https://bugzilla.novell.com/show_bug.cgi?id=410738
User clarkt@cnsp.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=410738#c5
Clark Tompsett changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEEDINFO |NEW
Info Provider|clarkt@cnsp.com |
--- Comment #5 from Clark Tompsett 2008-08-06 20:28:34 MDT ---
The 2.6.22 10.3 kernel works. The 2.6.25 from 11.0 and the 2.6.26 from 11.1A1
will not boot with out noacpi or hpet=disable. Without one of those two
options, the 25 and 26 kernels will lock on boot and you cannot proceed. Using
the noacpi results in only 1 of the 2 cpu cores being used. The 2nd is not
even shown.
Testing 11.1A1 with the 2.6.27-rc1 kernel works if you use the iommu=amd option
and the hpet timer is used.
I've been looking at the bugzilla reports for both kernel and Novell. There is
a thread in the kernel mailing list dealing with the new HP bio's and the fact
that they are not following how acpi tables are to be written and that HP (and
others) appears to be only working with 'bug' in each version of Windows and
setting bad or wrong values for linux. The kernel folks appear to be trying to
get the vendors and bios writers to tell them what changes are being done so
that the kernel can work with these changes and kluges.
I have tried building a custom dsdt for testing that removed the hpet segment
and that has not worked.
I have also posted questions on AMD forums and the only response from them was
to create a custom dsdt.
--
Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.