Some drivers also need a control like
V4L2_CID_MPEG_MFC51_VIDEO_FORCE_FRAME_TYPE to force an encoder
key frame. Add a general V4L2_CID_MPEG_VIDEO_FORCE_KEY_FRAME
so the new drivers and applications can use it.
Signed-off-by: Wu-Cheng Li <wuchengli@chromium.org>
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
(cherry picked from commit 0c8485ca3f2aaf7842d45ba24c667a9492c9900f)
Signed-off-by: Daniel Kurtz <djkurtz@chromium.org>
BUG=chromium:572825
TEST=Build and boot oak-rev5 to UI
TEST=emerge-smaug chromeos-kernel-3_18
Reviewed-on: https://chromium-review.googlesource.com/328870
Commit-Ready: Daniel Kurtz <djkurtz@chromium.org>
Tested-by: Daniel Kurtz <djkurtz@chromium.org>
Reviewed-by: Daniel Kurtz <djkurtz@chromium.org>
Change-Id: I45de3048d41edbe443b3d202c17e79f2d448213b
Signed-off-by: Jeffy Chen <jeffy.chen@rock-chips.com>
Signed-off-by: Yakir Yang <ykk@rock-chips.com>
DMA allocations might be subject to certain reqiurements specific to the
hardware using the buffers, such as availability of kernel mapping (for
contents fix-ups in the driver). The only entity that knows them is the
driver, so it must share this knowledge with vb2-dc.
This patch extends the alloc_ctx initialization interface to let the
driver specify DMA attrs, which are then stored inside the allocation
context and will be used for all allocations with that context.
As a side effect, all dma_*_coherent() calls are turned into
dma_*_attrs() calls, because the attributes need to be carried over
through all DMA operations.
BUG=chrome-os-partner:38873
TEST=compile
Signed-off-by: Tomasz Figa <tfiga@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/265363
Reviewed-by: Pawel Osciak <posciak@chromium.org>
Change-Id: I0e6c040cf820c194b6ca6f3e6355217496bd1532
Signed-off-by: Jeffy Chen <jeffy.chen@rock-chips.com>
Signed-off-by: Yakir Yang <ykk@rock-chips.com>
When queuing buffers allow for passing the configuration store ID that
should be associated with this buffer. Use the 'reserved2' field for this.
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
BUG=chrome-os-partner:33728
TEST=build
Signed-off-by: Pawel Osciak <posciak@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/232583
Trybot-Ready: Tomasz Figa <tfiga@chromium.org>
Tested-by: Tomasz Figa <tfiga@chromium.org>
Reviewed-by: Wu-cheng Li <wuchengli@chromium.org>
Commit-Queue: Tomasz Figa <tfiga@chromium.org>
Conflicts:
drivers/media/v4l2-core/videobuf2-core.c
[rebase44(groeck): fixed conflicts; structural changes to match v4.4]
Signed-off-by: Guenter Roeck <groeck@chromium.org>
Conflicts:
drivers/media/v4l2-core/v4l2-compat-ioctl32.c
Change-Id: Ibb823e9369bec79645e09651b0dda006ed53ecc5
Signed-off-by: Jeffy Chen <jeffy.chen@rock-chips.com>
Signed-off-by: Yakir Yang <ykk@rock-chips.com>
The ctrl_class is fairly pointless when used with drivers that use the control
framework: you can just fill in 0 and it will just work fine. There are still
some old unconverted drivers that do not support 0 and instead want the control
class there. The idea being that all controls in the list all belong to that
class. This was done to simplify drivers in the absence of the control framework.
When using the control framework the framework itself is smart enough to allow
controls of any class to be included in the control list.
Since configuration store IDs are in the range 1..255 (or so, in any case a relatively
small non-zero positive integer) it makes sense to effectively rename ctrl_class
to config_store. Set it to 0 and you get the normal behavior (you change the current
control value), set it to a configuration store ID and you get/set the control for
that store.
