Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.

Dismiss

stable-digest V5 #801

1 view

Skip to first unread message

owner-freebsd...@freebsd.org

unread,

Feb 23, 2003, 6:45:51 AM2/23/03

stable-digest Sunday, February 23 2003 Volume 05 : Number 801

In this issue:
Kernel Panic : ATA problem in STABLE.
Re: Kernel Panic : ATA problem in STABLE.
Re: Kernel Panic : (Not ATA related) problem in STABLE.
changes in FreeBSD-STABLE src/sys/i386/i386/nexus.c 1.26.2.9
Re: changes in FreeBSD-STABLE src/sys/i386/i386/nexus.c 1.26.2.9
Re: Maxtor Firewire Drives
Re: Maxtor Firewire Drives
4.8-PRERELEASE Kernel Panic
Re: 4.8-PRERELEASE Kernel Panic
Re: 4.8-PRERELEASE Kernel Panic
4.7 -> 4.8 woes. kernel now panics on bootup with "still using grody create_intr interface" message
RE: Raising KVA_PAGES breaks linux emulation?
Elaboração de cartas comerciais
panic: still using grody create_intr interface
RE: Raising KVA_PAGES breaks linux emulation?
RE: Raising KVA_PAGES breaks linux emulation?
RE: Raising KVA_PAGES breaks linux emulation?
Re: Maxtor Firewire Drives
Re: Compaq WL200 wireless card trouble
Re: STABLE buildworld/buildkernel on 4.5-R apparently worked
Re-post: 4-STABLE panics in usbd_transfer after plugging in PQI Travel Flash
Re: STABLE buildworld/buildkernel on 4.5-R apparently worked
Re: Maxtor Firewire Drives
Re: Maxtor Firewire Drives
Re: Maxtor Firewire Drives
Re: Maxtor Firewire Drives
Re: Maxtor Firewire Drives
Re: Raising KVA_PAGES breaks linux emulation?
Re: Compaq WL200 wireless card trouble
Re: Compaq WL200 wireless card trouble

----------------------------------------------------------------------

Date: Sat, 22 Feb 2003 07:33:57 -0300
From: "Fabio Vilan" <fa...@isec.com.br>
Subject: Kernel Panic : ATA problem in STABLE.

After "make kernel" using latest stable (as of Feb 22) and booting :

- ---
atapci0: <Intel PIIX4 ATA33 controller> port 0xf000-0xf00f at device 7.1 on
pci0
ata0: at 0x1f0 irq14 on atapci0
panic: still using grody create_intr interface
Uptime: 0s
Automatic reboot in 15 seconds - press a key on the console to abort.
- ---

There were no problems in stable of 3 days ago (Feb 19)
Might have been those recent (Feb 20) changes in ATA at
/src/sys/dev/ata/ata*
http://docs.freebsd.org/cgi/getmsg.cgi?fetch=2449713+0+current/cvs-all
http://docs.freebsd.org/cgi/getmsg.cgi?fetch=2447437+0+current/cvs-all
( I said might ... )

Thanks
- --
Fabio Vilan <fa...@isec.com.br>
"Any sufficiently advanced technology is indistinguishable from magic."
Arthur C. Clarke's Third Law, from Profiles of the Future

------------------------------

Date: Sat, 22 Feb 2003 17:12:53 +0600 (OMST)
From: El Vampiro <vam...@rusunix.org>
Subject: Re: Kernel Panic : ATA problem in STABLE.

FV> panic: still using grody create_intr interface
FV> Uptime: 0s
FV> Automatic reboot in 15 seconds - press a key on the console to abort.
FV> ---
FV>
FV> There were no problems in stable of 3 days ago (Feb 19)
FV> Might have been those recent (Feb 20) changes in ATA at
FV> /src/sys/dev/ata/ata*
FV> http://docs.freebsd.org/cgi/getmsg.cgi?fetch=2449713+0+current/cvs-all
FV> http://docs.freebsd.org/cgi/getmsg.cgi?fetch=2447437+0+current/cvs-all
this panic not caused by ATA-code
this is due to pretty changes in nexus.c (1.26.2.9)
using 1.26.2.8 version of nexus.c you can avoid panic

- --
http://aquatique.rusunix.org
http://rusunix.org

------------------------------

Date: Sat, 22 Feb 2003 08:46:07 -0300
From: "Fabio Vilan" <fa...@isec.com.br>
Subject: Re: Kernel Panic : (Not ATA related) problem in STABLE.

You're right Vampiro,

The ata changes were in CURRENT, not RELENG_4. I need some sleep :)
Anyway someone please have this problem fixed as soon as possible.

- --
Fabio Vilan <fa...@isec.com.br>
"Any sufficiently advanced technology is indistinguishable from magic."
Arthur C. Clarke's Third Law, from Profiles of the Future

------------------------------

Date: Sat, 22 Feb 2003 14:06:41 +0100
From: Oliver Eikemeier <eike...@fillmore-labs.com>
Subject: changes in FreeBSD-STABLE src/sys/i386/i386/nexus.c 1.26.2.9

After rebuilding my kernel with FreeBSD-STABLE booting fails with:

panic: still using grody create_intr interface

when loading the ata0 driver (FreeBSD 4.8-PRERELEASE, 440 LX based
motherboard)

After I changed line 495 of src/sys/i386/i386/nexus.c rev 1.26.2.9 to

switch (flags & ~INTR_TYPE_FAST) {

everything works as expected. Could this be a typo in the recent
changes? Otherwise I'm happy to post my dmesg.

