On Thu, Jun 14, 2012 at 4:53 PM, Russell King - ARM Linux
[off-list ref] wrote:

On Thu, Jun 14, 2012 at 04:48:56PM +0530, Shilimkar, Santosh wrote:

quoted

On Thu, Jun 14, 2012 at 3:50 PM, Felipe Balbi [off-list ref] wrote:

quoted

otherwise we could get our IRQ line disabled due
to many spurious IRQs.

Signed-off-by: Felipe Balbi <redacted>
---
?drivers/i2c/busses/i2c-omap.c | ? ?2 +-
?1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/i2c/busses/i2c-omap.c b/drivers/i2c/busses/i2c-omap.c
index fc5b8bc..5b78a73 100644
--- a/drivers/i2c/busses/i2c-omap.c
+++ b/drivers/i2c/busses/i2c-omap.c

@@ -1015,7 +1015,7 @@ omap_i2c_isr(int this_irq, void *dev_id)

? ? ? ? ? ? ? ?}
? ? ? ?} while (stat);

- ? ? ? return count ? IRQ_HANDLED : IRQ_NONE;
+ ? ? ? return IRQ_HANDLED;

no sure if this is correct. if you have IRQ flood and instead of _actually_
handling it, if you return handled, you still have interrupt pending, right?

The point of returning IRQ_NONE is to indicate to the interrupt layer that
the interrupt you received was not processed by any interrupt handler, and
therefore to provide a way of preventing the system being brought to a halt
though a stuck interrupt line.

So, if you do process an interrupt, you should always return IRQ_HANDLED
even if you couldn't complete its processing (eg, because you've serviced
it 100 times.)

That make sense. Thanks for explanation Russell.

Regards
santosh