> -----Original Message-----
> From: Stokes, Ian <[email protected]>
> Sent: Wednesday, July 7, 2021 3:38 PM
> To: Eelco Chaudron <[email protected]>; Van Haaren, Harry
> <[email protected]>; Amber, Kumar <[email protected]>
> Cc: Amber, Kumar <[email protected]>; Ferriter, Cian
> <[email protected]>; [email protected]; [email protected];
> [email protected]
> Subject: RE: [v6 00/11] MFEX Infrastructure + Optimizations
> 
> > On 7 Jul 2021, at 12:13, Van Haaren, Harry wrote:
> >
> > > Hi All,
> > >
> > > This thread has dissolved into unnecessary time-wasting on nitpick 
> > > changes.
> > There is no
> > > technical issue with uint32_t, so this patch remains as is, and this 
> > > should be
> > accepted for merge.
> > >
> > > If you feel differently, reply to this with a detailed description of a 
> > > genuine
> > technical bug.
> >
> > Reviews are not only about technical correctness but also about coding style
> > and consistency.
> > In this case the dp_packet_batch_size() API returns a size_t, so we should 
> > try
> to
> > use it.
> >
> +1

OK, lets work on a viable solution that OVS build-system & the compilers 
support.

> > But I leave it to the maintainers to decide if they accept this as is, or 
> > not :)
> 
> If the standard is to use size_t then I believe we should stick with that. I 
> think it's
> a separate conversation with regards to if size_t should be used and is best 
> for
> the dp_packet_batch_size API and probably beyond the scope of this patch
> series.

Agree that it's beyond scope of patchset. 

> For the moment I'd prefer to see a solution using size_t.

OK, based on coding-style.rst, the PRIuSIZE format spec. will be used.

> Regards
> Ian
> 
> > //Eelco

Thanks for input all.

<snip>
_______________________________________________
dev mailing list
[email protected]
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Reply via email to