Thanks
Oliver

------------------------------

Date: Sat, 22 Feb 2003 06:09:17 -0700 (MST)
From: "M. Warner Losh" <i...@bsdimp.com>
Subject: Re: changes in FreeBSD-STABLE src/sys/i386/i386/nexus.c 1.26.2.9

cTimeTag Unit 3 Channel 3:Got Tag 472221 84751:086171223:200000002 Correlated Timescale
In message: <3E5775E...@fillmore-labs.com>
Oliver Eikemeier <eike...@fillmore-labs.com> writes:
: After rebuilding my kernel with FreeBSD-STABLE booting fails with:
:
: panic: still using grody create_intr interface
:
: when loading the ata0 driver (FreeBSD 4.8-PRERELEASE, 440 LX based
: motherboard)
:
: After I changed line 495 of src/sys/i386/i386/nexus.c rev 1.26.2.9 to
:
: switch (flags & ~INTR_TYPE_FAST) {
:
: everything works as expected. Could this be a typo in the recent
: changes? Otherwise I'm happy to post my dmesg.

Nope, I botched the merge :-( 1.26.9.10 fixes this problem. Sorry
for the bother.

Warner

------------------------------

Date: Sat, 22 Feb 2003 07:33:19 -0600
From: "Aaron Wohl" <fre...@soith.com>
Subject: Re: Maxtor Firewire Drives

> The enclosure I have gets this ->
> firewire0: New S400 device ID:0030e001e0017089
> firewire0: Device SBP-II
> sbp0: <SBP2/SCSI over firewire> on firewire0
> da1 at sbp0 bus 0 target 0 lun 0
> da1: <Wise Adv Wise Advanced ID 0028> Fixed Simplified Direct Access
> SCSI-4 device
> da1: 50.000MB/s transfers
> da1: 76319MB (156301488 512 byte sectors: 255H 63S/T 9729C)
> sbp0:0:0 request timeout ... agent reset

I note this is NOT a maxtor dv5000, that doesnt mount at all. The error
you see in the log sbp0:0:0 request timeout ... agent reset you only have
one of. Depending on the timeing you get a lot more of them - and long
delays and trashed data. It may depend on machine speed as well most of
the machines we tried this on where 2600 mhz or 2800 mhz with fast disks.
I notice that disk your trying is an 80gig. We where trying larger
disks. I just put an 80 gig in an ADS enclosure and its getting a lot
less errors, but still some. Perhaps there are more problems for 120+
gigabyte disks. I don't think firewire disks are ready for use in
backups yet on freebsd... im having no problems using 120gb firewire
disks with retrospect remote under win xp though (on the same hardware
exactly we tried freebsd on).

------------------------------

Date: 23 Feb 2003 00:37:23 +1030
From: "Daniel O'Connor" <doco...@gsoft.com.au>
Subject: Re: Maxtor Firewire Drives

On Sun, 2003-02-23 at 00:03, Aaron Wohl wrote:
> I note this is NOT a maxtor dv5000, that doesnt mount at all. The error
> you see in the log sbp0:0:0 request timeout ... agent reset you only have
> one of. Depending on the timeing you get a lot more of them - and long
> delays and trashed data. It may depend on machine speed as well most of

Hmm, well I did notice it sometimes caused a delay, but I couldn't
really characterise it.

It seemed fairly repeatable by running iozone -a.

> the machines we tried this on where 2600 mhz or 2800 mhz with fast disks.
> I notice that disk your trying is an 80gig. We where trying larger
> disks. I just put an 80 gig in an ADS enclosure and its getting a lot

Hmm odd. This is on a Athlon XP2000+ the drive is a Western Digital
WD800BB 7200rpm.

I have a Mapower enclosure (http://www.mapower.com.tw/).
I tried it on my laptop (Inspiron 8000 with a 700Mhz PIII) and it seemed
to work fine.

> less errors, but still some. Perhaps there are more problems for 120+
> gigabyte disks. I don't think firewire disks are ready for use in
> backups yet on freebsd... im having no problems using 120gb firewire
> disks with retrospect remote under win xp though (on the same hardware
> exactly we tried freebsd on).

Well, hopefully it will get more patches. If you got panics etc, it
would be useful to see back traces and the like. Firewire, and USB to a
lesser extent, is fairly new, so it hasn't been tested as heavily as
other things.

The only way that happens is if end users try it and report what
happens.

I (well work is) am going to purchase a Maxtor enclosure so I will see
how it goes.

