* [PATCH v2] leds: introduce ordered workqueue for leds events instead of system_wq
@ 2024-09-03 22:39 Dmitry Rokosov
2024-09-16 11:18 ` Dmitry Rokosov
2024-11-06 8:11 ` (subset) " Lee Jones
0 siblings, 2 replies; 8+ messages in thread
From: Dmitry Rokosov @ 2024-09-03 22:39 UTC (permalink / raw)
To: pavel, lee
Cc: linux-kernel, linux-leds, kernel, rockosov, Dmitry Rokosov,
Alexey Romanov
This allows to setup ordered workqueue for leds events. This may be
useful, because default 'system_wq' does not guarantee execution order
of each work_struct, thus for several brightness update requests (for
multiple leds), real brightness switch could be in random order.
Yes, for sysfs-based leds we have flush_work() call inside
brightness_store() operation, but it's blocking call, so userspace
caller can be blocked at a long time, which means leds animation stream
can be broken.
Ordered workqueue has the same behaviour as system_wq + flush_work(),
but all scheduled works are async and userspace caller is not blocked,
which it better for userspace animation scheduling.
Signed-off-by: Alexey Romanov <avromanov@salutedevices.com>
Signed-off-by: Dmitry Rokosov <ddrokosov@salutedevices.com>
---
Changes v2 since v1 at [1]:
- replace "leds" with "LEDs" in the log message
Links:
[1] https://lore.kernel.org/all/20240820155407.32729-1-ddrokosov@salutedevices.com/
---
drivers/leds/led-class.c | 12 +++++++++++-
drivers/leds/led-core.c | 6 +++---
include/linux/leds.h | 1 +
3 files changed, 15 insertions(+), 4 deletions(-)
diff --git a/drivers/leds/led-class.c b/drivers/leds/led-class.c
index c66d1bead0a4..b5e28ad54f7f 100644
--- a/drivers/leds/led-class.c
+++ b/drivers/leds/led-class.c
@@ -25,6 +25,8 @@
static DEFINE_MUTEX(leds_lookup_lock);
static LIST_HEAD(leds_lookup_list);
+static struct workqueue_struct *leds_wq;
+
static ssize_t brightness_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
@@ -57,7 +59,6 @@ static ssize_t brightness_store(struct device *dev,
if (state == LED_OFF)
led_trigger_remove(led_cdev);
led_set_brightness(led_cdev, state);
- flush_work(&led_cdev->set_brightness_work);
ret = size;
unlock:
@@ -548,6 +549,8 @@ int led_classdev_register_ext(struct device *parent,
led_update_brightness(led_cdev);
+ led_cdev->wq = leds_wq;
+
led_init_core(led_cdev);
#ifdef CONFIG_LEDS_TRIGGERS
@@ -666,12 +669,19 @@ EXPORT_SYMBOL_GPL(devm_led_classdev_unregister);
static int __init leds_init(void)
{
+ leds_wq = alloc_ordered_workqueue("leds", 0);
+ if (!leds_wq) {
+ pr_err("failed to create LEDs ordered workqueue\n");
+ return -ENOMEM;
+ }
+
return class_register(&leds_class);
}
static void __exit leds_exit(void)
{
class_unregister(&leds_class);
+ destroy_workqueue(leds_wq);
}
subsys_initcall(leds_init);
diff --git a/drivers/leds/led-core.c b/drivers/leds/led-core.c
index 89c9806cc97f..9769ac49be20 100644
--- a/drivers/leds/led-core.c
+++ b/drivers/leds/led-core.c
@@ -266,7 +266,7 @@ void led_blink_set_nosleep(struct led_classdev *led_cdev, unsigned long delay_on
led_cdev->delayed_delay_on = delay_on;
led_cdev->delayed_delay_off = delay_off;
set_bit(LED_SET_BLINK, &led_cdev->work_flags);
- schedule_work(&led_cdev->set_brightness_work);
+ queue_work(led_cdev->wq, &led_cdev->set_brightness_work);
return;
}
@@ -297,7 +297,7 @@ void led_set_brightness(struct led_classdev *led_cdev, unsigned int brightness)
*/
if (!brightness) {
set_bit(LED_BLINK_DISABLE, &led_cdev->work_flags);
- schedule_work(&led_cdev->set_brightness_work);
+ queue_work(led_cdev->wq, &led_cdev->set_brightness_work);
} else {
set_bit(LED_BLINK_BRIGHTNESS_CHANGE,
&led_cdev->work_flags);
@@ -333,7 +333,7 @@ void led_set_brightness_nopm(struct led_classdev *led_cdev, unsigned int value)
set_bit(LED_SET_BRIGHTNESS_OFF, &led_cdev->work_flags);
}
- schedule_work(&led_cdev->set_brightness_work);
