On Mon, Aug 19, 2013 at 07:12:38PM +0200, Sebastian Andrzej Siewior wrote:
* Zubair Lutfullah | 2013-08-13 21:05:03 [+0100]:
quoted
diff --git a/drivers/iio/adc/ti_am335x_adc.c b/drivers/iio/adc/ti_am335x_adc.c
index 3ceac3e..0d7e313 100644
--- a/drivers/iio/adc/ti_am335x_adc.c
+++ b/drivers/iio/adc/ti_am335x_adc.c
…
quoted
+static irqreturn_t tiadc_irq(int irq, void *private)
+{
…
quoted
+ wake_up_interruptible(&adc_dev->wq_data_avail);
…
quoted
+}
…
quoted
+static irqreturn_t tiadc_trigger_h(int irq, void *p)
+{
…
quoted
+ schedule_work(&adc_dev->poll_work);
…
quoted
+}
…
quoted
+static void tiadc_adc_work(struct work_struct *work_s)
+{
…
quoted
+ wait_event_interruptible(adc_dev->wq_data_avail,
+ (adc_dev->data_avail == 1));
…
quoted
+}
This is not very nice. The problem is that you might sleep in a
workqueue and so the other jobs will wait until you are done. I'm think
This will change to a different style.
I'm looking into DMA support for this so once your code is working it
should be possible to switch to DMA instead reading byte wise from the
fifo.
Great.
How did you test the whole thing? Do you have a test program which
selects a few sources and reads them in continuous mode?
generic_buffer.c runs for this for sysfs trigger..
However. If you follow the other thread.
This is obsolete and trigger style is changing to a driver trigger.
I'll update when I'm done.
Thanks
ZubairLK