Typically, when not in HE mode, we will not perform well
with AMSDUs bigger than 8K.
Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Recently we started to send the WEP keys to the firmware so
that we could use hardware Tx encryption also on newer
devices that require the keys to be installed in the firmware
for encryption (as opposed to older devices that can get
the key in the Tx command for each Tx).
When we implemented that, we forgot to remove the key when
we remove a station leading to a situation where a station
that connects and disconnects a lot of times exhausts the
key database inside the firmware.
A fix was made for that, but we always removed the same
key: mvmvif->ap_wep_key which means that we removed the
same key entry in the firmware. This can make sense since
in WEP, the key is the same for all the stations, but the
internal implementation of iwl_mvm_set_sta_key and
iwl_mvm_remove_sta_key assumes that each station uses a
different key in the firmware's key database.
So now we got to the situation where we have a single
ieee80211_key_conf instance that means, a single
ieee80211_key_conf.hw_key_idx index for several stations
and hence for several keys.
ieee80211_key_conf.hw_key_idx is set to 0 when the first
station associates, and then it is overwritten to 1 when
the second station associates which is a buggy of course.
This led to the following message upon the removal of the
second station:
iwlwifi 0000:00:03.0: offset 1 not used in fw key table.
WARNING: CPU: 2 PID: 27883 at net/mac80211/sta_info.c:1122 __sta_info_destroy_part2+0x16b/0x180 [mac80211]
RIP: 0010:__sta_info_destroy_part2+0x16b/0x180 [mac80211]
Call Trace:
__sta_info_destroy+0x2a/0x40 [mac80211]
sta_info_destroy_addr_bss+0x38/0x60 [mac80211]
ieee80211_del_station+0x1d/0x30 [mac80211]
nl80211_del_station+0xe0/0x1f0 [cfg80211]
Fix this by copying the ieee80211_key_conf structure for
each and every station. This is the easiest way to properly
remove the keys with the right index. Another solution
would have been to allow several stations to use the same
key offset in the firmware. That would require to change
the way we track keys in iwlmvm and not really worth it.
Also, maintain correctly fw_key_table when we add a key
for the multicast station.
Remove the key when we remove the multicast station.
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Fixes: 337bfc9881 ("iwlwifi: mvm: set wep key for all stations in soft ap mode")
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Some switch-cases were missing a fall through comment, so the compiler
may warn. Fix that by adding the comments where needed. In other
cases there was more text in the comment, which the compiler doesn't
recognize, so either remove the extra text or move it to a separate
comment line as appropriate.
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Some case statements were missing a comment about falling through or
had more text than the compiler can recognize. Fix that.
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Align the documentation of the structs with the latest version in
the FW.
Signed-off-by: Shahar S Matityahu <shahar.s.matityahu@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
There are several dumping flows in the driver in case of a fail
prior to operational.
In some cases we get 2 dumps while in others we get none.
Fix this by uniting the different flows.
Add a different dump type to driver triggered dumps in case we want
a dump but did not got assert, and make all dumping go through
iwl_fw_dbg_collect_desc to avoid multiple dumps.
Signed-off-by: Shahar S Matityahu <shahar.s.matityahu@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
The firmware is changing the format of the beacon
notification to remove the dependency on the Tx response
format.
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
The driver should support 64 region ids. Update
IWL_FW_INI_MAX_REGION_ID accordingly.
Signed-off-by: Shahar S Matityahu <shahar.s.matityahu@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Align the triggers enum in the driver to the one in the FW.
Signed-off-by: Shahar S Matityahu <shahar.s.matityahu@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
There are two cases that can cause the alive flow to fail,
an assert or a timeout.
Currently we mask any incoming asserts when we wait for alive.
Solve this by differentiating between the two cases:
1. Let the regular error handling to handle a received assert
2. Do a dump collection in the case of a timeout
Signed-off-by: Shahar S Matityahu <shahar.s.matityahu@intel.com>
Fixes: f38efdb293 ("iwlwifi: add dump collection in case alive flow fails")
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
When we do channel switch, we used to schedule time events
ourselves. This was offloaded to FW. Support the new command
and flow.
Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Simplify some return conditions found by running a semantic patch
to detect unnecessary assignments to local variables.
Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Prior to gen2 we allocate the paging memory via alloc_pages
which requires passing ownership on the memory between the
cpu and the device using dma_sync_single_for_cpu and
dma_sync_single_for_device.