+ queue_work(led_cdev->wq, &led_cdev->set_brightness_work);
}
EXPORT_SYMBOL_GPL(led_set_brightness_nopm);
diff --git a/include/linux/leds.h b/include/linux/leds.h
index 6300313c46b7..7c9f1cb12ab9 100644
--- a/include/linux/leds.h
+++ b/include/linux/leds.h
@@ -169,6 +169,7 @@ struct led_classdev {
int new_blink_brightness;
void (*flash_resume)(struct led_classdev *led_cdev);
+ struct workqueue_struct *wq; /* LED workqueue */
struct work_struct set_brightness_work;
int delayed_set_value;
unsigned long delayed_delay_on;
--
2.43.0
^ permalink raw reply related [flat|nested] 8+ messages in thread* Re: [PATCH v2] leds: introduce ordered workqueue for leds events instead of system_wq 2024-09-03 22:39 [PATCH v2] leds: introduce ordered workqueue for leds events instead of system_wq Dmitry Rokosov @ 2024-09-16 11:18 ` Dmitry Rokosov 2024-09-17 8:04 ` Lee Jones 2024-11-06 8:11 ` (subset) " Lee Jones 1 sibling, 1 reply; 8+ messages in thread From: Dmitry Rokosov @ 2024-09-16 11:18 UTC (permalink / raw) To: pavel, lee; +Cc: linux-kernel, linux-leds, kernel, rockosov, Alexey Romanov Hello Lee! What are the next steps? Should I make any changes, or are we waiting for test results from the mailing list members? Sorry for the ping. On Wed, Sep 04, 2024 at 01:39:30AM +0300, Dmitry Rokosov wrote: > This allows to setup ordered workqueue for leds events. This may be > useful, because default 'system_wq' does not guarantee execution order > of each work_struct, thus for several brightness update requests (for > multiple leds), real brightness switch could be in random order. > > Yes, for sysfs-based leds we have flush_work() call inside > brightness_store() operation, but it's blocking call, so userspace > caller can be blocked at a long time, which means leds animation stream > can be broken. > > Ordered workqueue has the same behaviour as system_wq + flush_work(), > but all scheduled works are async and userspace caller is not blocked, > which it better for userspace animation scheduling. > > Signed-off-by: Alexey Romanov <avromanov@salutedevices.com> > Signed-off-by: Dmitry Rokosov <ddrokosov@salutedevices.com> > --- > Changes v2 since v1 at [1]: > - replace "leds" with "LEDs" in the log message > > Links: > [1] https://lore.kernel.org/all/20240820155407.32729-1-ddrokosov@salutedevices.com/ > --- > drivers/leds/led-class.c | 12 +++++++++++- > drivers/leds/led-core.c | 6 +++--- > include/linux/leds.h | 1 + > 3 files changed, 15 insertions(+), 4 deletions(-) > > diff --git a/drivers/leds/led-class.c b/drivers/leds/led-class.c > index c66d1bead0a4..b5e28ad54f7f 100644 > --- a/drivers/leds/led-class.c > +++ b/drivers/leds/led-class.c > @@ -25,6 +25,8 @@ > static DEFINE_MUTEX(leds_lookup_lock); > static LIST_HEAD(leds_lookup_list); > > +static struct workqueue_struct *leds_wq; > + > static ssize_t brightness_show(struct device *dev, > struct device_attribute *attr, char *buf) > { > @@ -57,7 +59,6 @@ static ssize_t brightness_store(struct device *dev, > if (state == LED_OFF) > led_trigger_remove(led_cdev); > led_set_brightness(led_cdev, state); > - flush_work(&led_cdev->set_brightness_work); > > ret = size; > unlock: > @@ -548,6 +549,8 @@ int led_classdev_register_ext(struct device *parent, > > led_update_brightness(led_cdev); > > + led_cdev->wq = leds_wq; > + > led_init_core(led_cdev); > > #ifdef CONFIG_LEDS_TRIGGERS > @@ -666,12 +669,19 @@ EXPORT_SYMBOL_GPL(devm_led_classdev_unregister); > > static int __init leds_init(void) > { > + leds_wq = alloc_ordered_workqueue("leds", 0); > + if (!leds_wq) { > + pr_err("failed to create LEDs ordered workqueue\n"); > + return -ENOMEM; > + } > + > return class_register(&leds_class); > } > > static void __exit leds_exit(void) > { > class_unregister(&leds_class); > + destroy_workqueue(leds_wq); > } > > subsys_initcall(leds_init); > diff --git a/drivers/leds/led-core.c b/drivers/leds/led-core.c > index 89c9806cc97f..9769ac49be20 100644 > --- a/drivers/leds/led-core.c > +++ b/drivers/leds/led-core.c > @@ -266,7 +266,7 @@ void led_blink_set_nosleep(struct led_classdev *led_cdev, unsigned long delay_on > led_cdev->delayed_delay_on = delay_on; > led_cdev->delayed_delay_off = delay_off; > set_bit(LED_SET_BLINK, &led_cdev->work_flags); > - schedule_work(&led_cdev->set_brightness_work); > + queue_work(led_cdev->wq, &led_cdev->set_brightness_work); > return; > } > > @@ -297,7 +297,7 @@ void led_set_brightness(struct led_classdev *led_cdev, unsigned int brightness) > */ > if (!brightness) { > set_bit(LED_BLINK_DISABLE, &led_cdev->work_flags); > - schedule_work(&led_cdev->set_brightness_work); > + queue_work(led_cdev->wq, &led_cdev->set_brightness_work); > } else { > set_bit(LED_BLINK_BRIGHTNESS_CHANGE, > &led_cdev->work_flags); > @@ -333,7 +333,7 @@ void led_set_brightness_nopm(struct led_classdev *led_cdev, unsigned int value) > set_bit(LED_SET_BRIGHTNESS_OFF, &led_cdev->work_flags); > } > > - schedule_work(&led_cdev->set_brightness_work); > + queue_work(led_cdev->wq, &led_cdev->set_brightness_work); > } > EXPORT_SYMBOL_GPL(led_set_brightness_nopm); > > diff --git a/include/linux/leds.h b/include/linux/leds.h > index 6300313c46b7..7c9f1cb12ab9 100644 > --- a/include/linux/leds.h > +++ b/include/linux/leds.h > @@ -169,6 +169,7 @@ struct led_classdev { > int new_blink_brightness; > void (*flash_resume)(struct led_classdev *led_cdev); > > + struct workqueue_struct *wq; /* LED workqueue */ > struct work_struct set_brightness_work; > int delayed_set_value; > unsigned long delayed_delay_on; > -- > 2.43.0 > -- Thank you, Dmitry ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH v2] leds: introduce ordered workqueue for leds events instead of system_wq 2024-09-16 11:18 ` Dmitry Rokosov @ 2024-09-17 8:04 ` Lee Jones 2024-10-22 14:22 ` Dmitry Rokosov 0 siblings, 1 reply; 8+ messages in thread From: Lee Jones @ 2024-09-17 8:04 UTC (permalink / raw) To: Dmitry Rokosov Cc: pavel, linux-kernel, linux-leds, kernel, rockosov, Alexey Romanov On Mon, 16 Sep 2024, Dmitry Rokosov wrote: > Hello Lee! > > What are the next steps? Should I make any changes, or are we waiting > for test results from the mailing list members? This is an intrusive core change that was submitted during -rc6. It's going to need some time on the list for people to respond. > Sorry for the ping. > > On Wed, Sep 04, 2024 at 01:39:30AM +0300, Dmitry Rokosov wrote: > > This allows to setup ordered workqueue for leds events. This may be > > useful, because default 'system_wq' does not guarantee execution order > > of each work_struct, thus for several brightness update requests (for > > multiple leds), real brightness switch could be in random order. > > > > Yes, for sysfs-based leds we have flush_work() call inside > > brightness_store() operation, but it's blocking call, so userspace > > caller can be blocked at a long time, which means leds animation stream > > can be broken. > > > > Ordered workqueue has the same behaviour as system_wq + flush_work(), > > but all scheduled works are async and userspace caller is not blocked, > > which it better for userspace animation scheduling. > > > > Signed-off-by: Alexey Romanov <avromanov@salutedevices.com> > > Signed-off-by: Dmitry Rokosov <ddrokosov@salutedevices.com> > > --- > > Changes v2 since v1 at [1]: > > - replace "leds" with "LEDs" in the log message > > > > Links: > > [1] https://lore.kernel.org/all/20240820155407.32729-1-ddrokosov@salutedevices.com/ > > --- > > drivers/leds/led-class.c | 12 +++++++++++- > > drivers/leds/led-core.c | 6 +++--- > > include/linux/leds.h | 1 + > > 3 files changed, 15 insertions(+), 4 deletions(-) > > > > diff --git