regulator/core: Trivial cleanups and improvements
cover regulator/core: Trivial cleanups and improvements
This is a random set of cleanups, dead code removal and minor optimizations. Most patches are applicable independently; those that aren't share a bit of context, but are nevertheless logically separate.
Michał Mirosław (9):
regulator/core: _regulator_get: simplify error returns regulator/core: set_consumer_device_supply: remove has_dev regulator/core: of_get_child_regulator: remove goto regulator/core: regulator_bulk_get: remove redundant NULL stores regulator/core: regulator_ena_gpio_ctrl: pull in ena_gpio state
handling
regulator/core: remove regulator_init callback regulator/core: remove regulator_get_init_drvdata() regulator/core: set_consumer_device_supply: avoid copying const data regulator/core: make regulator_class const
drivers/regulator/core.c | 93 ++++++++++--------------------- drivers/regulator/internal.h | 2 +- include/linux/regulator/driver.h | 1 - include/linux/regulator/machine.h | 7 +-- 4 files changed, 33 insertions(+), 70 deletions(-)
Signature · 1 line
2.39.2
PATCH 1/9 regulator/core: _regulator_get: simplify error returns
Remove unnecessary stores to regulator.
--- drivers/regulator/core.c | 18 ++++++------------ 1 file changed, 6 insertions(+), 12 deletions(-)
drivers/regulator/core.c +6 -12
@@ -2209,15 +2209,13 @@ struct regulator *_regulator_get(struct device *dev, const char *id, }if (rdev->exclusive) { - regulator =ERR_PTR (-EPERM );put_device (&rdev->dev); -return regulator; +return ERR_PTR (-EPERM ); }if (get_type ==EXCLUSIVE_GET && rdev->open_count) { - regulator =ERR_PTR (-EBUSY );put_device (&rdev->dev); -return regulator; +return ERR_PTR (-EBUSY ); }mutex_lock (®ulator_list_mutex);
@@ -2225,32 +2223,28 @@ struct regulator *_regulator_get(struct device *dev, const char *id,mutex_unlock (®ulator_list_mutex);if (ret !=0 ) { - regulator =ERR_PTR (-EPROBE_DEFER );put_device (&rdev->dev); -return regulator; +return ERR_PTR (-EPROBE_DEFER ); } ret =regulator_resolve_supply (rdev);if (ret <0 ) { - regulator =ERR_PTR (ret);put_device (&rdev->dev); -return regulator; +return ERR_PTR (ret); }if (!try_module_get (rdev->owner)) { - regulator =ERR_PTR (-EPROBE_DEFER );put_device (&rdev->dev); -return regulator; +return ERR_PTR (-EPROBE_DEFER ); }regulator_lock (rdev); regulator =create_regulator (rdev, dev, id);regulator_unlock (rdev);if (regulator ==NULL ) { - regulator =ERR_PTR (-ENOMEM );module_put (rdev->owner);put_device (&rdev->dev); -return regulator; +return ERR_PTR (-ENOMEM ); } rdev->open_count++;
Signature · 1 line
2.39.2
PATCH 2/9 regulator/core: set_consumer_device_supply: remove `has_dev`
In reply to Michał Mirosławhas_dev is only ever used once to check if the name is non-NULL. Inline the check and make the intent obvious.
--- drivers/regulator/core.c | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-)
drivers/regulator/core.c +1 -7
@@ -1726,16 +1726,10 @@ static int set_consumer_device_supply(struct regulator_dev *rdev, const char *supply) {struct regulator_map *node, *new_node; -int has_dev;if (supply ==NULL )return -EINVAL ; -if (consumer_dev_name !=NULL ) - has_dev =1 ; -else - has_dev =0 ; - new_node =kzalloc (sizeof (struct regulator_map ),GFP_KERNEL );if (new_node ==NULL )return -ENOMEM ;
@@ -1743,7 +1737,7 @@ static int set_consumer_device_supply(struct regulator_dev *rdev, new_node->regulator = rdev; new_node->supply = supply; -if (has_dev) { +if (consumer_dev_name !=NULL ) { new_node->dev_name =kstrdup (consumer_dev_name,GFP_KERNEL );if (new_node->dev_name ==NULL ) {kfree (new_node);
Signature · 1 line
2.39.2
PATCH 3/9 regulator/core: of_get_child_regulator: remove goto
In reply to Michał MirosławBecause of_node_put() handles NULL properly (like kfree() et al) we can call it also after the loop ends (due to child == NULL). This makes the gotos redundant.
--- drivers/regulator/core.c | 15 ++++++--------- 1 file changed, 6 insertions(+), 9 deletions(-)
drivers/regulator/core.c +6 -9
@@ -426,18 +426,15 @@ static struct device_node *of_get_child_regulator(struct device_node *parent,for_each_child_of_node (parent, child) { regnode =of_parse_phandle (child, prop_name,0 ); +if (regnode) +break ; -if (!regnode) { - regnode =of_get_child_regulator (child, prop_name); -if (regnode) - goto err_node_put; - }else { - goto err_node_put; - } + regnode =of_get_child_regulator (child, prop_name); +if (regnode) +break ; } -return NULL ; -err_node_put: +/* Release the node if the loop was exited early. */ of_node_put (child);return regnode; }
Signature · 1 line
2.39.2
PATCH 4/9 regulator/core: regulator_bulk_get: remove redundant NULL stores
In reply to Michał MirosławOn error, callers of regulator_bulk_get() pass the error up and don't use the pointers in consumers[]. The function is documented to release all regulators if any request fails.
Note: if an i-th regulator_get() failed only the i-th pointer was cleared. This is another suggestion that the clearing was unnecessary.
