HP DL385 "Gen 1" ACPI timing glitch in 10.0 BETA3 (and 9.2).

[bunny at tx.net]

On an HP DL385 first-generation system, 10.0 BETA3 and 9.2-RELEASE most of 
the time fail to boot, freezing the system after the following lines, 
requiring a cold boot:

  Timecounter "ACPI-fast" frequency 3579545 Hz quality 900
  acpi_timer0: <32-bit timer at 3.579545MHz> port 0x908-0x90b on acpi0
  pcib0: <ACPI Host-PCI bridge> on acpi0
  pci0: <ACPI PCI bus> on pcib0

The next lines would have been, if reached:

  pcib1: <ACPI PCI-PCI bridge> at device 3.0 on pci0
  pci1: <ACPI PCI bus> on pcib1


This behavior was never seen before 9.2.  It happens with the 10.0BETA3 
install disc, also, though not with the 9.2-R install disc.  No similar 
problems have been noticed on these systems with any level of 9.1-R or 
before.

When the Verbose boot option is "on", the system always completes 
the boot sequence in 9.2 and 10.0.

These items stand out from the verbose dmesg.boot from 10.0B3 disc1:

   WARNING: This architecture revision has known SMP hardware bugs which
            may cause random instability

   ACPI BIOS Warning (bug): Invalid length for FADT/Pm1aControlBlock: 32,
   using default 16 (20130823/tbfadt-682)

This similar message appears in an always-working 9-1-R boot of the same 
system:

   ACPI Warning: Invalid length for Pm1aControlBlock: 32,
   using default 16 (20110527/tbfadt-638)

We have quite a number of these DL385 gen1 systems, and they have, in 
practice, been very stable.  Are they going to be unstable under 10.0?  Is 
there a workaround?

Let me know what other information to supply.

stu