Add missing dma_sync_single_for_device in iwl_dump_paging
after copying the memory.
since gen2, we allocate the paging memory using dma_alloc_coherent
which does not need passing ownership between the cpu and device.
Remove unneeded call to dma_sync_single_for_cpu in
iwl_trans_pcie_dump_data prior to copying the memory.
Signed-off-by: Shahar S Matityahu <shahar.s.matityahu@intel.com>
Fixes: 5538409ba3 ("iwlwifi: pcie: support page dumping in wrt in gen2")
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Some of the region struct fields have misleading naming
change those fields to have an informative naming
Signed-off-by: Shahar S Matityahu <shahar.s.matityahu@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
change IWL_FW_INI_LOCATION_SRAM_INVALID into IWL_FW_INI_LOCATION_INVALID
Signed-off-by: Shahar S Matityahu <shahar.s.matityahu@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
update ini struct documentation to the structs being used in the FW
Signed-off-by: Shahar S Matityahu <shahar.s.matityahu@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Since commit c40dd2f766 ("powerpc: Add System RAM to /proc/iomem")
it is possible to use the generic walk_system_ram_range() and
the generic page_is_ram().
To enable the use of walk_system_ram_range() by the IBM EHEA ethernet
driver, we still need an export of the generic function.
As powerpc was the only user of CONFIG_ARCH_HAS_WALK_MEMORY, the
ifdef around the generic walk_system_ram_range() has become useless
and can be dropped.
Fixes: c40dd2f766 ("powerpc: Add System RAM to /proc/iomem")
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
[mpe: Keep the EXPORT_SYMBOL_GPL in powerpc code]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
This reverts commit abd7d0972a. This
change was already partially reverted by John Stultz in
commit 9c6d26df1f ("arm64: dts: hikey: Fix eMMC corruption regression").
This change appears to cause controller resets and block read failures
which prevents successful booting on some hikey boards.
Cc: Ryan Grachek <ryan@edited.us>
Cc: Wei Xu <xuwei5@hisilicon.com>
Cc: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Cc: Rob Herring <robh+dt@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: linux-arm-kernel@lists.infradead.org
Cc: devicetree@vger.kernel.org
Cc: stable <stable@vger.kernel.org> #4.17+
Signed-off-by: Alistair Strachan <astrachan@google.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Wei Xu <xuwei5@hisilicon.com>
Somewhere along recent changes to power control of the wl1835, power-on
became very unreliable on the hikey, failing like this:
wl1271_sdio: probe of mmc2:0001:1 failed with error -16
wl1271_sdio: probe of mmc2:0001:2 failed with error -16
After playing with some dt parameters and comparing to other users of
this chip, it turned out we need some power-on delay to make things
stable again. In contrast to those other users which define 200 ms, the
hikey would already be happy with 1 ms. Still, we use the safer 10 ms,
like on the Ultra96.
Fixes: ea45267873 ("arm64: dts: hikey: Fix WiFi support")
Cc: <stable@vger.kernel.org> #4.12+
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Acked-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Wei Xu <xuwei5@hisilicon.com>
Select IRQ_DOMAIN_HIERARCHY for spmi-gpio in Kconfig since this driver
is now setup as a hierarchical IRQ chip.
Signed-off-by: Brian Masney <masneyb@onstation.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Export the irq_chip_set_wake_parent symbol so that drivers with
hierarchical IRQ chips can be built as a module.
Signed-off-by: Brian Masney <masneyb@onstation.org>
Reported-by: Mark Brown <broonie@kernel.org>
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
When first enabling preemption, we hesitated from making it a free-for-all
where every higher priority client would force a preempt-to-idle cycle
and take over from all lower priority clients. We hesitated because we
were uncertain just how well preemption would work in practice, whether
the preemption latency itself would detract from the latency gains for
higher priority tasks and whether it would work at all. Since
introducing preemption, we have been enabling it for more common tasks,
even giving normal clients a small preemptive boost when they first
start (to aide fairness and improve interactivity). Now lets take one
step further and give permission for all normal (priority:0) clients to
preempt any idle (priority:<0) task so that users running long compute
jobs do not overly impact other jobs (i.e. their desktop) and the system
remains responsive under such idle loads.