--- drivers/regulator/core.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-)
drivers/regulator/core.c +1 -6
@@ -4829,11 +4829,7 @@ static int _notifier_call_chain(struct regulator_dev *rdev,int _regulator_bulk_get (struct device *dev,int num_consumers,struct regulator_bulk_data *consumers,enum regulator_get_type get_type) { -int i; -int ret; - -for (i =0 ; i < num_consumers; i++) - consumers[i].consumer =NULL ; +int ret, i;for (i =0 ; i < num_consumers; i++) { consumers[i].consumer = _regulator_get(dev,
@@ -4842,7 +4838,6 @@ int _regulator_bulk_get(struct device *dev, int num_consumers, ret =dev_err_probe (dev,PTR_ERR (consumers[i].consumer),"Failed to get supply '%s'" , consumers[i].supply); - consumers[i].consumer =NULL ; goto err; }
Signature · 1 line
2.39.2
PATCH 5/9 regulator/core: regulator_ena_gpio_ctrl: pull in ena_gpio state handling
In reply to Michał MirosławDeduplicate ena_gpio_state handling by pulling it into regulator_ena_gpio_ctrl().
--- drivers/regulator/core.c | 30 +++++++++++------------------- 1 file changed, 11 insertions(+), 19 deletions(-)
drivers/regulator/core.c +11 -19
@@ -2601,6 +2601,9 @@ static int regulator_ena_gpio_ctrl(struct regulator_dev *rdev, bool enable)if (!pin)return -EINVAL ; +if (rdev->ena_gpio_state == enable) +return 0 ; +if (enable) {/* Enable GPIO at initial use */ if (pin->enable_count ==0 )
@@ -2608,18 +2611,14 @@ static int regulator_ena_gpio_ctrl(struct regulator_dev *rdev, bool enable) pin->enable_count++; }else { -if (pin->enable_count >1 ) { - pin->enable_count--; -return 0 ; - } -/* Disable GPIO if not used */ -if (pin->enable_count <=1 ) { +if (pin->enable_count-- <=1 ) {gpiod_set_value_cansleep (pin->gpiod,0 ); pin->enable_count =0 ; } } + rdev->ena_gpio_state = enable;return 0 ; }
@@ -2720,12 +2719,9 @@ static int _regulator_do_enable(struct regulator_dev *rdev) }if (rdev->ena_pin) { -if (!rdev->ena_gpio_state) { - ret =regulator_ena_gpio_ctrl (rdev, true); -if (ret <0 ) -return ret; - rdev->ena_gpio_state =1 ; - } + ret =regulator_ena_gpio_ctrl (rdev, true); +if (ret <0 ) +return ret; }else if (rdev->desc->ops->enable) { ret = rdev->desc->ops->enable (rdev);if (ret <0 )
@@ -2938,13 +2934,9 @@ static int _regulator_do_disable(struct regulator_dev *rdev)trace_regulator_disable (rdev_get_name (rdev));if (rdev->ena_pin) { -if (rdev->ena_gpio_state) { - ret =regulator_ena_gpio_ctrl (rdev, false); -if (ret <0 ) -return ret; - rdev->ena_gpio_state =0 ; - } - + ret =regulator_ena_gpio_ctrl (rdev, false); +if (ret <0 ) +return ret; }else if (rdev->desc->ops->disable) { ret = rdev->desc->ops->disable (rdev);if (ret !=0 )
Signature · 1 line
2.39.2
Mark Brown
maintainerRe: [PATCH 5/9] regulator/core: regulator_ena_gpio_ctrl: pull in ena_gpio state handling
drivers/regulator/core.c static int regulator_ena_gpio_ctrl(struct regulator_dev *rdev, bool enable) · line 2615
Quoted 8 lines · Michał Mirosław
drivers/regulator/core.c +1
-6-if (pin->enable_count >1 ) { - pin->enable_count--; -return 0 ; - } -/* Disable GPIO if not used */ -if (pin->enable_count <=1 ) { +if (pin->enable_count-- <=1 ) {
The goal isn't to write the minimum number of lines possible - this just makes the logic harder to follow and for bonus points isn't obviously related to the chnages described in changelog.
Re: [PATCH 5/9] regulator/core: regulator_ena_gpio_ctrl: pull in ena_gpio state handling
Quoted 14 lines · Mark Brown
drivers/regulator/core.c static int regulator_ena_gpio_ctrl(struct regulator_dev *rdev, bool enable) · line 2615
Quoted 8 lines · Michał Mirosław
drivers/regulator/core.c +1
-6-if (pin->enable_count >1 ) { - pin->enable_count--; -return 0 ; - } -/* Disable GPIO if not used */ -if (pin->enable_count <=1 ) { +if (pin->enable_count-- <=1 ) {The goal isn't to write the minimum number of lines possible - this just makes the logic harder to follow and for bonus points isn't obviously related to the chnages described in changelog.
I see that I missed this fragment when splitting patches. I'll resend without this part if the other ones are good to go.
Best Regards
Michał Mirosław
A message that is not in the archive (sent off-list or not yet received): 1c102d15cb8f20141f0bfd1e42d5ac8a98947154.1693485621.git.mirq-linux@rere.qmqm.pl
Mark Brown
maintainerRe: [PATCH 5/9 fixed] regulator/core: regulator_ena_gpio_ctrl: pull in ena_gpio state handling
Quoted 7 lines · Michał Mirosław
Deduplicate
ena_gpio_statehandling by pulling it intoregulator_ena_gpio_ctrl().Signed-off-by Michał Mirosław <[email protected]>---
This is a fixed version of patch #5 with the extra code change removed. I'll resend the series if needed after other patches are reviewed.
Please don't do this, it just makes everything harder to follow. Send the whole series when resending.
PATCH 6/9 regulator/core: remove regulator_init callback
In reply to Michał MirosławThere are no in-tree users. The only usage went away in 2019 in commit 8c44e448583c ("regulator: stpmic1: Simplify regulators registration").
