headers
Ilya A. Kovalenko | 2 Aug 2005 21:32

typo in ppp(8) manpage

Index: usr.sbin/ppp/ppp/ppp.8.m4
===================================================================
RCS file: /cvs/src/usr.sbin/ppp/ppp/ppp.8.m4,v
retrieving revision 1.22
diff -u -r1.22 ppp.8.m4
--- usr.sbin/ppp/ppp/ppp.8.m4   2005/03/10 09:35:10     1.22
+++ usr.sbin/ppp/ppp/ppp.8.m4   2005/08/02 19:22:15
 <at>  <at>  -4594,7 +4594,7  <at>  <at> 
 .Pp
 For example (wrapped for readability):
 .Bd -literal -offset indent
-set login "TIMEOUT 5 \\"\\" \\"\\" login:--login: ppp \e
+set dial "TIMEOUT 5 \\"\\" \\"\\" login:--login: ppp \e
 word: ppp \\"!sh \\\\-c \\\\\\"echo \\\\-n label: >&2\\\\\\"\\" \e
 \\"!/bin/echo in\\" HELLO"
 .Ed

diff, maybe, is broken

Ilya A. Kovalenko            (mailto:shadow <at> oganer.net)
Permalink | Reply | 1 comment |
headers
Ilya A. Kovalenko | 2 Aug 2005 22:14

Re: typo in ppp(8) manpage

Called off, sorry.
Permalink | Reply | 0 comments |
headers
Oleg Safiullin | 3 Aug 2005 16:02
Picon

floppy support for amd64

Hi.
This is the patch makes floppy drives work on amd64...

Index: arch/amd64/amd64/machdep.c
===================================================================
RCS file: /cvs/src/sys/arch/amd64/amd64/machdep.c,v
retrieving revision 1.34
diff -d -u -r1.34 machdep.c
--- arch/amd64/amd64/machdep.c	2005/08/01 16:18:43	1.34
+++ arch/amd64/amd64/machdep.c	2005/08/03 13:15:04
 <at>  <at>  -1378,6 +1378,7  <at>  <at> 
  			printf("WARNING: CAN'T ALLOCATE EXTENDED MEMORY FROM "
  			    "IOMEM EXTENT MAP!\n");
  		}
+#if 0
  #if NISADMA > 0
  		/*
  		 * Some motherboards/BIOSes remap the 384K of RAM that would
 <at>  <at>  -1399,6 +1400,7  <at>  <at> 
  			    pbuf);
  			biosextmem = (15*1024);
  		}
+#endif
  #endif
  		mem_clusters[1].start = IOM_END;
  		mem_clusters[1].size = trunc_page(KBTOB(biosextmem));
Index: arch/amd64/conf/GENERIC
===================================================================
RCS file: /cvs/src/sys/arch/amd64/conf/GENERIC,v
retrieving revision 1.76
(Continue reading)

Permalink | Reply | 1 comment |
headers
Jesper Louis Andersen | 3 Aug 2005 23:06

Adding support for popa3d to only listen on certain devices

I am currently planning a bit to add support for popa3d to only listen 
on specific interfaces when running daemonized. How would that fare if I 
intended to submit a patch to OpenBSD? Is this regarded as bloat?

Here is the current plan:

1. Listen to Humppa! to get into correct hacking mood.
2. Copycat Henning Brauers OpenBGPd by reading source code:
   a. build a struct listen_addr suitable for <list.h> macros.
   b. add -L <ip> option, implying -D to getopt().
   c. for each -L; push on list.
   d. bind() and listen() to each given interface.
   e. fumble a bit around, accept traffic from each fd in play.
   f. Drink coffee.
   g. Provide patch for ok'ing after extensive debugging and review.
Permalink | Reply | 0 comments |
headers
Andrew M. Hoerter | 4 Aug 2005 23:26
Picon
Favicon

RFC: decouple YP client & server startup

Currently /etc/rc always tests for the existence of /var/yp/binding
before attempting to start ypbind and/or ypserv.  Occasionally though,
it might be desired to run ypserv and not ypbind, and yp(8) does not
mention any dependency on a server also being a client.  

The following patch changes this behavior so that if a YP domainname
is set, ypserv and ypbind are started independently of each other.

Index: etc/rc
===================================================================
RCS file: /cvs/src/etc/rc,v
retrieving revision 1.271
diff -u -r1.271 rc
--- etc/rc	25 Jul 2005 19:18:44 -0000	1.271
+++ etc/rc	4 Aug 2005 20:58:56 -0000
 <at>  <at>  -336,14 +336,18  <at>  <at> 
 	echo -n ' portmap';		portmap
 fi

-if [ -d /var/yp/binding -a X`domainname` != X ]; then
+# Initialize YP when a domainname is set.
+if [ X`domainname` != X ]; then
 	if [ -d /var/yp/`domainname` ]; then
-		# yp server capabilities needed...
+		# YP server capabilities needed.
 		echo -n ' ypserv';		ypserv ${ypserv_flags}
 		#echo -n ' ypxfrd';		ypxfrd
 	fi

-	echo -n ' ypbind';		ypbind
(Continue reading)

Permalink | Reply | 0 comments |
headers
Marc Espie | 5 Aug 2005 12:41
Favicon

Re: handle long names in packages

This is a fixed version of the first patch.
This requires -current pkg_add, to have the proper Ustar code.

