Re: bug: "device ether" no longer optional

[Maxim Sobolev]

"Andrey A. Chernov" wrote:

> Without "device ether" in config file kernel fails to compile and
> complains on undefined function ether_ifdetach() in if.c:if_detach()

Here too.

-Maxim



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

bug: kernel hangs at boot

[Andrey A. Chernov]

Fresh -current kernel with fresh -current bootblocks loaded ok but hangs in 
boot (no diagnostics even printed, just silent hang with "/" on the screen).

Kernel from Jun14 with fresh -current bootblocks (i.e. the same bootblocks) 
booted Ok.

-- 
Andrey A. Chernov
<[EMAIL PROTECTED]>
http://ache.pp.ru/


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

RE: Laptop boot problem in current with GENERIC & GENERIC.hints

[John Baldwin]

On 27-Jun-00 John Baldwin wrote:
> 
> On 27-Jun-00 Edwin Culp wrote:
>> I have 5 machines running current.  All are booting fine with the
>> changes made on June 12, with the exception of a k-6 laptop.  I have
>> been booting the kernel.works since the changes and can't seem to get
>> something right.  I have tried with my original kernel configuration and
>> now have been trying with the GENERIC configuration and GENERIC.hints.
>> When I try to boot with my new kernel I get:
>> 
>> BTX loader 1.0 BTX version is 1.01
>> Console: internal video/keyboard
>> BIOS drive A: is disk0
>> BIOS drive B: is disk1
>> BIOS 639kB/97280kB available memory
>> 
>> FreeBSD/i386 bootstrap loader, Revision 0.8
>> [EMAIL PROTECTED] Mon June 26 12:37:44 PDT 2000
>> 
>> int=000d err= eft=00010046 eip=92eb
>> eax=00fb ebx= ecx=0152 edx=03f6
>> esi=000fbc92 edi=000f ebp=17b4 esp=17b4
>> cs:eip=0f 32 89 45 1c 89 55 14-46 eb cf 80 65 31 fd eb
> 
> RDMSR instruction.  It's trying to read an MSR (Model
> Specific Register) from the CPU.  In this case, one that
> is not documented in Intel's manuals.  Thus, it is either
> an undocmented MSR or a K6-only MSR.
> 
>> ss:esp=08 00 00 00 1c 9c 00 00-fe 0e 00 00 d4 17 00 00
>> System halted
>> 
>> I copied this manually since I don't have a console on the laptop.
>> 
>> Any suggestions will be appreciated.
> 
> Rev 1.18 of sys/boot/i386/btx/btx/btx.s should fix this.  Try
> doing an ftp install of a really recent snap from
> current.FreeBSD.org.

Augh!  I found out that actually rev 1.18 contained a bug
causing this problem.  Rev 1.19 of btx.s should actually fix
this.  Sorry.

-- 

John Baldwin <[EMAIL PROTECTED]> -- http://www.FreeBSD.org/~jhb/
PGP Key: http://www.cslab.vt.edu/~jobaldwi/pgpkey.asc
"Power Users Use the Power to Serve!"  -  http://www.FreeBSD.org/


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Laptop boot problem in current with GENERIC & GENERIC.hints

[Andrey A. Chernov]

On Wed, Jun 28, 2000 at 10:14:27PM -0700, John Baldwin wrote:
> It's fixed.  It wasn't a kernel problem but a stupid bogon on my part
> in btx.s.  Make sure you have rev 1.19 of /sys/boot/i386/btx/btx/btx.s
> and rebuild your loader.
> 

It seems there is a different problem.  Fresh -current kernel with fresh 
-current bootblock loaded ok but hangs in boot (no diagnostics even 
printed).  Kernel from Jun14 with fresh -current bootblocks booted Ok.

-- 
Andrey A. Chernov
<[EMAIL PROTECTED]>
http://ache.pp.ru/


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

bug: "device ether" no longer optional

[Andrey A. Chernov]

Without "device ether" in config file kernel fails to compile and
complains on undefined function ether_ifdetach() in if.c:if_detach()

Please fix.

-- 
Andrey A. Chernov
<[EMAIL PROTECTED]>
http://ache.pp.ru/


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Laptop boot problem in current with GENERIC & GENERIC.hints

[John Baldwin]

On 27-Jun-00 Edwin Culp wrote:
> Donn Miller wrote:
> 
>> Did you compile your kernel with any optimizations, such as -Os?  If so,
>> try it recompiling it without optimizations.
> 
> Unfortunately, no.  I wish that I had, because I'm fresh out of things to
> try.

It's fixed.  It wasn't a kernel problem but a stupid bogon on my part
in btx.s.  Make sure you have rev 1.19 of /sys/boot/i386/btx/btx/btx.s
and rebuild your loader.

> Thanks,
>
> ed

-- 

John Baldwin <[EMAIL PROTECTED]> -- http://www.FreeBSD.org/~jhb/
PGP Key: http://www.cslab.vt.edu/~jobaldwi/pgpkey.asc
"Power Users Use the Power to Serve!"  -  http://www.FreeBSD.org/


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Laptop boot problem in current with GENERIC & GENERIC.hints

[John Baldwin]

On 28-Jun-00 Donn Miller wrote:
> On Wed, 28 Jun 2000, Warner Losh wrote:
> 
>> In message <[EMAIL PROTECTED]> Edwin Culp writes:
>> : int=000d err= eft=00010046 eip=92eb
>> : eax=00fb ebx= ecx=0152 edx=03f6
>> : esi=000fbc92 edi=000f ebp=17b4 esp=17b4
>> : cs:eip=0f 32 89 45 1c 89 55 14-46 eb cf 80 65 31 fd eb
>> : ss:esp=08 00 00 00 1c 9c 00 00-fe 0e 00 00 d4 17 00 00
>> : System halted
>> : 
>> : I copied this manually since I don't have a console on the laptop.
>> 
>> I'd try booting the old kernel and seeing if doing a completely clean
>> build fixes this.  I'd also try to reboot after power off.
> 
> It looks like the optimization bug that's been biting kernel builds
> lately.  When Peter Wemm made the overhaul to the kernel config, strange
> things started happening when optimization higher than -O was used to
> build the kernel.  For example, I got a Fatal Trap 12 immediately after
> the probe for available memory.  I was using -Os to compile my kernel at
> the time.  Backing down to -O for kernel builds solved this, and boots
> proceeded normally.

Ok, not directed at you personally, but for future reference:

This is a BTX fault.  Kernel traps do not look like this.  If
you see this, you haven't even loaded the kernel yet and thus
your problem does not come from the kernel.  Since we got into
the loader, that means that the loader is b0rked.

As per this problem, it was due to a stupid error on my part
and is fixed in rev 1.19 of /sys/boot/i386/btx/btx.s.  Please
update your sources and recompile all of /sys/boot/i386.

Thank you.

> - Donn

-- 

John Baldwin <[EMAIL PROTECTED]> -- http://www.FreeBSD.org/~jhb/
PGP Key: http://www.cslab.vt.edu/~jobaldwi/pgpkey.asc
"Power Users Use the Power to Serve!"  -  http://www.FreeBSD.org/


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

RE: The position of {boot,fixit}_crunch.conf

[John Baldwin]

On 28-Jun-00 Takahashi Yoshihiro wrote:
> Hi,
> 
> I'd like to move {boot,fixit}_crunch.conf from src/release to
> src/release/$MACHINE because it needs to use different configuration
> file between PC/AT and PC-98 to decrease the size of PC-98 fixit.flp
> to 1.2MB.
> 
> In addition, alpha machines don't require the fdisk command, so I
> think that it should be removed the fdisk command from
> fixit_crunch.conf for alpha.

Actually, I'd add the src/release/${MACHINE} dir, but only move
a conf file if you need to.  For instance, if you rip the unused
/usr/src/sbin/${MACHINE_ARCH} dir from boot_crunch.conf, it will
work on the alpha again and will be the same for all platforms.

Only the fixit floppy actually differs at the moment.

You can implement this by checking for src/release/${MACHINE}/${conf_file}
first and fallback to src/release/${conf_file} if you don't find
the first one.

> Comment?

-- 

John Baldwin <[EMAIL PROTECTED]> -- http://www.FreeBSD.org/~jhb/
PGP Key: http://www.cslab.vt.edu/~jobaldwi/pgpkey.asc
"Power Users Use the Power to Serve!"  -  http://www.FreeBSD.org/


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

RE: Laptop boot problem in current with GENERIC & GENERIC.hints

[John Baldwin]

On 27-Jun-00 Edwin Culp wrote:
> I have 5 machines running current.  All are booting fine with the
> changes made on June 12, with the exception of a k-6 laptop.  I have
> been booting the kernel.works since the changes and can't seem to get
> something right.  I have tried with my original kernel configuration and
> now have been trying with the GENERIC configuration and GENERIC.hints.
> When I try to boot with my new kernel I get:
> 
> BTX loader 1.0 BTX version is 1.01
> Console: internal video/keyboard
> BIOS drive A: is disk0
> BIOS drive B: is disk1
> BIOS 639kB/97280kB available memory
> 
> FreeBSD/i386 bootstrap loader, Revision 0.8
> [EMAIL PROTECTED] Mon June 26 12:37:44 PDT 2000
> 
> int=000d err= eft=00010046 eip=92eb
> eax=00fb ebx= ecx=0152 edx=03f6
> esi=000fbc92 edi=000f ebp=17b4 esp=17b4
> cs:eip=0f 32 89 45 1c 89 55 14-46 eb cf 80 65 31 fd eb

RDMSR instruction.  It's trying to read an MSR (Model
Specific Register) from the CPU.  In this case, one that
is not documented in Intel's manuals.  Thus, it is either
an undocmented MSR or a K6-only MSR.

> ss:esp=08 00 00 00 1c 9c 00 00-fe 0e 00 00 d4 17 00 00
> System halted
> 
> I copied this manually since I don't have a console on the laptop.
> 
> Any suggestions will be appreciated.

Rev 1.18 of sys/boot/i386/btx/btx/btx.s should fix this.  Try
doing an ftp install of a really recent snap from
current.FreeBSD.org.

> Thanks,
> 
> ed

-- 

John Baldwin <[EMAIL PROTECTED]> -- http://www.FreeBSD.org/~jhb/
PGP Key: http://www.cslab.vt.edu/~jobaldwi/pgpkey.asc
"Power Users Use the Power to Serve!"  -  http://www.FreeBSD.org/


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Release still broken with new Perl

[John Baldwin]

On 28-Jun-00 Mark Murray wrote:
>> ===> gnu/usr.bin/perl/library/B
>> make: don't know how to make distribute. Stop
> 
> OK - I know how to fix this.
> 
> Thanks for the report!

Thanks. :)

> M
> --
> Mark Murray
> Join the anti-SPAM movement: http://www.cauce.org

-- 

John Baldwin <[EMAIL PROTECTED]> -- http://www.FreeBSD.org/~jhb/
PGP Key: http://www.cslab.vt.edu/~jobaldwi/pgpkey.asc
"Power Users Use the Power to Serve!"  -  http://www.FreeBSD.org/


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: USB modems

[Mike Meyer]

Warner Losh writes:
> In message  Bob Bishop writes:
> : Can anyone give a quick synopsis of the current status of support for USB
> : modems? TIA
> They aren't supported yet.  There's at least one group that might be
> working on them.  The value of supporting them is well known.  Take
> care in your purcahse of a usb modem because some of them expect an
> isochronous audio stream...

Nick (and I, for that matter) have a umodem.c that works, for some
definition of "works". It seems to work fine on USR USB modems. On the
Supra I bought (because it was easily available), it works for dialout
and makes PPP connections, but outgoing IP connections fail under an
indeterminate set of conditions. It's not clear where the problem is -
I'll be investigating it as soon as I once again have free time (a
couple of weeks).

Nick has indicated he was going to try this version and commit it, but
it hasn't happened yet.

Re: XML driver config file to replace LINT

[Warner Losh]

In message <[EMAIL PROTECTED]> Nik Clayton writes:
: On Wed, Jun 28, 2000 at 11:27:12AM -0400, Thomas M. Sommers wrote:
: > Warner Losh wrote:
: > > 
: > > Any reason that the .c/.h files of the drivers couldn't be used to
: > > generate this information?  
: > 
: > Or perhaps the other way around.
: 
: That's what I'd prefer.  Any solution that relys on trying to parse
: "structured comments" like that is a kludge at best.  I'd rather get 
: all this information in to a usefully structured form now, and then
: process it to produce the various output formats we need.

I think that this will doom the information to always being obsolete.
If the information is in the .h or .c files, then it will be looked at
(and corrected) all the time by the programmers.  If not, then it will
rot as badly as LINT has been rotting.  It has taken much effort to
keep LINT as non-rotten as it has been kept.

Then again, there are some comments in the kernel that are in good
need of a swift kick to the head as well since they have rotted beyond
all usefulness in the source they are in.

Warner


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: XML driver config file to replace LINT

[Warner Losh]

In message <[EMAIL PROTECTED]> Nik
Clayton writes:
: Or does loading Japanese text in to a non-Japanese aware editor scramble
: the text?

It can, if the user editing the text isn't careful, or the editor
likes to do too many things automatically.  Generally speaking,
however, it shouldn't be a big problem.

Warner


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: XML driver config file to replace LINT

[Warner Losh]

In message <[EMAIL PROTECTED]> "Thomas M. Sommers" writes:
: Warner Losh wrote:
: > 
: > Any reason that the .c/.h files of the drivers couldn't be used to
: > generate this information?  
: 
: Or perhaps the other way around.

No.  I'm saying that the .c and .h files (likely .h) are the source to
the driver and driver maintanier are 1000% more likely to keep those
up to date than they are xxx.xml.  Right now, with modules, you can
easily not have to worry about any config issues outside of those
files.  Forcing a doc file just to get docs and breaking this is
undesirable.  Expanding what the driver writers are already doing
a little seems like a smarter move.

Warner


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: XML driver config file to replace LINT

[Warner Losh]

In message <[EMAIL PROTECTED]> Dan Moschuk writes:
: Perhaps it would be a good idea to assign someone the task of just maintaing
: the HARDWARE.TXT file, rather than expecting all developers to keep 
: documentation up to date?

This can be difficult to do.  It will take someone with enough cycles
to grok the new hardware support, as well as monitoring lists for
reports of trouble, etc.

: Warner is the gatekeeper for UPDATING, and it seems to work rather well.

Yes.  Sometimes less well than other times.  when I have about 2 hours
a week to spend on it or more, it works a lot better than when I'm
completely swamped...  I think it is working because I wanted to stop
the whining in -current about how hard it was to keep up with
current.  UPDATING has gone from a very terse warning bulletin to a
less terse, fairly useful document as the rough spots in the road get
written up well by others and I pick from the N different copies I
get.  Still, it is somewhat quirky.  I'd Kinda like to see it somehow
on the web pages updated in near real time (well, daily is close
enough :-)

: I'd take on the responsibility.

Good luck!

Warner



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: USB modems

[Warner Losh]

In message  Bob Bishop writes:
: Can anyone give a quick synopsis of the current status of support for USB
: modems? TIA

They aren't supported yet.  There's at least one group that might be
working on them.  The value of supporting them is well known.  Take
care in your purcahse of a usb modem because some of them expect an
isochronous audio stream...

Warner



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: XML driver config file to replace LINT

[Satoshi - Ports Wraith - Asami]

 * From: Nik Clayton <[EMAIL PROTECTED]>

 * Possibly.  I was thinking that the only thing that would be language
 * specific about each driver would be the comment section.
 * 
 * ...
 * 
 * All the other stuff is language independent.
 * 
 * That being the case, it wouldn't be too hard to do
 * 
 * ...
 * 
 * ...
 * 
 * and so on, would it?
 * 
 * Or does loading Japanese text in to a non-Japanese aware editor scramble
 * the text?

With ja_JP.eucJP, it certainly would, since that's an 8-bit encoding.
If you use JIS, it should be ok (as long as people pretend they don't
see all the ^['s hanging around and don't try to do anything cute).

Satoshi


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Strange make buildworld problem...

[Donn Miller]

Recently, the make world problems with perl have been fixed, and I can do
a make world all the way through, provided I do a make world the first
time only.  However, I did a cvsup to update my source tree again after
the first make world.  I did a make -DNOCLEAN buildworld with /usr/obj
untouched.  I get this:

>>> stage 2: build tools
--
cd /usr/src; MAKEOBJDIRPREFIX=/usr/obj  INSTALL="sh
/usr/src/tools/install.sh"
PATH=/usr/obj/usr/src/i386/usr/sbin:/usr/obj/usr/src/i386/usr/bin:/usr/obj/usr/s
rc/i386/usr/games:/sbin:/bin:/usr/sbin:/usr/bin make -f Makefile.inc1
build-tools
cd /usr/src/bin/sh; make build-tools
cd /usr/src/games/adventure; make build-tools
cd /usr/src/games/hack; make build-tools
cd /usr/src/games/phantasia; make build-tools
cd /usr/src/gnu/usr.bin/cc/cc_tools; make build-tools
cd /usr/src/gnu/usr.bin/cc/f771; make build-tools
cd /usr/src/lib/libncurses; make build-tools
cd /usr/src/share/syscons/scrnmaps; make build-tools
cd /usr/src/gnu/usr.bin/perl; make build-tools
cd /usr/src/gnu/usr.bin/perl/libperl && make build-tools
make: don't know how to make
/usr/obj/usr/src/i386/usr/include/sys/types.h. Stop
*** Error code 2
Stop in /usr/src/gnu/usr.bin/perl.
*** Error code 1
Stop in /usr/src.
*** Error code 1

Again, this is after re-making buildworld with -DNOCLEAN after a
successful first make buildworld.

- Donn



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Huh? ssh stopped working with new kernel.

[Brian Somers]

> * Julian Elischer <[EMAIL PROTECTED]> [000627 12:36] wrote:
> > So I cvsupped yesterday, and tehn made a new kernel.
> > so suddenly ssh doesn't work any more.
> 
> cvsup again and recompile and reload the randomdev module, it should
> resume working.

And ``sed -n /^2624:$/,/^$/p /usr/src/UPDATING''

> -Alfred

-- 
Brian <[EMAIL PROTECTED]>
     
Don't _EVER_ lose your sense of humour !




To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: What's going on with crypto all of a sudden ?