- --
Daniel O'Connor software and network engineer
for Genesis Software - http://www.gsoft.com.au
"The nice thing about standards is that there
are so many of them to choose from."
-- Andrew Tanenbaum
GPG Fingerprint - 9A8C 569F 685A D928 5140 AE4B 319B 41F4 5D17 FDD5

------------------------------

Date: Sat, 22 Feb 2003 10:06:39 -0500 (EST)
From: "Michael Sharp" <m...@probsd.org>
Subject: 4.8-PRERELEASE Kernel Panic

I pulled down RELENG_4 at 11:45pm ( Feb 21 ) EST via cvsup and 'made
world' like I have for the past two years with no problems. However, after
doing so last night, I get the following on a reboot:

panic: still using grody create_intr interface
uptime: 0s
Automatic reboot in 15 seconds...

I am able to boot into kernel.old after unloading the default /kernel

Has something changed in 4.8-PRERELEASE that I missed in regards to make
world or could someone point me towards finding out what " grody
create_intr interface " even is ?

Michael

------------------------------

Date: Sat, 22 Feb 2003 16:38:02 +0100
From: Oliver Brandmueller <o...@e-Gitt.NET>
Subject: Re: 4.8-PRERELEASE Kernel Panic

Hello.

On Sat, Feb 22, 2003 at 10:06:39AM -0500, Michael Sharp wrote:
> I pulled down RELENG_4 at 11:45pm ( Feb 21 ) EST via cvsup and 'made
> world' like I have for the past two years with no problems. However, after
> doing so last night, I get the following on a reboot:
>
> panic: still using grody create_intr interface
> uptime: 0s
> Automatic reboot in 15 seconds...
>
> I am able to boot into kernel.old after unloading the default /kernel
>
> Has something changed in 4.8-PRERELEASE that I missed in regards to make
> world or could someone point me towards finding out what " grody
> create_intr interface " even is ?

I have the same panic on one system, but not on another another (which
resp. has sources from yesterday, but I did not see kernel changes in
the meantime).

The machine with panic is:

CPU: AMD Athlon(TM) XP1500+ (1343.05-MHz 686-class CPU)
Origin = "AuthenticAMD" Id = 0x662 Stepping = 2
Features=0x383f9ff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,SEP,MTRR,PGE,MCA,CMOV,PA
T,PSE36,MMX,FXSR,SSE>
AMD Features=0xc0480000<MP,AMIE,DSP,3DNow!>
real memory = 536788992 (524208K bytes)
avail memory = 518713344 (506556K bytes)

It's on an Asus A7V266-E Motherboard and / is on /dev/ad4s1a which is
on the Promise controller.

The machine without the panic is an Athlon 1 GHz on an Asus A7V133
motherboard and / on the first IDE channel of the motherboard.

- - Oliver

------------------------------

Date: Sat, 22 Feb 2003 16:56:02 +0100
From: Thierry Herbelot <thi...@herbelot.com>
Subject: Re: 4.8-PRERELEASE Kernel Panic

Le Saturday 22 February 2003 16:38, Oliver Brandmueller a écrit :
> Hello.
>
> On Sat, Feb 22, 2003 at 10:06:39AM -0500, Michael Sharp wrote:
> > I pulled down RELENG_4 at 11:45pm ( Feb 21 ) EST via cvsup and 'made
> > world' like I have for the past two years with no problems. However,
> > after doing so last night, I get the following on a reboot:
> >
> > panic: still using grody create_intr interface
> > uptime: 0s
> > Automatic reboot in 15 seconds...
> >

already corrected : re-cvsup and make world

TfH

------------------------------

Date: Sun, 23 Feb 2003 03:21:21 +1100
From: Steve Horan <sjh-...@horan.net.au>
Subject: 4.7 -> 4.8 woes. kernel now panics on bootup with "still using grody create_intr interface" message

Hi,

I've just cvsup'd from 4.7-STABLE (built from Nov 15 2002 sources) to
today's 4.8-STABLE sources (Feb 22 2003)

The machine now refuses to boot.

The panic message is:
"still using grody create_intr interface"

Any pointers would be greatly appreciated

regards

sjh

------------------------------

Date: Sat, 22 Feb 2003 09:38:41 -0800
From: "Luoqi Chen" <lc...@briontech.com>
Subject: RE: Raising KVA_PAGES breaks linux emulation?

> Now, when I did this the first time, I got a bunch of thread issues, so
> Tor sent me a patch (included in previous email) to fix it in libc_r ...
> does the linux module make use of KVA_PAGES as defined in the kernel
> config file, or do I have to change KVA_PAGES in /sys/i386/include/pmap.h
> for the modules to recognize the change?
>
KVA_PAGES if defined in config goes to opt_global.h and I don't think
module builds uses this file. You may put it in make.conf if you don't
want to mess with pmap.h

- -lq

------------------------------

Date: Sat, 22 Feb 2003 14:11:20 -0300
From: "Redação Comercial" <redac...@bol.com.br>
Subject: Elaboração de cartas comerciais

COMUNICADO IMPORTANTE!!

Estamos lançando o KIT DE CARTAS COMERCIAIS, que sana suas dúvidas na
elaboração de: agradecimentos, atestados e declarações, avisos, cartas de
cobrança, cartas em inglês, comunicados, convites, contratos, propostas,
empregos, solicitações e pedidos, telegramas, cartas por e-mail, etc.
Composto de 02 (dois) disquetes com 150 modelos de documentos cada um,
mais livreto 20 páginas, com técnicas de redação comercial. Indicado para:
secretárias em geral, gerências, Rh, executivos, estudantes e empresas de
toda ordem.
Este kit possui um preço ínfimo em relação ao que poderá gerar no
aperfeiçoamento da comunicação de sua empresa.

