Yes, this is currently documented in the doxygen for the various set
headroom routines for the patch I posted last night.
>From your code snippet are you suggesting that some platforms have a fixed
headroom that cannot be changed?  The assumption we've had is that this is
settable at least on a per-interface level.

Bill

On Tue, Nov 11, 2014 at 12:39 AM, Shmulik Ladkani <[email protected]
> wrote:

> Hi,
>
> On Mon, 10 Nov 2014 22:49:25 -0600 Bill Fischofer <
> [email protected]> wrote:
> > No, the suggestion is that implementations understand that the
> application
> > assumes buffers are "naturally aligned" so the implementation can infer
> the
> > requested alignment from the specified headroom request.  So for example
> if
> > the user requests a hr of 10 that means the user expects the packet to be
> > offset by two bytes from an 8 byte alignment.  The implementation is free
> > to add to the requested hr as long as the result preserves the implied
> > alignment from the original request.  So giving an hr of 18 or 26 is OK.
>
> OK, this is fine.
>
> So implementor MUST preserve the implicit offset "hint" provided by
> user's headroom value (i.e. allowed to add extra number of headroom
> bytes which is a multiple of an agreed alignment factor, preferrably
> platform independent factor).
>
> Meaning, such implementation isn't allowed:
>
> #define IMPL_MAX_HEADROOM 128
>
>     if (user_headroom > IMPL_MAX_HEADROOM)
>         return -ERR_NOT_SUPP;
>     actual_headroom = IMPL_MAX_HEADROOM; /* XXX: for ease of
> implementation */
>
> This should be documented.
>
> Regards,
> Shmulik
>
_______________________________________________
lng-odp mailing list
[email protected]
http://lists.linaro.org/mailman/listinfo/lng-odp

Reply via email to