[Sidefx-houdini-list] houdini crashes...

[Mark Alexander]

After collecting system specs from a number of users that are 
experiencing lockups, it appears that the common theme is an older AMD 
motherboard chipset with AGP interfaces. I have a system based on the 
AMD 760MP chipset (approximately 2-3 years old) which exhibits the 
problem. Another developer here has a newer dual Opteron system which 
does not.

I've installed the 8178 drivers, and so far have yet to see a crash 
(however, I may not have tested long enough). I would do this as a first 
step - I have seen no problems with it so far (as long as you are on 
Linux -- 8178 has big CPU usage problems on Windows).

I also found some interesting notes in the Nvidia linux driver readme 
(that you can download in full from 
http://download.nvidia.com/XFree86_40/1.0-2880/README.txt).

This readme contains notes on solving stability problems. I have listed 
in order of easy to drastic changes (from that readme). I would 
recommend first installing the 8178 drivers, and then doing these fixes 
one by one until the problem stops (testing after each):

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

AGP Fast Writes

     This AGP feature may cause severe stability problems. It can be
     en/disabled in many system BIOSes.  If your BIOS does not offer
     this option, you can force support for AGP Fast Writes off with the
     NVreg_EnableAGPFW NVdriver module parameter.

     If you are inserting the module manually:

       insmod NVdriver NVreg_EnableAGPFW=0

     If you are using modprobe (/etc/modules.conf):

       alias char-major-195 NVdriver
       options NVdriver NVreg_EnableAGPFW=0

[MA: This only affects performance; you can't hose your system by 
changing this]

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

  AGP Rate

     You may want to decrease the AGP rate setting if you are seeing
     lockups with the value you are currently using. You can do so
     with the NVreg_ReqAGPRate NVdriver module parameter.

     If you are inserting the module manually:

       insmod NVdriver NVreg_ReqAGPRate=2   # force AGP Rate to 2x
       insmod NVdriver NVreg_ReqAGPRate=1   # force AGP Rate to 1x

     If you are using modprobe (/etc/modules.conf):

       alias char-major-195 NVdriver
       options NVdriver NVreg_ReqAGPRate=2  # force AGP Rate to 2x
       options NVdriver NVreg_ReqAGPRate=1  # force AGP Rate to 1x

       On Athlon motherboards with the VIA KX133 or 694X chip set, such 
as the ASUS K7V motherboard, NVIDIA drivers default to AGP 2x mode to 
work around insufficient drive strength on one of the signals.  You can 
force AGP 4x by setting NVreg_EnableVia4x to 1.  Note that this may 
cause the system to become unstable.

       On ALi1541 and ALi1647 chipsets, NVIDIA drivers disable AGP to 
work around timing issues and signal integrity issues. You can force AGP 
to be enabled on these chipsets by setting NVreg_EnableALiAGP to 1. Note 
that this may cause the system to become unstable.

[MA: This only affects performance; you can't make your system more 
unstable by changing this unless you increase the rate]

-----------

o AGP drive strength BIOS setting (Via based mainboards)

     Many Via based mainboards allow adjusting the AGP drive strength in
     the system BIOS. The setting of this option largely affects system
     stability, the range between 0xEA and 0xEE seems to work best for
     NVIDIA hardware. Setting either nibble to 0xF generally results in
     severe stability problems.

     If you decide to experiment with this, you need to be aware of
     the fact that you are doing so at your own risk and that you may
     render your system unbootable with improper settings until you
     reset the setting to a working value (w/ a PCI graphics card or
     by resetting the BIOS to its default values).

[MA: Try 0xEA, 0xEB, 0xEC, 0xED, or 0xEE. I have no idea what they mean, 
other than 0xEA is likely the weakest drive strength of the bunch, so 
you might want to start there]

-----------

  o System BIOS version

     Make sure to have the latest system BIOS provided by the board
     manufacturer.

[MA: This involves flashing the bios ROM on the motherboard. Do NOT do 
this unless you know what you're doing - you can render your motherboard 
completely inoperative]

-----------

o Support for the processor's Page Size Extension on Athlon Processors

     Similar to systems using Windows 2000 based operating systems, your
     Linux system may stop responding if you use applications that stress
     AGP (such as ViewPerf). This can often be solved by passing the
     "mem=nopentium" option to the Linux kernel, which disables support
     for the processor's Page Size Extension.  This may impact
     performance with some applications.  For further details, see
     Microsoft Knowledge Base Article Q270715.

[MA: MPlay can definitely stress AGP during playback, and so it sounds 
like a plausible cause. Unfortunately, this is a VERY drastic measure, 
as I suspect mantra and Houdini would happen to fit in the category of 
"some applications". The article is geared at MS Windows 2000 and 
doesn't really apply here, but if you're interested: 
http://support.microsoft.com/kb/q270715/ ]

----------


Hopefully this can solve the MPlay X-crashes people have been seeing. 
I'm sorry it's not a simple MPlay patch, but I will continue to keep an 
eye out for MPlay problems. Thanks for your patience, and good luck...

M.