Re: 4k mbuf alignment on intel?
Re: 4k mbuf alignment on intel?
- Subject: Re: 4k mbuf alignment on intel?
- From: Andrew Gallatin <email@hidden>
- Date: Mon, 23 Oct 2006 08:58:45 -0400 (EDT)
Terry Lambert writes:
> From past posts, he's already outside KPI-land for his application,
> and is OK with rev'ing the code on each SU to be able to do that.
Actually, all my drivers are with the KPI. Perhaps you're thinking
about my posts where I complain about the restrictions of the KPIs.
I'm complaining about them precisely because I'm using them!
> On Oct 20, 2006, at 6:03 PM, Josh Graessley wrote:
> > It is not possible to maintain your own buffer list and use M_EXT..
> > There is no way to do this using the KPI. If you absolutely need
> > this, file a bug report and it may be considered for some future
> > release.
> >
> > -josh
> >
Thank you for verifying that, Josh. The other big problem with M_EXT
buffers is that the last time I checked, IOMbufMemoryCursor was not
capable of returning a dma address for anything outside of
mcl_paddr[]. This means that private receive buffers allocated by a
driver would be hard to forward out another interface..
Drew
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Darwin-kernel mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden