.hajda@samsung.com> Cc: Sylwester Nawrocki <s.nawrocki@samsung.com> Cc: Sangwook Lee <sangwook.lee@linaro.org> 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>
The driver overrides the error codes returned by platform_get_irq() to
-ENOENT, so if it returns -EPROBE_DEFER, the driver would fail the probe
permanently instead of the deferred probing. Switch to propagating the
error code upstream, still checking/overriding IRQ0 as libata regards it
as "no IRQ" (thus polling) anyway...
The driver overrides the error codes returned by platform_get_irq() to
-EINVAL, so if it returns -EPROBE_DEFER, the driver would fail the probe
permanently instead of the deferred probing. Switch to propagating the
error code upstream, still checking/overriding IRQ0 as libata regards it
as "no IRQ" (thus polling) anyway...
Testing ixp4xx_crypto with CONFIG_DMA_API_DEBUG lead to the following error:
DMA-API: platform ixp4xx_crypto.0: device driver tries to free DMA memory it has not allocated [device address=0x0000000000000000] [size=24 bytes]
If pm_runtime resume logic fails, return the error code
provided by it, instead of -EAGAIN, as, depending on what
caused it to fail, it may not be something that would be
recovered.
Fixes: cbba45d43631 ("[media] smiapp: Use runtime PM") Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Acked-by: Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
There's a bug at s5p_cec_adap_enable(): if called to
disable the device, it should call pm_runtime_put()
instead of pm_runtime_disable(), as the goal here is to
decrement the usage_count and not to disable PM runtime.
The Venus code has a sort of watchdog that attempts to recover
from IP errors, implemented as a delayed work job, which
calls venus_sys_error_handler().
Right now, it has several issues:
1. It assumes that PM runtime resume never fails
2. It internally runs two while() loops that also assume that
PM runtime will never fail to go idle:
while (pm_runtime_active(core->dev_dec) || pm_runtime_active(core->dev_enc))
msleep(10);
...
while (core->pmdomains[0] && pm_runtime_active(core->pmdomains[0]))
usleep_range(1000, 1500);
3. It uses an OR to merge all return codes and then report to the user
4. If the hardware never recovers, it keeps running on every 10ms,
flooding the syslog with 2 messages (so, up to 200 messages
per second).
Rework the code, in order to prevent that, by:
1. check the return code from PM runtime resume;
2. don't let the while() loops run forever;
3. store the failed event;
4. use warn ratelimited when it fails to recover.
Fixes: af2c3834c8ca ("[media] media: venus: adding core part and helper functions") Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
ffz(), that has been used to count unused native CSs,
might cause undefined behaviour when called against ~0U.
To fix that, open code it with ffs(~value) - 1.
Fixes: 7d93aecdb58d ("spi: Add generic support for unused native cs with cs-gpios") Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Link: https://lore.kernel.org/r/20210420164425.40287-2-andriy.shevchenko@linux.intel.com Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The commit 7d93aecdb58d ("spi: Add generic support for unused native cs
with cs-gpios") excludes the valid case for the controllers that doesn't
need to switch native CS in order to perform the transfer, i.e. when
where <n> defines maximum of native CSs supported by the controller.
To allow this, bail out from spi_get_gpio_descs() conditionally for
the controllers which explicitly marked with SPI_MASTER_GPIO_SS.
Fixes: 7d93aecdb58d ("spi: Add generic support for unused native cs with cs-gpios") Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Link: https://lore.kernel.org/r/20210420164425.40287-1-andriy.shevchenko@linux.intel.com Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Commit c65e774fb3f6 ("x86/mm: Make PGDIR_SHIFT and PTRS_PER_P4D variable")
made PTRS_PER_P4D variable on x86 and introduced MAX_PTRS_PER_P4D as a
constant for cases which need a compile-time constant (e.g. fixed-size
arrays).
powerpc likewise has boot-time selectable MMU features which can cause
other mm "constants" to vary. For KASAN, we have some static
PTE/PMD/PUD/P4D arrays so we need compile-time maximums for all these
constants. Extend the MAX_PTRS_PER_ idiom, and place default definitions
in include/pgtable.h. These define MAX_PTRS_PER_x to be PTRS_PER_x unless
an architecture has defined MAX_PTRS_PER_x in its arch headers.
Clean up pgtable-nop4d.h and s390's MAX_PTRS_PER_P4D definitions while
we're at it: both can just pick up the default now.
isw_nr_in_flight is used to determine whether the inode switch queue
should be flushed from the umount path. Currently it's increased after
grabbing an inode and even scheduling the switch work. It means the
umount path can walk past cleanup_offline_cgwb() with active inode
references, which can result in a "Busy inodes after unmount." message and
use-after-free issues (with inode->i_sb which gets freed).
Fix it by incrementing isw_nr_in_flight before doing anything with the
inode and decrementing in the case when switching wasn't scheduled.
The problem hasn't yet been seen in the real life and was discovered by
Jan Kara by looking into the code.
Link: https://lkml.kernel.org/r/20210608230225.2078447-4-guro@fb.com Signed-off-by: Roman Gushchin <guro@fb.com> Suggested-by: Jan Kara <jack@suse.com> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Dave Chinner <dchinner@redhat.com> Cc: Dennis Zhou <dennis@kernel.org> Cc: Tejun Heo <tj@kernel.org> Cc: Jens Axboe <axboe@kernel.dk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
gcc points out a mistake in the mca driver that goes back to before the
git history:
arch/ia64/kernel/mca_drv.c: In function 'init_record_index_pools':
arch/ia64/kernel/mca_drv.c:346:54: error: expression does not compute the number of elements in this array; element typ
e is 'int', not 'size_t' {aka 'long unsigned int'} [-Werror=sizeof-array-div]
346 | for (i = 1; i < sizeof sal_log_sect_min_sizes/sizeof(size_t); i++)
| ^
This is the same as sizeof(size_t), which is two shorter than the actual
array. Use the ARRAY_SIZE() macro to get the correct calculation instead.
kthread_mod_delayed_work() might race with
kthread_cancel_delayed_work_sync() or another kthread_mod_delayed_work()
call. The function lets the other operation win when it sees
work->canceling counter set. And it returns @false.
