* [PATCH] da9052: ensure da9052_ts_input_open will not lock up on irq flood
@ 2016-03-03 10:52 Michael Grzeschik
2016-03-11 19:33 ` Dmitry Torokhov
0 siblings, 1 reply; 3+ messages in thread
From: Michael Grzeschik @ 2016-03-03 10:52 UTC (permalink / raw)
To: linux-input; +Cc: dmitry.torokhov, support.opensource, kernel
It is possible that da9052_ts_input_open is called and the touchscreen
is triggering continuous interrupts. In this case we will never leave
da9052_ts_input_open. The da9052_reg_update will never get the i2c bus
as the interrupt handler will continuously lock it.
This patch switch places of da9052_reg_update and the setting of
tsi->stopped. This way its ensured that da9052_ts_input_open will never
lockup on interrupt flood.
Signed-off-by: Michael Grzeschik <m.grzeschik@pengutronix.de>
---
drivers/input/touchscreen/da9052_tsi.c | 11 +++++++----
1 file changed, 7 insertions(+), 4 deletions(-)
diff --git a/drivers/input/touchscreen/da9052_tsi.c b/drivers/input/touchscreen/da9052_tsi.c
index 5a013bb..1b136b8 100644
--- a/drivers/input/touchscreen/da9052_tsi.c
+++ b/drivers/input/touchscreen/da9052_tsi.c
@@ -189,17 +189,20 @@ static int da9052_configure_tsi(struct da9052_tsi *tsi)
static int da9052_ts_input_open(struct input_dev *input_dev)
{
+ int val;
struct da9052_tsi *tsi = input_get_drvdata(input_dev);
- tsi->stopped = false;
- mb();
+ val = da9052_reg_update(tsi->da9052, DA9052_TSI_CONT_A_REG,
+ 1 << 1, 1 << 1);
/* Unmask PEN_DOWN event */
da9052_enable_irq(tsi->da9052, DA9052_IRQ_PENDOWN);
+ tsi->stopped = false;
+ mb();
+
/* Enable Pen Detect Circuit */
- return da9052_reg_update(tsi->da9052, DA9052_TSI_CONT_A_REG,
- 1 << 1, 1 << 1);
+ return val;
}
static void da9052_ts_input_close(struct input_dev *input_dev)
--
2.7.0
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH] da9052: ensure da9052_ts_input_open will not lock up on irq flood
2016-03-03 10:52 [PATCH] da9052: ensure da9052_ts_input_open will not lock up on irq flood Michael Grzeschik
@ 2016-03-11 19:33 ` Dmitry Torokhov
2016-03-19 9:12 ` Michael Grzeschik
0 siblings, 1 reply; 3+ messages in thread
From: Dmitry Torokhov @ 2016-03-11 19:33 UTC (permalink / raw)
To: Michael Grzeschik; +Cc: linux-input, support.opensource, kernel
On Thu, Mar 03, 2016 at 11:52:02AM +0100, Michael Grzeschik wrote:
> It is possible that da9052_ts_input_open is called and the touchscreen
> is triggering continuous interrupts. In this case we will never leave
> da9052_ts_input_open. The da9052_reg_update will never get the i2c bus
> as the interrupt handler will continuously lock it.
>
> This patch switch places of da9052_reg_update and the setting of
> tsi->stopped. This way its ensured that da9052_ts_input_open will never
> lockup on interrupt flood.
I understand what you are saying about updating register, however I do
not understand why we would want to swap unmasking pendown event ans
clearing the "stopped" flag.
Thanks.
>
> Signed-off-by: Michael Grzeschik <m.grzeschik@pengutronix.de>
> ---
> drivers/input/touchscreen/da9052_tsi.c | 11 +++++++----
> 1 file changed, 7 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/input/touchscreen/da9052_tsi.c b/drivers/input/touchscreen/da9052_tsi.c
> index 5a013bb..1b136b8 100644
> --- a/drivers/input/touchscreen/da9052_tsi.c
> +++ b/drivers/input/touchscreen/da9052_tsi.c
> @@ -189,17 +189,20 @@ static int da9052_configure_tsi(struct da9052_tsi *tsi)
>
> static int da9052_ts_input_open(struct input_dev *input_dev)
> {
> + int val;
> struct da9052_tsi *tsi = input_get_drvdata(input_dev);
>
> - tsi->stopped = false;
> - mb();
> + val = da9052_reg_update(tsi->da9052, DA9052_TSI_CONT_A_REG,
> + 1 << 1, 1 << 1);
>
> /* Unmask PEN_DOWN event */
> da9052_enable_irq(tsi->da9052, DA9052_IRQ_PENDOWN);
>
> + tsi->stopped = false;
> + mb();
> +
> /* Enable Pen Detect Circuit */
> - return da9052_reg_update(tsi->da9052, DA9052_TSI_CONT_A_REG,
> - 1 << 1, 1 << 1);
> + return val;
> }
>
> static void da9052_ts_input_close(struct input_dev *input_dev)
> --
> 2.7.0
>
--
Dmitry
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] da9052: ensure da9052_ts_input_open will not lock up on irq flood
2016-03-11 19:33 ` Dmitry Torokhov
@ 2016-03-19 9:12 ` Michael Grzeschik
0 siblings, 0 replies; 3+ messages in thread
From: Michael Grzeschik @ 2016-03-19 9:12 UTC (permalink / raw)
To: Dmitry Torokhov; +Cc: support.opensource, kernel, linux-input
On Fri, Mar 11, 2016 at 11:33:11AM -0800, Dmitry Torokhov wrote:
> On Thu, Mar 03, 2016 at 11:52:02AM +0100, Michael Grzeschik wrote:
> > It is possible that da9052_ts_input_open is called and the touchscreen
> > is triggering continuous interrupts. In this case we will never leave
> > da9052_ts_input_open. The da9052_reg_update will never get the i2c bus
> > as the interrupt handler will continuously lock it.
> >
> > This patch switch places of da9052_reg_update and the setting of
> > tsi->stopped. This way its ensured that da9052_ts_input_open will never
> > lockup on interrupt flood.
>
> I understand what you are saying about updating register, however I do
> not understand why we would want to swap unmasking pendown event ans
> clearing the "stopped" flag.
The "stopped" flag has an locking purpose in the irq handler. My patch
ensures that the irq handler will not trigger any da9052_reg_update
before da9052_ts_input_open has finished.
On second thought, it would probably be enough to swap the call of
da9052_reg_update and da9052_enable_irq in da9052_ts_input_open.
This will have the same effect. I will send an corrected v2.
Thanks,
Michael
--
Pengutronix e.K. | |
Industrial Linux Solutions | http://www.pengutronix.de/ |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2016-03-19 9:12 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2016-03-03 10:52 [PATCH] da9052: ensure da9052_ts_input_open will not lock up on irq flood Michael Grzeschik
2016-03-11 19:33 ` Dmitry Torokhov
2016-03-19 9:12 ` Michael Grzeschik
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).