Acesse nossa Home Page para mais detalhes:

http://www.redacaodecartas.ihp.com.br

Ps: Caso não queira receber novas mensagens e novidades sobre esse
assunto, acesse:

http://www.remova-me.ihp.com.br

------------------------------

Date: Sat, 22 Feb 2003 13:19:25 -0500 (EST)
From: Mikhail Teterin <m...@corbulon.video-collage.com>
Subject: panic: still using grody create_intr interface

Hello!

This is with today's 4.8-PRERELEASE boot.flp :-( The machine is an old
but working PPro-200 with two IDE drives (850Mb and 240Mb) and an ATAPI
CD-ROM.

I get the panic in Subject at boot, right after the:

atapci0: <Intel PIIX3 ATA controller> port 0xffa0-0xffaf at device 7.1 on pci0
ata0: at 0x1f0 irq 14 on atapci0

The ata1 is never given a chance :-\

Any clues? Thanks,

-mi

------------------------------

Date: Sat, 22 Feb 2003 15:30:57 -0400 (AST)
From: "Marc G. Fournier" <scr...@hub.org>
Subject: RE: Raising KVA_PAGES breaks linux emulation?

On Sat, 22 Feb 2003, Luoqi Chen wrote:

> > Now, when I did this the first time, I got a bunch of thread issues, so
> > Tor sent me a patch (included in previous email) to fix it in libc_r ...
> > does the linux module make use of KVA_PAGES as defined in the kernel
> > config file, or do I have to change KVA_PAGES in /sys/i386/include/pmap.h
> > for the modules to recognize the change?
> >
> KVA_PAGES if defined in config goes to opt_global.h and I don't think
> module builds uses this file. You may put it in make.conf if you don't
> want to mess with pmap.h

Okay, is there any good documentation on raising KVA_PAGES? The only
thing I've found is in LINT, and it mentioned nothing about setting it
anywhere but in the kernel config :(

For instance, since setting KVA_PAGES in the kernel config doesn't cover
everything, and, as I've experienced so far, causes some serious problems,
would it be better to suggest that someone sets it in make.conf? From a
quick perusal of /etc/defaults/make.conf, my guess is that I'd have to
set:

COPTFLAGS= -O -pipe -DKVA_PAGES=512
- -and-
CFLAGS= -O -pipe -DKVA_PAGES=512

Will that cover it all, or is there a problem with doing it that way? Or
is there a better way of doing this?

------------------------------

Date: Sat, 22 Feb 2003 11:56:20 -0800
From: "Luoqi Chen" <lc...@briontech.com>
Subject: RE: Raising KVA_PAGES breaks linux emulation?

> For instance, since setting KVA_PAGES in the kernel config doesn't cover
> everything, and, as I've experienced so far, causes some serious problems,
> would it be better to suggest that someone sets it in make.conf? From a
> quick perusal of /etc/defaults/make.conf, my guess is that I'd have to
> set:
>
> COPTFLAGS= -O -pipe -DKVA_PAGES=512
> -and-
> CFLAGS= -O -pipe -DKVA_PAGES=512
>
> Will that cover it all, or is there a problem with doing it that way? Or
> is there a better way of doing this?
>
Module builds seem to use CFLAGS (while kernel builds use COPTFLAGS).
A better way is to have kernel Makefile passing opt_global.h to the
modules. The best solution for this particular problem would probably
be changing USRSTACK to a global variable instead of a #define.

- -lq

------------------------------

Date: Sat, 22 Feb 2003 15:56:44 -0400 (AST)
From: "Marc G. Fournier" <scr...@hub.org>
Subject: RE: Raising KVA_PAGES breaks linux emulation?

On Sat, 22 Feb 2003, Luoqi Chen wrote:

> > For instance, since setting KVA_PAGES in the kernel config doesn't cover
> > everything, and, as I've experienced so far, causes some serious problems,
> > would it be better to suggest that someone sets it in make.conf? From a
> > quick perusal of /etc/defaults/make.conf, my guess is that I'd have to
> > set:
> >
> > COPTFLAGS= -O -pipe -DKVA_PAGES=512
> > -and-
> > CFLAGS= -O -pipe -DKVA_PAGES=512
> >
> > Will that cover it all, or is there a problem with doing it that way? Or
> > is there a better way of doing this?
> >
> Module builds seem to use CFLAGS (while kernel builds use COPTFLAGS).

Right, and, as was found out, this doesn't just affect kernel+modules, but
libc_r as well ... anything else it might affect? :(

> A better way is to have kernel Makefile passing opt_global.h to the
> modules. The best solution for this particular problem would probably
> be changing USRSTACK to a global variable instead of a #define.

I realize we're in a freeze for 4.8-RELEASE, but sicne KVA_PAGES is
documented, and its use does break things, any chance on getting this
fixed before the RELEASE?