[Kenneth Wayne Culver]

Nothing to be sorry about... just read a little more carefully :-)


=
| Kenneth Culver  | FreeBSD: The best NT upgrade|
| Unix Systems Administrator  | ICQ #: 24767726 |
| and student at The  | AIM: muythaibxr |
| The University of Maryland, | Website: (Under Construction)   |
| College Park.   | http://www.wam.umd.edu/~culverk/|
=

On Wed, 28 Jun 2000, Pascal Hofstee wrote:

> On Wed, Jun 28, 2000 at 04:28:24PM -0400, Kenneth Wayne Culver wrote:
> > if you're using -CURRENT, it's because you havn't followed the advice
> > about the new randomdev.
> 
> Stumbled over it right after i sent the email  sorry 
> 
> -- 
>   Pascal Hofstee  < daeron @ shadowmere . student . utwente . nl >
>   Managers know it must be good because the programmers hate it so much.
> 



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: What's going on with crypto all of a sudden ?

[Pascal Hofstee]

On Wed, Jun 28, 2000 at 04:28:24PM -0400, Kenneth Wayne Culver wrote:
> if you're using -CURRENT, it's because you havn't followed the advice
> about the new randomdev.

Stumbled over it right after i sent the email  sorry 

-- 
  Pascal Hofstee  < daeron @ shadowmere . student . utwente . nl >
  Managers know it must be good because the programmers hate it so much.


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: What's going on with crypto all of a sudden ?

[Kenneth Wayne Culver]

if you're using -CURRENT, it's because you havn't followed the advice
about the new randomdev.


=
| Kenneth Culver  | FreeBSD: The best NT upgrade|
| Unix Systems Administrator  | ICQ #: 24767726 |
| and student at The  | AIM: muythaibxr |
| The University of Maryland, | Website: (Under Construction)   |
| College Park.   | http://www.wam.umd.edu/~culverk/|
=

On Wed, 28 Jun 2000, Pascal Hofstee wrote:

> Just now after a new buildworld/installworld i am suddenly getting errors
> like the following when trying to use OpenSSH's version 1 protocoll:
> 
> ssh: no RSA support in libssl and libcrypto.  See ssl(8).
> Disabling protocol version 1
> Protocol major versions differ: 2 vs. 1
> 
> 
> man 8 sslresults into the following:
> No entry for ssl in section 8 of the manual
> 
> This has been working just fine for months before.
> I get the idea it has something to do with the /dev/(random|zero) update.
> 
> Anyone care to explain this ?
> 
> -- 
>   Pascal Hofstee  < daeron @ shadowmere . student . utwente . nl >
>   Managers know it must be good because the programmers hate it so much.
> 
> 
> To Unsubscribe: send mail to [EMAIL PROTECTED]
> with "unsubscribe freebsd-current" in the body of the message
> 



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

What's going on with crypto all of a sudden ?

[Pascal Hofstee]

Just now after a new buildworld/installworld i am suddenly getting errors
like the following when trying to use OpenSSH's version 1 protocoll:

ssh: no RSA support in libssl and libcrypto.  See ssl(8).
Disabling protocol version 1
Protocol major versions differ: 2 vs. 1


man 8 sslresults into the following:
No entry for ssl in section 8 of the manual

This has been working just fine for months before.
I get the idea it has something to do with the /dev/(random|zero) update.

Anyone care to explain this ?

-- 
  Pascal Hofstee  < daeron @ shadowmere . student . utwente . nl >
  Managers know it must be good because the programmers hate it so much.


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: XML driver config file to replace LINT

[Wilko Bulte]

On Wed, Jun 28, 2000 at 02:45:25PM -0400, Dan Moschuk wrote:
> 
> | > Perhaps it would be a good idea to assign someone the task of just maintaing
> | > the HARDWARE.TXT file, rather than expecting all developers to keep 
> | > documentation up to date?
> | 
> | That is what I am currently doing (at least trying to.. ;-) for
> | FreeBSD/alpha. But considering that the alpha community is considerably
> | smaller this might be a problem in the much bigger x86 world.
> 
> Right, but as someone that reads all the commit messages, catching new
> driver support and updating is more of a mundane task than anything.  If
> maintainers emailed when they added new support (like when an entry in
> UPDATING is needed) it's even easier.

We can always dream, can we? ;-)

I don't read all commit messages, rather procmail sifts:

# want commit messages related to alpha saved
:0:
* ^Subject:.*cvs commit.*alpha
$MAILDIR/commit-mail

the alpha stuff from them. 

-- 
Wilko Bulte http://www.freebsd.org  "Do, or do not. There is no try"
[EMAIL PROTECTED]   http://www.nlfug.nl Yoda - The Empire Strikes Back


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: The position of {boot,fixit}_crunch.conf

[Brandon D. Valentine]

On Wed, 28 Jun 2000, Wilko Bulte wrote:

>Rather, I mean a *small* bootable CD image like NetBSD uses, essentially 
>a 'super floppy'. Which would then be suitable for a Net install too.
>There is already a LS120 (?) image similar to that I think.

Which, btw, would be a nifty thing to have if you've got an alpha
capable of bootstrapping over the network.  I've just installed
NetBSD/sparc on a couple of sparcs using NetBSD's boot.net images and an
NFS mounted root directory as the install media.  It's very cool and
saved me the trouble of tracking down a SCSI floppy or CDROM.

Brandon D. Valentine
-- 
bandix at looksharp.net  |  bandix at structbio.vanderbilt.edu
"Truth suffers from too much analysis." -- Ancient Fremen Saying



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: XML driver config file to replace LINT

[Dan Moschuk]

| > Perhaps it would be a good idea to assign someone the task of just maintaing
| > the HARDWARE.TXT file, rather than expecting all developers to keep 
| > documentation up to date?
| 
| That is what I am currently doing (at least trying to.. ;-) for
| FreeBSD/alpha. But considering that the alpha community is considerably
| smaller this might be a problem in the much bigger x86 world.

Right, but as someone that reads all the commit messages, catching new
driver support and updating is more of a mundane task than anything.  If
maintainers emailed when they added new support (like when an entry in
UPDATING is needed) it's even easier.

-- 
Dan Moschuk ([EMAIL PROTECTED])
"Don't get even -- get odd!"


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: The position of {boot,fixit}_crunch.conf

[Wilko Bulte]

On Wed, Jun 28, 2000 at 11:19:20AM -0700, David O'Brien wrote:
> On Wed, Jun 28, 2000 at 08:11:22PM +0200, Wilko Bulte wrote:
> > Seconded. But Maybe at some point in time we should decide CD[R] is the load
> > medium of choice for Alpha. Or?
> 
> That would mostly prevent Net installs, and anyone installing snapshots
> from current.freebsd.org.  Do we really want to hang Alpha users out to
> dry like that?

Rather, I mean a *small* bootable CD image like NetBSD uses, essentially 
a 'super floppy'. Which would then be suitable for a Net install too.
There is already a LS120 (?) image similar to that I think.

Makes sense?

-- 
Wilko Bulte http://www.freebsd.org  "Do, or do not. There is no try"
[EMAIL PROTECTED]   http://www.nlfug.nl Yoda - The Empire Strikes Back


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Check for ports updates

[David O'Brien]

On Wed, Jun 28, 2000 at 08:08:07PM +0200, Leif Neland wrote:
> NO_UPDATE="major jump in version - keep old version too" 
> Or
> UPDATE_OK="1.0.3"
> meaning it's ok to upgrade if the previous version is 1.0.3 or higher.

You are now *way* into Ports issues.  Please move this discussion to
[EMAIL PROTECTED] where the audience is more versed on the implications
of this.

-- 
-- David  ([EMAIL PROTECTED])


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: XML driver config file to replace LINT

[Wilko Bulte]

On Wed, Jun 28, 2000 at 01:49:36PM -0400, Dan Moschuk wrote:
> 
> | I disagree.  We're not Linux, where people can throw in code without thought
> | to the wider consequences -- one of the commitments you should make (that's
> | a generic "you" there, not you specifically) as a FreeBSD committer is to 
> | maintain the documentation that's affected by your changes.  A look at
> | HARDWARE.TXT shows that (with a few notable exceptions) the FreeBSD Developer
> | Community at large is *not* keeping it up to date.
> 
> Perhaps it would be a good idea to assign someone the task of just maintaing
> the HARDWARE.TXT file, rather than expecting all developers to keep 
> documentation up to date?

That is what I am currently doing (at least trying to.. ;-) for
FreeBSD/alpha. But considering that the alpha community is considerably
smaller this might be a problem in the much bigger x86 world.

-- 
Wilko Bulte http://www.freebsd.org  "Do, or do not. There is no try"
[EMAIL PROTECTED]   http://www.nlfug.nl Yoda - The Empire Strikes Back


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: The position of {boot,fixit}_crunch.conf

[David O'Brien]

On Wed, Jun 28, 2000 at 08:11:22PM +0200, Wilko Bulte wrote:
> Seconded. But Maybe at some point in time we should decide CD[R] is the load
> medium of choice for Alpha. Or?

That would mostly prevent Net installs, and anyone installing snapshots
from current.freebsd.org.  Do we really want to hang Alpha users out to
dry like that?
 
-- 
-- David  ([EMAIL PROTECTED])


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Check for ports updates

[Leif Neland]

On Wed, 28 Jun 2000, Alexander Leidinger wrote:

> On 28 Jun, Leif Neland wrote:
> > Any reason not to put this into bsd.port.mk?
> > 
> > make update
> 
>  - it removes your config files in most cases.

Most sane install's either just installs foo.conf.sample, or won't
overwrite existing config-files.

>  - it may break binaries which depend upon a specific library.
>  - ...

Libraries should be marked non-autoupgradable somehow.

