linux-toradex.git/kernel/power/qos.c, branch v3.10.78

PM / QoS: remove duplicate call to pm_qos_update_target

2015-03-18T12:22:28+00:00

In 3.10.y backport patch 1dba303727f52ea062580b0a9b3f0c3b462769cf,
the logic to call pm_qos_update_target was moved to __pm_qos_update_request.
However, the original code was left in function pm_qos_update_request.

Currently, if pm_qos_update_request is called where new_value !=
req->node.prio then pm_qos_update_target will be called twice in a row.
Once in pm_qos_update_request and then again in the following call to
_pm_qos_update_request.

Removing the left over code from pm_qos_update_request stops this second
call to pm_qos_update_target where the work of removing / re-adding the
new_value in the constraints list would be duplicated.

Signed-off-by: Michael Scott 
Signed-off-by: Greg Kroah-Hartman

PM / QoS: Fix workqueue deadlock when using pm_qos_update_request_timeout()

2013-08-20T15:43:05+00:00

commit 40fea92ffb5fa0ef26d10ae0fe5688bc8e61c791 upstream.

pm_qos_update_request_timeout() updates a qos and then schedules
a delayed work item to bring the qos back down to the default
after the timeout. When the work item runs, pm_qos_work_fn() will
call pm_qos_update_request() and deadlock because it tries to
cancel itself via cancel_delayed_work_sync(). Future callers of
that qos will also hang waiting to cancel the work that is
canceling itself. Let's extract the little bit of code that does
the real work of pm_qos_update_request() and call it from the
work function so that we don't deadlock.

Before ed1ac6e (PM: don't use [delayed_]work_pending()) this didn't
happen because the work function wouldn't try to cancel itself.

[backport to 3.10 - gregkh]

Signed-off-by: Stephen Boyd 
Reviewed-by: Tejun Heo 
Signed-off-by: Rafael J. Wysocki 
Signed-off-by: Greg Kroah-Hartman

PM: don't use [delayed_]work_pending()

2013-01-25T23:39:11+00:00

There's no need to test whether a (delayed) work item is pending
before queueing, flushing or cancelling it, so remove work_pending()
tests used in those cases.

Signed-off-by: Tejun Heo 
Signed-off-by: Rafael J. Wysocki

Merge branch 'pm-sleep'

2012-11-29T20:46:48+00:00

* pm-sleep:
  PM / Freezer: Fixup compile error of try_to_freeze_nowarn()
  driver core / PM: move the calling to device_pm_remove behind the calling to bus_remove_device
  PM / Hibernate: use rb_entry
  PM / sysfs: replace strict_str* with kstrto*

PM / sysfs: replace strict_str* with kstrto*

2012-11-14T23:37:08+00:00

Replace strict_strtoul() with kstrtoul() in pm_async_store() and
pm_qos_power_write().

[rjw: Modified subject and changelog.]

Signed-off-by: Daniel Walter 
Acked-by: Pavel Machek 
Signed-off-by: Rafael J. Wysocki

PM / QoS: Introduce request and constraint data types for PM QoS flags

2012-10-22T23:07:46+00:00

Introduce struct pm_qos_flags_request and struct pm_qos_flags
representing PM QoS flags request type and PM QoS flags constraint
type, respectively.  With these definitions the data structures
will be arranged so that the list member of a struct pm_qos_flags
object will contain the head of a list of struct pm_qos_flags_request
objects representing all of the "flags" requests present for the
given device.  Then, the effective_flags member of a struct
pm_qos_flags object will contain the bitwise OR of the flags members
of all the struct pm_qos_flags_request objects in the list.

Additionally, introduce helper function pm_qos_update_flags()
allowing the caller to manage the list of struct pm_qos_flags_request
pointed to by the list member of struct pm_qos_flags.

The flags are of type s32 so that the request's "value" field
is always of the same type regardless of what kind of request it
is (latency requests already have value fields of type s32).

Signed-off-by: Rafael J. Wysocki 
Reviewed-by: Jean Pihet 
Acked-by: mark gross

PM / QoS: Add return code to pm_qos_get_value function.

2012-09-07T19:35:21+00:00

pm_qos_get_value don't return a return code in all cases. It's sure that
anything interesting happend after BUG() but this prevent any compilation
warning.

[rjw: Chaneged the new return value to PM_QOS_DEFAULT_VALUE.]

Signed-off-by: Luis Gonzalez Fernandez 
Signed-off-by: Rafael J. Wysocki

PM / QoS: add pm_qos_update_request_timeout() API

2012-03-28T21:31:24+00:00

The new API, pm_qos_update_request_timeout() is to provide a timeout
with pm_qos_update_request.

For example, pm_qos_update_request_timeout(req, 100, 1000), means that
QoS request on req with value 100 will be active for 1000 microseconds.
After 1000 microseconds, the QoS request thru req is reset. If there
were another pm_qos_update_request(req, x) during the 1000 us, this
new request with value x will override as this is another request on the
same req handle. A new request on the same req handle will always
override the previous request whether it is the conventional request or
it is the new timeout request.

Signed-off-by: MyungJoo Ham 
Signed-off-by: Kyungmin Park 
Acked-by: Mark Gross 
Signed-off-by: Rafael J. Wysocki

PM / QoS: Simplify PM QoS expansion/merge

2012-01-29T19:39:25+00:00

 - Replace class ID #define with enumeration
 - Loop through PM QoS objects during initialization (rather than
   initializing them one-by-one)

Signed-off-by: Alex Frid 
Reviewed-by: Antti Miettinen 
Reviewed-by: Diwakar Tundlam 
Reviewed-by: Scott Williams 
Reviewed-by: Yu-Huan Hsu 
Acked-by: markgross 
Signed-off-by: Rafael J. Wysocki

PM / QoS: Set cpu_dma_pm_qos->name

2011-11-07T22:02:24+00:00

Since commit 4a31a334, the name of this misc device is not initialized,
which leads to a funny device named /dev/(null) being created and
/proc/misc containing an entry with just a number but no name. The latter
leads to complaints by cryptsetup, which caused me to investigate this
matter.

Signed-off-by: Dominik Brodowski 
Signed-off-by: Rafael J. Wysocki