svn commit: r356755 - in head/sys: net netinet netinet6 netpfil/ipfw/nat64 sys

[Gleb Smirnoff]

On Thu, Jan 16, 2020 at 12:06:17PM -0800, Cy Schubert wrote:
C> On January 15, 2020 2:38:05 PM PST, Gleb Smirnoff <glebius at freebsd.org> wrote:
C> >On Wed, Jan 15, 2020 at 09:44:53AM -1000, Jeff Roberson wrote:
C> >J> On Wed, 15 Jan 2020, Gleb Smirnoff wrote:
C> >J> 
C> >J> > Author: glebius
C> >J> > Date: Wed Jan 15 06:05:20 2020
C> >J> > New Revision: 356755
C> >J> > URL: https://svnweb.freebsd.org/changeset/base/356755
C> >J> >
C> >J> > Log:
C> >J> >  Introduce NET_EPOCH_CALL() macro and use it everywhere where we
C> >free
C> >J> >  data based on the network epoch.   The macro reverses the
C> >argument
C> >J> >  order of epoch_call(9) - first function, then its argument. NFC
C> >J> 
C> >J> Is there some practical impact of changing the argument order or
C> >does it 
C> >J> just seem more natural to you?
C> >
C> >It is just more natural. I'm suggesting to change prototype of
C> >epoch_call()
C> >to the same order as well.
C> 
C> Yes.
C> 
C> Are there any ports that might be affected,  such as kms-drm or virtualbox-ose-kmod?

I'm not aware of any software outside /usr/src/sys that uses epoch.

-- 
Gleb Smirnoff