Re: IPv6 hop-by-hop check in parallel

2022-08-13 Thread Vitaliy Makkoveev
On Sat, Aug 13, 2022 at 09:08:36PM +0200, Alexander Bluhm wrote: > Hi, > > While running forwarding in parallel, I have introduced a hard > barrier for parallel local protocol processing. The packets are > requeued from shared to exclusive netlock. > > Unless we unlock all protocol input

Re: Split (*pr_usrreq)() by multiple handlers

2022-08-13 Thread Vitaliy Makkoveev
On Sat, Aug 13, 2022 at 09:28:46PM +0200, Alexander Bluhm wrote: > On Sat, Aug 13, 2022 at 09:58:16PM +0300, Vitaliy Makkoveev wrote: > > I propose to commit pru_*() hunks first. > > OK bluhm@ > The pr_usrreqs hunks. I replaced spaces by tab within `routesw' structure. > These are two

Re: Split (*pr_usrreq)() by multiple handlers

2022-08-13 Thread Alexander Bluhm
On Sat, Aug 13, 2022 at 09:58:16PM +0300, Vitaliy Makkoveev wrote: > I propose to commit pru_*() hunks first. OK bluhm@ > Index: sys/kern/sys_socket.c > === > RCS file: /cvs/src/sys/kern/sys_socket.c,v > retrieving revision 1.51 >

IPv6 hop-by-hop check in parallel

2022-08-13 Thread Alexander Bluhm
Hi, While running forwarding in parallel, I have introduced a hard barrier for parallel local protocol processing. The packets are requeued from shared to exclusive netlock. Unless we unlock all protocol input routines at once, we need some mechanism from to move from one queue to the other.

Re: Split (*pr_usrreq)() by multiple handlers

2022-08-13 Thread Vitaliy Makkoveev
On Sat, Aug 13, 2022 at 12:04:57PM +0200, Alexander Bluhm wrote: > On Mon, Aug 08, 2022 at 03:48:57PM +0300, Vitaliy Makkoveev wrote: > > We use if (error != 0) or if (m == NULL) idioms in the network stack, so > > I used them too in the trivial places. > > I nearly got them consistent in

Re: rpki-client: add ASPA profile-10 support

2022-08-13 Thread Theo Buehler
On Sat, Aug 13, 2022 at 01:06:38AM +, Job Snijders wrote: > Dear all, > > The below changeset extends rpki-client(8) to validate and emit > Autonomous System Provider Authorizations (ASPAs) in JSON format for > consumption in other routing stacks. > > The ASPA CMS protected content type is

Re: unhook mount_tmpfs?

2022-08-13 Thread Theo de Raadt
Because there is a hope someone will fix it. Solène Rapenne wrote: > hi > > is there a reason to keep mount_tmpfs available while it's deactived in > the kernel? > > Index: sbin/Makefile > === > RCS file: /cvs/src/sbin/Makefile,v

Re: rpki-client: disallow inherit in ROA EE IP Resources extension

2022-08-13 Thread Theo Buehler
On Fri, Aug 12, 2022 at 09:59:11PM +0200, Theo Buehler wrote: > On Wed, Aug 10, 2022 at 06:16:30PM +0200, Theo Buehler wrote: > > On Wed, Aug 10, 2022 at 03:10:19PM +, Job Snijders wrote: > > > Hi all, > > > > > > An errata exists for RFC 6482, which informs us: """The EE certificate > > >

Re: switch tcpbench to use TCP_INFO instead of kvm

2022-08-13 Thread Alexander Bluhm
On Sat, Aug 13, 2022 at 10:16:43AM +0200, Claudio Jeker wrote: > This diff changes tcpbench to use the TCP_INFO getsockopt to fetch > all the extra information instead of kvm. > The variable names change because of this. OK bluhm@ > - if (pledge("stdio unveil rpath dns inet unix id proc",

Re: Split (*pr_usrreq)() by multiple handlers

2022-08-13 Thread Alexander Bluhm
On Mon, Aug 08, 2022 at 03:48:57PM +0300, Vitaliy Makkoveev wrote: > We use if (error != 0) or if (m == NULL) idioms in the network stack, so > I used them too in the trivial places. I nearly got them consistent in netinet/netinet6 $ grep 'if (error)' netinet*/* | wc 106 347

switch tcpbench to use TCP_INFO instead of kvm

2022-08-13 Thread Claudio Jeker
This diff changes tcpbench to use the TCP_INFO getsockopt to fetch all the extra information instead of kvm. The variable names change because of this. -- :wq Claudio Index: Makefile === RCS file: