IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
In platform_device_register_full() the err_alloc label is
misleading, we only ever jump to it if the pdev is NULL,
but it then proceeds to free it, which is a no-op.
Remove the label and simply exit the function immediately.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This is intended for use with NVDIMMs that are physically persistent
(physically like flash) so that they can be used as a cost-effective
RAM replacement. Intel Optane DC persistent memory is one
implementation of this kind of NVDIMM.
Currently, a persistent memory region is "owned" by a device driver,
either the "Direct DAX" or "Filesystem DAX" drivers. These drivers
allow applications to explicitly use persistent memory, generally
by being modified to use special, new libraries. (DIMM-based
persistent memory hardware/software is described in great detail
here: Documentation/nvdimm/nvdimm.txt).
However, this limits persistent memory use to applications which
*have* been modified. To make it more broadly usable, this driver
"hotplugs" memory into the kernel, to be managed and used just like
normal RAM would be.
To make this work, management software must remove the device from
being controlled by the "Device DAX" infrastructure:
echo dax0.0 > /sys/bus/dax/drivers/device_dax/unbind
and then tell the new driver that it can bind to the device:
echo dax0.0 > /sys/bus/dax/drivers/kmem/new_id
After this, there will be a number of new memory sections visible
in sysfs that can be onlined, or that may get onlined by existing
udev-initiated memory hotplug rules.
This rebinding procedure is currently a one-way trip. Once memory
is bound to "kmem", it's there permanently and can not be
unbound and assigned back to device_dax.
The kmem driver will never bind to a dax device unless the device
is *explicitly* bound to the driver. There are two reasons for
this: One, since it is a one-way trip, it can not be undone if
bound incorrectly. Two, the kmem driver destroys data on the
device. Think of if you had good data on a pmem device. It
would be catastrophic if you compile-in "kmem", but leave out
the "device_dax" driver. kmem would take over the device and
write volatile data all over your good data.
This inherits any existing NUMA information for the newly-added
memory from the persistent memory device that came from the
firmware. On Intel platforms, the firmware has guarantees that
require each socket's persistent memory to be in a separate
memory-only NUMA node. That means that this patch is not expected
to create NUMA nodes, but will simply hotplug memory into existing
nodes.
Because NUMA nodes are created, the existing NUMA APIs and tools
are sufficient to create policies for applications or memory areas
to have affinity for or an aversion to using this memory.
There is currently some metadata at the beginning of pmem regions.
The section-size memory hotplug restrictions, plus this small
reserved area can cause the "loss" of a section or two of capacity.
This should be fixable in follow-on patches. But, as a first step,
losing 256MB of memory (worst case) out of hundreds of gigabytes
is a good tradeoff vs. the required code to fix this up precisely.
This calculation is also the reason we export
memory_block_size_bytes().
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Reviewed-by: Keith Busch <keith.busch@intel.com>
Cc: Dave Jiang <dave.jiang@intel.com>
Cc: Ross Zwisler <zwisler@kernel.org>
Cc: Vishal Verma <vishal.l.verma@intel.com>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: linux-nvdimm@lists.01.org
Cc: linux-kernel@vger.kernel.org
Cc: linux-mm@kvack.org
Cc: Huang Ying <ying.huang@intel.com>
Cc: Fengguang Wu <fengguang.wu@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: Yaowei Bai <baiyaowei@cmss.chinamobile.com>
Cc: Takashi Iwai <tiwai@suse.de>
Cc: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Vishal Verma <vishal.l.verma@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
If the provided fwnode is an OF node, set dev.of_node as well.
Also add an of_node_reused flag to struct platform_device_info and copy
this to the new device. This is needed to avoid pinctrl settings being
requested twice. See 4e75e1d7dac9 ("driver core: add helper to reuse a
device-tree node") for a longer explanation.
Some drivers are just shims that create extra "glue" devices with the
DT device as parent and have the real driver bind to these. In these
cases, the glue device needs to get a reference to the original DT node
in order for the main driver to access properties and child nodes.
For example, the sunxi-musb driver creates such a glue device using
platform_device_register_full(). Consequently, devices attached to
this USB interface don't get associated with DT nodes, if present,
the way they do with EHCI.
This change will allow sunxi-musb and similar drivers to easily
propagate the DT node to child devices as required.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When no file /path was found, the error code of -ENOENT
enumerated in errno-base.h, is returned. Stating clearly that
the file was not found is much more useful for debugging, So
let's be explicit about that.
Signed-off-by: John Zhao <yuankuiz@codeaurora.org>
Acked-by: Luis Chamberlain <mcgrof@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We only build devm_ioremap_resource() if HAS_IOMEM is selected, so this
dependency must cascade down to devm_platform_ioremap_resource().
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
There are currently 1200+ instances of using platform_get_resource()
and devm_ioremap_resource() together in the kernel tree.
This patch wraps these two calls in a single helper. Thanks to that
we don't have to declare a local variable for struct resource * and can
omit the redundant argument for resource type. We also have one
function call less.
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
When rpm_resume() desactivates the autosuspend timer, it should only
try to cancel hrtimer but not wait for the handler to finish, because
both rpm_resume() and pm_suspend_timer_fn() take the power.lock.
A deadlock is possible as follows:
CPU0 CPU1
rpm_resume()
spin_lock_irqsave
pm_suspend_timer_fn()
spin_lock_irqsave
pm_runtime_deactivate_timer()
hrtimer_cancel()
It is sufficient to call hrtimer_try_to_cancel() from
pm_runtime_deactivate_timer(), because dev->power.timer_expires
reset to 0 by it, so use that function instead of hrtimer_cancel().
Fixes: 8234f6734c5d ("PM-runtime: Switch autosuspend over to using hrtimers")
Reported-by: Sunzhaosheng Sun(Zhaosheng) <sunzhaosheng@hisilicon.com>
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
[ rjw: Changelog ]
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
We need 32ea33a04484 ("mei: bus: export to_mei_cl_device for mei
client devices drivers") for the mei-hdcp patches.
References: https://lkml.org/lkml/2019/2/19/356
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
This is where all the related code already lives.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 4c06c4e6cf63 ("driver core: Fix possible supplier PM-usage
counter imbalance") introduced a regression that causes suppliers
to be suspended prematurely for device links added during consumer
driver probe if the initial PM-runtime status of the consumer is
"suspended" and the consumer is resumed after adding the link and
before pm_runtime_put_suppliers() is called. In that case,
pm_runtime_put_suppliers() will drop the rpm_active refcount for
the link by one and (since rpm_active is equal to two after the
preceding consumer resume) the supplier's PM-runtime usage counter
will be decremented, which may cause the supplier to suspend even
though the consumer's PM-runtime status is "active".
For this reason, partially revert commit 4c06c4e6cf63 as the problem
it tried to fix needs to be addressed somewhat differently, and
change pm_runtime_get_suppliers() and pm_runtime_put_suppliers() so
that the latter only drops rpm_active references acquired by the
former. [This requires adding a new field to struct device_link,
but I coulnd't find a cleaner way to address the issue that would
work in all cases.]
This causes pm_runtime_put_suppliers() to effectively ignore device
links added during consumer probe, so device_link_add() doesn't need
to worry about ensuring that suppliers will remain active after
pm_runtime_put_suppliers() for links created with DL_FLAG_RPM_ACTIVE
set and it only needs to bump up rpm_active by one for those links,
so pm_runtime_active_link() is not necessary any more.
Fixes: 4c06c4e6cf63 ("driver core: Fix possible supplier PM-usage counter imbalance")
Reported-by: Jon Hunter <jonathanh@nvidia.com>
Tested-by: Jon Hunter <jonathanh@nvidia.com>
Tested-by: Ulf Hansson <ulf.hansson@linaro.org>
Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Tested-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Polish the kerneldoc a bit with suggestions from Randy.
v2: Randy found another typo: s/compent/component/
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Cc: "Rafael J. Wysocki" <rafael@kernel.org>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Ramalingam C <ramalingam.c@intel.com>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
All device objects in the driver model contain fields that control the
handling of various power management activities. However, it's not
always useful. There are few instances where pseudo devices are added
to the model just to take advantage of many other features like
kobjects, udev events, and so on. One such example is cpu devices and
their caches.
The sysfs for the cpu caches are managed by adding devices with cpu
as the parent in cpu_device_create() when secondary cpu is brought
online. Generally when the secondary CPUs are hotplugged back in as part
of resume from suspend-to-ram, we call cpu_device_create() from the cpu
hotplug state machine while the cpu device associated with that CPU is
not yet ready to be resumed as the device_resume() call happens bit
later. It's not really needed to set the flag is_prepared for cpu
devices as they are mostly pseudo device and hotplug framework deals
with state machine and not managed through the cpu device.
This often results in annoying warning when resuming:
Enabling non-boot CPUs ...
CPU1: Booted secondary processor
cache: parent cpu1 should not be sleeping
CPU1 is up
CPU2: Booted secondary processor
cache: parent cpu2 should not be sleeping
CPU2 is up
.... and so on.
So in order to fix these kind of errors, we could just completely avoid
doing any power management related initialisations and operations if
they are not used by these devices.
Add no_pm flags to indicate that the device doesn't require any sort of
PM activities and all of them can be completely skipped. We can use the
same flag to also avoid adding not used *power* sysfs entries for these
devices. For now, lets use this for cpu cache devices.
Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Sudeep Holla <sudeep.holla@arm.com>
Tested-by: Eugeniu Rosca <erosca@de.adit-jv.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
As of the patch ("PM / Domains: Mark "name" const in
genpd_dev_pm_attach_by_name()") it's clear that the name in
dev_pm_domain_attach_by_name() can be const. Mark it as so. This
allows drivers to pass in a name that was declared "const" in a
driver.
Fixes: 27dceb81f445 ("PM / Domains: Introduce dev_pm_domain_attach_by_name()")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Stephen Boyd <swboyd@chromium.org>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
The genpd_dev_pm_attach_by_name() simply takes the name and passes it
to of_property_match_string() where the argument is "const char *".
Adding a const here allows a later patch to add a const to
dev_pm_domain_attach_by_name() which allows drivers to pass in a name
that was declared "const" in a driver.
Fixes: 5d6be70add65 ("PM / Domains: Introduce option to attach a device by name to genpd")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Stephen Boyd <swboyd@chromium.org>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
This makes it possible to support drivers that use
fwnode_get_named_child_node() and device_get_named_child_node()
functions.
The node name is for now taken from a device property named
"name". That mimics the old style of naming of the nodes in
devicetree (though with modern flattened DT, the name is
matched against the actual node-name, it used to be done
with a property "name"). In Open Firmware DT the "name"
property is also still being used.
Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
If connections between devices are described in OF graph or
ACPI device graph, we can find them by using the
fwnode_graph_*() functions.
