10 messages, 4 authors, 2011-01-21 · open the first message on its own page
From: Philippe Langlais <hidden>
Date: 2011-01-12 10:04:08
This patch is needed to add support for SFH7741 proximity sensor on ux500 platform. Between ux500 SoC and SFH7741 sensor, we use a GPIO expander (MFD TC35892) which relays the GPIO IRQs.
From: Philippe Langlais <hidden>
Date: 2011-01-12 09:43:40
Signed-off-by: Philippe Langlais <redacted> --- drivers/input/keyboard/gpio_keys.c | 6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/input/keyboard/gpio_keys.c b/drivers/input/keyboard/gpio_keys.c
index 6069abe..eb30063 100644
--- a/drivers/input/keyboard/gpio_keys.c
+++ b/drivers/input/keyboard/gpio_keys.c@@ -322,7 +322,7 @@ static void gpio_keys_report_event(struct gpio_button_data *bdata) struct gpio_keys_button *button = bdata->button; struct input_dev *input = bdata->input; unsigned int type = button->type ?: EV_KEY; - int state = (gpio_get_value(button->gpio) ? 1 : 0) ^ button->active_low; + int state = (gpio_get_value_cansleep(button->gpio) ? 1 : 0) ^ button->active_low; input_event(input, type, button->code, !!state); input_sync(input);
@@ -410,8 +410,8 @@ static int __devinit gpio_keys_setup_key(struct platform_device *pdev, if (!button->can_disable) irqflags |= IRQF_SHARED; - error = request_irq(irq, gpio_keys_isr, irqflags, desc, bdata); - if (error) { + error = request_any_context_irq(irq, gpio_keys_isr, irqflags, desc, bdata); + if (error < 0) { dev_err(dev, "Unable to claim irq %d; error %d\n", irq, error); goto fail3;
--
1.7.3.1
From: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Date: 2011-01-20 07:53:55
Hi Philippe, On Wed, Jan 12, 2011 at 10:41:32AM +0100, Philippe Langlais wrote:
Signed-off-by: Philippe Langlais <redacted> --- drivers/input/keyboard/gpio_keys.c | 6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-)diff --git a/drivers/input/keyboard/gpio_keys.c b/drivers/input/keyboard/gpio_keys.c index 6069abe..eb30063 100644 --- a/drivers/input/keyboard/gpio_keys.c +++ b/drivers/input/keyboard/gpio_keys.c@@ -322,7 +322,7 @@ static void gpio_keys_report_event(struct gpio_button_data *bdata) struct gpio_keys_button *button = bdata->button; struct input_dev *input = bdata->input; unsigned int type = button->type ?: EV_KEY; - int state = (gpio_get_value(button->gpio) ? 1 : 0) ^ button->active_low; + int state = (gpio_get_value_cansleep(button->gpio) ? 1 : 0) ^ button->active_low; input_event(input, type, button->code, !!state); input_sync(input);@@ -410,8 +410,8 @@ static int __devinit gpio_keys_setup_key(struct platform_device *pdev, if (!button->can_disable) irqflags |= IRQF_SHARED; - error = request_irq(irq, gpio_keys_isr, irqflags, desc, bdata); - if (error) { + error = request_any_context_irq(irq, gpio_keys_isr, irqflags, desc, bdata); + if (error < 0) {
No, this it not correct. request_any_context_irq() means that you could either get a hardirq or a treaded one. However above you are changing to gpio_get_value_cansleep() which indicates that you can sleep. You need to either revert to gpio_get_value() or explicitly request threaded IRQ. Thanks. -- Dmitry
From: Trilok Soni <hidden>
Date: 2011-01-20 08:00:36
Hi Dmitry, On 1/20/2011 1:23 PM, Dmitry Torokhov wrote:
Hi Philippe, On Wed, Jan 12, 2011 at 10:41:32AM +0100, Philippe Langlais wrote:quoted
Signed-off-by: Philippe Langlais <redacted> --- drivers/input/keyboard/gpio_keys.c | 6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-)diff --git a/drivers/input/keyboard/gpio_keys.c b/drivers/input/keyboard/gpio_keys.c index 6069abe..eb30063 100644 --- a/drivers/input/keyboard/gpio_keys.c +++ b/drivers/input/keyboard/gpio_keys.c@@ -322,7 +322,7 @@ static void gpio_keys_report_event(struct gpio_button_data *bdata) struct gpio_keys_button *button = bdata->button; struct input_dev *input = bdata->input; unsigned int type = button->type ?: EV_KEY; - int state = (gpio_get_value(button->gpio) ? 