Smatch is not smart enough to see that "&stream->clock.lock" and
"&clock->lock" are the same thing so it complains about the locking
here. Let's make it more consistent.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
V4L2_CID_EXPOSURE_ABSOLUTE property does not return an updated value when
autoexposure (V4L2_CID_EXPOSURE_AUTO) is turned on. This patch fixes this
issue by adding the UVC_CTRL_FLAG_AUTO_UPDATE flag.
Tested on a C920 camera.
Signed-off-by: Anton V. Shokurov <shokurov.anton.v@yandex.ru>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
Smatch complains about where the au8293_data is placed:
drivers/media/pci/cx23885/cx23885-dvb.c:2174 dvb_register() info: 'a8293_pdata' is not actually initialized (unreached code).
It is not actually expected to have such initialization at
switch {
foo = bar;
case:
...
}
Not really sure how gcc does that, but this is something that I would
expect that different compilers would do different things.
David Howells checked with the compiler people: it's not really expected to
initialise as expected.
So, move the initialization outside the switch(), making smatch to
shut up one warning.
Acked-by: David Howells <dhowells@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
When pipeline is deleted, set the pipeline state to invalid state.
Signed-off-by: Jeeja KP <jeeja.kp@intel.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
This patch adds clean up routine to clear the stream registers and
calls this routine before setting up stream registers.
Signed-off-by: Jeeja KP <jeeja.kp@intel.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Currently the error path of function gfs2_inode_lookup calls function
gfs2_glock_put corresponding to an earlier call to gfs2_glock_get for
the inode glock. That's wrong because the error path also calls
iget_failed() which eventually calls iput, which eventually calls
gfs2_evict_inode, which does another gfs2_glock_put. This double-put
can cause the glock reference count to get off.
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Before this patch, when function try_rgrp_unlink queued a glock for
delete_work to reclaim the space, it used the inode glock to do so.
That's different from the iopen callback which uses the iopen glock
for the same purpose. We should be consistent and always use the
iopen glock. This may also save us reference counting problems with
the inode glock, since clear_glock does an extra glock_put() for the
inode glock.
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Some error cases in gfs2_create_inode were not unlocking the iopen
glock, getting the reference count off. This adds the proper unlock.
The error logic in function gfs2_create_inode was also convoluted,
so this patch simplifies it. It also takes care of a bug in
which gfs2_qa_delete() was not called in an error case.
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
In function gfs2_delete_inode() we write and flush the mapping for
a glock, among other things. We truncate the mapping for the inode,
but we never truncate the mapping for the glock. This patch makes it
also truncate the metamapping. This avoid cases where the glock is
reused by another process who is trying to recreate an inode in its
place using the same block.
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Acked-by: Steven Whitehouse <swhiteho@redhat.com>
This patch changes every glock_dq for iopen glocks into a dq_wait.
This makes sure that iopen glocks do not outlive the inode itself.
In turn, that ensures that anyone trying to unlink the glock will
be able to find the inode when it receives a remote iopen callback.
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Acked-by: Steven Whitehouse <swhiteho@redhat.com>
Add qspi memory mapped region entries for AM43xx based SoCs. Also,
update the binding documents for the controller to document this change.
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Vignesh R <vigneshr@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
Add qspi memory mapped region entries for DRA7xx based SoCs. Also,
update the binding documents for the controller to document this change.
Signed-off-by: Vignesh R <vigneshr@ti.com>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Tony Lindgren <tony@atomide.com>
When allocating a pciback device fails, clear the private
field. This could lead to an use-after free, however
the 'really_probe' takes care of setting
dev_set_drvdata(dev, NULL) in its failure path (which we would
exercise if the ->probe function failed), so we we
are OK. However lets be defensive as the code can change.
Going forward we should clean up the pci_set_drvdata(dev, NULL)
in the various code-base. That will be for another day.
Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Reported-by: Jonathan Creekmore <jonathan.creekmore@gmail.com>
Signed-off-by: Doug Goldstein <cardoe@cardoe.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
The sensor_template_group structures are never modified, so declare them as
const.
Done with the help of Coccinelle.
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Andreas Herrmann won't take the maintainer of fam15h_power driver. I
will take it and appreciate him for the great contributions on this
driver.