Leif




To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: The position of {boot,fixit}_crunch.conf

[Wilko Bulte]

On Wed, Jun 28, 2000 at 10:09:31AM -0700, David O'Brien wrote:
> On Wed, Jun 28, 2000 at 09:28:31PM +0900, Takahashi Yoshihiro wrote:
> > I'd like to move {boot,fixit}_crunch.conf from src/release to
> > src/release/$MACHINE because it needs to use different configuration
> > file between PC/AT and PC-98 to decrease the size of PC-98 fixit.flp
> > to 1.2MB.
> > 
> > In addition, alpha machines don't require the fdisk command, so I
> > think that it should be removed the fdisk command from
> > fixit_crunch.conf for alpha.
> > 
> > Comment?
> 
> I really like that idea.  Alpha binaries are larger than Intel ones, thus
> when someone tweaks the floppies, often the Alpha ones become too big.

Seconded. But Maybe at some point in time we should decide CD[R] is the load
medium of choice for Alpha. Or?

-- 
Wilko Bulte http://www.freebsd.org  "Do, or do not. There is no try"
[EMAIL PROTECTED]   http://www.nlfug.nl Yoda - The Empire Strikes Back


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Check for ports updates

[Leif Neland]

On Wed, 28 Jun 2000, Alexander Leidinger wrote:

> On 28 Jun, Neil Blakey-Milner wrote:
> 
> >> Any reason not to put this into bsd.port.mk?
> >> 
> >> make update
> > 
> > It will break the system at least 20% of the time.  Change 20% to 100%
> > for gnome, kde, xpm, png, tiff, jpeg, and so forth.
> 
> I've successfully updated png/tiff/jpeg and some gnome packages
> (core/libs, but not from 1.0.x to 1.2.x) this way. 100% is to much, but
> even 10% is to much for "make update".
> 
Can the portmaintainer add a flag: 

NO_UPDATE="major jump in version - keep old version too" 

Or

UPDATE_OK="1.0.3"
meaning it's ok to upgrade if the previous version is 1.0.3 or higher.

Leif




To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: XML driver config file to replace LINT

[Dan Moschuk]

| I disagree.  We're not Linux, where people can throw in code without thought
| to the wider consequences -- one of the commitments you should make (that's
| a generic "you" there, not you specifically) as a FreeBSD committer is to 
| maintain the documentation that's affected by your changes.  A look at
| HARDWARE.TXT shows that (with a few notable exceptions) the FreeBSD Developer
| Community at large is *not* keeping it up to date.

Perhaps it would be a good idea to assign someone the task of just maintaing
the HARDWARE.TXT file, rather than expecting all developers to keep 
documentation up to date?

Warner is the gatekeeper for UPDATING, and it seems to work rather well.

If there is no one explicitly on the doc project that you'd like doing this,
I'd take on the responsibility.

-- 
Dan Moschuk ([EMAIL PROTECTED])
"Don't get even -- get odd!"


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: new ftpd feature...

[Will Andrews]

On Wed, Jun 28, 2000 at 12:23:53PM -0500, Visigoth wrote:
>   I understand how IP_PORTRANGE_HIGH and all work, but there are
> other programs which use these as well, and in at least my application,
> having support in ftpd is nice...

I'd have to agree with Visigoth here; system controls should be for
EVERYTHING, not just controlling one server.  Visigoth could use it for
a short-term solution, but in the long term, I think his solution is
better (in principle, I didn't look at the code).

-- 
Will Andrews <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
GCS/E/S @d- s+:+>+:- a--->+++ C++ UB P+ L- E--- W+++ !N !o ?K w---
?O M+ V-- PS+ PE++ Y+ PGP+>+++ t++ 5 X++ R+ tv+ b++> DI+++ D+ 
G++>+++ e-> h! r-->+++ y?


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: new ftpd feature...

[Visigoth]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

> ftpd uses ports in the high range, just adjust the last two sysctls
> and you'll be fine.

I understand how IP_PORTRANGE_HIGH and all work, but there are
other programs which use these as well, and in at least my application,
having support in ftpd is nice...

Visigoth

Damieon Stark
Sr. Unix Systems Administrator
[EMAIL PROTECTED]

PGP Public Key: www.telemere.net/~visigoth/visigoth.asc


|
M$ -Where do you want to go today?  |
Linux -Where do you want to go tomorrow?|   FreeBSD - The POWER to serve
Freebsd -Are you guys comming or what?  |   http://www.freebsd.org
|
|
- 

-BEGIN PGP SIGNATURE-
Version: PGPfreeware 5.0i for non-commercial use
Charset: noconv

iQA/AwUBOVomnTnmC/+RTnGeEQJg+gCggr35YOhnCfd2//YEmTJzH/72zKYAn2c4
6lS/dzts0qx97t3NRpUflZti
=ZqQI
-END PGP SIGNATURE-



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: new ftpd feature...

[Dag-Erling Smorgrav]

Visigoth <[EMAIL PROTECTED]> writes:
> [patches to limit the range of ports used for passive FTP]

des@flood ~% sysctl -A | grep portrange
net.inet.ip.portrange.lowfirst: 1023
net.inet.ip.portrange.lowlast: 600
net.inet.ip.portrange.first: 1024
net.inet.ip.portrange.last: 5000
net.inet.ip.portrange.hifirst: 49152
net.inet.ip.portrange.hilast: 65535

ftpd uses ports in the high range, just adjust the last two sysctls
and you'll be fine.

DES
-- 
Dag-Erling Smorgrav - [EMAIL PROTECTED]


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: The position of {boot,fixit}_crunch.conf

[David O'Brien]

On Wed, Jun 28, 2000 at 09:28:31PM +0900, Takahashi Yoshihiro wrote:
> I'd like to move {boot,fixit}_crunch.conf from src/release to
> src/release/$MACHINE because it needs to use different configuration
> file between PC/AT and PC-98 to decrease the size of PC-98 fixit.flp
> to 1.2MB.
> 
> In addition, alpha machines don't require the fdisk command, so I
> think that it should be removed the fdisk command from
> fixit_crunch.conf for alpha.
> 
> Comment?

I really like that idea.  Alpha binaries are larger than Intel ones, thus
when someone tweaks the floppies, often the Alpha ones become too big.
 
-- 
-- David  ([EMAIL PROTECTED])


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

new ftpd feature...

[Visigoth]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1


Sup all?

In light of the recent WU-ftpd exploits and general security
concearns, I decided to change to a different ftpd.(duh)  The largest
feature about wu-ftpd that I needed was the ability to specify the passive
port range so as to be able to write firewall rulesets with default deny
and only a small range of open ports for passive ftp.

I decided to integrate the suppport for that into freebsd's native
ftpd so that I could use an ftpd that I had faith in.  So here it
is.  These patches are against a brand new -stable so they should work
well.  Flames, Praise, Wine are all welcome ;) (nothing like the napa
valley or a good aussie red)

I even updated the man page for easy integration.

If somebody commits this, please tell me...

thanks

Visigoth


Damieon Stark
Sr. Unix Systems Administrator
[EMAIL PROTECTED]

PGP Public Key: www.telemere.net/~visigoth/visigoth.asc


|
M$ -Where do you want to go today?  |
Linux -Where do you want to go tomorrow?|   FreeBSD - The POWER to serve
Freebsd -Are you guys comming or what?  |   http://www.freebsd.org
|
|
- 

-BEGIN PGP SIGNATURE-
Version: PGPfreeware 5.0i for non-commercial use
Charset: noconv

iQA/AwUBOVocbjnmC/+RTnGeEQLFUgCg5sKoRD2gj7P+hIssj4zujPgdd/IAoLPk
xosQMqmtehAjA6sQb8/DjTI7
=zbKS
-END PGP SIGNATURE-


--- /usr/src/libexec/ftpd/ftpd.c.oldWed Jun 28 10:08:42 2000
+++ /usr/src/libexec/ftpd/ftpd.cWed Jun 28 09:49:05 2000
@@ -125,8 +125,9 @@
 union sockunion his_addr;
 union sockunion pasv_addr;
 
 intdaemon_mode;
+int pass_port_opt,min_pport,max_pport;
 intdata;
 jmp_buferrcatch, urgcatch;
 intlogged_in;
 struct passwd *pw;
@@ -158,8 +159,9 @@
 char   tmpline[7];
 char   *hostname;
 #ifdef VIRTUAL_HOSTING
 char   *ftpuser;
+char*opt_ptr;
 
 intepsvall = 0;
 
 static struct ftphost {
@@ -289,9 +291,9 @@
LastArgv = envp[-1] + strlen(envp[-1]);
 #endif /* OLD_SETPROCTITLE */
 
 
-   while ((ch = getopt(argc, argv, "AdlDSURt:T:u:va:p:46")) != -1) {
+   while ((ch = getopt(argc, argv, "AdlDSURt:T:u:va:p:P:46")) != -1) {
switch (ch) {
case 'D':
daemon_mode++;
break;
@@ -335,8 +337,21 @@
case 'p':
pid_file = optarg;
break;
 
+case 'P':
+pass_port_opt++;
+if((min_pport = strtod(optarg, &opt_ptr )) == 0 )
+errx(1,"Unrecognized passive port number!\n");
+  
+opt_ptr++;
+if((max_pport = strtod(opt_ptr,NULL)) == 0 )
+errx(1,"Unrecognized passive port number!\n");
+
+if( min_pport > max_pport )
+errx(1,"Portrange from %d to %d 
+invalid!\n",min_pport,max_pport);
+break;
+
case 'u':
{
long val = 0;
 