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
BUG=chrome-os-partner:33728
TEST=build
Signed-off-by: Pawel Osciak <posciak@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/232582
Trybot-Ready: Tomasz Figa <tfiga@chromium.org>
Tested-by: Tomasz Figa <tfiga@chromium.org>
Reviewed-by: Wu-cheng Li <wuchengli@chromium.org>
Commit-Queue: Tomasz Figa <tfiga@chromium.org>
Conflicts:
include/uapi/linux/videodev2.h
Change-Id: I862bb5796e27bcbbd055e22202ac9a1ed0cc6f7d
Signed-off-by: Jeffy Chen <jeffy.chen@rock-chips.com>
Signed-off-by: Yakir Yang <ykk@rock-chips.com>
Set the newly added id for i2s_src, so that they can be called
in other parts.
Change-Id: Ie4ecc4d19e3ae64a07d1f2a80aa08d40f38d09ad
Signed-off-by: Finley Xiao <finley.xiao@rock-chips.com>
This misc update would try to fix below comments from my eDP thread[0],
and lucky to say this version is stable, and i'm start to perpare the
pull request to David with this version. So i guess it's time to create
a misc FIXUP patch to address the comments.
[0]: https://patchwork.kernel.org/patch/9175613/
- Correct the misspell of "marcos" in commit message (Dominik, reviewed at Google Gerrit)
- Write a kerneldoc-style comment explaining the chips data fields (Tomasz, reviewed at Google Gerrit)
- Drop the '.lcdcsel_mask' number in chips data field (Tomasz, reviewed at Google Gerrit)
- Make this hack code more clear (Tomasz, reviewed at Google Gerrit)
reg = ~reg & REF_CLK_MASK; ---> reg ^= REF_CLK_MASK;
- Give the "rk3399-edp" a separate line for clarity in document (Tomasz, reviewed at Google Gerrit)
- Move 'output_type' setting before the return statement (Tomasz, reviewed at Google Gerrit)
- Avoid to change any internal driver state in .mode_valid interface. (Tomasz, reviewed at Google Gerrit)
- Hook the connector's color_formats in .get_modes directly. (Tomasz, reviewed at Google Gerrit)
Change-Id: Ic35f166ebac04e417ff3d135e7bf4573bbca2004
Signed-off-by: Yakir Yang <ykk@rock-chips.com>
This patch adds a helper to parse the encoder endpoint connected to the
encoder's crtc and two helpers to return its id and port id.
This can be used to determine input mux setting from endpoint or port ids.
Change-Id: I48eb7c66edb951af40085e4e388afbd5d4b2c77b
Suggested-by: Daniel Kurtz <djkurtz@chromium.org>
Reviewed-by: Daniel Kurtz <djkurtz@chromium.org>
Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
(cherry pick from commit 4cacf91fcb)
commit d7591f0c41 upstream.
The three variants use same copy&pasted code, condense this into a
helper and use that.
Make sure info.name is 0-terminated.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit ce683e5f9d upstream.
We're currently asserting that targetoff + targetsize <= nextoff.
Extend it to also check that targetoff is >= sizeof(xt_entry).
Since this is generic code, add an argument pointing to the start of the
match/target, we can then derive the base structure size from the delta.
We also need the e->elems pointer in a followup change to validate matches.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit fc1221b3a1 upstream.
32bit rulesets have different layout and alignment requirements, so once
more integrity checks get added to xt_check_entry_offsets it will reject
well-formed 32bit rulesets.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 7d35812c32 upstream.
Currently arp/ip and ip6tables each implement a short helper to check that
the target offset is large enough to hold one xt_entry_target struct and
that t->u.target_size fits within the current rule.
Unfortunately these checks are not sufficient.
To avoid adding new tests to all of ip/ip6/arptables move the current
checks into a helper, then extend this helper in followup patches.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit dd5f1b049d upstream.
The INTID mask is wrong, and is made a signed value, which has
nteresting effects in the KVM emulation. Let's sanitize it.
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 7e059158d5 ]
Prior to 4.3, openvswitch tunnel vports (vxlan, gre and geneve) could
transmit vxlan packets of any size, constrained only by the ability to
send out the resulting packets. 4.3 introduced netdevs corresponding
to tunnel vports. These netdevs have an MTU, which limits the size of
a packet that can be successfully encapsulated. The default MTU
values are low (1500 or less), which is awkwardly small in the context
of physical networks supporting jumbo frames, and leads to a
conspicuous change in behaviour for userspace.
Instead, set the MTU on openvswitch-created netdevs to be the relevant
maximum (i.e. the maximum IP packet size minus any relevant overhead),
effectively restoring the behaviour prior to 4.3.
Signed-off-by: David Wragg <david@weave.works>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit da4ed55165 ]
The problem is that fib_info->nh is [0] so the struct fib_info
allocation size depends on number of nexthops. If we just copy fib_info,
we do not copy the nexthops info and driver accesses memory which is not
ours.
Given the fact that fib4 does not defer operations and therefore it does
not need copy, just pass the pointer down to drivers as it was done
before.
Fixes: 850d0cbc91 ("switchdev: remove pointers from switchdev objects")
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The hdmi-codec is a platform device driver to be registered from
drivers of external HDMI encoders with I2S and/or spdif interface. The
driver in turn registers an ASoC codec for the HDMI encoder's audio
functionality.
The structures and definitions in the API header are mostly redundant
copies of similar structures in ASoC headers. This is on purpose to
avoid direct dependencies to ASoC structures in video side driver.
Signed-off-by: Jyri Sarha <jsarha@ti.com>
Acked-by: Arnaud Pouliquen <arnaud.pouliquen@st.com>
Acked-by: PC Liao <pc.liao@mediatek.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
(cherry picked from commit 09184118a8)
Change-Id: I4fc0651b732c2604df58cb2e0ec5f5edeecdf412
Signed-off-by: Chris Zhong <zyw@rock-chips.com>
Add IEC958 channel status helper that gets the audio properties from
snd_pcm_hw_params instead of snd_pcm_runtime. This is needed to
produce the channel status bits already in audio stream configuration
phase.
Signed-off-by: Jyri Sarha <jsarha@ti.com>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Mark Brown <broonie@kernel.org>
(cherry picked from commit 4a4436573a)
Change-Id: Ie19500cd63fb311ec273035c336acc8c568d84db
Signed-off-by: Chris Zhong <zyw@rock-chips.com>
This patch exports related MAC clocks for dts reference.
Signed-off-by: Xing Zheng <zhengxing@rock-chips.com>
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
(cherry picked from git://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip.git
v4.8-shared/clkids commit 9ff59360b8)
Change-Id: Ib6f5f2a0ccd19a8b71c384abddacadbd4da291bb
Signed-off-by: Xing Zheng <zhengxing@rock-chips.com>
There are four clocks that vop module would need to operate:
DCLK_VOP, HCLK_VOP, SCLK_VOP, ACLK_VOP,
Signed-off-by: Yakir Yang <ykk@rock-chips.com>
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
(cherry picked from git://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip.git
v4.8-shared/clkids commit 31b1fed36e)
Change-Id: Iea57109b85928c4139283e366ce5d60430e8984b
Signed-off-by: Xing Zheng <zhengxing@rock-chips.com>
This patch adds 'SCLK_TSADC' and 'PCLK_TSADC' id found on rk3228 SoCs.
That will be needed by TSADC controller.
Signed-off-by: Caesar Wang <wxt@rock-chips.com>
Acked-by: Michael Turquette <mturquette@baylibre.com>
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
(cherry picked from git://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip.git
v4.8-clk/next commit 3629e70b8c)
Change-Id: Ia4c926d4decc9affd63510a794d7f2553f6be3a5
Signed-off-by: Xing Zheng <zhengxing@rock-chips.com>
Add the dt-bindings header for the rk3228, that gets shared between
the clock controller and the clock references in the dts.
Signed-off-by: Jeffy Chen <jeffy.chen@rock-chips.com>
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
(cherry picked from git://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip.git
v4.8-shared/clkids commit abf1296599)
Change-Id: Id164aa064046276f22dc763c746e9e85c344cd81
Signed-off-by: Xing Zheng <zhengxing@rock-chips.com>
Some information, like driver specific configuration, is found
in the perf event structure. As such pass a 'struct perf_event'
to function setup_aux() rather than just the CPU number so that
individual drivers can make the right configuration when setting
up a session.
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
It is entirely possible that some PMUs need specific configuration
that is currently not found in the perf options before a session
can be setup.
It is the case for the CoreSight PMU where a sink needs to be
provided. That sink doesn't fall in any of the current perf
options.
As such this patch adds the capability to receive driver
specific configuration using the existing ioctl() mechanism.
Once the configuration has been pushed down the kernel PMU
callbacks are used to deal with the information sent from user
space.
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
The DMAengine API has a long standing race condition that is inherent to
the API itself. Calling dmaengine_terminate_all() is supposed to stop and
abort any pending or active transfers that have previously been submitted.
Unfortunately it is possible that this operation races against a currently
running (or with some drivers also scheduled) completion callback.
Since the API allows dmaengine_terminate_all() to be called from atomic
context as well as from within a completion callback it is not possible to
synchronize to the execution of the completion callback from within
dmaengine_terminate_all() itself.
This means that a user of the DMAengine API does not know when it is safe
to free resources used in the completion callback, which can result in a
use-after-free race condition.
This patch addresses the issue by introducing an explicit synchronization
primitive to the DMAengine API called dmaengine_synchronize().
The existing dmaengine_terminate_all() is deprecated in favor of
dmaengine_terminate_sync() and dmaengine_terminate_async(). The former
aborts all pending and active transfers and synchronizes to the current
context, meaning it will wait until all running completion callbacks have
finished. This means it is only possible to call this function from
non-atomic context. The later function does not synchronize, but can still
be used in atomic context or from within a complete callback. It has to be
followed up by dmaengine_synchronize() before a client can free the
resources used in a completion callback.
In addition to this the semantics of the device_terminate_all() callback
are slightly relaxed by this patch. It is now OK for a driver to only
schedule the termination of the active transfer, but does not necessarily
have to wait until the DMA controller has completely stopped. The driver
must ensure though that the controller has stopped and no longer accesses
any memory when the device_synchronize() callback returns.
This was in part done since most drivers do not pay attention to this
anyway at the moment and to emphasize that this needs to be done when the
device_synchronize() callback is implemented. But it also helps with
implementing support for devices where stopping the controller can require
operations that may sleep.
Change-Id: Ica0822ecbe803ec9605787e30751dfb098bdbe80
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Signed-off-by: Caesar Wang <wxt@rock-chips.com>
(cherry picked from git.kernel.org next/linux-next.git master
commit b36f09c3c4)
When kernel.perf_event_open is set to 3 (or greater), disallow all
access to performance events by users without CAP_SYS_ADMIN.
Add a Kconfig symbol CONFIG_SECURITY_PERF_EVENTS_RESTRICT that
makes this value the default.
This is based on a similar feature in grsecurity
(CONFIG_GRKERNSEC_PERF_HARDEN). This version doesn't include making
the variable read-only. It also allows enabling further restriction
at run-time regardless of whether the default is changed.
https://lkml.org/lkml/2016/1/11/587
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Bug: 29054680
Change-Id: Iff5bff4fc1042e85866df9faa01bce8d04335ab8
* lsk-v4.4-eas-v5.2:
DEBUG: schedtune: add tracepoint for schedtune_tasks_update() values
DEBUG: schedtune: add tracepoint for CPU boost signal
DEBUG: schedtune: add tracepoint for SchedTune configuration update
DEBUG: sched: add energy procfs interface
DEBUG: sched,cpufreq: add cpu_capacity change tracepoint
DEBUG: sched: add tracepoint for CPU load/util signals
DEBUG: sched: add tracepoint for task load/util signals
DEBUG: sched: add tracepoint for cpu/freq scale invariance
sched/fair: filter energy_diff() based on energy_payoff value
sched/tune: add support to compute normalized energy
sched/fair: keep track of energy/capacity variations
sched/fair: add boosted task utilization
sched/{fair,tune}: track RUNNABLE tasks impact on per CPU boost value
sched/tune: compute and keep track of per CPU boost value
sched/tune: add initial support for CGroups based boosting
sched/fair: add boosted CPU usage
sched/fair: add function to convert boost value into "margin"
sched/tune: add sysctl interface to define a boost value
sched/tune: add detailed documentation
fixup! sched/fair: jump to max OPP when crossing UP threshold
fixup! sched: scheduler-driven cpu frequency selection
sched: rt scheduler sets capacity requirement
sched: deadline: use deadline bandwidth in scale_rt_capacity
sched: remove call of sched_avg_update from sched_rt_avg_update
sched/cpufreq_sched: add trace events
sched/fair: jump to max OPP when crossing UP threshold
sched/fair: cpufreq_sched triggers for load balancing
sched/{core,fair}: trigger OPP change request on fork()
sched/fair: add triggers for OPP change requests
sched: scheduler-driven cpu frequency selection
cpufreq: introduce cpufreq_driver_is_slow
sched: Consider misfit tasks when load-balancing
sched: Add group_misfit_task load-balance type
sched: Add per-cpu max capacity to sched_group_capacity
sched: Do eas idle balance regardless of the rq avg idle value
arm64: Enable max freq invariant scheduler load-tracking and capacity support
arm: Enable max freq invariant scheduler load-tracking and capacity support
sched: Update max cpu capacity in case of max frequency constraints
cpufreq: Max freq invariant scheduler load-tracking and cpu capacity support
arm64, topology: Updates to use DT bindings for EAS costing data
sched: Support for extracting EAS energy costs from DT
Documentation: DT bindings for energy model cost data required by EAS
sched: Disable energy-unfriendly nohz kicks
sched: Consider a not over-utilized energy-aware system as balanced
sched: Energy-aware wake-up task placement
sched: Determine the current sched_group idle-state
sched, cpuidle: Track cpuidle state index in the scheduler
sched: Add over-utilization/tipping point indicator
sched: Estimate energy impact of scheduling decisions
sched: Extend sched_group_energy to test load-balancing decisions
sched: Calculate energy consumption of sched_group
sched: Highest energy aware balancing sched_domain level pointer
sched: Relocated cpu_util() and change return type
sched: Compute cpu capacity available at current frequency
arm64: Cpu invariant scheduler load-tracking and capacity support
arm: Cpu invariant scheduler load-tracking and capacity support
sched: Introduce SD_SHARE_CAP_STATES sched_domain flag
sched: Initialize energy data structures
sched: Introduce energy data structures
sched: Make energy awareness a sched feature
sched: Documentation for scheduler energy cost model
sched: Prevent unnecessary active balance of single task in sched group
sched: Enable idle balance to pull single task towards cpu with higher capacity
sched: Consider spare cpu capacity at task wake-up
sched: Add cpu capacity awareness to wakeup balancing
sched: Store system-wide maximum cpu capacity in root domain
arm: Update arch_scale_cpu_capacity() to reflect change to define
arm64: Enable frequency invariant scheduler load-tracking support
arm: Enable frequency invariant scheduler load-tracking support
cpufreq: Frequency invariant scheduler load-tracking support
sched/fair: Fix new task's load avg removed from source CPU in wake_up_new_task()
Conflicts:
drivers/cpufreq/Kconfig
include/linux/cpufreq.h
include/trace/events/power.h
Change-Id: I0efa846911ea6b8d3f458115529cf67be73858e3
Signed-off-by: Huang, Tao <huangtao@rock-chips.com>
commit 310944d148 upstream.
