2019-05-27 08:55:01 +02:00
|
|
|
|
2019-06-17 18:16:52 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2010-12-21 23:49:31 +00:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 17:04:11 +09:00
|
|
|
|
2011-05-24 08:12:40 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-12-21 12:21:52 +00:00
|
|
|
|
2014-06-29 17:55:54 +01:00
|
|
|
|
2011-11-18 16:47:20 +05:30
|
|
|
|
2023-10-26 16:48:21 +02:00
|
|
|
|
2012-04-15 11:16:05 +01:00
|
|
|
|
2011-11-18 16:47:20 +05:30
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2011-07-17 16:28:23 -04:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2010-11-10 14:38:30 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2010-02-12 10:18:08 +00:00
|
|
|
|
2013-09-11 13:15:40 +01:00
|
|
|
|
2023-12-05 16:22:04 +05:30
|
|
|
|
2010-02-12 10:18:08 +00:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2013-02-18 06:50:39 +00:00
|
|
|
|
2013-10-15 20:14:20 +01:00
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
2010-12-21 23:30:07 +00:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
regulator: core: reresolve unresolved supplies when available
When a regulator A and its supply B are provided by different devices,
the driver implementing B might be last to probe (with A still
pending resolution of its supply B). While we try to resolve all
pending supplies for all regulators (including A) during
regulator_register() of B via regulator_register_resolve_supply(),
supply resolution will still not work for A as the driver for B hasn't
finished binding to the PMIC device corresponding to B at that stage
yet. The regulator core explicitly only allows supplies from other
devices to be used once the relevant driver has fully bound, mainly to
avoid having to deal with cases where B itself might -EPROBE_DEFER.
In this case, A's supply will only be resolved as part of the core's
regulator_init_complete_work_function(), which currently is scheduled
to run after 30s. This was added as a work-around in
commit 3827b64dba27 ("regulator: core: Resolve supplies before
disabling unused regulators") to cover this situation.
There are two problems with that approach:
* it potentially runs long after all our consumers have probed
* an upcoming change will allow regulator_register() to complete
successfully even when required supplies (e.g. due to always-on or
boot-on) are missing at register time, deferring full configuration
of the regulator (and usability by consumers, i.e. usually consumer
probe) until the supply becomes available.
Resolving supplies in the late work func can therefore make it
impossible for consumers to probe at all, as the driver core will not
know to reprobe consumers when supplies have resolved.
We could schedule an earlier work to try to resolve supplies sooner,
but that'd be racy as consumers of A might try to probe before A's
supply gets fully resolved via this extra work.
Instead, add a very simple regulator bus and add a dummy device with a
corresponding driver to it for each regulator that is missing its
supply during regulator_register(). This way, the driver core will call
our bus' probe whenever a new (regulator) device was successfully
bound, allowing us to retry resolving the supply during (our bus) probe
and to bind this dummy device if successful. In turn this means the
driver core will see a newly bound device and retry probing of all
pending consumers, if any.
With that in place, we can avoid walking the full list of all known
regulators to try resolve missing supplies during regulator_register(),
as the driver core will invoke the bus probe for regulators that are
still pending their supplies. We can also drop the code trying to
resolve supplies one last time before unused regulators get disabled,
as all supplies should have resolved at that point in time, and if they
haven't then there's no point in trying again, as the outcome won't
change.
Note: We can not reuse the existing struct device created for each
rail, as a device can not be part of a class and a bus simultaneously.
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-7-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:43 +00:00
|
|
|
|
|
|
|
|
|
2010-12-21 23:49:31 +00:00
|
|
|
|
|
|
|
|
|
2008-12-31 12:52:40 +00:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-06-17 17:56:39 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2013-02-18 06:50:39 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-06-29 17:55:54 +01:00
|
|
|
|
2013-02-18 06:50:39 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2013-10-15 20:14:20 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-10-01 12:56:49 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core: don't fail regulator_register() with missing required supply
Since commit 98e48cd9283d ("regulator: core: resolve supply for
boot-on/always-on regulators"), the regulator core returns
-EPROBE_DEFER if a supply can not be resolved at regulator_register()
time due to set_machine_constraints() requiring that supply (e.g.
because of always-on or boot-on).
In some hardware designs, multiple PMICs are used where individual
rails of each act as supplies for rails of the other, and vice-versa.
In such a design no PMIC driver can probe when registering one top-
level regulator device (as is common practice for almost all regulator
drivers in Linux) since that commit. Supplies are only considered when
their driver has fully bound, but because in a design like the above
two drivers / devices depend on each other, neither will have fully
bound while the other probes. The Google Pixel 6 and 6 Pro (oriole and
raven) are examples of such a design.
One way to make this work would be to register each rail as an
individual device, rather than just one top-level regulator device.
Then, fw-devlink and Linux' driver core could do their usual handling
of deferred device probe as each rail would be probed individually.
This approach was dismissed in [1] as each regulator driver would have
to take care of this itself.
Alternatively, we can change the regulator core to not fail
regulator_register() if a rail's required supply can not be resolved
while keeping the intended change from above mentioned commit, and
instead retry whenever a new rail is registered. This commit implements
such an approach:
If set_machine_constraints() requests probe deferral,
regulator_register() still succeeds and we retry setting
constraints as part of regulator_resolve_supply().
We still do not enable the regulator or allow consumers to use it
until constraints have been set (including resolution of the
supply) to prevent enabling of a regulator before its supply.
With this change, we keep track of regulators with missing required
supplies and can therefore try to resolve them again and try to set
the constraints again once more regulators become available.
Care has to be taken to not allow consumers to use regulators that
haven't had their constraints set yet. regulator_get() ensures that
and now returns -EPROBE_DEFER in that case.
The implementation is straight-forward, thanks to our newly introduced
regulator-bus. Locking in regulator_resolve_supply() has to be done
carefully, as a combination of regulator_(un)lock_two() and
regulator_(un)lock_dependent() is needed. The reason is that
set_machine_constraints() might call regulator_enable() which needs
rdev and all its dependents locked, but everything else requires to
only have rdev and its supply locked.
Link: https://lore.kernel.org/all/aRn_-o-vie_QoDXD@sirena.co.uk/ [1]
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-8-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:44 +00:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
2022-05-03 23:52:49 -07:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2014-08-28 12:36:04 -07:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2010-12-12 14:25:50 +00:00
|
|
|
|
|
|
|
|
|
regulator: core: Add voltage balancing mechanism
On Odroid XU3/4 and other Exynos5422 based boards there is a case, that
different devices on the board are supplied by different regulators
with non-fixed voltages. If one of these devices temporarily requires
higher voltage, there might occur a situation that the spread between
two devices' voltages is so high, that there is a risk of changing
'high' and 'low' states on the interconnection between devices powered
by those regulators.
Introduce new function regulator_balance_voltage(), which
keeps max_spread constraint fulfilled between a group of coupled
regulators. It should be called if a regulator changes its
voltage or after disabling or enabling. Disabled regulators should
follow changes of the enabled ones, but their consumers' demands
shouldn't be taken into account while calculating voltage of other
coupled regulators.
Find voltages, which are closest to suiting all the consumers' demands,
while fulfilling max_spread constraint, keeping the following rules:
- if one regulator is about to rise its voltage, rise others
voltages in order to keep the max_spread
- if a regulator, which has caused rising other regulators, is
lowered, lower other regulators if possible
- if one regulator is about to lower its voltage, but it hasn't caused
rising other regulators, change its voltage so that it doesn't break the
max_spread
Change regulators' voltages step by step, keeping max_spread constraint
fulfilled all the time. Function regulator_get_optimal_voltage()
should find the best possible change for the regulator, which doesn't
break max_spread constraint. In function regulator_balance_voltage()
optimize number of steps by finding highest voltage difference on
each iteration.
If a regulator, which is about to change its voltage, is not coupled,
method regulator_get_optimal_voltage() should simply return the lowest
voltage fulfilling consumers' demands.
Coupling should be checked only if the system is in PM_SUSPEND_ON state.
Signed-off-by: Maciej Purski <m.purski@samsung.com>
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2018-10-05 18:36:28 +03:00
|
|
|
|
|
|
|
|
|
2011-06-09 16:22:22 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-07-15 21:20:51 -07:00
|
|
|
|
2015-07-15 16:10:29 +02:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2019-06-24 00:08:32 +03:00
|
|
|
|
2009-10-22 16:31:32 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-03 08:41:37 +03:00
|
|
|
|
2009-10-22 16:31:32 +01:00
|
|
|
|
2013-11-27 16:13:10 +00:00
|
|
|
|
|
|
|
|
|
2013-11-27 16:22:53 +00:00
|
|
|
|
2013-11-27 16:13:10 +00:00
|
|
|
|
|
|
|
|
|
2016-04-23 15:11:05 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
kernel/locking: Add context to ww_mutex_trylock()
i915 will soon gain an eviction path that trylock a whole lot of locks
for eviction, getting dmesg failures like below:
BUG: MAX_LOCK_DEPTH too low!
turning off the locking correctness validator.
depth: 48 max: 48!
48 locks held by i915_selftest/5776:
#0: ffff888101a79240 (&dev->mutex){....}-{3:3}, at: __driver_attach+0x88/0x160
#1: ffffc900009778c0 (reservation_ww_class_acquire){+.+.}-{0:0}, at: i915_vma_pin.constprop.63+0x39/0x1b0 [i915]
#2: ffff88800cf74de8 (reservation_ww_class_mutex){+.+.}-{3:3}, at: i915_vma_pin.constprop.63+0x5f/0x1b0 [i915]
#3: ffff88810c7f9e38 (&vm->mutex/1){+.+.}-{3:3}, at: i915_vma_pin_ww+0x1c4/0x9d0 [i915]
#4: ffff88810bad5768 (reservation_ww_class_mutex){+.+.}-{3:3}, at: i915_gem_evict_something+0x110/0x860 [i915]
#5: ffff88810bad60e8 (reservation_ww_class_mutex){+.+.}-{3:3}, at: i915_gem_evict_something+0x110/0x860 [i915]
...
#46: ffff88811964d768 (reservation_ww_class_mutex){+.+.}-{3:3}, at: i915_gem_evict_something+0x110/0x860 [i915]
#47: ffff88811964e0e8 (reservation_ww_class_mutex){+.+.}-{3:3}, at: i915_gem_evict_something+0x110/0x860 [i915]
INFO: lockdep is turned off.
Fixing eviction to nest into ww_class_acquire is a high priority, but
it requires a rework of the entire driver, which can only be done one
step at a time.
As an intermediate solution, add an acquire context to
ww_mutex_trylock, which allows us to do proper nesting annotations on
the trylocks, making the above lockdep splat disappear.
This is also useful in regulator_lock_nested, which may avoid dropping
regulator_nesting_mutex in the uncontended path, so use it there.
TTM may be another user for this, where we could lock a buffer in a
fastpath with list locks held, without dropping all locks we hold.
[peterz: rework actual ww_mutex_trylock() implementations]
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/YUBGPdDDjKlxAuXJ@hirez.programming.kicks-ass.net
2021-09-09 11:32:18 +02:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-19 23:28:25 +02:00
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-19 23:28:25 +02:00
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
|
|
|
|
|
regulator: core: Avoid lockdep reports when resolving supplies
An automated bot told me that there was a potential lockdep problem
with regulators. This was on the chromeos-5.15 kernel, but I see
nothing that would be different downstream compared to upstream. The
bot said:
============================================
WARNING: possible recursive locking detected
5.15.104-lockdep-17461-gc1e499ed6604 #1 Not tainted
--------------------------------------------
kworker/u16:4/115 is trying to acquire lock:
ffffff8083110170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: create_regulator+0x398/0x7ec
but task is already holding lock:
ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(regulator_ww_class_mutex);
lock(regulator_ww_class_mutex);
*** DEADLOCK ***
May be due to missing lock nesting notation
4 locks held by kworker/u16:4/115:
#0: ffffff808006a948 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_one_work+0x520/0x1348
#1: ffffffc00e0a7cc0 ((work_completion)(&entry->work)){+.+.}-{0:0}, at: process_one_work+0x55c/0x1348
#2: ffffff80828a2260 (&dev->mutex){....}-{3:3}, at: __device_attach_async_helper+0xd0/0x2a4
#3: ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
stack backtrace:
CPU: 2 PID: 115 Comm: kworker/u16:4 Not tainted 5.15.104-lockdep-17461-gc1e499ed6604 #1 9292e52fa83c0e23762b2b3aa1bacf5787a4d5da
Hardware name: Google Quackingstick (rev0+) (DT)
Workqueue: events_unbound async_run_entry_fn
Call trace:
dump_backtrace+0x0/0x4ec
show_stack+0x34/0x50
dump_stack_lvl+0xdc/0x11c
dump_stack+0x1c/0x48
__lock_acquire+0x16d4/0x6c74
lock_acquire+0x208/0x750
__mutex_lock_common+0x11c/0x11f8
ww_mutex_lock+0xc0/0x440
create_regulator+0x398/0x7ec
regulator_resolve_supply+0x654/0x7c4
regulator_register_resolve_supply+0x30/0x120
class_for_each_device+0x1b8/0x230
regulator_register+0x17a4/0x1f40
devm_regulator_register+0x60/0xd0
reg_fixed_voltage_probe+0x728/0xaec
platform_probe+0x150/0x1c8
really_probe+0x274/0xa20
__driver_probe_device+0x1dc/0x3f4
driver_probe_device+0x78/0x1c0
__device_attach_driver+0x1ac/0x2c8
bus_for_each_drv+0x11c/0x190
__device_attach_async_helper+0x1e4/0x2a4
async_run_entry_fn+0xa0/0x3ac
process_one_work+0x638/0x1348
worker_thread+0x4a8/0x9c4
kthread+0x2e4/0x3a0
ret_from_fork+0x10/0x20
The problem was first reported soon after we made many of the
regulators probe asynchronously, though nothing I've seen implies that
the problems couldn't have also happened even without that.
I haven't personally been able to reproduce the lockdep issue, but the
issue does look somewhat legitimate. Specifically, it looks like in
regulator_resolve_supply() we are holding a "rdev" lock while calling
set_supply() -> create_regulator() which grabs the lock of a
_different_ "rdev" (the one for our supply). This is not necessarily
safe from a lockdep perspective since there is no documented ordering
between these two locks.
In reality, we should always be locking a regulator before the
supplying regulator, so I don't expect there to be any real deadlocks
in practice. However, the regulator framework in general doesn't
express this to lockdep.
Let's fix the issue by simply grabbing the two locks involved in the
same way we grab multiple locks elsewhere in the regulator framework:
using the "wound/wait" mechanisms.
Fixes: eaa7995c529b ("regulator: core: avoid regulator_resolve_supply() race condition")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20230329143317.RFC.v2.2.I30d8e1ca10cfbe5403884cdd192253a2e063eb9e@changeid
Signed-off-by: Mark Brown <broonie@kernel.org>
2023-03-29 14:33:54 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-13 17:34:17 -07:00
|
|
|
|
regulator: core: Avoid lockdep reports when resolving supplies
An automated bot told me that there was a potential lockdep problem
with regulators. This was on the chromeos-5.15 kernel, but I see
nothing that would be different downstream compared to upstream. The
bot said:
============================================
WARNING: possible recursive locking detected
5.15.104-lockdep-17461-gc1e499ed6604 #1 Not tainted
--------------------------------------------
kworker/u16:4/115 is trying to acquire lock:
ffffff8083110170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: create_regulator+0x398/0x7ec
but task is already holding lock:
ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(regulator_ww_class_mutex);
lock(regulator_ww_class_mutex);
*** DEADLOCK ***
May be due to missing lock nesting notation
4 locks held by kworker/u16:4/115:
#0: ffffff808006a948 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_one_work+0x520/0x1348
#1: ffffffc00e0a7cc0 ((work_completion)(&entry->work)){+.+.}-{0:0}, at: process_one_work+0x55c/0x1348
#2: ffffff80828a2260 (&dev->mutex){....}-{3:3}, at: __device_attach_async_helper+0xd0/0x2a4
#3: ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
stack backtrace:
CPU: 2 PID: 115 Comm: kworker/u16:4 Not tainted 5.15.104-lockdep-17461-gc1e499ed6604 #1 9292e52fa83c0e23762b2b3aa1bacf5787a4d5da
Hardware name: Google Quackingstick (rev0+) (DT)
Workqueue: events_unbound async_run_entry_fn
Call trace:
dump_backtrace+0x0/0x4ec
show_stack+0x34/0x50
dump_stack_lvl+0xdc/0x11c
dump_stack+0x1c/0x48
__lock_acquire+0x16d4/0x6c74
lock_acquire+0x208/0x750
__mutex_lock_common+0x11c/0x11f8
ww_mutex_lock+0xc0/0x440
create_regulator+0x398/0x7ec
regulator_resolve_supply+0x654/0x7c4
regulator_register_resolve_supply+0x30/0x120
class_for_each_device+0x1b8/0x230
regulator_register+0x17a4/0x1f40
devm_regulator_register+0x60/0xd0
reg_fixed_voltage_probe+0x728/0xaec
platform_probe+0x150/0x1c8
really_probe+0x274/0xa20
__driver_probe_device+0x1dc/0x3f4
driver_probe_device+0x78/0x1c0
__device_attach_driver+0x1ac/0x2c8
bus_for_each_drv+0x11c/0x190
__device_attach_async_helper+0x1e4/0x2a4
async_run_entry_fn+0xa0/0x3ac
process_one_work+0x638/0x1348
worker_thread+0x4a8/0x9c4
kthread+0x2e4/0x3a0
ret_from_fork+0x10/0x20
The problem was first reported soon after we made many of the
regulators probe asynchronously, though nothing I've seen implies that
the problems couldn't have also happened even without that.
I haven't personally been able to reproduce the lockdep issue, but the
issue does look somewhat legitimate. Specifically, it looks like in
regulator_resolve_supply() we are holding a "rdev" lock while calling
set_supply() -> create_regulator() which grabs the lock of a
_different_ "rdev" (the one for our supply). This is not necessarily
safe from a lockdep perspective since there is no documented ordering
between these two locks.
In reality, we should always be locking a regulator before the
supplying regulator, so I don't expect there to be any real deadlocks
in practice. However, the regulator framework in general doesn't
express this to lockdep.
Let's fix the issue by simply grabbing the two locks involved in the
same way we grab multiple locks elsewhere in the regulator framework:
using the "wound/wait" mechanisms.
Fixes: eaa7995c529b ("regulator: core: avoid regulator_resolve_supply() race condition")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20230329143317.RFC.v2.2.I30d8e1ca10cfbe5403884cdd192253a2e063eb9e@changeid
Signed-off-by: Mark Brown <broonie@kernel.org>
2023-03-29 14:33:54 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core: regulator_lock_two() should test for EDEADLK not EDEADLOCK
Compare against -EDEADLK, which is what ww_mutex_lock() actually
returns and what every other deadlock check in this file already uses.
Function regulator_lock_two() acquires two regulators via
regulator_lock_nested() -> ww_mutex_lock(). On contention,
ww_mutex_lock() returns -EDEADLK, which is the caller's signal to drop
the lock it holds and retry the acquisition in the canonical order.
However, regulator_lock_two() tests the return value against -EDEADLOCK
rather than -EDEADLK. On most architectures, EDEADLK and EDEADLOCK are
the same value, so the comparison happens to be correct and the bug is
invisible. But on MIPS, SPARC, and PowerPC, those two errors have
different values. The test is wrong: a genuine -EDEADLK backoff no
longer matches -EDEADLOCK, so instead of unlocking and retrying, the
code falls into WARN_ON(ret) and returns with only one of the two
regulators locked.
In practice, this is a bug only on MIPS, because the regulator core is
not built or used on the other two platforms.
In general, EDEADLK is preferred over EDEADLOCK for new code.
Fixes: cba6cfdc7c3f ("regulator: core: Avoid lockdep reports when resolving supplies")
Signed-off-by: Timur Tabi <ttabi@nvidia.com>
Link: https://patch.msgid.link/20260708235722.2953579-1-ttabi@nvidia.com
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-07-08 18:57:22 -05:00
|
|
|
|
regulator: core: Avoid lockdep reports when resolving supplies
An automated bot told me that there was a potential lockdep problem
with regulators. This was on the chromeos-5.15 kernel, but I see
nothing that would be different downstream compared to upstream. The
bot said:
============================================
WARNING: possible recursive locking detected
5.15.104-lockdep-17461-gc1e499ed6604 #1 Not tainted
--------------------------------------------
kworker/u16:4/115 is trying to acquire lock:
ffffff8083110170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: create_regulator+0x398/0x7ec
but task is already holding lock:
ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(regulator_ww_class_mutex);
lock(regulator_ww_class_mutex);
*** DEADLOCK ***
May be due to missing lock nesting notation
4 locks held by kworker/u16:4/115:
#0: ffffff808006a948 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_one_work+0x520/0x1348
#1: ffffffc00e0a7cc0 ((work_completion)(&entry->work)){+.+.}-{0:0}, at: process_one_work+0x55c/0x1348
#2: ffffff80828a2260 (&dev->mutex){....}-{3:3}, at: __device_attach_async_helper+0xd0/0x2a4
#3: ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
stack backtrace:
CPU: 2 PID: 115 Comm: kworker/u16:4 Not tainted 5.15.104-lockdep-17461-gc1e499ed6604 #1 9292e52fa83c0e23762b2b3aa1bacf5787a4d5da
Hardware name: Google Quackingstick (rev0+) (DT)
Workqueue: events_unbound async_run_entry_fn
Call trace:
dump_backtrace+0x0/0x4ec
show_stack+0x34/0x50
dump_stack_lvl+0xdc/0x11c
dump_stack+0x1c/0x48
__lock_acquire+0x16d4/0x6c74
lock_acquire+0x208/0x750
__mutex_lock_common+0x11c/0x11f8
ww_mutex_lock+0xc0/0x440
create_regulator+0x398/0x7ec
regulator_resolve_supply+0x654/0x7c4
regulator_register_resolve_supply+0x30/0x120
class_for_each_device+0x1b8/0x230
regulator_register+0x17a4/0x1f40
devm_regulator_register+0x60/0xd0
reg_fixed_voltage_probe+0x728/0xaec
platform_probe+0x150/0x1c8
really_probe+0x274/0xa20
__driver_probe_device+0x1dc/0x3f4
driver_probe_device+0x78/0x1c0
__device_attach_driver+0x1ac/0x2c8
bus_for_each_drv+0x11c/0x190
__device_attach_async_helper+0x1e4/0x2a4
async_run_entry_fn+0xa0/0x3ac
process_one_work+0x638/0x1348
worker_thread+0x4a8/0x9c4
kthread+0x2e4/0x3a0
ret_from_fork+0x10/0x20
The problem was first reported soon after we made many of the
regulators probe asynchronously, though nothing I've seen implies that
the problems couldn't have also happened even without that.
I haven't personally been able to reproduce the lockdep issue, but the
issue does look somewhat legitimate. Specifically, it looks like in
regulator_resolve_supply() we are holding a "rdev" lock while calling
set_supply() -> create_regulator() which grabs the lock of a
_different_ "rdev" (the one for our supply). This is not necessarily
safe from a lockdep perspective since there is no documented ordering
between these two locks.
In reality, we should always be locking a regulator before the
supplying regulator, so I don't expect there to be any real deadlocks
in practice. However, the regulator framework in general doesn't
express this to lockdep.
Let's fix the issue by simply grabbing the two locks involved in the
same way we grab multiple locks elsewhere in the regulator framework:
using the "wound/wait" mechanisms.
Fixes: eaa7995c529b ("regulator: core: avoid regulator_resolve_supply() race condition")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20230329143317.RFC.v2.2.I30d8e1ca10cfbe5403884cdd192253a2e063eb9e@changeid
Signed-off-by: Mark Brown <broonie@kernel.org>
2023-03-29 14:33:54 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-13 17:34:17 -07:00
|
|
|
|
|
|
|
|
|
regulator: core: Avoid lockdep reports when resolving supplies
An automated bot told me that there was a potential lockdep problem
with regulators. This was on the chromeos-5.15 kernel, but I see
nothing that would be different downstream compared to upstream. The
bot said:
============================================
WARNING: possible recursive locking detected
5.15.104-lockdep-17461-gc1e499ed6604 #1 Not tainted
--------------------------------------------
kworker/u16:4/115 is trying to acquire lock:
ffffff8083110170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: create_regulator+0x398/0x7ec
but task is already holding lock:
ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(regulator_ww_class_mutex);
lock(regulator_ww_class_mutex);
*** DEADLOCK ***
May be due to missing lock nesting notation
4 locks held by kworker/u16:4/115:
#0: ffffff808006a948 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_one_work+0x520/0x1348
#1: ffffffc00e0a7cc0 ((work_completion)(&entry->work)){+.+.}-{0:0}, at: process_one_work+0x55c/0x1348
#2: ffffff80828a2260 (&dev->mutex){....}-{3:3}, at: __device_attach_async_helper+0xd0/0x2a4
#3: ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
stack backtrace:
CPU: 2 PID: 115 Comm: kworker/u16:4 Not tainted 5.15.104-lockdep-17461-gc1e499ed6604 #1 9292e52fa83c0e23762b2b3aa1bacf5787a4d5da
Hardware name: Google Quackingstick (rev0+) (DT)
Workqueue: events_unbound async_run_entry_fn
Call trace:
dump_backtrace+0x0/0x4ec
show_stack+0x34/0x50
dump_stack_lvl+0xdc/0x11c
dump_stack+0x1c/0x48
__lock_acquire+0x16d4/0x6c74
lock_acquire+0x208/0x750
__mutex_lock_common+0x11c/0x11f8
ww_mutex_lock+0xc0/0x440
create_regulator+0x398/0x7ec
regulator_resolve_supply+0x654/0x7c4
regulator_register_resolve_supply+0x30/0x120
class_for_each_device+0x1b8/0x230
regulator_register+0x17a4/0x1f40
devm_regulator_register+0x60/0xd0
reg_fixed_voltage_probe+0x728/0xaec
platform_probe+0x150/0x1c8
really_probe+0x274/0xa20
__driver_probe_device+0x1dc/0x3f4
driver_probe_device+0x78/0x1c0
__device_attach_driver+0x1ac/0x2c8
bus_for_each_drv+0x11c/0x190
__device_attach_async_helper+0x1e4/0x2a4
async_run_entry_fn+0xa0/0x3ac
process_one_work+0x638/0x1348
worker_thread+0x4a8/0x9c4
kthread+0x2e4/0x3a0
ret_from_fork+0x10/0x20
The problem was first reported soon after we made many of the
regulators probe asynchronously, though nothing I've seen implies that
the problems couldn't have also happened even without that.
I haven't personally been able to reproduce the lockdep issue, but the
issue does look somewhat legitimate. Specifically, it looks like in
regulator_resolve_supply() we are holding a "rdev" lock while calling
set_supply() -> create_regulator() which grabs the lock of a
_different_ "rdev" (the one for our supply). This is not necessarily
safe from a lockdep perspective since there is no documented ordering
between these two locks.
In reality, we should always be locking a regulator before the
supplying regulator, so I don't expect there to be any real deadlocks
in practice. However, the regulator framework in general doesn't
express this to lockdep.
Let's fix the issue by simply grabbing the two locks involved in the
same way we grab multiple locks elsewhere in the regulator framework:
using the "wound/wait" mechanisms.
Fixes: eaa7995c529b ("regulator: core: avoid regulator_resolve_supply() race condition")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20230329143317.RFC.v2.2.I30d8e1ca10cfbe5403884cdd192253a2e063eb9e@changeid
Signed-off-by: Mark Brown <broonie@kernel.org>
2023-03-29 14:33:54 -07:00
|
|
|
|
2023-04-13 17:34:17 -07:00
|
|
|
|
regulator: core: Avoid lockdep reports when resolving supplies
An automated bot told me that there was a potential lockdep problem
with regulators. This was on the chromeos-5.15 kernel, but I see
nothing that would be different downstream compared to upstream. The
bot said:
============================================
WARNING: possible recursive locking detected
5.15.104-lockdep-17461-gc1e499ed6604 #1 Not tainted
--------------------------------------------
kworker/u16:4/115 is trying to acquire lock:
ffffff8083110170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: create_regulator+0x398/0x7ec
but task is already holding lock:
ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(regulator_ww_class_mutex);
lock(regulator_ww_class_mutex);
*** DEADLOCK ***
May be due to missing lock nesting notation
4 locks held by kworker/u16:4/115:
#0: ffffff808006a948 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_one_work+0x520/0x1348
#1: ffffffc00e0a7cc0 ((work_completion)(&entry->work)){+.+.}-{0:0}, at: process_one_work+0x55c/0x1348
#2: ffffff80828a2260 (&dev->mutex){....}-{3:3}, at: __device_attach_async_helper+0xd0/0x2a4
#3: ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
stack backtrace:
CPU: 2 PID: 115 Comm: kworker/u16:4 Not tainted 5.15.104-lockdep-17461-gc1e499ed6604 #1 9292e52fa83c0e23762b2b3aa1bacf5787a4d5da
Hardware name: Google Quackingstick (rev0+) (DT)
Workqueue: events_unbound async_run_entry_fn
Call trace:
dump_backtrace+0x0/0x4ec
show_stack+0x34/0x50
dump_stack_lvl+0xdc/0x11c
dump_stack+0x1c/0x48
__lock_acquire+0x16d4/0x6c74
lock_acquire+0x208/0x750
__mutex_lock_common+0x11c/0x11f8
ww_mutex_lock+0xc0/0x440
create_regulator+0x398/0x7ec
regulator_resolve_supply+0x654/0x7c4
regulator_register_resolve_supply+0x30/0x120
class_for_each_device+0x1b8/0x230
regulator_register+0x17a4/0x1f40
devm_regulator_register+0x60/0xd0
reg_fixed_voltage_probe+0x728/0xaec
platform_probe+0x150/0x1c8
really_probe+0x274/0xa20
__driver_probe_device+0x1dc/0x3f4
driver_probe_device+0x78/0x1c0
__device_attach_driver+0x1ac/0x2c8
bus_for_each_drv+0x11c/0x190
__device_attach_async_helper+0x1e4/0x2a4
async_run_entry_fn+0xa0/0x3ac
process_one_work+0x638/0x1348
worker_thread+0x4a8/0x9c4
kthread+0x2e4/0x3a0
ret_from_fork+0x10/0x20
The problem was first reported soon after we made many of the
regulators probe asynchronously, though nothing I've seen implies that
the problems couldn't have also happened even without that.
I haven't personally been able to reproduce the lockdep issue, but the
issue does look somewhat legitimate. Specifically, it looks like in
regulator_resolve_supply() we are holding a "rdev" lock while calling
set_supply() -> create_regulator() which grabs the lock of a
_different_ "rdev" (the one for our supply). This is not necessarily
safe from a lockdep perspective since there is no documented ordering
between these two locks.
In reality, we should always be locking a regulator before the
supplying regulator, so I don't expect there to be any real deadlocks
in practice. However, the regulator framework in general doesn't
express this to lockdep.
Let's fix the issue by simply grabbing the two locks involved in the
same way we grab multiple locks elsewhere in the regulator framework:
using the "wound/wait" mechanisms.