--- include/linux/regulator/machine.h | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-)
include/linux/regulator/machine.h +2 -5
@@ -256,8 +256,7 @@ struct regulator_consumer_supply {* @num_consumer_supplies: Number of consumer device supplies. * @consumer_supplies: Consumer device supply configuration. * -* @regulator_init: Callback invoked when the regulator has been registered. -* @driver_data: Data passed to regulator_init. +* @driver_data: Pointer copied to regulator_dev.reg_data. */ struct regulator_init_data {const char *supply_regulator;/* or NULL for system supply */
@@ -267,9 +266,7 @@ struct regulator_init_data {int num_consumer_supplies;struct regulator_consumer_supply *consumer_supplies; -/* optional regulator machine specific init */ -int (*regulator_init)(void *driver_data); -void *driver_data;/* core does not touch this */ +void *driver_data; };#ifdef CONFIG_REGULATOR
Signature · 1 line
2.39.2
kernel test robot
build botRe: [PATCH 6/9] regulator/core: remove regulator_init callback
kernel test robot noticed the following build errors:
Hi Michał,
kernel test robot noticed the following build errors:
[auto build test ERROR on broonie-regulator/for-next]
[also build test ERROR on linus/master v6.5 next-20230831] [If your patch is applied to the wrong git tree, kindly drop us a note. And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]
url: https://github.com/intel-lab-lkp/linux/commits/Micha-Miros-aw/regulator-core-_regulator_get-simplify-error-returns/20230831-065946 base: https://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator.git for-next patch link: https://lore.kernel.org/r/67b78825385762957b121db2c5d71b119517a7ce.1693431144.git.mirq-linux%40rere.qmqm.pl patch subject: [PATCH 6/9] regulator/core: remove regulator_init callback config: arc-allmodconfig (https://download.01.org/0day-ci/archive/20230901/[email protected]/config) compiler: arceb-elf-gcc (GCC) 13.2.0 reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20230901/[email protected]/reproduce)
If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <[email protected]>
| Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/
All errors (new ones prefixed by >>):
drivers/regulator/core.c: In function 'regulator_register':
>> drivers/regulator/core.c:5609:35: error: 'const struct regulator_init_data' has no member named 'regulator_init'
5609 | if (init_data && init_data->regulator_init) {
| ^~
drivers/regulator/core.c:5610:32: error: 'const struct regulator_init_data' has no member named 'regulator_init'
5610 | ret = init_data->regulator_init(rdev->reg_data);
| ^~vim +5609 drivers/regulator/core.c
d8ca7d184b33af Dmitry Osipenko 2019-06-24 5446
414c70cb91c445 Liam Girdwood 2008-04-30 5447 /**
414c70cb91c445 Liam Girdwood 2008-04-30 5448 * regulator_register - register regulator
8f3cbcd6b44003 ChiYuan Huang 2022-12-06 5449 * @dev: the device that drive the regulator
69279fb9a95051 Mark Brown 2008-12-31 5450 * @regulator_desc: regulator to register
f47531b1aa86e0 Krzysztof Kozlowski 2015-01-12 5451 * @cfg: runtime configuration for regulator
414c70cb91c445 Liam Girdwood 2008-04-30 5452 *
414c70cb91c445 Liam Girdwood 2008-04-30 5453 * Called by regulator drivers to register a regulator.
0384618a79ccfa Axel Lin 2013-01-03 5454 * Returns a valid pointer to struct regulator_dev on success
0384618a79ccfa Axel Lin 2013-01-03 5455 * or an ERR_PTR() on error.
414c70cb91c445 Liam Girdwood 2008-04-30 5456 */
65f26846b90611 Mark Brown 2012-04-03 5457 struct regulator_dev *
8f3cbcd6b44003 ChiYuan Huang 2022-12-06 5458 regulator_register(struct device *dev,
8f3cbcd6b44003 ChiYuan Huang 2022-12-06 5459 const struct regulator_desc *regulator_desc,
1b3de223385d6b Krzysztof Kozlowski 2015-01-05 5460 const struct regulator_config *cfg)
414c70cb91c445 Liam Girdwood 2008-04-30 5461 {
c172708d38a401 Mark Brown 2012-04-04 5462 const struct regulator_init_data *init_data;
1b3de223385d6b Krzysztof Kozlowski 2015-01-05 5463 struct regulator_config *config = NULL;
72dca06f62c504 Aniroop Mathur 2014-12-28 5464 static atomic_t regulator_no = ATOMIC_INIT(-1);
414c70cb91c445 Liam Girdwood 2008-04-30 5465 struct regulator_dev *rdev;
0edb040d416ab3 Linus Walleij 2018-12-06 5466 bool dangling_cfg_gpiod = false;
0edb040d416ab3 Linus Walleij 2018-12-06 5467 bool dangling_of_gpiod = false;
a5766f11cfd3a0 Liam Girdwood 2008-10-10 5468 int ret, i;
520fb178212d1d Christian Kohlschütter 2022-08-29 5469 bool resolved_early = false;