1 : 0) ^ button->active_low; + int state = (gpio_get_value_cansleep(button->gpio) ? 1 : 0) ^ button->active_low; input_event(input, type, button->code, !!state); input_sync(input);@@ -410,8 +410,8 @@ static int __devinit gpio_keys_setup_key(struct platform_device *pdev, if (!button->can_disable) irqflags |= IRQF_SHARED; - error = request_irq(irq, gpio_keys_isr, irqflags, desc, bdata); - if (error) { + error = request_any_context_irq(irq, gpio_keys_isr, irqflags, desc, bdata); + if (error < 0) {No, this it not correct. request_any_context_irq() means that you could either get a hardirq or a treaded one. However above you are changing to gpio_get_value_cansleep() which indicates that you can sleep. You need to either revert to gpio_get_value() or explicitly request threaded IRQ.
This could be problem. Say, we are doing request_any_context_irq(..) and gpio is over slow-bus, so we naturally get into the threaded code, and in the thread handler we are calling gpio_get_value(..), but in this case gpio_get_value could throw a warning that we should call gpio_get_value_cansleep(...) because sleep attribute is added into the gpiolib hook implementation for these gpios. Now suppose gpio is memory mapped and we enter into the hardirq and calling sleep variant into the handler code will be naturally bad. How about checking maysleep explicitly here while using the request_any_context_irq() call though the semantics of gpiolib framework discourages that. ---Trilok Soni -- Sent by a consultant of the Qualcomm Innovation Center, Inc. The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
From: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Date: 2011-01-20 08:13:26
Hi Trilok, On Thu, Jan 20, 2011 at 01:30:31PM +0530, Trilok Soni wrote:
Hi Dmitry, On 1/20/2011 1:23 PM, Dmitry Torokhov wrote:quoted
Hi Philippe, On Wed, Jan 12, 2011 at 10:41:32AM +0100, Philippe Langlais wrote:quoted
Signed-off-by: Philippe Langlais <redacted> --- drivers/input/keyboard/gpio_keys.c | 6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-)diff --git a/drivers/input/keyboard/gpio_keys.c b/drivers/input/keyboard/gpio_keys.c index 6069abe..eb30063 100644 --- a/drivers/input/keyboard/gpio_keys.c +++ b/drivers/input/keyboard/gpio_keys.c@@ -322,7 +322,7 @@ static void gpio_keys_report_event(struct gpio_button_data *bdata) struct gpio_keys_button *button = bdata->button; struct input_dev *input = bdata->input; unsigned int type = button->type ?: EV_KEY; - int state = (gpio_get_value(button->gpio) ? 1 : 0) ^ button->active_low; + int state = (gpio_get_value_cansleep(button->gpio) ? 1 : 0) ^ button->active_low; input_event(input, type, button->code, !!state); input_sync(input);@@ -410,8 +410,8 @@ static int __devinit gpio_keys_setup_key(struct platform_device *pdev, if (!button->can_disable) irqflags |= IRQF_SHARED; - error = request_irq(irq, gpio_keys_isr, irqflags, desc, bdata); - if (error) { + error = request_any_context_irq(irq, gpio_keys_isr, irqflags, desc, bdata); + if (error < 0) {No, this it not correct. request_any_context_irq() means that you could either get a hardirq or a treaded one. However above you are changing to gpio_get_value_cansleep() which indicates that you can sleep. You need to either revert to gpio_get_value() or explicitly request threaded IRQ.This could be problem. Say, we are doing request_any_context_irq(..) and gpio is over slow-bus, so we naturally get into the threaded code, and in the thread handler we are calling gpio_get_value(..), but in this case gpio_get_value could throw a warning that we should call gpio_get_value_cansleep(...) because sleep attribute is added into the gpiolib hook implementation for these gpios. Now suppose gpio is memory mapped and we enter into the hardirq and calling sleep variant into the handler code will be naturally bad. How about checking maysleep explicitly here while using the request_any_context_irq() call though the semantics of gpiolib framework discourages that.