But it should return @true as it is done by the related workqueue API, see
mod_delayed_work_on().
The reason is that the return value might be used for reference counting.
It has to distinguish the case when the number of queued works has changed
or stayed the same.
The change is safe. kthread_mod_delayed_work() return value is not
checked anywhere at the moment.
Link: https://lore.kernel.org/r/20210521163526.GA17916@redhat.com Link: https://lkml.kernel.org/r/20210610133051.15337-4-pmladek@suse.com Signed-off-by: Petr Mladek <pmladek@suse.com> Reported-by: Oleg Nesterov <oleg@redhat.com> Cc: Nathan Chancellor <nathan@kernel.org> Cc: Nick Desaulniers <ndesaulniers@google.com> Cc: Tejun Heo <tj@kernel.org> Cc: Minchan Kim <minchan@google.com> Cc: <jenhaochen@google.com> Cc: Martin Liu <liumartin@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
ll_new_hw_segment() is reached only in case of single range discard
merge, and we don't have max discard segment size limit actually, so
it is wrong to run the following check:
if (req->nr_phys_segments + nr_phys_segs > blk_rq_get_max_segments(req))
it may be always false since req->nr_phys_segments is initialized as
one, and bio's segment count is still 1, blk_rq_get_max_segments(reg)
is 1 too.
Fix the issue by not doing the check and bypassing the calculation of
discard request's nr_phys_segments.
Based on analysis from Wang Shanker.
Cc: Christoph Hellwig <hch@lst.de> Reported-by: Wang Shanker <shankerwangmiao@gmail.com> Signed-off-by: Ming Lei <ming.lei@redhat.com> Link: https://lore.kernel.org/r/20210628023312.1903255-1-ming.lei@redhat.com Signed-off-by: Jens Axboe <axboe@kernel.dk> Signed-off-by: Sasha Levin <sashal@kernel.org>
This is because MSM8939 has 3 APCS instances for Cluster0 (little cores),
Cluster1 (big cores) and CCI (Cache Coherent Interconnect). Although
only APCS of Cluster0 and Cluster1 have IPC bits, each of 3 APCS has
A53PLL clock control bits. That said, 3 'qcom-apcs-msm8916-clk' devices
need to be registered to instantiate all 3 clocks. Use PLATFORM_DEVID_AUTO
rather than PLATFORM_DEVID_NONE for platform_device_register_data() call
to fix the issue above.
In the other places where we update ses->status we protect the
updates via GlobalMid_Lock. So to be consistent add the same
locking around it in cifs_put_smb_ses where it was missing.
Addresses-Coverity: 1268904 ("Data race condition") Signed-off-by: Steve French <stfrench@microsoft.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
A custom DSDT file is mostly used during development or debugging,
and in that case it is quite likely to want to rebuild the kernel
after changing ONLY the content of the DSDT.
This patch adds the custom DSDT as a prerequisite to tables.o
to ensure a rebuild if the DSDT file is updated. Make will merge
the prerequisites from multiple rules for the same target.
Signed-off-by: Richard Fitzgerald <rf@opensource.cirrus.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The SPI core always reports a "MODALIAS=spi:<foo>", even if the device was
registered via OF. This means that this module won't auto-load if a DT has
for example has a node with a compatible "infineon,slb9670" string.
In that case kmod will expect a "MODALIAS=of:N*T*Cinfineon,slb9670" uevent
but instead will get a "MODALIAS=spi:slb9670", which is not present in the
kernel module aliases:
To workaround this issue, add in the SPI device ID table all the entries
that are present in the OF device ID table.
Reported-by: Alexander Wellbrock <a.wellbrock@mailbox.org> Signed-off-by: Javier Martinez Canillas <javierm@redhat.com> Tested-by: Peter Robinson <pbrobinson@gmail.com> Reviewed-by: Jarkko Sakkinen <jarkko@kernel.org> Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Some sorts of per-CPU clock sources have a history of going out of
synchronization with each other. However, this problem has purportedy been
solved in the past ten years. Except that it is all too possible that the
problem has instead simply been made less likely, which might mean that
some of the occasional "Marking clocksource 'tsc' as unstable" messages
might be due to desynchronization. How would anyone know?
Therefore apply CPU-to-CPU synchronization checking to newly unstable
clocksource that are marked with the new CLOCK_SOURCE_VERIFY_PERCPU flag.
Lists of desynchronized CPUs are printed, with the caveat that if it
is the reporting CPU that is itself desynchronized, it will appear that
all the other clocks are wrong. Just like in real life.
Reported-by: Chris Mason <clm@fb.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Acked-by: Feng Tang <feng.tang@intel.com> Link: https://lore.kernel.org/r/20210527190124.440372-2-paulmck@kernel.org Signed-off-by: Sasha Levin <sashal@kernel.org>
When the clocksource watchdog marks a clock as unstable, this might be due
to that clock being unstable or it might be due to delays that happen to
occur between the reads of the two clocks. Yes, interrupts are disabled
across those two reads, but there are no shortage of things that can delay
interrupts-disabled regions of code ranging from SMI handlers to vCPU
preemption. It would be good to have some indication as to why the clock
was marked unstable.
Therefore, re-read the watchdog clock on either side of the read from the
clock under test. If the watchdog clock shows an excessive time delta
between its pair of reads, the reads are retried.
The maximum number of retries is specified by a new kernel boot parameter
clocksource.max_cswd_read_retries, which defaults to three, that is, up to
four reads, one initial and up to three retries. If more than one retry
was required, a message is printed on the console (the occasional single
retry is expected behavior, especially in guest OSes). If the maximum
number of retries is exceeded, the clock under test will be marked
unstable. However, the probability of this happening due to various sorts
of delays is quite small. In addition, the reason (clock-read delays) for
the unstable marking will be apparent.