Fixes: eaa7995c529b ("regulator: core: avoid regulator_resolve_supply() race condition")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20230329143317.RFC.v2.2.I30d8e1ca10cfbe5403884cdd192253a2e063eb9e@changeid
Signed-off-by: Mark Brown <broonie@kernel.org>
2023-03-29 14:33:54 -07:00
|
|
|
|
2023-04-13 17:34:17 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core: Avoid lockdep reports when resolving supplies
An automated bot told me that there was a potential lockdep problem
with regulators. This was on the chromeos-5.15 kernel, but I see
nothing that would be different downstream compared to upstream. The
bot said:
============================================
WARNING: possible recursive locking detected
5.15.104-lockdep-17461-gc1e499ed6604 #1 Not tainted
--------------------------------------------
kworker/u16:4/115 is trying to acquire lock:
ffffff8083110170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: create_regulator+0x398/0x7ec
but task is already holding lock:
ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(regulator_ww_class_mutex);
lock(regulator_ww_class_mutex);
*** DEADLOCK ***
May be due to missing lock nesting notation
4 locks held by kworker/u16:4/115:
#0: ffffff808006a948 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_one_work+0x520/0x1348
#1: ffffffc00e0a7cc0 ((work_completion)(&entry->work)){+.+.}-{0:0}, at: process_one_work+0x55c/0x1348
#2: ffffff80828a2260 (&dev->mutex){....}-{3:3}, at: __device_attach_async_helper+0xd0/0x2a4
#3: ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
stack backtrace:
CPU: 2 PID: 115 Comm: kworker/u16:4 Not tainted 5.15.104-lockdep-17461-gc1e499ed6604 #1 9292e52fa83c0e23762b2b3aa1bacf5787a4d5da
Hardware name: Google Quackingstick (rev0+) (DT)
Workqueue: events_unbound async_run_entry_fn
Call trace:
dump_backtrace+0x0/0x4ec
show_stack+0x34/0x50
dump_stack_lvl+0xdc/0x11c
dump_stack+0x1c/0x48
__lock_acquire+0x16d4/0x6c74
lock_acquire+0x208/0x750
__mutex_lock_common+0x11c/0x11f8
ww_mutex_lock+0xc0/0x440
create_regulator+0x398/0x7ec
regulator_resolve_supply+0x654/0x7c4
regulator_register_resolve_supply+0x30/0x120
class_for_each_device+0x1b8/0x230
regulator_register+0x17a4/0x1f40
devm_regulator_register+0x60/0xd0
reg_fixed_voltage_probe+0x728/0xaec
platform_probe+0x150/0x1c8
really_probe+0x274/0xa20
__driver_probe_device+0x1dc/0x3f4
driver_probe_device+0x78/0x1c0
__device_attach_driver+0x1ac/0x2c8
bus_for_each_drv+0x11c/0x190
__device_attach_async_helper+0x1e4/0x2a4
async_run_entry_fn+0xa0/0x3ac
process_one_work+0x638/0x1348
worker_thread+0x4a8/0x9c4
kthread+0x2e4/0x3a0
ret_from_fork+0x10/0x20
The problem was first reported soon after we made many of the
regulators probe asynchronously, though nothing I've seen implies that
the problems couldn't have also happened even without that.
I haven't personally been able to reproduce the lockdep issue, but the
issue does look somewhat legitimate. Specifically, it looks like in
regulator_resolve_supply() we are holding a "rdev" lock while calling
set_supply() -> create_regulator() which grabs the lock of a
_different_ "rdev" (the one for our supply). This is not necessarily
safe from a lockdep perspective since there is no documented ordering
between these two locks.
In reality, we should always be locking a regulator before the
supplying regulator, so I don't expect there to be any real deadlocks
in practice. However, the regulator framework in general doesn't
express this to lockdep.
Let's fix the issue by simply grabbing the two locks involved in the
same way we grab multiple locks elsewhere in the regulator framework:
using the "wound/wait" mechanisms.
Fixes: eaa7995c529b ("regulator: core: avoid regulator_resolve_supply() race condition")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20230329143317.RFC.v2.2.I30d8e1ca10cfbe5403884cdd192253a2e063eb9e@changeid
Signed-off-by: Mark Brown <broonie@kernel.org>
2023-03-29 14:33:54 -07:00
|
|
|
|
regulator: core: regulator_lock_two() should test for EDEADLK not EDEADLOCK
Compare against -EDEADLK, which is what ww_mutex_lock() actually
returns and what every other deadlock check in this file already uses.
Function regulator_lock_two() acquires two regulators via
regulator_lock_nested() -> ww_mutex_lock(). On contention,
ww_mutex_lock() returns -EDEADLK, which is the caller's signal to drop
the lock it holds and retry the acquisition in the canonical order.
However, regulator_lock_two() tests the return value against -EDEADLOCK
rather than -EDEADLK. On most architectures, EDEADLK and EDEADLOCK are
the same value, so the comparison happens to be correct and the bug is
invisible. But on MIPS, SPARC, and PowerPC, those two errors have
different values. The test is wrong: a genuine -EDEADLK backoff no
longer matches -EDEADLOCK, so instead of unlocking and retrying, the
code falls into WARN_ON(ret) and returns with only one of the two
regulators locked.
In practice, this is a bug only on MIPS, because the regulator core is
not built or used on the other two platforms.
In general, EDEADLK is preferred over EDEADLOCK for new code.
Fixes: cba6cfdc7c3f ("regulator: core: Avoid lockdep reports when resolving supplies")
Signed-off-by: Timur Tabi <ttabi@nvidia.com>
Link: https://patch.msgid.link/20260708235722.2953579-1-ttabi@nvidia.com
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-07-08 18:57:22 -05:00
|
|
|
|
regulator: core: Avoid lockdep reports when resolving supplies
An automated bot told me that there was a potential lockdep problem
with regulators. This was on the chromeos-5.15 kernel, but I see
nothing that would be different downstream compared to upstream. The
bot said:
============================================
WARNING: possible recursive locking detected
5.15.104-lockdep-17461-gc1e499ed6604 #1 Not tainted
--------------------------------------------
kworker/u16:4/115 is trying to acquire lock:
ffffff8083110170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: create_regulator+0x398/0x7ec
but task is already holding lock:
ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(regulator_ww_class_mutex);
lock(regulator_ww_class_mutex);
*** DEADLOCK ***
May be due to missing lock nesting notation
4 locks held by kworker/u16:4/115:
#0: ffffff808006a948 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_one_work+0x520/0x1348
#1: ffffffc00e0a7cc0 ((work_completion)(&entry->work)){+.+.}-{0:0}, at: process_one_work+0x55c/0x1348
#2: ffffff80828a2260 (&dev->mutex){....}-{3:3}, at: __device_attach_async_helper+0xd0/0x2a4
#3: ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
stack backtrace:
CPU: 2 PID: 115 Comm: kworker/u16:4 Not tainted 5.15.104-lockdep-17461-gc1e499ed6604 #1 9292e52fa83c0e23762b2b3aa1bacf5787a4d5da
Hardware name: Google Quackingstick (rev0+) (DT)
Workqueue: events_unbound async_run_entry_fn
Call trace:
dump_backtrace+0x0/0x4ec
show_stack+0x34/0x50
dump_stack_lvl+0xdc/0x11c
dump_stack+0x1c/0x48
__lock_acquire+0x16d4/0x6c74
lock_acquire+0x208/0x750
__mutex_lock_common+0x11c/0x11f8
ww_mutex_lock+0xc0/0x440
create_regulator+0x398/0x7ec
regulator_resolve_supply+0x654/0x7c4
regulator_register_resolve_supply+0x30/0x120
class_for_each_device+0x1b8/0x230
regulator_register+0x17a4/0x1f40
devm_regulator_register+0x60/0xd0
reg_fixed_voltage_probe+0x728/0xaec
platform_probe+0x150/0x1c8
really_probe+0x274/0xa20
__driver_probe_device+0x1dc/0x3f4
driver_probe_device+0x78/0x1c0
__device_attach_driver+0x1ac/0x2c8
bus_for_each_drv+0x11c/0x190
__device_attach_async_helper+0x1e4/0x2a4
async_run_entry_fn+0xa0/0x3ac
process_one_work+0x638/0x1348
worker_thread+0x4a8/0x9c4
kthread+0x2e4/0x3a0
ret_from_fork+0x10/0x20
The problem was first reported soon after we made many of the
regulators probe asynchronously, though nothing I've seen implies that
the problems couldn't have also happened even without that.
I haven't personally been able to reproduce the lockdep issue, but the
issue does look somewhat legitimate. Specifically, it looks like in
regulator_resolve_supply() we are holding a "rdev" lock while calling
set_supply() -> create_regulator() which grabs the lock of a
_different_ "rdev" (the one for our supply). This is not necessarily
safe from a lockdep perspective since there is no documented ordering
between these two locks.
In reality, we should always be locking a regulator before the
supplying regulator, so I don't expect there to be any real deadlocks
in practice. However, the regulator framework in general doesn't
express this to lockdep.
Let's fix the issue by simply grabbing the two locks involved in the
same way we grab multiple locks elsewhere in the regulator framework:
using the "wound/wait" mechanisms.
Fixes: eaa7995c529b ("regulator: core: avoid regulator_resolve_supply() race condition")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20230329143317.RFC.v2.2.I30d8e1ca10cfbe5403884cdd192253a2e063eb9e@changeid
Signed-off-by: Mark Brown <broonie@kernel.org>
2023-03-29 14:33:54 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:18 +03:00
|
|
|
|
2015-09-30 16:05:42 +02:00
|
|
|
|
2018-11-19 00:56:18 +03:00
|
|
|
|
2015-11-20 15:24:39 +01:00
|
|
|
|
2015-09-30 16:05:42 +02:00
|
|
|
|
2018-11-19 00:56:18 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-08-31 23:43:35 +03:00
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-08-31 23:43:35 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-10-05 18:36:29 +03:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-09-30 16:05:42 +02:00
|
|
|
|
2018-10-05 18:36:29 +03:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2015-09-30 16:05:42 +02:00
|
|
|
|
2018-10-05 18:36:29 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-10-05 18:36:29 +03:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:18 +03:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-10-05 18:36:29 +03:00
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-09-30 16:05:42 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-10-05 18:36:29 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2018-10-05 18:36:29 +03:00
|
|
|
|
2019-01-08 13:12:33 +01:00
|
|
|
|
2015-09-30 16:05:42 +02:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
2015-09-30 16:05:42 +02:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
2015-09-30 16:05:42 +02:00
|
|
|
|
|
|
|
|
|
2018-10-05 18:36:29 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2018-10-05 18:36:29 +03:00
|
|
|
|
|
|
|
|
|
2019-01-08 13:12:33 +01:00
|
|
|
|
2018-10-05 18:36:29 +03:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
2018-10-05 18:36:29 +03:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-09-30 16:05:42 +02:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2015-09-30 16:05:42 +02:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-03-29 14:33:53 -07:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-09-30 16:05:42 +02:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2019-06-24 00:08:32 +03:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-04-23 15:11:05 +08:00
|
|
|
|
2015-09-16 12:10:26 -07:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-07-12 11:20:37 +09:00
|
|
|
|
|
|
|
|
|
2011-07-21 15:07:37 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2011-07-12 11:20:37 +09:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-26 21:08:47 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2010-12-03 11:31:07 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:32 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2010-12-03 11:31:07 +01:00
|
|
|
|
|
|
|
|
|
2018-01-26 21:08:44 +08:00
|
|
|
|
2010-12-03 11:31:07 +01:00
|
|
|
|
|
|
|
|
|
2018-01-26 21:08:44 +08:00
|
|
|
|
2011-05-14 13:42:34 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-26 21:08:44 +08:00
|
|
|
|
2011-05-14 13:42:34 -07:00
|
|
|
|
|
|
|
|
|
2018-01-26 21:08:44 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2010-12-03 11:31:07 +01:00
|
|
|
|
|
|
|
|
|
2012-11-28 17:09:27 +00:00
|
|
|
|
2013-02-14 04:46:33 -08:00
|
|
|
|
|
|
|
|
|
2010-12-03 11:31:07 +01:00
|
|
|
|
2012-11-28 17:09:27 +00:00
|
|
|
|
2010-12-03 11:31:07 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-04-23 15:11:05 +08:00
|
|
|
|
2015-09-16 12:10:26 -07:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-11-21 15:42:28 +01:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-07-12 11:20:37 +09:00
|
|
|
|
|
|
|
|
|
2011-07-21 15:07:37 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2011-07-12 11:20:37 +09:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-11-29 11:50:03 +00:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2011-03-30 06:29:12 +09:00
|
|
|
|
2008-11-16 11:46:56 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-07-12 11:20:37 +09:00
|
|
|
|
2008-11-16 11:46:56 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-04-23 15:11:05 +08:00
|
|
|
|
2015-09-16 12:10:26 -07:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2011-03-30 06:29:12 +09:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-04-20 09:17:18 +05:30
|
|
|
|
|
|
|
|
|
2011-03-30 06:29:12 +09:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2011-03-30 06:29:12 +09:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2018-01-26 21:08:47 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-08-04 00:08:37 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2019-09-10 21:22:29 +02:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2019-09-10 21:22:29 +02:00
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2019-09-10 21:22:29 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2013-07-24 15:05:21 -07:00
|
|
|
|
|
|
|
|
|
2008-10-10 15:33:20 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-10-22 16:31:32 +01:00
|
|
|
|
2008-10-10 15:33:20 +01:00
|
|
|
|
2013-07-24 15:05:21 -07:00
|
|
|
|
2008-10-10 15:33:20 +01:00
|
|
|
|
2018-08-16 13:28:01 -07:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-08-16 13:28:01 -07:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2018-08-16 13:28:01 -07:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2018-08-16 13:28:01 -07:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2018-08-16 13:28:01 -07:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2018-08-16 13:28:01 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2008-11-11 17:38:53 -08:00
|
|
|
|
|
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
2008-11-11 17:38:53 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
|
|
|
|
|
2008-11-11 17:38:53 -08:00
|
|
|
|
|
|
|
|
|
2009-08-03 18:49:56 +01:00
|
|
|
|
2008-11-11 17:38:53 -08:00
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2009-08-03 18:49:56 +01:00
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2009-08-03 18:49:56 +01:00
|
|
|
|
|
|
|
|
|
2008-11-11 17:38:53 -08:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
2008-11-11 17:38:53 -08:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
|
|
|
|
|
2009-01-14 23:03:17 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-08-31 10:36:37 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-07-12 13:53:35 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-01-14 23:03:17 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
2009-01-14 23:03:17 -08:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2013-07-24 15:05:21 -07:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2013-07-24 15:05:21 -07:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2013-07-24 15:05:21 -07:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2013-07-24 15:05:21 -07:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2008-11-11 17:38:53 -08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2008-11-11 17:38:53 -08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2008-11-11 17:38:53 -08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2008-11-11 17:38:53 -08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2008-11-11 17:38:53 -08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2008-11-11 17:38:53 -08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
2008-10-10 15:33:20 +01:00
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
|
|
|
|
|
2012-08-31 10:36:37 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-05-29 19:52:26 +08:00
|
|
|
|
2008-11-11 17:39:02 -08:00
|
|
|
|
2025-01-15 15:41:57 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-05-03 23:52:49 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2021-04-20 09:17:18 +05:30
|
|
|
|
|
|
|
|
|
2015-01-27 18:46:31 -08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-01-27 18:46:31 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-02-21 12:55:22 +01:00
|
|
|
|
|
|
|
|
|
2015-01-27 18:46:31 -08:00
|
|
|
|
2019-02-21 12:55:22 +01:00
|
|
|
|
2015-01-27 18:46:31 -08:00
|
|
|
|
2015-02-11 19:39:12 -08:00
|
|
|
|
|
|
|
|
|
2015-01-27 18:46:31 -08:00
|
|
|
|
|
|
|
|
|
2015-02-11 19:39:12 -08:00
|
|
|
|
|
|
|
|
|
2015-01-27 18:46:31 -08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2015-06-11 17:37:03 -07:00
|
|
|
|
|
|
|
|
|
2015-02-11 19:39:12 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
|
|
|
|
|
2015-02-11 19:39:12 -08:00
|
|
|
|
regulator: core: Don't err if allow-set-load but no allowed-modes
Apparently the device trees of some boards have the property
"regulator-allow-set-load" for some of their regulators but then they
don't specify anything for "regulator-allowed-modes". That's not
really legit, but...
...before commit efb0cb50c427 ("regulator: qcom-rpmh: Implement
get_optimum_mode(), not set_load()") they used to get away with it, at
least on boards using RPMH regulators. That's because when a regulator
driver implements set_load() then the core doesn't look at
"regulator-allowed-modes" when trying to automatically adjust things
in response to the regulator's load. The core doesn't know what mode
we'll end up in, so how could it validate it?
Said another way: before commit efb0cb50c427 ("regulator: qcom-rpmh:
Implement get_optimum_mode(), not set_load()") some boards _were_
having the regulator mode adjusted despite listing no allowed
modes. After commit efb0cb50c427 ("regulator: qcom-rpmh: Implement
get_optimum_mode(), not set_load()") these same boards were now
getting an error returned when trying to use their regulators, since
simply enabling a regulator tries to update its load and that was
failing.
We don't really want to go back to the behavior from before commit
efb0cb50c427 ("regulator: qcom-rpmh: Implement get_optimum_mode(), not
set_load()"). Boards shouldn't have been changing modes if no allowed
modes were listed. However, the behavior after commit efb0cb50c427
("regulator: qcom-rpmh: Implement get_optimum_mode(), not set_load()")
isn't the best because now boards can't even turn their regulators on.
Let's choose to detect this case and return "no error" from
drms_uA_update(). The net-result will be _different_ behavior than we
had before commit efb0cb50c427 ("regulator: qcom-rpmh: Implement
get_optimum_mode(), not set_load()"), but this new behavior seems more
correct. If a board truly needed the mode switched then its device
tree should be updated to list the allowed modes.
Reported-by: Andrew Halaney <ahalaney@redhat.com>
Fixes: efb0cb50c427 ("regulator: qcom-rpmh: Implement get_optimum_mode(), not set_load()")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Tested-by: Andrew Halaney <ahalaney@redhat.com>
Link: https://lore.kernel.org/r/20220824142229.RFT.v2.2.I6f77860e5cd98bf5c67208fa9edda4a08847c304@changeid
Signed-off-by: Mark Brown <broonie@kernel.org>
2022-08-24 14:22:57 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-09-19 14:46:54 -07:00
|
|
|
|
2019-06-24 00:08:32 +03:00
|
|
|
|
2022-08-24 14:22:56 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-09-19 14:46:54 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-12-15 11:46:46 +01:00
|
|
|
|
2016-09-19 14:46:54 -07:00
|
|
|
|
|
|
|
|
|
2022-08-24 14:22:56 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-09-19 14:46:54 -07:00
|
|
|
|
2015-02-11 19:39:12 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2015-02-11 19:39:12 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
|
|
|
|
|
2015-02-11 19:39:12 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
|
|
|
|
|
2015-01-27 18:46:31 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2020-08-04 00:08:37 -07:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2009-10-22 16:31:33 +01:00
|
|
|
|
2018-01-26 21:08:45 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2018-01-26 21:08:45 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2012-04-14 09:14:34 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-26 21:08:47 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-08-04 00:08:37 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-10-05 15:15:46 +02:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2015-06-09 20:09:42 +00:00
|
|
|
|
2015-06-10 06:13:15 +00:00
|
|
|
|
2009-10-22 16:31:31 +01:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2009-10-22 16:31:31 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2015-06-10 06:13:15 +00:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2015-06-10 06:13:15 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2009-10-22 16:31:31 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:32 +03:00
|
|
|
|
2009-10-22 16:31:31 +01:00
|
|
|
|
2015-06-10 06:13:15 +00:00
|
|
|
|
|
|
|
|
|
2009-10-22 16:31:31 +01:00
|
|
|
|
|
|
|
|
|
2011-05-08 22:13:37 +01:00
|
|
|
|
2015-06-10 06:13:15 +00:00
|
|
|
|
|
|
|
|
|
2011-05-08 22:13:37 +01:00
|
|
|
|
2009-10-22 16:31:31 +01:00
|
|
|
|
|
|
|
|
|
2015-06-10 06:13:15 +00:00
|
|
|
|
|
|
|
|
|
2009-10-22 16:31:31 +01:00
|
|
|
|
2015-06-10 06:13:15 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-10-22 16:31:31 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-06-10 06:13:15 +00:00
|
|
|
|
|
|
|
|
|
2009-10-22 16:31:31 +01:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2015-06-10 06:13:15 +00:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2015-06-10 06:13:15 +00:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2015-06-10 06:13:15 +00:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2025-01-15 15:41:57 +01:00
|
|
|
|
2026-01-09 08:38:37 +00:00
|
|
|
|
2025-01-15 15:41:57 +01:00
|
|
|
|
|
|
|
|
|
2012-08-07 21:01:37 +02:00
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-08-07 21:01:37 +02:00
|
|
|
|
2014-10-31 19:11:59 +00:00
|
|
|
|
2020-10-05 15:15:46 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-02-09 13:26:13 +00:00
|
|
|
|
|
|
|
|
|
2016-04-23 15:11:05 +08:00
|
|
|
|
2012-02-09 13:26:13 +00:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2009-10-19 15:53:50 +01:00
|
|
|
|
2009-10-22 16:31:32 +01:00
|
|
|
|
2009-10-19 15:53:50 +01:00
|
|
|
|
2014-08-13 19:33:38 +08:00
|
|
|
|
2009-10-22 16:31:30 +01:00
|
|
|
|
2026-07-20 21:25:17 +05:30
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-10-22 16:31:30 +01:00
|
|
|
|
|
|
|
|
|
2026-07-20 21:25:17 +05:30
|
|
|
|
2016-03-21 18:12:52 +00:00
|
|
|
|
2019-06-24 00:08:32 +03:00
|
|
|
|
2018-05-15 15:07:17 -07:00
|
|
|
|
|
|
|
|
|
2019-01-08 13:12:33 +01:00
|
|
|
|
2018-05-15 15:07:17 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:32 +03:00
|
|
|
|
2018-05-15 15:07:17 -07:00
|
|
|
|
|
|
|
|
|
2014-05-20 12:12:16 +02:00
|
|
|
|
2021-08-26 12:40:17 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-05-20 12:12:16 +02:00
|
|
|
|
|
|
|
|
|
2016-03-21 18:12:52 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-03-29 16:33:42 -07:00
|
|
|
|
|
|
|
|
|
2014-05-20 12:12:16 +02:00
|
|
|
|
2016-03-21 18:12:52 +00:00
|
|
|
|
2014-05-20 12:12:16 +02:00
|
|
|
|
|
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
|
|
|
|
|
2014-05-20 12:12:16 +02:00
|
|
|
|
|
|
|
|
|
2010-12-12 14:25:50 +00:00
|
|
|
|
2009-10-22 16:31:30 +01:00
|
|
|
|
2009-10-19 15:53:50 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2013-09-20 13:13:02 +05:30
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-08-13 19:33:38 +08:00
|
|
|
|
2013-09-20 13:13:02 +05:30
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-02-20 17:36:03 +01:00
|
|
|
|
|
|
|
|
|
2021-06-03 08:42:12 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
|
|
|
|
|
2008-12-31 12:52:41 +00:00
|
|
|
|
regulator: core: don't fail regulator_register() with missing required supply
Since commit 98e48cd9283d ("regulator: core: resolve supply for
boot-on/always-on regulators"), the regulator core returns
-EPROBE_DEFER if a supply can not be resolved at regulator_register()
time due to set_machine_constraints() requiring that supply (e.g.
because of always-on or boot-on).
In some hardware designs, multiple PMICs are used where individual
rails of each act as supplies for rails of the other, and vice-versa.
In such a design no PMIC driver can probe when registering one top-
level regulator device (as is common practice for almost all regulator
drivers in Linux) since that commit. Supplies are only considered when
their driver has fully bound, but because in a design like the above
two drivers / devices depend on each other, neither will have fully
bound while the other probes. The Google Pixel 6 and 6 Pro (oriole and
raven) are examples of such a design.
One way to make this work would be to register each rail as an
individual device, rather than just one top-level regulator device.
Then, fw-devlink and Linux' driver core could do their usual handling
of deferred device probe as each rail would be probed individually.
This approach was dismissed in [1] as each regulator driver would have
to take care of this itself.
Alternatively, we can change the regulator core to not fail
regulator_register() if a rail's required supply can not be resolved
while keeping the intended change from above mentioned commit, and
instead retry whenever a new rail is registered. This commit implements
such an approach:
If set_machine_constraints() requests probe deferral,
regulator_register() still succeeds and we retry setting
constraints as part of regulator_resolve_supply().
We still do not enable the regulator or allow consumers to use it
until constraints have been set (including resolution of the
supply) to prevent enabling of a regulator before its supply.
With this change, we keep track of regulators with missing required
supplies and can therefore try to resolve them again and try to set
the constraints again once more regulators become available.
Care has to be taken to not allow consumers to use regulators that
haven't had their constraints set yet. regulator_get() ensures that
and now returns -EPROBE_DEFER in that case.
The implementation is straight-forward, thanks to our newly introduced
regulator-bus. Locking in regulator_resolve_supply() has to be done
carefully, as a combination of regulator_(un)lock_two() and
regulator_(un)lock_dependent() is needed. The reason is that
set_machine_constraints() might call regulator_enable() which needs
rdev and all its dependents locked, but everything else requires to
only have rdev and its supply locked.
Link: https://lore.kernel.org/all/aRn_-o-vie_QoDXD@sirena.co.uk/ [1]
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-8-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:44 +00:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
regulator: core: don't fail regulator_register() with missing required supply
Since commit 98e48cd9283d ("regulator: core: resolve supply for
boot-on/always-on regulators"), the regulator core returns
-EPROBE_DEFER if a supply can not be resolved at regulator_register()
time due to set_machine_constraints() requiring that supply (e.g.
because of always-on or boot-on).
In some hardware designs, multiple PMICs are used where individual
rails of each act as supplies for rails of the other, and vice-versa.
In such a design no PMIC driver can probe when registering one top-
level regulator device (as is common practice for almost all regulator
drivers in Linux) since that commit. Supplies are only considered when
their driver has fully bound, but because in a design like the above
two drivers / devices depend on each other, neither will have fully
bound while the other probes. The Google Pixel 6 and 6 Pro (oriole and
raven) are examples of such a design.
One way to make this work would be to register each rail as an
individual device, rather than just one top-level regulator device.
Then, fw-devlink and Linux' driver core could do their usual handling
of deferred device probe as each rail would be probed individually.
This approach was dismissed in [1] as each regulator driver would have
to take care of this itself.
Alternatively, we can change the regulator core to not fail
regulator_register() if a rail's required supply can not be resolved
while keeping the intended change from above mentioned commit, and
instead retry whenever a new rail is registered. This commit implements
such an approach:
If set_machine_constraints() requests probe deferral,
regulator_register() still succeeds and we retry setting
constraints as part of regulator_resolve_supply().
We still do not enable the regulator or allow consumers to use it
until constraints have been set (including resolution of the
supply) to prevent enabling of a regulator before its supply.
With this change, we keep track of regulators with missing required
supplies and can therefore try to resolve them again and try to set
the constraints again once more regulators become available.
Care has to be taken to not allow consumers to use regulators that
haven't had their constraints set yet. regulator_get() ensures that
and now returns -EPROBE_DEFER in that case.
The implementation is straight-forward, thanks to our newly introduced
regulator-bus. Locking in regulator_resolve_supply() has to be done
carefully, as a combination of regulator_(un)lock_two() and
regulator_(un)lock_dependent() is needed. The reason is that
set_machine_constraints() might call regulator_enable() which needs
rdev and all its dependents locked, but everything else requires to
only have rdev and its supply locked.
Link: https://lore.kernel.org/all/aRn_-o-vie_QoDXD@sirena.co.uk/ [1]
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-8-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:44 +00:00
|
|
|
|
|
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
|
|
|
|
|
2014-08-13 19:33:38 +08:00
|
|
|
|
2008-09-09 16:21:19 +01:00
|
|
|
|
2026-01-09 08:38:39 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2010-11-29 15:55:17 +00:00
|
|
|
|
2009-10-19 15:53:50 +01:00
|
|
|
|
2016-01-26 16:38:59 +00:00
|
|
|
|
2009-02-26 11:48:36 -08:00
|
|
|
|
2013-09-20 13:13:02 +05:30
|
|
|
|
2009-10-19 15:53:50 +01:00
|
|
|
|
2016-01-26 16:38:59 +00:00
|
|
|
|
2009-02-26 11:48:36 -08:00
|
|
|
|
2015-06-11 17:37:06 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
2016-01-26 16:38:59 +00:00
|
|
|
|
2015-06-11 17:37:06 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2011-11-29 18:11:19 +00:00
|
|
|
|
2020-08-04 00:08:37 -07:00
|
|
|
|
2008-09-09 16:21:19 +01:00
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
2016-01-26 16:38:59 +00:00
|
|
|
|
2008-09-09 16:21:19 +01:00
|
|
|
|
|
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2011-11-29 18:11:19 +00:00
|
|
|
|
2009-02-26 19:24:19 +00:00
|
|
|
|
2010-11-30 05:53:48 -08:00
|
|
|
|
2016-01-26 16:38:59 +00:00
|
|
|
|
2009-02-26 19:24:19 +00:00
|
|
|
|
|
|
|
|
|
2010-11-29 15:55:17 +00:00
|
|
|
|
2009-02-26 19:24:19 +00:00
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
2016-01-26 16:38:59 +00:00
|
|
|
|
2009-02-26 19:24:19 +00:00
|
|
|
|
2018-11-26 09:08:27 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-02-26 19:24:19 +00:00
|
|
|
|
|
|
|
|
|
2013-06-29 18:21:15 +05:30
|
|
|
|
|
|
|
|
|
2012-06-11 17:41:08 +05:30
|
|
|
|
|
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
2016-01-26 16:38:59 +00:00
|
|
|
|
2012-06-11 17:41:08 +05:30
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-06-11 17:37:04 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
2016-01-26 16:38:59 +00:00
|
|
|
|
2015-06-11 17:37:04 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-06-11 17:37:05 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
2016-01-26 16:38:59 +00:00
|
|
|
|
2015-06-11 17:37:05 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-03 08:42:12 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026-03-23 10:03:35 -04:00
|
|
|
|
2021-06-03 08:42:12 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-07-17 14:41:54 -07:00
|
|
|
|
|
|
|
|
|
2021-06-03 08:42:12 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-07-17 14:41:54 -07:00
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
|
|
|
|
|
2016-01-26 16:38:59 +00:00
|
|
|
|
2015-07-17 14:41:54 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-03 08:42:12 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-03-02 16:24:46 +05:30
|
|
|
|
2025-08-28 20:27:37 +08:00
|
|
|
|
|
|
|
|
|
2016-03-02 16:24:46 +05:30
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
2016-03-02 16:24:46 +05:30
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-26 17:27:44 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-11-07 18:10:08 +01:00
|
|
|
|
|
|
|
|
|
2026-01-09 08:38:39 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-12-01 11:38:06 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core: don't fail regulator_register() with missing required supply
Since commit 98e48cd9283d ("regulator: core: resolve supply for
boot-on/always-on regulators"), the regulator core returns
-EPROBE_DEFER if a supply can not be resolved at regulator_register()
time due to set_machine_constraints() requiring that supply (e.g.
because of always-on or boot-on).
In some hardware designs, multiple PMICs are used where individual
rails of each act as supplies for rails of the other, and vice-versa.
In such a design no PMIC driver can probe when registering one top-
level regulator device (as is common practice for almost all regulator
drivers in Linux) since that commit. Supplies are only considered when
their driver has fully bound, but because in a design like the above
two drivers / devices depend on each other, neither will have fully
bound while the other probes. The Google Pixel 6 and 6 Pro (oriole and
raven) are examples of such a design.
One way to make this work would be to register each rail as an
individual device, rather than just one top-level regulator device.
Then, fw-devlink and Linux' driver core could do their usual handling
of deferred device probe as each rail would be probed individually.
This approach was dismissed in [1] as each regulator driver would have
to take care of this itself.
Alternatively, we can change the regulator core to not fail
regulator_register() if a rail's required supply can not be resolved
while keeping the intended change from above mentioned commit, and
instead retry whenever a new rail is registered. This commit implements
such an approach:
If set_machine_constraints() requests probe deferral,
regulator_register() still succeeds and we retry setting
constraints as part of regulator_resolve_supply().
We still do not enable the regulator or allow consumers to use it
until constraints have been set (including resolution of the
supply) to prevent enabling of a regulator before its supply.
With this change, we keep track of regulators with missing required
supplies and can therefore try to resolve them again and try to set
the constraints again once more regulators become available.
Care has to be taken to not allow consumers to use regulators that
haven't had their constraints set yet. regulator_get() ensures that
and now returns -EPROBE_DEFER in that case.
The implementation is straight-forward, thanks to our newly introduced
regulator-bus. Locking in regulator_resolve_supply() has to be done
carefully, as a combination of regulator_(un)lock_two() and
regulator_(un)lock_dependent() is needed. The reason is that
set_machine_constraints() might call regulator_enable() which needs
rdev and all its dependents locked, but everything else requires to
only have rdev and its supply locked.
Link: https://lore.kernel.org/all/aRn_-o-vie_QoDXD@sirena.co.uk/ [1]
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-8-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:44 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-12-06 14:23:18 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-11-07 18:10:08 +01:00
|
|
|
|
2018-12-06 14:23:18 -08:00
|
|
|
|
|
|
|
|
|
2018-11-26 17:27:44 +02:00
|
|
|
|
|
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
2025-11-07 18:10:08 +01:00
|
|
|
|
|
|
|
|
|
2018-11-26 17:27:44 +02:00
|
|
|
|
|
|
|
|
|
2019-11-13 11:27:37 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core: Shorten off-on-delay-us for always-on/boot-on by time since booted
This is very close to a straight revert of commit 218320fec294
("regulator: core: Fix off-on-delay-us for always-on/boot-on
regulators"). We've identified that patch as causing a boot speed
regression on sc7180-trogdor boards. While boot speed certainly isn't
more important than making sure that power sequencing is correct,
looking closely at the original change it doesn't seem to have been
fully justified. It mentions "cycling issues" without describing
exactly what the issues were. That means it's possible that the
cycling issues were really a problem that should be fixed in a
different way.