References: f6322eddaf ("drm/i915/preemption: Allow preemption between submission ports")
References: b16c765122 ("drm/i915: Priority boost for new clients")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: "Bloomfield, Jon" <jon.bloomfield@intel.com>
Cc: "Stead, Alan" <alan.stead@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190204084116.3013-1-chris@chris-wilson.co.uk
Originally snd_soc_init_platform was not cleaning up its pointers, this
was fixed to always reallocate dynamic memory but created a memory leak
when snd_soc_init_platform was called multiple times during the same
probe attempt and also threw away any changes made to the struct between
calls. In order to avoid reallocating memory that is still valid, the
behaviour will be changed to clear the dynamically set pointers on a
probe error and a unregister event and snd_soc_init_platform will go
back to its original behaviour of only allocating null pointers so it will
stop throwing away valid changes.
Signed-off-by: Curtis Malainey <cujomalainey@chromium.org>
Signed-off-by: Mark Brown <broonie@kernel.org>
The of_find_device_by_node() takes a reference to the underlying device
structure, we should release that reference.
Signed-off-by: Wen Yang <yellowriver2010@hotmil.com>
Cc: Timur Tabi <timur@kernel.org>
Cc: Nicolin Chen <nicoleotsuka@gmail.com>
Cc: Xiubo Li <Xiubo.Lee@gmail.com>
Cc: Fabio Estevam <festevam@gmail.com>
Cc: Liam Girdwood <lgirdwood@gmail.com>
Cc: Mark Brown <broonie@kernel.org>
Cc: Jaroslav Kysela <perex@perex.cz>
Cc: Takashi Iwai <tiwai@suse.com>
Cc: alsa-devel@alsa-project.org
Cc: linuxppc-dev@lists.ozlabs.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>
Add dma-channel-mask as a property for k3dma, it defines
available dma channels which a non-secure mode driver can use.
One sample usage of this is in Hi3660 SoC. DMA channel 0 is
reserved to lpm3, which is a coprocessor for power management. So
as a result, any request in kernel (which runs on main processor
and in non-secure mode) should start from at least channel 1.
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Vinod Koul <vkoul@kernel.org>
Cc: Tanglei Han <hantanglei@huawei.com>
Cc: Zhuangluan Su <suzhuangluan@hisilicon.com>
Cc: Ryan Grachek <ryan@edited.us>
Cc: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Cc: Guodong Xu <guodong.xu@linaro.org>
Cc: dmaengine@vger.kernel.org
Signed-off-by: Li Yu <liyu65@hisilicon.com>
[jstultz: Reworked to use a channel mask]
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Axi_config controls whether DMA resources can be accessed in non-secure
mode, such as linux kernel. The register should be set by the bootloader
stage and depends on the device.
Thus, this patch removes axi_config from k3dma driver.
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Vinod Koul <vkoul@kernel.org>
Cc: Tanglei Han <hantanglei@huawei.com>
Cc: Zhuangluan Su <suzhuangluan@hisilicon.com>
Cc: Ryan Grachek <ryan@edited.us>
Cc: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Cc: dmaengine@vger.kernel.org
Acked-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Signed-off-by: Li Yu <liyu65@hisilicon.com>
Signed-off-by: Guodong Xu <guodong.xu@linaro.org>
[jstultz: Minor tweaks to commit message]
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Vinod Koul <vkoul@kernel.org>
On the hi3660 hardware there are two (at least) DMA controllers,
the DMA-P (Peripheral DMA) and the DMA-A (Audio DMA). The
two blocks are similar, but have some slight differences. This
resulted in the vendor implementing two separate drivers, which
after review, they have been able to condense and re-use the
existing k3dma driver.
Thus, this patch adds support for the new "hisi-pcm-asp-dma-1.0"
compatible string in the binding.
One difference with the DMA-A controller, is that it does not
need to initialize a clock. So we skip this by adding and using
soc data flags.
After above this driver will support both k3 and hisi_asp dma
hardware.
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Vinod Koul <vkoul@kernel.org>
Cc: Zhuangluan Su <suzhuangluan@hisilicon.com>
Cc: Ryan Grachek <ryan@edited.us>
Cc: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Cc: dmaengine@vger.kernel.org
Acked-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Signed-off-by: Youlin Wang <wwx575822@notesmail.huawei.com>
Signed-off-by: Tanglei Han <hantanglei@huawei.com>
[jstultz: Reworked to use of_match_data, commit msg improvements]
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Some dma channels can be reserved for secure mode or other
hardware on the SoC, so provide a binding for a bitmask
listing the available channels for the kernel to use.