414c70cb91c445 Liam Girdwood 2008-04-30 5470
0edb040d416ab3 Linus Walleij 2018-12-06 5471 if (cfg == NULL)
414c70cb91c445 Liam Girdwood 2008-04-30 5472 return ERR_PTR(-EINVAL);
0edb040d416ab3 Linus Walleij 2018-12-06 5473 if (cfg->ena_gpiod)
0edb040d416ab3 Linus Walleij 2018-12-06 5474 dangling_cfg_gpiod = true;
0edb040d416ab3 Linus Walleij 2018-12-06 5475 if (regulator_desc == NULL) {
0edb040d416ab3 Linus Walleij 2018-12-06 5476 ret = -EINVAL;
0edb040d416ab3 Linus Walleij 2018-12-06 5477 goto rinse;
0edb040d416ab3 Linus Walleij 2018-12-06 5478 }
414c70cb91c445 Liam Girdwood 2008-04-30 5479
8f3cbcd6b44003 ChiYuan Huang 2022-12-06 5480 WARN_ON(!dev || !cfg->dev);
32c8fad438200b Mark Brown 2012-04-11 5481
0edb040d416ab3 Linus Walleij 2018-12-06 5482 if (regulator_desc->name == NULL || regulator_desc->ops == NULL) {
0edb040d416ab3 Linus Walleij 2018-12-06 5483 ret = -EINVAL;
0edb040d416ab3 Linus Walleij 2018-12-06 5484 goto rinse;
0edb040d416ab3 Linus Walleij 2018-12-06 5485 }
414c70cb91c445 Liam Girdwood 2008-04-30 5486
cd78dfc6c6e321 Diego Liziero 2009-04-14 5487 if (regulator_desc->type != REGULATOR_VOLTAGE &&
0edb040d416ab3 Linus Walleij 2018-12-06 5488 regulator_desc->type != REGULATOR_CURRENT) {
0edb040d416ab3 Linus Walleij 2018-12-06 5489 ret = -EINVAL;
0edb040d416ab3 Linus Walleij 2018-12-06 5490 goto rinse;
0edb040d416ab3 Linus Walleij 2018-12-06 5491 }
414c70cb91c445 Liam Girdwood 2008-04-30 5492
476c2d83c7ffb2 Mark Brown 2010-12-10 5493 /* Only one of each should be implemented */
476c2d83c7ffb2 Mark Brown 2010-12-10 5494 WARN_ON(regulator_desc->ops->get_voltage &&
476c2d83c7ffb2 Mark Brown 2010-12-10 5495 regulator_desc->ops->get_voltage_sel);
e8eef82b2c652d Mark Brown 2010-12-12 5496 WARN_ON(regulator_desc->ops->set_voltage &&
e8eef82b2c652d Mark Brown 2010-12-12 5497 regulator_desc->ops->set_voltage_sel);
476c2d83c7ffb2 Mark Brown 2010-12-10 5498
476c2d83c7ffb2 Mark Brown 2010-12-10 5499 /* If we're using selectors we must implement list_voltage. */
476c2d83c7ffb2 Mark Brown 2010-12-10 5500 if (regulator_desc->ops->get_voltage_sel &&
476c2d83c7ffb2 Mark Brown 2010-12-10 5501 !regulator_desc->ops->list_voltage) {
0edb040d416ab3 Linus Walleij 2018-12-06 5502 ret = -EINVAL;
0edb040d416ab3 Linus Walleij 2018-12-06 5503 goto rinse;
476c2d83c7ffb2 Mark Brown 2010-12-10 5504 }
e8eef82b2c652d Mark Brown 2010-12-12 5505 if (regulator_desc->ops->set_voltage_sel &&
e8eef82b2c652d Mark Brown 2010-12-12 5506 !regulator_desc->ops->list_voltage) {
0edb040d416ab3 Linus Walleij 2018-12-06 5507 ret = -EINVAL;
0edb040d416ab3 Linus Walleij 2018-12-06 5508 goto rinse;
e8eef82b2c652d Mark Brown 2010-12-12 5509 }
476c2d83c7ffb2 Mark Brown 2010-12-10 5510
414c70cb91c445 Liam Girdwood 2008-04-30 5511 rdev = kzalloc(sizeof(struct regulator_dev), GFP_KERNEL);
0edb040d416ab3 Linus Walleij 2018-12-06 5512 if (rdev == NULL) {
0edb040d416ab3 Linus Walleij 2018-12-06 5513 ret = -ENOMEM;
0edb040d416ab3 Linus Walleij 2018-12-06 5514 goto rinse;
0edb040d416ab3 Linus Walleij 2018-12-06 5515 }
d3c731564e09b6 Michał Mirosław 2020-08-12 5516 device_initialize(&rdev->dev);
7111c6d1b31b42 Matti Vaittinen 2021-06-03 5517 spin_lock_init(&rdev->err_lock);
414c70cb91c445 Liam Girdwood 2008-04-30 5518
1b3de223385d6b Krzysztof Kozlowski 2015-01-05 5519 /*
1b3de223385d6b Krzysztof Kozlowski 2015-01-05 5520 * Duplicate the config so the driver could override it after
1b3de223385d6b Krzysztof Kozlowski 2015-01-05 5521 * parsing init data.
1b3de223385d6b Krzysztof Kozlowski 2015-01-05 5522 */
1b3de223385d6b Krzysztof Kozlowski 2015-01-05 5523 config = kmemdup(cfg, sizeof(*cfg), GFP_KERNEL);
1b3de223385d6b Krzysztof Kozlowski 2015-01-05 5524 if (config == NULL) {
0edb040d416ab3 Linus Walleij 2018-12-06 5525 ret = -ENOMEM;
d3c731564e09b6 Michał Mirosław 2020-08-12 5526 goto clean;
1b3de223385d6b Krzysztof Kozlowski 2015-01-05 5527 }
1b3de223385d6b Krzysztof Kozlowski 2015-01-05 5528
bfa21a0dfe6915 Krzysztof Kozlowski 2015-01-05 5529 init_data = regulator_of_get_init_data(dev, regulator_desc, config,
a0c7b164ad115e Mark Brown 2014-09-09 5530 &rdev->dev.of_node);
f8970d341eec73 Marco Felsch 2019-09-17 5531
f8970d341eec73 Marco Felsch 2019-09-17 5532 /*
f8970d341eec73 Marco Felsch 2019-09-17 5533 * Sometimes not all resources are probed already so we need to take
f8970d341eec73 Marco Felsch 2019-09-17 5534 * that into account. This happens most the time if the ena_gpiod comes
f8970d341eec73 Marco Felsch 2019-09-17 5535 * from a gpio extender or something else.