Signed-off-by: Huang Rui <ray.huang@amd.com>
Acked-by: Andreas Herrmann <herrmann.der.user@googlemail.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Aravind Gopalakrishnan <Aravind.Gopalakrishnan@amd.com>
[groeck: Fixed typo in CREDITS]
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
AMD Family 15h Models 70h-7fh processors also support TDP power
reporting interface.
Signed-off-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
LTC3815 is a Monolithic Synchronous DC/DC Step-Down Converter.
Cc: Michael Jones <mike@proclivis.com>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
HTU21 is now supported by IIO, and can be instantiated as hwmon driver
using the iio-hwmon bridge. An explicit hwmon driver is no longer needed.
Cc: William Markezana <william.markezana@meas-spec.com>
Cc: Ludovic Tancerel <ludovic.tancerel@maplehightech.com>
Cc: Jonathan Cameron <jic23@kernel.org>
Reviewed-by: Jean Delvare <jdelvare@suse.de>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
If CONFIG_BITREVERSE is not built-in, the sht15 driver fails to link:
drivers/built-in.o: In function `sht15_crc8':
drivers/hwmon/sht15.c:195: undefined reference to `byte_rev_table'
This adds a Kconfig 'select' statement, like all other users of
bitrev.h have it.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: 33836ee985 ("hwmon:change sht15_reverse()")
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
As reported by smatch:
drivers/media/pci/cx23885/cx23885-dvb.c:2174 dvb_register() info: 'a8293_pdata' is not actually initialized (unreached code).
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
As reported by smatch:
drivers/media/v4l2-core/videobuf2-core.c:2415 __vb2_init_fileio() warn: possible memory leak of 'fileio'
While here, avoid the usage of sizeof(struct foo_struct).
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
Limit busywaiting only to the request currently being processed by the
GPU. If the request is not currently being processed by the GPU, there
is a very low likelihood of it being completed within the 2 microsecond
spin timeout and so we will just be wasting CPU cycles.
v2: Check for logical inversion when rebasing - we were incorrectly
checking for this request being active, and instead busywaiting for
when the GPU was not yet processing the request of interest.
v3: Try another colour for the seqno names.
v4: Another colour for the function names.
v5: Remove the forced coherency when checking for the active request. On
reflection and plenty of recent experimentation, the issue is not a
cache coherency problem - but an irq/seqno ordering problem (timing issue).
Here, we do not need the w/a to force ordering of the read with an
interrupt.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: "Rogozhkin, Dmitry V" <dmitry.v.rogozhkin@intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: Eero Tamminen <eero.t.tamminen@intel.com>
Cc: "Rantala, Valtteri" <valtteri.rantala@intel.com>
Cc: stable@vger.kernel.org
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/1449833608-22125-4-git-send-email-chris@chris-wilson.co.uk
When waiting for high frequency requests, the finite amount of time
required to set up the irq and wait upon it limits the response rate. By
busywaiting on the request completion for a short while we can service
the high frequency waits as quick as possible. However, if it is a slow
request, we want to sleep as quickly as possible. The tradeoff between
waiting and sleeping is roughly the time it takes to sleep on a request,
on the order of a microsecond. Based on measurements of synchronous
workloads from across big core and little atom, I have set the limit for
busywaiting as 10 microseconds. In most of the synchronous cases, we can
reduce the limit down to as little as 2 miscroseconds, but that leaves
quite a few test cases regressing by factors of 3 and more.
The code currently uses the jiffie clock, but that is far too coarse (on
the order of 10 milliseconds) and results in poor interactivity as the
CPU ends up being hogged by slow requests. To get microsecond resolution
we need to use a high resolution timer. The cheapest of which is polling
local_clock(), but that is only valid on the same CPU. If we switch CPUs
because the task was preempted, we can also use that as an indicator that
the system is too busy to waste cycles on spinning and we should sleep
instead.
__i915_spin_request was introduced in
commit 2def4ad99b [v4.2]
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date: Tue Apr 7 16:20:41 2015 +0100
drm/i915: Optimistically spin for the request completion
v2: Drop full u64 for unsigned long - the timer is 32bit wraparound safe,
so we can use native register sizes on smaller architectures. Mention
the approximate microseconds units for elapsed time and add some extra
comments describing the reason for busywaiting.
v3: Raise the limit to 10us
v4: Now 5us.
