Problems with a Acer Aspire

[Jung-uk Kim]

On Friday 11 February 2005 10:35 am, Astrodog wrote:
> On Fri, 11 Feb 2005 15:43:02 +0100, Kent Palmkvist 
<kentp at isy.liu.se> wrote:
> > Hello all,
> >
> > Here is a fix how to get the Acer Aspire 1522 WLMi to boot amd64
> > FreeBSD (tested with 5.3 and 6-current-snap from February 2005).
> >
> > Problem:
> > The boot process hangs after loading the kernel, but before
> > getting any output from the kernel boot.
> > In 5.3 this means that I get the daemon figure with the 7
> > choices, and independent of which of the
> > choices I make, the loader indicator will make a couple of turns
> > and then stop. This has been tested
> > with all the different versions of the amd64 FreeBSD that I have
> > been able to find (5.3-RELEASE,
> > 5.3-SNAP, 6-Current-SNAP etc.) dating from December 2004 to
> > February 2005. Also tested the one
> > by Kelly Black that solved a problem with the Compaq Presario
> > R3000z (http://blackk.union.edu/~black/freebsd/) but with same
> > disapointing results.
> >
> > Solution:
> > Disable the Legacy USB Support in the bios setup (at least
> > available in v1.09 of the BIOS). It boots
> > correctly into the installer for 6-current-snap from February
> > 2005 (do not need to select any special
> > boot options). For 5.3-RELEASE, it needs the ACPI to be disabled,
> > otherwise it will get a panic on
> > "vm_fault: fault on nofault entry" after cardbus detection. If
> > ACPI is disabled, 5.3-RELEASE will boot
> > nicely into the installer.
> >
> > good luck!
> >
> > /Kent
>
> Tried the R3000Z line of patches? Might relate.

If I am not mistaken completely, his problem is not related to my 
patches.  Acer Aspire 1522 WLMi has VIA chipsets and they have 
different issues.  In fact, I have Acer Ferrari 3400 and I had 
experienced similar problems before.  I am using 6-CURRENT on it now.

BTW, 'turning off USB legacy support fixed problem' sounds like he has 
C0 revision core (i. e., errata #93) without BIOS workaround.  You 
can see the CPU ID from dmesg:

CPU: AMD Athlon(tm) 64 Processor 3000+ (2000.08-MHz K8-class CPU)
  Origin = "AuthenticAMD"  Id = 0xf48  Stepping = 8

This is from my desktop.  If the Id is 0xf48, you have C0 revision 
core.  FYI, Ferrari 3400 has Oakville core!

CPU: Mobile AMD Athlon 64 Processor 3000+ (2004.56-MHz K8-class CPU)
  Origin = "AuthenticAMD"  Id = 0x10fc0  Stepping = 0

Jung-uk Kim