tar -N for numeric

i just discovered tar has a really sweet "feature" where it takes the name of a user or group in the archive and uses that to guess what uid and gid to create the file as when using p. it turned out this was not at all what i wanted, since i was processing the tar files on a system that had di

Re: wihap_data_input -> wi_start -> panic: m_copydata: len -7 < 0

Previous history is here: http://www.mail-archive.com/tech@openbsd.org/msg02735.html On Fri, Nov 19, 2010 at 11:19 PM, Greg Steuck wrote: > Believe it or not, I just got this reproduced with a patch to print > the mbuf (courtesy of Claudio). > > Running 4.8-sparc64 with the 2 patches below. Here

Re: wihap_data_input -> wi_start -> panic: m_copydata: len -7 < 0

Believe it or not, I just got this reproduced with a patch to print the mbuf (courtesy of Claudio). Running 4.8-sparc64 with the 2 patches below. Here's the output. Maybe this will give somebody an idea about why such weird mbufs are making it to wi_start. This one is the real offender that would

sys/types.h in sys/socket.h

openbsd is apparently among the last operating systems to require sys/types.h before sys/socket.h. posix doesn't require this and it runs contrary to current recommendations i think, so it's just one more weird thing to deal with when trying to get something to compile. i haven't really tested

too much stack use in localtime.c

I don't know if this was caused by an update, or just by nobody ever using time functions in a thread, but this union is too big for threaded stacks. Index: time/localtime.c === RCS file: /home/tedu/cvs/src/lib/libc/time/localtime.c,

posix mistake is here to stay

Speed up libc compiles by not copying a useless string into the command line of every cc invocation. :) Index: Makefile.inc === RCS file: /home/tedu/cvs/src/lib/libc/regex/Makefile.inc,v retrieving revision 1.6 diff -u -r1.6 Makefile

Re: fdisk(8) LBA only option

On Fri, Nov 19, 2010 at 09:24:37PM +, Miod Vallat wrote: > When editing MBR partitions under fdisk(8), you always get asked whether > you want to edit the MBR in C/H/S or LBA mode. > > On modern non-x86 platforms using MBR-style partitions, C/H/S doesn't > make any sense. What about adding a `

acpibat(4) diff to report charge/discharge rate neatly

Hello, t...@! Due to ACPI spec, _BST method returns the rate value in milliwats or milliamperes according to the unit reported by _BIF. Tweak acpibat so sysctl sees the data like this: hw.sensors.acpibat0.power0=14.95 W (rate) Index: acpibat.c

fdisk(8) LBA only option

When editing MBR partitions under fdisk(8), you always get asked whether you want to edit the MBR in C/H/S or LBA mode. On modern non-x86 platforms using MBR-style partitions, C/H/S doesn't make any sense. What about adding a `always use LBA' option to fdisk(8) (to later be used in the installatio

Re: crontab.5 diff

On Fri, Nov 19, 2010 at 06:12:20PM +0200, Vladimir Kirillov wrote: > Hello, tech@ > > There's a typo in crontab.5, which may be confusing and lead to parser > warnings like: > Nov 19 18:04:01 sun cron[24897]: (root) UNSAFE (`') > slightly different fix from millert ("") committed a few minutes a

Re: iostat and more than one core

On Thu, Nov 18, 2010 at 15:20 +0100, Mike Belopuhov wrote: > On Thu, Nov 18, 2010 at 14:42 +0100, Mike Belopuhov wrote: > > On Thu, Nov 18, 2010 at 01:01 -0500, Ted Unangst wrote: > > > 2010/11/17 Mike Belopuhov : > > > > I think this time around it's kernel that's wrong. top(1) calls CPTIME2 > >

crontab.5 diff

Hello, tech@ There's a typo in crontab.5, which may be confusing and lead to parser warnings like: Nov 19 18:04:01 sun cron[24897]: (root) UNSAFE (`') Index: crontab.5 === RCS file: /cvs/src/usr.sbin/cron/crontab.5,v retrieving revi

relayd icmp_check on ENETUNREACH

Hi All, I am testing a redundant firewall setup where I am trying to use relayd to make sure that the uplinks are reachable. The relayd.conf is very simple: --- # cat /etc/relayd.conf interval 5 table { B B B B 10.0.0.1 ip ttl 1 retry 2 } router "upli