headers
Lord Isildur | 2 Jan 2004 19:58
Picon

issues with >1g memory on miata?


So, i have had a weird phenomenon. On a first-gen (i.e. no USB, etc)
miata, 433au model, in both 1.4.x and 1.5.x, i have noticed that if there
is more than a gig of core in the system, it is rather flaky. It can use
all the memory, no problem, but at least in this system, the problem shows
up in the de driver, which will reliably panic the machine if one brings
the interface up and down twice, even over wuite a span of time, and will 
lose use of the interface after a while (roughly 30-45 minutes, it seems
that heavier use breaks it sooner) without a panic, but an attempt to
reset the interface from e.g. the console will panic it immediately.
Removing the memory above 1g stops the problem. It is not a matter of bad
memory, I've tried 4 different sets of dimms in many different
combinations of slots, and that doesn't change the 1g memory size being
the cause of the problem. 

When the interface just locks up, the kernel keeps printing this message: 

de1: unable to load tx map, error = 22

and the output from the most recent panic: 

fatal kernel trap:

    trap entry = 0x2 (memory management fault)
    a0         = 0x68
    a1         = 0x1
    a2         = 0x0
    pc         = 0xfffffc0000480b1c
    ra         = 0xfffffc0000480b50
    curproc    = 0xfffffc000b0fb408
(Continue reading)

Permalink | Reply | 0 comments |
headers
naoki | 4 Jan 2004 16:00

X doesn't work on 164LX

Hi,

I'm tring to run X server on my 164LX. (1.6ZG, around Jan 3 2004)

When X server started (xf86cfg, xinit, etc), display switched to blank
screen or blinking cursor on blue screen, and machine freeze soon.

I could get below (and some similar) kernel message,

  unexpected machine check:

      mces    = 0x1
      vector  = 0x670
      param   = 0xfffffc0000006068
      pc      = 0x160a81d88
      ra      = 0x120283dc0
      code    = 0x98
      curlwp = 0xfffffc0001347a00
          pid = 733.1, comm = XFree86

X sources are from caseydoodle.com (XFree86 4.3.0 and XFree 4.4.0 RC 2)
and from XFree86.org (latest XFree86 4.4.0 RC 2). All X servers were
same result.

My VGA card is,
  vga0 at pci0 dev 9 function 0: Matrox MGA Millennium II 2164W (rev. 0x00)
and I have this too,
  vga0 at pci0 dev 9 function 0: ATI Technologies Radeon 9200SE (rev. 0x01)

I got machine freeze on both cards.
(Continue reading)

Permalink | Reply | 0 comments |
headers
mel kravitz | 4 Jan 2004 17:48

Re : X doesn't work on my 164LX

ATI cards are NOT supported on this box , see
http://www.caseydoodle.com/cards.html
for supported cards.  In short try an Nvidia TNT2 or MX 440 of GeForce 2
card.
I have also used Matrox G400 succesfully in 4.4.0rc1 and 2 Matrox MII
cards also do not
work -known to crash in 4.2.0 and 4.3.0- have not this card with
4.4.0rc*
Best -Mel
Permalink | Reply | 2 comments |
headers
der Mouse | 5 Jan 2004 03:37
Picon

netbooting 3000/400S?

I just got a 3000/400S (from someone who rescued it from being trashed,
but was under the impression it was a VAX - perhaps because it has VMS
installed on it).

I want to netboot this, preparatory to putting it to more productive
uses than sitting around at a Username: prompt (I know no logins for
its VMS install).  The Alpha netbooting page
http://www.netbsd.org/Documentation/network/netboot/intro.alpha.html
says that any SRM machine can netboot, and goes on to describe
netbooting via bootp.  When I tell this machine "show device", the
network is called ESA0,and when I "boot esa0", tcpdump on the boot
server (using "ether host 8:0:2b:....") shows a bunch of stuff I don't
recognize, and apparently bootpd doesn't either (sent to MAC addresses
ab:0:0:1:0:0 and ab:0:0:2:0:0).

Is it possible that what I have here isn't SRM but some VMS-specific
console code, and I have to find SRM and update the firmware before I
can do anything NetBSDoid with this machine?  Or is the netboot
documentation incomplete, or what?

I coukld just pull the disk and move it to one of my existing alphas to
do the install.  But if there is a console firmware issue involved,
that may not be sufficient.

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML	       mouse <at> rodents.montreal.qc.ca
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B
(Continue reading)

Permalink | Reply | 3 comments |
headers
Ray Phillips | 5 Jan 2004 07:52
Picon
Picon

Re: netbooting 3000/400S?

I haven't tried examining the traffic with tcpdump but I netboot a 
DEC 3000 - M400 using

>>>  boot ez0

as per the user guide.  It's  show dev  also calls the 10baseT 
interface ESA0, so perhaps this will work for your 400S too.

I didn't attempt a netboot before upgrading to the latest firmware, 
which might be important.  If you want/need to upgrade, beware that 
it's vitally important to have an SROM with the correct version, as 
per

http://www.xray.mpe.mpg.de/mailing-lists/tru64-unix-managers/9507/msg00317.html

I guess you've also noticed it's necessary to have the  vm=rfc1048 
option in /etc/bootptab for these machines?

