svn commit: r266974 - in head/sys: dev/dc dev/fxp dev/mii dev/netmap kern net

[Gleb Smirnoff]

On Tue, Jun 03, 2014 at 06:05:50PM +0100, David Chisnall wrote:
D> On 3 Jun 2014, at 06:01, Adrian Chadd <adrian at FreeBSD.org> wrote:
D> 
D> > I wonder if in the short term we should just use inlines for now, at
D> > least so the methodization can get done without hurting people on
D> > ARM/MIPS.
D> 
D> It's probably worth thinking a bit more carefully about the KPI, since it's something we'll likely be stuck with for a long time...
D> 
D> If access to a particular field is on the hot path, and you're never going to want to do interposition, then the obvious thing to do is have a method that returns a pointer to the field so that you can cache it in the caller.  Some thought needs to be given there as to whether these things are _Atomic() qualified for when we (eventually) move to C11 for the kernel.

Yes, this is the plan. Some high performance drivers can cache pointers to
partial contents of an ifnet(9).

-- 
Totus tuus, Glebius.