a/drivers/leds/led-class.c b/drivers/leds/led-class.c > > index c66d1bead0a4..b5e28ad54f7f 100644 > > --- a/drivers/leds/led-class.c > > +++ b/drivers/leds/led-class.c > > @@ -25,6 +25,8 @@ > > static DEFINE_MUTEX(leds_lookup_lock); > > static LIST_HEAD(leds_lookup_list); > > > > +static struct workqueue_struct *leds_wq; > > + > > static ssize_t brightness_show(struct device *dev, > > struct device_attribute *attr, char *buf) > > { > > @@ -57,7 +59,6 @@ static ssize_t brightness_store(struct device *dev, > > if (state == LED_OFF) > > led_trigger_remove(led_cdev); > > led_set_brightness(led_cdev, state); > > - flush_work(&led_cdev->set_brightness_work); > > > > ret = size; > > unlock: > > @@ -548,6 +549,8 @@ int led_classdev_register_ext(struct device *parent, > > > > led_update_brightness(led_cdev); > > > > + led_cdev->wq = leds_wq; > > + > > led_init_core(led_cdev); > > > > #ifdef CONFIG_LEDS_TRIGGERS > > @@ -666,12 +669,19 @@ EXPORT_SYMBOL_GPL(devm_led_classdev_unregister); > > > > static int __init leds_init(void) > > { > > + leds_wq = alloc_ordered_workqueue("leds", 0); > > + if (!leds_wq) { > > + pr_err("failed to create LEDs ordered workqueue\n"); > > + return -ENOMEM; > > + } > > + > > return class_register(&leds_class); > > } > > > > static void __exit leds_exit(void) > > { > > class_unregister(&leds_class); > > + destroy_workqueue(leds_wq); > > } > > > > subsys_initcall(leds_init); > > diff --git a/drivers/leds/led-core.c b/drivers/leds/led-core.c > > index 89c9806cc97f..9769ac49be20 100644 > > --- a/drivers/leds/led-core.c > > +++ b/drivers/leds/led-core.c > > @@ -266,7 +266,7 @@ void led_blink_set_nosleep(struct led_classdev *led_cdev, unsigned long delay_on > > led_cdev->delayed_delay_on = delay_on; > > led_cdev->delayed_delay_off = delay_off; > > set_bit(LED_SET_BLINK, &led_cdev->work_flags); > > - schedule_work(&led_cdev->set_brightness_work); > > + queue_work(led_cdev->wq, &led_cdev->set_brightness_work); > > return; > > } > > > > @@ -297,7 +297,7 @@ void led_set_brightness(struct led_classdev *led_cdev, unsigned int brightness) > > */ > > if (!brightness) { > > set_bit(LED_BLINK_DISABLE, &led_cdev->work_flags); > > - schedule_work(&led_cdev->set_brightness_work); > > + queue_work(led_cdev->wq, &led_cdev->set_brightness_work); > > } else { > > set_bit(LED_BLINK_BRIGHTNESS_CHANGE, > > &led_cdev->work_flags); > > @@ -333,7 +333,7 @@ void led_set_brightness_nopm(struct led_classdev *led_cdev, unsigned int value) > > set_bit(LED_SET_BRIGHTNESS_OFF, &led_cdev->work_flags); > > } > > > > - schedule_work(&led_cdev->set_brightness_work); > > + queue_work(led_cdev->wq, &led_cdev->set_brightness_work); > > } > > EXPORT_SYMBOL_GPL(led_set_brightness_nopm); > > > > diff --git a/include/linux/leds.h b/include/linux/leds.h > > index 6300313c46b7..7c9f1cb12ab9 100644 > > --- a/include/linux/leds.h > > +++ b/include/linux/leds.h > > @@ -169,6 +169,7 @@ struct led_classdev { > > int new_blink_brightness; > > void (*flash_resume)(struct led_classdev *led_cdev); > > > > + struct workqueue_struct *wq; /* LED workqueue */ > > struct work_struct set_brightness_work; > > int delayed_set_value; > > unsigned long delayed_delay_on; > > -- > > 2.43.0 > > > > -- > Thank you, > Dmitry -- Lee Jones [李琼斯] ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH v2] leds: introduce ordered workqueue for leds events instead of system_wq 2024-09-17 8:04 ` Lee Jones @ 2024-10-22 14:22 ` Dmitry Rokosov 2024-10-29 20:06 ` Dmitry Rokosov 0 siblings, 1 reply; 8+ messages in thread From: Dmitry Rokosov @ 2024-10-22 14:22 UTC (permalink / raw) To: Lee Jones Cc: pavel, linux-kernel, linux-leds, kernel, rockosov, Alexey Romanov Hello Lee, I hope this message finds you well. I wanted to ask for your thoughts on whether there was enough time for testing and if we can