headers
Johnny C. Lam | 1 Jul 2005 01:45
Picon

Please test fix for dlopen()

Could someone please test the patch in the following PR?

     http://www.netbsd.org/cgi-bin/query-pr-single.pl?number=30570

It is supposed to fix the dlopen() call on NetBSD/amd64, and it would be 
nice to have that fixed for NetBSD-3.0.  I had meant to test it but I no 
longer have easy access to that machine.  There is a test case provided 
in that PR as well.

	Thanks,

	-- Johnny Lam <jlam <at> NetBSD.org>
Permalink | Reply | 0 comments |
headers
Stephen Degler | 1 Jul 2005 02:53

Re: Stable Opteron configurations with 16GB or more

Jens Andre Griepentrog wrote:

>Dear NetBSD AMD64 Community,
>
>I would like to get an Opteron compute server with a large amount
>of memory appropriate for three-dimensional numerical simulations.
>
>I have got some offer for a system with a Tyan S2882-D or S2892 board
>with two Opterons 248HE, U320 SCSI disks, and 16GB memory. But after
>tweaking out more information the same dealer concedes severe BIOS
>problems with the 90nm Opteron version connected to more than 2GB
>of memory! So one can feel a little bit irritated having in mind
>that all 248HE Opterons are manufactured in 90nm process ... 
>
>It would be nice to hear from experienced BSD users which combinations
>of Opteron mainboards and processors work stable with 16GB or more of
>installed memory.
>
>Best regards,
>Jens A. Griepentrog
>
>PS: I would like to thank the whole BSD Community for their efforts
>maintaining that nice operating system which I have been using now
>for over ten years.
>
>  
>
As a datapoint I work at a firm that uses numerous Tyan 2 and 4 way 
boards with up to 16Gb Memory.  Generally our 2 way systems have 8Gb and 
the 4 way have 16 (1Gb Dimms).  We recently have ordered a 2 way system
(Continue reading)

Permalink | Reply | 0 comments |
headers
Richard Rauch | 1 Jul 2005 10:17

Confirm problem?

I"m running NetBSD/amd64 3.99.3 here.  I took an i386 hard drive and
did a complete, fresh install of -current a little while back, and have
not updated it since then.

In /usr/pkgsrc/lang/erlang, I was trying to build it.  I was told by
others that it does not really require Java, so I tried commenting
out the java-vm requirement in the Makefile.

During configure it hangs up on this line:

checking for unreliable floating point execptions...

...I let it run for over 8 minutes and it never finished this test.
I looked at the code it was running for it, and I couldn't see any
loops in the .c file.  It was chewing up all of the CPU.  (top(1)
actually said 101% to 102% of the CPU, but we'll leave that alone
for now.  (^&)

Can someone confirm?  Or would this likely go away if I updated
my kernel and/or the compiler & libraries?

I'm a little fatigued at this point, so I'm not going to fiddle
more with this for now to try to isolate just what's walking off
into limbo, and am just wondering if anyone else can reproduce it.

--

-- 
  "I probably don't know what I'm talking about."  http://www.olib.org/~rkr/
Permalink | Reply | 0 comments |
headers
Rhialto | 1 Jul 2005 17:51
Picon

Re: building from pkgsrc on NetBSD/AMD64

On Thu 30 Jun 2005 at 21:15:06 +0200, Robert Cates wrote:
> either an nVidia or ATI graphics card.  How about TV card compatability?

I would stay away from proprietary nVidia stuff. I have an ATI Radeon
graphics card (whatever cheapest was available, "ATI Technologies Inc
Radeon RV100 QY [Radeon 7000/VE] rev 0"). I also have a Hauppauge tv
card ("bktr0: Hauppauge Model 56304 D" "bktr0: Hauppauge WinCast/TV,
Temic PAL tuner"), and unfortunately I haven't got xawtv to work, even
after fixing its 64-bit-uncleanness. Everything appears to work ok (as
far as I can tell without having seen it work before) but I just get a
black window instead of tv (or even instead of snow). fxtv dumps core;
probably also sloppy coding.