Let's take a careful look at how we should handle regulators that have
an off-on-delay and that are boot-on or always-on. Linux currently
doesn't have any way to identify whether a GPIO regulator was already
on when the kernel booted. That means that when the kernel boots we
probe a regulator, see that it wants boot-on / always-on we, and then
turn the regulator on. We could be in one of two cases when we do
this:
a) The regulator might have been left on by the bootloader and we're
ensuring that it stays on.
b) The regulator might have been left off by the bootloader and we're
just now turning it on.
For case a) we definitely don't need any sort of delay. For case b) we
_might_ need some delay in case the bootloader turned the regulator
off _right_ before booting the kernel. To get the proper delay for
case b) then we can just assume a `last_off` of 0, which is what it
gets initialized to by default.
As per above, we can't tell whether we're in case a) or case b) so
we'll assume the longer delay (case b). This basically puts the code
to how it was before commit 218320fec294 ("regulator: core: Fix
off-on-delay-us for always-on/boot-on regulators"). However, we add
one important change: we make sure that the delay is actually honored
if `last_off` is 0. Though the original "cycling issues" cited were
vague, I'm hopeful that this important extra change will be enough to
fix the issues that the initial commit mentioned.
With this fix, I've confined that on a sc7180-trogdor board the delay
at boot goes down from 500 ms to ~250 ms. That's not as good as the 0
ms that we had prior to commit 218320fec294 ("regulator: core: Fix
off-on-delay-us for always-on/boot-on regulators"), but it's probably
safer because we don't know if the bootloader turned the regulator off
right before booting.
One note is that it's possible that we could be in a state that's not
a) or b) if there are other issues in the kernel. The only one I can
think of is related to pinctrl. If the pinctrl driver being used on a
board isn't careful about avoiding glitches when setting up a pin then
it's possible that setting up a pin could cause the regulator to "turn
off" briefly immediately before the regulator probes. If this is
indeed causing problems then the pinctrl driver should be fixed,
perhaps in a similar way to what was done in commit d21f4b7ffc22
("pinctrl: qcom: Avoid glitching lines when we first mux to output")
Fixes: 218320fec294 ("regulator: core: Fix off-on-delay-us for always-on/boot-on regulators")
Cc: Christian Kohlschütter <christian@kohlschutter.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20230313111806.1.I2eaad872be0932a805c239a7c7a102233fb0b03b@changeid
Signed-off-by: Mark Brown <broonie@kernel.org>
2023-03-13 11:18:19 -07:00
|
|
|
|
|
|
|
|
|
2018-11-26 17:27:44 +02:00
|
|
|
|
|
|
|
|
|
2025-01-15 15:41:57 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2011-07-15 10:50:43 +08:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
|
|
|
|
|
2025-10-01 12:56:49 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
treewide: Replace kmalloc with kmalloc_obj for non-scalar types
This is the result of running the Coccinelle script from
scripts/coccinelle/api/kmalloc_objs.cocci. The script is designed to
avoid scalar types (which need careful case-by-case checking), and
instead replace kmalloc-family calls that allocate struct or union
object instances:
Single allocations: kmalloc(sizeof(TYPE), ...)
are replaced with: kmalloc_obj(TYPE, ...)
Array allocations: kmalloc_array(COUNT, sizeof(TYPE), ...)
are replaced with: kmalloc_objs(TYPE, COUNT, ...)
Flex array allocations: kmalloc(struct_size(PTR, FAM, COUNT), ...)
are replaced with: kmalloc_flex(*PTR, FAM, COUNT, ...)
(where TYPE may also be *VAR)
The resulting allocations no longer return "void *", instead returning
"TYPE *".
Signed-off-by: Kees Cook <kees@kernel.org>
2026-02-20 23:49:23 -08:00
|
|
|
|
2025-10-01 12:56:49 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core: don't fail regulator_register() with missing required supply
Since commit 98e48cd9283d ("regulator: core: resolve supply for
boot-on/always-on regulators"), the regulator core returns
-EPROBE_DEFER if a supply can not be resolved at regulator_register()
time due to set_machine_constraints() requiring that supply (e.g.
because of always-on or boot-on).
In some hardware designs, multiple PMICs are used where individual
rails of each act as supplies for rails of the other, and vice-versa.
In such a design no PMIC driver can probe when registering one top-
level regulator device (as is common practice for almost all regulator
drivers in Linux) since that commit. Supplies are only considered when
their driver has fully bound, but because in a design like the above
two drivers / devices depend on each other, neither will have fully
bound while the other probes. The Google Pixel 6 and 6 Pro (oriole and
raven) are examples of such a design.
One way to make this work would be to register each rail as an
individual device, rather than just one top-level regulator device.
Then, fw-devlink and Linux' driver core could do their usual handling
of deferred device probe as each rail would be probed individually.
This approach was dismissed in [1] as each regulator driver would have
to take care of this itself.
Alternatively, we can change the regulator core to not fail
regulator_register() if a rail's required supply can not be resolved
while keeping the intended change from above mentioned commit, and
instead retry whenever a new rail is registered. This commit implements
such an approach:
If set_machine_constraints() requests probe deferral,
regulator_register() still succeeds and we retry setting
constraints as part of regulator_resolve_supply().
We still do not enable the regulator or allow consumers to use it
until constraints have been set (including resolution of the
supply) to prevent enabling of a regulator before its supply.
With this change, we keep track of regulators with missing required
supplies and can therefore try to resolve them again and try to set
the constraints again once more regulators become available.
Care has to be taken to not allow consumers to use regulators that
haven't had their constraints set yet. regulator_get() ensures that
and now returns -EPROBE_DEFER in that case.
The implementation is straight-forward, thanks to our newly introduced
regulator-bus. Locking in regulator_resolve_supply() has to be done
carefully, as a combination of regulator_(un)lock_two() and
regulator_(un)lock_dependent() is needed. The reason is that
set_machine_constraints() might call regulator_enable() which needs
rdev and all its dependents locked, but everything else requires to
only have rdev and its supply locked.
Link: https://lore.kernel.org/all/aRn_-o-vie_QoDXD@sirena.co.uk/ [1]
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-8-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:44 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
|
|
|
|
|
regulator: core: Avoid lockdep reports when resolving supplies
An automated bot told me that there was a potential lockdep problem
with regulators. This was on the chromeos-5.15 kernel, but I see
nothing that would be different downstream compared to upstream. The
bot said:
============================================
WARNING: possible recursive locking detected
5.15.104-lockdep-17461-gc1e499ed6604 #1 Not tainted
--------------------------------------------
kworker/u16:4/115 is trying to acquire lock:
ffffff8083110170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: create_regulator+0x398/0x7ec
but task is already holding lock:
ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(regulator_ww_class_mutex);
lock(regulator_ww_class_mutex);
*** DEADLOCK ***
May be due to missing lock nesting notation
4 locks held by kworker/u16:4/115:
#0: ffffff808006a948 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_one_work+0x520/0x1348
#1: ffffffc00e0a7cc0 ((work_completion)(&entry->work)){+.+.}-{0:0}, at: process_one_work+0x55c/0x1348
#2: ffffff80828a2260 (&dev->mutex){....}-{3:3}, at: __device_attach_async_helper+0xd0/0x2a4
#3: ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
stack backtrace:
CPU: 2 PID: 115 Comm: kworker/u16:4 Not tainted 5.15.104-lockdep-17461-gc1e499ed6604 #1 9292e52fa83c0e23762b2b3aa1bacf5787a4d5da
Hardware name: Google Quackingstick (rev0+) (DT)
Workqueue: events_unbound async_run_entry_fn
Call trace:
dump_backtrace+0x0/0x4ec
show_stack+0x34/0x50
dump_stack_lvl+0xdc/0x11c
dump_stack+0x1c/0x48
__lock_acquire+0x16d4/0x6c74
lock_acquire+0x208/0x750
__mutex_lock_common+0x11c/0x11f8
ww_mutex_lock+0xc0/0x440
create_regulator+0x398/0x7ec
regulator_resolve_supply+0x654/0x7c4
regulator_register_resolve_supply+0x30/0x120
class_for_each_device+0x1b8/0x230
regulator_register+0x17a4/0x1f40
devm_regulator_register+0x60/0xd0
reg_fixed_voltage_probe+0x728/0xaec
platform_probe+0x150/0x1c8
really_probe+0x274/0xa20
__driver_probe_device+0x1dc/0x3f4
driver_probe_device+0x78/0x1c0
__device_attach_driver+0x1ac/0x2c8
bus_for_each_drv+0x11c/0x190
__device_attach_async_helper+0x1e4/0x2a4
async_run_entry_fn+0xa0/0x3ac
process_one_work+0x638/0x1348
worker_thread+0x4a8/0x9c4
kthread+0x2e4/0x3a0
ret_from_fork+0x10/0x20
The problem was first reported soon after we made many of the
regulators probe asynchronously, though nothing I've seen implies that
the problems couldn't have also happened even without that.
I haven't personally been able to reproduce the lockdep issue, but the
issue does look somewhat legitimate. Specifically, it looks like in
regulator_resolve_supply() we are holding a "rdev" lock while calling
set_supply() -> create_regulator() which grabs the lock of a
_different_ "rdev" (the one for our supply). This is not necessarily
safe from a lockdep perspective since there is no documented ordering
between these two locks.
In reality, we should always be locking a regulator before the
supplying regulator, so I don't expect there to be any real deadlocks
in practice. However, the regulator framework in general doesn't
express this to lockdep.
Let's fix the issue by simply grabbing the two locks involved in the
same way we grab multiple locks elsewhere in the regulator framework:
using the "wound/wait" mechanisms.
Fixes: eaa7995c529b ("regulator: core: avoid regulator_resolve_supply() race condition")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20230329143317.RFC.v2.2.I30d8e1ca10cfbe5403884cdd192253a2e063eb9e@changeid
Signed-off-by: Mark Brown <broonie@kernel.org>
2023-03-29 14:33:54 -07:00
|
|
|
|
|
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
|
|
|
|
|
2011-06-09 16:22:22 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-09-29 15:07:17 +01:00
|
|
|
|
2011-06-09 16:22:22 +01:00
|
|
|
|
2015-07-15 16:10:28 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-06-09 16:22:22 +01:00
|
|
|
|
2011-12-29 17:03:20 +08:00
|
|
|
|
2022-12-01 20:27:05 +08:00
|
|
|
|
2011-12-29 17:03:20 +08:00
|
|
|
|
2011-06-09 16:22:22 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2012-07-23 20:35:46 +05:30
|
|
|
|
2011-06-09 16:22:22 +01:00
|
|
|
|
|
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2010-11-18 15:02:26 -08:00
|
|
|
|
2008-12-31 12:52:41 +00:00
|
|
|
|
2009-06-17 17:56:39 +01:00
|
|
|
|
2008-12-31 12:52:41 +00:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
|
|
|
|
|
2012-02-02 00:10:51 +00:00
|
|
|
|
|
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2020-08-12 03:31:36 +02:00
|
|
|
|
2009-07-21 16:00:26 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-07-21 16:00:26 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026-02-21 16:37:42 -08:00
|
|
|
|
2020-08-12 03:31:36 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-12-31 12:54:19 +00:00
|
|
|
|
2010-04-29 10:55:09 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-12-31 12:54:19 +00:00
|
|
|
|
2010-04-29 10:55:09 +03:00
|
|
|
|
|
|
|
|
|
2008-12-31 12:54:19 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-02-02 00:10:51 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-08-12 03:31:36 +02:00
|
|
|
|
2008-12-31 12:54:19 +00:00
|
|
|
|
|
|
|
|
|
2020-08-12 03:31:36 +02:00
|
|
|
|
|
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
|
|
|
|
|
2020-08-12 03:31:36 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
|
|
|
|
|
2009-01-22 16:00:29 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-06-17 17:56:39 +01:00
|
|
|
|
2009-01-22 16:00:29 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-04-21 17:23:21 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-03-05 17:05:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-08-12 03:31:35 +02:00
|
|
|
|
regulator: core: Avoid lockdep reports when resolving supplies
An automated bot told me that there was a potential lockdep problem
with regulators. This was on the chromeos-5.15 kernel, but I see
nothing that would be different downstream compared to upstream. The
bot said:
============================================
WARNING: possible recursive locking detected
5.15.104-lockdep-17461-gc1e499ed6604 #1 Not tainted
--------------------------------------------
kworker/u16:4/115 is trying to acquire lock:
ffffff8083110170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: create_regulator+0x398/0x7ec
but task is already holding lock:
ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(regulator_ww_class_mutex);
lock(regulator_ww_class_mutex);
*** DEADLOCK ***
May be due to missing lock nesting notation
4 locks held by kworker/u16:4/115:
#0: ffffff808006a948 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_one_work+0x520/0x1348
#1: ffffffc00e0a7cc0 ((work_completion)(&entry->work)){+.+.}-{0:0}, at: process_one_work+0x55c/0x1348
#2: ffffff80828a2260 (&dev->mutex){....}-{3:3}, at: __device_attach_async_helper+0xd0/0x2a4
#3: ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
stack backtrace:
CPU: 2 PID: 115 Comm: kworker/u16:4 Not tainted 5.15.104-lockdep-17461-gc1e499ed6604 #1 9292e52fa83c0e23762b2b3aa1bacf5787a4d5da
Hardware name: Google Quackingstick (rev0+) (DT)
Workqueue: events_unbound async_run_entry_fn
Call trace:
dump_backtrace+0x0/0x4ec
show_stack+0x34/0x50
dump_stack_lvl+0xdc/0x11c
dump_stack+0x1c/0x48
__lock_acquire+0x16d4/0x6c74
lock_acquire+0x208/0x750
__mutex_lock_common+0x11c/0x11f8
ww_mutex_lock+0xc0/0x440
create_regulator+0x398/0x7ec
regulator_resolve_supply+0x654/0x7c4
regulator_register_resolve_supply+0x30/0x120
class_for_each_device+0x1b8/0x230
regulator_register+0x17a4/0x1f40
devm_regulator_register+0x60/0xd0
reg_fixed_voltage_probe+0x728/0xaec
platform_probe+0x150/0x1c8
really_probe+0x274/0xa20
__driver_probe_device+0x1dc/0x3f4
driver_probe_device+0x78/0x1c0
__device_attach_driver+0x1ac/0x2c8
bus_for_each_drv+0x11c/0x190
__device_attach_async_helper+0x1e4/0x2a4
async_run_entry_fn+0xa0/0x3ac
process_one_work+0x638/0x1348
worker_thread+0x4a8/0x9c4
kthread+0x2e4/0x3a0
ret_from_fork+0x10/0x20
The problem was first reported soon after we made many of the
regulators probe asynchronously, though nothing I've seen implies that
the problems couldn't have also happened even without that.
I haven't personally been able to reproduce the lockdep issue, but the
issue does look somewhat legitimate. Specifically, it looks like in
regulator_resolve_supply() we are holding a "rdev" lock while calling
set_supply() -> create_regulator() which grabs the lock of a
_different_ "rdev" (the one for our supply). This is not necessarily
safe from a lockdep perspective since there is no documented ordering
between these two locks.
In reality, we should always be locking a regulator before the
supplying regulator, so I don't expect there to be any real deadlocks
in practice. However, the regulator framework in general doesn't
express this to lockdep.
Let's fix the issue by simply grabbing the two locks involved in the
same way we grab multiple locks elsewhere in the regulator framework:
using the "wound/wait" mechanisms.
Fixes: eaa7995c529b ("regulator: core: avoid regulator_resolve_supply() race condition")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20230329143317.RFC.v2.2.I30d8e1ca10cfbe5403884cdd192253a2e063eb9e@changeid
Signed-off-by: Mark Brown <broonie@kernel.org>
2023-03-29 14:33:54 -07:00
|
|
|
|
|
|
|
|
|
2020-08-12 03:31:35 +02:00
|
|
|
|
regulator: core: Remove regulator supply_name length limit
When creating the regulator object, associated with a consumer device,
the supply_name is string formatted into a statically sized buffer on
the stack, then strdup()'ed onto the heap.
Not only is the dance on the stack unnecessary, but when the device's
name is long we might not fit the constructed supply_name in the fixed
64 byte buffer on the stack.
One such case can be seen on the Qualcomm Rb3Gen2 board, where we find a
PCIe controller, with a PCIe switch, with a USB controller, with a USB
hub, consuming a regulator. In this example the dev->kobj.name itself is
62 characters long.
Drop the temporary buffer on the stack and kasprintf() the string
directly on the heap, both to simplify the code, and to remove the
length limitation.
Signed-off-by: Bjorn Andersson <bjorn.andersson@oss.qualcomm.com>
Link: https://patch.msgid.link/20260211-regulator-supply-name-length-v1-1-3875541c1576@oss.qualcomm.com
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-02-11 21:37:14 -06:00
|
|
|
|
2020-08-12 03:31:35 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2026-02-21 16:37:42 -08:00
|
|
|
|
2020-08-12 03:31:35 +02:00
|
|
|
|
2022-11-23 11:46:16 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2020-08-12 03:31:35 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-04-19 13:19:07 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-04-23 15:11:05 +08:00
|
|
|
|
2012-04-19 13:19:07 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-12-21 12:21:52 +00:00
|
|
|
|
|
|
|
|
|
2013-09-18 18:18:02 +05:30
|
|
|
|
|
|
|
|
|
2019-05-04 11:33:36 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-12-21 12:21:52 +00:00
|
|
|
|
|
|
|
|
|
2013-10-15 20:14:20 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-11-27 10:57:16 +08:00
|
|
|
|
2013-10-15 20:14:20 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-11-27 10:57:16 +08:00
|
|
|
|
2013-10-15 20:14:20 +01:00
|
|
|
|
|
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-01-09 11:18:07 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
2017-02-04 10:19:21 -08:00
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2017-02-04 10:19:21 -08:00
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
2011-11-18 16:47:20 +05:30
|
|
|
|
2017-02-04 10:19:21 -08:00
|
|
|
|
2011-11-18 16:47:20 +05:30
|
|
|
|
2017-06-13 16:12:47 +01:00
|
|
|
|
2012-03-30 12:24:37 +01:00
|
|
|
|
|
|
|
|
|
2011-11-18 16:47:20 +05:30
|
|
|
|
2013-10-15 20:14:20 +01:00
|
|
|
|
|
|
|
|
|
2011-11-18 16:47:20 +05:30
|
|
|
|
2025-01-09 11:18:07 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-11-18 16:47:20 +05:30
|
|
|
|
|
|
|
|
|
2012-03-30 12:24:37 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
2012-03-30 12:24:37 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
|
|
|
|
|
2017-02-04 10:19:21 -08:00
|
|
|
|
|
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
2012-03-30 12:24:37 +01:00
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
2012-03-30 12:24:37 +01:00
|
|
|
|
2017-06-13 16:12:47 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-02-04 10:19:21 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-11-18 16:47:20 +05:30
|
|
|
|
|
|
|
|
|
2015-03-24 18:56:05 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core: Avoid lockdep reports when resolving supplies
An automated bot told me that there was a potential lockdep problem
with regulators. This was on the chromeos-5.15 kernel, but I see
nothing that would be different downstream compared to upstream. The
bot said:
============================================
WARNING: possible recursive locking detected
5.15.104-lockdep-17461-gc1e499ed6604 #1 Not tainted
--------------------------------------------
kworker/u16:4/115 is trying to acquire lock:
ffffff8083110170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: create_regulator+0x398/0x7ec
but task is already holding lock:
ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(regulator_ww_class_mutex);
lock(regulator_ww_class_mutex);
*** DEADLOCK ***
May be due to missing lock nesting notation
4 locks held by kworker/u16:4/115:
#0: ffffff808006a948 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_one_work+0x520/0x1348
#1: ffffffc00e0a7cc0 ((work_completion)(&entry->work)){+.+.}-{0:0}, at: process_one_work+0x55c/0x1348
#2: ffffff80828a2260 (&dev->mutex){....}-{3:3}, at: __device_attach_async_helper+0xd0/0x2a4
#3: ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
stack backtrace:
CPU: 2 PID: 115 Comm: kworker/u16:4 Not tainted 5.15.104-lockdep-17461-gc1e499ed6604 #1 9292e52fa83c0e23762b2b3aa1bacf5787a4d5da
Hardware name: Google Quackingstick (rev0+) (DT)
Workqueue: events_unbound async_run_entry_fn
Call trace:
dump_backtrace+0x0/0x4ec
show_stack+0x34/0x50
dump_stack_lvl+0xdc/0x11c
dump_stack+0x1c/0x48
__lock_acquire+0x16d4/0x6c74
lock_acquire+0x208/0x750
__mutex_lock_common+0x11c/0x11f8
ww_mutex_lock+0xc0/0x440
create_regulator+0x398/0x7ec
regulator_resolve_supply+0x654/0x7c4
regulator_register_resolve_supply+0x30/0x120
class_for_each_device+0x1b8/0x230
regulator_register+0x17a4/0x1f40
devm_regulator_register+0x60/0xd0
reg_fixed_voltage_probe+0x728/0xaec
platform_probe+0x150/0x1c8
really_probe+0x274/0xa20
__driver_probe_device+0x1dc/0x3f4
driver_probe_device+0x78/0x1c0
__device_attach_driver+0x1ac/0x2c8
bus_for_each_drv+0x11c/0x190
__device_attach_async_helper+0x1e4/0x2a4
async_run_entry_fn+0xa0/0x3ac
process_one_work+0x638/0x1348
worker_thread+0x4a8/0x9c4
kthread+0x2e4/0x3a0
ret_from_fork+0x10/0x20
The problem was first reported soon after we made many of the
regulators probe asynchronously, though nothing I've seen implies that
the problems couldn't have also happened even without that.
I haven't personally been able to reproduce the lockdep issue, but the
issue does look somewhat legitimate. Specifically, it looks like in
regulator_resolve_supply() we are holding a "rdev" lock while calling
set_supply() -> create_regulator() which grabs the lock of a
_different_ "rdev" (the one for our supply). This is not necessarily
safe from a lockdep perspective since there is no documented ordering
between these two locks.
In reality, we should always be locking a regulator before the
supplying regulator, so I don't expect there to be any real deadlocks
in practice. However, the regulator framework in general doesn't
express this to lockdep.
Let's fix the issue by simply grabbing the two locks involved in the
same way we grab multiple locks elsewhere in the regulator framework:
using the "wound/wait" mechanisms.
Fixes: eaa7995c529b ("regulator: core: avoid regulator_resolve_supply() race condition")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20230329143317.RFC.v2.2.I30d8e1ca10cfbe5403884cdd192253a2e063eb9e@changeid
Signed-off-by: Mark Brown <broonie@kernel.org>
2023-03-29 14:33:54 -07:00
|
|
|
|
regulator: core: don't fail regulator_register() with missing required supply
Since commit 98e48cd9283d ("regulator: core: resolve supply for
boot-on/always-on regulators"), the regulator core returns
-EPROBE_DEFER if a supply can not be resolved at regulator_register()
time due to set_machine_constraints() requiring that supply (e.g.
because of always-on or boot-on).
In some hardware designs, multiple PMICs are used where individual
rails of each act as supplies for rails of the other, and vice-versa.
In such a design no PMIC driver can probe when registering one top-
level regulator device (as is common practice for almost all regulator
drivers in Linux) since that commit. Supplies are only considered when
their driver has fully bound, but because in a design like the above
two drivers / devices depend on each other, neither will have fully
bound while the other probes. The Google Pixel 6 and 6 Pro (oriole and
raven) are examples of such a design.
One way to make this work would be to register each rail as an
individual device, rather than just one top-level regulator device.
Then, fw-devlink and Linux' driver core could do their usual handling
of deferred device probe as each rail would be probed individually.
This approach was dismissed in [1] as each regulator driver would have
to take care of this itself.
Alternatively, we can change the regulator core to not fail
regulator_register() if a rail's required supply can not be resolved
while keeping the intended change from above mentioned commit, and
instead retry whenever a new rail is registered. This commit implements
such an approach:
If set_machine_constraints() requests probe deferral,
regulator_register() still succeeds and we retry setting
constraints as part of regulator_resolve_supply().
We still do not enable the regulator or allow consumers to use it
until constraints have been set (including resolution of the
supply) to prevent enabling of a regulator before its supply.
With this change, we keep track of regulators with missing required
supplies and can therefore try to resolve them again and try to set
the constraints again once more regulators become available.
Care has to be taken to not allow consumers to use regulators that
haven't had their constraints set yet. regulator_get() ensures that
and now returns -EPROBE_DEFER in that case.
The implementation is straight-forward, thanks to our newly introduced
regulator-bus. Locking in regulator_resolve_supply() has to be done
carefully, as a combination of regulator_(un)lock_two() and
regulator_(un)lock_dependent() is needed. The reason is that
set_machine_constraints() might call regulator_enable() which needs
rdev and all its dependents locked, but everything else requires to
only have rdev and its supply locked.
Link: https://lore.kernel.org/all/aRn_-o-vie_QoDXD@sirena.co.uk/ [1]
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-8-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:44 +00:00
|
|
|
|
|
|
|
|
|
2021-01-07 17:16:02 -08:00
|
|
|
|
2015-03-24 18:56:05 -07:00
|
|
|
|
2019-01-08 13:12:33 +01:00
|
|
|
|
2015-03-24 18:56:05 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-01-07 17:16:02 -08:00
|
|
|
|
regulator: core: don't fail regulator_register() with missing required supply
Since commit 98e48cd9283d ("regulator: core: resolve supply for
boot-on/always-on regulators"), the regulator core returns
-EPROBE_DEFER if a supply can not be resolved at regulator_register()
time due to set_machine_constraints() requiring that supply (e.g.
because of always-on or boot-on).
In some hardware designs, multiple PMICs are used where individual
rails of each act as supplies for rails of the other, and vice-versa.
In such a design no PMIC driver can probe when registering one top-
level regulator device (as is common practice for almost all regulator
drivers in Linux) since that commit. Supplies are only considered when
their driver has fully bound, but because in a design like the above
two drivers / devices depend on each other, neither will have fully
bound while the other probes. The Google Pixel 6 and 6 Pro (oriole and
raven) are examples of such a design.
One way to make this work would be to register each rail as an
individual device, rather than just one top-level regulator device.
Then, fw-devlink and Linux' driver core could do their usual handling
of deferred device probe as each rail would be probed individually.
This approach was dismissed in [1] as each regulator driver would have
to take care of this itself.
Alternatively, we can change the regulator core to not fail
regulator_register() if a rail's required supply can not be resolved
while keeping the intended change from above mentioned commit, and
instead retry whenever a new rail is registered. This commit implements
such an approach:
If set_machine_constraints() requests probe deferral,
regulator_register() still succeeds and we retry setting
constraints as part of regulator_resolve_supply().
We still do not enable the regulator or allow consumers to use it
until constraints have been set (including resolution of the
supply) to prevent enabling of a regulator before its supply.
With this change, we keep track of regulators with missing required
supplies and can therefore try to resolve them again and try to set
the constraints again once more regulators become available.
Care has to be taken to not allow consumers to use regulators that
haven't had their constraints set yet. regulator_get() ensures that
and now returns -EPROBE_DEFER in that case.
The implementation is straight-forward, thanks to our newly introduced
regulator-bus. Locking in regulator_resolve_supply() has to be done
carefully, as a combination of regulator_(un)lock_two() and
regulator_(un)lock_dependent() is needed. The reason is that
set_machine_constraints() might call regulator_enable() which needs
rdev and all its dependents locked, but everything else requires to
only have rdev and its supply locked.
Link: https://lore.kernel.org/all/aRn_-o-vie_QoDXD@sirena.co.uk/ [1]
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-8-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:44 +00:00
|
|
|
|
2015-03-24 18:56:05 -07:00
|
|
|
|
|
|
|
|
|
2025-01-09 11:18:07 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-02-04 10:19:21 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-10-01 10:59:48 +01:00
|
|
|
|
|
|
|
|
|
2021-01-07 17:16:02 -08:00
|
|
|
|
2015-10-01 10:59:48 +01:00
|
|
|
|
2015-07-14 11:17:26 +01:00
|
|
|
|
|
|
|
|
|
2025-03-13 11:27:39 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
2015-07-14 11:17:26 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-01-07 17:16:02 -08:00
|
|
|
|
|
|
|
|
|
2015-07-14 11:17:26 +01:00
|
|
|
|
2015-03-24 18:56:05 -07:00
|
|
|
|
|
|
|
|
|
2020-11-13 01:20:28 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-01-07 17:16:02 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-11-13 01:20:28 +01:00
|
|
|
|
2025-03-13 11:27:39 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-11-13 01:20:28 +01:00
|
|
|
|
2020-11-13 01:20:28 +01:00
|
|
|
|
|
|
|
|
|
2017-01-11 17:44:25 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-01-07 17:16:02 -08:00
|
|
|
|
|
|
|
|
|
2017-01-11 17:44:25 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-03-24 18:56:05 -07:00
|
|
|
|
|
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
|
|
|
|
|
2021-01-07 17:16:02 -08:00
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
2015-03-24 18:56:05 -07:00
|
|
|
|
2021-01-22 13:20:42 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core: Avoid lockdep reports when resolving supplies
An automated bot told me that there was a potential lockdep problem
with regulators. This was on the chromeos-5.15 kernel, but I see
nothing that would be different downstream compared to upstream. The
bot said:
============================================
WARNING: possible recursive locking detected
5.15.104-lockdep-17461-gc1e499ed6604 #1 Not tainted
--------------------------------------------
kworker/u16:4/115 is trying to acquire lock:
ffffff8083110170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: create_regulator+0x398/0x7ec
but task is already holding lock:
ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(regulator_ww_class_mutex);
lock(regulator_ww_class_mutex);
*** DEADLOCK ***
May be due to missing lock nesting notation
4 locks held by kworker/u16:4/115:
#0: ffffff808006a948 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_one_work+0x520/0x1348
#1: ffffffc00e0a7cc0 ((work_completion)(&entry->work)){+.+.}-{0:0}, at: process_one_work+0x55c/0x1348
#2: ffffff80828a2260 (&dev->mutex){....}-{3:3}, at: __device_attach_async_helper+0xd0/0x2a4
#3: ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
stack backtrace:
CPU: 2 PID: 115 Comm: kworker/u16:4 Not tainted 5.15.104-lockdep-17461-gc1e499ed6604 #1 9292e52fa83c0e23762b2b3aa1bacf5787a4d5da
Hardware name: Google Quackingstick (rev0+) (DT)
Workqueue: events_unbound async_run_entry_fn
Call trace:
dump_backtrace+0x0/0x4ec
show_stack+0x34/0x50
dump_stack_lvl+0xdc/0x11c
dump_stack+0x1c/0x48
__lock_acquire+0x16d4/0x6c74
lock_acquire+0x208/0x750
__mutex_lock_common+0x11c/0x11f8
ww_mutex_lock+0xc0/0x440
create_regulator+0x398/0x7ec
regulator_resolve_supply+0x654/0x7c4
regulator_register_resolve_supply+0x30/0x120
class_for_each_device+0x1b8/0x230
regulator_register+0x17a4/0x1f40
devm_regulator_register+0x60/0xd0
reg_fixed_voltage_probe+0x728/0xaec
platform_probe+0x150/0x1c8
really_probe+0x274/0xa20
__driver_probe_device+0x1dc/0x3f4
driver_probe_device+0x78/0x1c0
__device_attach_driver+0x1ac/0x2c8
bus_for_each_drv+0x11c/0x190
__device_attach_async_helper+0x1e4/0x2a4
async_run_entry_fn+0xa0/0x3ac
process_one_work+0x638/0x1348
worker_thread+0x4a8/0x9c4
kthread+0x2e4/0x3a0
ret_from_fork+0x10/0x20
The problem was first reported soon after we made many of the
regulators probe asynchronously, though nothing I've seen implies that
the problems couldn't have also happened even without that.
I haven't personally been able to reproduce the lockdep issue, but the
issue does look somewhat legitimate. Specifically, it looks like in
regulator_resolve_supply() we are holding a "rdev" lock while calling
set_supply() -> create_regulator() which grabs the lock of a
_different_ "rdev" (the one for our supply). This is not necessarily
safe from a lockdep perspective since there is no documented ordering
between these two locks.
