videomode.h: add int hskew member to struct videomode

Some ARM systems have decoupled their display/LCD controllers from their display transmitters, such as the am335x/beaglebone black whose LCD controller (on am335x SoC) breaks out to a TDA19988 IC (on PCB). This sometimes causes sync timing problems in that the timings purported by the LCD

if_iwmreg.h enums

This gets rid of almost all the silly enums in iwm's header file. Replace them with #define to better match the style of other drivers. No functional change intended. This conversion was semi-automatic and additional eyeballing is very much appreciated. Index: if_iwm.c

Re: Kernel panic pf.c during halting

> > > The key is really being able to reproduce the problem, Lampshade do you > > already know which service or config triggers this panic? Could you try > > to figure out by simplifying your setup? > > I don't know. > Problem is that even with most complicated config it happens a few times >

Remove mention of deprecated flags in ping6.8

I don't think the -s option is ever ignored now. OK? Index: sbin/ping6/ping6.8 === RCS file: /cvs/src/sbin/ping6/ping6.8,v retrieving revision 1.60 diff -u -p -r1.60 ping6.8 --- sbin/ping6/ping6.8 25 Oct 2015 14:43:48 -

Re: send fewer router solicitations

On Fri, Sep 02, 2016 at 05:49:22PM +0100, Stuart Henderson wrote: > On 2016/09/02 10:37, Florian Obser wrote: > > To stop naddy from pestering me about this at every hackathon (rightly > > so!), let's base the timeout on the prefixes pltime. ;) > > Just a thought, are we going to need to cap this

Periodically sync RTC

I would like to sync the system time periodically back to the RTC. Currently we update the RTC (1) when the time is set with clock_settime() or settimeofday(), which never happens for a typical ntpd setup; (2) before suspend; (3) when the system is properly shut down. This means if a machine

Re: Mention the maximum packet size in ping.8

Michal Mazurek(akf...@jasminek.net) on 2016.09.02 18:41:27 +0200: > Index: sbin/ping/ping.8 > === > RCS file: /cvs/src/sbin/ping/ping.8,v > retrieving revision 1.52 > diff -u -p -r1.52 ping.8 > --- sbin/ping/ping.8 24 Mar 2014

Re: Mention the maximum packet size in ping.8

Michal Mazurek wrote: whee! > Index: sbin/ping/ping.8 > === > RCS file: /cvs/src/sbin/ping/ping.8,v > retrieving revision 1.52 > diff -u -p -r1.52 ping.8 > --- sbin/ping/ping.8 24 Mar 2014 11:11:49 - 1.52 > +++

Re: mbuf cluster and socket buffer limits

On Fri, Sep 2, 2016 at 6:59 PM, Alexander Bluhm wrote: > Hi, > > To move our network performance to modern high bandwith and high > latency characteristics, we have to increase the socket buffer size > limit. That also implies more mbuf clusters to avoid running out > of

Re: mbuf cluster and socket buffer limits

On Fri, Sep 02, 2016 at 06:59:21PM +0200, Alexander Bluhm wrote: > Hi, > > To move our network performance to modern high bandwith and high > latency characteristics, we have to increase the socket buffer size > limit. That also implies more mbuf clusters to avoid running out > of them. > >

relayd fork+exec

Hi, after all the preparation, the following diff adds support for fork+exec in relayd; based on rzalamena@'s work for httpd. Notes: - proc.c is identical to the version that is already in httpd. - The recvfd pledge in some procs is only needed once, could be dropped later after receiving

Re: NAT-on-enc on iked(8)

On 1 September 2016 at 13:35, Vincent Gross wrote: > This diff adds the missing bits to support NAT-on-enc in iked(8). > > See OUTGOING NETWORK ADDRESS TRANSLATION in iked.conf(5), and also > http://undeadly.org/cgi?action=article=20090127205841. > > Ok ? > OK mikeb

mbuf cluster and socket buffer limits

Hi, To move our network performance to modern high bandwith and high latency characteristics, we have to increase the socket buffer size limit. That also implies more mbuf clusters to avoid running out of them. This diff includes several aspects that somehow belong together. - Increase the

Re: send fewer router solicitations

On 2016/09/02 10:37, Florian Obser wrote: > To stop naddy from pestering me about this at every hackathon (rightly > so!), let's base the timeout on the prefixes pltime. ;) Just a thought, are we going to need to cap this to the RDNSS time too when we start caring about that?

Mention the maximum packet size in ping.8

Index: sbin/ping/ping.8 === RCS file: /cvs/src/sbin/ping/ping.8,v retrieving revision 1.52 diff -u -p -r1.52 ping.8 --- sbin/ping/ping.824 Mar 2014 11:11:49 - 1.52 +++ sbin/ping/ping.82 Sep 2016 16:35:50 - @@

Re: refactor gzip methods

On Fri, 02 Sep 2016 12:02:20 -0400, "Ted Unangst" wrote: > this diff doesn't change any behavior, but creates separate read > and write open functions (which currently share quite a bit of code) > that only take the necessary arguments. > > it also deletes a never used zopen function. and

new style signing coming soon

I've been working on a new way to sign archives, in preparation to a change to pkg tools, with feedback from Theo and Tedu. The actual problem happened in FreeBSD. They got a hole in their gunzip pipeline a few weeks ago (rather subtle one). So if your workflow is: 1/ fetch data -> 2/ uncompress

Re: Better wording in ping error messages

On 12:12:11, 2.09.16, Ted Unangst wrote: > Michal Mazurek wrote: > > worms(6) does this more concisely: > > > > errx(1, "length (2-1024) is %s: %s", errstr, > > there's multiple schools of thought here, but i've always preferred somewhat > shorter error messages. we tell the user it's too

Re: Better wording in ping error messages

Michal Mazurek wrote: > On 12:12:11, 2.09.16, Ted Unangst wrote: > > Michal Mazurek wrote: > > > worms(6) does this more concisely: > > > > > > errx(1, "length (2-1024) is %s: %s", errstr, > > > > there's multiple schools of thought here, but i've always preferred somewhat > > shorter error

Re: constify auth functions

Todd C. Miller wrote: > On Thu, 01 Sep 2016 13:27:49 -0400, "Ted Unangst" wrote: > > > Ted Unangst wrote: > > > I'm getting annoying warnings because these functions aren't const. > > > > so > > > > turns out several of these functions *do* modify their parameters. which is > > probably

Re: Better wording in ping error messages

Michal Mazurek wrote: > worms(6) does this more concisely: > > errx(1, "length (2-1024) is %s: %s", errstr, there's multiple schools of thought here, but i've always preferred somewhat shorter error messages. we tell the user it's too small or too large already. if they need to know the

Better wording in ping error messages

worms(6) does this more concisely: errx(1, "length (2-1024) is %s: %s", errstr, Index: ping.c === RCS file: /cvs/src/sbin/ping/ping.c,v retrieving revision 1.142 diff -u -p -r1.142 ping.c --- ping.c 30 Aug 2016

refactor gzip methods

the code in gzip uses a function pointer abstraction but tries to shove too much code into the same functions. this is impeding work to refactor the code to fork + pledge helper processes. in many cases, the duplication results absurd code like this: error = (decomp ? dodecompress :

acpixtu: Intel Extreme Tuning driver

Hi, Here is a work-in-progress driver for the Intel eXtreme Tuning Utility. The main purpose is to add more sensors to the existing pool. I am very interested in tests on recent Intel K- and X-series processors! None the less, please test on all Intel cpus and if you see acpixtu attaching please

Re: mg(1) Initialize stack variables to zero before memmove()

On Fri, 02 Sep 2016 07:10:38 -, Mark Lumsden wrote: > Source Joachim Nilsson: > > Coverity Scan reported these two stack variables as uninitialized, in > particular the .r_lineno struct member was uninitialized. This patch > clears the 'struct region' rather than setting

Re: Diff: Fix wrong bzero call in snmpd

On Fri, Sep 02, 2016 at 16:56 +0200, Jan Klemkow wrote: > Hi, > > When a new IP address shows up on an interface, the function > if_newaddr() save this IP address and its mask and boardcast address > into a data structure. If the corresponding IP mask or broad cast > address is not set, their

Diff: Fix wrong bzero call in snmpd

Hi, When a new IP address shows up on an interface, the function if_newaddr() save this IP address and its mask and boardcast address into a data structure. If the corresponding IP mask or broad cast address is not set, their data fields in this structure are bzero()-ed. But, the code wipes the

Re: switch the cubie miniroot to cubieboard2

- Original Message - > si...@slackware.it [si...@slackware.it] wrote: > > Speaking as a Cubieboard owner here ;-) > > Would it be too much hassle to provide both images? (and a pony!) > > > > It's fairly easy to take a miniroot image for a similar board, and > adapt it to your board.

Re: Futexes for OpenBSD

The new files should have the ISC/OpenBSD license inserted at the top. IMHO, the ticket changes are a separate diff and you will be able to commit that part first. Also, replacing struct _spinlock with int is also a separate diff. Those changes don't clash with futexes, so less chance of being

acpi: remove unused argument in aml_showvalue()

Hi, This diff removes the second argument of aml_showvalue which was probably supposed to set the verbosity level through dnprintf() but in fact it does nothing. OK? Index: acpicpu.c === RCS file: /cvs/src/sys/dev/acpi/acpicpu.c,v

Futexes for OpenBSD

Here is a working futex implementation for OpenBSD. This diff touches the kernel and librthread. * get rid of tickets from rthreads, they were getting in the way and are unused anyway * replace all struct _spinlock with int * use futexes instead of spinlocks everywhere within librthread *

trade the rb tree in pools for subr_tree code

its not quite an even trade, we lose some bytes still cos the subr_tree code is a bit bigger than what RB_GENERATE produced. we'll be ahead after the next conversion though. ok? Index: conf/files === RCS file:

