On Wed, Apr 18, 2018 at 11:31:02PM +0200, Alexander Bluhm wrote:
> On Wed, Apr 18, 2018 at 05:03:04PM +0200, Florian Obser wrote:
> > @@ -1158,9 +1158,9 @@ rt_ifa_del(struct ifaddr *ifa, int flags, struct
> > sockaddr *dst)
> > error = rtrequest_delete(&info, prio, ifp, &rt, rtableid);
> >
like a small handful of man pages, kern.9 puts prototypes in places
other than the SYNOPSIS section. of these, it is probably the most
straightforward to fix. kern.9 documents a bunch of different apis
provided by "libkern", but they can be split out into manpages for
each type of api. this diff ta
On Wed, Apr 18, 2018 at 05:03:04PM +0200, Florian Obser wrote:
> @@ -1158,9 +1158,9 @@ rt_ifa_del(struct ifaddr *ifa, int flags, struct
> sockaddr *dst)
> error = rtrequest_delete(&info, prio, ifp, &rt, rtableid);
> if (error == 0) {
> rtm_send(rt, RTM_DELETE, 0, rtableid
On Fri, Apr 13, 2018 at 02:38:07PM +0200, Klemens Nanni wrote:
> On Thu, Apr 12, 2018 at 10:07:58AM +0200, Peter Hessler wrote:
> > On 2018 Apr 11 (Wed) at 23:01:45 +0200 (+0200), Klemens Nanni wrote:
> > :On Wed, Apr 11, 2018 at 09:28:03AM +0200, Peter Hessler wrote:
> > :> No, all of these uses a
diff --git route.c route.c
index 9c85f70de78..a9ee414c562 100644
--- route.c
+++ route.c
@@ -1257,7 +1257,8 @@ char *msgtypes[] = {
"RTM_DESYNC: route socket overflow",
"RTM_INVALIDATE: invalidate cache of L2 route",
"RTM_BFD: bidirectional forwarding detection",
- "RT
On Wed, Apr 18, 2018 at 05:05:59PM +0200, Florian Obser wrote:
> This is to inform userland (i.e. slaacd(8)) when duplicate address
> detection finishes.
>
> Not a big fan of the lock/unlock dance but I guess it can't be helped
> for now.
>
> Comments, OKs?
Theo points out that I suck at naming
On Wed, Apr 18, 2018 at 07:13:49PM +0200, Florian Obser wrote:
> So I was puzzled why route(8) pledges rpath. As far as I can work out
> it's because asr tries to open /etc/networks in getnetnamadr_async.c
> (line 183).
>
> If we whitelist that file like all the other things asr needs we can
> dro
Well the other approach would be to finally delete support
for /etc/networks
Florian Obser wrote:
> So I was puzzled why route(8) pledges rpath. As far as I can work out
> it's because asr tries to open /etc/networks in getnetnamadr_async.c
> (line 183).
>
> If we whitelist that file li
So I was puzzled why route(8) pledges rpath. As far as I can work out
it's because asr tries to open /etc/networks in getnetnamadr_async.c
(line 183).
If we whitelist that file like all the other things asr needs we can
drop the rpath pledge in route(8).
Thoughts?
diff --git sbin/route/route.c s
On Wed, Apr 18, 2018 at 05:03:04PM +0200, Florian Obser wrote:
> some dude commited it like this in '95.
It used to be used until mpi@ started passing ifa:
"Pass the configured ``ifa'' to rt_sendaddrmsg() instead of
getting it via ``rt->rt_ifa'' later"
http://cvsweb.openbsd.org/cg
teach route(8) about RTM_CHGADDR
OK?
diff --git route.c route.c
index 9c85f70de78..fee99dcd24b 100644
--- route.c
+++ route.c
@@ -1257,7 +1257,8 @@ char *msgtypes[] = {
"RTM_DESYNC: route socket overflow",
"RTM_INVALIDATE: invalidate cache of L2 route",
"RTM_BFD: bidirecti
This is to inform userland (i.e. slaacd(8)) when duplicate address
detection finishes.
Not a big fan of the lock/unlock dance but I guess it can't be helped
for now.
Comments, OKs?
diff --git net/route.h net/route.h
index 3c89348cb43..3d958719049 100644
--- net/route.h
+++ net/route.h
@@ -241,6
some dude commited it like this in '95.
I came accress it because I want to use it in a place where I don't
have a rtentry allocated already.
OK?
diff --git net/route.c net/route.c
index 30c8def301d..a0c5738d831 100644
--- net/route.c
+++ net/route.c
@@ -1103,7 +1103,7 @@ rt_ifa_add(struct ifadd
Run duplicate address detection again if an existing address gets
updated from userland that was marked duplicated or tentative.
Otherwise we would just lose the duplicated / tentative state and assume
that the address is now unique and usable.
OK?
diff --git in6.c in6.c
index 1c5ec065aa5..b347e
Diff below gets rid of the (un)famous XXX in NFS by implementing proper
locking for NFS nodes, similar to the inode one.
It needs *a lot* of tests. If you run into a problem, try to reproduce
it by defining VFSLCKDEBUG and WITNESS in your kernel config file. If
that's not enough, you can toggle
On 16/04/18(Mon) 12:33, Martin Pieuchot wrote:
> Diff below is a rewrite/cleanup of fdcopy() to avoid using memcpy().
>
> The problem with the memcpys is that they introduce a window where
> some 'struct file *' are duplicated without being refcounted. We
> should instead use the following common
16 matches
Mail list logo
