snmpd OPENBSD-PF-MIB table 'match' counters

2013-09-02 Thread Joel Knight
Hi, This diff adds the table packet/byte counters for "match" rules to PF-MIB. In case gmail mucks up the formatting, the diff is here too: http://www.packetmischief.ca/files/openbsd/patches/snmpd.match.diff ok? .joel Index: OPENBSD-PF-MIB.txt =

Re: pms: synaptics touchpad resume fix

2013-09-02 Thread Stefan Sperling
On Tue, Sep 03, 2013 at 12:22:32AM +0600, Alexandr Shadchin wrote: > Maybe it makes sense to try to increase the response time Changing these timeouts doesn't seem to make any difference. It's still retrying 3 or 4 times after resume, then works.

Re: pms: synaptics touchpad resume fix

2013-09-02 Thread Alexandr Shadchin
On Mon, Sep 02, 2013 at 06:51:36PM +0200, Stefan Sperling wrote: > On Mon, Sep 02, 2013 at 02:32:50PM +0200, Martin Pieuchot wrote: > > You might also rely on the fact that if you have a "sc->synaptics" > > already allocated to try harder. Because in this case you know > > that you have a synaptic

Re: pms: synaptics touchpad resume fix

2013-09-02 Thread Stefan Sperling
On Mon, Sep 02, 2013 at 02:32:50PM +0200, Martin Pieuchot wrote: > You might also rely on the fact that if you have a "sc->synaptics" > already allocated to try harder. Because in this case you know > that you have a synaptic touchpad. That works. Here's a simpler diff that fixes my issue, too.

Re: pflow(4) without flowsrc

2013-09-02 Thread Florian Obser
On Mon, Sep 02, 2013 at 11:11:43AM +0200, Martin Pieuchot wrote: > On 31/08/13(Sat) 04:28, Nathanael Rensen wrote: > > If no flowsrc is specified on a pflow(4) interface then the src address > > is determined by ip_output(). However prior to calling ip_output() pflow(4) > > has already calculated t

Re: useradd with empty -k doesn't chown/chmod new home directory

2013-09-02 Thread Craig R. Skinner
On 2013-08-31 Sat 11:45 AM |, patrick keshishian wrote: > On Sat, Aug 31, 2013 at 06:23:25AM -0600, Todd C. Miller wrote: > > Assuming we want to make this a non-fatal error the following should > > do. > > You meant non-existent skel dir, not empty. Unless you > meant empty argument for -k option

Re: pms: synaptics touchpad resume fix

2013-09-02 Thread Alexey E. Suslikov
Stefan Sperling openbsd.org> writes: > I've got a synaptics touchpad which is taking a relatively large > amount of time to respond to the synaptics magic query during resume. As seen in Linux[1], USB devices also require longer than normal timeouts after being woken up. [1] http://marc.info/?l

Re: Introduce rt_msg() (was nd6_rtmsg)

2013-09-02 Thread Kenneth R Westerback
On Mon, Sep 02, 2013 at 12:43:51PM +0200, Martin Pieuchot wrote: > Diff below is just a small refactoring of two similar code chunks to > inform user processes that something changed regarding a route. > > I'd like to get this in because it removes one use of rt_addrinfo in > netinet6. > > There'

Re: pms: synaptics touchpad resume fix

2013-09-02 Thread Martin Pieuchot
On 02/09/13(Mon) 11:44, Stefan Sperling wrote: > I've got a synaptics touchpad which is taking a relatively large > amount of time to respond to the synaptics magic query during resume. > > pms0 at pckbc0 (aux slot) > wsmouse0 at pms0 mux 0 > wsmouse1 at pms0 mux 0 > pms0: Synaptics clickpad, firm

Re: Introduce rt_msg() (was nd6_rtmsg)

2013-09-02 Thread Martin Pieuchot
On 02/09/13(Mon) 07:54, Kenneth R Westerback wrote: > On Mon, Sep 02, 2013 at 12:43:51PM +0200, Martin Pieuchot wrote: > > Diff below is just a small refactoring of two similar code chunks to > > inform user processes that something changed regarding a route. > > > > I'd like to get this in becaus

Introduce rt_msg() (was nd6_rtmsg)

2013-09-02 Thread Martin Pieuchot
Diff below is just a small refactoring of two similar code chunks to inform user processes that something changed regarding a route. I'd like to get this in because it removes one use of rt_addrinfo in netinet6. There's no functional change, ok? Index: net/route.c ===

Re: defer routing table updates on link state changes

2013-09-02 Thread Mike Belopuhov
On Mon, Aug 26, 2013 at 13:36 +0200, Mike Belopuhov wrote: > hi, > > in order to make our life a bit easier and prevent rogue > accesses to the routing table from the hardware interrupt > context violating all kinds of spl assumptions we would > like if_link_state_change that is called by network

Re: in_var.h incudes in6_var.h

2013-09-02 Thread Martin Pieuchot
On 30/08/13(Fri) 21:50, Alexander Bluhm wrote: > Hi, > > The file netinet/in_var.h includes netinet6/in6_var.h. This creates > a bunch of useless dependencies. For an upcomming change in in6_var.h > I would like to split that up. > > Is this a good idea? comments/ok? I like the idea but we sho

pms: synaptics touchpad resume fix

2013-09-02 Thread Stefan Sperling
I've got a synaptics touchpad which is taking a relatively large amount of time to respond to the synaptics magic query during resume. pms0 at pckbc0 (aux slot) wsmouse0 at pms0 mux 0 wsmouse1 at pms0 mux 0 pms0: Synaptics clickpad, firmware 8.0 The pms(4) driver gives up on it quickly, and then

Re: pflow(4) without flowsrc

2013-09-02 Thread Martin Pieuchot
On 31/08/13(Sat) 04:28, Nathanael Rensen wrote: > If no flowsrc is specified on a pflow(4) interface then the src address > is determined by ip_output(). However prior to calling ip_output() pflow(4) > has already calculated the UPD pseudo-header checksum based on INADDR_ANY. > This results in a ba