Acked-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Reviewed-by: Jun Li <jun.li@nxp.com>
Tested-by: Jun Li <jun.li@nxp.com>
Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When the connections are defined in firmware, struct
device_connection will have the fwnode member pointing to
the device node (struct fwnode_handle) of the requested
device. The endpoint member for the device names will not be
used at all in that case.
Acked-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Reviewed-by: Jun Li <jun.li@nxp.com>
Tested-by: Jun Li <jun.li@nxp.com>
Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Asynchronous driver probing can help much on kernel fastboot, and
this option can provide a flexible way to optimize and quickly verify
async driver probe.
Also it will help in below cases:
* Some driver actually covers several families of HWs, some of which
could use async probing while others don't. So we can't simply
turn on the PROBE_PREFER_ASYNCHRONOUS flag in driver, but use this
cmdline option, like igb driver async patch discussed at
https://www.spinics.net/lists/netdev/msg545986.html
* For SOC (System on Chip) with multiple spi or i2c controllers, most
of the slave spi/i2c devices will be assigned with fixed controller
number, while async probing may make those controllers get different
index for each boot, which prevents those controller drivers to be
async probed. For platforms not using these spi/i2c slave devices,
they can use this cmdline option to benefit from the async probing.
Suggested-by: Alexander Duyck <alexander.h.duyck@linux.intel.com>
Cc: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Feng Tang <feng.tang@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Devices using the direct-complete optimization are not present it
debug messages printed by the core device suspend and resume code,
which sometimes makes it difficult to diagnose problems related to
them, so add debug messages for those devices.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
If a stateless device link to a certain supplier with
DL_FLAG_PM_RUNTIME set in the flags is added and then removed by the
consumer driver's probe callback, the supplier's PM-runtime usage
counter will be nonzero after that which effectively causes the
supplier to remain "always on" going forward.
Namely, device_link_add() called to add the link invokes
device_link_rpm_prepare() which notices that the consumer driver is
probing, so it increments the supplier's PM-runtime usage counter
with the assumption that the link will stay around until
pm_runtime_put_suppliers() is called by driver_probe_device(),
but if the link goes away before that point, the supplier's
PM-runtime usage counter will remain nonzero.
To prevent that from happening, first rework pm_runtime_get_suppliers()
and pm_runtime_put_suppliers() to use the rpm_active refounts of device
links and make the latter only drop rpm_active and the supplier's
PM-runtime usage counter for each link by one, unless rpm_active is
one already for it. Next, modify device_link_add() to bump up the
new link's rpm_active refcount and the suppliers PM-runtime usage
counter by two, to prevent pm_runtime_put_suppliers(), if it is
called subsequently, from suspending the supplier prematurely (in
case its PM-runtime usage counter goes down to 0 in there).
Due to the way rpm_put_suppliers() works, this change does not
affect runtime suspend of the consumer ends of new device links (or,
generally, device links for which DL_FLAG_PM_RUNTIME has just been
set).
Fixes: e2f3cd831a28 ("driver core: Fix handling of runtime PM flags in device_link_add()")
Reported-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
Tested-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 4080ab083000 ("PM-runtime: Take suppliers into account in
__pm_runtime_set_status()") introduced a race condition that may
trigger if __pm_runtime_set_status() is used incorrectly (that is,
if it is called when PM-runtime is enabled for the target device
and working).
In that case, if the original PM-runtime status of the device is
RPM_SUSPENDED, a runtime resume of the device may occur after
__pm_runtime_set_status() has dropped its power.lock spinlock
and before deactivating its suppliers, so the suppliers may be
deactivated while the device is PM-runtime-active which may lead
to functional issues.
To avoid that, modify __pm_runtime_set_status() to check whether
or not PM-runtime is enabled for the device before activating its
suppliers (if the new status is RPM_ACTIVE) and either return an
error if that's the case or increment the device's disable_depth
counter to prevent PM-runtime from being enabled for it while
the remaining part of the function is running (disable_depth is
then decremented on the way out).
Fixes: 4080ab083000 ("PM-runtime: Take suppliers into account in __pm_runtime_set_status()")
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
ACPI 5 added support for GpioInt resources as a way to provide
information about interrupts mediated via a GPIO controller.
Several device buses (e.g. SPI, I2C) have support for retrieving
an IRQ specified via this type of resource, and providing it
directly to the driver as an IRQ number.
This is not currently done for the platform drivers, as platform_get_irq()
does not try to parse GpioInt() resources. This requires drivers to
either have to support only one possible IRQ resource, or to have code
in place to try both as a failsafe.
While there is a possibility of ambiguity for devices that exposes
multiple IRQs, it is easy and feasible to support the common case
of devices that only expose one IRQ which would be of either type
depending on the underlying system's architecture.
This commit adds support for parsing a GpioInt resource in order
to fulfill a request for the index 0 IRQ for a platform device.
Signed-off-by: Enrico Granata <egranata@chromium.org>
Reviewed-by: Dmitry Torokhov <dtor@chromium.org>
Acked-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This is needed by the new MEI-HDCP support in i915, so will need to go
in through drm and drivers-misc trees at least.
-----BEGIN PGP SIGNATURE-----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=IhHa
-----END PGP SIGNATURE-----
Merge tag 'topic/component-typed-2019-02-11' of git://anongit.freedesktop.org/drm/drm-intel into driver-core-next
Daniel writes:
typed componented support + i915/snd-hda changes
This is needed by the new MEI-HDCP support in i915, so will need to go
in through drm and drivers-misc trees at least.