proceed with merging this patchset. I appreciate any feedback about it. On Tue, Sep 17, 2024 at 09:04:12AM +0100, Lee Jones wrote: > On Mon, 16 Sep 2024, Dmitry Rokosov wrote: > > > Hello Lee! > > > > What are the next steps? Should I make any changes, or are we waiting > > for test results from the mailing list members? > > This is an intrusive core change that was submitted during -rc6. > > It's going to need some time on the list for people to respond. > > > Sorry for the ping. > > > > On Wed, Sep 04, 2024 at 01:39:30AM +0300, Dmitry Rokosov wrote: > > > This allows to setup ordered workqueue for leds events. This may be > > > useful, because default 'system_wq' does not guarantee execution order > > > of each work_struct, thus for several brightness update requests (for > > > multiple leds), real brightness switch could be in random order. > > > > > > Yes, for sysfs-based leds we have flush_work() call inside > > > brightness_store() operation, but it's blocking call, so userspace > > > caller can be blocked at a long time, which means leds animation stream > > > can be broken. > > > > > > Ordered workqueue has the same behaviour as system_wq + flush_work(), > > > but all scheduled works are async and userspace caller is not blocked, > > > which it better for userspace animation scheduling. > > > > > > Signed-off-by: Alexey Romanov <avromanov@salutedevices.com> > > > Signed-off-by: Dmitry Rokosov <ddrokosov@salutedevices.com> > > > --- > > > Changes v2 since v1 at [1]: > > > - replace "leds" with "LEDs" in the log message > > > > > > Links: > > > [1] https://lore.kernel.org/all/20240820155407.32729-1-ddrokosov@salutedevices.com/ > > > --- > > > drivers/leds/led-class.c | 12 +++++++++++- > > > drivers/leds/led-core.c | 6 +++--- > > > include/linux/leds.h | 1 + > > > 3 files changed, 15 insertions(+), 4 deletions(-) > > > > > > diff --git a/drivers/leds/led-class.c b/drivers/leds/led-class.c > > > index c66d1bead0a4..b5e28ad54f7f 100644 > > > --- a/drivers/leds/led-class.c > > > +++ b/drivers/leds/led-class.c > > > @@ -25,6 +25,8 @@ > > > static DEFINE_MUTEX(leds_lookup_lock); > > > static LIST_HEAD(leds_lookup_list); > > > > > > +static struct workqueue_struct *leds_wq; > > > + > > > static ssize_t brightness_show(struct device *dev, > > > struct device_attribute *attr, char *buf) > > > { > > > @@ -57,7 +59,6 @@ static ssize_t brightness_store(struct device *dev, > > > if (state == LED_OFF) > > > led_trigger_remove(led_cdev); > > > led_set_brightness(led_cdev, state); > > > - flush_work(&led_cdev->set_brightness_work); > > > > > > ret = size; > > > unlock: > > > @@ -548,6 +549,8 @@ int led_classdev_register_ext(struct device *parent, > > > > > > led_update_brightness(led_cdev); > > > > > > + led_cdev->wq = leds_wq; > > > + > > > led_init_core(led_cdev); > > > > > > #ifdef CONFIG_LEDS_TRIGGERS > > > @@ -666,12 +669,19 @@ EXPORT_SYMBOL_GPL(devm_led_classdev_unregister); > > > > > > static int __init leds_init(void) > > > { > > > + leds_wq = alloc_ordered_workqueue("leds", 0); > > > + if (!leds_wq) { > > > + pr_err("failed to create LEDs ordered workqueue\n"); > > > + return -ENOMEM; > > > + } > > > + > > > return class_register(&leds_class); > > > } > > > > > > static void __exit leds_exit(void) > > > { > > > class_unregister(&leds_class); > > > + destroy_workqueue(leds_wq); > > > } > > > > > > subsys_initcall(leds_init); > > > diff --git a/drivers/leds/led-core.c b/drivers/leds/led-core.c > > > index 89c9806cc97f..9769ac49be20 100644 > > > --- a/drivers/leds/led-core.c > > > +++ b/drivers/leds/led-core.c > > > @@ -266,7 +266,7 @@ void led_blink_set_nosleep(struct led_classdev *led_cdev, unsigned long delay_on > > > led_cdev->delayed_delay_on = delay_on; > > > led_cdev->delayed_delay_off = delay_off; > > > set_bit(LED_SET_BLINK, &led_cdev->work_flags); > > > - schedule_work(&led_cdev->set_brightness_work); > > > + queue_work(led_cdev->wq, &led_cdev->set_brightness_work); > > > return; > > > } > > > > > > @@ -297,7 +297,7 @@ void led_set_brightness(struct led_classdev *led_cdev, unsigned int brightness) > > > */ > > > if (!brightness) { > > > set_bit(LED_BLINK_DISABLE, &led_cdev->work_flags); > > > - schedule_work(&led_cdev->set_brightness_work); > > > + queue_work(led_cdev->wq, &led_cdev->set_brightness_work); > > > } else { > > > set_bit(LED_BLINK_BRIGHTNESS_CHANGE, > > > &led_cdev->work_flags); > > > @@ -333,7 +333,7 @@ void led_set_brightness_nopm(struct led_classdev *led_cdev, unsigned int value) > > > set_bit(LED_SET_BRIGHTNESS_OFF, &led_cdev->work_flags); > > > } > > > > > > - schedule_work(&led_cdev->set_brightness_work); > > > + queue_work(led_cdev->wq, &led_cdev->set_brightness_work); > > > } > > > EXPORT_SYMBOL_GPL(led_set_brightness_nopm); > > > > > > diff --git a/include/linux/leds.h b/include/linux/leds.h > > > index 6300313c46b7..7c9f1cb12ab9 100644 > > > --- a/include/linux/leds.h > > > +++ b/include/linux/leds.h > > > @@ -169,6 +169,7 @@ struct led_classdev { > > > int new_blink_brightness; > > > void (*flash_resume)(struct led_classdev *led_cdev); > > > > > > + struct workqueue_struct *wq; /* LED workqueue */ > > > struct work_struct set_brightness_work; > > > int delayed_set_value; > > > unsigned long delayed_delay_on; > > > -- > > > 2.43.0 > > > > > > > -- > > Thank you, > > Dmitry > > -- > Lee Jones [李琼斯] -- Thank you, Dmitry ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH v2] leds: introduce ordered workqueue for leds events instead of system_wq 2024-10-22 14:22 ` Dmitry Rokosov @ 2024-10-29 20:06 ` Dmitry Rokosov 2024-10-31 10:28 ` Lee Jones 0 siblings, 1 reply; 8+ messages in thread From: Dmitry Rokosov @ 2024-10-29 20:06 UTC (permalink / raw) To: Lee Jones Cc: pavel, linux-kernel, linux-leds, kernel, rockosov, Alexey Romanov Hello Lee, Kindly reminder. On Tue, Oct 22, 2024 at 05:22:48PM +0300, Dmitry Rokosov wrote: > Hello Lee, > > I hope this message finds you well. > I wanted to ask for your thoughts on whether there was enough time for > testing and if we can proceed with merging this patchset. > > I appreciate any feedback about it. > > On Tue, Sep 17, 2024 at 09:04:12AM +0100, Lee Jones wrote: > > On Mon, 16 Sep 2024, Dmitry Rokosov wrote: > > > > > Hello Lee! > > > > > > What are the next steps? Should I make any changes, or are we waiting > > > for test results from the mailing list members? > > > > This is an intrusive core change that was submitted during -rc6. > > > > It's going to need some time on the list for people to respond. > > > > > Sorry for the ping. > > > > > > On Wed, Sep 04, 2024 at 01:39:30AM +0300, Dmitry Rokosov wrote: > > > > This allows to setup ordered workqueue for leds events. This may be > > > > useful, because default 'system_wq' does not guarantee execution order > > > > of each work_struct, thus for several brightness update requests (for > > > > multiple leds), real brightness switch could be in random order. > > > > > > > > Yes, for sysfs-based leds we have flush_work() call inside > > > > brightness_store() operation, but it's blocking call, so userspace > > > > caller can be blocked at a long time, which means leds animation stream > > > > can be broken. > > > > > > > > Ordered workqueue has the same behaviour as system_wq + flush_work(), > > > > but all scheduled works are async and userspace caller is not blocked, > > > > which it better for userspace animation scheduling. > > > > > > > > Signed-off-by: Alexey Romanov <avromanov@salutedevices.com> > > > > Signed-off-by: Dmitry Rokosov <ddrokosov@salutedevices.com> > > > > --- > > > > Changes v2 since v1 at [1]: > > > > - replace "leds" with "LEDs" in the log message > > > > > > > > Links: > > > > [1] https://lore.kernel.org/all/20240820155407.32729-1-ddrokosov@salutedevices.com/ > > > > --- > > > > drivers/leds/led-class.c | 12 +++++++++++- > > > > drivers/leds/led-core.c | 6 +++--- > > > > include/linux/leds.h | 1 + > > > > 3 files changed, 15 insertions(+), 4 deletions(-) > > > > > > > > diff --git a/drivers/leds/led-class.c b/drivers/leds/led-class.c > > > > index c66d1bead0a4..b5e28ad54f7f 100644 > > > > --- a/drivers/leds/led-class.c > > > > +++ b/drivers/leds/led-class.c > > > > @@ -25,6 +25,8 @@ > > > > static DEFINE_MUTEX(leds_lookup_lock); > > > > static LIST_HEAD(leds_lookup_list); > > > > > > > > +static struct workqueue_struct *leds_wq; > > > > + > > > > static ssize_t brightness_show(struct device *dev, > > > > struct device_attribute *attr, char *buf) > > > > { > > > > @@ -57,7 +59,6 @@ static ssize_t brightness_store(struct device *dev, > > > > if (state == LED_OFF) > > > > led_trigger_remove(led_cdev); > > > > led_set_brightness(led_cdev, state); > > > > - flush_work(&led_cdev->set_brightness_work); > > > > > > > > ret = size; > > > > unlock: > > > > @@ -548,6 +549,8 @@ int led_classdev_register_ext(struct device *parent, > > > > > > > > led_update_brightness(led_cdev); > > > > > > > > + led_cdev->wq = leds_wq; > > > > + > > > > led_init_core(led_cdev); > > > > > > > > #ifdef CONFIG_LEDS_TRIGGERS > > > > @@ -666,12 +669,19 @@ EXPORT_SYMBOL_GPL(devm_led_classdev_unregister); > > > > > > > > static int __init leds_init(void) > > > > { > > > > + leds_wq = alloc_ordered_workqueue("leds", 0); > > > > + if (!leds_wq) { > > > > + pr_err("failed to create LEDs ordered workqueue\n"); > > > > + return -ENOMEM; > > > > + } > > > > + > > > > return class_register(&leds_class); > > > > } > > > > > > > > static void __exit leds_exit(void) > > > > { > > > > class_unregister(&leds_class); > > > > + destroy_workqueue(leds_wq); > > > > } > > > > > > > > subsys_initcall(leds_init); > > > > diff --git a/drivers/leds/led-core.c b/drivers/leds/led-core.c > > > > index 89c9806cc97f..9769ac49be20 100644 > > > > --- a/drivers/leds/led-core.c > > > > +++ b/drivers/leds/led-core.c > > > > @@ -266,7 +266,7 @@ void led_blink_set_nosleep(struct led_classdev *led_cdev, unsigned long delay_on > > > > led_cdev->delayed_delay_on = delay_on; > > > > led_cdev->delayed_delay_off = delay_off; > > > > set_bit(LED_SET_BLINK, &led_cdev->work_flags); > > > > - schedule_work(&led_cdev->set_brightness_work); > > > > + queue_work(led_cdev->wq, &led_cdev->set_brightness_work); > > > > return; > > > > } > > > > > > > > @@ -297,7 +297,7 @@ void led_set_brightness(struct led_classdev *led_cdev, unsigned int brightness) > > > > */ > > > > if (!brightness) { > > > > set_bit(LED_BLINK_DISABLE, &led_cdev->work_flags); > > > > - schedule_work(&led_cdev->set_brightness_work); > > > > + queue_work(led_cdev->wq, &led_cdev->set_brightness_work); > > > > } else { > > > > set_bit(LED_BLINK_BRIGHTNESS_CHANGE, > > > > &led_cdev->work_flags); > > > > @@ -333,7 +333,7 @@ void led_set_brightness_nopm(struct led_classdev *led_cdev, unsigned int value) > > > > set_bit(LED_SET_BRIGHTNESS_OFF, &led_cdev->work_flags); > > > > } > > > > > > > > - schedule_work(&led_cdev->set_brightness_work); > > > > + queue_work(led_cdev->wq, &led_cdev->set_brightness_work); > > > > } > > > > EXPORT_SYMBOL_GPL(led_set_brightness_nopm); > > > > > > > > diff --git a/include/linux/leds.h b/include/linux/leds.h > > > > index 6300313c46b7..7c9f1cb12ab9 100644 > > > > --- a/include/linux/leds.h > > > > +++ b/include/linux/leds.h > > > > @@ -169,6 +169,7 @@ struct led_classdev { > > > > int new_blink_brightness; > > > > void (*flash_resume)(struct led_classdev *led_cdev); > > > > > > > > + struct workqueue_struct *wq; /* LED workqueue */ > > > > struct