Re: [PATCH net-next v2 0/2] skbuff: Fix applications not being woken for errors

[Andrew Lunn <andrew@lunn.ch>]

On Wed, Mar 14, 2018 at 02:54:35PM -0700, Vinicius Costa Gomes wrote:
> Hi,
> 
> Changes from v1:
>  - Fixed comments from Willem de Bruijn, about the order of the
>  options passed to getopt();
>  - Added Reviewed-by and Fixes tags to patch (2);
> 
> Changes from the RFC:
>  - tweaked commit messages;
> 
> Original cover letter:
> 
> This is actually a "bug report"-RFC instead of the more usual "new
> feature"-RFC.
> 
> We are developing an application that uses TX hardware timestamping to
> make some measurements, and during development Randy Witt initially
> reported that the application poll() never unblocked when TX hardware
> timestamping was enabled.

Hi Vinicius

This sounds a lot like an issue i was chasing a month ago with ptp4l
and the newly added PTP support for Marvell switch chips. I was seeing
poll not unblocking, and when it did, the sequence numbers for PTP
messages indicating they were old. But it suddenly started working,
and i've not been able to reproduce it again.

> (my hypothesis is that only triggers when the error is reported from
> a different task context than the application).

The case i was having problems with, it is a kernel task which does
the error reporting, which is clearly a different context to the ptp4l
application. So your hypothesis could be correct.

	     Andrew