Reported-by: Jens Axboe <axboe@kernel.dk>
Link: https://lkml.org/lkml/2015/11/12/621
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: "Rogozhkin, Dmitry V" <dmitry.v.rogozhkin@intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: Eero Tamminen <eero.t.tamminen@intel.com>
Cc: "Rantala, Valtteri" <valtteri.rantala@intel.com>
Cc: stable@vger.kernel.org
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/1449833608-22125-3-git-send-email-chris@chris-wilson.co.uk
The busywait in __i915_spin_request() does not respect pending signals
and so may consume the entire timeslice for the task instead of
returning to userspace to handle the signal.
In the worst case this could cause a delay in signal processing of 20ms,
which would be a noticeable jitter in cursor tracking. If a higher
resolution signal was being used, for example to provide fairness of a
server timeslices between clients, we could expect to detect some
unfairness between clients (i.e. some windows not updating as fast as
others). This issue was noticed when inspecting a report of poor
interactivity resulting from excessively high __i915_spin_request usage.
Fixes regression from
commit 2def4ad99b [v4.2]
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date: Tue Apr 7 16:20:41 2015 +0100
drm/i915: Optimistically spin for the request completion
v2: Try to assess the impact of the bug
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc; "Rogozhkin, Dmitry V" <dmitry.v.rogozhkin@intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: Eero Tamminen <eero.t.tamminen@intel.com>
Cc: "Rantala, Valtteri" <valtteri.rantala@intel.com>
Cc: stable@vger.kernel.org
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/1449833608-22125-2-git-send-email-chris@chris-wilson.co.uk
The realview multiplatform series has a trivial conflict with
one of the treewide cleanups, let's just merge that in to
avoid having to resolve this later.
* treewide/cleanup:
ARM: use "depends on" for SoC configs instead of "if" after prompt
ARM/clocksource: use automatic DT probing for ux500 PRCMU
ARM: use const and __initconst for smp_operations
ARM: hisi: do not export smp_operations structures
Conflicts:
arch/arm/mach-integrator/Kconfig
The handling of q->plane_sizes was wrong in vb2_core_create_bufs().
The q->plane_sizes array was global and it was overwritten by create_bufs.
So if reqbufs was called with e.g. size 100000 then q->plane_sizes[0] would
be set to 100000. If create_bufs was called afterwards with size 200000,
then q->plane_sizes[0] would be overwritten with the new value. Calling
create_bufs again for size 100000 would cause an error since 100000 is now
less than q->plane_sizes[0].
This patch fixes this problem by 1) removing q->plane_sizes and using the
vb->planes[].length field instead, and 2) by introducing a min_length field
in struct vb2_plane. This field is set to the plane size as returned by
the queue_setup op and is the minimum required plane size. So user pointers
or dmabufs should all be at least this size.
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Reported-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
Ensure that the offsets are correct before buf_init() is called.
As a consequence the __setup_offsets() function now sets up the
offsets for the given buffer instead of for all new buffers.
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
The platsmp-dt.c file does not build when CONFIG_SMP is disabled:
platsmp-dt.c:84:2: error: unknown field 'smp_prepare_cpus' specified in initializer
This changes the Makefile to build it conditionally on CONFIG_SMP.
The legacy platsmp.c file is only used for ATAGS based builds, so
we can move it into the CONFIG_ATAGS conditional.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fill in q->bufs[vb->index] before the call to buf_init: it makes
sense that this is initialized correctly.
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
Rather than setting up the lengths at the end, set them up when
the vb2_buffer is allocated. This also ensures that buf_init()
sees the right length values.
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
This ops can never fail, so make these void functions.
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
This module is no longer V4L2 specific, so update the module description
accordingly.
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
After several open/close sai test with ctrl+c, there will be
I/O error. The SAI can't work anymore, can't recover. There
will be no frame clock. With adding the software reset in
trigger stop, the issue can be fixed.
This is a hardware bug/errata and reset is the only option.
According to the reference manual, the software reset doesn't
reset any control register but only internal hardware logics
such as bit clock generator, status flags, and FIFO pointers.
(Our purpose is just to reset the clock generator while the
software reset is the only way to do that.)
Since slave mode doesn't use the clock generator, only apply
the reset procedure to the master mode.
For asynchronous mode, TX will not be reset when RX is still
running. In this case, i can't reproduce this issue.
Signed-off-by: Zidan Wang <zidan.wang@freescale.com>
Acked-by: Nicolin Chen <nicoleotsuka@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Move things related with vb2 file I/O and vb2_thread without doing any
functional changes. After that, videobuf2-internal.h is removed because
it is not necessary any more.