@@ -2323,9 +2338,9 @@
  */
 void
 passive()
 {
-   int len;
+   int len,next;
char *p, *a;
 
if (pdata >= 0) /* close old port if one set */
close(pdata);
@@ -2337,33 +2352,57 @@
}
 
(void) seteuid((uid_t)0);
 
+if (pass_port_opt < 1) {  /* if passive ports are in use don't bother
+with IP[V6]_PORTRANGE.  Doesn't affect port
+asignment, but saves little cpu? ;) */
+
 #ifdef IP_PORTRANGE
-   if (ctrl_addr.su_family == AF_INET) {
-   int on = restricted_data_ports ? IP_PORTRANGE_HIGH
+   if (ctrl_addr.su_family == AF_INET) {
+   int on = restricted_data_ports ? IP_PORTRANGE_HIGH
   : IP_PORTRANGE_DEFAULT;
 
-   if (setsockopt(pdata, IPPROTO_IP, IP_PORTRANGE,
+   if (setsockopt(pdata, IPPROTO_IP, IP_PORTRANGE,
(char *)&on, sizeof(on)) < 0)
goto pasv_error;
-   }
+   }
 #endif
 #ifdef IPV6_PORTRANGE
-   if (ctrl_addr.su_family == AF_INET6) {
-   int on = restricted_data_ports ? IPV6_PORTRANGE_HIGH
+   if (ctrl_addr.su_family == AF_INET6) {
+   int on = restricted_data_ports ? IPV6_PORTRANGE_HIGH
   : IPV6_PORTRANGE_DEFAULT;
 
-   if (setsockopt(pdata, IPPROTO_IPV6, IPV6_PORTRANGE,
+   if (setsockopt(pdata, IPPROTO_IPV6,

Re: XML driver config file to replace LINT

[Nik Clayton]

On Wed, Jun 28, 2000 at 11:27:12AM -0400, Thomas M. Sommers wrote:
> Warner Losh wrote:
> > 
> > Any reason that the .c/.h files of the drivers couldn't be used to
> > generate this information?  
> 
> Or perhaps the other way around.

That's what I'd prefer.  Any solution that relys on trying to parse
"structured comments" like that is a kludge at best.  I'd rather get 
all this information in to a usefully structured form now, and then
process it to produce the various output formats we need.

N
-- 
Internet connection, $19.95 a month.  Computer, $799.95.  Modem, $149.95.
Telephone line, $24.95 a month.  Software, free.  USENET transmission,
hundreds if not thousands of dollars.  Thinking before posting, priceless.
Somethings in life you can't buy.  For everything else, there's MasterCard.
  -- Graham Reed, in the Scary Devil Monastery


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: vnode_if.h: how should it be done ?

[Warner Losh]

In message <[EMAIL PROTECTED]> Kris 
Kennaway writes:
: On Sun, 25 Jun 2000, Warner Losh wrote:
: 
: > They aren't installed.  You must have a kernel tree to build the kld.
: > Just add vnode_if.h to your SRCS and the rest happens automatically
: > via bsd.kmod.mk.  It works great.  I do it all the time.  In fact, I
: > added SYSDIR support to bsd.kmod.mk at Timing Solutions so that we
: > could compile our drivers outside of the tree.
: 
: As I understand it, this is the only file which is required from the
: kernel sources in order to build the arla port. ISTR FiST also needs it to
: be present, and I also think I once came across another port that needed
: it too. Is it really so wrong to make the lives of these third-party FS
: tools harder by requiring full kernel sources to be present?

Yes.  They must be present.  That's the API we have, for good or for
ill.  In order to get the right interfaces and the right generated
files, they have to use the right proceedures.  If you have stale or
incorrect files, the ports will mysteriously not work.  I'll take a
look at the ports in question.  Likely they are buggy by not using
bsd.kmod.mk.  Using bsd.kmod.mk is mandatory for kernel modules.

Warner



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: XML driver config file to replace LINT

[Nik Clayton]

On Tue, Jun 27, 2000 at 11:49:34PM +0900, Hiroki Sato wrote:
> Nik Clayton <[EMAIL PROTECTED]> wrote
>  in <[EMAIL PROTECTED]>:
> 
> > [ That schema is not set in stone, and certainly requires more work.  In
> >   particular, we probably need "lang" and "encoding" options on the
> >element, to support comments in more than one language. ]
> > 
> > LINT would then become a skeletal file for things which don't fit this
> > sort of pattern, and the full LINT would be generated by a script which
> > parsed the above and the skeletal file to generate the full LINT.
> 
>  I think it is difficult to maintain the files because few editors
>  can handle various languages/encodings at the same time.
>  So, especially for translators, it is better that the .xml files
>  are separated on a encoding/language basis.

Possibly.  I was thinking that the only thing that would be language
specific about each driver would be the comment section.

...

All the other stuff is language independent.

That being the case, it wouldn't be too hard to do

...

...

and so on, would it?

Or does loading Japanese text in to a non-Japanese aware editor scramble
the text?
 
N
-- 
Internet connection, $19.95 a month.  Computer, $799.95.  Modem, $149.95.
Telephone line, $24.95 a month.  Software, free.  USENET transmission,
hundreds if not thousands of dollars.  Thinking before posting, priceless.
Somethings in life you can't buy.  For everything else, there's MasterCard.
  -- Graham Reed, in the Scary Devil Monastery


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Check for ports updates

[Alexander Leidinger]

On 28 Jun, Neil Blakey-Milner wrote:

>> Any reason not to put this into bsd.port.mk?
>> 
>> make update
> 
> It will break the system at least 20% of the time.  Change 20% to 100%
> for gnome, kde, xpm, png, tiff, jpeg, and so forth.

I've successfully updated png/tiff/jpeg and some gnome packages
(core/libs, but not from 1.0.x to 1.2.x) this way. 100% is to much, but
even 10% is to much for "make update".

Bye,
Alexander.

-- 
   I believe the technical term is "Oops!"

http://www.Leidinger.net  Alexander+Home @ Leidinger.net
  GPG fingerprint = 7423 F3E6 3A7E B334 A9CC  B10A 1F5F 130A A638 6E7E



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Check for ports updates

[Alexander Leidinger]

On 28 Jun, Leif Neland wrote:
> Any reason not to put this into bsd.port.mk?
> 
> make update

 - it removes your config files in most cases.
 - it may break binaries which depend upon a specific library.
 - ...

Bye,
Alexander.

-- 
Secret hacker rule #11: hackers read manuals.

http://www.Leidinger.net  Alexander+Home @ Leidinger.net
  GPG fingerprint = 7423 F3E6 3A7E B334 A9CC  B10A 1F5F 130A A638 6E7E



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Boot Oddities

[Dampure, Pierre Y.]

Apologies to all, this was a pilot error. /boot/defaults/loader.conf was
referencing disk1s1a -- it should have been disk2s1a (I forgot the
floppy was included in the unit count).

After a few interesting floppy boots / fdisk / disklabel sessions, I got
the thingie to work.

One point, though: the GENERIC kernel on the snapshot install was
(obvioulsy) unaware of my peculiar setup and thus  refused to load
/boot/loader. This might prove a bit puzzling for newbies like me... and
thus might deserve an FAQ entry.

All in all, absence of answers proved to be the best incentive to hack
around. Oh well, you learn everyday :-)


Best Regards,

PYD


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: XML driver config file to replace LINT

[Thomas M. Sommers]

Warner Losh wrote:
> 
> Any reason that the .c/.h files of the drivers couldn't be used to
> generate this information?  

Or perhaps the other way around.


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: perl5 ports and perl5.006

[Ade Lovett]

On Tue, Jun 27, 2000 at 07:05:39PM -0500, Ade Lovett wrote:
> Take a perl5 port at random.. I chose converters/p5-Convert-UU
> I had to make the following (kludgy) hack to the PLIST to make
> things package properly:
> 
> Index: PLIST
> ===
> RCS file: /home/src/FreeBSD/ports/converters/p5-Convert-UU/pkg/PLIST,v
> retrieving revision 1.6
> diff -u -r1.6 PLIST
> --- PLIST 1999/01/11 15:33:32 1.6
> +++ PLIST 2000/06/27 23:54:22
> @@ -1,5 +1,5 @@
>  bin/puudecode
>  bin/puuencode
>  lib/perl5/site_perl/%%PERL_VER%%/Convert/UU.pm
> -lib/perl5/site_perl/%%PERL_VER%%/%%PERL_ARCH%%/auto/Convert/UU/.packlist
> -@dirrm lib/perl5/site_perl/%%PERL_VER%%/%%PERL_ARCH%%/auto/Convert/UU
> +lib/perl5/site_perl/%%PERL_VER%%/mach/auto/Convert/UU/.packlist
> +@dirrm lib/perl5/site_perl/%%PERL_VER%%/mach/auto/Convert/UU
> 
> Notice how things appear to be being installed in a mach/ subdirectory,
> as opposed to the previous {i386,alpha}-freebsd/ subdirectories.


Bingo!  Found the problem.

Comparing src/gnu/usr.bin/perl/libperl/config.SH-elf.i386 between
-STABLE and -CURRENT we find:

(-CURRENT)
installsitearch='/usr/local/lib/perl5/site_perl/5.006/mach'

(-STABLE)
installsitearch='/usr/local/lib/perl5/site_perl/5.005/i386-freebsd'


This is what's causing the port PLISTs to malfunction.. -current
needs to be fixed appropriately, replacing 'mach' with
'{i386,alpha}-freebsd' as appropriate.