This follows the pre-existing bcm,dma-channel-mask binding.
Cc: Vinod Koul <vkoul@kernel.org>
Cc: Rob Herring <robh+dt@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Tanglei Han <hantanglei@huawei.com>
Cc: Zhuangluan Su <suzhuangluan@hisilicon.com>
Cc: Ryan Grachek <ryan@edited.us>
Cc: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Cc: dmaengine@vger.kernel.org
Cc: devicetree@vger.kernel.org
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Looks like refactoring didn't go well and left ALDO2, DLDO2 and ELDO3
definitions broken for AXP803 - now they are using register address
instead of mask. Fix it by using mask where necessary.
Fixes: db4a555f7c ("regulator: axp20x: use defines for masks")
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com>
Reviewed-by: Chen-Yu Tsai <wens@csie.org>
Signed-off-by: Mark Brown <broonie@kernel.org>
The function used to probe the peripheral clock controller of the arm64
amlogic SoCs is mostly the same. We now have 3 of those controllers so
it is time to factorize things a bit.
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Reviewed-by: Neil Armstrong <narmstrong@baylibre.com>
Signed-off-by: Neil Armstrong <narmstrong@baylibre.com>
Link: https://lkml.kernel.org/r/20190201145345.6795-5-jbrunet@baylibre.com
Add the peripheral clock controller found in the g12a SoC family
Signed-off-by: Jian Hu <jian.hu@amlogic.com>
Signed-off-by: Neil Armstrong <narmstrong@baylibre.com>
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Reviewed-by: Neil Armstrong <narmstrong@baylibre.com>
Link: https://lkml.kernel.org/r/20190201145345.6795-4-jbrunet@baylibre.com
Add new clock controller compatible and dt-bindings header for the
Everything-Else domain of the g12a SoC
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Jian Hu <jian.hu@amlogic.com>
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Reviewed-by: Neil Armstrong <narmstrong@baylibre.com>
Signed-off-by: Neil Armstrong <narmstrong@baylibre.com>
Link: https://lkml.kernel.org/r/20190201145345.6795-3-jbrunet@baylibre.com
The g12a use fractional parameter of 17 useful bits. At the moment, this
parameter in encoded using u16 value. Use this opportunity to switch all
the pll to parameter to unsigned int. This should save us some annoying
trouble shooting when and m and n field eventually grow as well.
This patch also introduce pll multiplier range. On the g12a, the hifi and
gp0 plls are able to lock as long as the following condition is met:
55 <= m/n <= 255.
The param table describing this would be huge which is a waste of memory.
Using ranges, we can save memory. Ranges also help find the best pll
parameter significantly faster since we don't have to try all the possible
settings.
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Reviewed-by: Neil Armstrong <narmstrong@baylibre.com>
Signed-off-by: Neil Armstrong <narmstrong@baylibre.com>
[jbrunet: fixed fix pll settings calculation with arm32]
Link: https://lkml.kernel.org/r/20190201145345.6795-2-jbrunet@baylibre.com
Sargun reported a crash:
"I picked up c40f7d74c7 sched/fair: Fix
infinite loop in update_blocked_averages() by reverting a9e7f6544b
and put it on top of 4.19.13. In addition to this, I uninlined
list_add_leaf_cfs_rq for debugging.
This revealed a new bug that we didn't get to because we kept getting
crashes from the previous issue. When we are running with cgroups that
are rapidly changing, with CFS bandwidth control, and in addition
using the cpusets cgroup, we see this crash. Specifically, it seems to
occur with cgroups that are throttled and we change the allowed
cpuset."
The algorithm used to order cfs_rq in rq->leaf_cfs_rq_list assumes that
it will walk down to root the 1st time a cfs_rq is used and we will finish
to add either a cfs_rq without parent or a cfs_rq with a parent that is
already on the list. But this is not always true in presence of throttling.
Because a cfs_rq can be throttled even if it has never been used but other CPUs
of the cgroup have already used all the bandwdith, we are not sure to go down to
the root and add all cfs_rq in the list.