f8970d341eec73 Marco Felsch 2019-09-17 5536 */
f8970d341eec73 Marco Felsch 2019-09-17 5537 if (PTR_ERR(init_data) == -EPROBE_DEFER) {
f8970d341eec73 Marco Felsch 2019-09-17 5538 ret = -EPROBE_DEFER;
d3c731564e09b6 Michał Mirosław 2020-08-12 5539 goto clean;
f8970d341eec73 Marco Felsch 2019-09-17 5540 }
f8970d341eec73 Marco Felsch 2019-09-17 5541
0edb040d416ab3 Linus Walleij 2018-12-06 5542 /*
0edb040d416ab3 Linus Walleij 2018-12-06 5543 * We need to keep track of any GPIO descriptor coming from the
0edb040d416ab3 Linus Walleij 2018-12-06 5544 * device tree until we have handled it over to the core. If the
0edb040d416ab3 Linus Walleij 2018-12-06 5545 * config that was passed in to this function DOES NOT contain
0edb040d416ab3 Linus Walleij 2018-12-06 5546 * a descriptor, and the config after this call DOES contain
48f1b4efd67c92 Krzysztof Kozlowski 2019-01-08 5547 * a descriptor, we definitely got one from parsing the device
0edb040d416ab3 Linus Walleij 2018-12-06 5548 * tree.
0edb040d416ab3 Linus Walleij 2018-12-06 5549 */
0edb040d416ab3 Linus Walleij 2018-12-06 5550 if (!cfg->ena_gpiod && config->ena_gpiod)
0edb040d416ab3 Linus Walleij 2018-12-06 5551 dangling_of_gpiod = true;
a0c7b164ad115e Mark Brown 2014-09-09 5552 if (!init_data) {
a0c7b164ad115e Mark Brown 2014-09-09 5553 init_data = config->init_data;
a0c7b164ad115e Mark Brown 2014-09-09 5554 rdev->dev.of_node = of_node_get(config->of_node);
a0c7b164ad115e Mark Brown 2014-09-09 5555 }
a0c7b164ad115e Mark Brown 2014-09-09 5556
f8702f9e4aa7b4 Dmitry Osipenko 2018-11-19 5557 ww_mutex_init(&rdev->mutex, ®ulator_ww_class);
c172708d38a401 Mark Brown 2012-04-04 5558 rdev->reg_data = config->driver_data;
414c70cb91c445 Liam Girdwood 2008-04-30 5559 rdev->owner = regulator_desc->owner;
414c70cb91c445 Liam Girdwood 2008-04-30 5560 rdev->desc = regulator_desc;
3a4b0a07fa69cb Mark Brown 2012-05-08 5561 if (config->regmap)
65b19ce6c22328 Mark Brown 2012-04-15 5562 rdev->regmap = config->regmap;
52b84dac436a68 AnilKumar Ch 2012-09-07 5563 else if (dev_get_regmap(dev, NULL))
3a4b0a07fa69cb Mark Brown 2012-05-08 5564 rdev->regmap = dev_get_regmap(dev, NULL);
52b84dac436a68 AnilKumar Ch 2012-09-07 5565 else if (dev->parent)
52b84dac436a68 AnilKumar Ch 2012-09-07 5566 rdev->regmap = dev_get_regmap(dev->parent, NULL);
414c70cb91c445 Liam Girdwood 2008-04-30 5567 INIT_LIST_HEAD(&rdev->consumer_list);
414c70cb91c445 Liam Girdwood 2008-04-30 5568 INIT_LIST_HEAD(&rdev->list);
414c70cb91c445 Liam Girdwood 2008-04-30 5569 BLOCKING_INIT_NOTIFIER_HEAD(&rdev->notifier);
da07ecd93b1968 Mark Brown 2011-09-11 5570 INIT_DELAYED_WORK(&rdev->disable_work, regulator_disable_work);
414c70cb91c445 Liam Girdwood 2008-04-30 5571
520fb178212d1d Christian Kohlschütter 2022-08-29 5572 if (init_data && init_data->supply_regulator)
520fb178212d1d Christian Kohlschütter 2022-08-29 5573 rdev->supply_name = init_data->supply_regulator;
520fb178212d1d Christian Kohlschütter 2022-08-29 5574 else if (regulator_desc->supply_name)
520fb178212d1d Christian Kohlschütter 2022-08-29 5575 rdev->supply_name = regulator_desc->supply_name;
daad134d66492a Krzysztof Adamski 2016-02-22 5576
a5766f11cfd3a0 Liam Girdwood 2008-10-10 5577 /* register with sysfs */
a5766f11cfd3a0 Liam Girdwood 2008-10-10 5578 rdev->dev.class = ®ulator_class;
0debed5b117d11 ChiYuan Huang 2022-12-14 5579 rdev->dev.parent = config->dev;
72dca06f62c504 Aniroop Mathur 2014-12-28 5580 dev_set_name(&rdev->dev, "regulator.%lu",
39138818a4f5c6 Aniroop Mathur 2014-12-29 5581 (unsigned long) atomic_inc_return(®ulator_no));
9177514ce34902 Vladimir Zapolskiy 2020-07-24 5582 dev_set_drvdata(&rdev->dev, rdev);
414c70cb91c445 Liam Girdwood 2008-04-30 5583
74f544c1fc0339 Mike Rapoport 2008-11-25 5584 /* set regulator constraints */
9a8f5e07200dd8 Mark Brown 2011-11-29 5585 if (init_data)
57a6ad482af256 Michał Mirosław 2020-11-13 5586 rdev->constraints = kmemdup(&init_data->constraints,
57a6ad482af256 Michał Mirosław 2020-11-13 5587 sizeof(*rdev->constraints),
57a6ad482af256 Michał Mirosław 2020-11-13 5588 GFP_KERNEL);
57a6ad482af256 Michał Mirosław 2020-11-13 5589 else