Signed-off-by: Junghak Sung <jh1009.sung@samsung.com>
Signed-off-by: Geunyoung Kim <nenggun.kim@samsung.com>
Acked-by: Seung-Woo Kim <sw0312.kim@samsung.com>
Acked-by: Inki Dae <inki.dae@samsung.com>
Acked-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Hans Verkuil <hansverk@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
Replace v4l2-stuffs with common things in struct vb2_fileio_data and
vb2_thread().
Signed-off-by: Junghak Sung <jh1009.sung@samsung.com>
Signed-off-by: Geunyoung Kim <nenggun.kim@samsung.com>
Acked-by: Seung-Woo Kim <sw0312.kim@samsung.com>
Acked-by: Inki Dae <inki.dae@samsung.com>
Acked-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Hans Verkuil <hansverk@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
The location in which last_buffer_queued is set is moved to fill_v4l2_buffer().
So, __vb2_perform_fileio() can use vb2_core_dqbuf() instead of
vb2_internal_dqbuf().
Signed-off-by: Junghak Sung <jh1009.sung@samsung.com>
Signed-off-by: Geunyoung Kim <nenggun.kim@samsung.com>
Acked-by: Seung-Woo Kim <sw0312.kim@samsung.com>
Acked-by: Inki Dae <inki.dae@samsung.com>
Acked-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Hans Verkuil <hansverk@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
Add copy_timestamp to struct vb2_queue as a flag set if vb2-core should
copy timestamps.
Signed-off-by: Junghak Sung <jh1009.sung@samsung.com>
Signed-off-by: Geunyoung Kim <nenggun.kim@samsung.com>
Acked-by: Seung-Woo Kim <sw0312.kim@samsung.com>
Acked-by: Inki Dae <inki.dae@samsung.com>
Signed-off-by: Hans Verkuil <hansverk@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
Now that we have multiplatform support, let the RealView
defconfigs select all the RealView boards so we boot out of
the box like before. This updates both defconfigs.
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Move timestamp from struct vb2_v4l2_buffer to struct vb2_buffer
for common use, and change its type to u64 in order to handling
y2038 problem. This patch also includes all device drivers' changes related to
this restructuring.
Signed-off-by: Junghak Sung <jh1009.sung@samsung.com>
Signed-off-by: Geunyoung Kim <nenggun.kim@samsung.com>
Acked-by: Seung-Woo Kim <sw0312.kim@samsung.com>
Acked-by: Inki Dae <inki.dae@samsung.com>
Signed-off-by: Hans Verkuil <hansverk@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
The timestamp of a v4l2_buffer was advertised as being CLOCK_MONOTONIC,
but instead a timestamp from a header field was used. This is inconsistent
and not what applications expect. Use v4l2_get_timestamp to properly
set the timestamp.
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
During the Seoul media workshop we decided to relax the VIDIOC_CREATE_BUFS
specification so it would no longer require drivers to validate the format
field since almost no driver did that anyway.
Instead drivers use the buffer size(s) based on the format type and the
corresponding format fields and will ignore any other fields. If the size
cannot be used an error is returned, otherwise the size is used as-is.
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
commit f598282f51 ("PCI: Fix the NIU MSI-X problem in a better way")
teaches us that dealing with MSI-X can be troublesome.
Further checks in the MSI-X architecture shows that if the
PCI_COMMAND_MEMORY bit is turned of in the PCI_COMMAND we
may not be able to access the BAR (since they are memory regions).
Since the MSI-X tables are located in there.. that can lead
to us causing PCIe errors. Inhibit us performing any
operation on the MSI-X unless the MEMORY bit is set.
Note that Xen hypervisor with:
"x86/MSI-X: access MSI-X table only after having enabled MSI-X"
will return:
xen_pciback: 0000:0a:00.1: error -6 enabling MSI-X for guest 3!
When the generic MSI code tries to setup the PIRQ without
MEMORY bit set. Which means with later versions of Xen
(4.6) this patch is not neccessary.
This is part of XSA-157
CC: stable@vger.kernel.org
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Otherwise just continue on, returning the same values as
previously (return of 0, and op->result has the PIRQ value).
This does not change the behavior of XEN_PCI_OP_disable_msi[|x].
The pci_disable_msi or pci_disable_msix have the checks for
msi_enabled or msix_enabled so they will error out immediately.