------------------------------

Date: Sat, 22 Feb 2003 15:48:29 -0500 (EST)
From: Charles Sprickman <sp...@inch.com>
Subject: Re: Maxtor Firewire Drives

On Fri, 22 Feb 2003, Daniel O'Connor wrote:

> On Sat, 2003-02-22 at 07:03, Aaron Wohl wrote:
> > The maxtor 5000xx series does not work on 4.7 or 5.0 for all variations
> > of stable/currrent/standard. This is for USB1 USB2 and firewire.

> Hmm.. can you explain more about the firewire bit?
> The enclosure I have gets this ->
> firewire0: New S400 device ID:0030e001e0017089
> firewire0: Device SBP-I

I've asked before, but maybe things have changed since then... What cards
are people who are doing devel work using? Since so many are "no name" or
not identified as a brand on many online stores, can someone post store
and model info?

As for drive enclosures, I was planning on using something with the Oxford
911 chipset. Might this work?

I'm planning on using one enclosure to share my CD-R drive amongst
machines, and using the other for a hard drive (for backup purposes at
home). Is this a bad idea?

Thanks,

Charles

> sbp0: <SBP2/SCSI over firewire> on firewire0
> da1 at sbp0 bus 0 target 0 lun 0
> da1: <Wise Adv Wise Advanced ID 0028> Fixed Simplified Direct Access SCSI-4 device
> da1: 50.000MB/s transfers
> da1: 76319MB (156301488 512 byte sectors: 255H 63S/T 9729C)
> sbp0:0:0 request timeout ... agent reset
>
> It seems to work fine too.
>
> > Some of the older enclosers you add your own drive to kind of work with
> > firewire. However id advise you do transfer some large files (say
> > 200megabyte) to the disk make many copies then check the md5 checksums.
> > Also check the dmesg logs. The other enclosures I tried worked ok for
> > smaller files but go timeouts and trashed data on largers files.
>
> OK, I just copied a 600Mb file over and it verifies properly.
>
> > The adaptec fireconnect and universal (usb2+firewirecards) seem to work
> > fine by themselfs. All of controller cards seemed ok. Its just something
> > messed up with using them for disks.
>
> Bummer :(
>
> > Tests where conducted on 4.7 (standard,stable), 5.0 (standard, current)
> > using a 120gb maxtor DV5000 a 250gb maxtor DV5000, and ADS enclosure with
> > a maxtor disk added to it, a USB2 enclosure "USB2.0 slim its labled" from
> > compusa. This with 5 different computers (various motherboards).
>
> OK.. Under -stable I had to get a patch for the sbp code to use M_NOWAIT
> otherwise it would panic fairly easily..
>
> > We ended up going with tape backups.
>
> Not an option for me :)
>
> --
> Daniel O'Connor software and network engineer
> for Genesis Software - http://www.gsoft.com.au
> "The nice thing about standards is that there
> are so many of them to choose from."
> -- Andrew Tanenbaum
> GPG Fingerprint - 9A8C 569F 685A D928 5140 AE4B 319B 41F4 5D17 FDD5
>
>
> To Unsubscribe: send mail to majo...@FreeBSD.org
> with "unsubscribe freebsd-stable" in the body of the message
>

------------------------------

Date: Sat, 22 Feb 2003 14:48:36 -0700 (MST)
From: "M. Warner Losh" <i...@bsdimp.com>
Subject: Re: Compaq WL200 wireless card trouble

In message: <2003022121...@freebie.xs4all.nl>
Wilko Bulte <w...@freebie.xs4all.nl> writes:
: pcic0: <Cirrus Logic PD6729/6730 PCI-PCMCIA Bridge> port 0xcc00-0xcc03 irq

This doesn't work before 4.7-stable (and maybe I haven't MFC'd all the
changes form current to make it work).

Warner

------------------------------

Date: Sat, 22 Feb 2003 14:48:11 -0800 (PST)
From: Brian O'Shea <b_o...@yahoo.com>
Subject: Re: STABLE buildworld/buildkernel on 4.5-R apparently worked

- --- Peter Wu <pet...@canada.com> wrote:
> "Kevin Oberman" <obe...@es.net> writes:
>
[...]
> > Yes. Be very sure that you run 'mergemaster -p' before you attempt to
> > install the new system. If you don't, installworld might fail in a
> > very ugly way.
>
> 4.5-RELEASE does not have a mergemaster that knows -p switch.

I used src/usr.sbin/mergemaster/mergemaster.sh from the cvsup'ed directory
instead and it worked. In -STABLE, the -p option tells mergemaster to
only compare "crucial files". The only differences that it reported were
in /etc/master.passwd, /etc/group, and /etc/make.conf (make.conf didn't exist
on my 4.5-RELEASE system; there was only an /etc/defaults/make.conf).

FWIW, the upgrade went cleanly and my system works (except for a panic,
which I reported on this list earlier today but apparently hasn't been
posted yet [???]).

Thanks to all who replied,
- -brian

__________________________________________________
Do you Yahoo!?
Yahoo! Tax Center - forms, calculators, tips, more
http://taxes.yahoo.com/

------------------------------

Date: Sat, 22 Feb 2003 15:19:23 -0800 (PST)
From: Brian O'Shea <b_o...@yahoo.com>
Subject: Re-post: 4-STABLE panics in usbd_transfer after plugging in PQI Travel Flash

(re-posting sans URLs and attachments since the original hasn't posted yet)

4-STABLE panics in usbd_transfer after plugging in PQI Travel Flash (USB
compact flash reader).