work_struct set_brightness_work; > > > > int delayed_set_value; > > > > unsigned long delayed_delay_on; > > > > -- > > > > 2.43.0 -- Thank you, Dmitry ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH v2] leds: introduce ordered workqueue for leds events instead of system_wq 2024-10-29 20:06 ` Dmitry Rokosov @ 2024-10-31 10:28 ` Lee Jones 0 siblings, 0 replies; 8+ messages in thread From: Lee Jones @ 2024-10-31 10:28 UTC (permalink / raw) To: Dmitry Rokosov Cc: pavel, linux-kernel, linux-leds, kernel, rockosov, Alexey Romanov On Tue, 29 Oct 2024, Dmitry Rokosov wrote: > Hello Lee, > > Kindly reminder. It's on the list. There is no need to contentless pings. -- Lee Jones [李琼斯] ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: (subset) [PATCH v2] leds: introduce ordered workqueue for leds events instead of system_wq 2024-09-03 22:39 [PATCH v2] leds: introduce ordered workqueue for leds events instead of system_wq Dmitry Rokosov 2024-09-16 11:18 ` Dmitry Rokosov @ 2024-11-06 8:11 ` Lee Jones 2024-11-06 8:12 ` Lee Jones 1 sibling, 1 reply; 8+ messages in thread From: Lee Jones @ 2024-11-06 8:11 UTC (permalink / raw) To: pavel, lee, Dmitry Rokosov Cc: linux-kernel, linux-leds, kernel, rockosov, Alexey Romanov On Wed, 04 Sep 2024 01:39:30 +0300, Dmitry Rokosov wrote: > This allows to setup ordered workqueue for leds events. This may be > useful, because default 'system_wq' does not guarantee execution order > of each work_struct, thus for several brightness update requests (for > multiple leds), real brightness switch could be in random order. > > Yes, for sysfs-based leds we have flush_work() call inside > brightness_store() operation, but it's blocking call, so userspace > caller can be blocked at a long time, which means leds animation stream > can be broken. > > [...] Applied, thanks! [1/1] leds: introduce ordered workqueue for leds events instead of system_wq commit: 32360bf6a5d4016669c3545e7b0ec939937f5331 -- Lee Jones [李琼斯] ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: (subset) [PATCH v2] leds: introduce ordered workqueue for leds events instead of system_wq 2024-11-06 8:11 ` (subset) " Lee Jones @ 2024-11-06 8:12 ` Lee Jones 0 siblings, 0 replies; 8+ messages in thread From: Lee Jones @ 2024-11-06 8:12 UTC (permalink / raw) To: pavel, Dmitry Rokosov Cc: linux-kernel, linux-leds, kernel, rockosov, Alexey Romanov On Wed, 06 Nov 2024, Lee Jones wrote: > On Wed, 04 Sep 2024 01:39:30 +0300, Dmitry Rokosov wrote: > > This allows to setup ordered workqueue for leds events. This may be > > useful, because default 'system_wq' does not guarantee execution order > > of each work_struct, thus for several brightness update requests (for > > multiple leds), real brightness switch could be in random order. > > > > Yes, for sysfs-based leds we have flush_work() call inside > > brightness_store() operation, but it's blocking call, so userspace > > caller can be blocked at a long time, which means leds animation stream > > can be broken. > > > > [...] > > Applied, thanks! > > [1/1] leds: introduce ordered workqueue for leds events instead of system_wq > commit: 32360bf6a5d4016669c3545e7b0ec939937f5331 With a couple of style fix-ups. -- Lee Jones [李琼斯] ^ permalink raw reply [flat|nested] 8+ messages in thread
end of thread, other threads:[~2024-11-06 8:12 UTC | newest] Thread overview: 8+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2024-09-03 22:39 [PATCH v2] leds: introduce ordered workqueue for leds events instead of system_wq Dmitry Rokosov 2024-09-16 11:18 ` Dmitry Rokosov 2024-09-17 8:04 ` Lee Jones 2024-10-22 14:22 ` Dmitry Rokosov 2024-10-29 20:06 ` Dmitry Rokosov 2024-10-31 10:28 ` Lee Jones 2024-11-06 8:11 ` (subset) " Lee Jones 2024-11-06 8:12 ` Lee Jones
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox