On Thu, Dec 08, 2016 at 01:04:11PM -0600, Grygorii Strashko wrote:

huh. Seems this is not really good idea, because MISC Irq will be 
triggered for *any* CPTS event and there is no way to enable it just for
HW_TS_PUSH.

So what?  That is not a problem.

So, this doesn't work will with current code for RX/TX timestamping
(which uses polling mode).

Why doesn't it work?

+ runtime overhead in net RX/TX caused by 
triggering more interrupts. 

This is not relevant.  Without HW_TS_PUSH, there is no need for
enabling the interrupt simply because we don't need it.  Now, with
HW_TS_PUSH, we do need it.
 

May be, overflow check/polling timeout can be made configurable (module parameter). 

No, it should just work without any user space fiddling.

I getting a bit tired of your half-baked implementations of the
ancillary clock functions.  Either do it right, or just leave it
unsupported.

Thanks,
Richard