Reported-by: Chris Mason <clm@fb.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Acked-by: Feng Tang <feng.tang@intel.com> Link: https://lore.kernel.org/r/20210527190124.440372-1-paulmck@kernel.org Signed-off-by: Sasha Levin <sashal@kernel.org>
On HP Pavilion Gaming Laptop 15-cx0xxx, the ECDT EC and DSDT EC share
the same port addresses but different GPEs. And the DSDT GPE is the
right one to use.
The current code duplicates DSDT EC with ECDT EC if the port addresses
are the same, and uses ECDT GPE as a result, which breaks this machine.
Introduce a new quirk for the HP laptop to trust the DSDT GPE,
and avoid duplicating even if the port addresses are the same.
Link: https://bugzilla.kernel.org/show_bug.cgi?id=209989 Reported-and-tested-by: Shao Fu, Chen <leo881003@gmail.com> Signed-off-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Although in practice this can not occur (since IPv4 and IPv6 are the
only two cases currently supported), it is cleaner to avoid uninitialized
variable warnings.
Addresses smatch warning:
fs/cifs/cifs_swn.c:468 cifs_swn_store_swn_addr() error: uninitialized symbol 'port'.
Reported-by: kernel test robot <lkp@intel.com> Reported-by: Dan Carpenter <dan.carpenter@oracle.com> CC: Samuel Cabrero <scabrero@suse.de> Signed-off-by: Steve French <stfrench@microsoft.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Interlink is a special type of DFS link that resolves to a different
DFS domain-based namespace. To determine whether it is an interlink
or not, check if ReferralServers and StorageServers bits are set to 1
and 0 respectively in ReferralHeaderFlags, as specified in MS-DFSC
3.1.5.4.5 Determining Whether a Referral Response is an Interlink.
Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Aurelien Aptel <aaptel@suse.com> Signed-off-by: Steve French <stfrench@microsoft.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
We don't have a real fallocate in the SMB2 protocol so we used to emulate fallocate
by simply switching the file to become non-sparse. But as that could potantially consume
a lot more data than we intended to fallocate (large sparse file and fallocating a thin
slice in the middle) we would only do this IFF the fallocate request was for virtually
the entire file.
This patch improves this and starts allowing us to fallocate smaller chunks of a file by
overwriting the region with 0, for the parts that are unallocated.
The method used is to first query the server for FSCTL_QUERY_ALLOCATED_RANGES to find what
is unallocated in the fallocate range and then to only overwrite-with-zero the unallocated
ranges to fill in the holes.
As overwriting-with-zero is different from just allocating blocks, and potentially much
more expensive, we limit this to only allow fallocate ranges up to 1Mb in size.
Reported-by: kernel test robot <lkp@intel.com> Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Acked-by: Aurelien Aptel <aaptel@suse.com> Acked-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Steve French <stfrench@microsoft.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Add check for hv_is_hyperv_initialized() at the top of
init_hv_pci_drv(), so if the pci-hyperv driver is force-loaded on non
Hyper-V platforms, the init_hv_pci_drv() will exit immediately, without
any side effects, like assignments to hvpci_block_ops, etc.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com> Reported-and-tested-by: Mohammad Alqayeem <mohammad.alqyeem@nutanix.com> Reviewed-by: Wei Liu <wei.liu@kernel.org> Link: https://lore.kernel.org/r/1621984653-1210-1-git-send-email-haiyangz@microsoft.com Signed-off-by: Wei Liu <wei.liu@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The uncore memory frequency value from the mailbox command
CONFIG_TDP_GET_MEM_FREQ needs to be scaled based on the platform for
display. There is no single constant multiplier.
This change introduces CPU model specific memory frequency multiplier.
Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com> Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
There's little to no point in loading an EDAC driver running in a guest:
1) The CPU model reported by CPUID may not represent actual h/w
2) The hypervisor likely does not pass in access to memory controller devices
3) Hypervisors generally do not pass corrected error details to guests
Add a check in each of the Intel EDAC drivers for X86_FEATURE_HYPERVISOR
and simply return -ENODEV in the init routine.
When parsing a request in nvmet_fc_handle_fcp_rqst() we should not
check for invalid target ports; if we do the command is aborted
from the fcp layer, causing the host to assume a transport error.
Rather we should still forward this request to the nvmet layer, which
will then correctly fail the command with an appropriate error status.
Signed-off-by: Hannes Reinecke <hare@suse.de> Reviewed-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Sasha Levin <sashal@kernel.org>
The quirks added to asus-nb-wmi for the ASUS ROG Zephyrus G14 and G15 are
wrong, they tell the asus-wmi code to use the vendor specific WMI backlight
interface. But there is no such interface on these laptops.
As a side effect, these quirks stop the acpi_video driver to register since
they make acpi_video_get_backlight_type() return acpi_backlight_vendor,
leaving only the native AMD backlight driver in place, which is the one we
want. This happy coincidence is being replaced with a new quirk in
drivers/acpi/video_detect.c which actually sets the backlight_type to
acpi_backlight_native fixinf this properly. This reverts
commit 13bceda68fb9 ("platform/x86: asus-nb-wmi: add support for ASUS ROG
Zephyrus G14 and G15").
This is a preparation revert for reverting the "add support for ASUS ROG
Zephyrus G14 and G15" change. This reverts
commit 67186653c903 ("platform/x86: asus-nb-wmi: Drop duplicate DMI quirk
structures")
[16687.001777] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000008
...
[16687.163549] pc : __rq_qos_track+0x38/0x60
or
[ 997.690455] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000020
...