57a6ad482af256 Michał Mirosław 2020-11-13 5590 rdev->constraints = kzalloc(sizeof(*rdev->constraints),
57a6ad482af256 Michał Mirosław 2020-11-13 5591 GFP_KERNEL);
57a6ad482af256 Michał Mirosław 2020-11-13 5592 if (!rdev->constraints) {
57a6ad482af256 Michał Mirosław 2020-11-13 5593 ret = -ENOMEM;
57a6ad482af256 Michał Mirosław 2020-11-13 5594 goto wash;
57a6ad482af256 Michał Mirosław 2020-11-13 5595 }
9a8f5e07200dd8 Mark Brown 2011-11-29 5596
8a866d527ac044 Christian Kohlschütter 2022-08-18 5597 if ((rdev->supply_name && !rdev->supply) &&
8a866d527ac044 Christian Kohlschütter 2022-08-18 5598 (rdev->constraints->always_on ||
8a866d527ac044 Christian Kohlschütter 2022-08-18 5599 rdev->constraints->boot_on)) {
8a866d527ac044 Christian Kohlschütter 2022-08-18 5600 ret = regulator_resolve_supply(rdev);
8a866d527ac044 Christian Kohlschütter 2022-08-18 5601 if (ret)
8a866d527ac044 Christian Kohlschütter 2022-08-18 5602 rdev_dbg(rdev, "unable to resolve supply early: %pe\n",
8a866d527ac044 Christian Kohlschütter 2022-08-18 5603 ERR_PTR(ret));
520fb178212d1d Christian Kohlschütter 2022-08-29 5604
520fb178212d1d Christian Kohlschütter 2022-08-29 5605 resolved_early = true;
8a866d527ac044 Christian Kohlschütter 2022-08-18 5606 }
8a866d527ac044 Christian Kohlschütter 2022-08-18 5607
8a866d527ac044 Christian Kohlschütter 2022-08-18 5608 /* perform any regulator specific init */
9a8f5e07200dd8 Mark Brown 2011-11-29 @5609 if (init_data && init_data->regulator_init) {
a5766f11cfd3a0 Liam Girdwood 2008-10-10 5610 ret = init_data->regulator_init(rdev->reg_data);
4fca9545d17b99 David Brownell 2008-11-11 5611 if (ret < 0)
520fb178212d1d Christian Kohlschütter 2022-08-29 5612 goto wash;
414c70cb91c445 Liam Girdwood 2008-04-30 5613 }
414c70cb91c445 Liam Girdwood 2008-04-30 5614
541d052d721506 Linus Walleij 2019-01-29 5615 if (config->ena_gpiod) {
daad134d66492a Krzysztof Adamski 2016-02-22 5616 ret = regulator_ena_gpio_request(rdev, config);
daad134d66492a Krzysztof Adamski 2016-02-22 5617 if (ret != 0) {
61aab5ad27d551 Michał Mirosław 2020-09-26 5618 rdev_err(rdev, "Failed to request enable GPIO: %pe\n",
61aab5ad27d551 Michał Mirosław 2020-09-26 5619 ERR_PTR(ret));
520fb178212d1d Christian Kohlschütter 2022-08-29 5620 goto wash;
daad134d66492a Krzysztof Adamski 2016-02-22 5621 }
0edb040d416ab3 Linus Walleij 2018-12-06 5622 /* The regulator core took over the GPIO descriptor */
0edb040d416ab3 Linus Walleij 2018-12-06 5623 dangling_cfg_gpiod = false;
0edb040d416ab3 Linus Walleij 2018-12-06 5624 dangling_of_gpiod = false;
daad134d66492a Krzysztof Adamski 2016-02-22 5625 }
0178f3e28e2166 Mark Brown 2010-04-26 5626
57a6ad482af256 Michał Mirosław 2020-11-13 5627 ret = set_machine_constraints(rdev);
520fb178212d1d Christian Kohlschütter 2022-08-29 5628 if (ret == -EPROBE_DEFER && !resolved_early) {
aea6cb99703e17 Michał Mirosław 2020-09-26 5629 /* Regulator might be in bypass mode and so needs its supply
69b8821e293aa8 Shubhankar Kuranagatti 2021-04-20 5630 * to set the constraints
69b8821e293aa8 Shubhankar Kuranagatti 2021-04-20 5631 */
aea6cb99703e17 Michał Mirosław 2020-09-26 5632 /* FIXME: this currently triggers a chicken-and-egg problem
aea6cb99703e17 Michał Mirosław 2020-09-26 5633 * when creating -SUPPLY symlink in sysfs to a regulator
69b8821e293aa8 Shubhankar Kuranagatti 2021-04-20 5634 * that is just being created
69b8821e293aa8 Shubhankar Kuranagatti 2021-04-20 5635 */
0917c9db23accb Michał Mirosław 2020-11-13 5636 rdev_dbg(rdev, "will resolve supply early: %s\n",
0917c9db23accb Michał Mirosław 2020-11-13 5637 rdev->supply_name);
aea6cb99703e17 Michał Mirosław 2020-09-26 5638 ret = regulator_resolve_supply(rdev);
aea6cb99703e17 Michał Mirosław 2020-09-26 5639 if (!ret)
57a6ad482af256 Michał Mirosław 2020-11-13 5640 ret = set_machine_constraints(rdev);
aea6cb99703e17 Michał Mirosław 2020-09-26 5641 else
aea6cb99703e17 Michał Mirosław 2020-09-26 5642 rdev_dbg(rdev, "unable to resolve supply early: %pe\n",
aea6cb99703e17 Michał Mirosław 2020-09-26 5643 ERR_PTR(ret));
aea6cb99703e17 Michał Mirosław 2020-09-26 5644 }
45389c47526d1e Jon Hunter 2016-04-26 5645 if (ret < 0)
