Re: rtfree(): "rt->rt_refcnt > 0" assertion

On Sat, Oct 02, 2021 at 11:27:48AM +0200, Martin Pieuchot wrote: > On 15/09/21(Wed) 01:23, Vitaliy Makkoveev wrote: > > We have weird `rt_refcnt' check in rtfree(): > > > > > 497 rtfree(struct rtentry *rt) > > 498 { > > ... > > 504 refcnt = (int)atomic_dec_int_nv(&rt->rt_refcnt); > >

Re: forwarding in parallel ipsec workaround

Hrvoje Popovski [hrv...@srce.hr] wrote: > > box didn't panic, just stopped forwarding traffic through tunnel. any chance any progress has been made here? is there any newer versions of these diffs floating around?

Re: Variable type fix in parse.y (all of them)

Here's another attempt, incorporating millert's feedback and adding a few more casts: Index: bin/chio/parse.y === RCS file: /cvs/src/bin/chio/parse.y,v retrieving revision 1.23 diff -u -p -r1.23 parse.y --- bin/chio/parse.y15 Oct

Re: mi_switch() & setting `p_stat'

On Sat, Oct 2, 2021 at 8:57 AM Martin Pieuchot wrote: > When a thread running on a CPU schedules itself out, it does the following > (pseudo_code): > > SCHED_LOCK() > curproc->p_stat = SSLEEP; > // some more operations > mi_switch() > > The problem with this is tha

Re: mi_switch() & setting `p_stat'

> Date: Sat, 2 Oct 2021 20:35:41 +0200 > From: Martin Pieuchot > > On 02/10/21(Sat) 20:24, Mark Kettenis wrote: > > > Date: Sat, 2 Oct 2021 19:55:49 +0200 > > > From: Martin Pieuchot > > > > > > When a thread running on a CPU schedules itself out, it does the following > > > (pseudo_code): > >

Re: Relayd daily crash ca_dispatch_relay invalid

abyx...@mnetic.ch(abyx...@mnetic.ch) on 2021.10.01 09:56:32 -0400: > On Fri, Oct 1, 2021, at 09:44, Stuart Henderson wrote: > > On 2021/10/01 14:43, Stuart Henderson wrote: > >> On 2021/10/01 09:29, abyx...@mnetic.ch wrote: > >> > I'm getting a daily crash (call to fatalx). No clue what triggers it

Re: mi_switch() & setting `p_stat'

On 02/10/21(Sat) 20:24, Mark Kettenis wrote: > > Date: Sat, 2 Oct 2021 19:55:49 +0200 > > From: Martin Pieuchot > > > > When a thread running on a CPU schedules itself out, it does the following > > (pseudo_code): > > > > SCHED_LOCK() > > curproc->p_stat = SSLEEP; > > // some more op

Re: mi_switch() & setting `p_stat'

> Date: Sat, 2 Oct 2021 19:55:49 +0200 > From: Martin Pieuchot > > When a thread running on a CPU schedules itself out, it does the following > (pseudo_code): > > SCHED_LOCK() > curproc->p_stat = SSLEEP; > // some more operations > mi_switch() > > The problem with this i

mi_switch() & setting `p_stat'

When a thread running on a CPU schedules itself out, it does the following (pseudo_code): SCHED_LOCK() curproc->p_stat = SSLEEP; // some more operations mi_switch() The problem with this is that any instrumentation between setting `p_stat' and cpu_switchto() is inc

Re: [PATCH] usr.sbin/ldapd: Match bind DN by suffix instead of complete DN.

On Sat Oct 2, 2021 at 6:36 PM CEST, Raf Czlonka wrote: > On Sat, Oct 02, 2021 at 02:15:53PM BST, vifino wrote: > > Index: ldapd.conf.5 > > === > > RCS file: /cvs/src/usr.sbin/ldapd/ldapd.conf.5,v > > retrieving revision 1.27 > > diff -

Re: [PATCH] usr.sbin/ldapd: Match bind DN by suffix instead of complete DN.

On Sat, Oct 02, 2021 at 02:15:53PM BST, vifino wrote: > Index: ldapd.conf.5 > === > RCS file: /cvs/src/usr.sbin/ldapd/ldapd.conf.5,v > retrieving revision 1.27 > diff -u -p -u -p -r1.27 ldapd.conf.5 > --- ldapd.conf.5 24 Jun 2020

[PATCH] usr.sbin/ldapd: Match bind DN by suffix instead of complete DN.

Hello. While using ldapd(8), I noticed that the allow/deny rules match the bound DN in a (to me) inconvenient way. Assuming I had a subtree ou=humans,dc=example,dc=org, users in that subtree and I wanted to allow all users to get read access to other users info, I would have to create a rule for

Re: Alternate cpu policy on battery

Hello, I've been using this diff for a week now. I didn't encounter any issue. As my laptop has a quite old battery, I noticed improvements : I can work 30min longer on battery now. The systemp is a bit slower though (ie libreoffice longer to start), but that was expected and I must say I didn't

Re: patch: remove dead variable from sys___realpath()

On Sat, Oct 02, 2021 at 12:31:29PM +0200, Sebastien Marie wrote: > Hi, > > The following diff removes a dead variable `c' which is not used. > > it is a leftover from LOCKPARENT removal in NDINIT(). > > See: > - > https://github.com/openbsd/src/commit/0eff3f09cea339207a839f716e75649e24b667b9 >

patch: remove dead variable from sys___realpath()

Hi, The following diff removes a dead variable `c' which is not used. it is a leftover from LOCKPARENT removal in NDINIT(). See: - https://github.com/openbsd/src/commit/0eff3f09cea339207a839f716e75649e24b667b9 - https://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sys/kern/vfs_syscalls.c.diff?r1=1.

Re: rtfree(): "rt->rt_refcnt > 0" assertion

On 15/09/21(Wed) 01:23, Vitaliy Makkoveev wrote: > We have weird `rt_refcnt' check in rtfree(): > > 497 rtfree(struct rtentry *rt) > 498 { > ... > 504 refcnt = (int)atomic_dec_int_nv(&rt->rt_refcnt); > 505 if (refcnt <= 0) { > 506 KASSERT(!ISSET(rt->rt_flags,

patch: vfs: merge *_badop to vop_generic_badop

Hi, The following diff was suggested by mpi@ some months ago. It replaces spec_badop, fifo_badop, dead_badop and mfs_badop, which are only calls to panic(9), to one unique function vop_generic_badop(). No intented behaviour changes (outside the panic message which isn't the same). Comments or O

Switch to kqueue based select(2)

As discussed during k2k21 I'd like to switch to the new select(2) implementation early during this release cycle. I'd like to first make sure there's no regression for select(2) and poll(2) then work towards improving the latency and removing the contention on those syscalls. This has been large