The component master driver imx-drm-core matches component devices using
their of_node. Since commit 950b410dd1ab ("gpu: ipu-v3: Fix imx-ipuv3-crtc
module autoloading"), the imx-ipuv3-crtc dev->of_node is not set during
probing. Before that, of_node was set and caused an of: modalias to be
used instead of the platform: modalias, which broke module autoloading.
On the other hand, if dev->of_node is not set yet when the imx-ipuv3-crtc
probe function calls component_add, component matching in imx-drm-core
fails. While dev->of_node will be set once the next component tries to
bring up the component master, imx-drm-core component binding will never
succeed if one of the crtc devices is probed last.
Add of_node to the component platform data and match against the
pdata->of_node instead of dev->of_node in imx-drm-core to work around
this problem.
Fixes: 950b410dd1ab ("gpu: ipu-v3: Fix imx-ipuv3-crtc module autoloading")
Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
Tested-by: Fabio Estevam <fabio.estevam@nxp.com>
Tested-by: Lothar Waßmann <LW@KARO-electronics.de>
Tested-by: Heiko Schocher <hs@denx.de>
Tested-by: Chris Ruehl <chris.ruehl@gtsys.com.hk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 759c01142a upstream.
On no-so-small systems, it is possible for a single process to cause an
OOM condition by filling large pipes with data that are never read. A
typical process filling 4000 pipes with 1 MB of data will use 4 GB of
memory. On small systems it may be tricky to set the pipe max size to
prevent this from happening.
This patch makes it possible to enforce a per-user soft limit above
which new pipes will be limited to a single page, effectively limiting
them to 4 kB each, as well as a hard limit above which no new pipes may
be created for this user. This has the effect of protecting the system
against memory abuse without hurting other users, and still allowing
pipes to work correctly though with less data at once.
The limit are controlled by two new sysctls : pipe-user-pages-soft, and
pipe-user-pages-hard. Both may be disabled by setting them to zero. The
default soft limit allows the default number of FDs per process (1024)
to create pipes of the default size (64kB), thus reaching a limit of 64MB
before starting to create only smaller pipes. With 256 processes limited
to 1024 FDs each, this results in 1024*64kB + (256*1024 - 1024) * 4kB =
1084 MB of memory allocated for a user. The hard limit is disabled by
default to avoid breaking existing applications that make intensive use
of pipes (eg: for splicing).
Reported-by: socketpair@gmail.com
Reported-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Mitigates: CVE-2013-4312 (Linux 2.0+)
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Willy Tarreau <w@1wt.eu>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Cc: Moritz Muehlenhoff <moritz@wikimedia.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 4b50bcc7ed upstream.
Since commit 92923ca3aa ("mm: meminit: only set page reserved in the
memblock region") the reserved bit is set on reserved memblock regions.
However start and end address are passed as unsigned long. This is only
32bit on i386, so it can end up marking the wrong pages reserved for
ranges at 4GB and above.
This was observed on a 32bit Xen dom0 which was booted with initial
memory set to a value below 4G but allowing to balloon in memory
(dom0_mem=1024M for example). This would define a reserved bootmem
region for the additional memory (for example on a 8GB system there was
a reverved region covering the 4GB-8GB range). But since the addresses
were passed on as unsigned long, this was actually marking all pages
from 0 to 4GB as reserved.
Fixes: 92923ca3aa ("mm: meminit: only set page reserved in the memblock region")
Link: http://lkml.kernel.org/r/1463491221-10573-1-git-send-email-stefan.bader@canonical.com
Signed-off-by: Stefan Bader <stefan.bader@canonical.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch add max_burst to dma_get_slave_caps for clients
to get the burst capability of slave dma controller.
Conflicts:
include/linux/dmaengine.h
Change-Id: I7ffcf775ad48247ee0bfa9e18c8ee3b4b256eab2
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
Signed-off-by: Caesar Wang <wxt@rock-chips.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
(cherry picked from commit 6d5bbed30f)