Ensure that all cfs_rq will be added in the list even if they are throttled.
[ mingo: Fix !CGROUPS build. ]
Reported-by: Sargun Dhillon <sargun@sargun.me>
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: tj@kernel.org
Fixes: 9c2791f936 ("Fix hierarchical order in rq->leaf_cfs_rq_list")
Link: https://lkml.kernel.org/r/1548825767-10799-1-git-send-email-vincent.guittot@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The magic in list_add_leaf_cfs_rq() requires that at the end of
enqueue_task_fair():
rq->tmp_alone_branch == &rq->lead_cfs_rq_list
If this is violated, list integrity is compromised for list entries
and the tmp_alone_branch pointer might dangle.
Also, reflow list_add_leaf_cfs_rq() while there. This looses one
indentation level and generates a form that's convenient for the next
patch.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
move_queued_task() synchronizes with task_rq_lock() as follows:
move_queued_task() task_rq_lock()
[S] ->on_rq = MIGRATING [L] rq = task_rq()
WMB (__set_task_cpu()) ACQUIRE (rq->lock);
[S] ->cpu = new_cpu [L] ->on_rq
where "[L] rq = task_rq()" is ordered before "ACQUIRE (rq->lock)" by an
address dependency and, in turn, "ACQUIRE (rq->lock)" is ordered before
"[L] ->on_rq" by the ACQUIRE itself.
Use READ_ONCE() to load ->cpu in task_rq() (c.f., task_cpu()) to honor
this address dependency. Also, mark the accesses to ->cpu and ->on_rq
with READ_ONCE()/WRITE_ONCE() to comply with the LKMM.
Signed-off-by: Andrea Parri <andrea.parri@amarulasolutions.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul E. McKenney <paulmck@linux.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Will Deacon <will.deacon@arm.com>
Link: https://lkml.kernel.org/r/20190121155240.27173-1-andrea.parri@amarulasolutions.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
register_sched_domain_sysctl() copies the cpu_possible_mask into
sd_sysctl_cpus, but only if sd_sysctl_cpus hasn't already been
allocated (ie, CONFIG_CPUMASK_OFFSTACK is set). However, when
CONFIG_CPUMASK_OFFSTACK is not set, sd_sysctl_cpus is left
uninitialized (all zeroes) and the kernel may fail to initialize
sched_domain sysctl entries for all possible CPUs.
This is visible to the user if the kernel is booted with maxcpus=n, or
if ACPI tables have been modified to leave CPUs offline, and then
checking for missing /proc/sys/kernel/sched_domain/cpu* entries.
Fix this by separating the allocation and initialization, and adding a
flag to initialize the possible CPU entries while system booting only.
Tested-by: Syuuichirou Ishii <ishii.shuuichir@jp.fujitsu.com>
Tested-by: Tarumizu, Kohei <tarumizu.kohei@jp.fujitsu.com>
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Masayoshi Mizuma <m.mizuma@jp.fujitsu.com>
Acked-by: Joe Lawrence <joe.lawrence@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Masayoshi Mizuma <msys.mizuma@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: https://lkml.kernel.org/r/20190129151245.5073-1-msys.mizuma@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
util_est is mainly meant to be a lower-bound for tasks utilization.
That's why task_util_est() returns the actual util_avg when it's higher
than the estimated utilization.
With new invaraince signal and without any special check on samples
collection, if a task is limited because of thermal capping for
example, we could end up overestimating its utilization and thus
perhaps generating an unwanted frequency spike when the capping is
relaxed... and (even worst) it will take some more activations for the
estimated utilization to converge back to the actual utilization.
Since we cannot easily know if there is idle time in a CPU when a task
completes an activation with a utilization higher then the CPU capacity,
we skip the sampling when utilization is higher than CPU's capacity.
Suggested-by: Patrick Bellasi <patrick.bellasi@arm.com>
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Morten.Rasmussen@arm.com
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: bsegall@google.com
Cc: dietmar.eggemann@arm.com
Cc: pjt@google.com
Cc: pkondeti@codeaurora.org
Cc: quentin.perret@arm.com
Cc: rjw@rjwysocki.net
Cc: srinivas.pandruvada@linux.intel.com
Cc: thara.gopinath@linaro.org
Link: https://lkml.kernel.org/r/1548257214-13745-4-git-send-email-vincent.guittot@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The current implementation of load tracking invariance scales the
contribution with current frequency and uarch performance (only for
utilization) of the CPU. One main result of this formula is that the
figures are capped by current capacity of CPU. Another one is that the
load_avg is not invariant because not scaled with uarch.
