On Mon, 2017-07-17 at 08:51 -0400, Josh Boyer wrote:
> On Mon, Jul 17, 2017 at 8:40 AM, Sérgio Basto <ser...@serjux.com>
> wrote:
> > On Fri, 2017-07-14 at 08:03 -0400, Josh Boyer wrote:
> > > On Fri, Jul 14, 2017 at 7:46 AM, Sérgio Basto <ser...@serjux.com>
> > > wrote:
> > > > Hi,
> > > > I have a bug report that can't build virtualbox kmods for
> > > > kernels
> > > > on
> > > > rawhide
> > > > 
> > > > Larry Finger for opensuse wrote:
> > > > 
> > > > Yes, it does not work for kernel 4.11. The "#ifndef" will
> > > > eventually be
> > > > replaced
> > > > by "#if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 13, 0)", but
> > > > that
> > > > will
> > > > not work
> > > > until kernel 4.13-rc1 is released. You asked about 4.13-rc0,
> > > > but
> > > > that
> > > > entity
> > > > does not exist here.
> > > > If your kernel Makefile does indeed have 4.13, then use the
> > > > kernel
> > > > version test.
> > > > 
> > > > So the main question is kernel Makefile does indeed have 4.13 ?
> > > 
> > > No, and it won't until upstream releases 4.13-rc1.  Merge window
> > > kernels are called 4.12+ until that point.  The RPM versioning in
> > > Fedora is a construct to reflect that we are working on what will
> > > be
> > > the 4.13 code base.
> > 
> > As I completely understand that we test it , I ask if is it
> > possible
> > change kernel Makefile to reflect the new release , seems to me
> > more
> > correct ... .
> 
> I'd recommend against it.  It's an invented version that would only
> live for 2 weeks at a time and be a patch that needs to be applied.
> Upstream already expressed dislike for our invented 2.6.40 that we
> carried for a while because it causes confusion the other way too
> (e.g. "what is 2.6.40??  that's not a real version").
> 
> Instead, I'd recommend not building anything external against any
> merge window kernel.  That is the time of maximal code churn and
> given
> that upstream doesn't provide a stable kernel ABI/API anyway, it
> seems
> to be the worst possible time to try and build something.  Just wait
> for -rc1.

OK , JFTR [1] was not my fault :) 

[1]
https://bugzilla.rpmfusion.org/show_bug.cgi?id=4594

> josh
> _______________________________________________
> kernel mailing list -- kernel@lists.fedoraproject.org
> To unsubscribe send an email to kernel-le...@lists.fedoraproject.org
-- 
Sérgio M. B.
_______________________________________________
kernel mailing list -- kernel@lists.fedoraproject.org
To unsubscribe send an email to kernel-le...@lists.fedoraproject.org

Reply via email to