However the guest can still call these operations and cause
us to disable the 'ack_intr'. That means the backend IRQ handler
for the legacy interrupt will not respond to interrupts anymore.
This will lead to (if the device is causing an interrupt storm)
for the Linux generic code to disable the interrupt line.
Naturally this will only happen if the device in question
is plugged in on the motherboard on shared level interrupt GSI.
This is part of XSA-157
CC: stable@vger.kernel.org
Reviewed-by: David Vrabel <david.vrabel@citrix.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Otherwise an guest can subvert the generic MSI code to trigger
an BUG_ON condition during MSI interrupt freeing:
for (i = 0; i < entry->nvec_used; i++)
BUG_ON(irq_has_action(entry->irq + i));
Xen PCI backed installs an IRQ handler (request_irq) for
the dev->irq whenever the guest writes PCI_COMMAND_MEMORY
(or PCI_COMMAND_IO) to the PCI_COMMAND register. This is
done in case the device has legacy interrupts the GSI line
is shared by the backend devices.
To subvert the backend the guest needs to make the backend
to change the dev->irq from the GSI to the MSI interrupt line,
make the backend allocate an interrupt handler, and then command
the backend to free the MSI interrupt and hit the BUG_ON.
Since the backend only calls 'request_irq' when the guest
writes to the PCI_COMMAND register the guest needs to call
XEN_PCI_OP_enable_msi before any other operation. This will
cause the generic MSI code to setup an MSI entry and
populate dev->irq with the new PIRQ value.
Then the guest can write to PCI_COMMAND PCI_COMMAND_MEMORY
and cause the backend to setup an IRQ handler for dev->irq
(which instead of the GSI value has the MSI pirq). See
'xen_pcibk_control_isr'.
Then the guest disables the MSI: XEN_PCI_OP_disable_msi
which ends up triggering the BUG_ON condition in 'free_msi_irqs'
as there is an IRQ handler for the entry->irq (dev->irq).
Note that this cannot be done using MSI-X as the generic
code does not over-write dev->irq with the MSI-X PIRQ values.
The patch inhibits setting up the IRQ handler if MSI or
MSI-X (for symmetry reasons) code had been called successfully.
P.S.
Xen PCIBack when it sets up the device for the guest consumption
ends up writting 0 to the PCI_COMMAND (see xen_pcibk_reset_device).
XSA-120 addendum patch removed that - however when upstreaming said
addendum we found that it caused issues with qemu upstream. That
has now been fixed in qemu upstream.
This is part of XSA-157
CC: stable@vger.kernel.org
Reviewed-by: David Vrabel <david.vrabel@citrix.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
The guest sequence of:
a) XEN_PCI_OP_enable_msix
b) XEN_PCI_OP_enable_msix
results in hitting an NULL pointer due to using freed pointers.
The device passed in the guest MUST have MSI-X capability.
The a) constructs and SysFS representation of MSI and MSI groups.
The b) adds a second set of them but adding in to SysFS fails (duplicate entry).
'populate_msi_sysfs' frees the newly allocated msi_irq_groups (note that
in a) pdev->msi_irq_groups is still set) and also free's ALL of the
MSI-X entries of the device (the ones allocated in step a) and b)).
The unwind code: 'free_msi_irqs' deletes all the entries and tries to
delete the pdev->msi_irq_groups (which hasn't been set to NULL).
However the pointers in the SysFS are already freed and we hit an
NULL pointer further on when 'strlen' is attempted on a freed pointer.
The patch adds a simple check in the XEN_PCI_OP_enable_msix to guard
against that. The check for msi_enabled is not stricly neccessary.
This is part of XSA-157
CC: stable@vger.kernel.org
Reviewed-by: David Vrabel <david.vrabel@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
The queue_setup callback has a void pointer that is just for V4L2
and is the pointer to the v4l2_format struct that was passed to
VIDIOC_CREATE_BUFS. The idea was that drivers would use the information
from that struct to buffers suitable for the requested format.
After the vb2 split series this pointer is now a void pointer,
which is ugly, and the reality is that all existing drivers will
effectively just look at the sizeimage field of v4l2_format.
To make this more generic the queue_setup callback is changed:
the void pointer is dropped, instead if the *num_planes argument
is 0, then use the current format size, if it is non-zero, then
it contains the number of requested planes and the sizes array
contains the requested sizes. If either is unsupported, then return
-EINVAL, otherwise use the requested size(s).
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