Ray
Permalink | Reply | 1 comment |
headers
der Mouse | 5 Jan 2004 07:59
Picon

Re: netbooting 3000/400S?

> I haven't tried examining the traffic with tcpdump but I netboot a 
> DEC 3000 - M400 using

> >>>  boot ez0

> as per the user guide.

e*z*?  I'll have to try that.  I found a PDF (I loathe PDFs; give me
good old plain text any day) that appeared to say ex0.  But trying to
boot from ex0 failed, and trying to boot from kw0 (which I invented on
the spot) failed in the same way, so I assumed it was just a question
of what kind of Ethernet was in the machine.  But the PDF's resolution
on the screen was quite poor, so it could well have said ez0 instead.

> It's  show dev  also calls the 10baseT interface ESA0, so perhaps
> this will work for your 400S too.

Turns out it does.  Apparently booting from esa0 uses MOP and ez0 uses
tftp/bootp.

> I didn't attempt a netboot before upgrading to the latest firmware,
> which might be important.  If you want/need to upgrade, beware that
> it's vitally important to have an SROM with the correct version,

Yes, I'd noticed that.  But if I'm raeding SHOW CONFIG output
correctly, I already have the latest version:

>>> show config

DEC 3000 - M400
(Continue reading)

Permalink | Reply | 0 comments |
headers
naoki | 5 Jan 2004 17:19

Re: X doesn't work on my 164LX

Hi,

At Sun, 04 Jan 2004 16:48:41 +0000,
mel kravitz wrote:
> ATI cards are NOT supported on this box , see
> http://www.caseydoodle.com/cards.html
> for supported cards.  In short try an Nvidia TNT2 or MX 440 of GeForce 2
> card.

Thank you!

I bought Nvidia FX5200 today. It works fine :)

I made alpha_video.c patch for latest source from XFree86.org CVS repo.
(File attached)

I have a question. My machine (164LX) has BWX extension

  cpu0 at mainbus0: ID 0 (primary), 21164A-2
  cpu0: Architecture extensions: 1<BWX>

but XFree86 log says

  (II) has_bwx = 0
  (--) Machine needs sparse mapping

Why?

This is XFree86.0.log:
(Continue reading)

Permalink | Reply | 0 comments |
headers
Eric Schnoebelen | 6 Jan 2004 01:19
Favicon

Re: netbooting 3000/400S?


der Mouse writes:
- I want to netboot this, preparatory to putting it to more productive
- uses than sitting around at a Username: prompt (I know no logins for
- its VMS install).  The Alpha netbooting page
- http://www.netbsd.org/Documentation/network/netboot/intro.alpha.html
- says that any SRM machine can netboot, and goes on to describe
- netbooting via bootp.  When I tell this machine "show device", the
- network is called ESA0,and when I "boot esa0", tcpdump on the boot
- server (using "ether host 8:0:2b:....") shows a bunch of stuff I don't
- recognize, and apparently bootpd doesn't either (sent to MAC addresses
- ab:0:0:1:0:0 and ab:0:0:2:0:0).
- 
- Is it possible that what I have here isn't SRM but some VMS-specific
- console code, and I have to find SRM and update the firmware before I
- can do anything NetBSDoid with this machine?  Or is the netboot
- documentation incomplete, or what?

	I would bet it's trying to MOP boot, rather than BOOTP
boot.  As I remember (and I'm verifying by booting another
DS10L, SRM supports both MOP and BOOTP.  Set the SRM variable
esa0_protocols BOOTP to get it to use BOOTP, or set up the MOP
server.  

	Warning, it's undocumented (other than the source) but
_don't_ add the file extension on the MOP boot request.
Additionally, the file to be booted needs to be named in all
caps, and with a .SYS extension.. (at least NetBSD's MOP daemon
enforces these restrictions.
(Continue reading)

Permalink | Reply | 2 comments |
headers
naoki | 6 Jan 2004 17:48

Re: xsrc/20560 (was Re: X doesn't work on my 164LX)

Hi,

At Mon, 05 Jan 2004 20:46:07 -0800,
David Hopper wrote:
> I think this patch is much cleaner than the (now dysfunctional) patch from
> xsrc/20560.  I'm appending it to the bug report FWIW.

More simple patch (comment out "msb_set" again) is below.

But my patch is for CVS HEAD version of X source from XFree86.org. It
may not be usable until xsrc is updated to XF86 4.4.0.

# Should send-pr to XFree86.org ?

And I found a problem on my machine.

Exec XFree86 twice gets screen corrupt.

e.g.
  startx -> exit -> startx
    or
  startx -> switch console (Ctrl+Alt+Fn) -> switch back to X

  => X screen corrupted and lost control of video card (no screen update)
     + Under some condition(?), XFree86 eats cpu power around 100% ;)

No screen update, but keyboard still works. So I can do:

  Ctrl+Alt+Fn (screen still broken) -> exec shutdown -r now
    -> machine reboot and return normal state
(Continue reading)

Permalink | Reply | 1 comment |
headers
Arto Huusko | 6 Jan 2004 17:55
Picon
Picon

OpenLDAP on -current

Hi,

does anyone know whether OpenLDAP is supposed to work on -current?
For me it coredumps when I try to search something from a
database (I tried with an empty one).

GDB isn't being to helpful, either.

  -- Arto
Permalink | Reply | 0 comments |

Gmane