Re: add in6 multicast support to vxlan(4) ; question on mbufs

On Wed, 5 Oct 2016 14:30:27 +0200 Mike Belopuhov wrote: > On Wed, Oct 05, 2016 at 10:58 +0900, YASUOKA Masahiko wrote: >> On Tue, 4 Oct 2016 17:27:12 +0200 >> Mike Belopuhov wrote: >> > On Tue, Oct 04, 2016 at 01:07 +0200, Vincent Gross wrote: >> >> On

Re: syslogd milliseconds timestamp

> On 6 Oct 2016, at 07:07, Alexander Bluhm wrote: > > Hi, > > RFC 5424 says you should add 1 to 6 digits fractions of a second > to each syslog timestamp. As we do not measure the time in syslog(3), > it takes 100 microseconds on my laptop before syslogd(8) adds the >

syslogd milliseconds timestamp

Hi, RFC 5424 says you should add 1 to 6 digits fractions of a second to each syslog timestamp. As we do not measure the time in syslog(3), it takes 100 microseconds on my laptop before syslogd(8) adds the timestamp. So I have decided to use only 3 digits with millisecond precision. Note that

Re: hide wpi firmware error log

On Wed, Oct 05, 2016 at 10:35:50PM +0200, Stefan Sperling wrote: > I don't think any normal human being understands the lines > printed by the wpi(4) driver after 'fatal firmware error'. > > wpi0: fatal firmware error > firmware error log (count=1): > error type = "UNKNOWN" (0x0013) >

hide wpi firmware error log

I don't think any normal human being understands the lines printed by the wpi(4) driver after 'fatal firmware error'. wpi0: fatal firmware error firmware error log (count=1): error type = "UNKNOWN" (0x0013) error data = 0x0070 branch link = 0x08B60274 interrupt

Can we get rid of _SUBDIRUSE for both "all" and "regress"?

Hi, a Makefile including bsd.regress.mk cannot reasonably set both SUBDIR and REGRESS_TARGETS right now, or "make all" will first run "make regress" in all subdirs, and then run "make all" in all subdirs, because "all" depends on "regress", and both depend on _SUBDIRUSE. So all the SUBDIR tests

Re: add in6 multicast support to vxlan(4) ; question on mbufs

On Wed, Oct 05, 2016 at 20:36 +0200, Mike Belopuhov wrote: > On Wed, Oct 05, 2016 at 14:30 +0200, Mike Belopuhov wrote: > > On Wed, Oct 05, 2016 at 10:58 +0900, YASUOKA Masahiko wrote: > > > On Tue, 4 Oct 2016 17:27:12 +0200 > > > Mike Belopuhov wrote: > > > > On Tue, Oct 04,

Re: Help me testing the netlock

panic: rw_enter: netlock locking against myself TIDPID UIDPRFLAGS 12705 12705 1041 0x23 *62630 62630 1000 0x32 PFLAGS CPU COMMAND 0 1chrome 0 0 Xorg panic() rw_enter() rw_enter_write() rw_enter_timer() timeout_run() softlock()

Re: add in6 multicast support to vxlan(4) ; question on mbufs

On Wed, Oct 05, 2016 at 14:30 +0200, Mike Belopuhov wrote: > On Wed, Oct 05, 2016 at 10:58 +0900, YASUOKA Masahiko wrote: > > On Tue, 4 Oct 2016 17:27:12 +0200 > > Mike Belopuhov wrote: > > > On Tue, Oct 04, 2016 at 01:07 +0200, Vincent Gross wrote: > > >> On Sat, 24 Sep 2016

Re: httpd(8): dup2() fix for proc.c

On Wed, Oct 05, 2016 at 06:56:54PM +0200, Rafael Zalamena wrote: > This diff fixes the same problem ntpd(8) had with the dup2() when oldd == > newd. > > Quick background: > when you dup2(oldd, newd) and oldd == newd the CLOEXEC flag won't be removed > by the descriptor. We could use dup3() to

Re: syslogd fork+exec

On Sun, Oct 02, 2016 at 11:07:21PM +0200, Alexander Bluhm wrote: > On Sat, Oct 01, 2016 at 07:41:13PM +0200, Rafael Zalamena wrote: > > This could be replaced with "closefrom(4);". > > Updated diff: > - use closefrom(2) > - use execvp(3) to allow starting syslogd(8) without full path > - add a