* tag 'topic/component-typed-2019-02-11' of git://anongit.freedesktop.org/drm/drm-intel:
i915/snd_hdac: I915 subcomponent for the snd_hdac
components: multiple components for a device
component: Add documentation
Here are some driver core fixes for 5.0-rc6.
Well, not so much "driver core" as "debugfs". There's a lot of
outstanding debugfs cleanup patches coming in through different
subsystem trees, and in that process the debugfs core was found that it
really should return errors when something bad happens, to prevent
random files from showing up in the root of debugfs afterward. So
debugfs was fixed up to handle this properly, and then two fixes for
the relay and blk-mq code was needed as it was making invalid
assumptions about debugfs return values.
There's also a cacheinfo fix in here that resolves a tiny issue.
All of these have been in linux-next for over a week with no reported
problems.
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-----BEGIN PGP SIGNATURE-----
iG0EABECAC0WIQT0tgzFv3jCIUoxPcsxR9QN2y37KQUCXF069g8cZ3JlZ0Brcm9h
aC5jb20ACgkQMUfUDdst+yk0+gCgy9PTVAJR5ZbYtWTJOTdBnd7pfqMAoMuGxc+6
LLEbfSykLRxEf5SeOJun
=KP8e
-----END PGP SIGNATURE-----
Merge tag 'driver-core-5.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core
Pull driver core fixes from Greg KH:
"Here are some driver core fixes for 5.0-rc6.
Well, not so much "driver core" as "debugfs". There's a lot of
outstanding debugfs cleanup patches coming in through different
subsystem trees, and in that process the debugfs core was found that
it really should return errors when something bad happens, to prevent
random files from showing up in the root of debugfs afterward. So
debugfs was fixed up to handle this properly, and then two fixes for
the relay and blk-mq code was needed as it was making invalid
assumptions about debugfs return values.
There's also a cacheinfo fix in here that resolves a tiny issue.
All of these have been in linux-next for over a week with no reported
problems"
* tag 'driver-core-5.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core:
blk-mq: protect debugfs_create_files() from failures
relay: check return of create_buf_file() properly
debugfs: debugfs_lookup() should return NULL if not found
debugfs: return error values, not NULL
debugfs: fix debugfs_rename parameter checking
cacheinfo: Keep the old value if of_property_read_u32 fails
Component framework is extended to support multiple components for
a struct device. These will be matched with different masters based on
its sub component value.
We are introducing this, as I915 needs two different components
with different subcomponent value, which will be matched to two
different component masters(Audio and HDCP) based on the subcomponent
values.
v2: Add documenation.
v3: Rebase on top of updated documenation.
v4: Review from Rafael:
- Remove redundant "This" from kerneldoc (also in the previous patch)
- Streamline the logic in find_component() a bit.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> (v1 code)
Signed-off-by: Ramalingam C <ramalingam.c@intel.com> (v1 commit message)
Cc: Ramalingam C <ramalingam.c@intel.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Russell King <rmk+kernel@arm.linux.org.uk>
Cc: Rafael J. Wysocki <rafael@kernel.org>
Cc: Jaroslav Kysela <perex@perex.cz>
Cc: Takashi Iwai <tiwai@suse.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20190207232759.14553-2-daniel.vetter@ffwll.ch
While typing these I think doing an s/component_master/aggregate/
would be useful:
- it's shorter :-)
- I think component/aggregate is much more meaningful naming than
component/puppetmaster or something like that. At least to my
English ear "aggregate" emphasizes much more the "assemble a pile of
things into something bigger" aspect, and there's not really much
of a control hierarchy between aggregate and constituing components.
But that's way more than a quick doc typing exercise ...
Thanks to Ram for commenting on an initial draft of these docs.
v2: Review from Rafael:
- git add Documenation/driver-api/component.rst
- lots of polish to the wording + spelling fixes.
v3: Review from Russell:
- s/framework/helper
- clarify the documentation for component_match_add functions.
v4: Remove a few superflous "This".
Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: "C, Ramalingam" <ramalingam.c@intel.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Russell King <rmk+kernel@arm.linux.org.uk>
Cc: Rafael J. Wysocki <rafael@kernel.org>
Cc: Jaroslav Kysela <perex@perex.cz>
Cc: Takashi Iwai <tiwai@suse.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190207232759.14553-1-daniel.vetter@ffwll.ch
When unbinding the (IOMMU-enabled) R-Car SATA device on Salvator-XS
(R-Car H3 ES2.0), in preparation of rebinding against vfio-platform for
device pass-through for virtualization:
echo ee300000.sata > /sys/bus/platform/drivers/sata_rcar/unbind
the kernel crashes with:
Unable to handle kernel paging request at virtual address ffffffbf029ffffc
Mem abort info:
ESR = 0x96000006
Exception class = DABT (current EL), IL = 32 bits
SET = 0, FnV = 0
EA = 0, S1PTW = 0
Data abort info:
ISV = 0, ISS = 0x00000006
CM = 0, WnR = 0
swapper pgtable: 4k pages, 39-bit VAs, pgdp = 000000007e8c586c
[ffffffbf029ffffc] pgd=000000073bfc6003, pud=000000073bfc6003, pmd=0000000000000000
Internal error: Oops: 96000006 [#1] SMP
Modules linked in:
CPU: 0 PID: 1098 Comm: bash Not tainted 5.0.0-rc5-salvator-x-00452-g37596f884f4318ef #287
Hardware name: Renesas Salvator-X 2nd version board based on r8a7795 ES2.0+ (DT)
pstate: 60400005 (nZCv daif +PAN -UAO)
pc : __free_pages+0x8/0x58
lr : __dma_direct_free_pages+0x50/0x5c
sp : ffffff801268baa0
x29: ffffff801268baa0 x28: 0000000000000000
x27: ffffffc6f9c60bf0 x26: ffffffc6f9c60bf0
x25: ffffffc6f9c60810 x24: 0000000000000000
x23: 00000000fffff000 x22: ffffff8012145000
x21: 0000000000000800 x20: ffffffbf029fffc8
x19: 0000000000000000 x18: ffffffc6f86c42c8
x17: 0000000000000000 x16: 0000000000000070
x15: 0000000000000003 x14: 0000000000000000
x13: ffffff801103d7f8 x12: 0000000000000028
x11: ffffff8011117604 x10: 0000000000009ad8
x9 : ffffff80110126d0 x8 : ffffffc6f7563000
x7 : 6b6b6b6b6b6b6b6b x6 : 0000000000000018
x5 : ffffff8011cf3cc8 x4 : 0000000000004000
x3 : 0000000000080000 x2 : 0000000000000001
x1 : 0000000000000000 x0 : ffffffbf029fffc8
Process bash (pid: 1098, stack limit = 0x00000000c38e3e32)
Call trace:
__free_pages+0x8/0x58
__dma_direct_free_pages+0x50/0x5c
arch_dma_free+0x1c/0x98
dma_direct_free+0x14/0x24
dma_free_attrs+0x9c/0xdc
dmam_release+0x18/0x20
release_nodes+0x25c/0x28c
devres_release_all+0x48/0x4c
device_release_driver_internal+0x184/0x1f0
device_release_driver+0x14/0x1c
unbind_store+0x70/0xb8
drv_attr_store+0x24/0x34
sysfs_kf_write+0x4c/0x64
kernfs_fop_write+0x154/0x1c4
__vfs_write+0x34/0x164
vfs_write+0xb4/0x16c
ksys_write+0x5c/0xbc
__arm64_sys_write+0x14/0x1c
el0_svc_common+0x98/0x114
el0_svc_handler+0x1c/0x24
el0_svc+0x8/0xc
Code: d51b4234 17fffffa a9bf7bfd 910003fd (b9403404)
---[ end trace 8c564cdd3a1a840f ]---
While I've bisected this to commit e8e683ae9a736407 ("iommu/of: Fix
probe-deferral"), and reverting that commit on post-v5.0-rc4 kernels
does fix the problem, this turned out to be a red herring.
On arm64, arch_teardown_dma_ops() resets dev->dma_ops to NULL.
Hence if a driver has used a managed DMA allocation API, the allocated
DMA memory will be freed using the direct DMA ops, while it may have
been allocated using a custom DMA ops (iommu_dma_ops in this case).
Fix this by reversing the order of the calls to devres_release_all() and
arch_teardown_dma_ops().
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Acked-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: stable <stable@vger.kernel.org>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If the target device has any suppliers, as reflected by device links
to them, __pm_runtime_set_status() does not take them into account,
which is not consistent with the other parts of the PM-runtime
framework and may lead to programming mistakes.
Modify __pm_runtime_set_status() to take suppliers into account by
activating them upfront if the new status is RPM_ACTIVE and
deactivating them on exit if the new status is RPM_SUSPENDED.
If the activation of one of the suppliers fails, the new status
will be RPM_SUSPENDED and the (remaining) suppliers will be
deactivated on exit (the child count of the device's parent
will be dropped too then).
Of course, adding device links locking to __pm_runtime_set_status()
means that it cannot be run fron interrupt context, so make it use
spin_lock_irq() and spin_unlock_irq() instead of spin_lock_irqsave()
and spin_unlock_irqrestore(), respectively.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Update the accounting_timestamp field only when PM runtime is enabled
and don't forget to account the last state before disabling it.
Suggested-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
[ rjw: Minor cleanups ]
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Similar to what happened whith autosuspend, a deadlock has been
reported with PM-runtime accounting in the call path:
change_clocksource
...
write_seqcount_begin
...
timekeeping_update
...
sh_cmt_clocksource_enable
...
rpm_resume
update_pm_runtime_accounting
ktime_get
do
read_seqcount_begin
while read_seqcount_retry
....
write_seqcount_end
Make PM-runtime accounting use ktime_get_mono_fast_ns() to avoid this
problem.
With ktime_get_mono_fast_ns(), the timestamp is not guaranteed to be
monotonic across an update of timekeeping and as a result time can go
backward. Add a test to skip accounting for such situation which should
stay exceptional.
Fixes: a08c2a5a3194 ("PM-runtime: Replace jiffies-based accounting with ktime-based accounting")
Reported-by: Biju Das <biju.das@bp.renesas.com>
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
[ rjw: Subject, changelog, comment cleanup ]
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
pm_runtime_autosuspend_expiration calls ktime_get_mono_fast_ns()
even when its returned value may be unused. Therefore get the
current time later and remove gotos while there.
Signed-off-by: Ladislav Michl <ladis@linux-mips.org>
Acked-by: Tony Lindgren <tony@atomide.com>
Acked-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Add a new device link flag, DL_FLAG_AUTOPROBE_CONSUMER, to request the
driver core to probe for a consumer driver automatically after binding
a driver to the supplier device on a persistent managed device link.
As unbinding the supplier driver on a managed device link causes the
consumer driver to be detached from its device automatically, this
flag provides a complementary mechanism which is needed to address
some "composite device" use cases.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Even though stateful device links are managed by the driver core in
principle, their creators are allowed and sometimes even expected
to drop references to them via device_link_del() or
device_link_remove(), but that doesn't really play well with the
"persistent" link concept.
If "persistent" managed device links are created from driver
probe callbacks, device_link_add() called to do that will take a
new reference on the link each time the callback runs and those
references will never be dropped, which kind of isn't nice.
This issues arises because of the link reference counting carried
out by device_link_add() for existing links, but that is only done to
avoid deleting device links that may still be necessary, which
shouldn't be a concern for managed (stateful) links. These device
links are managed by the driver core and whoever creates one of them
will need it at least as long as until the consumer driver is detached
from its device and deleting it may be left to the driver core just
fine.
For this reason, rework device_link_add() to apply the reference
counting to stateless links only and make device_link_del() and
device_link_remove() drop references to stateless links only too.
After this change, if called to add a stateful device link for
a consumer-supplier pair for which a stateful device link is
present already, device_link_add() will return the existing link
without incrementing its reference counter. Accordingly,
device_link_del() and device_link_remove() will WARN() and do
nothing when called to drop a reference to a stateful link. Thus,
effectively, all stateful device links will be owned by the driver
core.
In addition, clean up the handling of the link management flags,
DL_FLAG_AUTOREMOVE_CONSUMER and DL_FLAG_AUTOREMOVE_SUPPLIER, so that
(a) they are never set at the same time and (b) if device_link_add()
is called for a consumer-supplier pair with an existing stateful link
between them, the flags of that link will be combined with the flags
passed to device_link_add() to ensure that the life time of the link
is sufficient for all of the callers of device_link_add() for the
same consumer-supplier pair.
Update the device_link_add() kerneldoc comment to reflect the
above changes.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Calling rpm_put_suppliers() from pm_runtime_drop_link() is excessive
as it affects all suppliers of the consumer device and not just the
one pointed to by the device link being dropped. Worst case it may
cause the consumer device to stop working unexpectedly. Moreover, in
principle it is racy with respect to runtime PM of the consumer
device.
To avoid these problems drop runtime PM references on the particular
supplier pointed to by the link in question only and do that after
the link has been dropped from the consumer device's list of links to
suppliers, which is in device_link_free().
Fixes: a0504aecba76 ("PM / runtime: Drop usage count for suppliers at device link removal")
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Currently, it is not valid to add a device link from a consumer
driver ->probe callback to a supplier that is still probing too, but
generally this is a valid use case. For example, if the consumer has
just acquired a resource that can only be available if the supplier
is functional, adding a device link to that supplier right away
should be safe (and even desirable arguably), but device_link_add()
doesn't handle that case correctly and the initial state of the link
created by it is wrong then.
To address this problem, change the initial state of device links
added between a probing supplier and a probing consumer to
DL_STATE_CONSUMER_PROBE and update device_links_driver_bound() to
skip such links on the supplier side.
With this change, if the supplier probe completes first,
device_links_driver_bound() called for it will skip the link state
update and when it is called for the consumer, the link state will
be updated to "active". In turn, if the consumer probe completes
first, device_links_driver_bound() called for it will change the
state of the link to "active" and when it is called for the
supplier, the link status update will be skipped.
However, in principle the supplier or consumer probe may still fail
after the link has been added, so modify device_links_no_driver() to
change device links in the "active" or "consumer probe" state to
"dormant" on the supplier side and update __device_links_no_driver()
to change the link state to "available" only if it is "consumer
probe" or "active".
Then, if the supplier probe fails first, the leftover link to the
probing consumer will become "dormant" and device_links_no_driver()
called for the consumer (when its probe fails) will clean it up.
In turn, if the consumer probe fails first, it will either drop the
link, or change its state to "available" and, in the latter case,
when device_links_no_driver() is called for the supplier, it will
update the link state to "dormant". [If the supplier probe fails,
but the consumer probe succeeds, which should not happen as long as
the consumer driver is correct, the link still will be around, but
it will be "dormant" until the supplier is probed again.]
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
After commit ead18c23c263 ("driver core: Introduce device links
reference counting"), if there is a link between the given supplier
and the given consumer already, device_link_add() will refcount it
and return it unconditionally without updating its flags. It is
possible, however, that the second (or any subsequent) caller of
device_link_add() for the same consumer-supplier pair will pass
DL_FLAG_PM_RUNTIME, possibly along with DL_FLAG_RPM_ACTIVE, in flags
to it and the existing link may not behave as expected then.
First, if DL_FLAG_PM_RUNTIME is not set in the existing link's flags
at all, it needs to be set like during the original initialization of
the link.
Second, if DL_FLAG_RPM_ACTIVE is passed to device_link_add() in flags
(in addition to DL_FLAG_PM_RUNTIME), the existing link should to be
updated to reflect the "active" runtime PM configuration of the
consumer-supplier pair and extra care must be taken here to avoid
possible destructive races with runtime PM of the consumer.
To that end, redefine the rpm_active field in struct device_link
as a refcount, initialize it to 1 and make rpm_resume() (for the
consumer) and device_link_add() increment it whenever they acquire
a runtime PM reference on the supplier device. Accordingly, make
rpm_suspend() (for the consumer) and pm_runtime_clean_up_links()
decrement it and drop runtime PM references to the supplier
device in a loop until rpm_active becones 1 again.