> Robert
-Olaf.
--

-- 
___ Olaf 'Rhialto' Seibert                            --  rhialto/at/falu.nl
\X/ Hi! I'm a signature virus! Copy me to your .signature to help me spread!
>
Permalink | Reply | 0 comments |
headers
Karthik Subramanian | 4 Jul 2005 06:53
Picon

NetBSD install fails on ASUS K8S-MX motherboard/AMD Athlon 2800+

Hi Folks,

I'm a *total* NetBSD newbie, have some experience running linux.

I'm trying to install NetBSD 2.0.2 (AMD64) / NetBSD 2.0 (i386) on my
box which has an ASUS K8S-MX motherboard
(http://www.asus.com/products4.aspx?l1=3&l2=14&l3=0&model=377&modelmenu=1).

I seem to have exactly the same problem as msmith <at> edtech.mcc.edu, as
indicated in his email to port-amd64 on 02/10/2005, I get the
following messages when the NetBSD-2.0.2 (AMD64) install CD boots:

>siside0:0:0: lost interrupt
>type: ata tc_bcount: 512 tc_skip: 0
>siside0:0:0: lost interrupt
>type: ata tc_bcount: 512 tc_skip: 0

Trouble is, it hangs right there and proceeds no further for me.
What's more, I tried installing NetBSD-2.0 (the i386 port) to try and
see if it worked for me like it did for msmith - to no avail, that
install cd hung on the boot as well.

I am at my wits end, I'm not able to run any OS on my box except MS-crap.
Could Someone please help! I would like to get NetBSD up and running on my box.

Thanks in Advance,
Karthik.
Permalink | Reply | 0 comments |
headers
tlaronde | 4 Jul 2005 12:51

Re: NetBSD install fails on ASUS K8S-MX motherboard/AMD Athlon 2800+

Hello,

On Mon, Jul 04, 2005 at 10:23:45AM +0530, Karthik Subramanian wrote:
> Hi Folks,
> 
> I'm a *total* NetBSD newbie, have some experience running linux.
> 
> I'm trying to install NetBSD 2.0.2 (AMD64) / NetBSD 2.0 (i386) on my
> box which has an ASUS K8S-MX motherboard
> (http://www.asus.com/products4.aspx?l1=3&l2=14&l3=0&model=377&modelmenu=1).
> 
> I seem to have exactly the same problem as msmith <at> edtech.mcc.edu, as
> indicated in his email to port-amd64 on 02/10/2005, I get the
> following messages when the NetBSD-2.0.2 (AMD64) install CD boots:
> 
> >siside0:0:0: lost interrupt
> >type: ata tc_bcount: 512 tc_skip: 0
> >siside0:0:0: lost interrupt
> >type: ata tc_bcount: 512 tc_skip: 0
> 

It is just for the record and to give some supplementary tracks for
developers.
This same kind of error happens on an ASUS notebook (L3) when ACPI is
enable when accessing a slow device (which is the case with a CDROM).
Please note that this is _i386_ so it is not related to AMD64 in my
case. It might be in this case too that the problem is with some bios
stuff and not the arch per se.

Since (needs to be verified) an install kernel is unlikely to have ACPI
(Continue reading)

Permalink | Reply | 0 comments |
headers
Manuel Bouyer | 4 Jul 2005 20:21

Re: NetBSD install fails on ASUS K8S-MX motherboard/AMD Athlon 2800+

On Mon, Jul 04, 2005 at 10:23:45AM +0530, Karthik Subramanian wrote:
> Hi Folks,
> 
> I'm a *total* NetBSD newbie, have some experience running linux.
> 
> I'm trying to install NetBSD 2.0.2 (AMD64) / NetBSD 2.0 (i386) on my
> box which has an ASUS K8S-MX motherboard
> (http://www.asus.com/products4.aspx?l1=3&l2=14&l3=0&model=377&modelmenu=1).
> 
> I seem to have exactly the same problem as msmith <at> edtech.mcc.edu, as
> indicated in his email to port-amd64 on 02/10/2005, I get the
> following messages when the NetBSD-2.0.2 (AMD64) install CD boots:
> 
> >siside0:0:0: lost interrupt
> >type: ata tc_bcount: 512 tc_skip: 0
> >siside0:0:0: lost interrupt
> >type: ata tc_bcount: 512 tc_skip: 0
> 
> Trouble is, it hangs right there and proceeds no further for me.
> What's more, I tried installing NetBSD-2.0 (the i386 port) to try and
> see if it worked for me like it did for msmith - to no avail, that
> install cd hung on the boot as well.
> 
> I am at my wits end, I'm not able to run any OS on my box except MS-crap.
> Could Someone please help! I would like to get NetBSD up and running on my box.