45389c47526d1e Jon Hunter 2016-04-26 5646 goto wash;
45389c47526d1e Jon Hunter 2016-04-26 5647
f9503385b1877a Dmitry Osipenko 2018-10-05 5648 ret = regulator_init_coupling(rdev);
f9503385b1877a Dmitry Osipenko 2018-10-05 5649 if (ret < 0)
d3d64537c33956 Maciej Purski 2018-04-23 5650 goto wash;
d3d64537c33956 Maciej Purski 2018-04-23 5651
a5766f11cfd3a0 Liam Girdwood 2008-10-10 5652 /* add consumers devices */
9a8f5e07200dd8 Mark Brown 2011-11-29 5653 if (init_data) {
a5766f11cfd3a0 Liam Girdwood 2008-10-10 5654 for (i = 0; i < init_data->num_consumer_supplies; i++) {
a5766f11cfd3a0 Liam Girdwood 2008-10-10 5655 ret = set_consumer_device_supply(rdev,
40f9244f4da897 Mark Brown 2009-06-17 5656 init_data->consumer_supplies[i].dev_name,
a5766f11cfd3a0 Liam Girdwood 2008-10-10 5657 init_data->consumer_supplies[i].supply);
23c2f041efa891 Mark Brown 2011-02-24 5658 if (ret < 0) {
23c2f041efa891 Mark Brown 2011-02-24 5659 dev_err(dev, "Failed to set supply %s\n",
23c2f041efa891 Mark Brown 2011-02-24 5660 init_data->consumer_supplies[i].supply);
d4033b54fc9122 Jani Nikula 2010-04-29 5661 goto unset_supplies;
414c70cb91c445 Liam Girdwood 2008-04-30 5662 }
23c2f041efa891 Mark Brown 2011-02-24 5663 }
45389c47526d1e Jon Hunter 2016-04-26 5664 }
5e3ca2b349b1e2 Javier Martinez Canillas 2016-03-23 5665
fd086045559d90 Matthias Kaehlcke 2017-03-27 5666 if (!rdev->desc->ops->get_voltage &&
fd086045559d90 Matthias Kaehlcke 2017-03-27 5667 !rdev->desc->ops->list_voltage &&
fd086045559d90 Matthias Kaehlcke 2017-03-27 5668 !rdev->desc->fixed_uV)
fd086045559d90 Matthias Kaehlcke 2017-03-27 5669 rdev->is_switch = true;
fd086045559d90 Matthias Kaehlcke 2017-03-27 5670
9177514ce34902 Vladimir Zapolskiy 2020-07-24 5671 ret = device_add(&rdev->dev);
9177514ce34902 Vladimir Zapolskiy 2020-07-24 5672 if (ret != 0)
c438b9d017362b Jon Hunter 2016-04-21 5673 goto unset_supplies;
414c70cb91c445 Liam Girdwood 2008-04-30 5674
1130e5b3ff4a7f Mark Brown 2010-12-21 5675 rdev_init_debugfs(rdev);
5e3ca2b349b1e2 Javier Martinez Canillas 2016-03-23 5676
f9503385b1877a Dmitry Osipenko 2018-10-05 5677 /* try to resolve regulators coupling since a new one was registered */
f9503385b1877a Dmitry Osipenko 2018-10-05 5678 mutex_lock(®ulator_list_mutex);
f9503385b1877a Dmitry Osipenko 2018-10-05 5679 regulator_resolve_coupling(rdev);
f9503385b1877a Dmitry Osipenko 2018-10-05 5680 mutex_unlock(®ulator_list_mutex);
f9503385b1877a Dmitry Osipenko 2018-10-05 5681
5e3ca2b349b1e2 Javier Martinez Canillas 2016-03-23 5682 /* try to resolve regulators supply since a new one was registered */
5e3ca2b349b1e2 Javier Martinez Canillas 2016-03-23 5683 class_for_each_device(®ulator_class, NULL, NULL,
5e3ca2b349b1e2 Javier Martinez Canillas 2016-03-23 5684 regulator_register_resolve_supply);
1b3de223385d6b Krzysztof Kozlowski 2015-01-05 5685 kfree(config);
a5766f11cfd3a0 Liam Girdwood 2008-10-10 5686 return rdev;
4fca9545d17b99 David Brownell 2008-11-11 5687
d4033b54fc9122 Jani Nikula 2010-04-29 5688 unset_supplies:
45389c47526d1e Jon Hunter 2016-04-26 5689 mutex_lock(®ulator_list_mutex);
d4033b54fc9122 Jani Nikula 2010-04-29 5690 unset_regulator_supplies(rdev);
d8ca7d184b33af Dmitry Osipenko 2019-06-24 5691 regulator_remove_coupling(rdev);
45389c47526d1e Jon Hunter 2016-04-26 5692 mutex_unlock(®ulator_list_mutex);
32165230eb6e62 Krzysztof Adamski 2016-02-24 5693 wash:
ba62319a42c50e Yang Yingliang 2022-12-02 5694 regulator_put(rdev->supply);
26c2c997aa1a6c Dmitry Osipenko 2019-10-25 5695 kfree(rdev->coupling_desc.coupled_rdevs);
45389c47526d1e Jon Hunter 2016-04-26 5696 mutex_lock(®ulator_list_mutex);
32165230eb6e62 Krzysztof Adamski 2016-02-24 5697 regulator_ena_gpio_free(rdev);
45389c47526d1e Jon Hunter 2016-04-26 5698 mutex_unlock(®ulator_list_mutex);
5f4b204b6b8153 Zeng Heng 2022-11-16 5699 put_device(&rdev->dev);
5f4b204b6b8153 Zeng Heng 2022-11-16 5700 rdev = NULL;
4fca9545d17b99 David Brownell 2008-11-11 5701 clean:
0edb040d416ab3 Linus Walleij 2018-12-06 5702 if (dangling_of_gpiod)
0edb040d416ab3 Linus Walleij 2018-12-06 5703 gpiod_put(config->ena_gpiod);
5f4b204b6b8153 Zeng Heng 2022-11-16 5704 if (rdev && rdev->dev.of_node)
