Re: delete pltime and vltime

From: Florian Obser Date: Wed, 12 Apr 2023 14:45:10 +0200 I had overlooked SIOCAIFADDR_IN6. > On 2023-04-12 20:13 +09, Masato Asou wrote: >> Hi, >> >> SIOCSIFALIFETIME_IN6 has been removed from sys/netinet6/in6_var.h with >> the following commit: >> >> commit f487585d711456156cf95432fac5a11ff78

Re: efi(4): Support for EFI variables and tables in the kernel

> Date: Tue, 18 Apr 2023 19:11:40 +0300 > From: Sergii Dmytruk > > Hi there, > > Pinging to say that this thread is silent for 3 months. EFI runtime and > bootloader changes are there, /dev/efi is the last piece left to be > committed for fwupd port to work. > > Best regards, > Sergii Hi Serg

in*_control: Drop error variable to return directly

On Wed, Apr 19, 2023 at 05:59:24PM +, Klemens Nanni wrote: > No unlock except for the NULL case, but this lets us clean up the callers > a little. In, here's the cleanup. OK? Index: netinet6/in6.c === RCS file: /cvs/src/sys/neti

Re: efi(4): Support for EFI variables and tables in the kernel

Hi there, Pinging to say that this thread is silent for 3 months. EFI runtime and bootloader changes are there, /dev/efi is the last piece left to be committed for fwupd port to work. Best regards, Sergii On Sun, Jan 08, 2023 at 12:48:05PM +0900, YASUOKA Masahiko wrote: > Hi, > > On Wed, 04 Jan

Re: bgpd add imsg code for bgpctl support

On Wed, Apr 19, 2023 at 06:05:28PM +0200, Claudio Jeker wrote: > Implement IMSG_CTL_SHOW_FLOWSPEC and IMSG_FLOWSPEC_FLUSH and add bits for > IMSG_FLOWSPEC_ADD and IMSG_FLOWSPEC_REMOVE received from bgpctl via SE. > > This is mostly streight forward code (copying the bits which are already > around

move kernel lock into multicast ioctl handler

No unlock except for the NULL case, but this lets us clean up the callers a little. OK? Index: netinet6/ip6_mroute.c === RCS file: /cvs/src/sys/netinet6/ip6_mroute.c,v retrieving revision 1.135 diff -u -p -r1.135 ip6_mroute.c --- net

bgpd add imsg code for bgpctl support

Implement IMSG_CTL_SHOW_FLOWSPEC and IMSG_FLOWSPEC_FLUSH and add bits for IMSG_FLOWSPEC_ADD and IMSG_FLOWSPEC_REMOVE received from bgpctl via SE. This is mostly streight forward code (copying the bits which are already around). -- :wq Claudio Index: bgpd.h ===

tmpfs: eject it

tmpfs was disabled in GENERIC a bit shy of 7 years ago, is it time to kill it? Skimming sbin/mount_tmpfs and sys/tmpfs, it looks like the vast majority of the touches were part of wide-encompassing diffs, along with a couple tmpfs panic fixes from the mailing lists. My cvs diff-fu is not strong

Re: rtfree(9): kernel lock is not required for free(9) calls

Forgot to say, rt_timer_remove_all() and ifafree() could be moved out of kernel lock too, but with next diff. On Wed, Apr 19, 2023 at 05:03:00PM +0300, Vitaliy Makkoveev wrote: > Index: sys/net/route.c > === > RCS file: /cvs/src/sys/n

rtfree(9): kernel lock is not required for free(9) calls

Index: sys/net/route.c === RCS file: /cvs/src/sys/net/route.c,v retrieving revision 1.416 diff -u -p -r1.416 route.c --- sys/net/route.c 28 Jan 2023 10:17:16 - 1.416 +++ sys/net/route.c 19 Apr 2023 14:00:31 - @@ -5

Current state of BFD (RFC 5880 / 5881)

Hi, I'm just wondering what the current state of BFD - Bi-directional Forwarding Detection (RFC 5880 / 5881) - is? There was a lot of "noise" (sorry for the bad wording) about it between 2016 and 2018/2019 and several commits too as phessler@ was talking about and working on it. Currently it

Re: bgpd announce flowspec

On Wed, Apr 19, 2023 at 12:53:56PM +0200, Claudio Jeker wrote: > With this diffs all the bits I committed before are connected and allow > for flowspec to be announced to other systems. > > I first thought I will put the flowspec_valid() inside flowspec_add() but > I changed my mind now and moved

bgpd announce flowspec

With this diffs all the bits I committed before are connected and allow for flowspec to be announced to other systems. I first thought I will put the flowspec_valid() inside flowspec_add() but I changed my mind now and moved it into the imsg handler. flowspec_add() and flowspec_delete() are built

Re: bgpd remove fatalx calls from flowspec.c

On Wed, Apr 19, 2023 at 11:00:46AM +0200, Claudio Jeker wrote: > I want to use this code also in bgpctl (like util.c) and since bgpctl > has no fatalx() and "library" code should not abort. > > The comparison function can not return an error so instead sort invalid > objects in a deterministic way

bgpd remove fatalx calls from flowspec.c

I want to use this code also in bgpctl (like util.c) and since bgpctl has no fatalx() and "library" code should not abort. The comparison function can not return an error so instead sort invalid objects in a deterministic way. flowspec_cmp() should only be called on flowspec NLRI that have previou

Re: bgpd mrt support for flowspec

On Wed, Apr 19, 2023 at 10:49:38AM +0200, Claudio Jeker wrote: > Implement the minimum to support flowspec in mrt mp table dumps. > This adds a dummy nexthop encoding for flowspec (there is no nexthop). > > Not sure if this matters (dumping tables in anything than the v2 table > format for non-IPv

bgpd mrt support for flowspec

Implement the minimum to support flowspec in mrt mp table dumps. This adds a dummy nexthop encoding for flowspec (there is no nexthop). Not sure if this matters (dumping tables in anything than the v2 table format for non-IPv4/IPv6 prefixes is not common) but it prevents a log_warnx() if used. --

Re: bgpd handle flowspec in update path

On Wed, Apr 19, 2023 at 09:23:01AM +0200, Claudio Jeker wrote: > Add handling for flowspec in the update path. > > Flowspec has no nexthop so adjust up_prep_adjout() to handle a NULL nexthop. > Add the MP_REACH encoding in up_generate_mp_reach for flowspec. ok tb

bgpd handle flowspec in update path

Add handling for flowspec in the update path. Flowspec has no nexthop so adjust up_prep_adjout() to handle a NULL nexthop. Add the MP_REACH encoding in up_generate_mp_reach for flowspec. -- :wq Claudio Index: rde_update.c =