Ca you try a 3.0_BETA kernel ? You have a snapshot of 3.0_BETA at
ftp://ftp.netbsd.org/pub/NetBSD-daily/netbsd-3/200506230000Z/

This may be something that was fixed last month.
(Continue reading)

Permalink | Reply | 0 comments |
headers
Robert Cates | 4 Jul 2005 20:39

Re: NetBSD install fails on ASUS K8S-MX motherboard/AMD Athlon 2800+

Hi,

I'm curious, did you change any settings in the BIOS, at any time, like to 
overclock the CPU?  If so, you might want to set the BIOS back to the 
default settings, and try again.

-Robert

----- Original Message ----- 
From: "Manuel Bouyer" <bouyer <at> antioche.eu.org>
To: "Karthik Subramanian" <karthik301176 <at> gmail.com>
Cc: <port-amd64 <at> NetBSD.org>; <netbsd-users <at> NetBSD.org>
Sent: Monday, July 04, 2005 8:21 PM
Subject: Re: NetBSD install fails on ASUS K8S-MX motherboard/AMD Athlon 
2800+

> On Mon, Jul 04, 2005 at 10:23:45AM +0530, Karthik Subramanian wrote:
>> Hi Folks,
>>
>> I'm a *total* NetBSD newbie, have some experience running linux.
>>
>> I'm trying to install NetBSD 2.0.2 (AMD64) / NetBSD 2.0 (i386) on my
>> box which has an ASUS K8S-MX motherboard
>> (http://www.asus.com/products4.aspx?l1=3&l2=14&l3=0&model=377&modelmenu=1).
>>
>> I seem to have exactly the same problem as msmith <at> edtech.mcc.edu, as
>> indicated in his email to port-amd64 on 02/10/2005, I get the
>> following messages when the NetBSD-2.0.2 (AMD64) install CD boots:
>>
>> >siside0:0:0: lost interrupt
(Continue reading)

Permalink | Reply | 0 comments |
headers
Chris | 4 Jul 2005 22:29
Picon

amd64: Installation crashes when using SATA disk

Hi all,

I tried to install the amd64 port and it instantly reboots when it tries
to access my SATA disk. It boots without problems when I unplug the SATA
disk. I used boot images for netbsd 2.0, 3.0 and current, all showed the
same behaviour. There are no problems with the i386 port, however.

Some hardware information:
Athlon64, Asus A8N-E (Nforce4 Ultra), Western Digital 200GB Sata drive
WD2000J(?)

Any hints, except using an IDE drive?

Cheers,
Chris
Permalink | Reply | 1 comment |
headers
Christopher SEKIYA | 5 Jul 2005 02:31

Re: NetBSD install fails on ASUS K8S-MX motherboard/AMD Athlon 2800+

On Mon, Jul 04, 2005 at 10:23:45AM +0530, Karthik Subramanian wrote:

> >siside0:0:0: lost interrupt
> >type: ata tc_bcount: 512 tc_skip: 0
> >siside0:0:0: lost interrupt
> >type: ata tc_bcount: 512 tc_skip: 0

ioapic interrupts are not being routed correctly.

You need to use an installation kernel that was compiled one of the following
ways:

* with ioapic support commented out (suboptimal), or
* with ACPI and PCI_INTR_FIXUP options defined.

Frank, now that the cdboot code is in-tree, would it make sense to turn on
the ACPI fixups in INSTALL and GENERIC?
--

-- 

-- Chris
	GPG key FEB9DE7F (91AF 4534 4529 4BCC 31A5  938E 023E EEFB FEB9 DE7F)
Permalink | Reply | 0 comments |

Gmane