On 2016 Sep 17 (Sat) at 12:05:54 +0200 (+0200), Peter Hessler wrote:
:This is a 2nd pass at having BFD send route messages.
:
:I fixed the things pointed out in the first thread, with the following
:comments:
:
: route(8) for the ramdisks is not built with SMALL, so adding SMALL
:won't help.
:
: rt
On 2016 Sep 18 (Sun) at 20:29:57 +0200 (+0200), Peter Hessler wrote:
:On 2016 Sep 17 (Sat) at 13:15:56 +0200 (+0200), Peter Hessler wrote:
::Tested on an amd64 bsd.rd. dhclient, route add, route del, route show,
::all still work, and it's slightly smaller.
::
:
:All options, commands, modifiers, an
On 2016 Sep 17 (Sat) at 13:15:56 +0200 (+0200), Peter Hessler wrote:
:On 2016 Sep 17 (Sat) at 04:50:29 -0600 (-0600), Theo de Raadt wrote:
::> route(8) for the ramdisks is not built with SMALL, so adding SMALL
::> won't help.
::
::Then you'll help it. By compiling them SMALL.
:
:Happilly.
:
:Tes
On 2016 Sep 17 (Sat) at 13:15:56 +0200 (+0200), Peter Hessler wrote:
:On 2016 Sep 17 (Sat) at 04:50:29 -0600 (-0600), Theo de Raadt wrote:
::> route(8) for the ramdisks is not built with SMALL, so adding SMALL
::> won't help.
::
::Then you'll help it. By compiling them SMALL.
:
:Happilly.
:
:Tes
On 2016 Sep 17 (Sat) at 04:50:29 -0600 (-0600), Theo de Raadt wrote:
:> route(8) for the ramdisks is not built with SMALL, so adding SMALL
:> won't help.
:
:Then you'll help it. By compiling them SMALL.
Happilly.
Tested on an amd64 bsd.rd. dhclient, route add, route del, route show,
all still
> route(8) for the ramdisks is not built with SMALL, so adding SMALL
> won't help.
Then you'll help it. By compiling them SMALL. Otherwise they
won't fit, and bfd gets ripped out.
This is a 2nd pass at having BFD send route messages.
I fixed the things pointed out in the first thread, with the following
comments:
route(8) for the ramdisks is not built with SMALL, so adding SMALL
won't help.
rt_bfdmsg() is named in the same style, and is in the same place, as
the other f
On 2016 Sep 15 (Thu) at 16:30:50 +0200 (+0200), Peter Hessler wrote:
:Based on the above, and a few private comments, I will change the name
:away from bfd_softc.
Here is a diff doing a massive rename. It is mostly mechanical,
functional changes will come later. I can still establish and bring do
On 2016 Sep 15 (Thu) at 14:47:24 +0200 (+0200), Peter Hessler wrote:
:On 2016 Sep 15 (Thu) at 13:28:10 +0200 (+0200), Martin Pieuchot wrote:
::On 14/09/16(Wed) 15:50, Peter Hessler wrote:
::> +
::> + bfd = pool_get(&bfd_pool_msghdr, PR_WAITOK | PR_ZERO);
::
::This doesn't make sense you're alloca
On 2016 Sep 15 (Thu) at 16:18:46 +0200 (+0200), Henning Brauer wrote:
:* Peter Hessler [2016-09-15 14:48]:
:> On 2016 Sep 15 (Thu) at 13:28:10 +0200 (+0200), Martin Pieuchot wrote:
:> :By the way 'softc' is generally for device drivers, you might prefer
:> :something like bfdd (d for descriptor).
* Peter Hessler [2016-09-15 14:48]:
> On 2016 Sep 15 (Thu) at 13:28:10 +0200 (+0200), Martin Pieuchot wrote:
> :By the way 'softc' is generally for device drivers, you might prefer
> :something like bfdd (d for descriptor).
> There's be a whole lot of bikeshedding about this topic, I would prefer
> :By the way 'softc' is generally for device drivers, you might prefer
> :something like bfdd (d for descriptor).
>
> There's be a whole lot of bikeshedding about this topic, I would prefer
> not to change it for now.
It should be changed. mpi is right. That variable naming should only
refer p
On 2016 Sep 15 (Thu) at 13:28:10 +0200 (+0200), Martin Pieuchot wrote:
:On 14/09/16(Wed) 15:50, Peter Hessler wrote:
:> This is a work-in-progress diff that I would like to commit. I can print
:> a few things, but there is a problem when trying to bring in more
:> fields. Printing is also ugly, b
On 14/09/16(Wed) 15:50, Peter Hessler wrote:
> This is a work-in-progress diff that I would like to commit. I can print
> a few things, but there is a problem when trying to bring in more
> fields. Printing is also ugly, but I can fix that in-tree.
Well you're mixing multiples things in one diff
This is a work-in-progress diff that I would like to commit. I can print
a few things, but there is a problem when trying to bring in more
fields. Printing is also ugly, but I can fix that in-tree.
While here, I print the descr's as ints, the same way Juniper does it.
I also had to add RTM_INVAL
15 matches
Mail list logo