The util_avg of a periodic task that runs r time slots every p time slots
varies in the range :
U * (1-y^r)/(1-y^p) * y^i < Utilization < U * (1-y^r)/(1-y^p)
with U is the max util_avg value = SCHED_CAPACITY_SCALE
At a lower capacity, the range becomes:
U * C * (1-y^r')/(1-y^p) * y^i' < Utilization < U * C * (1-y^r')/(1-y^p)
with C reflecting the compute capacity ratio between current capacity and
max capacity.
so C tries to compensate changes in (1-y^r') but it can't be accurate.
Instead of scaling the contribution value of PELT algo, we should scale the
running time. The PELT signal aims to track the amount of computation of
tasks and/or rq so it seems more correct to scale the running time to
reflect the effective amount of computation done since the last update.
In order to be fully invariant, we need to apply the same amount of
running time and idle time whatever the current capacity. Because running
at lower capacity implies that the task will run longer, we have to ensure
that the same amount of idle time will be applied when system becomes idle
and no idle time has been "stolen". But reaching the maximum utilization
value (SCHED_CAPACITY_SCALE) means that the task is seen as an
always-running task whatever the capacity of the CPU (even at max compute
capacity). In this case, we can discard this "stolen" idle times which
becomes meaningless.
In order to achieve this time scaling, a new clock_pelt is created per rq.
The increase of this clock scales with current capacity when something
is running on rq and synchronizes with clock_task when rq is idle. With
this mechanism, we ensure the same running and idle time whatever the
current capacity. This also enables to simplify the pelt algorithm by
removing all references of uarch and frequency and applying the same
contribution to utilization and loads. Furthermore, the scaling is done
only once per update of clock (update_rq_clock_task()) instead of during
each update of sched_entities and cfs/rt/dl_rq of the rq like the current
implementation. This is interesting when cgroup are involved as shown in
the results below:
On a hikey (octo Arm64 platform).
Performance cpufreq governor and only shallowest c-state to remove variance
generated by those power features so we only track the impact of pelt algo.
each test runs 16 times:
./perf bench sched pipe
(higher is better)
kernel tip/sched/core + patch
ops/seconds ops/seconds diff
cgroup
root 59652(+/- 0.18%) 59876(+/- 0.24%) +0.38%
level1 55608(+/- 0.27%) 55923(+/- 0.24%) +0.57%
level2 52115(+/- 0.29%) 52564(+/- 0.22%) +0.86%
hackbench -l 1000
(lower is better)
kernel tip/sched/core + patch
duration(sec) duration(sec) diff
cgroup
root 4.453(+/- 2.37%) 4.383(+/- 2.88%) -1.57%
level1 4.859(+/- 8.50%) 4.830(+/- 7.07%) -0.60%
level2 5.063(+/- 9.83%) 4.928(+/- 9.66%) -2.66%
Then, the responsiveness of PELT is improved when CPU is not running at max
capacity with this new algorithm. I have put below some examples of
duration to reach some typical load values according to the capacity of the
CPU with current implementation and with this patch. These values has been
computed based on the geometric series and the half period value:
Util (%) max capacity half capacity(mainline) half capacity(w/ patch)
972 (95%) 138ms not reachable 276ms
486 (47.5%) 30ms 138ms 60ms
256 (25%) 13ms 32ms 26ms
On my hikey (octo Arm64 platform) with schedutil governor, the time to
reach max OPP when starting from a null utilization, decreases from 223ms
with current scale invariance down to 121ms with the new algorithm.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Morten.Rasmussen@arm.com
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: bsegall@google.com
Cc: dietmar.eggemann@arm.com
Cc: patrick.bellasi@arm.com
Cc: pjt@google.com
Cc: pkondeti@codeaurora.org
Cc: quentin.perret@arm.com
Cc: rjw@rjwysocki.net
Cc: srinivas.pandruvada@linux.intel.com
Cc: thara.gopinath@linaro.org
Link: https://lkml.kernel.org/r/1548257214-13745-3-git-send-email-vincent.guittot@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
If a file with capability set (and hence security.capability xattr) is
written kernel clears security.capability xattr. For overlay, during file
copy up if xattrs are copied up first and then data is, copied up. This
means data copy up will result in clearing of security.capability xattr
file on lower has. And this can result into surprises. If a lower file has
CAP_SETUID, then it should not be cleared over copy up (if nothing was
actually written to file).
