| Message ID | d3cdb3fef8434663c95b7b2f799c5ee4446a5231.1528397238.git.leonard.crestez@nxp.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On 2018-06-07 20:52, Leonard Crestez wrote: > Interrupts are ignored if no event bit is set in the status status > register and this breaks the buffer interface. No data is shown when > running "iio_generic_buffer -n mma8451 -a" and interrupt counts go > crazy. > > Fix by not returning IRQ_NONE if DRDY is set. > > Fixes: 605f72de137a ("iio: accel: mma8452: improvements to handle > multiple events") > > Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com> At least this does no harm to events. So if this solves your problem: Acked-by: Martin Kepplinger <martink@posteo.de> thanks, martin > > --- > drivers/iio/accel/mma8452.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > Perhaps this whole early-exit check could be dropped? It is not clear > how it helps. > > If for some models we want to ignore unsupported events then maybe this > should be checked for each individual bit. Instead of > > if (src & MMA8452_INT_FF_MT) { > > Check for: > > if ((src & MMA8452_INT_FF_MT) && (data->chip_info->enabled_events & MMA8452_INT_FF_MT)) > > diff --git a/drivers/iio/accel/mma8452.c b/drivers/iio/accel/mma8452.c > index 7a2da7f9d4dc..5485b35fe553 100644 > --- a/drivers/iio/accel/mma8452.c > +++ b/drivers/iio/accel/mma8452.c > @@ -1032,11 +1032,11 @@ static irqreturn_t mma8452_interrupt(int irq, void *p) > > src = i2c_smbus_read_byte_data(data->client, MMA8452_INT_SRC); > if (src < 0) > return IRQ_NONE; > > - if (!(src & data->chip_info->enabled_events)) > + if (!(src & (data->chip_info->enabled_events | MMA8452_INT_DRDY))) > return IRQ_NONE; > > if (src & MMA8452_INT_DRDY) { > iio_trigger_poll_chained(indio_dev->trig); > ret = IRQ_HANDLED; >
On 2018-06-07 20:52, Leonard Crestez wrote: > Interrupts are ignored if no event bit is set in the status status > register and this breaks the buffer interface. No data is shown when > running "iio_generic_buffer -n mma8451 -a" and interrupt counts go > crazy. > > Fix by not returning IRQ_NONE if DRDY is set. > > Fixes: 605f72de137a ("iio: accel: mma8452: improvements to handle > multiple events") > > Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com> Thanks for catching it. Looks good to me. Acked-by: Harinath Nampally <harinath922@gmail.com> Thanks, Harinath On Wed, Jun 20, 2018 at 5:15 AM, Martin Kepplinger <martink@posteo.de> wrote: > On 2018-06-07 20:52, Leonard Crestez wrote: >> Interrupts are ignored if no event bit is set in the status status >> register and this breaks the buffer interface. No data is shown when >> running "iio_generic_buffer -n mma8451 -a" and interrupt counts go >> crazy. >> >> Fix by not returning IRQ_NONE if DRDY is set. >> >> Fixes: 605f72de137a ("iio: accel: mma8452: improvements to handle >> multiple events") >> >> Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com> > > At least this does no harm to events. So if this solves your problem: > > Acked-by: Martin Kepplinger <martink@posteo.de> > > thanks, > > martin > >> >> --- >> drivers/iio/accel/mma8452.c | 2 +- >> 1 file changed, 1 insertion(+), 1 deletion(-) >> >> Perhaps this whole early-exit check could be dropped? It is not clear >> how it helps. >> >> If for some models we want to ignore unsupported events then maybe this >> should be checked for each individual bit. Instead of >> >> if (src & MMA8452_INT_FF_MT) { >> >> Check for: >> >> if ((src & MMA8452_INT_FF_MT) && (data->chip_info->enabled_events & MMA8452_INT_FF_MT)) >> >> diff --git a/drivers/iio/accel/mma8452.c b/drivers/iio/accel/mma8452.c >> index 7a2da7f9d4dc..5485b35fe553 100644 >> --- a/drivers/iio/accel/mma8452.c >> +++ b/drivers/iio/accel/mma8452.c >> @@ -1032,11 +1032,11 @@ static irqreturn_t mma8452_interrupt(int irq, void *p) >> >> src = i2c_smbus_read_byte_data(data->client, MMA8452_INT_SRC); >> if (src < 0) >> return IRQ_NONE; >> >> - if (!(src & data->chip_info->enabled_events)) >> + if (!(src & (data->chip_info->enabled_events | MMA8452_INT_DRDY))) >> return IRQ_NONE; >> >> if (src & MMA8452_INT_DRDY) { >> iio_trigger_poll_chained(indio_dev->trig); >> ret = IRQ_HANDLED; >> > -- To unsubscribe from this list: send the line "unsubscribe linux-iio" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/drivers/iio/accel/mma8452.c b/drivers/iio/accel/mma8452.c index 7a2da7f9d4dc..5485b35fe553 100644 --- a/drivers/iio/accel/mma8452.c +++ b/drivers/iio/accel/mma8452.c @@ -1032,11 +1032,11 @@ static irqreturn_t mma8452_interrupt(int irq, void *p) src = i2c_smbus_read_byte_data(data->client, MMA8452_INT_SRC); if (src < 0) return IRQ_NONE; - if (!(src & data->chip_info->enabled_events)) + if (!(src & (data->chip_info->enabled_events | MMA8452_INT_DRDY))) return IRQ_NONE; if (src & MMA8452_INT_DRDY) { iio_trigger_poll_chained(indio_dev->trig); ret = IRQ_HANDLED;
Interrupts are ignored if no event bit is set in the status status register and this breaks the buffer interface. No data is shown when running "iio_generic_buffer -n mma8451 -a" and interrupt counts go crazy. Fix by not returning IRQ_NONE if DRDY is set. Fixes: 605f72de137a ("iio: accel: mma8452: improvements to handle multiple events") Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com> --- drivers/iio/accel/mma8452.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) Perhaps this whole early-exit check could be dropped? It is not clear how it helps. If for some models we want to ignore unsupported events then maybe this should be checked for each individual bit. Instead of if (src & MMA8452_INT_FF_MT) { Check for: if ((src & MMA8452_INT_FF_MT) && (data->chip_info->enabled_events & MMA8452_INT_FF_MT))