Any drawbacks for explicitly switching to threaded IRQ and keeping gpio_get_value_cansleep()? (That was the 2nd option I mentioned BTW). -- Dmitry
From: Trilok Soni <hidden>
Date: 2011-01-20 08:56:43
Hi Dmitry, On 1/20/2011 1:43 PM, Dmitry Torokhov wrote:
quoted
quoted
No, this it not correct. request_any_context_irq() means that you could either get a hardirq or a treaded one. However above you are changing to gpio_get_value_cansleep() which indicates that you can sleep. You need to either revert to gpio_get_value() or explicitly request threaded IRQ.This could be problem. Say, we are doing request_any_context_irq(..) and gpio is over slow-bus, so we naturally get into the threaded code, and in the thread handler we are calling gpio_get_value(..), but in this case gpio_get_value could throw a warning that we should call gpio_get_value_cansleep(...) because sleep attribute is added into the gpiolib hook implementation for these gpios. Now suppose gpio is memory mapped and we enter into the hardirq and calling sleep variant into the handler code will be naturally bad. How about checking maysleep explicitly here while using the request_any_context_irq() call though the semantics of gpiolib framework discourages that.Any drawbacks for explicitly switching to threaded IRQ and keeping gpio_get_value_cansleep()? (That was the 2nd option I mentioned BTW).
Only drawback is that you get "dedicated thread" in case the irqs are not chained handler. In worst-case it would introduce the latencies otherwise fine. ---Trilok Soni -- Sent by a consultant of the Qualcomm Innovation Center, Inc. The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
From: Rabin Vincent <hidden>
Date: 2011-01-20 09:24:06
On Thu, Jan 20, 2011 at 08:53:48 +0100, Dmitry Torokhov wrote:
On Wed, Jan 12, 2011 at 10:41:32AM +0100, Philippe Langlais wrote:quoted
diff --git a/drivers/input/keyboard/gpio_keys.c b/drivers/input/keyboard/gpio_keys.c index 6069abe..eb30063 100644 --- a/drivers/input/keyboard/gpio_keys.c +++ b/drivers/input/keyboard/gpio_keys.c@@ -322,7 +322,7 @@ static void gpio_keys_report_event(struct gpio_button_data *bdata) struct gpio_keys_button *button = bdata->button; struct input_dev *input = bdata->input; unsigned int type = button->type ?: EV_KEY; - int state = (gpio_get_value(button->gpio) ? 1 : 0) ^ button->active_low; + int state = (gpio_get_value_cansleep(button->gpio) ? 1 : 0) ^ button->active_low; input_event(input, type, button->code, !!state); input_sync(input);@@ -410,8 +410,8 @@ static int __devinit gpio_keys_setup_key(struct platform_device *pdev, if (!button->can_disable) irqflags |= IRQF_SHARED; - error = request_irq(irq, gpio_keys_isr, irqflags, desc, bdata); - if (error) { + error = request_any_context_irq(irq, gpio_keys_isr, irqflags, desc, bdata); + if (error < 0) {No, this it not correct. request_any_context_irq() means that you could either get a hardirq or a treaded one. However above you are changing to gpio_get_value_cansleep() which indicates that you can sleep.