In reality, we should always be locking a regulator before the
supplying regulator, so I don't expect there to be any real deadlocks
in practice. However, the regulator framework in general doesn't
express this to lockdep.
Let's fix the issue by simply grabbing the two locks involved in the
same way we grab multiple locks elsewhere in the regulator framework:
using the "wound/wait" mechanisms.
Fixes: eaa7995c529b ("regulator: core: avoid regulator_resolve_supply() race condition")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20230329143317.RFC.v2.2.I30d8e1ca10cfbe5403884cdd192253a2e063eb9e@changeid
Signed-off-by: Mark Brown <broonie@kernel.org>
2023-03-29 14:33:54 -07:00
|
|
|
|
2021-01-22 13:20:42 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core: don't fail regulator_register() with missing required supply
Since commit 98e48cd9283d ("regulator: core: resolve supply for
boot-on/always-on regulators"), the regulator core returns
-EPROBE_DEFER if a supply can not be resolved at regulator_register()
time due to set_machine_constraints() requiring that supply (e.g.
because of always-on or boot-on).
In some hardware designs, multiple PMICs are used where individual
rails of each act as supplies for rails of the other, and vice-versa.
In such a design no PMIC driver can probe when registering one top-
level regulator device (as is common practice for almost all regulator
drivers in Linux) since that commit. Supplies are only considered when
their driver has fully bound, but because in a design like the above
two drivers / devices depend on each other, neither will have fully
bound while the other probes. The Google Pixel 6 and 6 Pro (oriole and
raven) are examples of such a design.
One way to make this work would be to register each rail as an
individual device, rather than just one top-level regulator device.
Then, fw-devlink and Linux' driver core could do their usual handling
of deferred device probe as each rail would be probed individually.
This approach was dismissed in [1] as each regulator driver would have
to take care of this itself.
Alternatively, we can change the regulator core to not fail
regulator_register() if a rail's required supply can not be resolved
while keeping the intended change from above mentioned commit, and
instead retry whenever a new rail is registered. This commit implements
such an approach:
If set_machine_constraints() requests probe deferral,
regulator_register() still succeeds and we retry setting
constraints as part of regulator_resolve_supply().
We still do not enable the regulator or allow consumers to use it
until constraints have been set (including resolution of the
supply) to prevent enabling of a regulator before its supply.
With this change, we keep track of regulators with missing required
supplies and can therefore try to resolve them again and try to set
the constraints again once more regulators become available.
Care has to be taken to not allow consumers to use regulators that
haven't had their constraints set yet. regulator_get() ensures that
and now returns -EPROBE_DEFER in that case.
The implementation is straight-forward, thanks to our newly introduced
regulator-bus. Locking in regulator_resolve_supply() has to be done
carefully, as a combination of regulator_(un)lock_two() and
regulator_(un)lock_dependent() is needed. The reason is that
set_machine_constraints() might call regulator_enable() which needs
rdev and all its dependents locked, but everything else requires to
only have rdev and its supply locked.
Link: https://lore.kernel.org/all/aRn_-o-vie_QoDXD@sirena.co.uk/ [1]
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-8-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:44 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-03-24 18:56:05 -07:00
|
|
|
|
regulator: core: Avoid lockdep reports when resolving supplies
An automated bot told me that there was a potential lockdep problem
with regulators. This was on the chromeos-5.15 kernel, but I see
nothing that would be different downstream compared to upstream. The
bot said:
============================================
WARNING: possible recursive locking detected
5.15.104-lockdep-17461-gc1e499ed6604 #1 Not tainted
--------------------------------------------
kworker/u16:4/115 is trying to acquire lock:
ffffff8083110170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: create_regulator+0x398/0x7ec
but task is already holding lock:
ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(regulator_ww_class_mutex);
lock(regulator_ww_class_mutex);
*** DEADLOCK ***
May be due to missing lock nesting notation
4 locks held by kworker/u16:4/115:
#0: ffffff808006a948 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_one_work+0x520/0x1348
#1: ffffffc00e0a7cc0 ((work_completion)(&entry->work)){+.+.}-{0:0}, at: process_one_work+0x55c/0x1348
#2: ffffff80828a2260 (&dev->mutex){....}-{3:3}, at: __device_attach_async_helper+0xd0/0x2a4
#3: ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
stack backtrace:
CPU: 2 PID: 115 Comm: kworker/u16:4 Not tainted 5.15.104-lockdep-17461-gc1e499ed6604 #1 9292e52fa83c0e23762b2b3aa1bacf5787a4d5da
Hardware name: Google Quackingstick (rev0+) (DT)
Workqueue: events_unbound async_run_entry_fn
Call trace:
dump_backtrace+0x0/0x4ec
show_stack+0x34/0x50
dump_stack_lvl+0xdc/0x11c
dump_stack+0x1c/0x48
__lock_acquire+0x16d4/0x6c74
lock_acquire+0x208/0x750
__mutex_lock_common+0x11c/0x11f8
ww_mutex_lock+0xc0/0x440
create_regulator+0x398/0x7ec
regulator_resolve_supply+0x654/0x7c4
regulator_register_resolve_supply+0x30/0x120
class_for_each_device+0x1b8/0x230
regulator_register+0x17a4/0x1f40
devm_regulator_register+0x60/0xd0
reg_fixed_voltage_probe+0x728/0xaec
platform_probe+0x150/0x1c8
really_probe+0x274/0xa20
__driver_probe_device+0x1dc/0x3f4
driver_probe_device+0x78/0x1c0
__device_attach_driver+0x1ac/0x2c8
bus_for_each_drv+0x11c/0x190
__device_attach_async_helper+0x1e4/0x2a4
async_run_entry_fn+0xa0/0x3ac
process_one_work+0x638/0x1348
worker_thread+0x4a8/0x9c4
kthread+0x2e4/0x3a0
ret_from_fork+0x10/0x20
The problem was first reported soon after we made many of the
regulators probe asynchronously, though nothing I've seen implies that
the problems couldn't have also happened even without that.
I haven't personally been able to reproduce the lockdep issue, but the
issue does look somewhat legitimate. Specifically, it looks like in
regulator_resolve_supply() we are holding a "rdev" lock while calling
set_supply() -> create_regulator() which grabs the lock of a
_different_ "rdev" (the one for our supply). This is not necessarily
safe from a lockdep perspective since there is no documented ordering
between these two locks.
In reality, we should always be locking a regulator before the
supplying regulator, so I don't expect there to be any real deadlocks
in practice. However, the regulator framework in general doesn't
express this to lockdep.
Let's fix the issue by simply grabbing the two locks involved in the
same way we grab multiple locks elsewhere in the regulator framework:
using the "wound/wait" mechanisms.
Fixes: eaa7995c529b ("regulator: core: avoid regulator_resolve_supply() race condition")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20230329143317.RFC.v2.2.I30d8e1ca10cfbe5403884cdd192253a2e063eb9e@changeid
Signed-off-by: Mark Brown <broonie@kernel.org>
2023-03-29 14:33:54 -07:00
|
|
|
|
2021-01-22 13:20:42 +00:00
|
|
|
|
regulator: core: don't fail regulator_register() with missing required supply
Since commit 98e48cd9283d ("regulator: core: resolve supply for
boot-on/always-on regulators"), the regulator core returns
-EPROBE_DEFER if a supply can not be resolved at regulator_register()
time due to set_machine_constraints() requiring that supply (e.g.
because of always-on or boot-on).
In some hardware designs, multiple PMICs are used where individual
rails of each act as supplies for rails of the other, and vice-versa.
In such a design no PMIC driver can probe when registering one top-
level regulator device (as is common practice for almost all regulator
drivers in Linux) since that commit. Supplies are only considered when
their driver has fully bound, but because in a design like the above
two drivers / devices depend on each other, neither will have fully
bound while the other probes. The Google Pixel 6 and 6 Pro (oriole and
raven) are examples of such a design.
One way to make this work would be to register each rail as an
individual device, rather than just one top-level regulator device.
Then, fw-devlink and Linux' driver core could do their usual handling
of deferred device probe as each rail would be probed individually.
This approach was dismissed in [1] as each regulator driver would have
to take care of this itself.
Alternatively, we can change the regulator core to not fail
regulator_register() if a rail's required supply can not be resolved
while keeping the intended change from above mentioned commit, and
instead retry whenever a new rail is registered. This commit implements
such an approach:
If set_machine_constraints() requests probe deferral,
regulator_register() still succeeds and we retry setting
constraints as part of regulator_resolve_supply().
We still do not enable the regulator or allow consumers to use it
until constraints have been set (including resolution of the
supply) to prevent enabling of a regulator before its supply.
With this change, we keep track of regulators with missing required
supplies and can therefore try to resolve them again and try to set
the constraints again once more regulators become available.
Care has to be taken to not allow consumers to use regulators that
haven't had their constraints set yet. regulator_get() ensures that
and now returns -EPROBE_DEFER in that case.
The implementation is straight-forward, thanks to our newly introduced
regulator-bus. Locking in regulator_resolve_supply() has to be done
carefully, as a combination of regulator_(un)lock_two() and
regulator_(un)lock_dependent() is needed. The reason is that
set_machine_constraints() might call regulator_enable() which needs
rdev and all its dependents locked, but everything else requires to
only have rdev and its supply locked.
Link: https://lore.kernel.org/all/aRn_-o-vie_QoDXD@sirena.co.uk/ [1]
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-8-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:44 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-01-22 13:20:42 +00:00
|
|
|
|
|
|
|
|
|
2025-10-01 12:56:49 +02:00
|
|
|
|
regulator: core: don't fail regulator_register() with missing required supply
Since commit 98e48cd9283d ("regulator: core: resolve supply for
boot-on/always-on regulators"), the regulator core returns
-EPROBE_DEFER if a supply can not be resolved at regulator_register()
time due to set_machine_constraints() requiring that supply (e.g.
because of always-on or boot-on).
In some hardware designs, multiple PMICs are used where individual
rails of each act as supplies for rails of the other, and vice-versa.
In such a design no PMIC driver can probe when registering one top-
level regulator device (as is common practice for almost all regulator
drivers in Linux) since that commit. Supplies are only considered when
their driver has fully bound, but because in a design like the above
two drivers / devices depend on each other, neither will have fully
bound while the other probes. The Google Pixel 6 and 6 Pro (oriole and
raven) are examples of such a design.
One way to make this work would be to register each rail as an
individual device, rather than just one top-level regulator device.
Then, fw-devlink and Linux' driver core could do their usual handling
of deferred device probe as each rail would be probed individually.
This approach was dismissed in [1] as each regulator driver would have
to take care of this itself.
Alternatively, we can change the regulator core to not fail
regulator_register() if a rail's required supply can not be resolved
while keeping the intended change from above mentioned commit, and
instead retry whenever a new rail is registered. This commit implements
such an approach:
If set_machine_constraints() requests probe deferral,
regulator_register() still succeeds and we retry setting
constraints as part of regulator_resolve_supply().
We still do not enable the regulator or allow consumers to use it
until constraints have been set (including resolution of the
supply) to prevent enabling of a regulator before its supply.
With this change, we keep track of regulators with missing required
supplies and can therefore try to resolve them again and try to set
the constraints again once more regulators become available.
Care has to be taken to not allow consumers to use regulators that
haven't had their constraints set yet. regulator_get() ensures that
and now returns -EPROBE_DEFER in that case.
The implementation is straight-forward, thanks to our newly introduced
regulator-bus. Locking in regulator_resolve_supply() has to be done
carefully, as a combination of regulator_(un)lock_two() and
regulator_(un)lock_dependent() is needed. The reason is that
set_machine_constraints() might call regulator_enable() which needs
rdev and all its dependents locked, but everything else requires to
only have rdev and its supply locked.
Link: https://lore.kernel.org/all/aRn_-o-vie_QoDXD@sirena.co.uk/ [1]
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-8-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:44 +00:00
|
|
|
|
|
|
|
|
|
2025-10-01 12:56:49 +02:00
|
|
|
|
regulator: core: don't fail regulator_register() with missing required supply
Since commit 98e48cd9283d ("regulator: core: resolve supply for
boot-on/always-on regulators"), the regulator core returns
-EPROBE_DEFER if a supply can not be resolved at regulator_register()
time due to set_machine_constraints() requiring that supply (e.g.
because of always-on or boot-on).
In some hardware designs, multiple PMICs are used where individual
rails of each act as supplies for rails of the other, and vice-versa.
In such a design no PMIC driver can probe when registering one top-
level regulator device (as is common practice for almost all regulator
drivers in Linux) since that commit. Supplies are only considered when
their driver has fully bound, but because in a design like the above
two drivers / devices depend on each other, neither will have fully
bound while the other probes. The Google Pixel 6 and 6 Pro (oriole and
raven) are examples of such a design.
One way to make this work would be to register each rail as an
individual device, rather than just one top-level regulator device.
Then, fw-devlink and Linux' driver core could do their usual handling
of deferred device probe as each rail would be probed individually.
This approach was dismissed in [1] as each regulator driver would have
to take care of this itself.
Alternatively, we can change the regulator core to not fail
regulator_register() if a rail's required supply can not be resolved
while keeping the intended change from above mentioned commit, and
instead retry whenever a new rail is registered. This commit implements
such an approach:
If set_machine_constraints() requests probe deferral,
regulator_register() still succeeds and we retry setting
constraints as part of regulator_resolve_supply().
We still do not enable the regulator or allow consumers to use it
until constraints have been set (including resolution of the
supply) to prevent enabling of a regulator before its supply.
With this change, we keep track of regulators with missing required
supplies and can therefore try to resolve them again and try to set
the constraints again once more regulators become available.
Care has to be taken to not allow consumers to use regulators that
haven't had their constraints set yet. regulator_get() ensures that
and now returns -EPROBE_DEFER in that case.
The implementation is straight-forward, thanks to our newly introduced
regulator-bus. Locking in regulator_resolve_supply() has to be done
carefully, as a combination of regulator_(un)lock_two() and
regulator_(un)lock_dependent() is needed. The reason is that
set_machine_constraints() might call regulator_enable() which needs
rdev and all its dependents locked, but everything else requires to
only have rdev and its supply locked.
Link: https://lore.kernel.org/all/aRn_-o-vie_QoDXD@sirena.co.uk/ [1]
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-8-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:44 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026-01-09 08:38:42 +00:00
|
|
|
|
regulator: core: don't fail regulator_register() with missing required supply
Since commit 98e48cd9283d ("regulator: core: resolve supply for
boot-on/always-on regulators"), the regulator core returns
-EPROBE_DEFER if a supply can not be resolved at regulator_register()
time due to set_machine_constraints() requiring that supply (e.g.
because of always-on or boot-on).
In some hardware designs, multiple PMICs are used where individual
rails of each act as supplies for rails of the other, and vice-versa.
In such a design no PMIC driver can probe when registering one top-
level regulator device (as is common practice for almost all regulator
drivers in Linux) since that commit. Supplies are only considered when
their driver has fully bound, but because in a design like the above
two drivers / devices depend on each other, neither will have fully
bound while the other probes. The Google Pixel 6 and 6 Pro (oriole and
raven) are examples of such a design.
One way to make this work would be to register each rail as an
individual device, rather than just one top-level regulator device.
Then, fw-devlink and Linux' driver core could do their usual handling
of deferred device probe as each rail would be probed individually.
This approach was dismissed in [1] as each regulator driver would have
to take care of this itself.
Alternatively, we can change the regulator core to not fail
regulator_register() if a rail's required supply can not be resolved
while keeping the intended change from above mentioned commit, and
instead retry whenever a new rail is registered. This commit implements
such an approach:
If set_machine_constraints() requests probe deferral,
regulator_register() still succeeds and we retry setting
constraints as part of regulator_resolve_supply().
We still do not enable the regulator or allow consumers to use it
until constraints have been set (including resolution of the
supply) to prevent enabling of a regulator before its supply.
With this change, we keep track of regulators with missing required
supplies and can therefore try to resolve them again and try to set
the constraints again once more regulators become available.
Care has to be taken to not allow consumers to use regulators that
haven't had their constraints set yet. regulator_get() ensures that
and now returns -EPROBE_DEFER in that case.
The implementation is straight-forward, thanks to our newly introduced
regulator-bus. Locking in regulator_resolve_supply() has to be done
carefully, as a combination of regulator_(un)lock_two() and
regulator_(un)lock_dependent() is needed. The reason is that
set_machine_constraints() might call regulator_enable() which needs
rdev and all its dependents locked, but everything else requires to
only have rdev and its supply locked.
Link: https://lore.kernel.org/all/aRn_-o-vie_QoDXD@sirena.co.uk/ [1]
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-8-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:44 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-10-01 12:56:49 +02:00
|
|
|
|
regulator: core: don't fail regulator_register() with missing required supply
Since commit 98e48cd9283d ("regulator: core: resolve supply for
boot-on/always-on regulators"), the regulator core returns
-EPROBE_DEFER if a supply can not be resolved at regulator_register()
time due to set_machine_constraints() requiring that supply (e.g.
because of always-on or boot-on).
In some hardware designs, multiple PMICs are used where individual
rails of each act as supplies for rails of the other, and vice-versa.
In such a design no PMIC driver can probe when registering one top-
level regulator device (as is common practice for almost all regulator
drivers in Linux) since that commit. Supplies are only considered when
their driver has fully bound, but because in a design like the above
two drivers / devices depend on each other, neither will have fully
bound while the other probes. The Google Pixel 6 and 6 Pro (oriole and
raven) are examples of such a design.
One way to make this work would be to register each rail as an
individual device, rather than just one top-level regulator device.
Then, fw-devlink and Linux' driver core could do their usual handling
of deferred device probe as each rail would be probed individually.
This approach was dismissed in [1] as each regulator driver would have
to take care of this itself.
Alternatively, we can change the regulator core to not fail
regulator_register() if a rail's required supply can not be resolved
while keeping the intended change from above mentioned commit, and
instead retry whenever a new rail is registered. This commit implements
such an approach:
If set_machine_constraints() requests probe deferral,
regulator_register() still succeeds and we retry setting
constraints as part of regulator_resolve_supply().
We still do not enable the regulator or allow consumers to use it
until constraints have been set (including resolution of the
supply) to prevent enabling of a regulator before its supply.
With this change, we keep track of regulators with missing required
supplies and can therefore try to resolve them again and try to set
the constraints again once more regulators become available.
Care has to be taken to not allow consumers to use regulators that
haven't had their constraints set yet. regulator_get() ensures that
and now returns -EPROBE_DEFER in that case.
The implementation is straight-forward, thanks to our newly introduced
regulator-bus. Locking in regulator_resolve_supply() has to be done
carefully, as a combination of regulator_(un)lock_two() and
regulator_(un)lock_dependent() is needed. The reason is that
set_machine_constraints() might call regulator_enable() which needs
rdev and all its dependents locked, but everything else requires to
only have rdev and its supply locked.
Link: https://lore.kernel.org/all/aRn_-o-vie_QoDXD@sirena.co.uk/ [1]
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-8-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:44 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026-01-09 08:38:42 +00:00
|
|
|
|
regulator: core: don't fail regulator_register() with missing required supply
Since commit 98e48cd9283d ("regulator: core: resolve supply for
boot-on/always-on regulators"), the regulator core returns
-EPROBE_DEFER if a supply can not be resolved at regulator_register()
time due to set_machine_constraints() requiring that supply (e.g.
because of always-on or boot-on).
In some hardware designs, multiple PMICs are used where individual
rails of each act as supplies for rails of the other, and vice-versa.
In such a design no PMIC driver can probe when registering one top-
level regulator device (as is common practice for almost all regulator
drivers in Linux) since that commit. Supplies are only considered when
their driver has fully bound, but because in a design like the above
two drivers / devices depend on each other, neither will have fully
bound while the other probes. The Google Pixel 6 and 6 Pro (oriole and
raven) are examples of such a design.
One way to make this work would be to register each rail as an
individual device, rather than just one top-level regulator device.
Then, fw-devlink and Linux' driver core could do their usual handling
of deferred device probe as each rail would be probed individually.
This approach was dismissed in [1] as each regulator driver would have
to take care of this itself.
Alternatively, we can change the regulator core to not fail
regulator_register() if a rail's required supply can not be resolved
while keeping the intended change from above mentioned commit, and
instead retry whenever a new rail is registered. This commit implements
such an approach:
If set_machine_constraints() requests probe deferral,
regulator_register() still succeeds and we retry setting
constraints as part of regulator_resolve_supply().
We still do not enable the regulator or allow consumers to use it
until constraints have been set (including resolution of the
supply) to prevent enabling of a regulator before its supply.
With this change, we keep track of regulators with missing required
supplies and can therefore try to resolve them again and try to set
the constraints again once more regulators become available.
Care has to be taken to not allow consumers to use regulators that
haven't had their constraints set yet. regulator_get() ensures that
and now returns -EPROBE_DEFER in that case.
The implementation is straight-forward, thanks to our newly introduced
regulator-bus. Locking in regulator_resolve_supply() has to be done
carefully, as a combination of regulator_(un)lock_two() and
regulator_(un)lock_dependent() is needed. The reason is that
set_machine_constraints() might call regulator_enable() which needs
rdev and all its dependents locked, but everything else requires to
only have rdev and its supply locked.
Link: https://lore.kernel.org/all/aRn_-o-vie_QoDXD@sirena.co.uk/ [1]
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-8-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:44 +00:00
|
|
|
|
2026-01-09 08:38:42 +00:00
|
|
|
|
regulator: core: don't fail regulator_register() with missing required supply
Since commit 98e48cd9283d ("regulator: core: resolve supply for
boot-on/always-on regulators"), the regulator core returns
-EPROBE_DEFER if a supply can not be resolved at regulator_register()
time due to set_machine_constraints() requiring that supply (e.g.
because of always-on or boot-on).
In some hardware designs, multiple PMICs are used where individual
rails of each act as supplies for rails of the other, and vice-versa.
In such a design no PMIC driver can probe when registering one top-
level regulator device (as is common practice for almost all regulator
drivers in Linux) since that commit. Supplies are only considered when
their driver has fully bound, but because in a design like the above
two drivers / devices depend on each other, neither will have fully
bound while the other probes. The Google Pixel 6 and 6 Pro (oriole and
raven) are examples of such a design.
One way to make this work would be to register each rail as an
individual device, rather than just one top-level regulator device.
Then, fw-devlink and Linux' driver core could do their usual handling
of deferred device probe as each rail would be probed individually.
This approach was dismissed in [1] as each regulator driver would have
to take care of this itself.
Alternatively, we can change the regulator core to not fail
regulator_register() if a rail's required supply can not be resolved
while keeping the intended change from above mentioned commit, and
instead retry whenever a new rail is registered. This commit implements
such an approach:
If set_machine_constraints() requests probe deferral,
regulator_register() still succeeds and we retry setting
constraints as part of regulator_resolve_supply().
We still do not enable the regulator or allow consumers to use it
until constraints have been set (including resolution of the
supply) to prevent enabling of a regulator before its supply.
With this change, we keep track of regulators with missing required
supplies and can therefore try to resolve them again and try to set
the constraints again once more regulators become available.
Care has to be taken to not allow consumers to use regulators that
haven't had their constraints set yet. regulator_get() ensures that
and now returns -EPROBE_DEFER in that case.
The implementation is straight-forward, thanks to our newly introduced
regulator-bus. Locking in regulator_resolve_supply() has to be done
carefully, as a combination of regulator_(un)lock_two() and
regulator_(un)lock_dependent() is needed. The reason is that
set_machine_constraints() might call regulator_enable() which needs
rdev and all its dependents locked, but everything else requires to
only have rdev and its supply locked.
Link: https://lore.kernel.org/all/aRn_-o-vie_QoDXD@sirena.co.uk/ [1]
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-8-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:44 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026-01-09 08:38:42 +00:00
|
|
|
|
regulator: core: don't fail regulator_register() with missing required supply
Since commit 98e48cd9283d ("regulator: core: resolve supply for
boot-on/always-on regulators"), the regulator core returns
-EPROBE_DEFER if a supply can not be resolved at regulator_register()
time due to set_machine_constraints() requiring that supply (e.g.
because of always-on or boot-on).
In some hardware designs, multiple PMICs are used where individual
rails of each act as supplies for rails of the other, and vice-versa.
In such a design no PMIC driver can probe when registering one top-
level regulator device (as is common practice for almost all regulator
drivers in Linux) since that commit. Supplies are only considered when
their driver has fully bound, but because in a design like the above
two drivers / devices depend on each other, neither will have fully
bound while the other probes. The Google Pixel 6 and 6 Pro (oriole and
raven) are examples of such a design.
One way to make this work would be to register each rail as an
individual device, rather than just one top-level regulator device.
Then, fw-devlink and Linux' driver core could do their usual handling
of deferred device probe as each rail would be probed individually.
This approach was dismissed in [1] as each regulator driver would have
to take care of this itself.
Alternatively, we can change the regulator core to not fail
regulator_register() if a rail's required supply can not be resolved
while keeping the intended change from above mentioned commit, and
instead retry whenever a new rail is registered. This commit implements
such an approach:
If set_machine_constraints() requests probe deferral,
regulator_register() still succeeds and we retry setting
constraints as part of regulator_resolve_supply().
We still do not enable the regulator or allow consumers to use it
until constraints have been set (including resolution of the
supply) to prevent enabling of a regulator before its supply.
With this change, we keep track of regulators with missing required
supplies and can therefore try to resolve them again and try to set
the constraints again once more regulators become available.
Care has to be taken to not allow consumers to use regulators that
haven't had their constraints set yet. regulator_get() ensures that
and now returns -EPROBE_DEFER in that case.
The implementation is straight-forward, thanks to our newly introduced
regulator-bus. Locking in regulator_resolve_supply() has to be done
carefully, as a combination of regulator_(un)lock_two() and
regulator_(un)lock_dependent() is needed. The reason is that
set_machine_constraints() might call regulator_enable() which needs
rdev and all its dependents locked, but everything else requires to
only have rdev and its supply locked.
Link: https://lore.kernel.org/all/aRn_-o-vie_QoDXD@sirena.co.uk/ [1]
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-8-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:44 +00:00
|
|
|
|
2026-01-09 08:38:42 +00:00
|
|
|
|
regulator: core: don't fail regulator_register() with missing required supply
Since commit 98e48cd9283d ("regulator: core: resolve supply for
boot-on/always-on regulators"), the regulator core returns
-EPROBE_DEFER if a supply can not be resolved at regulator_register()
time due to set_machine_constraints() requiring that supply (e.g.
because of always-on or boot-on).
In some hardware designs, multiple PMICs are used where individual
rails of each act as supplies for rails of the other, and vice-versa.
In such a design no PMIC driver can probe when registering one top-
level regulator device (as is common practice for almost all regulator
drivers in Linux) since that commit. Supplies are only considered when
their driver has fully bound, but because in a design like the above
two drivers / devices depend on each other, neither will have fully
bound while the other probes. The Google Pixel 6 and 6 Pro (oriole and
raven) are examples of such a design.
One way to make this work would be to register each rail as an
individual device, rather than just one top-level regulator device.
Then, fw-devlink and Linux' driver core could do their usual handling
of deferred device probe as each rail would be probed individually.
This approach was dismissed in [1] as each regulator driver would have
to take care of this itself.
Alternatively, we can change the regulator core to not fail
regulator_register() if a rail's required supply can not be resolved
while keeping the intended change from above mentioned commit, and
instead retry whenever a new rail is registered. This commit implements
such an approach:
If set_machine_constraints() requests probe deferral,
regulator_register() still succeeds and we retry setting
constraints as part of regulator_resolve_supply().
We still do not enable the regulator or allow consumers to use it
until constraints have been set (including resolution of the
supply) to prevent enabling of a regulator before its supply.
With this change, we keep track of regulators with missing required
supplies and can therefore try to resolve them again and try to set
the constraints again once more regulators become available.
Care has to be taken to not allow consumers to use regulators that
haven't had their constraints set yet. regulator_get() ensures that
and now returns -EPROBE_DEFER in that case.
The implementation is straight-forward, thanks to our newly introduced
regulator-bus. Locking in regulator_resolve_supply() has to be done
carefully, as a combination of regulator_(un)lock_two() and
regulator_(un)lock_dependent() is needed. The reason is that
set_machine_constraints() might call regulator_enable() which needs
rdev and all its dependents locked, but everything else requires to
only have rdev and its supply locked.
Link: https://lore.kernel.org/all/aRn_-o-vie_QoDXD@sirena.co.uk/ [1]
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-8-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:44 +00:00
|
|
|
|
|
|
|
|
|
2021-01-22 13:20:42 +00:00
|
|
|
|
2025-03-05 17:05:04 +01:00
|
|
|
|
regulator: core: don't fail regulator_register() with missing required supply
Since commit 98e48cd9283d ("regulator: core: resolve supply for
boot-on/always-on regulators"), the regulator core returns
-EPROBE_DEFER if a supply can not be resolved at regulator_register()
time due to set_machine_constraints() requiring that supply (e.g.
because of always-on or boot-on).
In some hardware designs, multiple PMICs are used where individual
rails of each act as supplies for rails of the other, and vice-versa.
In such a design no PMIC driver can probe when registering one top-
level regulator device (as is common practice for almost all regulator
drivers in Linux) since that commit. Supplies are only considered when
their driver has fully bound, but because in a design like the above
two drivers / devices depend on each other, neither will have fully
bound while the other probes. The Google Pixel 6 and 6 Pro (oriole and
raven) are examples of such a design.
One way to make this work would be to register each rail as an
individual device, rather than just one top-level regulator device.
Then, fw-devlink and Linux' driver core could do their usual handling
of deferred device probe as each rail would be probed individually.
This approach was dismissed in [1] as each regulator driver would have
to take care of this itself.
Alternatively, we can change the regulator core to not fail
regulator_register() if a rail's required supply can not be resolved
while keeping the intended change from above mentioned commit, and
instead retry whenever a new rail is registered. This commit implements
such an approach:
If set_machine_constraints() requests probe deferral,
regulator_register() still succeeds and we retry setting
constraints as part of regulator_resolve_supply().
We still do not enable the regulator or allow consumers to use it
until constraints have been set (including resolution of the
supply) to prevent enabling of a regulator before its supply.
With this change, we keep track of regulators with missing required
supplies and can therefore try to resolve them again and try to set
the constraints again once more regulators become available.
Care has to be taken to not allow consumers to use regulators that
haven't had their constraints set yet. regulator_get() ensures that
and now returns -EPROBE_DEFER in that case.
The implementation is straight-forward, thanks to our newly introduced
regulator-bus. Locking in regulator_resolve_supply() has to be done
carefully, as a combination of regulator_(un)lock_two() and
regulator_(un)lock_dependent() is needed. The reason is that
set_machine_constraints() might call regulator_enable() which needs
rdev and all its dependents locked, but everything else requires to
only have rdev and its supply locked.
Link: https://lore.kernel.org/all/aRn_-o-vie_QoDXD@sirena.co.uk/ [1]
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-8-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:44 +00:00
|
|
|
|
2025-03-05 17:05:04 +01:00
|
|
|
|
2021-01-07 17:16:02 -08:00
|
|
|
|
|
|
|
|
|
regulator: core: don't fail regulator_register() with missing required supply
Since commit 98e48cd9283d ("regulator: core: resolve supply for
boot-on/always-on regulators"), the regulator core returns
-EPROBE_DEFER if a supply can not be resolved at regulator_register()
time due to set_machine_constraints() requiring that supply (e.g.
because of always-on or boot-on).
In some hardware designs, multiple PMICs are used where individual
rails of each act as supplies for rails of the other, and vice-versa.
In such a design no PMIC driver can probe when registering one top-
level regulator device (as is common practice for almost all regulator
drivers in Linux) since that commit. Supplies are only considered when
their driver has fully bound, but because in a design like the above
two drivers / devices depend on each other, neither will have fully
bound while the other probes. The Google Pixel 6 and 6 Pro (oriole and
raven) are examples of such a design.
One way to make this work would be to register each rail as an
individual device, rather than just one top-level regulator device.
Then, fw-devlink and Linux' driver core could do their usual handling
of deferred device probe as each rail would be probed individually.
This approach was dismissed in [1] as each regulator driver would have
to take care of this itself.
Alternatively, we can change the regulator core to not fail
regulator_register() if a rail's required supply can not be resolved
while keeping the intended change from above mentioned commit, and
instead retry whenever a new rail is registered. This commit implements
such an approach:
If set_machine_constraints() requests probe deferral,
regulator_register() still succeeds and we retry setting
constraints as part of regulator_resolve_supply().
