Re: Testing bigmem properly on amd64?

Previous thread: Re: OpenBSD culture? by VICTOR TARABOLA CORTIANO on Wednesday, April 14, 2010 - 6:01 pm. (2 messages)

Next thread: Re: OpenBSD culture? by VICTOR TARABOLA CORTIANO on Wednesday, April 14, 2010 - 6:19 pm. (1 message)
From: Peter
Date: Wednesday, April 14, 2010 - 5:12 pm

I know bigmem is still in a state of flux and can be enabled by editing 
machdep.c and compiling a custom kernel.

I also realise that it may or may not work on OpenBSD AMD64 due to the 
absence of an iommu on non VT-d capable Intel systems, aside from the 
AGP/PCI-e GART.

What's the best way to test and report this? (GNATS etc?) Is booting up 
sufficient, or does it require extensive testing of memory transfers 
from each device to above the 4GB line?

FWIW, with the latest kernel UHCI and AHC (specifically an Adaptec 29160 
running in a 32 bit PCI slot) don't work on a 975X/Core2Quad system. 
Once that's fixed the system boots. (iic also needs to be disabled, but 
that's an issue specific to the D975XBX2 motherboard, not to bigmem).

OpenBSD 4.7-current (GENERIC.MP) #0: Thu Apr 15 01:15:12 BST 2010
     
root@Gladstone.syllopsium.com:/usr/src/sys/arch/amd64/compile/GENERIC.MP
real mem = 8585592832 (8187MB)
avail mem = 8359256064 (7972MB)
User Kernel Config
UKC> disable iic
318 iic* disabled
UKC> disable ahc
  61 ahc* disabled
UKC> disable uhci
132 uhci* disabled
133 uhci* disabled
UKC> exit
Continuing...
RTC BIOS diagnostic error 80<clock_battery>
mainbus0 at root
bios0 at mainbus0: SMBIOS rev. 2.4 @ 0xe42e0 (39 entries)
bios0: vendor Intel Corp. version "BX97520J.86A.2836.2008.0728.1946" 
date 07/28/2008
bios0: Intel Corporation D975XBX2
acpi0 at bios0: rev 0
acpi0: tables DSDT FACP APIC WDDT MCFG ASF! SSDT SSDT SSDT SSDT SSDT
acpi0: wakeup devices SLPB(S4) P32_(S4) UAR1(S4) PEX0(S4) PEX1(S4) 
PEX2(S4) PEX3(S4) PEX4(S4) PEX5(S4) UHC1(S3) UHC2(S3) UHC3(S3) UHC4(S3) 
EHCI(S3) AC9M(S4) AZAL(S4)
acpitimer0 at acpi0: 3579545 Hz, 24 bits
acpimadt0 at acpi0 addr 0xfee00000: PC-AT compat
cpu0 at mainbus0: apid 0 (boot processor)
cpu0: Intel(R) Core(TM)2 Quad CPU Q6700 @ 2.66GHz, 2667.10 MHz
cpu0: ...
From: Henning Brauer
Date: Thursday, April 15, 2010 - 3:34 am

none. bigmem is known broken, otherwise it would be enabled by
default. and tests proving the fact, well, you could test that water
is wet, too.

-- 
Henning Brauer, hb@bsws.de, henning@openbsd.org
BS Web Services, http://bsws.de
Full-Service ISP - Secure Hosting, Mail and DNS Services
Dedicated Servers, Rootservers, Application Hosting

From: Peter Kay (Syllopsium)
Date: Thursday, April 15, 2010 - 3:44 am

Fair enough. I'll wait until it's deemed testable again, then.

PK 

Previous thread: Re: OpenBSD culture? by VICTOR TARABOLA CORTIANO on Wednesday, April 14, 2010 - 6:01 pm. (2 messages)

Next thread: Re: OpenBSD culture? by VICTOR TARABOLA CORTIANO on Wednesday, April 14, 2010 - 6:19 pm. (1 message)