[ 997.850347] pc : __rq_qos_done+0x2c/0x50
Turns out it is caused by race between adding rq qos(wbt) and normal IO
because rq_qos_add can be run when IO is being submitted, fix this issue
by freezing queue before adding/deleting rq qos to queue.
rq_qos_exit() needn't to freeze queue because it is called after queue
has been frozen.
iolatency calls rq_qos_add() during allocating queue, so freezing won't
add delay because queue usage refcount works at atomic mode at that
time.
iocost calls rq_qos_add() when writing cgroup attribute file, that is
fine to freeze queue at that time since we usually freeze queue when
storing to queue sysfs attribute, meantime iocost only exists on the
root cgroup.
wbt_init calls it in blk_register_queue() and queue sysfs attribute
store(queue_wb_lat_store() when write it 1st time in case of !BLK_WBT_MQ),
the following patch will speedup the queue freezing in wbt_init.
Reported-by: Yi Zhang <yi.zhang@redhat.com> Cc: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Ming Lei <ming.lei@redhat.com> Reviewed-by: Bart Van Assche <bvanassche@acm.org> Tested-by: Yi Zhang <yi.zhang@redhat.com> Link: https://lore.kernel.org/r/20210609015822.103433-2-ming.lei@redhat.com Signed-off-by: Jens Axboe <axboe@kernel.dk> Signed-off-by: Sasha Levin <sashal@kernel.org>
This commit fixes a freeze on insertion of a Guitar Hero Live PS3/WiiU
USB dongle. Indeed, with the current implementation, inserting one of
those USB dongles will lead to a hard freeze. I apologize for not
catching this earlier, it didn't occur on my old laptop.
While the issue was isolated to memory alloc/free, I could not figure
out why it causes a freeze. So this patch fixes this issue by
simplifying memory allocation and usage.
We remind that for the dongle to work properly, a control URB needs to
be sent periodically. We used to alloc/free the URB each time this URB
needed to be sent.
With this patch, the memory for the URB is allocated on the probe, reused
for as long as the dongle is plugged in, and freed once the dongle is
unplugged.
The Elantech touchscreen/digitizer in the Surface Go mistakenly reports
having a battery. This results in a low battery message every time you
try to use the pen.
This patch adds a quirk to ignore the non-existent battery and
gets rid of the false low battery messages.
The laptop keyboard doesn't work on many MEDION notebooks, but the
keyboard works well under Windows and Unix.
Through debugging, we found this log in the dmesg:
ACPI: IRQ 1 override to edge, high
pnp 00:03: Plug and Play ACPI device, IDs PNP0303 (active)
And we checked the IRQ definition in the DSDT, it is:
IRQ (Level, ActiveLow, Exclusive, )
{1}
So the BIOS defines the keyboard IRQ to Level_Low, but the Linux
kernel override it to Edge_High. If the Linux kernel is modified
to skip the IRQ override, the keyboard will work normally.
From the existing comment in acpi_dev_get_irqresource(), the override
function only needs to be called when IRQ() or IRQNoFlags() is used
to populate the resource descriptor, and according to Section 6.4.2.1
of ACPI 6.4 [1], if IRQ() is empty or IRQNoFlags() is used, the IRQ
is High true, edge sensitive and non-shareable. ACPICA also assumes
that to be the case (see acpi_rs_set_irq[] in rsirq.c).
In accordance with the above, check 3 additional conditions
(EdgeSensitive, ActiveHigh and Exclusive) when deciding whether or
not to treat an ACPI_RESOURCE_TYPE_IRQ resource as "legacy", in which
case the IRQ override is applicable to it.
Although the system will not be in a good condition or it will not
boot if acpi_bus_init() fails, it is still necessary to put the
kobject in the error path before returning to avoid leaking memory.
Signed-off-by: Hanjun Guo <guohanjun@huawei.com>
[ rjw: Subject and changelog edits ] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
According to the ACPI spec, _CID returns a package containing
hardware ID's. Each element of an ASL package contains a reference
count from the parent package as well as the element itself.
Name (TEST, Package() {
"String object" // this package element has a reference count of 2
})
A memory leak was caused in the _CID repair function because it did
not decrement the reference count created by the package. Fix the
memory leak by calling acpi_ut_remove_reference on _CID package elements
that represent a hardware ID (_HID).
Link: https://github.com/acpica/acpica/commit/180cb539 Tested-by: Shawn Guo <shawn.guo@linaro.org> Signed-off-by: Erik Kaneda <erik.kaneda@intel.com> Signed-off-by: Bob Moore <robert.moore@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
I got some kmemleak report when a node was fenced. The user space tool
dlm_controld will therefore run some rmdir() in dlm configfs which was
triggering some memleaks. This patch stores the sps and cms attributes
which stores some handling for subdirectories of the configfs cluster
entry and free them if they get released as the parent directory gets
freed.
This patch fixes the error path handling in lowcomms_start(). We need to
cleanup some static allocated data structure and cleanup possible
workqueue if these have started.
Signed-off-by: Alexander Aring <aahringo@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The new openat2() syscall verifies that no unknown O-flag values are
set and returns an error to userspace if they are while the older open
syscalls like open() and openat() simply ignore unknown flag values:
Fix this by preventing the immediate truncation in build_open_flags().
There's a snafu here though stripping FMODE_* directly from flags would
cause the upper 32 bits to be truncated as well due to integer promotion
rules since FMODE_* is unsigned int, O_* are signed ints (yuck).
In addition, struct open_flags currently defines flags to be 32 bit
which is reasonable. If we simply were to bump it to 64 bit we would
need to change a lot of code preemptively which doesn't seem worth it.
So simply add a compile-time check verifying that all currently known
O_* flags are within the 32 bit range and fail to build if they aren't
anymore.
This change shouldn't regress old open syscalls since they silently
truncate any unknown values anyway. It is a tiny semantic change for
openat2() but it is very unlikely people pass ing > 32 bit unknown flags
and the syscall is relatively new too.