We still do not enable the regulator or allow consumers to use it
until constraints have been set (including resolution of the
supply) to prevent enabling of a regulator before its supply.
With this change, we keep track of regulators with missing required
supplies and can therefore try to resolve them again and try to set
the constraints again once more regulators become available.
Care has to be taken to not allow consumers to use regulators that
haven't had their constraints set yet. regulator_get() ensures that
and now returns -EPROBE_DEFER in that case.
The implementation is straight-forward, thanks to our newly introduced
regulator-bus. Locking in regulator_resolve_supply() has to be done
carefully, as a combination of regulator_(un)lock_two() and
regulator_(un)lock_dependent() is needed. The reason is that
set_machine_constraints() might call regulator_enable() which needs
rdev and all its dependents locked, but everything else requires to
only have rdev and its supply locked.
Link: https://lore.kernel.org/all/aRn_-o-vie_QoDXD@sirena.co.uk/ [1]
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-8-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:44 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-03-24 18:56:05 -07:00
|
|
|
|
|
|
|
|
|
2024-09-11 15:27:41 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2017-02-03 13:56:02 -08:00
|
|
|
|
|
|
|
|
|
2024-09-11 15:27:41 +08:00
|
|
|
|
2017-02-03 13:56:02 -08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2024-08-22 15:20:44 +08:00
|
|
|
|
2024-09-11 15:27:41 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2024-09-11 15:27:41 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-02-06 19:56:14 -08:00
|
|
|
|
|
|
|
|
|
2009-06-17 17:56:39 +01:00
|
|
|
|
2017-02-06 19:56:14 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-01-27 17:34:07 +00:00
|
|
|
|
2017-02-06 19:56:14 -08:00
|
|
|
|
|
|
|
|
|
2024-05-07 12:47:02 +02:00
|
|
|
|
2017-02-06 19:56:14 -08:00
|
|
|
|
|
|
|
|
|
2011-11-18 16:47:20 +05:30
|
|
|
|
2017-02-06 19:56:14 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-03-13 11:27:39 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-02-06 19:56:14 -08:00
|
|
|
|
|
|
|
|
|
2013-09-23 16:12:52 +01:00
|
|
|
|
2017-02-06 19:56:14 -08:00
|
|
|
|
|
|
|
|
|
2024-05-07 12:47:02 +02:00
|
|
|
|
2020-08-23 17:36:59 -05:00
|
|
|
|
2013-04-16 16:45:16 -05:00
|
|
|
|
2017-02-06 19:56:14 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2010-02-12 10:18:08 +00:00
|
|
|
|
|
|
|
|
|
2009-07-21 16:00:23 +01:00
|
|
|
|
|
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
|
|
|
|
|
2009-07-21 16:00:23 +01:00
|
|
|
|
|
|
|
|
|
2017-02-03 13:56:02 -08:00
|
|
|
|
2009-07-21 16:00:23 +01:00
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
|
|
|
|
|
2009-07-21 16:00:23 +01:00
|
|
|
|
|
|
|
|
|
2018-10-05 18:36:31 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-03-24 18:56:05 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
|
|
|
|
|
2015-03-24 18:56:05 -07:00
|
|
|
|
|
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
2017-02-03 13:56:00 -08:00
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
regulator: core: Avoid lockdep reports when resolving supplies
An automated bot told me that there was a potential lockdep problem
with regulators. This was on the chromeos-5.15 kernel, but I see
nothing that would be different downstream compared to upstream. The
bot said:
============================================
WARNING: possible recursive locking detected
5.15.104-lockdep-17461-gc1e499ed6604 #1 Not tainted
--------------------------------------------
kworker/u16:4/115 is trying to acquire lock:
ffffff8083110170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: create_regulator+0x398/0x7ec
but task is already holding lock:
ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(regulator_ww_class_mutex);
lock(regulator_ww_class_mutex);
*** DEADLOCK ***
May be due to missing lock nesting notation
4 locks held by kworker/u16:4/115:
#0: ffffff808006a948 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_one_work+0x520/0x1348
#1: ffffffc00e0a7cc0 ((work_completion)(&entry->work)){+.+.}-{0:0}, at: process_one_work+0x55c/0x1348
#2: ffffff80828a2260 (&dev->mutex){....}-{3:3}, at: __device_attach_async_helper+0xd0/0x2a4
#3: ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
stack backtrace:
CPU: 2 PID: 115 Comm: kworker/u16:4 Not tainted 5.15.104-lockdep-17461-gc1e499ed6604 #1 9292e52fa83c0e23762b2b3aa1bacf5787a4d5da
Hardware name: Google Quackingstick (rev0+) (DT)
Workqueue: events_unbound async_run_entry_fn
Call trace:
dump_backtrace+0x0/0x4ec
show_stack+0x34/0x50
dump_stack_lvl+0xdc/0x11c
dump_stack+0x1c/0x48
__lock_acquire+0x16d4/0x6c74
lock_acquire+0x208/0x750
__mutex_lock_common+0x11c/0x11f8
ww_mutex_lock+0xc0/0x440
create_regulator+0x398/0x7ec
regulator_resolve_supply+0x654/0x7c4
regulator_register_resolve_supply+0x30/0x120
class_for_each_device+0x1b8/0x230
regulator_register+0x17a4/0x1f40
devm_regulator_register+0x60/0xd0
reg_fixed_voltage_probe+0x728/0xaec
platform_probe+0x150/0x1c8
really_probe+0x274/0xa20
__driver_probe_device+0x1dc/0x3f4
driver_probe_device+0x78/0x1c0
__device_attach_driver+0x1ac/0x2c8
bus_for_each_drv+0x11c/0x190
__device_attach_async_helper+0x1e4/0x2a4
async_run_entry_fn+0xa0/0x3ac
process_one_work+0x638/0x1348
worker_thread+0x4a8/0x9c4
kthread+0x2e4/0x3a0
ret_from_fork+0x10/0x20
The problem was first reported soon after we made many of the
regulators probe asynchronously, though nothing I've seen implies that
the problems couldn't have also happened even without that.
I haven't personally been able to reproduce the lockdep issue, but the
issue does look somewhat legitimate. Specifically, it looks like in
regulator_resolve_supply() we are holding a "rdev" lock while calling
set_supply() -> create_regulator() which grabs the lock of a
_different_ "rdev" (the one for our supply). This is not necessarily
safe from a lockdep perspective since there is no documented ordering
between these two locks.
In reality, we should always be locking a regulator before the
supplying regulator, so I don't expect there to be any real deadlocks
in practice. However, the regulator framework in general doesn't
express this to lockdep.
Let's fix the issue by simply grabbing the two locks involved in the
same way we grab multiple locks elsewhere in the regulator framework:
using the "wound/wait" mechanisms.
Fixes: eaa7995c529b ("regulator: core: avoid regulator_resolve_supply() race condition")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20230329143317.RFC.v2.2.I30d8e1ca10cfbe5403884cdd192253a2e063eb9e@changeid
Signed-off-by: Mark Brown <broonie@kernel.org>
2023-03-29 14:33:54 -07:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
regulator: core: Avoid lockdep reports when resolving supplies
An automated bot told me that there was a potential lockdep problem
with regulators. This was on the chromeos-5.15 kernel, but I see
nothing that would be different downstream compared to upstream. The
bot said:
============================================
WARNING: possible recursive locking detected
5.15.104-lockdep-17461-gc1e499ed6604 #1 Not tainted
--------------------------------------------
kworker/u16:4/115 is trying to acquire lock:
ffffff8083110170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: create_regulator+0x398/0x7ec
but task is already holding lock:
ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(regulator_ww_class_mutex);
lock(regulator_ww_class_mutex);
*** DEADLOCK ***
May be due to missing lock nesting notation
4 locks held by kworker/u16:4/115:
#0: ffffff808006a948 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_one_work+0x520/0x1348
#1: ffffffc00e0a7cc0 ((work_completion)(&entry->work)){+.+.}-{0:0}, at: process_one_work+0x55c/0x1348
#2: ffffff80828a2260 (&dev->mutex){....}-{3:3}, at: __device_attach_async_helper+0xd0/0x2a4
#3: ffffff808378e170 (regulator_ww_class_mutex){+.+.}-{3:3}, at: ww_mutex_trylock+0x3c/0x7b8
stack backtrace:
CPU: 2 PID: 115 Comm: kworker/u16:4 Not tainted 5.15.104-lockdep-17461-gc1e499ed6604 #1 9292e52fa83c0e23762b2b3aa1bacf5787a4d5da
Hardware name: Google Quackingstick (rev0+) (DT)
Workqueue: events_unbound async_run_entry_fn
Call trace:
dump_backtrace+0x0/0x4ec
show_stack+0x34/0x50
dump_stack_lvl+0xdc/0x11c
dump_stack+0x1c/0x48
__lock_acquire+0x16d4/0x6c74
lock_acquire+0x208/0x750
__mutex_lock_common+0x11c/0x11f8
ww_mutex_lock+0xc0/0x440
create_regulator+0x398/0x7ec
regulator_resolve_supply+0x654/0x7c4
regulator_register_resolve_supply+0x30/0x120
class_for_each_device+0x1b8/0x230
regulator_register+0x17a4/0x1f40
devm_regulator_register+0x60/0xd0
reg_fixed_voltage_probe+0x728/0xaec
platform_probe+0x150/0x1c8
really_probe+0x274/0xa20
__driver_probe_device+0x1dc/0x3f4
driver_probe_device+0x78/0x1c0
__device_attach_driver+0x1ac/0x2c8
bus_for_each_drv+0x11c/0x190
__device_attach_async_helper+0x1e4/0x2a4
async_run_entry_fn+0xa0/0x3ac
process_one_work+0x638/0x1348
worker_thread+0x4a8/0x9c4
kthread+0x2e4/0x3a0
ret_from_fork+0x10/0x20
The problem was first reported soon after we made many of the
regulators probe asynchronously, though nothing I've seen implies that
the problems couldn't have also happened even without that.
I haven't personally been able to reproduce the lockdep issue, but the
issue does look somewhat legitimate. Specifically, it looks like in
regulator_resolve_supply() we are holding a "rdev" lock while calling
set_supply() -> create_regulator() which grabs the lock of a
_different_ "rdev" (the one for our supply). This is not necessarily
safe from a lockdep perspective since there is no documented ordering
between these two locks.
In reality, we should always be locking a regulator before the
supplying regulator, so I don't expect there to be any real deadlocks
in practice. However, the regulator framework in general doesn't
express this to lockdep.
Let's fix the issue by simply grabbing the two locks involved in the
same way we grab multiple locks elsewhere in the regulator framework:
using the "wound/wait" mechanisms.
Fixes: eaa7995c529b ("regulator: core: avoid regulator_resolve_supply() race condition")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20230329143317.RFC.v2.2.I30d8e1ca10cfbe5403884cdd192253a2e063eb9e@changeid
Signed-off-by: Mark Brown <broonie@kernel.org>
2023-03-29 14:33:54 -07:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-11-24 22:58:35 +08:00
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2025-03-05 17:05:04 +01:00
|
|
|
|
|
|
|
|
|
2009-07-21 16:00:23 +01:00
|
|
|
|
2017-02-03 13:56:02 -08:00
|
|
|
|
2009-07-21 16:00:23 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-05-04 21:31:52 -07:00
|
|
|
|
2009-07-21 16:00:23 +01:00
|
|
|
|
2022-05-04 21:31:52 -07:00
|
|
|
|
2024-03-12 10:16:38 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-05-04 21:31:52 -07:00
|
|
|
|
2009-07-21 16:00:23 +01:00
|
|
|
|
2022-05-04 21:31:52 -07:00
|
|
|
|
|
|
|
|
|
2009-07-21 16:00:23 +01:00
|
|
|
|
|
|
|
|
|
2019-11-14 16:04:38 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-07-05 14:25:56 +00:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2009-07-21 16:00:23 +01:00
|
|
|
|
2024-09-11 15:27:41 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-07-21 16:00:23 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-01-03 16:55:41 +00:00
|
|
|
|
2009-07-21 16:00:23 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:23 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-07-21 16:00:23 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-02-03 13:56:02 -08:00
|
|
|
|
2009-07-21 16:00:23 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2009-07-21 16:00:23 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:23 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-07-21 16:00:23 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-01-03 16:55:41 +00:00
|
|
|
|
2009-07-21 16:00:23 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:23 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-07-21 16:00:23 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-02-03 13:56:02 -08:00
|
|
|
|
2009-07-21 16:00:23 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2013-07-29 21:42:42 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-01-03 16:55:41 +00:00
|
|
|
|
2013-07-29 21:42:42 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:23 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2013-07-29 21:42:42 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-02-03 13:56:02 -08:00
|
|
|
|
2013-07-29 21:42:42 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-07-15 21:20:51 -07:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2020-07-15 21:20:51 -07:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2011-06-19 13:33:16 +01:00
|
|
|
|
|
|
|
|
|
2018-07-05 14:25:56 +00:00
|
|
|
|
2019-11-14 16:04:38 -08:00
|
|
|
|
|
|
|
|
|
2018-07-05 14:25:56 +00:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2018-07-05 14:25:56 +00:00
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2009-07-21 16:00:23 +01:00
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2009-07-21 16:00:23 +01:00
|
|
|
|
2017-03-16 18:07:14 -07:00
|
|
|
|
2015-08-07 21:19:26 +01:00
|
|
|
|
2020-07-15 21:20:51 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-08-07 21:19:26 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2019-11-24 22:58:35 +08:00
|
|
|
|
2012-11-14 09:39:31 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2013-10-15 20:14:20 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2013-10-15 20:14:20 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-11-27 10:57:16 +08:00
|
|
|
|
2013-10-15 20:14:20 +01:00
|
|
|
|
2026-02-21 16:37:42 -08:00
|
|
|
|
2025-11-27 10:57:16 +08:00
|
|
|
|
2013-10-15 20:14:20 +01:00
|
|
|
|
|
|
|
|
|
2025-11-27 10:57:16 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2013-10-15 20:14:20 +01:00
|
|
|
|
2025-11-27 10:57:16 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2013-10-15 20:14:20 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-11-27 10:57:16 +08:00
|
|
|
|
2013-10-15 20:14:20 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-11-27 10:57:16 +08:00
|
|
|
|
2013-10-15 20:14:20 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:23 +08:00
|
|
|
|
|
|
|
|
|
2013-10-15 20:14:20 +01:00
|
|
|
|
2014-05-23 16:44:10 +01:00
|
|
|
|
|
|
|
|
|
2013-10-15 20:14:20 +01:00
|
|
|
|
2014-05-23 16:44:10 +01:00
|
|
|
|
2013-10-15 20:14:20 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-05-23 16:44:10 +01:00
|
|
|
|
2013-10-15 20:14:20 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2013-02-18 06:50:39 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-08-12 03:31:34 +02:00
|
|
|
|
2014-06-29 17:55:54 +01:00
|
|
|
|
2013-02-18 06:50:39 +00:00
|
|
|
|
2019-01-29 11:31:56 +01:00
|
|
|
|
2026-02-21 16:37:42 -08:00
|
|
|
|
2020-08-12 03:31:34 +02:00
|
|
|
|
|
|
|
|
|
2014-06-29 17:55:54 +01:00
|
|
|
|
2025-11-12 14:55:39 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2013-02-18 06:50:39 +00:00
|
|
|
|
2025-04-07 09:08:15 +02:00
|
|
|
|
2019-01-29 11:31:56 +01:00
|
|
|
|
2013-02-18 06:50:39 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-08-12 03:31:34 +02:00
|
|
|
|
|
|
|
|
|
2013-02-18 06:50:39 +00:00
|
|
|
|
2020-08-12 03:31:34 +02:00
|
|
|
|
|
|
|
|
|
2025-11-12 14:55:39 +01:00
|
|
|
|
2020-08-12 03:31:34 +02:00
|
|
|
|
|
|
|
|
|
2013-02-18 06:50:39 +00:00
|
|
|
|
2014-06-29 17:55:54 +01:00
|
|
|
|
2013-02-18 06:50:39 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-08-12 03:31:34 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2013-02-18 06:50:39 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-08-12 03:31:37 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2013-02-18 06:50:39 +00:00
|
|
|
|
2020-08-12 03:31:37 +02:00
|
|
|
|
|
|
|
|
|
2013-02-18 06:50:39 +00:00
|
|
|
|
|
|
|
|
|
2013-02-18 06:50:48 +00:00
|
|
|
|
2013-05-02 10:19:11 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2013-02-18 06:50:48 +00:00
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2013-02-18 06:50:48 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-12-03 09:47:37 +01:00
|
|
|
|
2013-02-18 06:50:48 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-12-03 09:47:37 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2013-02-18 06:50:48 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-12-03 09:47:37 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2013-02-18 06:50:48 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-06-22 18:11:07 +05:30
|
|
|
|
2024-08-29 16:51:21 +08:00
|
|
|
|
|
|
|
|
|
2020-06-22 18:11:07 +05:30
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-06-27 13:21:26 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
2012-06-27 13:21:26 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core: Shorten off-on-delay-us for always-on/boot-on by time since booted
This is very close to a straight revert of commit 218320fec294
("regulator: core: Fix off-on-delay-us for always-on/boot-on
regulators"). We've identified that patch as causing a boot speed
regression on sc7180-trogdor boards. While boot speed certainly isn't
more important than making sure that power sequencing is correct,
looking closely at the original change it doesn't seem to have been
fully justified. It mentions "cycling issues" without describing
exactly what the issues were. That means it's possible that the
cycling issues were really a problem that should be fixed in a
different way.
Let's take a careful look at how we should handle regulators that have
an off-on-delay and that are boot-on or always-on. Linux currently
doesn't have any way to identify whether a GPIO regulator was already
on when the kernel booted. That means that when the kernel boots we
probe a regulator, see that it wants boot-on / always-on we, and then
turn the regulator on. We could be in one of two cases when we do
this:
a) The regulator might have been left on by the bootloader and we're
ensuring that it stays on.
b) The regulator might have been left off by the bootloader and we're
just now turning it on.
For case a) we definitely don't need any sort of delay. For case b) we
_might_ need some delay in case the bootloader turned the regulator
off _right_ before booting the kernel. To get the proper delay for
case b) then we can just assume a `last_off` of 0, which is what it
gets initialized to by default.
As per above, we can't tell whether we're in case a) or case b) so
we'll assume the longer delay (case b). This basically puts the code
to how it was before commit 218320fec294 ("regulator: core: Fix
off-on-delay-us for always-on/boot-on regulators"). However, we add
one important change: we make sure that the delay is actually honored
if `last_off` is 0. Though the original "cycling issues" cited were
vague, I'm hopeful that this important extra change will be enough to
fix the issues that the initial commit mentioned.
With this fix, I've confined that on a sc7180-trogdor board the delay
at boot goes down from 500 ms to ~250 ms. That's not as good as the 0
ms that we had prior to commit 218320fec294 ("regulator: core: Fix
off-on-delay-us for always-on/boot-on regulators"), but it's probably
safer because we don't know if the bootloader turned the regulator off
right before booting.
One note is that it's possible that we could be in a state that's not
a) or b) if there are other issues in the kernel. The only one I can
think of is related to pinctrl. If the pinctrl driver being used on a
board isn't careful about avoiding glitches when setting up a pin then
it's possible that setting up a pin could cause the regulator to "turn
off" briefly immediately before the regulator probes. If this is
indeed causing problems then the pinctrl driver should be fixed,
perhaps in a similar way to what was done in commit d21f4b7ffc22
("pinctrl: qcom: Avoid glitching lines when we first mux to output")
Fixes: 218320fec294 ("regulator: core: Fix off-on-delay-us for always-on/boot-on regulators")
Cc: Christian Kohlschütter <christian@kohlschutter.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20230313111806.1.I2eaad872be0932a805c239a7c7a102233fb0b03b@changeid
Signed-off-by: Mark Brown <broonie@kernel.org>
2023-03-13 11:18:19 -07:00
|
|
|
|
2014-08-13 19:33:40 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-04-23 13:45:24 +02:00
|
|
|
|
2023-02-23 00:33:30 +00:00
|
|
|
|
2014-08-13 19:33:40 +08:00
|
|
|
|
2021-04-23 13:45:24 +02:00
|
|
|
|
2024-10-14 10:22:28 +02:00
|
|
|
|
2014-08-13 19:33:40 +08:00
|
|
|
|
|
|
|
|
|
2013-02-18 06:50:48 +00:00
|
|
|
|
2015-03-03 15:20:47 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-06-27 14:14:38 +01:00
|
|
|
|
2012-06-27 13:21:26 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-04-20 09:17:18 +05:30
|
|
|
|
|
|
|
|
|
2012-06-27 13:21:26 +01:00
|
|
|
|
|
|
|
|
|
2020-06-22 18:11:07 +05:30
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-20 14:22:12 -07:00
|
|
|
|
|
|
|
|
|
2020-06-22 18:11:07 +05:30
|
|
|
|
|
|
|
|
|
2022-09-09 14:59:53 +02:00
|
|
|
|
2020-06-22 18:11:07 +05:30
|
|
|
|
|
|
|
|
|
2024-10-14 10:22:28 +02:00
|
|
|
|
2020-06-22 18:11:07 +05:30
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-10-14 10:22:28 +02:00
|
|
|
|
2020-06-22 18:11:07 +05:30
|
|
|
|
2012-06-27 13:21:26 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:23 +08:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2022-08-19 14:43:36 -05:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-08-19 14:43:36 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:23 +08:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-06-27 13:21:26 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
2018-11-20 09:52:54 -08:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-06-29 09:04:43 +09:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2009-08-03 18:49:57 +01:00
|
|
|
|
2021-04-21 05:52:37 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-08-03 18:49:57 +01:00
|
|
|
|
|
|
|
|
|
2016-04-23 15:11:05 +08:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2012-06-27 13:21:26 +01:00
|
|
|
|
2009-12-21 12:21:52 +00:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
2009-12-21 12:21:52 +00:00
|
|
|
|
2017-02-23 17:06:52 +00:00
|
|
|
|
|
|
|
|
|
2009-08-26 12:54:04 +02:00
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2009-08-26 12:54:04 +02:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2023-11-03 15:42:31 +08:00
|
|
|
|
|
|
|
|
|
2009-08-03 18:49:57 +01:00
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2018-11-20 09:52:54 -08:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-12-31 12:52:43 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2008-12-31 12:52:43 +00:00
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-11-16 11:44:46 -08:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
2012-04-19 13:19:07 +01:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2011-06-09 16:22:22 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-06-27 13:21:26 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2013-02-18 06:50:48 +00:00
|
|
|
|
2015-03-03 15:20:47 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-06-27 13:21:26 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-08-13 19:33:40 +08:00
|
|
|
|
2023-02-23 00:33:30 +00:00
|
|
|
|
2014-08-13 19:33:40 +08:00
|
|
|
|
2012-06-27 13:21:26 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2015-06-29 09:04:43 +09:00
|
|
|
|
2023-11-03 15:42:31 +08:00
|
|
|
|
2010-12-06 14:05:19 -08:00
|
|
|
|
2009-03-11 16:43:34 -08:00
|
|
|
|
|
|
|
|
|
2023-11-03 15:42:31 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2023-11-03 15:42:31 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-11-24 14:10:52 +00:00
|
|
|
|
2023-11-03 15:42:31 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-11-24 14:10:52 +00:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2023-11-03 15:42:31 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2011-06-09 16:22:22 +01:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-20 09:52:54 -08:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-12-31 12:52:43 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-12-31 12:52:41 +00:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2008-12-31 12:52:43 +00:00
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-11-16 11:44:46 -08:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
2012-04-19 13:19:07 +01:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2010-10-08 14:49:19 -05:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-06-09 16:22:22 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2015-06-29 09:04:43 +09:00
|
|
|
|
2014-11-24 14:10:52 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-02-20 17:36:04 +01:00
|
|
|
|
|
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
2014-11-24 14:10:52 +00:00
|
|
|
|
|
|
|
|
|
2014-02-20 17:36:04 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2014-02-20 17:36:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-05-09 11:41:02 +02:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
2011-06-09 16:22:22 +01:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
2018-10-05 18:36:29 +03:00
|
|
|
|
|
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-20 09:52:54 -08:00
|
|
|
|
|
|
|
|
|
2010-10-08 14:49:19 -05:00
|
|
|
|
2018-11-20 09:52:54 -08:00
|
|
|
|
2010-10-08 14:49:19 -05:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-09-11 09:53:50 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2011-09-11 09:53:50 +01:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
|
|
|
|
|
2011-09-11 09:53:50 +01:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2011-09-11 09:53:50 +01:00
|
|
|
|
2017-07-12 17:08:13 +05:30
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
|
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
2011-09-11 09:53:50 +01:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
2011-09-11 09:53:50 +01:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-09-11 09:53:50 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-01-08 13:12:33 +01:00
|
|
|
|
2011-09-11 09:53:50 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2011-09-11 09:53:50 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-09-07 11:00:53 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
2017-07-12 17:08:13 +05:30
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2011-09-11 09:53:50 +01:00
|
|
|
|
2016-01-07 13:03:08 +03:00
|
|
|
|
2011-09-11 09:53:50 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2012-06-27 14:14:38 +01:00
|
|
|
|
2013-02-18 06:50:55 +00:00
|
|
|
|
2012-06-27 14:14:38 +01:00
|
|
|
|
|
|
|
|
|
2010-02-11 17:22:45 +00:00
|
|
|
|
2009-08-03 18:49:56 +01:00
|
|
|
|
2010-02-11 17:22:45 +00:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2009-08-03 18:49:56 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2018-01-22 15:30:07 +01:00
|
|
|
|
|
|
|
|
|
2015-09-30 16:05:43 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-11-25 19:25:47 +02:00
|
|
|
|
|
|
|
|
|
2015-09-30 16:05:43 +02:00
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2015-09-30 16:05:43 +02:00
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2017-03-27 16:54:12 -07:00
|
|
|
|
2018-01-22 15:30:07 +01:00
|
|
|
|
|
|
|
|
|
2015-09-30 16:05:43 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-11-16 11:44:46 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:23 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-08-03 18:49:56 +01:00
|
|
|
|
|
|
|
|
|
2012-04-19 13:19:07 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2009-08-03 18:49:56 +01:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2009-08-03 18:49:56 +01:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-02-26 11:48:36 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:23 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-02-26 11:48:36 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-07-29 18:28:56 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-03-27 16:54:12 -07:00
|
|
|
|
2014-07-29 18:28:56 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-02-26 11:48:36 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:23 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-02-26 11:48:36 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-22 15:30:07 +01:00
|
|
|
|
2009-02-26 11:48:36 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-07-21 18:38:48 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:23 +08:00
|
|
|
|
|
|
|
|
|
2014-07-21 18:38:48 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-06-10 20:55:32 +01:00
|
|
|
|
2014-07-21 18:38:48 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-07-21 18:38:48 +03:00
|
|
|
|
2024-08-29 16:51:30 +08:00
|
|
|
|
2014-07-21 18:38:48 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-08-19 18:07:41 +08:00
|
|
|
|
|
|
|
|
|
2014-07-21 18:38:48 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-12-07 13:30:09 +00:00
|
|
|
|
|
|
|
|
|
2014-07-21 18:38:48 +03:00
|
|
|
|
2020-09-20 15:24:54 +01:00
|
|
|
|
2014-07-21 18:38:48 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:23 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-07-21 18:38:48 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-08-19 18:07:41 +08:00
|
|
|
|
|
|
|
|
|
2014-07-21 18:38:48 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-11-25 19:25:47 +02:00
|
|
|
|
|
|
|
|
|
2014-07-21 18:38:48 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-06-16 11:53:55 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:23 +08:00
|
|
|
|
2024-06-16 11:53:55 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2013-06-07 08:06:56 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:23 +08:00
|
|
|
|
|
|
|
|
|
2013-06-07 08:06:56 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-07-21 16:00:24 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:22 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-07-21 16:00:24 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-07-02 15:00:18 +01:00
|
|
|
|
2009-07-21 16:00:24 +01:00
|
|
|
|
|
|
|
|
|
2012-07-02 15:00:18 +01:00
|
|
|
|
2016-04-23 15:11:05 +08:00
|
|
|
|
2012-07-02 15:00:18 +01:00
|
|
|
|
|
|
|
|
|
2014-01-08 10:02:16 +09:00
|
|
|
|
2012-07-02 15:00:18 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-09-24 18:56:53 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-07-21 16:00:24 +01:00
|
|
|
|
|
|
|
|
|
2019-04-22 19:28:24 +02:00
|
|
|
|
2009-07-21 16:00:24 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-12-28 12:48:45 +00:00
|
|
|
|
2009-07-21 16:00:24 +01:00
|
|
|
|
2015-10-20 14:37:27 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-09-14 11:31:36 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-10-20 14:37:27 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-08-28 12:36:04 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:32 +03:00
|
|
|
|
2014-08-28 12:36:04 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:32 +03:00
|
|
|
|
2014-08-28 12:36:04 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-07-03 18:10:34 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-09-14 09:52:08 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-04-04 18:59:50 +05:30
|
|
|
|
|
|
|
|
|
2017-05-16 11:43:43 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-09-14 09:52:08 -07:00
|
|
|
|
2022-08-20 13:14:21 +00:00
|
|
|
|
2016-09-14 09:52:08 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2010-12-12 14:25:50 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-03-17 13:24:36 +01:00
|
|
|
|
2012-06-26 10:57:51 +01:00
|
|
|
|
2010-12-12 14:25:50 +00:00
|
|
|
|
2012-04-04 10:32:10 +08:00
|
|
|
|
2016-09-14 09:52:05 -07:00
|
|
|
|
2019-06-24 00:08:32 +03:00
|
|
|
|
2010-12-12 14:25:50 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-05-08 22:13:37 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-04-04 10:32:10 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-05-21 09:37:52 +08:00
|
|
|
|
2016-09-14 09:52:05 -07:00
|
|
|
|
|
|
|
|
|
2012-04-04 10:32:10 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-09-14 09:52:05 -07:00
|
|
|
|
2014-08-28 12:36:04 -07:00
|
|
|
|
|
|
|
|
|
2012-06-26 10:57:51 +01:00
|
|
|
|
|
|
|
|
|
2016-09-14 09:52:05 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-06-26 10:57:51 +01:00
|
|
|
|
2019-06-24 00:08:32 +03:00
|
|
|
|
2012-06-26 10:57:51 +01:00
|
|
|
|
|
|
|
|
|
2016-09-14 09:52:05 -07:00
|
|
|
|
2015-10-20 14:37:27 +02:00
|
|
|
|
2012-05-09 21:16:06 +01:00
|
|
|
|
2016-09-14 09:52:05 -07:00
|
|
|
|
2012-06-26 10:57:51 +01:00
|
|
|
|
|
|
|
|
|
2013-02-06 11:09:48 +08:00
|
|
|
|
|
|
|
|
|
2019-07-03 18:10:34 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2013-02-06 11:09:48 +08:00
|
|
|
|
2014-08-28 12:36:04 -07:00
|
|
|
|
|
|
|
|
|
2012-06-26 10:57:51 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2010-12-12 14:36:17 +00:00
|
|
|
|
2010-12-12 14:25:50 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-09-14 09:52:06 -07:00
|
|
|
|
|
|
|
|
|
2012-04-04 10:32:10 +08:00
|
|
|
|
2016-09-14 09:52:08 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-06-14 15:07:56 -07:00
|
|
|
|
2011-03-17 13:24:36 +01:00
|
|
|
|
|
|
|
|
|
2016-09-14 09:52:08 -07:00
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
2016-09-14 09:52:08 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-10-14 10:22:28 +02:00
|
|
|
|
2011-03-17 13:24:36 +01:00
|
|
|
|
2016-09-14 09:52:06 -07:00
|
|
|
|
2012-08-06 23:44:19 +08:00
|
|
|
|
|
|
|
|
|
2010-12-16 13:59:10 +00:00
|
|
|
|
2012-08-06 23:44:19 +08:00
|
|
|
|
|
|
|
|
|
2010-12-16 13:59:10 +00:00
|
|
|
|
2016-09-14 09:52:06 -07:00
|
|
|
|
2012-04-04 10:32:10 +08:00
|
|
|
|
2010-12-12 14:25:50 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-26 21:08:47 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-07-18 16:11:36 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-10-13 12:45:25 +02:00
|
|
|
|
2018-01-26 21:08:44 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2018-01-26 21:08:44 +08:00
|
|
|
|
2010-12-16 15:49:37 +00:00
|
|
|
|
2025-07-18 16:11:36 +02:00
|
|
|
|
2012-12-13 10:13:00 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2010-12-16 15:49:37 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-26 21:08:44 +08:00
|
|
|
|
2010-12-16 15:49:37 +00:00
|
|
|
|
|
|
|
|
|
2014-02-05 12:30:26 -08:00
|
|
|
|
2015-08-13 18:09:49 +05:30
|
|
|
|
2014-02-05 12:30:26 -08:00
|
|
|
|
|
|
|
|
|
2016-04-23 15:11:05 +08:00
|
|
|
|
2019-06-24 00:08:32 +03:00
|
|
|
|
2014-02-05 12:30:26 -08:00
|
|
|
|
2018-01-26 21:08:44 +08:00
|
|
|
|
|
|
|
|
|
2014-02-05 12:30:26 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2010-12-12 14:36:17 +00:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-01-08 10:02:16 +09:00
|
|
|
|
2012-12-13 10:13:00 +01:00
|
|
|
|
2018-01-26 21:08:44 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2010-11-10 14:38:29 +00:00
|
|
|
|
2018-10-05 18:36:29 +03:00
|
|
|
|
|
|
|
|
|
2019-03-18 16:17:57 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2010-12-03 11:31:07 +01:00
|
|
|
|
2025-07-18 16:11:36 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-07-29 11:50:57 +02:00
|
|
|
|
2025-07-18 16:11:36 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-10-05 18:36:29 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:32 +03:00
|
|
|
|
|
|
|
|
|
2018-10-05 18:36:29 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-04-13 18:36:59 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-05-25 15:33:17 +05:30
|
|
|
|
|
|
|
|
|
2015-10-20 14:37:28 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-10-05 18:36:29 +03:00
|
|
|
|
2015-10-20 14:37:28 +02:00
|
|
|
|
|
|
|
|
|
2018-01-26 17:55:30 +00:00
|
|
|
|
2015-10-20 14:37:28 +02:00
|
|
|
|
|
|
|
|
|
2018-10-05 18:36:29 +03:00
|
|
|
|
2015-10-20 14:37:28 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:32 +03:00
|
|
|
|
2015-10-20 14:37:28 +02:00
|
|
|
|
|
|
|
|
|
2018-10-05 18:36:29 +03:00
|
|
|
|
2015-10-20 14:37:28 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-26 21:08:44 +08:00
|
|
|
|
2015-10-20 14:37:28 +02:00
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
|
|
|
|
|
2018-10-05 18:36:29 +03:00
|
|
|
|
2015-10-20 14:37:28 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-26 21:08:47 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-12-13 10:13:00 +01:00
|
|
|
|
2018-10-05 18:36:29 +03:00
|
|
|
|
2014-01-08 10:02:16 +09:00
|
|
|
|
2015-10-20 14:37:28 +02:00
|
|
|
|
|
|
|
|
|
2018-01-26 21:08:44 +08:00
|
|
|
|
2015-10-20 14:37:28 +02:00
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
|
|
|
|
|
2015-10-20 14:37:28 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2015-10-13 12:45:25 +02:00
|
|
|
|
regulator: core: Add voltage balancing mechanism
On Odroid XU3/4 and other Exynos5422 based boards there is a case, that
different devices on the board are supplied by different regulators
with non-fixed voltages. If one of these devices temporarily requires
higher voltage, there might occur a situation that the spread between
two devices' voltages is so high, that there is a risk of changing
'high' and 'low' states on the interconnection between devices powered
by those regulators.