5f4b204b6b8153 Zeng Heng 2022-11-16 5705 of_node_put(rdev->dev.of_node);
5f4b204b6b8153 Zeng Heng 2022-11-16 5706 kfree(rdev);
a2151374230820 Jon Hunter 2016-03-30 5707 kfree(config);
0edb040d416ab3 Linus Walleij 2018-12-06 5708 rinse:
0edb040d416ab3 Linus Walleij 2018-12-06 5709 if (dangling_cfg_gpiod)
0edb040d416ab3 Linus Walleij 2018-12-06 5710 gpiod_put(cfg->ena_gpiod);
a2151374230820 Jon Hunter 2016-03-30 5711 return ERR_PTR(ret);
414c70cb91c445 Liam Girdwood 2008-04-30 5712 }
a5766f11cfd3a0 Liam Girdwood 2008-10-10 5713 EXPORT_SYMBOL_GPL(regulator_register);
414c70cb91c445 Liam Girdwood 2008-04-30 5714 Signature · 2 lines
0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki
PATCH 7/9 regulator/core: remove regulator_get_init_drvdata()
In reply to Michał MirosławThere are no in-tree users.
--- drivers/regulator/core.c | 6 ------ include/linux/regulator/driver.h | 1 - 2 files changed, 7 deletions(-)
drivers/regulator/core.c +0 -6
@@ -5881,12 +5881,6 @@ struct regmap *rdev_get_regmap(struct regulator_dev *rdev) }EXPORT_SYMBOL_GPL (rdev_get_regmap); -void *regulator_get_init_drvdata (struct regulator_init_data *reg_init_data) -{ -return reg_init_data->driver_data; -} -EXPORT_SYMBOL_GPL (regulator_get_init_drvdata); -#ifdef CONFIG_DEBUG_FS static int supply_map_show (struct seq_file *sf,void *data) {
include/linux/regulator/driver.h +0 -1
@@ -758,7 +758,6 @@ int regulator_set_active_discharge_regmap(struct regulator_dev *rdev,int regulator_set_current_limit_regmap (struct regulator_dev *rdev,int min_uA,int max_uA);int regulator_get_current_limit_regmap (struct regulator_dev *rdev); -void *regulator_get_init_drvdata (struct regulator_init_data *reg_init_data);int regulator_find_closest_bigger (unsigned int target,const unsigned int *table,unsigned int num_sel,unsigned int *sel);int regulator_set_ramp_delay_regmap (struct regulator_dev *rdev,int ramp_delay);
Signature · 1 line
2.39.2
PATCH 8/9 regulator/core: set_consumer_device_supply: avoid copying const data
In reply to Michał MirosławAs consumer_dev_name might as well be const, don't copy it if not required.
--- drivers/regulator/core.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/regulator/core.c +3 -3
@@ -1735,7 +1735,7 @@ static int set_consumer_device_supply(struct regulator_dev *rdev, new_node->supply = supply;if (consumer_dev_name !=NULL ) { - new_node->dev_name =kstrdup (consumer_dev_name,GFP_KERNEL ); + new_node->dev_name =kstrdup_const (consumer_dev_name,GFP_KERNEL );if (new_node->dev_name ==NULL ) {kfree (new_node);return -ENOMEM ;
@@ -1770,7 +1770,7 @@ static int set_consumer_device_supply(struct regulator_dev *rdev, fail:mutex_unlock (®ulator_list_mutex); -kfree (new_node->dev_name); +kfree_const (new_node->dev_name);kfree (new_node);return -EBUSY ; }
@@ -1782,7 +1782,7 @@ static void unset_regulator_supplies(struct regulator_dev *rdev)list_for_each_entry_safe (node, n, ®ulator_map_list, list) {if (rdev == node->regulator) {list_del (&node->list); -kfree (node->dev_name); +kfree_const (node->dev_name);kfree (node); } }
Signature · 1 line
2.39.2
PATCH 9/9 regulator/core: make regulator_class const
In reply to Michał MirosławThe struct class is passed only to class_register() and a const pointer there. Make the data also const.
--- drivers/regulator/core.c | 3 ++- drivers/regulator/internal.h | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-)
drivers/regulator/core.c +2 -1
@@ -5797,7 +5797,7 @@ static const struct dev_pm_ops __maybe_unused regulator_pm_ops = { };#endif -struct class regulator_class = { +const struct class regulator_class = { .name ="regulator" , .dev_release = regulator_dev_release, .dev_groups = regulator_dev_groups,
@@ -5805,6 +5805,7 @@ struct class regulator_class = { .pm = ®ulator_pm_ops,#endif }; +/** * regulator_has_full_constraints - the system has fully specified constraints *
drivers/regulator/internal.h +1 -1
@@ -58,7 +58,7 @@ struct regulator {struct dentry *debugfs; }; -extern struct class regulator_class; +extern const struct class regulator_class;static inline struct regulator_dev *dev_to_rdev (struct device *dev) {
Signature · 1 line
2.39.2