Re: httpd(8): dup2() fix for proc.c

On Wed, Oct 05, 2016 at 06:56:54PM +0200, Rafael Zalamena wrote: > This diff fixes the same problem ntpd(8) had with the dup2() when oldd == > newd. > > Quick background: > when you dup2(oldd, newd) and oldd == newd the CLOEXEC flag won't be removed > by the descriptor. We could use dup3() to

httpd(8): dup2() fix for proc.c

This diff fixes the same problem ntpd(8) had with the dup2() when oldd == newd. Quick background: when you dup2(oldd, newd) and oldd == newd the CLOEXEC flag won't be removed by the descriptor. We could use dup3() to detect this, but it is easier/faster just to compare the fds and do the fcntl()

Re: timeout_set_proc(9)

On 5 October 2016 at 18:26, Ted Unangst wrote: > Christiano F. Haesbaert wrote: >> There is another bug, the thread runs outside of IPL_SOFTCLOCK, the >> interrupt handler already runs at IPL_SOFTCLOCK so it did not need to raise >> it, but the thread does. >> >> The mutex is

Re: timeout_set_proc(9)

Christiano F. Haesbaert wrote: > There is another bug, the thread runs outside of IPL_SOFTCLOCK, the > interrupt handler already runs at IPL_SOFTCLOCK so it did not need to raise > it, but the thread does. > > The mutex is not enough as it will drop before running the handler, this > can cause

wsfont.c: remove reference to non existing font

Hi tech@, Include file with font data (courier11x18.h) was removed from NetBSD due to licensing concerns [1] before wsfont was imported into OpenBSD. I renumbered the cookie values for consistency, and verified that no code in tree used hardcoded cookie values. Comments? OK? [1]

Re: timeout_set_proc(9)

Am Montag, 26. September 2016 schrieb David Gwynne : > > > On 26 Sep 2016, at 13:36, Ted Unangst > wrote: > > > > David Gwynne wrote: > >> +mtx_enter(_mutex); > >> +while (!CIRCQ_EMPTY(_proc)) { > >> +to =

Re: regress for etherip and vxlan

On Wed, Oct 05, 2016 at 04:48:55PM +0900, YASUOKA Masahiko wrote: > The diff add regress scripts for vxlan(4) and etherip(4). > > This will be my first commit to regress/. ok? I does not run with an obj directory. root@ot1:.../etherip# make obj /usr/src/regress/sys/net/etherip/obj ->

Re: add in6 multicast support to vxlan(4) ; question on mbufs

On Wed, Oct 05, 2016 at 10:58 +0900, YASUOKA Masahiko wrote: > On Tue, 4 Oct 2016 17:27:12 +0200 > Mike Belopuhov wrote: > > On Tue, Oct 04, 2016 at 01:07 +0200, Vincent Gross wrote: > >> On Sat, 24 Sep 2016 10:58:10 +0200 > >> Vincent Gross wrote: > >> >

Outdated remarks about partitioning on sgi

With the disklabel read logic in the sgi bootblocks, the 'a' partition no longer has to be the first partition on a disk, so the positioning remarks in the install notes and the installer seem outdated. The patch below removes them. OK? Index: distrib/notes/sgi/install

Re: Smarter OpenBSD/sgi boot blocks

On Tue, Oct 04, 2016 at 08:08:32PM +, Miod Vallat wrote: > The sgi boot blocks use the PROM (ARCBios or ARCS) for its I/O routines. > When using a disk-based path, these routines are using the partition > table found in the ``volume header''. > > In order to be able to use 16 partitions per

vmd: virtual switches and enhanced interface configuration

Hi, the following diff enhances vmd(8)'s network configuration. Before, you could only set "interfaces N" or -i N for the number of desired interfaces. The diff introduces the concept of virtual switches: bridge(4) or switch(4) devices that are partially managed by vmd(8) with the following

Re: Help me testing the netlock

On 10/04/16 16:44, Martin Pieuchot wrote: On 10/03/16 16:43, Martin Pieuchot wrote: Diff below introduces a single write lock that will be used to serialize access to ip_output(). This lock will be then split in multiple readers and writers to allow multiple forwarding paths to run in parallel

regress for etherip and vxlan

The diff add regress scripts for vxlan(4) and etherip(4). This will be my first commit to regress/. ok? diff --git a/regress/sys/net/Makefile b/regress/sys/net/Makefile index 40f49cc..f46c3dd 100644 --- a/regress/sys/net/Makefile +++ b/regress/sys/net/Makefile @@ -1,5 +1,6 @@ # $OpenBSD:

Re: Help me testing the netlock

On Tue, Oct 04, 2016 at 04:44:29PM +0200, Martin Pieuchot wrote: > On 10/03/16 16:43, Martin Pieuchot wrote: > > Diff below introduces a single write lock that will be used to serialize > > access to ip_output(). > > > > This lock will be then split in multiple readers and writers to allow > >