Introduce new function regulator_balance_voltage(), which
keeps max_spread constraint fulfilled between a group of coupled
regulators. It should be called if a regulator changes its
voltage or after disabling or enabling. Disabled regulators should
follow changes of the enabled ones, but their consumers' demands
shouldn't be taken into account while calculating voltage of other
coupled regulators.
Find voltages, which are closest to suiting all the consumers' demands,
while fulfilling max_spread constraint, keeping the following rules:
- if one regulator is about to rise its voltage, rise others
voltages in order to keep the max_spread
- if a regulator, which has caused rising other regulators, is
lowered, lower other regulators if possible
- if one regulator is about to lower its voltage, but it hasn't caused
rising other regulators, don't change its voltage if it breaks the
max_spread
Change regulators' voltages step by step, keeping max_spread constraint
fulfilled all the time. Function regulator_get_optimal_voltage()
should find the best possible change for the regulator, which doesn't
break max_spread constraint. In function regulator_balance_voltage()
optimize number of steps by finding highest voltage difference on
each iteration.
If a regulator, which is about to change its voltage, is not coupled,
method regulator_get_optimal_voltage() should simply return the lowest
voltage fulfilling consumers' demands.
Coupling should be checked only if the system is in PM_SUSPEND_ON state.
Signed-off-by: Maciej Purski <m.purski@samsung.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2018-04-23 16:33:41 +02:00
|
|
|
|
2020-01-20 13:39:21 +01:00
|
|
|
|
2015-10-13 12:45:25 +02:00
|
|
|
|
2018-10-05 18:36:35 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:32 +03:00
|
|
|
|
2018-10-05 18:36:35 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core: Add voltage balancing mechanism
On Odroid XU3/4 and other Exynos5422 based boards there is a case, that
different devices on the board are supplied by different regulators
with non-fixed voltages. If one of these devices temporarily requires
higher voltage, there might occur a situation that the spread between
two devices' voltages is so high, that there is a risk of changing
'high' and 'low' states on the interconnection between devices powered
by those regulators.
Introduce new function regulator_balance_voltage(), which
keeps max_spread constraint fulfilled between a group of coupled
regulators. It should be called if a regulator changes its
voltage or after disabling or enabling. Disabled regulators should
follow changes of the enabled ones, but their consumers' demands
shouldn't be taken into account while calculating voltage of other
coupled regulators.
Find voltages, which are closest to suiting all the consumers' demands,
while fulfilling max_spread constraint, keeping the following rules:
- if one regulator is about to rise its voltage, rise others
voltages in order to keep the max_spread
- if a regulator, which has caused rising other regulators, is
lowered, lower other regulators if possible
- if one regulator is about to lower its voltage, but it hasn't caused
rising other regulators, change its voltage so that it doesn't break the
max_spread
Change regulators' voltages step by step, keeping max_spread constraint
fulfilled all the time. Function regulator_get_optimal_voltage()
should find the best possible change for the regulator, which doesn't
break max_spread constraint. In function regulator_balance_voltage()
optimize number of steps by finding highest voltage difference on
each iteration.
If a regulator, which is about to change its voltage, is not coupled,
method regulator_get_optimal_voltage() should simply return the lowest
voltage fulfilling consumers' demands.
Coupling should be checked only if the system is in PM_SUSPEND_ON state.
Signed-off-by: Maciej Purski <m.purski@samsung.com>
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2018-10-05 18:36:28 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
regulator: core: Add voltage balancing mechanism
On Odroid XU3/4 and other Exynos5422 based boards there is a case, that
different devices on the board are supplied by different regulators
with non-fixed voltages. If one of these devices temporarily requires
higher voltage, there might occur a situation that the spread between
two devices' voltages is so high, that there is a risk of changing
'high' and 'low' states on the interconnection between devices powered
by those regulators.
Introduce new function regulator_balance_voltage(), which
keeps max_spread constraint fulfilled between a group of coupled
regulators. It should be called if a regulator changes its
voltage or after disabling or enabling. Disabled regulators should
follow changes of the enabled ones, but their consumers' demands
shouldn't be taken into account while calculating voltage of other
coupled regulators.
Find voltages, which are closest to suiting all the consumers' demands,
while fulfilling max_spread constraint, keeping the following rules:
- if one regulator is about to rise its voltage, rise others
voltages in order to keep the max_spread
- if a regulator, which has caused rising other regulators, is
lowered, lower other regulators if possible
- if one regulator is about to lower its voltage, but it hasn't caused
rising other regulators, change its voltage so that it doesn't break the
max_spread
Change regulators' voltages step by step, keeping max_spread constraint
fulfilled all the time. Function regulator_get_optimal_voltage()
should find the best possible change for the regulator, which doesn't
break max_spread constraint. In function regulator_balance_voltage()
optimize number of steps by finding highest voltage difference on
each iteration.
If a regulator, which is about to change its voltage, is not coupled,
method regulator_get_optimal_voltage() should simply return the lowest
voltage fulfilling consumers' demands.
Coupling should be checked only if the system is in PM_SUSPEND_ON state.
Signed-off-by: Maciej Purski <m.purski@samsung.com>
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2018-10-05 18:36:28 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
regulator: core: Add voltage balancing mechanism
On Odroid XU3/4 and other Exynos5422 based boards there is a case, that
different devices on the board are supplied by different regulators
with non-fixed voltages. If one of these devices temporarily requires
higher voltage, there might occur a situation that the spread between
two devices' voltages is so high, that there is a risk of changing
'high' and 'low' states on the interconnection between devices powered
by those regulators.
Introduce new function regulator_balance_voltage(), which
keeps max_spread constraint fulfilled between a group of coupled
regulators. It should be called if a regulator changes its
voltage or after disabling or enabling. Disabled regulators should
follow changes of the enabled ones, but their consumers' demands
shouldn't be taken into account while calculating voltage of other
coupled regulators.
Find voltages, which are closest to suiting all the consumers' demands,
while fulfilling max_spread constraint, keeping the following rules:
- if one regulator is about to rise its voltage, rise others
voltages in order to keep the max_spread
- if a regulator, which has caused rising other regulators, is
lowered, lower other regulators if possible
- if one regulator is about to lower its voltage, but it hasn't caused
rising other regulators, change its voltage so that it doesn't break the
max_spread
Change regulators' voltages step by step, keeping max_spread constraint
fulfilled all the time. Function regulator_get_optimal_voltage()
should find the best possible change for the regulator, which doesn't
break max_spread constraint. In function regulator_balance_voltage()
optimize number of steps by finding highest voltage difference on
each iteration.
If a regulator, which is about to change its voltage, is not coupled,
method regulator_get_optimal_voltage() should simply return the lowest
voltage fulfilling consumers' demands.
Coupling should be checked only if the system is in PM_SUSPEND_ON state.
Signed-off-by: Maciej Purski <m.purski@samsung.com>
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2018-10-05 18:36:28 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
|
|
|
|
|
regulator: core: Add voltage balancing mechanism
On Odroid XU3/4 and other Exynos5422 based boards there is a case, that
different devices on the board are supplied by different regulators
with non-fixed voltages. If one of these devices temporarily requires
higher voltage, there might occur a situation that the spread between
two devices' voltages is so high, that there is a risk of changing
'high' and 'low' states on the interconnection between devices powered
by those regulators.
Introduce new function regulator_balance_voltage(), which
keeps max_spread constraint fulfilled between a group of coupled
regulators. It should be called if a regulator changes its
voltage or after disabling or enabling. Disabled regulators should
follow changes of the enabled ones, but their consumers' demands
shouldn't be taken into account while calculating voltage of other
coupled regulators.
Find voltages, which are closest to suiting all the consumers' demands,
while fulfilling max_spread constraint, keeping the following rules:
- if one regulator is about to rise its voltage, rise others
voltages in order to keep the max_spread
- if a regulator, which has caused rising other regulators, is
lowered, lower other regulators if possible
- if one regulator is about to lower its voltage, but it hasn't caused
rising other regulators, change its voltage so that it doesn't break the
max_spread
Change regulators' voltages step by step, keeping max_spread constraint
fulfilled all the time. Function regulator_get_optimal_voltage()
should find the best possible change for the regulator, which doesn't
break max_spread constraint. In function regulator_balance_voltage()
optimize number of steps by finding highest voltage difference on
each iteration.
If a regulator, which is about to change its voltage, is not coupled,
method regulator_get_optimal_voltage() should simply return the lowest
voltage fulfilling consumers' demands.
Coupling should be checked only if the system is in PM_SUSPEND_ON state.
Signed-off-by: Maciej Purski <m.purski@samsung.com>
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2018-10-05 18:36:28 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:32 +03:00
|
|
|
|
regulator: core: Add voltage balancing mechanism
On Odroid XU3/4 and other Exynos5422 based boards there is a case, that
different devices on the board are supplied by different regulators
with non-fixed voltages. If one of these devices temporarily requires
higher voltage, there might occur a situation that the spread between
two devices' voltages is so high, that there is a risk of changing
'high' and 'low' states on the interconnection between devices powered
by those regulators.
Introduce new function regulator_balance_voltage(), which
keeps max_spread constraint fulfilled between a group of coupled
regulators. It should be called if a regulator changes its
voltage or after disabling or enabling. Disabled regulators should
follow changes of the enabled ones, but their consumers' demands
shouldn't be taken into account while calculating voltage of other
coupled regulators.
Find voltages, which are closest to suiting all the consumers' demands,
while fulfilling max_spread constraint, keeping the following rules:
- if one regulator is about to rise its voltage, rise others
voltages in order to keep the max_spread
- if a regulator, which has caused rising other regulators, is
lowered, lower other regulators if possible
- if one regulator is about to lower its voltage, but it hasn't caused
rising other regulators, change its voltage so that it doesn't break the
max_spread
Change regulators' voltages step by step, keeping max_spread constraint
fulfilled all the time. Function regulator_get_optimal_voltage()
should find the best possible change for the regulator, which doesn't
break max_spread constraint. In function regulator_balance_voltage()
optimize number of steps by finding highest voltage difference on
each iteration.
If a regulator, which is about to change its voltage, is not coupled,
method regulator_get_optimal_voltage() should simply return the lowest
voltage fulfilling consumers' demands.
Coupling should be checked only if the system is in PM_SUSPEND_ON state.
Signed-off-by: Maciej Purski <m.purski@samsung.com>
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2018-10-05 18:36:28 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-10-05 18:36:35 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core: Add voltage balancing mechanism
On Odroid XU3/4 and other Exynos5422 based boards there is a case, that
different devices on the board are supplied by different regulators
with non-fixed voltages. If one of these devices temporarily requires
higher voltage, there might occur a situation that the spread between
two devices' voltages is so high, that there is a risk of changing
'high' and 'low' states on the interconnection between devices powered
by those regulators.
Introduce new function regulator_balance_voltage(), which
keeps max_spread constraint fulfilled between a group of coupled
regulators. It should be called if a regulator changes its
voltage or after disabling or enabling. Disabled regulators should
follow changes of the enabled ones, but their consumers' demands
shouldn't be taken into account while calculating voltage of other
coupled regulators.
Find voltages, which are closest to suiting all the consumers' demands,
while fulfilling max_spread constraint, keeping the following rules:
- if one regulator is about to rise its voltage, rise others
voltages in order to keep the max_spread
- if a regulator, which has caused rising other regulators, is
lowered, lower other regulators if possible
- if one regulator is about to lower its voltage, but it hasn't caused
rising other regulators, change its voltage so that it doesn't break the
max_spread
Change regulators' voltages step by step, keeping max_spread constraint
fulfilled all the time. Function regulator_get_optimal_voltage()
should find the best possible change for the regulator, which doesn't
break max_spread constraint. In function regulator_balance_voltage()
optimize number of steps by finding highest voltage difference on
each iteration.
If a regulator, which is about to change its voltage, is not coupled,
method regulator_get_optimal_voltage() should simply return the lowest
voltage fulfilling consumers' demands.
Coupling should be checked only if the system is in PM_SUSPEND_ON state.
Signed-off-by: Maciej Purski <m.purski@samsung.com>
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2018-10-05 18:36:28 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:32 +03:00
|
|
|
|
regulator: core: Add voltage balancing mechanism
On Odroid XU3/4 and other Exynos5422 based boards there is a case, that
different devices on the board are supplied by different regulators
with non-fixed voltages. If one of these devices temporarily requires
higher voltage, there might occur a situation that the spread between
two devices' voltages is so high, that there is a risk of changing
'high' and 'low' states on the interconnection between devices powered
by those regulators.
Introduce new function regulator_balance_voltage(), which
keeps max_spread constraint fulfilled between a group of coupled
regulators. It should be called if a regulator changes its
voltage or after disabling or enabling. Disabled regulators should
follow changes of the enabled ones, but their consumers' demands
shouldn't be taken into account while calculating voltage of other
coupled regulators.
Find voltages, which are closest to suiting all the consumers' demands,
while fulfilling max_spread constraint, keeping the following rules:
- if one regulator is about to rise its voltage, rise others
voltages in order to keep the max_spread
- if a regulator, which has caused rising other regulators, is
lowered, lower other regulators if possible
- if one regulator is about to lower its voltage, but it hasn't caused
rising other regulators, change its voltage so that it doesn't break the
max_spread
Change regulators' voltages step by step, keeping max_spread constraint
fulfilled all the time. Function regulator_get_optimal_voltage()
should find the best possible change for the regulator, which doesn't
break max_spread constraint. In function regulator_balance_voltage()
optimize number of steps by finding highest voltage difference on
each iteration.
If a regulator, which is about to change its voltage, is not coupled,
method regulator_get_optimal_voltage() should simply return the lowest
voltage fulfilling consumers' demands.
Coupling should be checked only if the system is in PM_SUSPEND_ON state.
Signed-off-by: Maciej Purski <m.purski@samsung.com>
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2018-10-05 18:36:28 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-05-29 14:49:39 +02:00
|
|
|
|
|
|
|
|
|
regulator: core: Add voltage balancing mechanism
On Odroid XU3/4 and other Exynos5422 based boards there is a case, that
different devices on the board are supplied by different regulators
with non-fixed voltages. If one of these devices temporarily requires
higher voltage, there might occur a situation that the spread between
two devices' voltages is so high, that there is a risk of changing
'high' and 'low' states on the interconnection between devices powered
by those regulators.
Introduce new function regulator_balance_voltage(), which
keeps max_spread constraint fulfilled between a group of coupled
regulators. It should be called if a regulator changes its
voltage or after disabling or enabling. Disabled regulators should
follow changes of the enabled ones, but their consumers' demands
shouldn't be taken into account while calculating voltage of other
coupled regulators.
Find voltages, which are closest to suiting all the consumers' demands,
while fulfilling max_spread constraint, keeping the following rules:
- if one regulator is about to rise its voltage, rise others
voltages in order to keep the max_spread
- if a regulator, which has caused rising other regulators, is
lowered, lower other regulators if possible
- if one regulator is about to lower its voltage, but it hasn't caused
rising other regulators, change its voltage so that it doesn't break the
max_spread
Change regulators' voltages step by step, keeping max_spread constraint
fulfilled all the time. Function regulator_get_optimal_voltage()
should find the best possible change for the regulator, which doesn't
break max_spread constraint. In function regulator_balance_voltage()
optimize number of steps by finding highest voltage difference on
each iteration.
If a regulator, which is about to change its voltage, is not coupled,
method regulator_get_optimal_voltage() should simply return the lowest
voltage fulfilling consumers' demands.
Coupling should be checked only if the system is in PM_SUSPEND_ON state.
Signed-off-by: Maciej Purski <m.purski@samsung.com>
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2018-10-05 18:36:28 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
|
|
|
|
|
regulator: core: Add voltage balancing mechanism
On Odroid XU3/4 and other Exynos5422 based boards there is a case, that
different devices on the board are supplied by different regulators
with non-fixed voltages. If one of these devices temporarily requires
higher voltage, there might occur a situation that the spread between
two devices' voltages is so high, that there is a risk of changing
'high' and 'low' states on the interconnection between devices powered
by those regulators.
Introduce new function regulator_balance_voltage(), which
keeps max_spread constraint fulfilled between a group of coupled
regulators. It should be called if a regulator changes its
voltage or after disabling or enabling. Disabled regulators should
follow changes of the enabled ones, but their consumers' demands
shouldn't be taken into account while calculating voltage of other
coupled regulators.
Find voltages, which are closest to suiting all the consumers' demands,
while fulfilling max_spread constraint, keeping the following rules:
- if one regulator is about to rise its voltage, rise others
voltages in order to keep the max_spread
- if a regulator, which has caused rising other regulators, is
lowered, lower other regulators if possible
- if one regulator is about to lower its voltage, but it hasn't caused
rising other regulators, change its voltage so that it doesn't break the
max_spread
Change regulators' voltages step by step, keeping max_spread constraint
fulfilled all the time. Function regulator_get_optimal_voltage()
should find the best possible change for the regulator, which doesn't
break max_spread constraint. In function regulator_balance_voltage()
optimize number of steps by finding highest voltage difference on
each iteration.
If a regulator, which is about to change its voltage, is not coupled,
method regulator_get_optimal_voltage() should simply return the lowest
voltage fulfilling consumers' demands.
Coupling should be checked only if the system is in PM_SUSPEND_ON state.
Signed-off-by: Maciej Purski <m.purski@samsung.com>
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2018-10-05 18:36:28 +03:00
|
|
|
|
|
|
|
|
|
2020-05-29 14:49:39 +02:00
|
|
|
|
regulator: core: Add voltage balancing mechanism
On Odroid XU3/4 and other Exynos5422 based boards there is a case, that
different devices on the board are supplied by different regulators
with non-fixed voltages. If one of these devices temporarily requires
higher voltage, there might occur a situation that the spread between
two devices' voltages is so high, that there is a risk of changing
'high' and 'low' states on the interconnection between devices powered
by those regulators.
Introduce new function regulator_balance_voltage(), which
keeps max_spread constraint fulfilled between a group of coupled
regulators. It should be called if a regulator changes its
voltage or after disabling or enabling. Disabled regulators should
follow changes of the enabled ones, but their consumers' demands
shouldn't be taken into account while calculating voltage of other
coupled regulators.
Find voltages, which are closest to suiting all the consumers' demands,
while fulfilling max_spread constraint, keeping the following rules:
- if one regulator is about to rise its voltage, rise others
voltages in order to keep the max_spread
- if a regulator, which has caused rising other regulators, is
lowered, lower other regulators if possible
- if one regulator is about to lower its voltage, but it hasn't caused
rising other regulators, change its voltage so that it doesn't break the
max_spread
Change regulators' voltages step by step, keeping max_spread constraint
fulfilled all the time. Function regulator_get_optimal_voltage()
should find the best possible change for the regulator, which doesn't
break max_spread constraint. In function regulator_balance_voltage()
optimize number of steps by finding highest voltage difference on
each iteration.
If a regulator, which is about to change its voltage, is not coupled,
method regulator_get_optimal_voltage() should simply return the lowest
voltage fulfilling consumers' demands.
Coupling should be checked only if the system is in PM_SUSPEND_ON state.
Signed-off-by: Maciej Purski <m.purski@samsung.com>
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2018-10-05 18:36:28 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
regulator: core: Add voltage balancing mechanism
On Odroid XU3/4 and other Exynos5422 based boards there is a case, that
different devices on the board are supplied by different regulators
with non-fixed voltages. If one of these devices temporarily requires
higher voltage, there might occur a situation that the spread between
two devices' voltages is so high, that there is a risk of changing
'high' and 'low' states on the interconnection between devices powered
by those regulators.
Introduce new function regulator_balance_voltage(), which
keeps max_spread constraint fulfilled between a group of coupled
regulators. It should be called if a regulator changes its
voltage or after disabling or enabling. Disabled regulators should
follow changes of the enabled ones, but their consumers' demands
shouldn't be taken into account while calculating voltage of other
coupled regulators.
Find voltages, which are closest to suiting all the consumers' demands,
while fulfilling max_spread constraint, keeping the following rules:
- if one regulator is about to rise its voltage, rise others
voltages in order to keep the max_spread
- if a regulator, which has caused rising other regulators, is
lowered, lower other regulators if possible
- if one regulator is about to lower its voltage, but it hasn't caused
rising other regulators, change its voltage so that it doesn't break the
max_spread
Change regulators' voltages step by step, keeping max_spread constraint
fulfilled all the time. Function regulator_get_optimal_voltage()
should find the best possible change for the regulator, which doesn't
break max_spread constraint. In function regulator_balance_voltage()
optimize number of steps by finding highest voltage difference on
each iteration.
If a regulator, which is about to change its voltage, is not coupled,
method regulator_get_optimal_voltage() should simply return the lowest
voltage fulfilling consumers' demands.
Coupling should be checked only if the system is in PM_SUSPEND_ON state.
Signed-off-by: Maciej Purski <m.purski@samsung.com>
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2018-10-05 18:36:28 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-10-05 18:36:29 +03:00
|
|
|
|
regulator: core: Add voltage balancing mechanism
On Odroid XU3/4 and other Exynos5422 based boards there is a case, that
different devices on the board are supplied by different regulators
with non-fixed voltages. If one of these devices temporarily requires
higher voltage, there might occur a situation that the spread between
two devices' voltages is so high, that there is a risk of changing
'high' and 'low' states on the interconnection between devices powered
by those regulators.
Introduce new function regulator_balance_voltage(), which
keeps max_spread constraint fulfilled between a group of coupled
regulators. It should be called if a regulator changes its
voltage or after disabling or enabling. Disabled regulators should
follow changes of the enabled ones, but their consumers' demands
shouldn't be taken into account while calculating voltage of other
coupled regulators.
Find voltages, which are closest to suiting all the consumers' demands,
while fulfilling max_spread constraint, keeping the following rules:
- if one regulator is about to rise its voltage, rise others
voltages in order to keep the max_spread
- if a regulator, which has caused rising other regulators, is
lowered, lower other regulators if possible
- if one regulator is about to lower its voltage, but it hasn't caused
rising other regulators, change its voltage so that it doesn't break the
max_spread
Change regulators' voltages step by step, keeping max_spread constraint
fulfilled all the time. Function regulator_get_optimal_voltage()
should find the best possible change for the regulator, which doesn't
break max_spread constraint. In function regulator_balance_voltage()
optimize number of steps by finding highest voltage difference on
each iteration.
If a regulator, which is about to change its voltage, is not coupled,
method regulator_get_optimal_voltage() should simply return the lowest
voltage fulfilling consumers' demands.
Coupling should be checked only if the system is in PM_SUSPEND_ON state.
Signed-off-by: Maciej Purski <m.purski@samsung.com>
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2018-10-05 18:36:28 +03:00
|
|
|
|
|
|
|
|
|
2018-10-05 18:36:29 +03:00
|
|
|
|
regulator: core: Add voltage balancing mechanism
On Odroid XU3/4 and other Exynos5422 based boards there is a case, that
different devices on the board are supplied by different regulators
with non-fixed voltages. If one of these devices temporarily requires
higher voltage, there might occur a situation that the spread between
two devices' voltages is so high, that there is a risk of changing
'high' and 'low' states on the interconnection between devices powered
by those regulators.
Introduce new function regulator_balance_voltage(), which
keeps max_spread constraint fulfilled between a group of coupled
regulators. It should be called if a regulator changes its
voltage or after disabling or enabling. Disabled regulators should
follow changes of the enabled ones, but their consumers' demands
shouldn't be taken into account while calculating voltage of other
coupled regulators.
Find voltages, which are closest to suiting all the consumers' demands,
while fulfilling max_spread constraint, keeping the following rules:
- if one regulator is about to rise its voltage, rise others
voltages in order to keep the max_spread
- if a regulator, which has caused rising other regulators, is
lowered, lower other regulators if possible
- if one regulator is about to lower its voltage, but it hasn't caused
rising other regulators, change its voltage so that it doesn't break the
max_spread
Change regulators' voltages step by step, keeping max_spread constraint
fulfilled all the time. Function regulator_get_optimal_voltage()
should find the best possible change for the regulator, which doesn't
break max_spread constraint. In function regulator_balance_voltage()
optimize number of steps by finding highest voltage difference on
each iteration.
If a regulator, which is about to change its voltage, is not coupled,
method regulator_get_optimal_voltage() should simply return the lowest
voltage fulfilling consumers' demands.
Coupling should be checked only if the system is in PM_SUSPEND_ON state.
Signed-off-by: Maciej Purski <m.purski@samsung.com>
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2018-10-05 18:36:28 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
|
|
|
|
|
regulator: core: Add voltage balancing mechanism
On Odroid XU3/4 and other Exynos5422 based boards there is a case, that
different devices on the board are supplied by different regulators
with non-fixed voltages. If one of these devices temporarily requires
higher voltage, there might occur a situation that the spread between
two devices' voltages is so high, that there is a risk of changing
'high' and 'low' states on the interconnection between devices powered
by those regulators.
Introduce new function regulator_balance_voltage(), which
keeps max_spread constraint fulfilled between a group of coupled
regulators. It should be called if a regulator changes its
voltage or after disabling or enabling. Disabled regulators should
follow changes of the enabled ones, but their consumers' demands
shouldn't be taken into account while calculating voltage of other
coupled regulators.
Find voltages, which are closest to suiting all the consumers' demands,
while fulfilling max_spread constraint, keeping the following rules:
- if one regulator is about to rise its voltage, rise others
voltages in order to keep the max_spread
- if a regulator, which has caused rising other regulators, is
lowered, lower other regulators if possible
- if one regulator is about to lower its voltage, but it hasn't caused
rising other regulators, change its voltage so that it doesn't break the
max_spread
Change regulators' voltages step by step, keeping max_spread constraint
fulfilled all the time. Function regulator_get_optimal_voltage()
should find the best possible change for the regulator, which doesn't
break max_spread constraint. In function regulator_balance_voltage()
optimize number of steps by finding highest voltage difference on
each iteration.
If a regulator, which is about to change its voltage, is not coupled,
method regulator_get_optimal_voltage() should simply return the lowest
voltage fulfilling consumers' demands.
Coupling should be checked only if the system is in PM_SUSPEND_ON state.
Signed-off-by: Maciej Purski <m.purski@samsung.com>
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2018-10-05 18:36:28 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core: Add voltage balancing mechanism
On Odroid XU3/4 and other Exynos5422 based boards there is a case, that
different devices on the board are supplied by different regulators
with non-fixed voltages. If one of these devices temporarily requires
higher voltage, there might occur a situation that the spread between
two devices' voltages is so high, that there is a risk of changing
'high' and 'low' states on the interconnection between devices powered
by those regulators.
Introduce new function regulator_balance_voltage(), which
keeps max_spread constraint fulfilled between a group of coupled
regulators. It should be called if a regulator changes its
voltage or after disabling or enabling. Disabled regulators should
follow changes of the enabled ones, but their consumers' demands
shouldn't be taken into account while calculating voltage of other
coupled regulators.
Find voltages, which are closest to suiting all the consumers' demands,
while fulfilling max_spread constraint, keeping the following rules:
- if one regulator is about to rise its voltage, rise others
voltages in order to keep the max_spread
- if a regulator, which has caused rising other regulators, is
lowered, lower other regulators if possible
- if one regulator is about to lower its voltage, but it hasn't caused
rising other regulators, don't change its voltage if it breaks the
max_spread
Change regulators' voltages step by step, keeping max_spread constraint
fulfilled all the time. Function regulator_get_optimal_voltage()
should find the best possible change for the regulator, which doesn't
break max_spread constraint. In function regulator_balance_voltage()
optimize number of steps by finding highest voltage difference on
each iteration.
If a regulator, which is about to change its voltage, is not coupled,
method regulator_get_optimal_voltage() should simply return the lowest
voltage fulfilling consumers' demands.
Coupling should be checked only if the system is in PM_SUSPEND_ON state.