# uname -a
FreeBSD apsara 4.8-PRERELEASE FreeBSD 4.8-PRERELEASE #0: Fri Feb 21 18:13:15
PST 2003 root@apsara:/usr/local/cvsup/4-STABLE/src/sys/compile/APSARA_DBG
i386

#16 0xc0262c59 in usbd_transfer (xfer=0xc0e23b80) at ../../dev/usb/usbdi.c:270
270 err = pipe->methods->transfer(xfer);
(kgdb) p pipe
$1 = 0x6

If anyone would like to take a crack at debugging it, I can arrange to
send you the debug kernel and core file (if those are too big, I can
just send you the kernel config file, GDB backtrace, dmesg output, or
any other information that you might want).

I'm taking a look at it too, but I'm not much of a kernel hacker. If
you think that I should file a PR, let me know and I'll do it.

Thanks,
- -brian

__________________________________________________
Do you Yahoo!?
Yahoo! Tax Center - forms, calculators, tips, more
http://taxes.yahoo.com/

------------------------------

Date: Sat, 22 Feb 2003 15:27:39 -0800
From: "Kevin Oberman" <obe...@es.net>
Subject: Re: STABLE buildworld/buildkernel on 4.5-R apparently worked

> "Kevin Oberman" <obe...@es.net> writes:
>
> >> Greetings,
> >>
> >> I am in the process of upgrading my FreeBSD 4.5-RELEASE system to -STABLE.
> >>
> >> Last night (Feb 21 03:30 UTC) I cvsup'ed RELENG_4, did a buildworld and
> >> buildkernel, and it seemed to go without a hitch. I'm used to running
> >> into a hitch! :-)
> >>
> >> Before I go ahead and installkernel/installworld, are there any known
> >> problems with this upgrade path? Except for the pam.conf issue, I don't
> >> see any pertinent problems in the -stable archives or the UPDATING file
> >> (at least not with any of the software that I use -- I'm not running
> >> sendmail, for example). Am I missing any sneaky little problems?
> >
> > Yes. Be very sure that you run 'mergemaster -p' before you attempt to
> > install the new system. If you don't, installworld might fail in a
> > very ugly way.
>
> 4.5-RELEASE does not have a mergemaster that knows -p switch.

Sorry. It's amazing how easy it is to forget the obvious once it's no longer an issue. The '-p' option was added to mergemaster to deal with the problem that will bite you. So it's not there, yet. But it is in your source tree, so you can run it from there.

You need to explicitly invoke it with the full path which should be:
/usr/obj/usr/src/usr.sbin/mergemaster

Of course, once you have completed this update, you will not have to do this again as '-p' is now a permanent part of mergemaster.

- --
R. Kevin Oberman, Network Engineer
Energy Sciences Network (ESnet)
Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab)
E-mail: obe...@es.net Phone: +1 510 486-8634

------------------------------

Date: 23 Feb 2003 11:26:40 +1030
From: "Daniel O'Connor" <doco...@gsoft.com.au>
Subject: Re: Maxtor Firewire Drives

On Sun, 2003-02-23 at 07:18, Charles Sprickman wrote:
> I've asked before, but maybe things have changed since then... What cards
> are people who are doing devel work using? Since so many are "no name" or
> not identified as a brand on many online stores, can someone post store
> and model info?
>
> As for drive enclosures, I was planning on using something with the Oxford
> 911 chipset. Might this work?

My card has an Agere chip on it, the one in my laptop is Ti..
Agere -> fwohci0: <Lucent FW322/323> mem 0xdc042000-0xdc042fff irq 10 at device 8.0 on pci0
Ti -> fwohci0: <1394 Open Host Controller Interface> mem 0xf6ff8000-0xf6ffbfff,0xf6ffd800-0xf6ffdfff irq 11 at device 15.2 on pci2

> I'm planning on using one enclosure to share my CD-R drive amongst
> machines, and using the other for a hard drive (for backup purposes at
> home). Is this a bad idea?

I think so, but I've never tried it.

It _should_ appear as a SCSI cdrom..

------------------------------

Date: Sat, 22 Feb 2003 21:57:06 -0500 (EST)
From: Charles Sprickman <sp...@inch.com>
Subject: Re: Maxtor Firewire Drives

On Sat, 23 Feb 2003, Daniel O'Connor wrote:

> My card has an Agere chip on it, the one in my laptop is Ti..
> Agere -> fwohci0: <Lucent FW322/323> mem 0xdc042000-0xdc042fff irq 10 at device 8.0 on pci0
> Ti -> fwohci0: <1394 Open Host Controller Interface> mem 0xf6ff8000-0xf6ffbfff,0xf6ffd800-0xf6ffdfff irq 11 at device 15.2 on pci2

Thanks, I'll see if I can track down the lucent.

On a tangent, this is a very neat little box with Firewire built-in:

http://www.checkercube.com/store/ViaMBComp/ViaMBComp.html#EPIA-M

But it looks like the FW chipset is not supported (it's the Via 6307S).
5.0 claims support for the 6306, I wonder how different they are?

> > I'm planning on using one enclosure to share my CD-R drive amongst
> > machines, and using the other for a hard drive (for backup purposes at
> > home). Is this a bad idea?
>
> I think so, but I've never tried it.
>
> It _should_ appear as a SCSI cdrom..
>
> --
> Daniel O'Connor software and network engineer
> for Genesis Software - http://www.gsoft.com.au
> "The nice thing about standards is that there
> are so many of them to choose from."
> -- Andrew Tanenbaum
> GPG Fingerprint - 9A8C 569F 685A D928 5140 AE4B 319B 41F4 5D17 FDD5
>
>
> To Unsubscribe: send mail to majo...@FreeBSD.org
> with "unsubscribe freebsd-stable" in the body of the message
>

------------------------------

Date: Sat, 22 Feb 2003 22:44:28 -0500
From: Mike Tancsa <mi...@sentex.net>
Subject: Re: Maxtor Firewire Drives

At 03:48 PM 22/02/2003 -0500, Charles Sprickman wrote:
>On Fri, 22 Feb 2003, Daniel O'Connor wrote:
>
> > On Sat, 2003-02-22 at 07:03, Aaron Wohl wrote:
> > > The maxtor 5000xx series does not work on 4.7 or 5.0 for all variations
> > > of stable/currrent/standard. This is for USB1 USB2 and firewire.
>
> > Hmm.. can you explain more about the firewire bit?
> > The enclosure I have gets this ->
> > firewire0: New S400 device ID:0030e001e0017089
> > firewire0: Device SBP-I
>
>I've asked before, but maybe things have changed since then... What cards
>are people who are doing devel work using? Since so many are "no name" or
>not identified as a brand on many online stores, can someone post store
>and model info?

I am trying out

fwohci0: <VIA VT6306> port 0xa800-0xa87f mem 0xfa000000-0xfa0007ff irq 11
at device 13.0 on pci0
fwohci0: PCI bus latency was changing to 250.
fwohci0: OHCI version 1.0 (ROM=1)
fwohci0: No. of Isochronous channel is 8.
fwohci0: EUI64 00:11:06:00:00:99:9a:d6
fwohci0: Phy 1394a available S400, 3 ports.
fwohci0: Link S400, max_rec 2048 bytes.
firewire0: <IEEE1394(FireWire) bus> on fwohci0
sbp0: <SBP2/SCSI over firewire> on firewire0
fwohci0: Initiate bus reset
fwohci0: BUS reset
fwohci0: node_id = 0xc800ffc0, CYCLEMASTER mode

I added an IDE drive to an external enclosure and it came up as

Feb 18 09:20:53 amd /kernel: fw_attach_dev: 2 pending handlers called
Feb 18 09:20:53 amd /kernel: retry_count = 1
Feb 18 09:20:53 amd /kernel: da0 at sbp0 bus 0 target 0 lun 0
Feb 18 09:20:53 amd /kernel: da0: <Oxford S OXFORD IDE Devic 0035> Fixed
Simplified Direct Access SCSI-4 device
Feb 18 09:20:53 amd /kernel: da0: 50.000MB/s transfers
Feb 18 09:20:53 amd /kernel: da0: 28629MB (58633344 512 byte sectors: 255H
63S/T 3649C)
Feb 18 09:20:55 amd /kernel: fwohci0: Initiate bus reset
Feb 18 09:20:55 amd /kernel: fwohci0: BUS reset
Feb 18 09:20:55 amd /kernel: fwohci0: node_id = 0xc800ffc1, CYCLEMASTER mode
Feb 18 09:20:55 amd /kernel: firewire0: 2 nodes, maxhop <= 1, cable IRM = 1
(me)

The speed of the drive was pretty close to that of it being directly on the
IDE cable. I havent done any extensive testing yet but so far so good!

---Mike

------------------------------

Date: Sat, 22 Feb 2003 19:52:06 -0800 (PST)
From: Tenebrae <tenebr...@niceboots.com>
Subject: Re: Maxtor Firewire Drives

On Sat, 22 Feb 2003, Charles Sprickman wrote:

> I've asked before, but maybe things have changed since then... What cards
> are people who are doing devel work using? Since so many are "no name" or
> not identified as a brand on many online stores, can someone post store
> and model info?

I was thinking about getting a Coolmax 3.5" external enclosure that could
connect via USB 2.0 or Firewire and putting a Maxtor DiamondMax 80GB IDE
drive in it...The vendor I usually use sells a "PCI to FireWire IEEE 1394
3+1 Ports Host Adapter" that they use different brands interchangeably on
as well...
I have never used Firewire anything before.

> I'm planning on using one enclosure to share my CD-R drive amongst
> machines, and using the other for a hard drive (for backup purposes at
> home). Is this a bad idea?

Wish I knew.
I'm looking at using aforementioned hardware to back up my server. There
aren't any more available IDE channels or even places inside the case to
put another drive, so external is what I'm looking at.
My favourite vendor also has an external PARALLEL drive enclosure. With
the potential data corruption mentioned with Firewire, would I be better
off with getting the parallel enclosure?
I have too much data to back up to tape.
-Tenebrae.
- ---
The sending of any unsolicited email advertising messages to this domain
may result in the imposition of civil liability against you in accordance
with Cal. Bus. & Prof. Code Section 17538.45.

