Re: malloc canaries for > page sized objects

On Sun, Oct 16, 2016 at 01:14:35PM +0200, Otto Moerbeek wrote: > Hi, > > this diff is somewhat big since I decided to rewrite wrterror() to be > able to get better error messages. > > Please review and test this with malloc option C (and other flags). An > example run: > > a.out(85360) in

Re: let PF to send challenge ack

On Fri, Sep 30, 2016 at 11:55:48PM +0200, Alexandr Nedvedicky wrote: > The patch makes PF to send 'challenge ACK' for SYN packet, which matches > session in established state. The pf diff is OK bluhm@ > The patch also comes with test case. I've just learned it's bit tricky > to use scapy for

Re: splassert with if_run

On Mon, Oct 17, 2016 at 08:56:09PM +0200, egorenar-...@posteo.net wrote: > i recently updated from 6.0 to a snapshot and > since then i see splassert sometimes in my dmesg when i use if_run USB > network device. ... > splassert_check() at splassert_check+0x78 > sowakeup() at sowakeup+0x2c >

pf afto icmp checksum

Hi, The checksum of a ICMP "need to frag" packet for TCP was wrong when created from a ICMP6 "too big" packet. The function pf_change_icmp_af() has code to adjust the pseudo-header checksum in the ICMP6 case, but pf_test_state_icmp() changed the proto before the case was entered. So call

splassert with if_run

Hi, i recently updated from 6.0 to a snapshot and since then i see splassert sometimes in my dmesg when i use if_run USB network device. -- Regards Alexander Egorenkov OpenBSD 6.0-current (obj) #0: Sat Oct 15 18:53:54 CEST 2016

cc -Wshadow: don't warn about variable-vs-global-function

On Sun, 16 Oct 2016, Philip Guenther wrote: > So let's fix that and make our gcc a bit more like new ones. Written > without peeking at the new ones and tested against the .c file at bottom > to verify that it doesn't fail or crash on some weird combo of > shadowing. > > oks? Updated diff

Re: ksh vi mode: fix region deletion and yanking

Hi Theo, Theo Buehler wrote on Fri, Oct 14, 2016 at 09:09:34PM +0200: > The patch reads fine to me and behaves well in some testing. > It is ok tb@ as is. Committed, thanks for checking. > I wonder if for the sake of consistency it would be worth merging the > 'b' and 'B' cases as well as the

Re: FAQ entry for vmm

Hi Reyk, On Mon, Oct 17, 2016 at 06:20:25PM +0200, Reyk Floeter wrote: > Sorry for the late reply - thanks for doing this. No objections, OK > from me as well. Thanks. The general consensus was to try to reduce this a lot before commit though. I will probably only include the wireless setup in

Re: FAQ entry for vmm

On Sat, Oct 15, 2016 at 05:11:49PM +0100, Edd Barrett wrote: > Hi, > > Here's an intial stab at a FAQ entry for vmm. > > It covers two common setups: > * a vmm guest with network access via the host's wired network > * a vmm guest with network access via the host's wireless network > >

Re: vmd: simplify fatal/fatalx errno handling

On Sun, Oct 16, 2016 at 02:55:39PM -0700, Philip Guenther wrote: > > Instead of using errno as a hidden argument to vfatal(), make it an > _actual_ argument named 'code', ala the errc/warnc family, and rename it > to vfatalc() to match the pattern set. > > ok? > OK, makes sense please sync

Re: vmd: simplify fatal/fatalx errno handling

On Sun, Oct 16, 2016 at 08:03:05PM -0600, Theo de Raadt wrote: > > On Sun, Oct 16, 2016 at 02:55:39PM -0700, Philip Guenther wrote: > > > > > > Instead of using errno as a hidden argument to vfatal(), make it an > > > _actual_ argument named 'code', ala the errc/warnc family, and rename it > >

Re: minor diff for ldapd.conf.5

Rob Pierce writes: > Fix a couple of grammar mistakes, remove a redundant word, and add a FILES > reference for the /etc/ldap/certs directory. Committed, thanks. I adjusted the FILES entries so that ldapd.conf is still listed first, and so that the description of /etc/ldap/certs

Re: ffs/msdosfs: Flush cache when updating mount to R/O

On Sun, Oct 16, 2016 at 10:24:06AM +0200, Stefan Fritsch wrote: > On Sun, 16 Oct 2016, Stefan Fritsch wrote: > > > * When a R/W mount is updated to R/O. I will send patches for this in a > > > separate mail. > > Part 2: Use it > > > msdosfs & ffs: flush cache if updating mount to r/o > >

Re: Add ioctl for disk cache flush

On Sun, Oct 16, 2016 at 10:19:53AM +0200, Stefan Fritsch wrote: > On Sun, 16 Oct 2016, Stefan Fritsch wrote: > > * When a R/W mount is updated to R/O. I will send patches for this in a > > separate mail. > > Part 1: Add an ioctl: > > > add DIOCCACHESYNC > > Add an ioctl to tell storage

Re: vmd: simplify fatal/fatalx errno handling

Theo de Raadt(dera...@openbsd.org) on 2016.10.16 20:03:05 -0600: > > On Sun, Oct 16, 2016 at 02:55:39PM -0700, Philip Guenther wrote: > > > > > > Instead of using errno as a hidden argument to vfatal(), make it an > > > _actual_ argument named 'code', ala the errc/warnc family, and rename it >

Re: landisk: simply asid lookup

Philip Guenther writes: > It used to be that access to a process's vmspace (and thus its pmap where > the ASID can be found) was only via a thread's p_vmspace member. Now that > the process itself has a copy of that, the loops in the sh code that > iterated across threads

opencvs - show branch revision in status

Hi, Let's bring status a bit more inline with its GNU cvs counter part. This diff adds the branch revision for the sticky tag if set. .joris Index: status.c === RCS file: /cvs/src/usr.bin/cvs/status.c,v retrieving revision 1.96

Re: ifconfig(8): fix set switch(4) datapath id

> Am 17.10.2016 um 14:33 schrieb Rafael Zalamena : > >> On Mon, Oct 17, 2016 at 02:30:41PM +0400, Reyk Floeter wrote: >> >>> Am 17.10.2016 um 14:16 schrieb Rafael Zalamena : >>> >>> There are two inconsistencies with the ifconfig(8) switch(4)

Re: ifconfig(8): fix set switch(4) datapath id

> Am 17.10.2016 um 14:16 schrieb Rafael Zalamena : > > There are two inconsistencies with the ifconfig(8) switch(4) configuring: > 1) Datapath ID is an unsigned 64 bit integer, not a signed one; I think we should use strtonum instead of strtoull here. > 2) ifconfig(8) man

ifconfig(8): fix set switch(4) datapath id

There are two inconsistencies with the ifconfig(8) switch(4) configuring: 1) Datapath ID is an unsigned 64 bit integer, not a signed one; 2) ifconfig(8) man pages says that the parameter is "datapath" not "datapathid"; This diff fixes both problems and let us configure the datapath id

Re: make obj in config

On Sun, Oct 16, 2016 at 11:52:45AM +0200, Martin Natano wrote: > We don't need to re-run 'make obj' when obj exists. Ok? > > natano > > > Index: main.c > === > RCS file: /cvs/src/usr.sbin/config/main.c,v > retrieving revision 1.55

Re: use x2apic if it is enabled by BIOS

> Date: Sat, 15 Oct 2016 09:55:05 +0200 (CEST) > From: s...@openbsd.org > > On Fri, 14 Oct 2016, Mike Larkin wrote: > > > On Fri, Oct 14, 2016 at 04:49:31PM +0900, YASUOKA Masahiko wrote: > > > Hi, > > > > > > I'm working on NEC Express5800/R110h-1 (dmesg is attached). On this > > > machine,

sparc64: delete old "traptrace" bits

In 2003(!) a pile of code was deleted from sparc64's locore.s: revision 1.35 date: 2003/05/17 07:09:08; author: art; state: Exp; lines: +1 -957; Get rid of lots of hairy ifdefs that we'll most likely never use. TRAPTRACE, TRAPSTATS, FLTTRACE and SCHED_DEBUG. mdw@