Re: send fewer router solicitations

On 2016/09/02 10:37, Florian Obser wrote: > Our kernel based rtsol code is like this little child. We bring up > the interface, send our first solicitation and get an advertisment > back with a pltime of a week or so. > > We lean back, quite happy that we can do v6 now, but after 60 seconds > we

Re: installer: do proper checks for MBR/GPT and OpenBSD area

On Fri, Sep 02, 2016 at 01:57:21PM +0300, Paul Irofti wrote: > Hi, > > While reinstalling OpenBSD on the x260 I saw a weird message when > selecting the disk (that had a valid MBR) > > 'No valid MBR or GPT' > > which lead to the following installer diff. > > The idea is to do two checks

installer: do proper checks for MBR/GPT and OpenBSD area

Hi, While reinstalling OpenBSD on the x260 I saw a weird message when selecting the disk (that had a valid MBR) 'No valid MBR or GPT' which lead to the following installer diff. The idea is to do two checks (similar to i386): first see if there's an MBR or GPT present on the disk, and if

Re: Kernel panic pf.c during halting

Another crash. I should note that this kernel is build by me with patches to GENERIC (HZ from 100 to 300), pci and acpi files. Both previous reports were from official kernel builds (snapshots). Source code is based on: cvs -d$CVSROOT up -D "2016-08-26 12:45" -Pd As always with this panic I

Re: in6_selectroute should never get AF_INET filled struct route *

OK florian@ On Fri, Sep 02, 2016 at 11:21:33AM +0200, Vincent Gross wrote: > in6_selectroute() checks whether the struct route it received contains > a valid route whose AF is not AF_INET6, "in case the cache is shared". > Well, is this cache shared or not ? > > There's only two ways to get to

send fewer router solicitations

Our kernel based rtsol code is like this little child. We bring up the interface, send our first solicitation and get an advertisment back with a pltime of a week or so. We lean back, quite happy that we can do v6 now, but after 60 seconds we wake up, oh shit, better check if that prefix is

Re: Let iked specify its source address when sending

On Wed, Aug 31, 2016 at 03:57:45PM +0200, Vincent Gross wrote: > On Wed, 31 Aug 2016 15:26:53 +0200 > Vincent Gross wrote: > > > On Thu, 11 Aug 2016 16:57:27 +0100 > > Stuart Henderson wrote: > > > > > On 2016/06/27 13:00, J?r?mie Courr?ges-Anglas

in6_selectroute should never get AF_INET filled struct route *

in6_selectroute() checks whether the struct route it received contains a valid route whose AF is not AF_INET6, "in case the cache is shared". Well, is this cache shared or not ? There's only two ways to get to in6_selectroute() 1) in6_pcbselsrc() -> in6_selectif() -> in6_selectroute() It is

Re: Drop IPSec traffic that should be encapsulated but is not

On 1 September 2016 at 10:31, Vincent Gross wrote: > Our IPSec stack rejects UDP-encapsulated traffic using a non > encapsulating SA, but not the other way around. This diff adds > the missing check and the corresponding stat counter. > > Ok ? > Go for it. OK mikeb

Re: Let iked specify its source address when sending

Objections anyone ? On Wed, 31 Aug 2016 15:57:45 +0200 Vincent Gross wrote: > On Wed, 31 Aug 2016 15:26:53 +0200 > Vincent Gross wrote: > > > On Thu, 11 Aug 2016 16:57:27 +0100 > > Stuart Henderson wrote: > > > > > On

Re: arm: store curcpu pointer in thread id register

On Sat, Feb 27, 2016 at 05:03:46PM +0100, Patrick Wildt wrote: > On Fri, Feb 26, 2016 at 08:59:41PM -0800, Philip Guenther wrote: > > On Sat, Feb 20, 2016 at 1:49 PM, Patrick Wildt wrote: > > > since ARMv6 the coprocessor provides special registers to store software > > >

Re: sxidog hang

> Date: Fri, 2 Sep 2016 08:31:20 +0200 > From: Marcus Glocker > > Trying to disable the watchdog on the allwinner,sun5i-r8 breaks the > reboot ('reboot failed; spinning'). Can we change this back at least > for the A10? Sorry for dropping the ball here. But does it work if

mg(1) Initialize stack variables to zero before memmove()

Source Joachim Nilsson: Coverity Scan reported these two stack variables as uninitialized, in particular the .r_lineno struct member was uninitialized. This patch clears the 'struct region' rather than setting .r_lineno because if any more struct members are added in the future

sxidog hang

Trying to disable the watchdog on the allwinner,sun5i-r8 breaks the reboot ('reboot failed; spinning'). Can we change this back at least for the A10? Index: sxidog.c === RCS file: /cvs/src/sys/arch/armv7/sunxi/sxidog.c,v retrieving