Link: https://lore.kernel.org/r/20210528092417.3942079-3-brauner@kernel.org Cc: Christoph Hellwig <hch@lst.de> Cc: Aleksa Sarai <cyphar@cyphar.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: linux-fsdevel@vger.kernel.org Reported-by: Richard Guy Briggs <rgb@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Aleksa Sarai <cyphar@cyphar.com> Reviewed-by: Richard Guy Briggs <rgb@redhat.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
include/linux/prandom.h:114:45: sparse: sparse: cast truncates bits from
constant value
This is because the 64-bit seed value is manipulated and then placed in a
u32, causing an implicit cast and truncation. A forced cast to u32 doesn't
prevent this warning, which is reasonable because a typecast doesn't prove
that truncation was expected.
Logical-AND the value with 0xffffffff to make explicit that truncation to
32-bit is intended.
Reported-by: kernel test robot <lkp@intel.com> Signed-off-by: Richard Fitzgerald <rf@opensource.cirrus.com> Reviewed-by: Petr Mladek <pmladek@suse.com> Signed-off-by: Petr Mladek <pmladek@suse.com> Link: https://lore.kernel.org/r/20210525122012.6336-3-rf@opensource.cirrus.com Signed-off-by: Sasha Levin <sashal@kernel.org>
These rx tx flags arguments are for signaling close_connection() from
which worker they are called. Obviously the receive worker cannot cancel
itself and vice versa for swork. For the othercon the receive worker
should only be used, however to avoid deadlocks we should pass the same
flags as the original close_connection() was called.
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 will change the reconnect handling that if an error occurs
if a socket error callback is occurred. This will also handle reconnects
in a non blocking connecting case which is currently missing. If error
ECONNREFUSED is reported we delay the reconnect by one second.
Signed-off-by: Alexander Aring <aahringo@redhat.com> Signed-off-by: David Teigland <teigland@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Grab rq->refcount before calling ->fn in blk_mq_tagset_busy_iter(), and
this way will prevent the request from being re-used when ->fn is
running. The approach is same as what we do during handling timeout.
Fix request use-after-free(UAF) related with completion race or queue
releasing:
- If one rq is referred before rq->q is frozen, then queue won't be
frozen before the request is released during iteration.
- If one rq is referred after rq->q is frozen, refcount_inc_not_zero()
will return false, and we won't iterate over this request.
However, still one request UAF not covered: refcount_inc_not_zero() may
read one freed request, and it will be handled in next patch.
Tested-by: John Garry <john.garry@huawei.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Ming Lei <ming.lei@redhat.com> Link: https://lore.kernel.org/r/20210511152236.763464-3-ming.lei@redhat.com Signed-off-by: Jens Axboe <axboe@kernel.dk> Signed-off-by: Sasha Levin <sashal@kernel.org>
block_dump is an old debugging interface, one of it's functions is used
to print the information about who write which file on disk. If we
enable block_dump through /proc/sys/vm/block_dump and turn on debug log
level, we can gather information about write process name, target file
name and disk from kernel message. This feature is realized in
block_dump___mark_inode_dirty(), it print above information into kernel
message directly when marking inode dirty, so it is noisy and can easily
trigger log storm. At the same time, get the dentry refcount is also not
safe, we found it will lead to deadlock on ext4 file system with
data=journal mode.
After tracepoints has been introduced into the kernel, we got a
tracepoint in __mark_inode_dirty(), which is a better replacement of
block_dump___mark_inode_dirty(). The only downside is that it only trace
the inode number and not a file name, but it probably doesn't matter
because the original printed file name in block_dump is not accurate in
some cases, and we can still find it through the inode number and device
id. So this patch delete the dirting inode part of block_dump feature.
Signed-off-by: zhangyi (F) <yi.zhang@huawei.com> Reviewed-by: Jan Kara <jack@suse.cz> Reviewed-by: Christoph Hellwig <hch@lst.de> Link: https://lore.kernel.org/r/20210313030146.2882027-2-yi.zhang@huawei.com Signed-off-by: Jens Axboe <axboe@kernel.dk> Signed-off-by: Sasha Levin <sashal@kernel.org>
More ASUS laptops have the _GPE define in the DSDT table with a
different value than the _GPE number in the ECDT.
This is causing media keys not working on ASUS X505BA/BP, X542BA/BP
Add model info to the quirks list.
Signed-off-by: Chris Chiu <chris.chiu@canonical.com> Signed-off-by: Jian-Hong Pan <jhp@endlessos.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The Bay Trail Glavey TM800A550L tablet, which ships with Android installed
from the factory, uses a GT912 touchscreen controller which needs to have
its firmware uploaded by the OS to work (this is a first for a x86 based
device with a Goodix touchscreen controller).
Add a touchscreen_dmi entry for this which specifies the filenames
to use for the firmware and config files needed for this.
Note this matches on a GDIX1001 ACPI HID, while the original DSDT uses
a HID of GODX0911. For the touchscreen to work on these devices a DSDT
override is necessary to fix a missing IRQ and broken GPIO settings in
the ACPI-resources for the touchscreen. This override also changes the
HID to the standard GDIX1001 id typically used for Goodix touchscreens.
The DSDT override is available here:
https://fedorapeople.org/~jwrdegoede/glavey-tm800a550l-dsdt-override/
Teclast X89 tablets come in 2 versions, with Windows pre-installed and with
Android pre-installed. These 2 versions have different DMI strings.
Add a match for the DMI strings used by the Android version BIOS.
Note the Android version BIOS has a bug in the DSDT where no IRQ is
provided, so for the touchscreen to work a DSDT override fixing this
is necessary as well.
Move the DMI quirks for upside-down mounted Goodix touchscreens from
drivers/input/touchscreen/goodix.c to
drivers/platform/x86/touchscreen_dmi.c,
where all the other x86 touchscreen quirks live.
Note the touchscreen_dmi.c code attaches standard touchscreen
device-properties to an i2c-client device based on a combination of a
DMI match + a device-name match. I've verified that the: Teclast X98 Pro,
WinBook TW100 and WinBook TW700 uses an ACPI devicename of "GDIX1001:00"
based on acpidumps and/or dmesg output available on the web.
The existing code attempted to handle numbers by doing a strto[u]l(),
ignoring the field width, and then repeatedly dividing to extract the
field out of the full converted value. If the string contains a run of
valid digits longer than will fit in a long or long long, this would
overflow and no amount of dividing can recover the correct value.
This patch fixes vsscanf() to obey number field widths when parsing
the number.
A new _parse_integer_limit() is added that takes a limit for the number
of characters to parse. The number field conversion in vsscanf is changed
to use this new function.
If a number starts with a radix prefix, the field width must be long
enough for at last one digit after the prefix. If not, it will be handled
like this:
sscanf("0x4", "%1i", &i): i=0, scanning continues with the 'x'
sscanf("0x4", "%2i", &i): i=0, scanning continues with the '4'
This is consistent with the observed behaviour of userland sscanf.
Note that this patch does NOT fix the problem of a single field value
overflowing the target type. So for example:
Generally, the C-state latency is provided by the _CST method or
FADT, but some OEM platforms using AMD Picasso, Renoir, Van Gogh,
and Cezanne set the C2 latency greater than C3's which causes the
C2 state to be skipped.
That will block the core entering PC6, which prevents S0ix working
properly on Linux systems.
In other operating systems, the latency values are not validated and
this does not cause problems by skipping states.
To avoid this issue on Linux, detect when latencies are not an
arithmetic progression and sort them.
These are supposedly not required for AMD platforms,
but at least some HP laptops seem to require it to
properly turn off the keyboard backlight.
Based on a patch from Marcin Bachry <hegel666@gmail.com>.
Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1230 Reviewed-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Action of unbinding driver from a device is not cancellable and should not
fail, and driver core does not pay attention to the result of "remove"
method, therefore using down_interruptible() in hid_device_remove() does
not make sense.
Force backlight control in these models to use the native interface
at /sys/class/backlight/amdgpu_bl0.
Signed-off-by: Luke D. Jones <luke@ljones.dev> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
select SND_USB_AUDIO_USE_MEDIA_CONTROLLER if MEDIA_CONTROLLER &&
(MEDIA_SUPPORT=y || MEDIA_SUPPORT=SND_USB_AUDIO)
The following config check in include/media/media-dev-allocator.h is
in place to enable the API only when CONFIG_MEDIA_CONTROLLER and
CONFIG_USB are enabled.
This check doesn't work as intended when CONFIG_USB=m. When CONFIG_USB=m,
CONFIG_USB_MODULE is defined and CONFIG_USB is not. The above config check
doesn't catch that CONFIG_USB is defined as a module and disables the API.
This results in sound/usb enabling Media Controller specific ALSA driver
code, while Media disables the Media Controller API.
Fix the problem requires two changes:
1. Change the check to use IS_ENABLED to detect when CONFIG_USB is enabled
as a module or static. Since CONFIG_MEDIA_CONTROLLER is a bool, leave
the check unchanged to be consistent with drivers/media/Makefile.
2. Change the drivers/media/mc/Makefile to include mc-dev-allocator.o
in mc-objs when CONFIG_USB is enabled.
Using register asm statements has been proven to be very error prone,
especially when using code instrumentation where gcc may add function
calls, which clobbers register contents in an unexpected way.
Therefore get rid of register asm statements in kvm code, even though
there is currently nothing wrong with them. This way we know for sure
that this bug class won't be introduced here.
In the step #3 of check_irq_usage(), we seach backwards to find a lock
whose usage conflicts the usage of @target_entry1 on safe/unsafe.
However, we should only keep the irq-unsafe usage of @target_entry1 into
consideration, because it could be a case where a lock is hardirq-unsafe
but soft-safe, and in check_irq_usage() we find it because its
hardirq-unsafe could result into a hardirq-safe-unsafe deadlock, but
currently since we don't filter out the other usage bits, so we may find
a lock dependency path softirq-unsafe -> softirq-safe, which in fact
doesn't cause a deadlock. And this may cause misleading lockdep splats.
Fix this by only keeping LOCKF_ENABLED_IRQ_ALL bits when we try the
backwards search.
Reported-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Boqun Feng <boqun.feng@gmail.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lore.kernel.org/r/20210618170110.3699115-4-boqun.feng@gmail.com Signed-off-by: Sasha Levin <sashal@kernel.org>
We use the same code to print backwards lock dependency path as the
forwards lock dependency path, and this could result into incorrect
printing because for a backwards lock_list ->trace is not the call trace
where the lock of ->class is acquired.
Fix this by introducing a separate function on printing the backwards
dependency path. Also add a few comments about the printing while we are
at it.
Reported-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Boqun Feng <boqun.feng@gmail.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lore.kernel.org/r/20210618170110.3699115-2-boqun.feng@gmail.com Signed-off-by: Sasha Levin <sashal@kernel.org>
With a config having PAGE_SIZE set to 256K, BTRFS build fails
with the following message
include/linux/compiler_types.h:326:38: error: call to
'__compiletime_assert_791' declared with attribute error:
BUILD_BUG_ON failed: (BTRFS_MAX_COMPRESSED % PAGE_SIZE) != 0
BTRFS_MAX_COMPRESSED being 128K, BTRFS cannot support platforms with
256K pages at the time being.
There are two platforms that can select 256K pages:
- hexagon
- powerpc
Disable BTRFS when 256K page size is selected. Supporting this would
require changes to the subpage mode that's currently being developed.
Given that 256K is many times larger than page sizes commonly used and
for what the algorithms and structures have been tuned, it's out of
scope and disabling build is a reasonable option.
Since @offset is 0, btrfs_invalidate() will try to invalidate the full
page, and finally call clear_page_extent_mapped() which will detach
subpage structure from the page.
And since the page no longer has subpage structure, the subpage dirty
bitmap will be cleared, preventing the dirty range from being written
back, thus no way to wake up the ordered extent.