Signed-off-by: Maciej Purski <m.purski@samsung.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2018-04-23 16:33:41 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-05-29 14:49:39 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-10-13 12:45:25 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2015-10-13 12:45:25 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
2015-10-13 12:45:25 +02:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2015-10-13 12:45:25 +02:00
|
|
|
|
2018-01-26 21:08:44 +08:00
|
|
|
|
|
|
|
|
|
2015-10-13 12:45:25 +02:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2015-10-13 12:45:25 +02:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-26 21:08:47 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-09-03 16:49:37 +02:00
|
|
|
|
2018-01-26 21:08:47 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
2018-01-26 21:08:47 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2018-01-26 21:08:47 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2018-01-26 21:08:47 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-03-17 13:24:52 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2011-03-17 13:24:52 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-08-13 19:33:38 +08:00
|
|
|
|
|
|
|
|
|
2011-03-17 13:24:52 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-09-14 09:52:08 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-03-17 13:24:52 +01:00
|
|
|
|
2016-09-14 09:52:08 -07:00
|
|
|
|
2011-03-17 13:24:52 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-11-13 17:21:05 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-11-13 19:56:04 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-03-17 13:24:52 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-06-09 16:40:38 +05:30
|
|
|
|
2012-08-18 17:44:14 -07:00
|
|
|
|
|
|
|
|
|
2012-06-09 16:40:38 +05:30
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-06-19 09:38:45 +08:00
|
|
|
|
2012-06-18 10:11:28 +08:00
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2012-06-09 16:40:38 +05:30
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-06-19 09:38:45 +08:00
|
|
|
|
2012-06-18 10:11:28 +08:00
|
|
|
|
2012-06-19 09:38:45 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-06-18 10:11:28 +08:00
|
|
|
|
2012-06-19 09:38:45 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-09-14 09:52:08 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-06-09 16:40:38 +05:30
|
|
|
|
2012-06-23 11:34:20 +01:00
|
|
|
|
2012-06-09 16:40:38 +05:30
|
|
|
|
2021-05-28 02:54:00 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2010-12-16 15:49:36 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2010-12-16 15:49:36 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-26 21:08:44 +08:00
|
|
|
|
2010-12-16 15:49:36 +00:00
|
|
|
|
|
|
|
|
|
2021-10-21 21:33:08 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2010-12-16 15:49:36 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-26 21:08:44 +08:00
|
|
|
|
2010-12-16 15:49:36 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-26 21:08:44 +08:00
|
|
|
|
|
|
|
|
|
2010-12-16 15:49:36 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-26 21:08:44 +08:00
|
|
|
|
2010-12-16 15:49:36 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-01-22 20:43:11 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2010-12-16 15:49:36 +00:00
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2010-12-16 15:49:36 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:32 +03:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2011-05-08 22:13:37 +01:00
|
|
|
|
2016-04-07 16:22:36 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core: Add early supply resolution for regulators
The call to set_machine_constraints() in regulator_register(), will
attempt to get the voltage for the regulator. If a regulator is in
bypass will fail to get the voltage (ie. it's bypass voltage) and
hence register the regulator, because the supply for the regulator has
not been resolved yet.
To fix this, add a call to regulator_resolve_supply() before we call
set_machine_constraints(). If the call to regulator_resolve_supply()
fails, rather than returning an error at this point, allow the
registration of the regulator to continue because for some regulators
resolving the supply at this point may not be necessary and it will be
resolved later as more regulators are added. Furthermore, if the supply
is still not resolved for a bypassed regulator, this will be detected
when we attempt to get the voltage for the regulator and an error will
be propagated at this point.
If a bypassed regulator does not have a supply when we attempt to get
the voltage, rather than returing -EINVAL, return -EPROBE_DEFER instead
to allow the registration of the regulator to be deferred and tried
again later.
Please note that regulator_resolve_supply() will call
regulator_dev_lookup() which may acquire the regulator_list_mutex. To
avoid any deadlocks we cannot hold the regulator_list_mutex when calling
regulator_resolve_supply(). Therefore, rather than holding the lock
around a large portion of the registration code, just hold the lock when
aquiring any GPIOs and setting up supplies because these sections may
add entries to the regulator_map_list and regulator_ena_gpio_list,
respectively.
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2016-04-26 11:29:45 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-04-07 16:22:36 +02:00
|
|
|
|
2019-06-24 00:08:32 +03:00
|
|
|
|
2016-04-07 16:22:36 +02:00
|
|
|
|
|
|
|
|
|
2010-12-10 17:28:07 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-05-08 22:13:37 +01:00
|
|
|
|
2011-05-23 20:08:10 +08:00
|
|
|
|
2011-05-08 22:13:37 +01:00
|
|
|
|
2012-08-09 16:42:19 +01:00
|
|
|
|
|
|
|
|
|
2013-09-10 15:45:05 +05:30
|
|
|
|
|
|
|
|
|
2014-07-29 18:28:55 +02:00
|
|
|
|
2019-06-24 00:08:32 +03:00
|
|
|
|
2020-11-02 22:27:27 +01:00
|
|
|
|
|
|
|
|
|
2011-05-23 20:08:10 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2011-05-23 20:08:10 +08:00
|
|
|
|
2011-05-08 22:13:37 +01:00
|
|
|
|
2011-05-23 20:08:10 +08:00
|
|
|
|
|
|
|
|
|
2011-05-08 22:13:37 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2020-01-20 13:39:21 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:23 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2019-06-24 00:08:32 +03:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2013-05-21 15:04:07 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-08-16 13:28:03 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2018-08-16 13:28:03 -07:00
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2018-08-16 13:28:03 -07:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:23 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-01-15 15:41:57 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2010-05-17 21:24:48 +05:30
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2010-05-17 21:24:48 +05:30
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2011-04-01 18:25:25 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-08-16 13:28:03 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2018-08-16 13:28:03 -07:00
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2018-08-16 13:28:03 -07:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-03 08:41:55 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-11-03 12:11:42 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-03 08:41:55 +03:00
|
|
|
|
2016-11-03 12:11:42 +01:00
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2016-11-03 12:11:42 +01:00
|
|
|
|
2021-06-03 08:41:55 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-11-03 12:11:42 +01:00
|
|
|
|
2021-06-03 08:41:55 +03:00
|
|
|
|
|
|
|
|
|
2016-11-03 12:11:42 +01:00
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2021-06-03 08:41:55 +03:00
|
|
|
|
2016-11-03 12:11:42 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2016-11-03 12:11:42 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2015-02-11 19:35:27 -08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:23 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2015-02-11 19:35:27 -08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
|
|
|
|
|
2012-07-02 19:21:06 -07:00
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2015-01-27 18:46:31 -08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2015-02-11 19:35:27 -08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2012-08-31 10:36:37 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2013-02-28 18:44:54 -06:00
|
|
|
|
2012-08-31 10:36:37 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-08-31 10:36:37 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-05-29 16:22:16 +01:00
|
|
|
|
2012-08-31 10:36:37 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-04-23 15:11:05 +08:00
|
|
|
|
2012-08-31 10:36:37 -07:00
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2012-08-31 10:36:37 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-05-29 16:22:16 +01:00
|
|
|
|
|
|
|
|
|
2012-08-31 10:36:37 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-05-29 16:22:16 +01:00
|
|
|
|
|
|
|
|
|
2012-08-31 10:36:37 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-05-29 16:22:16 +01:00
|
|
|
|
|
|
|
|
|
2012-08-31 10:36:37 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-05-29 16:22:16 +01:00
|
|
|
|
|
|
|
|
|
2012-08-31 10:36:37 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2012-08-31 10:36:37 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-12-31 12:52:41 +00:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-12-31 12:52:41 +00:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-01-19 18:20:58 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-08-28 12:36:04 -07:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-12-05 16:22:04 +05:30
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2022-10-31 16:37:02 -07:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-10-31 16:37:02 -07:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2022-07-26 10:38:21 -07:00
|
|
|
|
2025-01-22 08:20:19 +01:00
|
|
|
|
2022-07-26 10:38:21 -07:00
|
|
|
|
2022-08-09 14:27:45 -07:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2022-07-26 10:38:21 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-02-20 10:32:16 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-10-31 16:37:02 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:23 +08:00
|
|
|
|
|
|
|
|
|
2022-10-31 16:37:02 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2011-05-24 08:12:40 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:23 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-07-09 19:33:25 -07:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2011-05-24 08:12:40 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2012-04-19 13:19:07 +01:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
|
|
|
|
|
2012-04-19 13:19:07 +01:00
|
|
|
|
2011-05-24 08:12:40 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2011-05-24 08:12:40 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2011-05-24 08:12:40 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2013-03-01 12:24:05 +01:00
|
|
|
|
|
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
|
|
|
|
|
2013-03-01 12:24:05 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-01-25 12:35:38 +01:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2024-08-29 16:51:23 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-03-28 21:36:38 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2012-01-25 12:35:38 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
2012-03-28 21:36:38 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
|
|
|
|
|
2012-03-28 21:36:38 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-01-03 16:22:03 +09:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:23 +08:00
|
|
|
|
|
|
|
|
|
2012-01-03 16:22:03 +09:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-02-03 15:16:17 -08:00
|
|
|
|
2012-01-03 16:22:03 +09:00
|
|
|
|
2017-02-03 15:16:17 -08:00
|
|
|
|
2012-01-03 16:22:03 +09:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-02-03 15:16:17 -08:00
|
|
|
|
|
|
|
|
|
2012-01-03 16:22:03 +09:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-10-26 16:48:21 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-02-17 21:39:48 +01:00
|
|
|
|
|
|
|
|
|
2023-10-26 16:48:21 +02:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2008-12-31 12:52:41 +00:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2008-12-31 12:52:41 +00:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2020-08-10 06:33:32 +02:00
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-10-26 16:48:21 +02:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-08-04 20:09:52 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2009-08-04 20:09:52 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-07-12 11:50:38 +01:00
|
|
|
|
2009-08-04 20:09:52 +02:00
|
|
|
|
|
|
|
|
|
2012-07-12 13:53:35 +01:00
|
|
|
|
2009-08-04 20:09:52 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-01-30 20:29:31 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-05-03 23:52:49 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-01-30 20:29:31 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-01-15 15:41:57 +01:00
|
|
|
|
|
|
|
|
|
2015-01-30 20:29:31 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-11-11 17:39:02 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-01-30 20:29:31 +01:00
|
|
|
|
|
|
|
|
|
2008-11-11 17:39:02 -08:00
|
|
|
|
2015-01-30 20:29:31 +01:00
|
|
|
|
2016-01-05 22:07:55 +08:00
|
|
|
|
2014-08-13 19:33:38 +08:00
|
|
|
|
2015-01-30 20:29:31 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-11-11 17:39:02 -08:00
|
|
|
|
|
|
|
|
|
2015-01-30 20:29:31 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-08-31 10:36:37 -07:00
|
|
|
|
2008-11-11 17:39:02 -08:00
|
|
|
|
2015-01-30 20:29:31 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-05-03 23:52:49 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-11-11 17:39:02 -08:00
|
|
|
|
2015-01-30 20:29:31 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-11-11 17:39:02 -08:00
|
|
|
|
2015-01-30 20:29:31 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-11-11 17:39:02 -08:00
|
|
|
|
2015-01-30 20:29:31 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-11-11 17:39:02 -08:00
|
|
|
|
2015-01-30 20:29:31 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-11-11 17:39:02 -08:00
|
|
|
|
2015-01-30 20:29:31 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-01-15 15:41:57 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-01-30 20:29:31 +01:00
|
|
|
|
2008-11-11 17:39:02 -08:00
|
|
|
|
|
|
|
|
|
2015-01-30 20:29:31 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-08-07 20:01:32 +01:00
|
|
|
|
2022-11-16 11:37:06 +08:00
|
|
|
|
2015-08-07 20:01:32 +01:00
|
|
|
|
|
|
|
|
|
2015-01-30 20:29:31 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2010-12-21 23:49:31 +00:00
|
|
|
|
|
|
|
|
|
2014-10-17 08:17:03 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-25 13:13:59 +02:00
|
|
|
|
|
|
|
|
|
2010-12-21 23:49:31 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2012-08-31 10:36:37 -07:00
|
|
|
|
|
|
|
|
|
2010-12-21 23:49:31 +00:00
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-10-05 18:36:30 +03:00
|
|
|
|
2018-04-23 16:33:40 +02:00
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
2018-04-23 16:33:40 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-10-05 18:36:30 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-10-05 18:36:30 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:40 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-10-05 18:36:36 +03:00
|
|
|
|
2018-04-23 16:33:40 +02:00
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
2018-10-05 18:36:36 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
2018-04-23 16:33:40 +02:00
|
|
|
|
2018-10-05 18:36:36 +03:00
|
|
|
|
2018-04-23 16:33:40 +02:00
|
|
|
|
2018-10-05 18:36:36 +03:00
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:40 +02:00
|
|
|
|
2018-10-05 18:36:36 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
|
|
|
|
|
2025-06-26 08:38:09 +00:00
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:40 +02:00
|
|
|
|
|
|
|
|
|
2018-10-05 18:36:30 +03:00
|
|
|
|
2018-04-23 16:33:40 +02:00
|
|
|
|
2020-08-09 21:21:16 +02:00
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
2018-04-23 16:33:40 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026-02-21 16:37:42 -08:00
|
|
|
|
2020-08-09 21:21:16 +02:00
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
2018-04-23 16:33:40 +02:00
|
|
|
|
2020-08-09 21:21:16 +02:00
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:40 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-08-12 03:31:34 +02:00
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
2020-08-12 03:31:34 +02:00
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
|
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:40 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-10-25 03:22:40 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2022-12-06 15:22:21 +08:00
|
|
|
|
2008-12-31 12:52:41 +00:00
|
|
|
|
2015-01-12 09:01:39 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:23 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2012-04-03 20:46:53 +01:00
|
|
|
|
2022-12-06 15:22:21 +08:00
|
|
|
|
|
|
|
|
|
2015-01-05 12:48:41 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2012-04-04 00:50:22 +01:00
|
|
|
|
2015-01-05 12:48:41 +01:00
|
|
|
|
2014-12-28 22:08:38 +05:30
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
regulator: core: reresolve unresolved supplies when available
When a regulator A and its supply B are provided by different devices,
the driver implementing B might be last to probe (with A still
pending resolution of its supply B). While we try to resolve all
pending supplies for all regulators (including A) during
regulator_register() of B via regulator_register_resolve_supply(),
supply resolution will still not work for A as the driver for B hasn't
finished binding to the PMIC device corresponding to B at that stage
yet. The regulator core explicitly only allows supplies from other
devices to be used once the relevant driver has fully bound, mainly to
avoid having to deal with cases where B itself might -EPROBE_DEFER.
In this case, A's supply will only be resolved as part of the core's
regulator_init_complete_work_function(), which currently is scheduled
to run after 30s. This was added as a work-around in
commit 3827b64dba27 ("regulator: core: Resolve supplies before
disabling unused regulators") to cover this situation.
There are two problems with that approach:
* it potentially runs long after all our consumers have probed
* an upcoming change will allow regulator_register() to complete
successfully even when required supplies (e.g. due to always-on or
boot-on) are missing at register time, deferring full configuration
of the regulator (and usability by consumers, i.e. usually consumer
probe) until the supply becomes available.
Resolving supplies in the late work func can therefore make it
impossible for consumers to probe at all, as the driver core will not
know to reprobe consumers when supplies have resolved.
We could schedule an earlier work to try to resolve supplies sooner,
but that'd be racy as consumers of A might try to probe before A's
supply gets fully resolved via this extra work.
Instead, add a very simple regulator bus and add a dummy device with a
corresponding driver to it for each regulator that is missing its
supply during regulator_register(). This way, the driver core will call
our bus' probe whenever a new (regulator) device was successfully
bound, allowing us to retry resolving the supply during (our bus) probe
and to bind this dummy device if successful. In turn this means the
driver core will see a newly bound device and retry probing of all
pending consumers, if any.
With that in place, we can avoid walking the full list of all known
regulators to try resolve missing supplies during regulator_register(),
as the driver core will invoke the bus probe for regulators that are
still pending their supplies. We can also drop the code trying to
resolve supplies one last time before unused regulators get disabled,
as all supplies should have resolved at that point in time, and if they
haven't then there's no point in trying again, as the outcome won't
change.
Note: We can not reuse the existing struct device created for each
rail, as a device can not be part of a class and a bus simultaneously.
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-7-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:43 +00:00
|
|
|
|
2018-12-06 13:43:37 +01:00
|
|
|
|
|
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2018-12-06 13:43:37 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2018-12-06 13:43:37 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2022-12-06 15:22:21 +08:00
|
|
|
|
2012-04-11 10:19:12 +01:00
|
|
|
|
2018-12-06 13:43:37 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2009-04-14 03:04:47 +02:00
|
|
|
|
2018-12-06 13:43:37 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2010-12-10 17:28:07 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2010-12-12 14:36:17 +00:00
|
|
|
|
|
|
|
|
|
2010-12-10 17:28:07 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-12-06 13:43:37 +01:00
|
|
|
|
|
|
|
|
|
2010-12-10 17:28:07 +00:00
|
|
|
|
2010-12-12 14:36:17 +00:00
|
|
|
|
|
|
|
|
|
2018-12-06 13:43:37 +01:00
|
|
|
|
|
|
|
|
|
2010-12-12 14:36:17 +00:00
|
|
|
|
2010-12-10 17:28:07 +00:00
|
|
|
|
2026-02-21 16:37:42 -08:00
|
|
|
|
2018-12-06 13:43:37 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-08-12 03:31:36 +02:00
|
|
|
|
2023-09-19 00:50:26 +02:00
|
|
|
|
|
|
|
|
|
2021-06-03 08:41:55 +03:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2015-01-05 12:48:41 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-12-06 13:43:37 +01:00
|
|
|
|
2020-08-12 03:31:36 +02:00
|
|
|
|
2015-01-05 12:48:41 +01:00
|
|
|
|
|
|
|
|
|
2025-02-11 16:48:06 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-09-17 17:40:21 +02:00
|
|
|
|
2025-02-11 16:48:06 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-09-09 23:13:57 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
2012-04-04 00:50:22 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2012-05-08 18:10:45 +01:00
|
|
|
|
|
|
|
|
|
2012-09-07 20:45:05 +05:30
|
|
|
|
2012-05-08 18:10:45 +01:00
|
|
|
|
2012-09-07 20:45:05 +05:30
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-09-11 09:53:50 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2022-08-29 16:55:43 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-02-22 09:24:00 +01:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2022-12-14 09:37:11 +08:00
|
|
|
|
2014-12-28 22:08:38 +05:30
|
|
|
|
2014-12-29 22:36:48 +05:30
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2008-11-25 14:53:53 +02:00
|
|
|
|
2011-11-29 18:11:19 +00:00
|
|
|
|
2020-11-13 01:20:27 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026-02-21 16:37:42 -08:00
|
|
|
|
2020-11-13 01:20:27 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-11-29 18:11:19 +00:00
|
|
|
|
2024-10-08 18:07:02 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-01-29 11:31:56 +01:00
|
|
|
|
2016-02-22 09:24:00 +01:00
|
|
|
|
|
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
|
|
|
|
|
2022-08-29 16:55:43 +00:00
|
|
|
|
2016-02-22 09:24:00 +01:00
|
|
|
|
2018-12-06 13:43:37 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-02-22 09:24:00 +01:00
|
|
|
|
2011-11-18 16:47:20 +05:30
|
|
|
|
regulator: core: don't fail regulator_register() with missing required supply
Since commit 98e48cd9283d ("regulator: core: resolve supply for
boot-on/always-on regulators"), the regulator core returns
-EPROBE_DEFER if a supply can not be resolved at regulator_register()
time due to set_machine_constraints() requiring that supply (e.g.
because of always-on or boot-on).
In some hardware designs, multiple PMICs are used where individual
rails of each act as supplies for rails of the other, and vice-versa.
In such a design no PMIC driver can probe when registering one top-
level regulator device (as is common practice for almost all regulator
drivers in Linux) since that commit. Supplies are only considered when
their driver has fully bound, but because in a design like the above
two drivers / devices depend on each other, neither will have fully
bound while the other probes. The Google Pixel 6 and 6 Pro (oriole and
raven) are examples of such a design.
One way to make this work would be to register each rail as an
individual device, rather than just one top-level regulator device.
Then, fw-devlink and Linux' driver core could do their usual handling
of deferred device probe as each rail would be probed individually.
This approach was dismissed in [1] as each regulator driver would have
to take care of this itself.
Alternatively, we can change the regulator core to not fail
regulator_register() if a rail's required supply can not be resolved
while keeping the intended change from above mentioned commit, and
instead retry whenever a new rail is registered. This commit implements
such an approach:
If set_machine_constraints() requests probe deferral,
regulator_register() still succeeds and we retry setting
constraints as part of regulator_resolve_supply().
We still do not enable the regulator or allow consumers to use it
until constraints have been set (including resolution of the
supply) to prevent enabling of a regulator before its supply.
With this change, we keep track of regulators with missing required
supplies and can therefore try to resolve them again and try to set
the constraints again once more regulators become available.
Care has to be taken to not allow consumers to use regulators that
haven't had their constraints set yet. regulator_get() ensures that
and now returns -EPROBE_DEFER in that case.
The implementation is straight-forward, thanks to our newly introduced
regulator-bus. Locking in regulator_resolve_supply() has to be done
carefully, as a combination of regulator_(un)lock_two() and
regulator_(un)lock_dependent() is needed. The reason is that
set_machine_constraints() might call regulator_enable() which needs
rdev and all its dependents locked, but everything else requires to
only have rdev and its supply locked.
Link: https://lore.kernel.org/all/aRn_-o-vie_QoDXD@sirena.co.uk/ [1]
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-8-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:44 +00:00
|
|
|
|
2026-01-09 08:38:40 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-04-20 09:17:18 +05:30
|
|
|
|
2020-09-26 23:32:41 +02:00
|
|
|
|
|
|
|
|
|
2021-04-20 09:17:18 +05:30
|
|
|
|
|
|
|
|
|
2020-11-13 01:20:27 +01:00
|
|
|
|
|
|
|
|
|
2020-09-26 23:32:41 +02:00
|
|
|
|
|
|
|
|
|
regulator: core: don't fail regulator_register() with missing required supply
Since commit 98e48cd9283d ("regulator: core: resolve supply for
boot-on/always-on regulators"), the regulator core returns
-EPROBE_DEFER if a supply can not be resolved at regulator_register()
time due to set_machine_constraints() requiring that supply (e.g.
because of always-on or boot-on).
In some hardware designs, multiple PMICs are used where individual
rails of each act as supplies for rails of the other, and vice-versa.
In such a design no PMIC driver can probe when registering one top-
level regulator device (as is common practice for almost all regulator
drivers in Linux) since that commit. Supplies are only considered when
their driver has fully bound, but because in a design like the above
two drivers / devices depend on each other, neither will have fully
bound while the other probes. The Google Pixel 6 and 6 Pro (oriole and
raven) are examples of such a design.
One way to make this work would be to register each rail as an
individual device, rather than just one top-level regulator device.
Then, fw-devlink and Linux' driver core could do their usual handling
of deferred device probe as each rail would be probed individually.
This approach was dismissed in [1] as each regulator driver would have
to take care of this itself.
Alternatively, we can change the regulator core to not fail
regulator_register() if a rail's required supply can not be resolved
while keeping the intended change from above mentioned commit, and
instead retry whenever a new rail is registered. This commit implements
such an approach:
If set_machine_constraints() requests probe deferral,
regulator_register() still succeeds and we retry setting
constraints as part of regulator_resolve_supply().
We still do not enable the regulator or allow consumers to use it
until constraints have been set (including resolution of the
supply) to prevent enabling of a regulator before its supply.
With this change, we keep track of regulators with missing required
supplies and can therefore try to resolve them again and try to set
the constraints again once more regulators become available.
Care has to be taken to not allow consumers to use regulators that
haven't had their constraints set yet. regulator_get() ensures that
and now returns -EPROBE_DEFER in that case.
The implementation is straight-forward, thanks to our newly introduced
regulator-bus. Locking in regulator_resolve_supply() has to be done
carefully, as a combination of regulator_(un)lock_two() and
regulator_(un)lock_dependent() is needed. The reason is that
set_machine_constraints() might call regulator_enable() which needs
rdev and all its dependents locked, but everything else requires to
only have rdev and its supply locked.
Link: https://lore.kernel.org/all/aRn_-o-vie_QoDXD@sirena.co.uk/ [1]
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-8-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:44 +00:00
|
|
|
|
2020-09-26 23:32:41 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core: reresolve unresolved supplies when available
When a regulator A and its supply B are provided by different devices,
the driver implementing B might be last to probe (with A still
pending resolution of its supply B). While we try to resolve all
pending supplies for all regulators (including A) during
regulator_register() of B via regulator_register_resolve_supply(),
supply resolution will still not work for A as the driver for B hasn't
finished binding to the PMIC device corresponding to B at that stage
yet. The regulator core explicitly only allows supplies from other
devices to be used once the relevant driver has fully bound, mainly to
avoid having to deal with cases where B itself might -EPROBE_DEFER.
In this case, A's supply will only be resolved as part of the core's
regulator_init_complete_work_function(), which currently is scheduled
to run after 30s. This was added as a work-around in
commit 3827b64dba27 ("regulator: core: Resolve supplies before
disabling unused regulators") to cover this situation.
There are two problems with that approach:
* it potentially runs long after all our consumers have probed
* an upcoming change will allow regulator_register() to complete
successfully even when required supplies (e.g. due to always-on or
boot-on) are missing at register time, deferring full configuration
of the regulator (and usability by consumers, i.e. usually consumer
probe) until the supply becomes available.
Resolving supplies in the late work func can therefore make it
impossible for consumers to probe at all, as the driver core will not
know to reprobe consumers when supplies have resolved.
We could schedule an earlier work to try to resolve supplies sooner,
but that'd be racy as consumers of A might try to probe before A's
supply gets fully resolved via this extra work.
Instead, add a very simple regulator bus and add a dummy device with a
corresponding driver to it for each regulator that is missing its
supply during regulator_register(). This way, the driver core will call
our bus' probe whenever a new (regulator) device was successfully
bound, allowing us to retry resolving the supply during (our bus) probe
and to bind this dummy device if successful. In turn this means the
driver core will see a newly bound device and retry probing of all
pending consumers, if any.
With that in place, we can avoid walking the full list of all known
regulators to try resolve missing supplies during regulator_register(),
as the driver core will invoke the bus probe for regulators that are
still pending their supplies. We can also drop the code trying to
resolve supplies one last time before unused regulators get disabled,
as all supplies should have resolved at that point in time, and if they
haven't then there's no point in trying again, as the outcome won't
change.
Note: We can not reuse the existing struct device created for each
rail, as a device can not be part of a class and a bus simultaneously.
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-7-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:43 +00:00
|
|
|
|
2020-09-26 23:32:41 +02:00
|
|
|
|
regulator: core: don't fail regulator_register() with missing required supply
Since commit 98e48cd9283d ("regulator: core: resolve supply for
boot-on/always-on regulators"), the regulator core returns
-EPROBE_DEFER if a supply can not be resolved at regulator_register()
time due to set_machine_constraints() requiring that supply (e.g.
because of always-on or boot-on).
In some hardware designs, multiple PMICs are used where individual
rails of each act as supplies for rails of the other, and vice-versa.
In such a design no PMIC driver can probe when registering one top-
level regulator device (as is common practice for almost all regulator
drivers in Linux) since that commit. Supplies are only considered when
their driver has fully bound, but because in a design like the above
two drivers / devices depend on each other, neither will have fully
bound while the other probes. The Google Pixel 6 and 6 Pro (oriole and
raven) are examples of such a design.
One way to make this work would be to register each rail as an
individual device, rather than just one top-level regulator device.
Then, fw-devlink and Linux' driver core could do their usual handling
of deferred device probe as each rail would be probed individually.
This approach was dismissed in [1] as each regulator driver would have
to take care of this itself.
Alternatively, we can change the regulator core to not fail
regulator_register() if a rail's required supply can not be resolved
while keeping the intended change from above mentioned commit, and
instead retry whenever a new rail is registered. This commit implements
such an approach:
If set_machine_constraints() requests probe deferral,
regulator_register() still succeeds and we retry setting
constraints as part of regulator_resolve_supply().
We still do not enable the regulator or allow consumers to use it
until constraints have been set (including resolution of the
supply) to prevent enabling of a regulator before its supply.
With this change, we keep track of regulators with missing required
supplies and can therefore try to resolve them again and try to set
the constraints again once more regulators become available.
Care has to be taken to not allow consumers to use regulators that
haven't had their constraints set yet. regulator_get() ensures that
and now returns -EPROBE_DEFER in that case.
The implementation is straight-forward, thanks to our newly introduced
regulator-bus. Locking in regulator_resolve_supply() has to be done
carefully, as a combination of regulator_(un)lock_two() and
regulator_(un)lock_dependent() is needed. The reason is that
set_machine_constraints() might call regulator_enable() which needs
rdev and all its dependents locked, but everything else requires to
only have rdev and its supply locked.
Link: https://lore.kernel.org/all/aRn_-o-vie_QoDXD@sirena.co.uk/ [1]
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-8-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:44 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core: Add early supply resolution for regulators
The call to set_machine_constraints() in regulator_register(), will
attempt to get the voltage for the regulator. If a regulator is in
bypass will fail to get the voltage (ie. it's bypass voltage) and
hence register the regulator, because the supply for the regulator has
not been resolved yet.
To fix this, add a call to regulator_resolve_supply() before we call
set_machine_constraints(). If the call to regulator_resolve_supply()
fails, rather than returning an error at this point, allow the
registration of the regulator to continue because for some regulators
resolving the supply at this point may not be necessary and it will be
resolved later as more regulators are added. Furthermore, if the supply
is still not resolved for a bypassed regulator, this will be detected
when we attempt to get the voltage for the regulator and an error will
be propagated at this point.
If a bypassed regulator does not have a supply when we attempt to get
the voltage, rather than returing -EINVAL, return -EPROBE_DEFER instead
to allow the registration of the regulator to be deferred and tried
again later.
Please note that regulator_resolve_supply() will call
regulator_dev_lookup() which may acquire the regulator_list_mutex. To
avoid any deadlocks we cannot hold the regulator_list_mutex when calling
regulator_resolve_supply(). Therefore, rather than holding the lock
around a large portion of the registration code, just hold the lock when
aquiring any GPIOs and setting up supplies because these sections may
add entries to the regulator_map_list and regulator_ena_gpio_list,
respectively.
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2016-04-26 11:29:45 +01:00
|
|
|
|
2018-10-05 18:36:30 +03:00
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:40 +02:00
|
|
|
|
|
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
2011-11-29 18:11:19 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-02-24 17:39:09 +00:00
|
|
|
|
2011-11-29 18:11:19 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-02-24 17:39:09 +00:00
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
|
|
|
|
|
2017-03-27 16:54:12 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-07-24 03:50:13 +03:00
|
|
|
|
|
|
|
|
|
2016-04-21 17:11:59 +01:00
|
|
|
|
|
|
|
|
|
regulator: core: reresolve unresolved supplies when available
When a regulator A and its supply B are provided by different devices,
the driver implementing B might be last to probe (with A still
pending resolution of its supply B). While we try to resolve all
pending supplies for all regulators (including A) during
regulator_register() of B via regulator_register_resolve_supply(),
supply resolution will still not work for A as the driver for B hasn't
finished binding to the PMIC device corresponding to B at that stage
yet. The regulator core explicitly only allows supplies from other
devices to be used once the relevant driver has fully bound, mainly to
avoid having to deal with cases where B itself might -EPROBE_DEFER.
In this case, A's supply will only be resolved as part of the core's
regulator_init_complete_work_function(), which currently is scheduled
to run after 30s. This was added as a work-around in
commit 3827b64dba27 ("regulator: core: Resolve supplies before
disabling unused regulators") to cover this situation.
There are two problems with that approach:
* it potentially runs long after all our consumers have probed
* an upcoming change will allow regulator_register() to complete
successfully even when required supplies (e.g. due to always-on or
boot-on) are missing at register time, deferring full configuration
of the regulator (and usability by consumers, i.e. usually consumer
probe) until the supply becomes available.
Resolving supplies in the late work func can therefore make it
impossible for consumers to probe at all, as the driver core will not
know to reprobe consumers when supplies have resolved.
We could schedule an earlier work to try to resolve supplies sooner,
but that'd be racy as consumers of A might try to probe before A's
supply gets fully resolved via this extra work.
Instead, add a very simple regulator bus and add a dummy device with a
corresponding driver to it for each regulator that is missing its
supply during regulator_register(). This way, the driver core will call
our bus' probe whenever a new (regulator) device was successfully
bound, allowing us to retry resolving the supply during (our bus) probe
and to bind this dummy device if successful. In turn this means the
driver core will see a newly bound device and retry probing of all
pending consumers, if any.
With that in place, we can avoid walking the full list of all known
regulators to try resolve missing supplies during regulator_register(),
as the driver core will invoke the bus probe for regulators that are
still pending their supplies. We can also drop the code trying to
resolve supplies one last time before unused regulators get disabled,
as all supplies should have resolved at that point in time, and if they
haven't then there's no point in trying again, as the outcome won't
change.
Note: We can not reuse the existing struct device created for each
rail, as a device can not be part of a class and a bus simultaneously.
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-7-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:43 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2010-12-21 23:49:31 +00:00
|
|
|
|
2016-03-23 20:59:34 -03:00
|
|
|
|
2018-10-05 18:36:30 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-01-05 12:48:41 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2008-11-11 17:38:53 -08:00
|
|
|
|
regulator: core: reresolve unresolved supplies when available
When a regulator A and its supply B are provided by different devices,
the driver implementing B might be last to probe (with A still
pending resolution of its supply B). While we try to resolve all
pending supplies for all regulators (including A) during
regulator_register() of B via regulator_register_resolve_supply(),
supply resolution will still not work for A as the driver for B hasn't
finished binding to the PMIC device corresponding to B at that stage
yet. The regulator core explicitly only allows supplies from other
devices to be used once the relevant driver has fully bound, mainly to
avoid having to deal with cases where B itself might -EPROBE_DEFER.