Note that gpio_get_value_cansleep() is not being called from the interrupt handler (threaded or not), but from a work queue. Rabin
From: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Date: 2011-01-20 10:04:26
On Thu, Jan 20, 2011 at 02:53:30PM +0530, Rabin Vincent wrote:
On Thu, Jan 20, 2011 at 08:53:48 +0100, Dmitry Torokhov wrote:quoted
On Wed, Jan 12, 2011 at 10:41:32AM +0100, Philippe Langlais wrote:quoted
diff --git a/drivers/input/keyboard/gpio_keys.c b/drivers/input/keyboard/gpio_keys.c index 6069abe..eb30063 100644 --- a/drivers/input/keyboard/gpio_keys.c +++ b/drivers/input/keyboard/gpio_keys.c@@ -322,7 +322,7 @@ static void gpio_keys_report_event(struct gpio_button_data *bdata) struct gpio_keys_button *button = bdata->button; struct input_dev *input = bdata->input; unsigned int type = button->type ?: EV_KEY; - int state = (gpio_get_value(button->gpio) ? 1 : 0) ^ button->active_low; + int state = (gpio_get_value_cansleep(button->gpio) ? 1 : 0) ^ button->active_low; input_event(input, type, button->code, !!state); input_sync(input);@@ -410,8 +410,8 @@ static int __devinit gpio_keys_setup_key(struct platform_device *pdev, if (!button->can_disable) irqflags |= IRQF_SHARED; - error = request_irq(irq, gpio_keys_isr, irqflags, desc, bdata); - if (error) { + error = request_any_context_irq(irq, gpio_keys_isr, irqflags, desc, bdata); + if (error < 0) {No, this it not correct. request_any_context_irq() means that you could either get a hardirq or a treaded one. However above you are changing to gpio_get_value_cansleep() which indicates that you can sleep.Note that gpio_get_value_cansleep() is not being called from the interrupt handler (threaded or not), but from a work queue.
Ah, then everything is good, sorry about the noise. -- Dmitry
From: Philippe Langlais <hidden>
Date: 2011-01-20 12:17:20
Dmitry, Sorry, I arrive after the war, thanks Rabin for your answer. When can I hope this patch merged in kernel? Philippe On 01/20/11 11:04, Dmitry Torokhov wrote:
On Thu, Jan 20, 2011 at 02:53:30PM +0530, Rabin Vincent wrote:quoted
On Thu, Jan 20, 2011 at 08:53:48 +0100, Dmitry Torokhov wrote:quoted
On Wed, Jan 12, 2011 at 10:41:32AM +0100, Philippe Langlais wrote:quoted
diff --git a/drivers/input/keyboard/gpio_keys.c b/drivers/input/keyboard/gpio_keys.c index 6069abe..eb30063 100644 --- a/drivers/input/keyboard/gpio_keys.c +++ b/drivers/input/keyboard/gpio_keys.c@@ -322,7 +322,7 @@ static void gpio_keys_report_event(struct gpio_button_data *bdata) struct gpio_keys_button *button = bdata->button; struct input_dev *input = bdata->input; unsigned int type = button->type ?: EV_KEY; - int state = (gpio_get_value(button->gpio) ? 1 : 0) ^ button->active_low; + int state = (gpio_get_value_cansleep(button->gpio) ? 1 : 0) ^ button->active_low; input_event(input, type, button->code, !!state); input_sync(input);@@ -410,8 +410,8 @@ static int __devinit gpio_keys_setup_key(struct platform_device *pdev, if (!button->can_disable) irqflags |= IRQF_SHARED; - error = request_irq(irq, gpio_keys_isr, irqflags, desc, bdata); - if (error) { + error = request_any_context_irq(irq, gpio_keys_isr, irqflags, desc, bdata); + if (error< 0) {No, this it not correct. request_any_context_irq() means that you could either get a hardirq or a treaded one. However above you are changing to gpio_get_value_cansleep() which indicates that you can sleep.Note that gpio_get_value_cansleep() is not being called from the interrupt handler (threaded or not), but from a work queue.Ah, then everything is good, sorry about the noise.
From: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Date: 2011-01-21 07:10:35
On Thu, Jan 20, 2011 at 01:14:50PM +0100, Philippe Langlais wrote:
Dmitry, Sorry, I arrive after the war, thanks Rabin for your answer. When can I hope this patch merged in kernel?
Applied and will be included in the next pull request. Thanks! -- Dmitry