Bernd noticed the first patch was dysfunctional wrt long file names,
and indeed there were a few problems.

Hardlinking to long names also didn't work. All these should be fixed now.

Please test/comment/okay.

Index: Makefile
===================================================================
RCS file: /cvs/src/usr.sbin/pkg_add/Makefile,v
retrieving revision 1.24
diff -u -p -r1.24 Makefile
--- Makefile	28 Jun 2005 20:34:55 -0000	1.24
+++ Makefile	5 Aug 2005 10:37:06 -0000
 <at>  <at>  -8,6 +8,7  <at>  <at>  POD2MAN=/usr/bin/pod2man

 PACKAGES= \
     OpenBSD/Add.pm \
+    OpenBSD/ArcCheck.pm \
     OpenBSD/CollisionReport.pm \
     OpenBSD/Delete.pm \
     OpenBSD/Error.pm \
Index: pkg_create
===================================================================
RCS file: /cvs/src/usr.sbin/pkg_add/pkg_create,v
retrieving revision 1.35
diff -u -p -r1.35 pkg_create
(Continue reading)

Permalink | Reply | 0 comments |
headers
Artur Grabowski | 4 Aug 2005 11:59

Re: floppy support for amd64

Oleg Safiullin <form <at> pdp-11.org.ru> writes:

> Hi.
> This is the patch makes floppy drives work on amd64...

>  <at>  <at>  -153,7 +153,7  <at>  <at> 
> 
>   puc*	at pci?				# PCI "universal" communication device
> 
> -#isadma0	at isa?
> +isadma0	at isa?
> 
>   option	WSDISPLAY_COMPAT_USL		# VT handling
>   option	WSDISPLAY_COMPAT_RAWKBD		# can get raw scancodes

Over my dead, rotten, violated body. Just because it's doable doesn't mean
it's a good idea. Some things are not meant to be done.

NO!

The memory handling code that allows isadma to work on amd64 will be
removed any day now. You might try to implement isdma with the gart,
but the old i386 isadma code will not touch amd64.

NO!

Throw the isa floppy into the trash. Get an usb floppy. Get a
cd-rw. Get a usb key.

NO!
(Continue reading)

Permalink | Reply | 0 comments |
headers
Ober Heim | 5 Aug 2005 18:00

Howto for Ethereal on OpenBSD 3.7

Now before everyone goes off half cocked about 'ohh, I read that 
application is insecure!' rant, be aware that process separation is
currently in the works by me.

With that said, I put together a simple how-to on getting ethereal 0.10.11 
to compile properly.

http://www.linbsd.org/ethereal_on_openbsd37.html

As far as security goes, it goes without saying:
Dont run ethereal in capture/decode mode as root.
Capture with tcpdump to a file and read with ethereal as a non privileged 
user.
You can even chown _ethereal /dev/bpf* if you are really inclined to run 
it as the primary capture device. Limit your captures to specific 
protocols you need and nothing else.
Most of the time that there are exploits, it happens to deal with specific 
dissectors that you do not care about.
So, say  if you only want to capture web traffic then do so.
ethereal -R "http"
Or limit it with capture filters to specific ports
ethereal port 80

</rant>
"I am not your puppet. Since when? Now, get your spongy pink ass out 
there, and dance for the cameras."  -Death to Smoochy
Permalink | Reply | 0 comments |
headers
Marc Espie | 6 Aug 2005 12:54
Favicon

crunchgen changes

Just a reminder for the few people using make release: crunchgen changed,
you must reinstall it before you mkr.
Permalink | Reply | 0 comments |
headers
Marc Espie | 7 Aug 2005 13:03
Favicon

LC_CTYPE diff, what's left

After getting prompted by theo to commit most of the LC_CTYPE diff, and
not activate it, here's the remaining diff, small enough to put here:

Index: include/ctype.h
===================================================================
RCS file: /mirrors/openbsd/cvs/src/include/ctype.h,v
retrieving revision 1.17
diff -u -p -r1.17 ctype.h
--- include/ctype.h	2 Apr 2004 22:37:12 -0000	1.17
+++ include/ctype.h	7 Aug 2005 10:45:36 -0000
 <at>  <at>  -147,14 +147,14  <at>  <at>  __CTYPE_INLINE int isxdigit(int c)

 __CTYPE_INLINE int tolower(int c)
 {
-	if ((unsigned int)c > 0177)
+	if ((unsigned int)c > 255)
 		return (c);
 	return ((_tolower_tab_ + 1)[c]);
 }

 __CTYPE_INLINE int toupper(int c)
 {
-	if ((unsigned int)c > 0177)
+	if ((unsigned int)c > 255)
 		return (c);
 	return ((_toupper_tab_ + 1)[c]);
 }
Index: lib/libc/shlib_version
===================================================================
RCS file: /mirrors/openbsd/cvs/src/lib/libc/shlib_version,v
(Continue reading)

Permalink | Reply | 0 comments |

Gmane