Fixes: ead18c23c263 ("driver core: Introduce device links reference counting")
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
It is incorrect to call pm_runtime_get_sync() under
device_links_write_lock(), because it may end up trying to take
device_links_read_lock() while resuming the target device and that
will deadlock in the non-SRCU case, so avoid that by resuming the
supplier device in device_link_add() before calling
device_links_write_lock().
Fixes: 21d5c57b3726 ("PM / runtime: Use device links")
Fixes: baa8809f6097 ("PM / runtime: Optimize the use of device links")
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
After commit ead18c23c263 ("driver core: Introduce device links
reference counting"), if there is a link between the given supplier
and the given consumer already, device_link_add() will refcount it
and return it unconditionally. However, if the flags passed to
it on the second (or any subsequent) attempt to create a device
link between the same consumer-supplier pair are not compatible with
the existing link's flags, that is incorrect.
First off, if the existing link is stateless and the next caller of
device_link_add() for the same consumer-supplier pair wants a
stateful one, or the other way around, the existing link cannot be
returned, because it will not match the expected behavior, so make
device_link_add() dump the stack and return NULL in that case.
Moreover, if the DL_FLAG_AUTOREMOVE_CONSUMER flag is passed to
device_link_add(), its caller will expect its reference to the link
to be dropped automatically on consumer driver removal, which will
not happen if that flag is not set in the link's flags (and
analogously for DL_FLAG_AUTOREMOVE_SUPPLIER). For this reason, make
device_link_add() update the existing link's flags accordingly
before returning it to the caller.
Fixes: ead18c23c263 ("driver core: Introduce device links reference counting")
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Change the list walk in device_links_driver_cleanup() to a safe one
to avoid use-after-free when dropping a link from the list during the
walk.
Also, while at it, fix device_link_add() to refuse to create
stateless device links with DL_FLAG_AUTOREMOVE_SUPPLIER set, which is
an invalid combination (setting that flag means that the driver core
should manage the link, so it cannot be stateless), and extend the
kerneldoc comment of device_link_add() to cover the
DL_FLAG_AUTOREMOVE_SUPPLIER flag properly too.
Fixes: 1689cac5b32a ("driver core: Add flag to autoremove device link on supplier unbind")
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Silence warnings (triggered at W=1) by adding relevant __printf
attributes.
drivers/base/cpu.c:432:2: warning: function '__cpu_device_create' might be a candidate for 'gnu_printf' format attribute [-Wsuggest-attribute=format]
Signed-off-by: Mathieu Malaterre <malat@debian.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The current async_probe test code is only testing one device allocated
prior to driver load and only loading one device afterwards. Instead of
doing things this way it makes much more sense to load one device per CPU
in order to actually stress the async infrastructure. By doing this we
should see delays significantly increase in the event of devices being
serialized.
In addition I have updated the test to verify that we are trying to place
the work on the correct NUMA node when we are running in async mode. By
doing this we can verify the best possible outcome for device and driver
load times.
I have added a timeout value that is used to disable the sleep and instead
cause the probe routine to report an error indicating it timed out. By
doing this we limit the maximum runtime for the test to 20 seconds or less.
The last major change in this set is that I have gone through and tuned it
for handling the massive number of possible events that will be scheduled.
Instead of reporting the sleep for each individual device it is moved to
only being displayed if we enable debugging.
With this patch applied below are what a failing test and a passing test
should look like. I elided a few hundred lines in the failing test that
were duplicated since the system I was testing on had a massive number of
CPU cores:
-- Failing --
[ 243.524697] test_async_driver_probe: registering first set of asynchronous devices...
[ 243.535625] test_async_driver_probe: registering asynchronous driver...
[ 243.543038] test_async_driver_probe: registration took 0 msecs
[ 243.549559] test_async_driver_probe: registering second set of asynchronous devices...
[ 243.568350] platform test_async_driver.447: registration took 9 msecs
[ 243.575544] test_async_driver_probe: registering first synchronous device...
[ 243.583454] test_async_driver_probe: registering synchronous driver...
[ 248.825920] test_async_driver_probe: registration took 5235 msecs
[ 248.825922] test_async_driver_probe: registering second synchronous device...
[ 248.825928] test_async_driver test_async_driver.443: NUMA node mismatch 3 != 1
[ 248.825932] test_async_driver test_async_driver.445: NUMA node mismatch 3 != 1
[ 248.825935] test_async_driver test_async_driver.446: NUMA node mismatch 3 != 1
[ 248.825939] test_async_driver test_async_driver.440: NUMA node mismatch 3 != 1
[ 248.825943] test_async_driver test_async_driver.441: NUMA node mismatch 3 != 1
...
[ 248.827150] test_async_driver test_async_driver.229: NUMA node mismatch 0 != 1
[ 248.827158] test_async_driver test_async_driver.228: NUMA node mismatch 0 != 1
[ 248.827220] test_async_driver test_async_driver.281: NUMA node mismatch 2 != 1
[ 248.827229] test_async_driver test_async_driver.282: NUMA node mismatch 2 != 1
[ 248.827240] test_async_driver test_async_driver.280: NUMA node mismatch 2 != 1
[ 253.945834] test_async_driver test_async_driver.1: NUMA node mismatch 0 != 1
[ 253.945878] test_sync_driver test_sync_driver.1: registration took 5119 msecs
[ 253.961693] test_async_driver_probe: async events still pending, forcing timeout and synchronize
[ 259.065839] test_async_driver test_async_driver.2: NUMA node mismatch 0 != 1
[ 259.073786] test_async_driver test_async_driver.3: async probe took too long
[ 259.081669] test_async_driver test_async_driver.3: NUMA node mismatch 0 != 1
[ 259.089569] test_async_driver test_async_driver.4: async probe took too long
[ 259.097451] test_async_driver test_async_driver.4: NUMA node mismatch 0 != 1
[ 259.105338] test_async_driver test_async_driver.5: async probe took too long
[ 259.113204] test_async_driver test_async_driver.5: NUMA node mismatch 0 != 1
[ 259.121089] test_async_driver test_async_driver.6: async probe took too long
[ 259.128961] test_async_driver test_async_driver.6: NUMA node mismatch 0 != 1
[ 259.136850] test_async_driver test_async_driver.7: async probe took too long
...