This also creates problems with chown logic where it first copies up file
and then tries to clear setuid bit. But by that time security.capability
xattr is already gone (due to data copy up), and caller gets -ENODATA.
This has been reported by Giuseppe here.
https://github.com/containers/libpod/issues/2015#issuecomment-447824842
Fix this by copying up data first and then metadta. This is a regression
which has been introduced by my commit as part of metadata only copy up
patches.
TODO: There will be some corner cases where a file is copied up metadata
only and later data copy up happens and that will clear security.capability
xattr. Something needs to be done about that too.
Fixes: bd64e57586 ("ovl: During copy up, first copy up metadata and then data")
Cc: <stable@vger.kernel.org> # v4.19+
Reported-by: Giuseppe Scrivano <gscrivan@redhat.com>
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
This adds an smp_acquire__after_ctrl_dep() barrier on successful
decrease of refcounter value from 1 to 0 for refcount_dec(sub)_and_test
variants and therefore gives stronger memory ordering guarantees than
prior versions of these functions.
Co-developed-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Elena Reshetova <elena.reshetova@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Andrea Parri <andrea.parri@amarulasolutions.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Will Deacon <will.deacon@arm.com>
Cc: dvyukov@google.com
Cc: keescook@chromium.org
Cc: stern@rowland.harvard.edu
Link: https://lkml.kernel.org/r/1548847131-27854-2-git-send-email-elena.reshetova@intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Track the number of slowpath locking operations that are being done
without any MCS node available as well renaming lock_index[123] to make
them more descriptive.
Using these stat counters is one way to find out if a code path is
being exercised.
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: James Morse <james.morse@arm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: SRINIVAS <srinivas.eeda@oracle.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Zhenzhong Duan <zhenzhong.duan@oracle.com>
Link: https://lkml.kernel.org/r/1548798828-16156-3-git-send-email-longman@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Four queue nodes per CPU are allocated to enable up to 4 nesting levels
using the per-CPU nodes. Nested NMIs are possible in some architectures.
Still it is very unlikely that we will ever hit more than 4 nested
levels with contention in the slowpath.
When that rare condition happens, however, it is likely that the system
will hang or crash shortly after that. It is not good and we need to
handle this exception case.
This is done by spinning directly on the lock using repeated trylock.
This alternative code path should only be used when there is nested
NMIs. Assuming that the locks used by those NMI handlers will not be
heavily contended, a simple TAS locking should work out.
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Will Deacon <will.deacon@arm.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: James Morse <james.morse@arm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: SRINIVAS <srinivas.eeda@oracle.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Zhenzhong Duan <zhenzhong.duan@oracle.com>
Link: https://lkml.kernel.org/r/1548798828-16156-2-git-send-email-longman@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Some users, specifically futexes and rwsems, required fixes
that allowed the callers to be safe when wakeups occur before
they are expected by wake_up_q(). Such scenarios also play
games and rely on reference counting, and until now were
pivoting on wake_q doing it. With the wake_q_add() call being
moved down, this can no longer be the case. As such we end up
with a a double task refcounting overhead; and these callers
care enough about this (being rather core-ish).
This patch introduces a wake_q_add_safe() call that serves
for callers that have already done refcounting and therefore the
task is 'safe' from wake_q point of view (int that it requires
reference throughout the entire queue/>wakeup cycle). In the one
case it has internal reference counting, in the other case it
consumes the reference counting.
Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Waiman Long <longman@redhat.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Xie Yongji <xieyongji@baidu.com>
Cc: Yongji Xie <elohimes@gmail.com>
Cc: andrea.parri@amarulasolutions.com
Cc: lilin24@baidu.com
Cc: liuqi16@baidu.com
Cc: nixun@baidu.com
Cc: yuanlinsi01@baidu.com
Cc: zhangyu31@baidu.com
Link: https://lkml.kernel.org/r/20181218195352.7orq3upiwfdbrdne@linux-r8p5
Signed-off-by: Ingo Molnar <mingo@kernel.org>
