Hey people,

I have exactly the same problem. My system is an AMD Athlon 1 GHz, with 512 Mb ram and the same amount of swap space.
Could it be an AMD-only problem?
I've tried disabling UDMA, but it still happens. (haven't tried disabling DMA yet)
The problem can occur very early, but sometimes the system will run for hours and then crash on some small disk-access.
In NetBSD 1.5 this never happened, so I guess the problem has been introduced along the way to 1.6. (For the record, I'm using 1.6 first release)

I've disabled any drivers I'm not using, and the ones I _am_ using are the same as in NetBSD 1.5, with the exception of the NTFS drivers (I moved from Win'95 to Win2000).
I mostly get pagefaults, but sometimes I get some really strange error saying that the kernel is 'locking against myself'.
Also, sometimes a file is 'wrong'. Then I build and the compiler complains about some error in a system header. Upon investigation, nothing is wrong. When compiling again, it simply gets past that stage.
So the filesystem is temporarily corrupt, so to say (or the compiler is buggy).
These errors happen when compiling pkgsrc or the kernel itself.
A kernel trace in the debugger shows different things everytime. (ie everytime different things cause it)

If it matters: I'm not using any funky things, not even X, when this happens.
Just a default install, with no extremely special daemons. (for a ps list, just ask me)

I'll try getting current one of the next days, and I'll report if it's still happening.
But it is really frustrating, since 1.6 is supposed to be stable.

If you need more information (like my kernel configuration file, or dmesg output), please ask.
Also, please do a CC to me when replying, since I'm not a member of tech-kern.

Greetings,
Peter Bex

>> But first running it under 1.5.3.  I finally determined that a
>> software system that I wrote seemed to be related to the problem.
>> When that system wasn't running, I could go for weeks w/o any
>> crashes.  So, that system is currently off.

> Oh. That's interesting. If it's all userland, it shouldn't be able to
> crash the system. oops! If you get a panic and you have ddb in the kernel,
> try "t" to get a stack trace.

>> As you can see from the above, I'm trying to migrate to 1.6 in the
>> hopes that some of the uvm problems had been found/fixed.  As it
>> turns out, 1.6 is even less stable than 1.5.3 was!  I can't even
>> compile a new kernel without having the compiler fault and/or having
>> the system hang or get a uvm_fault.  I can push through the errors,
>> and ultimately get a bootable kernel, but I've tried 60 times to get
>> mozilla to compile, and can try to push past the errors about five
>> times before hanging/getting a uvm_pagefault.  I have two or three
>> system dumps of the problem, if that would be a help.
>>
>> I'm more than happy to help whoever try and track down/debug the
>> problem.  I just need the instructions on what information I need to
>> provide.

> That's very strange. I have a SIS 735-based Athlon system, and I'm not
> seeing those problems. Does the box also run Windows? If so, try it and
> see if it too has problems. With 512 MB of RAM, you should be fine for
> compiling most things, even Mozilla I think.

> Do you have any swap configured?

> Take care,

> Bill

"The process of preparing programs for a digital computer
is especially attractive, not only because it can be economically
and scientifically rewarding, but also because it can be an aesthetic
experience much like composing poetry or music."
        --Donald Knuth