In this case, A's supply will only be resolved as part of the core's
regulator_init_complete_work_function(), which currently is scheduled
to run after 30s. This was added as a work-around in
commit 3827b64dba27 ("regulator: core: Resolve supplies before
disabling unused regulators") to cover this situation.
There are two problems with that approach:
* it potentially runs long after all our consumers have probed
* an upcoming change will allow regulator_register() to complete
successfully even when required supplies (e.g. due to always-on or
boot-on) are missing at register time, deferring full configuration
of the regulator (and usability by consumers, i.e. usually consumer
probe) until the supply becomes available.
Resolving supplies in the late work func can therefore make it
impossible for consumers to probe at all, as the driver core will not
know to reprobe consumers when supplies have resolved.
We could schedule an earlier work to try to resolve supplies sooner,
but that'd be racy as consumers of A might try to probe before A's
supply gets fully resolved via this extra work.
Instead, add a very simple regulator bus and add a dummy device with a
corresponding driver to it for each regulator that is missing its
supply during regulator_register(). This way, the driver core will call
our bus' probe whenever a new (regulator) device was successfully
bound, allowing us to retry resolving the supply during (our bus) probe
and to bind this dummy device if successful. In turn this means the
driver core will see a newly bound device and retry probing of all
pending consumers, if any.
With that in place, we can avoid walking the full list of all known
regulators to try resolve missing supplies during regulator_register(),
as the driver core will invoke the bus probe for regulators that are
still pending their supplies. We can also drop the code trying to
resolve supplies one last time before unused regulators get disabled,
as all supplies should have resolved at that point in time, and if they
haven't then there's no point in trying again, as the outcome won't
change.
Note: We can not reuse the existing struct device created for each
rail, as a device can not be part of a class and a bus simultaneously.
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-7-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:43 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2010-04-29 10:55:11 +03:00
|
|
|
|
regulator: core: Add early supply resolution for regulators
The call to set_machine_constraints() in regulator_register(), will
attempt to get the voltage for the regulator. If a regulator is in
bypass will fail to get the voltage (ie. it's bypass voltage) and
hence register the regulator, because the supply for the regulator has
not been resolved yet.
To fix this, add a call to regulator_resolve_supply() before we call
set_machine_constraints(). If the call to regulator_resolve_supply()
fails, rather than returning an error at this point, allow the
registration of the regulator to continue because for some regulators
resolving the supply at this point may not be necessary and it will be
resolved later as more regulators are added. Furthermore, if the supply
is still not resolved for a bypassed regulator, this will be detected
when we attempt to get the voltage for the regulator and an error will
be propagated at this point.
If a bypassed regulator does not have a supply when we attempt to get
the voltage, rather than returing -EINVAL, return -EPROBE_DEFER instead
to allow the registration of the regulator to be deferred and tried
again later.
Please note that regulator_resolve_supply() will call
regulator_dev_lookup() which may acquire the regulator_list_mutex. To
avoid any deadlocks we cannot hold the regulator_list_mutex when calling
regulator_resolve_supply(). Therefore, rather than holding the lock
around a large portion of the registration code, just hold the lock when
aquiring any GPIOs and setting up supplies because these sections may
add entries to the regulator_map_list and regulator_ena_gpio_list,
respectively.
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2016-04-26 11:29:45 +01:00
|
|
|
|
2010-04-29 10:55:11 +03:00
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
regulator: core: Add early supply resolution for regulators
The call to set_machine_constraints() in regulator_register(), will
attempt to get the voltage for the regulator. If a regulator is in
bypass will fail to get the voltage (ie. it's bypass voltage) and
hence register the regulator, because the supply for the regulator has
not been resolved yet.
To fix this, add a call to regulator_resolve_supply() before we call
set_machine_constraints(). If the call to regulator_resolve_supply()
fails, rather than returning an error at this point, allow the
registration of the regulator to continue because for some regulators
resolving the supply at this point may not be necessary and it will be
resolved later as more regulators are added. Furthermore, if the supply
is still not resolved for a bypassed regulator, this will be detected
when we attempt to get the voltage for the regulator and an error will
be propagated at this point.
If a bypassed regulator does not have a supply when we attempt to get
the voltage, rather than returing -EINVAL, return -EPROBE_DEFER instead
to allow the registration of the regulator to be deferred and tried
again later.
Please note that regulator_resolve_supply() will call
regulator_dev_lookup() which may acquire the regulator_list_mutex. To
avoid any deadlocks we cannot hold the regulator_list_mutex when calling
regulator_resolve_supply(). Therefore, rather than holding the lock
around a large portion of the registration code, just hold the lock when
aquiring any GPIOs and setting up supplies because these sections may
add entries to the regulator_map_list and regulator_ena_gpio_list,
respectively.
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2016-04-26 11:29:45 +01:00
|
|
|
|
2016-02-24 11:52:50 +01:00
|
|
|
|
regulator: core: fix resource leak in regulator_register()
I got some resource leak reports while doing fault injection test:
OF: ERROR: memory leak, expected refcount 1 instead of 100,
of_node_get()/of_node_put() unbalanced - destroy cset entry:
attach overlay node /i2c/pmic@64/regulators/buck1
unreferenced object 0xffff88810deea000 (size 512):
comm "490-i2c-rt5190a", pid 253, jiffies 4294859840 (age 5061.046s)
hex dump (first 32 bytes):
00 00 00 00 ad 4e ad de ff ff ff ff 00 00 00 00 .....N..........
ff ff ff ff ff ff ff ff a0 1e 00 a1 ff ff ff ff ................
backtrace:
[<00000000d78541e2>] kmalloc_trace+0x21/0x110
[<00000000b343d153>] device_private_init+0x32/0xd0
[<00000000be1f0c70>] device_add+0xb2d/0x1030
[<00000000e3e6344d>] regulator_register+0xaf2/0x12a0
[<00000000e2f5e754>] devm_regulator_register+0x57/0xb0
[<000000008b898197>] rt5190a_probe+0x52a/0x861 [rt5190a_regulator]
unreferenced object 0xffff88810b617b80 (size 32):
comm "490-i2c-rt5190a", pid 253, jiffies 4294859904 (age 5060.983s)
hex dump (first 32 bytes):
72 65 67 75 6c 61 74 6f 72 2e 32 38 36 38 2d 53 regulator.2868-S
55 50 50 4c 59 00 ff ff 29 00 00 00 2b 00 00 00 UPPLY...)...+...
backtrace:
[<000000009da9280d>] __kmalloc_node_track_caller+0x44/0x1b0
[<0000000025c6a4e5>] kstrdup+0x3a/0x70
[<00000000790efb69>] create_regulator+0xc0/0x4e0
[<0000000005ed203a>] regulator_resolve_supply+0x2d4/0x440
[<0000000045796214>] regulator_register+0x10b3/0x12a0
[<00000000e2f5e754>] devm_regulator_register+0x57/0xb0
[<000000008b898197>] rt5190a_probe+0x52a/0x861 [rt5190a_regulator]
After calling regulator_resolve_supply(), the 'rdev->supply' is set
by set_supply(), after this set, in the error path, the resources
need be released, so call regulator_put() to avoid the leaks.
Fixes: aea6cb99703e ("regulator: resolve supply after creating regulator")
Fixes: 8a866d527ac0 ("regulator: core: Resolve supply name earlier to prevent double-init")
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Link: https://lore.kernel.org/r/20221202025111.496402-1-yangyingliang@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>
2022-12-02 10:51:11 +08:00
|
|
|
|
2019-10-25 03:22:39 +03:00
|
|
|
|
regulator: core: Add early supply resolution for regulators
The call to set_machine_constraints() in regulator_register(), will
attempt to get the voltage for the regulator. If a regulator is in
bypass will fail to get the voltage (ie. it's bypass voltage) and
hence register the regulator, because the supply for the regulator has
not been resolved yet.
To fix this, add a call to regulator_resolve_supply() before we call
set_machine_constraints(). If the call to regulator_resolve_supply()
fails, rather than returning an error at this point, allow the
registration of the regulator to continue because for some regulators
resolving the supply at this point may not be necessary and it will be
resolved later as more regulators are added. Furthermore, if the supply
is still not resolved for a bypassed regulator, this will be detected
when we attempt to get the voltage for the regulator and an error will
be propagated at this point.
If a bypassed regulator does not have a supply when we attempt to get
the voltage, rather than returing -EINVAL, return -EPROBE_DEFER instead
to allow the registration of the regulator to be deferred and tried
again later.
Please note that regulator_resolve_supply() will call
regulator_dev_lookup() which may acquire the regulator_list_mutex. To
avoid any deadlocks we cannot hold the regulator_list_mutex when calling
regulator_resolve_supply(). Therefore, rather than holding the lock
around a large portion of the registration code, just hold the lock when
aquiring any GPIOs and setting up supplies because these sections may
add entries to the regulator_map_list and regulator_ena_gpio_list,
respectively.
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2016-04-26 11:29:45 +01:00
|
|
|
|
2016-02-24 11:52:50 +01:00
|
|
|
|
regulator: core: Add early supply resolution for regulators
The call to set_machine_constraints() in regulator_register(), will
attempt to get the voltage for the regulator. If a regulator is in
bypass will fail to get the voltage (ie. it's bypass voltage) and
hence register the regulator, because the supply for the regulator has
not been resolved yet.
To fix this, add a call to regulator_resolve_supply() before we call
set_machine_constraints(). If the call to regulator_resolve_supply()
fails, rather than returning an error at this point, allow the
registration of the regulator to continue because for some regulators
resolving the supply at this point may not be necessary and it will be
resolved later as more regulators are added. Furthermore, if the supply
is still not resolved for a bypassed regulator, this will be detected
when we attempt to get the voltage for the regulator and an error will
be propagated at this point.
If a bypassed regulator does not have a supply when we attempt to get
the voltage, rather than returing -EINVAL, return -EPROBE_DEFER instead
to allow the registration of the regulator to be deferred and tried
again later.
Please note that regulator_resolve_supply() will call
regulator_dev_lookup() which may acquire the regulator_list_mutex. To
avoid any deadlocks we cannot hold the regulator_list_mutex when calling
regulator_resolve_supply(). Therefore, rather than holding the lock
around a large portion of the registration code, just hold the lock when
aquiring any GPIOs and setting up supplies because these sections may
add entries to the regulator_map_list and regulator_ena_gpio_list,
respectively.
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
2016-04-26 11:29:45 +01:00
|
|
|
|
2008-11-11 17:38:53 -08:00
|
|
|
|
2018-12-06 13:43:37 +01:00
|
|
|
|
|
|
|
|
|
2016-03-30 17:09:13 +01:00
|
|
|
|
2023-09-19 00:50:27 +02:00
|
|
|
|
2018-12-06 13:43:37 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-03-30 17:09:13 +01:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-12-31 12:52:41 +00:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2013-07-08 09:14:45 +01:00
|
|
|
|
2025-10-01 12:56:49 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2013-07-08 09:14:45 +01:00
|
|
|
|
|
|
|
|
|
2012-03-28 21:17:55 +01:00
|
|
|
|
2013-07-08 09:14:45 +01:00
|
|
|
|
2018-11-19 00:56:19 +03:00
|
|
|
|
2019-04-04 16:32:18 +01:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2018-11-19 00:56:19 +03:00
|
|
|
|
2009-07-21 16:00:25 +01:00
|
|
|
|
2018-10-05 18:36:36 +03:00
|
|
|
|
2009-01-22 16:00:29 +02:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2013-02-18 06:50:39 +00:00
|
|
|
|
regulator: core: reresolve unresolved supplies when available
When a regulator A and its supply B are provided by different devices,
the driver implementing B might be last to probe (with A still
pending resolution of its supply B). While we try to resolve all
pending supplies for all regulators (including A) during
regulator_register() of B via regulator_register_resolve_supply(),
supply resolution will still not work for A as the driver for B hasn't
finished binding to the PMIC device corresponding to B at that stage
yet. The regulator core explicitly only allows supplies from other
devices to be used once the relevant driver has fully bound, mainly to
avoid having to deal with cases where B itself might -EPROBE_DEFER.
In this case, A's supply will only be resolved as part of the core's
regulator_init_complete_work_function(), which currently is scheduled
to run after 30s. This was added as a work-around in
commit 3827b64dba27 ("regulator: core: Resolve supplies before
disabling unused regulators") to cover this situation.
There are two problems with that approach:
* it potentially runs long after all our consumers have probed
* an upcoming change will allow regulator_register() to complete
successfully even when required supplies (e.g. due to always-on or
boot-on) are missing at register time, deferring full configuration
of the regulator (and usability by consumers, i.e. usually consumer
probe) until the supply becomes available.
Resolving supplies in the late work func can therefore make it
impossible for consumers to probe at all, as the driver core will not
know to reprobe consumers when supplies have resolved.
We could schedule an earlier work to try to resolve supplies sooner,
but that'd be racy as consumers of A might try to probe before A's
supply gets fully resolved via this extra work.
Instead, add a very simple regulator bus and add a dummy device with a
corresponding driver to it for each regulator that is missing its
supply during regulator_register(). This way, the driver core will call
our bus' probe whenever a new (regulator) device was successfully
bound, allowing us to retry resolving the supply during (our bus) probe
and to bind this dummy device if successful. In turn this means the
driver core will see a newly bound device and retry probing of all
pending consumers, if any.
With that in place, we can avoid walking the full list of all known
regulators to try resolve missing supplies during regulator_register(),
as the driver core will invoke the bus probe for regulators that are
still pending their supplies. We can also drop the code trying to
resolve supplies one last time before unused regulators get disabled,
as all supplies should have resolved at that point in time, and if they
haven't then there's no point in trying again, as the outcome won't
change.
Note: We can not reuse the existing struct device created for each
rail, as a device can not be part of a class and a bus simultaneously.
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-7-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:43 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-11-28 15:38:37 +01:00
|
|
|
|
2018-11-19 00:56:19 +03:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-26 21:08:47 +08:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2018-07-05 14:25:57 +00:00
|
|
|
|
2018-09-04 08:30:05 -07:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2018-07-05 14:25:57 +00:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2018-09-03 16:49:36 +02:00
|
|
|
|
2018-01-26 21:08:47 +08:00
|
|
|
|
2018-09-03 16:49:36 +02:00
|
|
|
|
2020-08-04 00:08:37 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2018-09-03 16:49:36 +02:00
|
|
|
|
2020-08-04 00:08:37 -07:00
|
|
|
|
2018-09-03 16:49:36 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
2018-03-22 15:23:35 +08:00
|
|
|
|
2018-09-03 16:49:36 +02:00
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
2018-09-03 16:49:36 +02:00
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
2018-01-26 21:08:47 +08:00
|
|
|
|
2018-09-03 16:49:36 +02:00
|
|
|
|
2018-01-26 21:08:47 +08:00
|
|
|
|
2018-09-03 16:49:36 +02:00
|
|
|
|
2018-01-26 21:08:47 +08:00
|
|
|
|
2018-02-13 10:37:59 +01:00
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
2020-08-04 00:08:37 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
2020-08-04 00:08:37 -07:00
|
|
|
|
|
|
|
|
|
2018-07-05 14:25:57 +00:00
|
|
|
|
2018-01-26 21:08:47 +08:00
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
2018-01-26 21:08:47 +08:00
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
2018-01-26 21:08:47 +08:00
|
|
|
|
|
|
|
|
|
2018-07-05 14:25:57 +00:00
|
|
|
|
|
|
|
|
|
2018-01-26 21:08:47 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-07-05 14:25:57 +00:00
|
|
|
|
|
|
|
|
|
2018-01-26 21:08:47 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-03-05 16:00:45 -03:00
|
|
|
|
2018-01-26 21:08:47 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core: reresolve unresolved supplies when available
When a regulator A and its supply B are provided by different devices,
the driver implementing B might be last to probe (with A still
pending resolution of its supply B). While we try to resolve all
pending supplies for all regulators (including A) during
regulator_register() of B via regulator_register_resolve_supply(),
supply resolution will still not work for A as the driver for B hasn't
finished binding to the PMIC device corresponding to B at that stage
yet. The regulator core explicitly only allows supplies from other
devices to be used once the relevant driver has fully bound, mainly to
avoid having to deal with cases where B itself might -EPROBE_DEFER.
In this case, A's supply will only be resolved as part of the core's
regulator_init_complete_work_function(), which currently is scheduled
to run after 30s. This was added as a work-around in
commit 3827b64dba27 ("regulator: core: Resolve supplies before
disabling unused regulators") to cover this situation.
There are two problems with that approach:
* it potentially runs long after all our consumers have probed
* an upcoming change will allow regulator_register() to complete
successfully even when required supplies (e.g. due to always-on or
boot-on) are missing at register time, deferring full configuration
of the regulator (and usability by consumers, i.e. usually consumer
probe) until the supply becomes available.
Resolving supplies in the late work func can therefore make it
impossible for consumers to probe at all, as the driver core will not
know to reprobe consumers when supplies have resolved.
We could schedule an earlier work to try to resolve supplies sooner,
but that'd be racy as consumers of A might try to probe before A's
supply gets fully resolved via this extra work.
Instead, add a very simple regulator bus and add a dummy device with a
corresponding driver to it for each regulator that is missing its
supply during regulator_register(). This way, the driver core will call
our bus' probe whenever a new (regulator) device was successfully
bound, allowing us to retry resolving the supply during (our bus) probe
and to bind this dummy device if successful. In turn this means the
driver core will see a newly bound device and retry probing of all
pending consumers, if any.
With that in place, we can avoid walking the full list of all known
regulators to try resolve missing supplies during regulator_register(),
as the driver core will invoke the bus probe for regulators that are
still pending their supplies. We can also drop the code trying to
resolve supplies one last time before unused regulators get disabled,
as all supplies should have resolved at that point in time, and if they
haven't then there's no point in trying again, as the outcome won't
change.
Note: We can not reuse the existing struct device created for each
rail, as a device can not be part of a class and a bus simultaneously.
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-7-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:43 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-03-16 19:36:34 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2008-12-31 12:52:41 +00:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-10-23 18:33:13 +02:00
|
|
|
|
2008-12-31 12:52:41 +00:00
|
|
|
|
2024-08-29 16:51:24 +08:00
|
|
|
|
|
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-01-09 18:44:00 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
regulator: core - Rework machine API to remove string based functions.
This improves the machine level API in order to configure
regulator constraints and consumers as platform data and removes the
old string based API that required several calls to set up each regulator.
The intention is to create a struct regulator_init_data, populate
it's fields with constraints, consumers devices, etc and then register
the regulator device from board.c in the standard Linux way.
e.g. regulator LDO2 (supplying codec and sim) platform data.
/* regulator LDO2 consumer devices */
static struct regulator_consumer_supply ldo2_consumers[] = {
{
.dev = &platform_audio_device.dev,
.supply = "codec_avdd",
},
{
.dev = &platform_sim_device.dev,
.supply = "sim_vcc",
}
};
/* regulator LDO2 constraints */
static struct regulator_init_data ldo2_data = {
.constraints = {
.min_uV = 3300000,
.max_uV = 3300000,
.valid_modes_mask = REGULATOR_MODE_NORMAL,
.apply_uV = 1,
},
.num_consumer_supplies = ARRAY_SIZE(ldo2_consumers),
.consumer_supplies = ldo2_consumers,
};
/* machine regulator devices with thier consumers and constraints */
static struct platform_device wm8350_regulator_devices[] = {
{
.name = "wm8350-regulator",
.id = WM8350_LDO_2,
.dev = {
.platform_data = &ldo2_data,
},
},
};
Changes in detail:-
o Removed all const char* regulator config functions in machine API.
o Created new struct regulator_init_data to contain regulator
machine configuration constraints and consmuers.
o Changed set_supply(), set_machine_constraints(),
set_consumer_device_supply() to remove their string identifier
parameters. Also made them static and moved functions nearer top of
core.c.
o Removed no longer used inline func to_rdev()
o Added regulator_get_init_drvdata() to retrieve init data.
o Added struct device* as parameter to regulator_register().
o Changed my email address.
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
2008-10-10 13:22:20 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-08-23 17:39:10 +01:00
|
|
|
|
2017-06-30 11:43:42 +08:00
|
|
|
|
2011-08-23 17:39:10 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-30 11:43:42 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2011-08-23 17:39:10 +01:00
|
|
|
|
|
|
|
|
|
2017-06-30 11:43:42 +08:00
|
|
|
|
|
|
|
|
|
2018-12-15 03:21:48 -05:00
|
|
|
|
2011-08-23 17:39:10 +01:00
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-04-07 16:16:39 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
2018-08-16 13:28:03 -07:00
|
|
|
|
2015-04-07 16:16:39 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-08-16 13:28:03 -07:00
|
|
|
|
2018-08-16 13:28:01 -07:00
|
|
|
|
2015-04-07 16:16:39 +02:00
|
|
|
|
|
|
|
|
|
2018-08-16 13:28:01 -07:00
|
|
|
|
2018-08-16 13:28:03 -07:00
|
|
|
|
2015-04-07 16:16:39 +02:00
|
|
|
|
2019-06-24 00:08:32 +03:00
|
|
|
|
2018-08-16 13:28:03 -07:00
|
|
|
|
|
|
|
|
|
2015-04-07 16:16:39 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-02-14 17:31:03 +02:00
|
|
|
|
2015-04-07 16:16:39 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-02-14 17:31:03 +02:00
|
|
|
|
2020-04-04 02:11:57 +02:00
|
|
|
|
2017-02-14 17:31:03 +02:00
|
|
|
|
2015-04-07 16:16:39 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
|
|
|
|
|
2018-08-16 13:28:02 -07:00
|
|
|
|
2018-11-20 09:52:53 -08:00
|
|
|
|
|
|
|
|
|
2018-01-26 21:08:44 +08:00
|
|
|
|
|
|
|
|
|
2015-04-07 16:16:39 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-04-07 16:16:39 +02:00
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-08-16 13:28:03 -07:00
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:19 +03:00
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-03-29 14:33:53 -07:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:19 +03:00
|
|
|
|
|
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-04-07 16:16:39 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
2018-08-16 13:28:01 -07:00
|
|
|
|
|
|
|
|
|
2015-04-07 16:16:39 +02:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
|
|
|
|
|
2015-04-07 16:16:39 +02:00
|
|
|
|
2018-11-19 00:56:17 +03:00
|
|
|
|
|
|
|
|
|
2015-04-07 16:16:39 +02:00
|
|
|
|
|
|
|
|
|
2018-12-15 03:21:48 -05:00
|
|
|
|
|
|
|
|
|
2015-04-07 16:16:39 +02:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
2010-02-12 10:18:08 +00:00
|
|
|
|
|
|
|
|
|
regulator: core: reresolve unresolved supplies when available
When a regulator A and its supply B are provided by different devices,
the driver implementing B might be last to probe (with A still
pending resolution of its supply B). While we try to resolve all
pending supplies for all regulators (including A) during
regulator_register() of B via regulator_register_resolve_supply(),
supply resolution will still not work for A as the driver for B hasn't
finished binding to the PMIC device corresponding to B at that stage
yet. The regulator core explicitly only allows supplies from other
devices to be used once the relevant driver has fully bound, mainly to
avoid having to deal with cases where B itself might -EPROBE_DEFER.
In this case, A's supply will only be resolved as part of the core's
regulator_init_complete_work_function(), which currently is scheduled
to run after 30s. This was added as a work-around in
commit 3827b64dba27 ("regulator: core: Resolve supplies before
disabling unused regulators") to cover this situation.
There are two problems with that approach:
* it potentially runs long after all our consumers have probed
* an upcoming change will allow regulator_register() to complete
successfully even when required supplies (e.g. due to always-on or
boot-on) are missing at register time, deferring full configuration
of the regulator (and usability by consumers, i.e. usually consumer
probe) until the supply becomes available.
Resolving supplies in the late work func can therefore make it
impossible for consumers to probe at all, as the driver core will not
know to reprobe consumers when supplies have resolved.
We could schedule an earlier work to try to resolve supplies sooner,
but that'd be racy as consumers of A might try to probe before A's
supply gets fully resolved via this extra work.
Instead, add a very simple regulator bus and add a dummy device with a
corresponding driver to it for each regulator that is missing its
supply during regulator_register(). This way, the driver core will call
our bus' probe whenever a new (regulator) device was successfully
bound, allowing us to retry resolving the supply during (our bus) probe
and to bind this dummy device if successful. In turn this means the
driver core will see a newly bound device and retry probing of all
pending consumers, if any.
With that in place, we can avoid walking the full list of all known
regulators to try resolve missing supplies during regulator_register(),
as the driver core will invoke the bus probe for regulators that are
still pending their supplies. We can also drop the code trying to
resolve supplies one last time before unused regulators get disabled,
as all supplies should have resolved at that point in time, and if they
haven't then there's no point in trying again, as the outcome won't
change.
Note: We can not reuse the existing struct device created for each
rail, as a device can not be part of a class and a bus simultaneously.
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-7-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:43 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2010-02-12 10:18:08 +00:00
|
|
|
|
regulator: core: reresolve unresolved supplies when available
When a regulator A and its supply B are provided by different devices,
the driver implementing B might be last to probe (with A still
pending resolution of its supply B). While we try to resolve all
pending supplies for all regulators (including A) during
regulator_register() of B via regulator_register_resolve_supply(),
supply resolution will still not work for A as the driver for B hasn't
finished binding to the PMIC device corresponding to B at that stage
yet. The regulator core explicitly only allows supplies from other
devices to be used once the relevant driver has fully bound, mainly to
avoid having to deal with cases where B itself might -EPROBE_DEFER.
In this case, A's supply will only be resolved as part of the core's
regulator_init_complete_work_function(), which currently is scheduled
to run after 30s. This was added as a work-around in
commit 3827b64dba27 ("regulator: core: Resolve supplies before
disabling unused regulators") to cover this situation.
There are two problems with that approach:
* it potentially runs long after all our consumers have probed
* an upcoming change will allow regulator_register() to complete
successfully even when required supplies (e.g. due to always-on or
boot-on) are missing at register time, deferring full configuration
of the regulator (and usability by consumers, i.e. usually consumer
probe) until the supply becomes available.
Resolving supplies in the late work func can therefore make it
impossible for consumers to probe at all, as the driver core will not
know to reprobe consumers when supplies have resolved.
We could schedule an earlier work to try to resolve supplies sooner,
but that'd be racy as consumers of A might try to probe before A's
supply gets fully resolved via this extra work.
Instead, add a very simple regulator bus and add a dummy device with a
corresponding driver to it for each regulator that is missing its
supply during regulator_register(). This way, the driver core will call
our bus' probe whenever a new (regulator) device was successfully
bound, allowing us to retry resolving the supply during (our bus) probe
and to bind this dummy device if successful. In turn this means the
driver core will see a newly bound device and retry probing of all
pending consumers, if any.
With that in place, we can avoid walking the full list of all known
regulators to try resolve missing supplies during regulator_register(),
as the driver core will invoke the bus probe for regulators that are
still pending their supplies. We can also drop the code trying to
resolve supplies one last time before unused regulators get disabled,
as all supplies should have resolved at that point in time, and if they
haven't then there's no point in trying again, as the outcome won't
change.
Note: We can not reuse the existing struct device created for each
rail, as a device can not be part of a class and a bus simultaneously.
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-7-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:43 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2010-02-12 10:18:08 +00:00
|
|
|
|
2010-12-21 23:49:31 +00:00
|
|
|
|
2023-05-15 22:29:38 +05:00
|
|
|
|
2023-05-25 13:13:59 +02:00
|
|
|
|
2011-08-23 17:39:10 +01:00
|
|
|
|
2018-12-15 03:21:48 -05:00
|
|
|
|
2012-02-20 21:01:04 +00:00
|
|
|
|
|
|
|
|
|
2010-12-21 23:49:31 +00:00
|
|
|
|
2015-04-07 16:16:39 +02:00
|
|
|
|
2015-09-21 16:02:47 +02:00
|
|
|
|
2018-12-15 03:21:48 -05:00
|
|
|
|
2010-02-12 10:18:08 +00:00
|
|
|
|
|
|
|
|
|
2019-06-24 00:08:31 +03:00
|
|
|
|
|
|
|
|
|
regulator: core: reresolve unresolved supplies when available
When a regulator A and its supply B are provided by different devices,
the driver implementing B might be last to probe (with A still
pending resolution of its supply B). While we try to resolve all
pending supplies for all regulators (including A) during
regulator_register() of B via regulator_register_resolve_supply(),
supply resolution will still not work for A as the driver for B hasn't
finished binding to the PMIC device corresponding to B at that stage
yet. The regulator core explicitly only allows supplies from other
devices to be used once the relevant driver has fully bound, mainly to
avoid having to deal with cases where B itself might -EPROBE_DEFER.
In this case, A's supply will only be resolved as part of the core's
regulator_init_complete_work_function(), which currently is scheduled
to run after 30s. This was added as a work-around in
commit 3827b64dba27 ("regulator: core: Resolve supplies before
disabling unused regulators") to cover this situation.
There are two problems with that approach:
* it potentially runs long after all our consumers have probed
* an upcoming change will allow regulator_register() to complete
successfully even when required supplies (e.g. due to always-on or
boot-on) are missing at register time, deferring full configuration
of the regulator (and usability by consumers, i.e. usually consumer
probe) until the supply becomes available.
Resolving supplies in the late work func can therefore make it
impossible for consumers to probe at all, as the driver core will not
know to reprobe consumers when supplies have resolved.
We could schedule an earlier work to try to resolve supplies sooner,
but that'd be racy as consumers of A might try to probe before A's
supply gets fully resolved via this extra work.
Instead, add a very simple regulator bus and add a dummy device with a
corresponding driver to it for each regulator that is missing its
supply during regulator_register(). This way, the driver core will call
our bus' probe whenever a new (regulator) device was successfully
bound, allowing us to retry resolving the supply during (our bus) probe
and to bind this dummy device if successful. In turn this means the
driver core will see a newly bound device and retry probing of all
pending consumers, if any.
With that in place, we can avoid walking the full list of all known
regulators to try resolve missing supplies during regulator_register(),
as the driver core will invoke the bus probe for regulators that are
still pending their supplies. We can also drop the code trying to
resolve supplies one last time before unused regulators get disabled,
as all supplies should have resolved at that point in time, and if they
haven't then there's no point in trying again, as the outcome won't
change.
Note: We can not reuse the existing struct device created for each
rail, as a device can not be part of a class and a bus simultaneously.
Signed-off-by: André Draszik <andre.draszik@linaro.org>
Link: https://patch.msgid.link/20260109-regulators-defer-v2-7-1a25dc968e60@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
2026-01-09 08:38:43 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2010-02-12 10:18:08 +00:00
|
|
|
|
2008-04-30 15:59:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-03-16 19:36:34 +00:00
|
|
|
|
2019-09-04 13:42:50 +01:00
|
|
|
|
2009-03-16 19:36:34 +00:00
|
|
|
|
2015-08-10 19:43:47 +01:00
|
|
|
|
|
|
|
|
|
2022-02-08 09:46:45 +01:00
|
|
|
|
2009-03-16 19:36:34 +00:00
|
|
|
|
2015-08-10 19:43:47 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-04-23 15:11:05 +08:00
|
|
|
|
2015-08-10 19:43:47 +01:00
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2015-08-10 19:43:47 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-02-08 09:46:45 +01:00
|
|
|
|
|
|
|
|
|
2015-08-10 19:43:47 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-04-20 09:17:18 +05:30
|
|
|
|
|
|
|
|
|
2015-08-10 19:43:47 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-26 23:32:40 +02:00
|
|
|
|
2015-08-10 19:43:47 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-04-23 16:33:37 +02:00
|
|
|
|
2015-08-10 19:43:47 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-11-07 20:09:18 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-09-04 13:42:50 +01:00
|
|
|
|
2015-08-10 19:43:47 +01:00
|
|
|
|
2023-11-07 20:09:18 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-03-16 19:36:34 +00:00
|
|
|
|
2014-06-01 19:15:16 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-03-16 19:36:34 +00:00
|
|
|
|
2015-08-10 19:43:47 +01:00
|
|
|
|
|
|
|
|
|
2019-09-04 13:42:50 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-04-29 17:23:49 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-09-04 13:42:50 +01:00
|
|
|
|
2020-04-29 17:23:49 +00:00
|
|
|
|
|
|
|
|
|
2009-03-16 19:36:34 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-04-23 18:10:50 -05:00
|
|
|
|