[ 262.124062] test_async_driver test_async_driver.221: async probe took too long
[ 262.132130] test_async_driver test_async_driver.221: NUMA node mismatch 3 != 1
[ 262.140206] test_async_driver test_async_driver.222: async probe took too long
[ 262.148277] test_async_driver test_async_driver.222: NUMA node mismatch 3 != 1
[ 262.156351] test_async_driver test_async_driver.223: async probe took too long
[ 262.164419] test_async_driver test_async_driver.223: NUMA node mismatch 3 != 1
[ 262.172630] test_async_driver_probe: Test failed with 222 errors and 336 warnings
-- Passing --
[ 105.419247] test_async_driver_probe: registering first set of asynchronous devices...
[ 105.432040] test_async_driver_probe: registering asynchronous driver...
[ 105.439718] test_async_driver_probe: registration took 0 msecs
[ 105.446239] test_async_driver_probe: registering second set of asynchronous devices...
[ 105.477986] platform test_async_driver.447: registration took 22 msecs
[ 105.485276] test_async_driver_probe: registering first synchronous device...
[ 105.493169] test_async_driver_probe: registering synchronous driver...
[ 110.597981] test_async_driver_probe: registration took 5097 msecs
[ 110.604806] test_async_driver_probe: registering second synchronous device...
[ 115.707490] test_sync_driver test_sync_driver.1: registration took 5094 msecs
[ 115.715478] test_async_driver_probe: completed successfully
Signed-off-by: Alexander Duyck <alexander.h.duyck@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Use the device specific version of the async_schedule commands to defer
various tasks related to power management. By doing this we should see a
slight improvement in performance as any device that is sensitive to
latency/locality in the setup will now be initializing on the node closest
to the device.
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Alexander Duyck <alexander.h.duyck@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Call the asynchronous probe routines on a CPU local to the device node. By
doing this we should be able to improve our initialization time
significantly as we can avoid having to access the device from a remote
node which may introduce higher latency.
For example, in the case of initializing memory for NVDIMM this can have a
significant impact as initialing 3TB on remote node can take up to 39
seconds while initialing it on a local node only takes 23 seconds. It is
situations like this where we will see the biggest improvement.
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Alexander Duyck <alexander.h.duyck@linux.intel.com>
Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Probe devices asynchronously instead of the driver. This results in us
seeing the same behavior if the device is registered before the driver or
after. This way we can avoid serializing the initialization should the
driver not be loaded until after the devices have already been added.
The motivation behind this is that if we have a set of devices that
take a significant amount of time to load we can greatly reduce the time to
load by processing them in parallel instead of one at a time. In addition,
each device can exist on a different node so placing a single thread on one
CPU to initialize all of the devices for a given driver can result in poor
performance on a system with multiple nodes.
This approach can reduce the time needed to scan SCSI LUNs significantly.
The only way to realize that speedup is by enabling more concurrency which
is what is achieved with this patch.
To achieve this it was necessary to add a new member "async_driver" to the
device_private structure to store the driver pointer while we wait on the
deferred probe call.
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Alexander Duyck <alexander.h.duyck@linux.intel.com>
Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Try to consolidate all of the locking and unlocking of both the parent and
device when attaching or removing a driver from a given device.
To do that I first consolidated the lock pattern into two functions
__device_driver_lock and __device_driver_unlock. After doing that I then
created functions specific to attaching and detaching the driver while
acquiring these locks. By doing this I was able to reduce the number of
spots where we touch need_parent_lock from 12 down to 4.
This patch should produce no functional changes, it is meant to be a code
clean-up/consolidation only.
Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Alexander Duyck <alexander.h.duyck@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Add an additional bit flag to the device_private struct named "dead".
This additional flag provides a guarantee that when a device_del is
executed on a given interface an async worker will not attempt to attach
the driver following the earlier device_del call. Previously this
guarantee was not present and could result in the device_del call
attempting to remove a driver from an interface only to have the async
worker attempt to probe the driver later when it finally completes the
asynchronous probe call.
One additional change added was that I pulled the check for dev->driver
out of the __device_attach_driver call and instead placed it in the
__device_attach_async_helper call. This was motivated by the fact that the
only other caller of this, __device_attach, had already taken the
device_lock() and checked for dev->driver. Instead of testing for this
twice in this path it makes more sense to just consolidate the dev->dead
and dev->driver checks together into one set of checks.
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Alexander Duyck <alexander.h.duyck@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Replace jiffies-based accounting for runtime_active_time and
runtime_suspended_time with ktime-based accounting. This makes the
runtime debug counters inline with genpd and other PM subsytems which
use ktime-based accounting.
Timekeeping is initialized before driver_init(). It's only at that time
that PM-runtime can be enabled.
Signed-off-by: Thara Gopinath <thara.gopinath@linaro.org>
[switch from ktime to raw nsec]
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