It also affects the threaded config file in the same directory.

No idea on the exact fix..  The perl mangling seems to be
"interesting" to say the least :)

Perhaps this will give Mark the ammo he needs to fix it.

That, and the patch to bsd.port.mk I gave should fix the perl
transition problem for ports.

-aDe

-- 
Ade Lovett, Austin, TX. [EMAIL PROTECTED]
FreeBSD: The Power to Serve http://www.FreeBSD.org/


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Laptop boot problem in current with GENERIC & GENERIC.hints

[Edwin Culp]

Thanks, Warner and Donn.  I have been successfully building a new world and
kernel every day, except with the latest perl problems the last two days.
Today's world is in process and looks like it is going to be successful.
This means that everything is up to date.   I haven't been able to boot any
of the kernels since the configuration changes.  My old kernel.works, as
suggested in the changes email, is keeping me going. I'm beginning to think
about and look for conflicts from very old current stuff that may be
obsolete.  (I've been running current on this machine for two years with a
daily cvsup and make world.)  I am also thinking that this may be processor
specific.  The machine is an older K6-2 at 300 Mhz.

I haven't optimized the compilation.  As I mentioned before, I wish that
were the problem.

Thanks very much for the suggestions and help.

Donn Miller wrote:

> On Wed, 28 Jun 2000, Warner Losh wrote:
>
> > In message <[EMAIL PROTECTED]> Edwin Culp writes:
> > : int=000d err= eft=00010046 eip=92eb
> > : eax=00fb ebx= ecx=0152 edx=03f6
> > : esi=000fbc92 edi=000f ebp=17b4 esp=17b4
> > : cs:eip=0f 32 89 45 1c 89 55 14-46 eb cf 80 65 31 fd eb
> > : ss:esp=08 00 00 00 1c 9c 00 00-fe 0e 00 00 d4 17 00 00
> > : System halted
> > :
> > : I copied this manually since I don't have a console on the laptop.
> >
> > I'd try booting the old kernel and seeing if doing a completely clean
> > build fixes this.  I'd also try to reboot after power off.
>
> It looks like the optimization bug that's been biting kernel builds
> lately.  When Peter Wemm made the overhaul to the kernel config, strange
> things started happening when optimization higher than -O was used to
> build the kernel.  For example, I got a Fatal Trap 12 immediately after
> the probe for available memory.  I was using -Os to compile my kernel at
> the time.  Backing down to -O for kernel builds solved this, and boots
> proceeded normally.
>
> - Donn
>
> To Unsubscribe: send mail to [EMAIL PROTECTED]
> with "unsubscribe freebsd-current" in the body of the message



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: XML driver config file to replace LINT

[Nik Clayton]

On Mon, Jun 26, 2000 at 03:02:07PM -0600, Warner Losh wrote:
> In message <[EMAIL PROTECTED]> Nik Clayton writes:
> : Another script would parse the above and generate HARDWARE.TXT.  And another 
> : could parse the above and spit out DocBook for the Handbook and FAQ.
> 
> There's some problems witht his.  the ed driver supports a whole raft
> of cards, but who can list them all?  

Initially, the driver's author.  Over time, user submissions will indicate
whether a particular driver supports a particular card.

All I want to do is make sure that this information only needs to be
maintained in one place.

N
-- 
Internet connection, $19.95 a month.  Computer, $799.95.  Modem, $149.95.
Telephone line, $24.95 a month.  Software, free.  USENET transmission,
hundreds if not thousands of dollars.  Thinking before posting, priceless.
Somethings in life you can't buy.  For everything else, there's MasterCard.
  -- Graham Reed, in the Scary Devil Monastery


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: XML driver config file to replace LINT

[Nik Clayton]

On Mon, Jun 26, 2000 at 03:02:07PM -0600, Warner Losh wrote:
> In message <[EMAIL PROTECTED]> Nik Clayton writes:
> : Another script would parse the above and generate HARDWARE.TXT.  And another 
> : could parse the above and spit out DocBook for the Handbook and FAQ.
> 
> There's some problems witht his.  the ed driver supports a whole raft
> of cards, but who can list them all?  

Initially, the driver's author.  Over time, user submissions will indicate
whether a particular driver supports a particular card.

All I want to do is make sure that this information only needs to be
maintained in one place.

N
-- 
Internet connection, $19.95 a month.  Computer, $799.95.  Modem, $149.95.
Telephone line, $24.95 a month.  Software, free.  USENET transmission,
hundreds if not thousands of dollars.  Thinking before posting, priceless.
Somethings in life you can't buy.  For everything else, there's MasterCard.
  -- Graham Reed, in the Scary Devil Monastery


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: XML driver config file to replace LINT

[Nik Clayton]

On Mon, Jun 26, 2000 at 03:06:57PM -0600, Warner Losh wrote:
> In message <[EMAIL PROTECTED]> Nik Clayton writes:
> : In my world, this XML file would be a replacement for many of the files
> : in src/sys/conf/.  Or, at the very least, those files would be generated
> : from this XML file.  As a developer, if you don't update the file the 
> : system won't even know about your driver (or option).
> 
> I'm not sure how well this would work.  Modules already obviate the
> need to update stuff in sys/conf.

S'fine, that's why this was cc'd to -current.  I need input from people 
involved in the future technical direction of FreeBSD.

How are we going to enumerate FreeBSD's supported hardware list in 5.x
and beyond?

N
-- 
Internet connection, $19.95 a month.  Computer, $799.95.  Modem, $149.95.
Telephone line, $24.95 a month.  Software, free.  USENET transmission,
hundreds if not thousands of dollars.  Thinking before posting, priceless.
Somethings in life you can't buy.  For everything else, there's MasterCard.
  -- Graham Reed, in the Scary Devil Monastery


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: XML driver config file to replace LINT

[Nik Clayton]

On Mon, Jun 26, 2000 at 03:06:57PM -0600, Warner Losh wrote:
> In message <[EMAIL PROTECTED]> Nik Clayton writes:
> : In my world, this XML file would be a replacement for many of the files
> : in src/sys/conf/.  Or, at the very least, those files would be generated
> : from this XML file.  As a developer, if you don't update the file the 
> : system won't even know about your driver (or option).
> 
> I'm not sure how well this would work.  Modules already obviate the
> need to update stuff in sys/conf.

S'fine, that's why this was cc'd to -current.  I need input from people 
involved in the future technical direction of FreeBSD.

How are we going to enumerate FreeBSD's supported hardware list in 5.x
and beyond?

N
-- 
Internet connection, $19.95 a month.  Computer, $799.95.  Modem, $149.95.
Telephone line, $24.95 a month.  Software, free.  USENET transmission,
hundreds if not thousands of dollars.  Thinking before posting, priceless.
Somethings in life you can't buy.  For everything else, there's MasterCard.
  -- Graham Reed, in the Scary Devil Monastery


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Check for ports updates

[Neil Blakey-Milner]

On Wed 2000-06-28 (15:13), Leif Neland wrote:
> Any reason not to put this into bsd.port.mk?
> 
> make update

It will break the system at least 20% of the time.  Change 20% to 100%
for gnome, kde, xpm, png, tiff, jpeg, and so forth.

Neil
-- 
Neil Blakey-Milner
Sunesi Clinical Systems
[EMAIL PROTECTED]


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Check for ports updates

[Leif Neland]

Any reason not to put this into bsd.port.mk?

make update




On Wed, 28 Jun 2000, Alexander Leidinger wrote:

> On 28 Jun, Leif Neland wrote:
> 
> > The steps needed for upgrading a package (from ports) would be:
> > 
> > make install
> > 
> > append foo-1.1/+REQUIRED_BY to foo-1.2/+REQUIRED_BY
> > 
> > traverse /var/db/pkg/* and remove foo-1.1/replace with foo-1.2
> > 
> > "subtract" foo-1.2/+CONTENT from foo-1.1/+CONTENT, only removing files
> > needed only in foo-1.1
> > 
> > remove /var/db/pkg/foo-1.1
> 
> cd /usr/ports/foo/bar
> make
> mv /var/db/pkg/bar-x.y.z/+REQUIRED_BY .
> pkg_delete bar-a.b.c
> make install
> mv +REQUIRED_BY /var/db/pkg/bar-d.e.f
> port_update.sh bar a.b.c d.e.f
> 
> port_update.sh:
> ---snip---
> #!/bin/sh
> 
> # Syntax: port_update.sh   
> # yes, I know it's ugly
> 
> if [ ! $3  ]; then
>   echo "$0   "
>   exit 1
> fi
> 
> for i in `cat /var/db/pkg/${1}-${3}/+REQUIRED_BY`; do
>   echo -n "Updating ${i}... "
>   perl -npi -e "s:\@pkgdep\ ${1}\-${2}:\@pkgdep\ ${1}\-${3}:g" 
>/var/db/pkg/${i}/+CONTENTS
>   echo "done"
> done
> ---snip---
> 
> Lazy disclaimer:
>  - works for me
>  - I think I know what I'm doing here (and what the limitations are
>(e.g. bar == libxyz && libmajorversion(old) !=
>libmajorversion(new))).
> 
> Bye,
> Alexander.
> 
> 



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Check for ports updates

[Alexander Leidinger]

On 28 Jun, Leif Neland wrote:

> The steps needed for upgrading a package (from ports) would be:
> 
> make install
> 
> append foo-1.1/+REQUIRED_BY to foo-1.2/+REQUIRED_BY
> 
> traverse /var/db/pkg/* and remove foo-1.1/replace with foo-1.2
> 
> "subtract" foo-1.2/+CONTENT from foo-1.1/+CONTENT, only removing files
> needed only in foo-1.1
> 
> remove /var/db/pkg/foo-1.1

cd /usr/ports/foo/bar
make
mv /var/db/pkg/bar-x.y.z/+REQUIRED_BY .
pkg_delete bar-a.b.c
make install
mv +REQUIRED_BY /var/db/pkg/bar-d.e.f
port_update.sh bar a.b.c d.e.f

port_update.sh:
---snip---
#!/bin/sh

# Syntax: port_update.sh   
# yes, I know it's ugly

if [ ! $3  ]; then
  echo "$0   "
  exit 1
fi

for i in `cat /var/db/pkg/${1}-${3}/+REQUIRED_BY`; do
  echo -n "Updating ${i}... "
  perl -npi -e "s:\@pkgdep\ ${1}\-${2}:\@pkgdep\ ${1}\-${3}:g" 
/var/db/pkg/${i}/+CONTENTS
  echo "done"
done
---snip---

Lazy disclaimer:
 - works for me
 - I think I know what I'm doing here (and what the limitations are
   (e.g. bar == libxyz && libmajorversion(old) !=
   libmajorversion(new))).