[FIX]
Just follow other filesystems, only to invalidate the page if the range
covers the full page.
There are cases like truncate_setsize() which can call
btrfs_invalidatepage() with offset == 0 and length != 0 for the last
page of an inode.
Although the old code will still try to invalidate the full page, we are
still safe to just wait for ordered extent to finish.
So it shouldn't cause extra problems.
The type of discard_bitmap_bytes and discard_extent_bytes is u64 so the
format should be %llu, though the actual values would hardly ever
overflow to negative values.
While stress testing our error handling I noticed that sometimes we
would still commit the transaction even though we had aborted the
transaction.
Currently we track if a trans handle has dirtied any metadata, and if it
hasn't we mark the filesystem as having an error (so no new transactions
can be started), but we will allow the current transaction to complete
as we do not mark the transaction itself as having been aborted.
This sounds good in theory, but we were not properly tracking IO errors
in btrfs_finish_ordered_io, and thus committing the transaction with
bogus free space data. This isn't necessarily a problem per-se with the
free space cache, as the other guards in place would have kept us from
accepting the free space cache as valid, but highlights a real world
case where we had a bug and could have corrupted the filesystem because
of it.
This "skip abort on empty trans handle" is nice in theory, but assumes
we have perfect error handling everywhere, which we clearly do not.
Also we do not allow further transactions to be started, so all this
does is save the last transaction that was happening, which doesn't
necessarily gain us anything other than the potential for real
corruption.
Remove this particular bit of code, if we decide we need to abort the
transaction then abort the current one and keep us from doing real harm
to the file system, regardless of whether this specific trans handle
dirtied anything or not.
Signed-off-by: Josef Bacik <josef@toxicpanda.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
If we fail to update the delayed inode we need to abort the transaction,
because we could leave an inode with the improper counts or some other
such corruption behind.
Signed-off-by: Josef Bacik <josef@toxicpanda.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
If we get an error while looking up the inode item we'll simply bail
without cleaning up the delayed node. This results in this style of
warning happening on commit:
Because the iref isn't dropped and this leaves an elevated node->count,
so any release just re-queues it onto the delayed inodes list. Fix this
by going to the out label to handle the proper cleanup of the delayed
node.
Signed-off-by: Josef Bacik <josef@toxicpanda.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The POWER9 vCPU TLB management code assumes all threads in a core share
a TLB, and that TLBIEL execued by one thread will invalidate TLBs for
all threads. This is not the case for SMT8 capable POWER9 and POWER10
(big core) processors, where the TLB is split between groups of threads.
This results in TLB multi-hits, random data corruption, etc.
Fix this by introducing cpu_first_tlb_thread_sibling etc., to determine
which siblings share TLBs, and use that in the guest TLB flushing code.
[npiggin@gmail.com: add changelog and comment]
Signed-off-by: Paul Mackerras <paulus@ozlabs.org> Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/20210602040441.3984352-1-npiggin@gmail.com Signed-off-by: Sasha Levin <sashal@kernel.org>
Restoring a guest with an active virtual PMU results in no perf
counters being instanciated on the host side. Not quite what
you'd expect from a restore.
In order to fix this, force a writeback of PMCR_EL0 on the first
run of a vcpu (using a new request so that it happens once the
vcpu has been loaded). This will in turn create all the host-side
counters that were missing.
Energy Aware Scheduling (EAS) needs to be able to predict the frequency
requests made by the SchedUtil governor to properly estimate energy used
in the future. It has to take into account CPUs utilization and forecast
Performance Domain (PD) frequency. There is a corner case when the max
allowed frequency might be reduced due to thermal. SchedUtil is aware of
that reduced frequency, so it should be taken into account also in EAS
estimations.
SchedUtil, as a CPUFreq governor, knows the maximum allowed frequency of
a CPU, thanks to cpufreq_driver_resolve_freq() and internal clamping
to 'policy::max'. SchedUtil is responsible to respect that upper limit
while setting the frequency through CPUFreq drivers. This effective
frequency is stored internally in 'sugov_policy::next_freq' and EAS has
to predict that value.
In the existing code the raw value of arch_scale_cpu_capacity() is used
for clamping the returned CPU utilization from effective_cpu_util().
This patch fixes issue with too big single CPU utilization, by introducing
clamping to the allowed CPU capacity. The allowed CPU capacity is a CPU
capacity reduced by thermal pressure raw value.
Thanks to knowledge about allowed CPU capacity, we don't get too big value
for a single CPU utilization, which is then added to the util sum. The
util sum is used as a source of information for estimating whole PD energy.
To avoid wrong energy estimation in EAS (due to capped frequency), make
sure that the calculation of util sum is aware of allowed CPU capacity.
This thermal pressure might be visible in scenarios where the CPUs are not
heavily loaded, but some other component (like GPU) drastically reduced
available power budget and increased the SoC temperature. Thus, we still
use EAS for task placement and CPUs are not over-utilized.
Signed-off-by: Lukasz Luba <lukasz.luba@arm.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org> Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com> Link: https://lore.kernel.org/r/20210614191128.22735-1-lukasz.luba@arm.com Signed-off-by: Sasha Levin <sashal@kernel.org>
0: Fan control disabled. Duty cycle is fixed to 0%
1: Fan control enabled, pwm mode. Duty cycle is determined by
values written into Target Duty Cycle registers.
2: Fan control enabled, rpm mode
Duty cycle is adjusted such that fan speed matches
the values in Target Count registers
The current code does not do this; instead, it mixes pwm control
configuration with fan speed monitoring configuration. Worse, it
reports that pwm control would be disabled (pwmX_enable==0) when
it is in fact enabled in pwm mode. Part of the problem may be that
the chip sets the "TACH input enable" bit on its own whenever the
mode bit is set to RPM mode, but that doesn't mean that "TACH input
enable" accurately reflects the pwm mode.
Fix it up and only handle pwm control with the pwmX_enable attributes.
