Page MenuHomeFreeBSD
Feed Advanced Search

Thu, Oct 8

roy_marples.name updated the diff for D26636: Implement SIOCGIFALIAS.

Break out of the loop early on matching an address.

Thu, Oct 8, 8:14 PM
roy_marples.name added inline comments to D26636: Implement SIOCGIFALIAS.
Thu, Oct 8, 3:09 PM

Mon, Oct 5

roy_marples.name updated the diff for D26652: Implement SO_RERROR.

Move SO_RERROR so it sits within the correct place.

Mon, Oct 5, 7:31 AM

Sun, Oct 4

roy_marples.name added a comment to D5469: Announce INET6 address flag changes to userland.

@melifaro any update on this?

Sun, Oct 4, 8:50 PM
roy_marples.name added a comment to D26652: Implement SO_RERROR.
In D26652#593989, @pi wrote:

I had cases where quagga missed routing updates, which caused inconsistent routing between different bgp speakers. This is probably still possible with frr7, and therefore I would welcome a way to at least get some indication that data was lost.

Sun, Oct 4, 7:35 PM
roy_marples.name updated the diff for D26652: Implement SO_RERROR.

Use SO_RERROR in route(8).
Warn on any errors returned by read(2) rather than assuming we always get a route message.

Sun, Oct 4, 4:30 PM
roy_marples.name updated the diff for D26652: Implement SO_RERROR.

Adjusted man page as requested.

Sun, Oct 4, 4:04 PM
roy_marples.name added a comment to D26652: Implement SO_RERROR.

So, it looks like even the feature has been present in Linux for 8+ years, it hasn't been adopted by the relevant software.

Sun, Oct 4, 3:50 PM
roy_marples.name added a comment to D26652: Implement SO_RERROR.
In D26652#593926, @gnn wrote:

While I applaud this idea for route(4)ing sockets I think that applying it broadly to other socket types has issues that need to be considered.

Sun, Oct 4, 2:16 PM
roy_marples.name added a comment to D26652: Implement SO_RERROR.

Why would you want to limit the scope?

Because an application writer might get the impression that he/she will be notified if an incoming packet was dropped. This is not true since this patch only covers one of many reasons. If any application wants to detect this, it should add some sequence numbers to the data and it will know.

Sun, Oct 4, 1:55 PM
roy_marples.name added a comment to D26652: Implement SO_RERROR.

what other OSes share this API now?

Sun, Oct 4, 3:51 AM

Sat, Oct 3

roy_marples.name added a comment to D26652: Implement SO_RERROR.

what other OSes share this API now?

I originally implemented on NetBSD and ported it to DragonFly BSD.
OpenBSD has an API which dhcpcd also uses which is specific to the route(4) API where it sends a RTM_DESYNC message.
The irony being it has to send a message on a socket which has already overflowed.

Maybe one can force that such a message (only one) is appended to the socket buffer even if it is full.

Sat, Oct 3, 9:14 PM
roy_marples.name added a comment to D26652: Implement SO_RERROR.

what other OSes share this API now?

Sat, Oct 3, 8:31 PM
roy_marples.name requested review of D26652: Implement SO_RERROR.
Sat, Oct 3, 6:53 PM

Fri, Oct 2

roy_marples.name updated the diff for D26636: Implement SIOCGIFALIAS.

Use NET_EPOCH_{ENTER,EXIT}

Fri, Oct 2, 4:24 PM
roy_marples.name added inline comments to D26636: Implement SIOCGIFALIAS.
Fri, Oct 2, 8:15 AM
roy_marples.name requested review of D26636: Implement SIOCGIFALIAS.
Fri, Oct 2, 2:50 AM

Thu, Sep 24

roy_marples.name added a comment to D26538: Add SIOCGIFDATA.

Looks ok to me, thanks. This seems like it would be useful for programs that run under Capsicum.

Thu, Sep 24, 2:24 AM

Wed, Sep 23

roy_marples.name added a comment to D26538: Add SIOCGIFDATA.

Hrm, @markj pointed out on IRC https://people.freebsd.org/~emaste/patches/SIOCGIFDATA.diff

One downside of this approach (vs NetBSD) is that we'll have trouble if we want to grow if_data in the future, although we could deal with that if/when it happens.

Wed, Sep 23, 10:43 PM
roy_marples.name added a comment to D26538: Add SIOCGIFDATA.

it looks like OpenBSD and NetBSD implemented this differently:

Wed, Sep 23, 10:02 PM
roy_marples.name updated the diff for D26538: Add SIOCGIFDATA.

_IOW is fine.

Wed, Sep 23, 9:29 PM
roy_marples.name added inline comments to D26538: Add SIOCGIFDATA.
Wed, Sep 23, 9:28 PM
roy_marples.name updated the diff for D26538: Add SIOCGIFDATA.

Ensure no uninitialised padding is leaked.

Wed, Sep 23, 9:06 PM
roy_marples.name updated the diff for D26538: Add SIOCGIFDATA.

Hopefully this is now enough context.

Wed, Sep 23, 8:46 PM
roy_marples.name requested review of D26538: Add SIOCGIFDATA.
Wed, Sep 23, 8:31 PM

May 11 2020

roy_marples.name added a comment to D5469: Announce INET6 address flag changes to userland.

Generally looking good, though my primary concern is interaction with the existing applications.

Both bird and frr ignore value of ifam_flags.

Will update the patch and look at the behaviour for both of them.

May 11 2020, 9:19 PM

Jul 12 2016

roy_marples.name added a comment to D5469: Announce INET6 address flag changes to userland.

Let me clarify.

Jul 12 2016, 7:42 PM
roy_marples.name added a comment to D5469: Announce INET6 address flag changes to userland.
In D5469#149446, @hrs wrote:

Looks reasonable to add notification when ia6_flags is changed, but why is RTM_ADD used instead of RTM_CHANGE? rt_addrmsg() is for addition/removal of an address and RTM_ADD is translated to RTM_NEWADDR there.

Jul 12 2016, 8:30 AM

Feb 27 2016

roy_marples.name retitled D5469: Announce INET6 address flag changes to userland from to Announce INET6 address flag changes to userland.
Feb 27 2016, 2:30 AM