Bye,
Alexander.

-- 
  The best things in life are free, but the
expensive ones are still worth a look.

http://www.Leidinger.net  Alexander+Home @ Leidinger.net
  GPG fingerprint = 7423 F3E6 3A7E B334 A9CC  B10A 1F5F 130A A638 6E7E



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

The position of {boot,fixit}_crunch.conf

[Takahashi Yoshihiro]

Hi,

I'd like to move {boot,fixit}_crunch.conf from src/release to
src/release/$MACHINE because it needs to use different configuration
file between PC/AT and PC-98 to decrease the size of PC-98 fixit.flp
to 1.2MB.

In addition, alpha machines don't require the fdisk command, so I
think that it should be removed the fdisk command from
fixit_crunch.conf for alpha.

Comment?

---
Takahashi Yoshihiro
The Center for Information Science, Kogakuin Univ.


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: let badsect recog IFCHR in 5.0-C

[Alexander Langer]

Thus spake Warner Losh ([EMAIL PROTECTED]):

> /* That's OK */
> #if 0
> That's not
> #endif

Lol. I'd like to see one who does such things :)

Alex
-- 
cat: /home/alex/.sig: No such file or directory


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Check for ports updates

[Leif Neland]

On Tue, 6 Jun 2000, Thomas Schuerger wrote:

> Hi!
> 
> Is there already a tool that checks the installed ports for available
> updates in /usr/ports?
> 
> I've written such a tool, which seems to work fine already. Anyone
> interested?
> 
Now that such a tool (yours?) is in /etc/periodic/weekly, how does one
update a package?

If I update foo-1.1 to foo-1.2 by just doing a make install, I now have
both foo-1.1 and foo-1.2 listed in /var/db/pkg

And if I remove foo-1.1 first, it will complain that bar-2.3 requires foo.

If I remove foo-1.1 after installing foo-1.2, /usr/bin/foo will be
removed, I fear. Or will the MD5 values in +CONTENT check, that only files
belonging to foo-1.1 be removed? That can't be, because some files might
be the same in both packages.

The steps needed for upgrading a package (from ports) would be:

make install

append foo-1.1/+REQUIRED_BY to foo-1.2/+REQUIRED_BY

traverse /var/db/pkg/* and remove foo-1.1/replace with foo-1.2

"subtract" foo-1.2/+CONTENT from foo-1.1/+CONTENT, only removing files
needed only in foo-1.1

remove /var/db/pkg/foo-1.1

Does this look sane?

Have I overlooked the pkg_add --update command, or is it now "send
patches-time"?

Leif



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: let badsect recog IFCHR in 5.0-C

[Clive Lin]

On Wed, Jun 28, 2000 at 11:31:55AM +1000, Bruce Evans wrote:
> 1) Don't "comment out" dead code.  Remove it.
> 2) Use only #if 0 to "comment out code.  Comments are unsuitable for
>"commenting out" code in general, because the code might contain
>comments, and they are harder to edit and read.
> 
> Bruce

Okay .. the style(9) may like the diff now :-)

Besides the original bogus style,
is there any side effect/potential risk along with the code ?

Regards,
Clive
-- 
CirX - This site doesnt' exist.
9c  k9o h9 s1bg s1f, 7v  .y xqx a  sj m8r ffg1 vg5 a6 asox tmul h38 .
ant sj m8r ob ? 1fj mwby a1 tao vg5. soq df v ' .a. CirX.


--- badsect.c~  Tue Jun 27 22:30:57 2000
+++ badsect.c   Wed Jun 28 17:20:55 2000
@@ -118,11 +118,17 @@
err(3, "%s", name);
name_dir_end = name + strlen(name);
while ((dp = readdir(dirp)) != NULL) {
+   /*
+* Opening of a mounted on device is not allowed.
+* Attempt to open the raw device instead.
+*/
+   if (dp->d_name[0] != 'r')
+   continue ; 
strcpy(name_dir_end, dp->d_name);
if (lstat(name, &devstat) < 0)
err(4, "%s", name);
if (stbuf.st_dev == devstat.st_rdev &&
-   (devstat.st_mode & IFMT) == IFBLK)
+   (devstat.st_mode & IFMT) == IFCHR)
break;
}
closedir(dirp);
@@ -131,12 +137,6 @@
(u_long)stbuf.st_rdev, argv[1]);
exit(5);
}
-   /*
-* Opening of a mounted on device is not allowed.
-* Attempt to open the raw device instead.
-*/
-   memcpy(name_dir_end + 1, name_dir_end, strlen(name_dir_end) + 1);
-   *name_dir_end = 'r';
if ((fsi = open(name, O_RDONLY)) < 0)
err(6, "%s", name);
fs = &sblock;

Re: vnode_if.h: how should it be done ?

[Kris Kennaway]

On Sun, 25 Jun 2000, Warner Losh wrote:

> They aren't installed.  You must have a kernel tree to build the kld.
> Just add vnode_if.h to your SRCS and the rest happens automatically
> via bsd.kmod.mk.  It works great.  I do it all the time.  In fact, I
> added SYSDIR support to bsd.kmod.mk at Timing Solutions so that we
> could compile our drivers outside of the tree.

As I understand it, this is the only file which is required from the
kernel sources in order to build the arla port. ISTR FiST also needs it to
be present, and I also think I once came across another port that needed
it too. Is it really so wrong to make the lives of these third-party FS
tools harder by requiring full kernel sources to be present?

Kris

--
In God we Trust -- all others must submit an X.509 certificate.
-- Charles Forsythe <[EMAIL PROTECTED]>



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Bootstrapping perl ...

[George W. Dinolt]

Mark:

Got through the buildworld with a recent cvsup. I am now running with
this "world". Many thanks.

I am not sure who is supposed to manage the fact that
/usr/ports/Mk/bsd.ports.mk doesn't yet know about perl-5.006.   One
appears to need to set PERL_VER and PERL_VERSION. I just added an ."if
{OSVERSION} >= 56" , but this clearly isn't enough for people who
have yet to upgrade to the newest "current'. Does this mean that the
OSVERSION needs to be incremented to indicate the perl upgrade?

Isn't life wonderful. Solutions to problems always lead to new problems.

Regards, George Dinolt







To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: perl5 ports and perl5.006

[Doug Barton]

Ade Lovett wrote:
> 
> Got a couple of issues with perl5.006 that just went into -current
> in terms of perl5 ports.
> 
> The first one is relatively straightforward, being a simple
> versioning problem.  Following patch fixes that (OSVERSION
> may be slightly wrong, but it corresponds to what I have
> in my up-to-date src tree):
> 
> Index: bsd.port.mk
> ===
> RCS file: /home/src/FreeBSD/ports/Mk/bsd.port.mk,v
> retrieving revision 1.340
> diff -u -r1.340 bsd.port.mk
> --- bsd.port.mk 2000/06/16 21:52:40 1.340
> +++ bsd.port.mk 2000/06/27 23:59:03
> @@ -764,12 +764,17 @@
>  .endif
>  .endif
> 
> +.if ${OSVERSION} >= 56
> +PERL_VERSION=  5.006
> +PERL_VER=  5.006
> +.else
> +PERL_VER=  5.005
>  .if ${OSVERSION} >= 30
>  PERL_VERSION=  5.00503
>  .else
>  PERL_VERSION=  5.00502
> +.endif
>  .endif
> -PERL_VER=  5.005
>  PERL_ARCH= ${ARCH}-freebsd
>  PLIST_SUB+=PERL_VERSION=${PERL_VERSION} \
> PERL_VER=${PERL_VER} \

Second this one... My mrtg died as soon as I install'ed the world since
perl is now looking for /usr/local/lib/perl5/site_perl/5.006.

Doug
-- 
"Live free or die"
- State motto of my ancestral homeland, New Hampshire

Do YOU Yahoo!?