In the documentation, clarify that disabling pwm control (pwmX_enable=0)
sets the pwm duty cycle to 0%. In the code, explain why TACH_INPUT_EN
is set together with RPM_MODE.
While at it, only update the configuration register if the configuration
has changed, and only update the cached configuration if updating the
chip configuration was successful.
The MAX31790 has two sets of registers for pwm duty cycles, one to request
a duty cycle and one to read the actual current duty cycle. Both do not
have to be the same.
When reporting the pwm duty cycle to the user, the actual pwm duty cycle
from pwm duty cycle registers needs to be reported. When setting it, the
pwm target duty cycle needs to be written. Since we don't know the actual
pwm duty cycle after a target pwm duty cycle has been written, set the
valid flag to false to indicate that actual pwm duty cycle should be read
from the chip instead of using cached values.
Some BT.656 sensors (e.g. ADV718x) transmit frames with unstable BT.656
sync codes after initial power on. This confuses the imx CSI,resulting
in vertical and/or horizontal sync issues. Skip the first 20 frames
to avoid the unstable sync codes.
[fabio: fixed checkpatch warning and increased the frame skipping to 20]
Signed-off-by: Steve Longerbeam <slongerbeam@gmail.com> Signed-off-by: Fabio Estevam <festevam@gmail.com> Reviewed-by: Tim Harvey <tharvey@gateworks.com> 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>
drivers/media/dvb-core/dvbdev.c: drivers/media/dvb-core/dvbdev.c:510 dvb_register_device() warn: '&dvbdev->list_head' not removed from list
drivers/media/dvb-core/dvbdev.c: drivers/media/dvb-core/dvbdev.c:530 dvb_register_device() warn: '&dvbdev->list_head' not removed from list
drivers/media/dvb-core/dvbdev.c: drivers/media/dvb-core/dvbdev.c:545 dvb_register_device() warn: '&dvbdev->list_head' not removed from list
The error logic inside dvb_register_device() doesn't remove
devices from the dvb_adapter_list in case of errors.
The risk of especulation is actually almost-non-existing here,
as there are very few users of TCP/IP using the DVB stack,
as, this is mainly used with DVB-S/S2 cards, and only by people
that receives TCP/IP from satellite connections, which limits
a lot the number of users of such feature(*).
(*) In thesis, DVB-C cards could also benefit from it, but I'm
yet to see a hardware that supports it.
crypto_shash_alg_has_setkey() is implemented by testing whether the
.setkey() member of a struct shash_alg points to the default version,
called shash_no_setkey(). As crypto_shash_alg_has_setkey() is a static
inline, this requires shash_no_setkey() to be exported to modules.
Unfortunately, when building with CFI, function pointers are routed
via CFI stubs which are private to each module (or to the kernel proper)
and so this function pointer comparison may fail spuriously.
Let's fix this by turning crypto_shash_alg_has_setkey() into an out of
line function.
Cc: Sami Tolvanen <samitolvanen@google.com> Cc: Eric Biggers <ebiggers@kernel.org> Signed-off-by: Ard Biesheuvel <ardb@kernel.org> Reviewed-by: Eric Biggers <ebiggers@google.com> Reviewed-by: Sami Tolvanen <samitolvanen@google.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: Sasha Levin <sashal@kernel.org>
The card timing and the bus frequency are not changed atomically with
respect to calls to the set_clock() callback in the driver. The result
is the driver sees a transient state where there's a mismatch between
the two and thus the inputs to the phase correction calculation
formula are garbage.
Switch from dev_warn() to dev_dbg() to avoid noise in the normal case,
though the change does make bad configurations less likely to be
noticed.
Reported-by: Joel Stanley <joel@jms.id.au> Signed-off-by: Andrew Jeffery <andrew@aj.id.au> Reviewed-by: Joel Stanley <joel@jms.id.au> Link: https://lore.kernel.org/r/20210607013020.85885-1-andrew@aj.id.au Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
This patch fixes the following issues:
1. memstick_free_host() will free the host, so the use of ms_dev(host) after
it will be a problem. To fix this, move memstick_free_host() after when we
are done with ms_dev(host).
2. In rtsx_usb_ms_drv_remove(), pm need to be disabled before we remove
and free host otherwise memstick_check will be called and UAF will
happen.
When the driver fails to talk with the hardware with dvb_usb_generic_rw,
it will return an error to dvb_usb_adapter_frontend_init. However, the
driver forgets to free the resource (e.g., struct cinergyt2_fe_state),
which leads to a memory leak.
Fix this by freeing struct cinergyt2_fe_state when dvb_usb_generic_rw
fails in cinergyt2_frontend_attach.
GDB produces the following warning when debugging kernels built with
CONFIG_RELR:
BFD: /android0/linux-next/vmlinux: unknown type [0x13] section `.relr.dyn'
when loading a kernel built with CONFIG_RELR into GDB. It can also
prevent debugging symbols using such relocations.
Peter sugguests:
[That flag] means that lld will use dynamic tags and section type
numbers in the OS-specific range rather than the generic range. The
kernel itself doesn't care about these numbers; it determines the
location of the RELR section using symbols defined by a linker script.
When ctx_id >= HVA_MAX_INSTANCES in hva_hw_its_irq_thread() it tries to
access fields of ctx that is NULL at that point. The patch gets rid of
these accesses.
Found by Linux Driver Verification project (linuxtesting.org).
In 'bt878_irq', the driver calls 'tasklet_schedule', but this tasklet is
set in 'dvb_bt8xx_load_card' of another driver 'dvb-bt8xx'.
However, this two drivers are separate. The user may not load the
'dvb-bt8xx' driver when loading the 'bt8xx' driver, that is, the tasklet
has not been initialized when 'tasklet_schedule' is called, so it is
necessary to check whether the tasklet is initialized in 'bt878_probe'.
Fix this by adding a check at the end of bt878_probe.
projects / thirdparty / kernel / stable.git / log