------------------------------

Date: 23 Feb 2003 15:56:31 +1030
From: "Daniel O'Connor" <doco...@gsoft.com.au>
Subject: Re: Maxtor Firewire Drives

On Sun, 2003-02-23 at 13:27, Charles Sprickman wrote:
> On a tangent, this is a very neat little box with Firewire built-in:
>
> http://www.checkercube.com/store/ViaMBComp/ViaMBComp.html#EPIA-M
>
> But it looks like the FW chipset is not supported (it's the Via 6307S).
> 5.0 claims support for the 6306, I wonder how different they are?

Probably not much :)

I think most Firewire controllers are OHCI compliant, so they should
work with the existing driver.

------------------------------

Date: Sat, 22 Feb 2003 23:03:29 -0800
From: David Schultz <d...@FreeBSD.ORG>
Subject: Re: Raising KVA_PAGES breaks linux emulation?

Thus spake Marc G. Fournier <scr...@hub.org>:
> On Sat, 22 Feb 2003, Luoqi Chen wrote:
> > Module builds seem to use CFLAGS (while kernel builds use COPTFLAGS).
>
> Right, and, as was found out, this doesn't just affect kernel+modules, but
> libc_r as well ... anything else it might affect? :(
>
> > A better way is to have kernel Makefile passing opt_global.h to the
> > modules. The best solution for this particular problem would probably
> > be changing USRSTACK to a global variable instead of a #define.
>
> I realize we're in a freeze for 4.8-RELEASE, but sicne KVA_PAGES is
> documented, and its use does break things, any chance on getting this
> fixed before the RELEASE?

For libc_r, we can do better than documenting the problem: we can
fix it! All we have to do is ask the kernel for the value of
KVA_PAGES via sysctl instead of relying on the static value. Most
of the work is already done, so the fix is a 2-line patch. I'm
going to run it by the libc_r maintainer ASAP if he hasn't dealt
with it already, but I don't know about getting the fix into 4.8-RELEASE.

I don't know what modules in general or the linuxolator
specifically need KVA_PAGES for, but it may be possible to have
them extract the information from a global symbol in the kernel
proper, instead of using a compile-time constant.

------------------------------

Date: Sun, 23 Feb 2003 11:17:09 +0100
From: Wilko Bulte <w...@freebie.xs4all.nl>
Subject: Re: Compaq WL200 wireless card trouble

On Sat, Feb 22, 2003 at 02:48:36PM -0700, M. Warner Losh wrote:
> In message: <2003022121...@freebie.xs4all.nl>
> Wilko Bulte <w...@freebie.xs4all.nl> writes:
> : pcic0: <Cirrus Logic PD6729/6730 PCI-PCMCIA Bridge> port 0xcc00-0xcc03 irq
>
> This doesn't work before 4.7-stable (and maybe I haven't MFC'd all the
> changes form current to make it work).

Ah, so I understand it is the PCI-PCMCIA bridge that is most likely
the source of the watchdog timeouts? The BP6 machine is at 4.7-stable
as of 2 days back or so.

I tried 5.0 on another machine (!!) and there the whole machine locks up
as soon as it sees the wi. I.e. when it displays the wi0 probe message.
To be fair, 4.7R does the same there.

<puzzled>
Wilko

- --
| / o / /_ _ wi...@FreeBSD.org
|/|/ / / /( (_) Bulte

------------------------------

Date: Sun, 23 Feb 2003 12:43:57 +0100 (CET)
From: Dirk-Willem van Gulik <di...@webweaving.org>
Subject: Re: Compaq WL200 wireless card trouble

On Sun, 23 Feb 2003, Wilko Bulte wrote:

> Ah, so I understand it is the PCI-PCMCIA bridge that is most likely
> the source of the watchdog timeouts? The BP6 machine is at 4.7-stable
> as of 2 days back or so.
..
> I tried 5.0 on another machine (!!) and there the whole machine locks up
> as soon as it sees the wi. I.e. when it displays the wi0 probe message.
> To be fair, 4.7R does the same there.

The commit you need is:

http://www.freebsd.org/cgi/cvsweb.cgi/src/sys/pccard/pcic_pci.c.diff?r1=1.122&r2=1.123

the two lines after ' Tell the chip to do its routing thing'.

static void
-pcic_pci_pd67xx_init(device_t dev)
+pcic_pci_pd6729_init(device_t dev)
{
struct pcic_softc *sc = device_get_softc(dev);

- if (sc->csc_route == pcic_iw_pci || sc->func_route == pcic_iw_pci)
- device_printf(dev, "PD67xx maybe broken for PCI routing.\n");
+ /*
+ * Tell the chip to do its routing thing.
+ */
+ pcic_pci_pd6729_func(&sc->slots[0], sc->func_route);
+ pcic_pci_pd6729_csc(&sc->slots[0], sc->csc_route);
}

Dw.

------------------------------

End of stable-digest V5 #801
****************************

To Unsubscribe: send mail to majo...@FreeBSD.org
with unsubscribe freebsd-stable-digest in the body of the message

0 new messages