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Bootstrapping perl (Re: cvs commit: src/gnu/usr.bin/perl Mak

[Neil Blakey-Milner]

On Tue 2000-06-27 (02:41), John Baldwin wrote:
> 
> On 26-Jun-00 David O'Brien wrote:
> > On Mon, Jun 26, 2000 at 10:28:53PM +0200, Mark Murray wrote:
> >> > Since I'm now through it, I don't know the latest problem, but the
> >> > last thing I saw that the old lib got used with the new perl (or the
> >> > other way round) and that looks like it can be fixed with some path
> >> > adjustments.
> >> 
> >> The problem here is that miniperl needs to be built early enough
> >> to be in the path perl "proper" is done.
> >> 
> >> I thought build-tools was the answer; sadly that seems to be wrong.
> >> Now I'm looking at cross-tools (out of src/makefile.inc1).
> > 
> > Adding something to bootstrap-tools implies that we can't use the
> > installed miniperl (backward compatibility problem) or the host doesn't
> > have miniperl.  The bootstrap-tools built miniperl would then be used
> > throughout the build and install stages.
> 
> I think that's what we have in this case.  We need the newer version of
> miniperl to build perl, correct?

Yes, exactly.  We were only building libperl (statically) and miniperl
in the build-tools stage, so that miniperl (statically linked with
libperl) is available for the building later.

The choice to use crosstools is easier, since it by default installs the
tool into the "strict path", but Mark used build-tools and a path to
miniperl to do it instead, presumably since it is restricted to a very
minor bit of the tree.

(This reminds me that I have a skeleton document about 'make world' that
needs serious attention.  Too many projects.)

Neil
-- 
Neil Blakey-Milner
Sunesi Clinical Systems
[EMAIL PROTECTED]


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Bootstrapping perl (Re: cvs commit: src/gnu/usr.bin/perl Mak

[John Baldwin]

On 26-Jun-00 David O'Brien wrote:
> On Mon, Jun 26, 2000 at 10:28:53PM +0200, Mark Murray wrote:
>> > Since I'm now through it, I don't know the latest problem, but the
>> > last thing I saw that the old lib got used with the new perl (or the
>> > other way round) and that looks like it can be fixed with some path
>> > adjustments.
>> 
>> The problem here is that miniperl needs to be built early enough
>> to be in the path perl "proper" is done.
>> 
>> I thought build-tools was the answer; sadly that seems to be wrong.
>> Now I'm looking at cross-tools (out of src/makefile.inc1).
> 
> Adding something to bootstrap-tools implies that we can't use the
> installed miniperl (backward compatibility problem) or the host doesn't
> have miniperl.  The bootstrap-tools built miniperl would then be used
> throughout the build and install stages.

I think that's what we have in this case.  We need the newer version of
miniperl to build perl, correct?

-- 

John Baldwin <[EMAIL PROTECTED]> -- http://www.FreeBSD.org/~jhb/
PGP Key: http://www.cslab.vt.edu/~jobaldwi/pgpkey.asc
"Power Users Use the Power to Serve!"  -  http://www.FreeBSD.org/


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: -current of 3 hours ago, can't get GENERIC kernel compiled

[Martin Cracauer]

In <[EMAIL PROTECTED]>, Andreas Klemm wrote: 
> current of today, very recent.
> 
> Just to drop you a note.
> 
> cc -pipe -O -nostdinc -I/usr/src/sys/compile/GENERIC/../.. -I. -I/usr/include-o 
>aicasm aicasm_gram.o aicasm_scan.o aicasm.o aicasm_symbol.o  -ll
> ./aicasm -nostdinc -I- -I. -I../.. -I../../../include -o aic7xxx_seq.h -r 
>aic7xxx_reg.h ../../dev/aic7xxx/aic7xxx.seq
> ./aicasm: 725 instructions used
> make: don't know how to make ../../crypto/blowfish/bf_cbc.c. Stop
> 5.922u 1.575s 0:11.54 64.9% 1350+1091k 312+4io 103pf+0w

You need the full crypto stuff, including src/sys/crypto.  See example
cvsup files.

Yes, a HEADS up or an entry to /usr/src/UPDATE would have been great.

Martin
-- 
%
Martin Cracauer <[EMAIL PROTECTED]> http://www.cons.org/cracauer/
BSD User Group Hamburg, Germany http://www.bsdhh.org/


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Laptop boot problem in current with GENERIC & GENERIC.hints

[Donn Miller]

On Wed, 28 Jun 2000, Warner Losh wrote:

> In message <[EMAIL PROTECTED]> Edwin Culp writes:
> : int=000d err= eft=00010046 eip=92eb
> : eax=00fb ebx= ecx=0152 edx=03f6
> : esi=000fbc92 edi=000f ebp=17b4 esp=17b4
> : cs:eip=0f 32 89 45 1c 89 55 14-46 eb cf 80 65 31 fd eb
> : ss:esp=08 00 00 00 1c 9c 00 00-fe 0e 00 00 d4 17 00 00
> : System halted
> : 
> : I copied this manually since I don't have a console on the laptop.
> 
> I'd try booting the old kernel and seeing if doing a completely clean
> build fixes this.  I'd also try to reboot after power off.

It looks like the optimization bug that's been biting kernel builds
lately.  When Peter Wemm made the overhaul to the kernel config, strange
things started happening when optimization higher than -O was used to
build the kernel.  For example, I got a Fatal Trap 12 immediately after
the probe for available memory.  I was using -Os to compile my kernel at
the time.  Backing down to -O for kernel builds solved this, and boots
proceeded normally.

- Donn



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: XML driver config file to replace LINT

[Warner Losh]

Any reason that the .c/.h files of the drivers couldn't be used to
generate this information?  I think it would greatly enhance the
ability of the aintainer to update it.

Warner


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: let badsect recog IFCHR in 5.0-C

[Warner Losh]

In message <[EMAIL PROTECTED]> Alexander Langer writes:
: Thus spake Dan Papasian ([EMAIL PROTECTED]):
: 
: > > > +   // un-needed.
: > > > +   // memcpy(name_dir_end + 1, name_dir_end, strlen(name_dir_end) + 1);
: > > > +   // *name_dir_end = 'r';
: > > Use /* */ comments or #if 0.
: > Actually, there was just a discussion on comp.lang.c about how
: > they aren't interchangable.  The 'un-needed' part will have
: > to be commented with /* */ because the preprocessor will
: > parse tokens before it evaluates conditionals.
: 
: Could you explain that paragraph? I really can't figure out what you
: try to say or what it has to do with my comment, respectively :)

/* That's OK */
#if 0
That's not
#endif

Warner


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Laptop boot problem in current with GENERIC & GENERIC.hints

[Warner Losh]

In message <[EMAIL PROTECTED]> Edwin Culp writes:
: int=000d err= eft=00010046 eip=92eb
: eax=00fb ebx= ecx=0152 edx=03f6
: esi=000fbc92 edi=000f ebp=17b4 esp=17b4
: cs:eip=0f 32 89 45 1c 89 55 14-46 eb cf 80 65 31 fd eb
: ss:esp=08 00 00 00 1c 9c 00 00-fe 0e 00 00 d4 17 00 00
: System halted
: 
: I copied this manually since I don't have a console on the laptop.

I'd try booting the old kernel and seeing if doing a completely clean
build fixes this.  I'd also try to reboot after power off.

Warner


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Re: Boot Oddities

[Dampure, Pierre Y.]

"Dampure, Pierre Y." wrote:
> 
> Apologies if this ought to be in -questions rather than -current, but
> since I first noticed the problem on 5.0-2621-CURRENT...
> 
> The above snapshot was installed on a brand new system (OR840, 2x733EB,
> 512MB RDRAM, 1xAHA2940UW, 4x9.1Gb Barracuda). The disk organisation is
> as follows:
> 
> da0:one NTFS partition (W2K), FreeBSD multiboot MBR installed
> da1:fully dedicated, FreeBSD, has the root partition (da1s1a)
> da2:fully dedicated, FreeBSD
> da3:fully dedicated, FreeBSD
> 
> Since this is a recent snapshot, I assume it defaults to use loader(8).
> 
> When the system boots, I get the usual F1/F5 choice, as follows:
> 
> F1  ??
> F5  Drive 1
> 
> Default: F5
> 
> Selecting F5 (or letting the timer run to completion) leads to the
> following:
> 
> Disk error 0x1 (lba=0xe0040)
> No /boot/loader
> 
> >> FreeBSD/i386 BOOT
> Default: 1:da(1,a)/kernel
> boot:
> 
> then proceeding to load /kernel, rather than /boot/loader (which is why
> I posted about problems with nlist (aka top/vmstat/systat) a few days
> ago).
> 
> If I hit a key early enough, I get:
> 
> >> FreeBSD/i386 BOOT
> Default: 1:da(1,a)/boot/loader
> boot:
> 
> Hitting RETURN then leads to:
> 
> Disk error 0x1 (lba=0xe0040)
> No /boot/loader
> 
> after which I need to enter 1:da(1,a)/kernel to proceed.
> 
> Needless to say, /boot/loader IS on da1s1a (with its usual assortment of
> .rc and .conf in required places), which is why I am a bit surprised.
> 
> Can anyone shed a light on the above? I now know how to get around the
> nlist issue, but this is not the recommended method...
> 
> Configuration files available upon request.
> 
> Best Regards,
> 
> PYD

FWIW, I DID have a look at the archives before posting... the only
thread that looks remotely related is:

http://www.FreeBSD.org/cgi/getmsg.cgi?fetch=3392426+3394340+/usr/local/www/db/text/1999/freebsd-current/19990509.freebsd-current

but this is in the case of IDE drives, not SCSI... CAM reports my drives
as having a (255H 63S/T 1106C) geometry, so even with a 512Mb root
partition I am still under the 1023 cylinders limit (if even applicable
to SCSI devices).

Best Regards,

PYD


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message