Hello Daniel,
Just want to make sure that we are on the same page: I'm talking about
i386. It seems from below that your concern is more about amd64, but I
didn't really try it, because my CPU isn't even a Xeon.
Wednesday, September 19, 2007, 6:00:16 PM, you wrote:
I have pretty
Boris Goldberg wrote:
Hello Daniel,
Just want to make sure that we are on the same page: I'm talking about
i386. It seems from below that your concern is more about amd64, but I
didn't really try it, because my CPU isn't even a Xeon.
You are 100% right. An oversight on my part here.
Hello Daniel,
Monday, September 17, 2007, 3:14:05 PM, you wrote:
DO Now that is working do me a favor and try to compile the userland and
DO kernel with that bsd.mp acpi enable kernel.
DO Also, try if possible to make transfer of huge files between two boxes
DO well connected to try to at a
Boris Goldberg wrote:
I have pretty much the same picture with HP ProLiant 320 G5 (Dual Core
Pentium-D 925). The server is new and passes all tests from the HP
maintenance CD.
I couldn't make what BIOS version you were actually running there, but
you did check to make sure you
On 2007/09/19 19:00, Daniel Ouellet wrote:
Le me know how it goes with current, I am curious as so far all feedback I
got is no one yet can get an AMD64.mp stable at this time
this must be hardware-dependent, my main desktop is amd64 MP
(opteron 175 i.e. dual-core) and gives no trouble on
Stuart Henderson wrote:
On 2007/09/19 19:00, Daniel Ouellet wrote:
Le me know how it goes with current, I am curious as so far all feedback I
got is no one yet can get an AMD64.mp stable at this time
this must be hardware-dependent, my main desktop is amd64 MP
(opteron 175 i.e. dual-core) and
load averages: 0.30, 0.08, 0.03
05:22:12
15 processes: 14 idle, 1 on processor
CPU0 states: 0.3% user, 0.0% nice, 0.7% system, 0.1% interrupt, 98.9%
idle
CPU1 states: 0.0% user, 0.0% nice, 0.1% system, 0.0% interrupt, 99.9%
idle
CPU2 states: 0.0% user, 0.0% nice, 0.0% system,
On 9/16/07, Daniel Ouellet [EMAIL PROTECTED] wrote:
Paul Taulborg wrote:
I appologize for not including this, here is the dmesg of a successful
boot of the amd 4.2 DEFAULT kernel:
Paul,
Not sure all the tests you did, but first do not run AMD64 on Intel
processor. I would do this first
Paul Taulborg wrote:
Booya! Updated my BIOS to the latest version (44), and applied the patch
that was kindly provided to me here:
http://marc.info/?l=openbsd-techm=118975639013313w=2
I also enabled acpi0 in the kernel by default (required to see the other
processors), and tada!
I had to
Also Paul,
Now that is working do me a favor and try to compile the userland and
kernel with that bsd.mp acpi enable kernel.
Also, try if possible to make transfer of huge files between two boxes
well connected to try to at a minimum get close to 100Mb/sec of
transfer, or more if you have
I appologize for not including this, here is the dmesg of a successful
boot of the amd 4.2 DEFAULT kernel:
OpenBSD 4.2-current (GENERIC) #1191: Thu Sep 13 14:19:37 MDT 2007
[EMAIL PROTECTED]:/usr/src/sys/arch/amd64/compile/GENERIC
real mem = 2142949376 (2043MB)
avail mem = 2069811200
Paul Taulborg wrote:
I appologize for not including this, here is the dmesg of a successful
boot of the amd 4.2 DEFAULT kernel:
Paul,
Not sure all the tests you did, but first do not run AMD64 on Intel
processor. I would do this first thing if you haven't done already.
- Go into BIOS and
Hi Daniel,
Kind of bummer, as I will be losing 64 bit support by use i386. This is an
Intel Xeon, which should be compatible with the amd64 branch.
In any case; when attempting to run the i386 bsd.mp it hangs here:
mtrr: Pentium Pro MTRR support
It's a hard freeze-up, the keyboard will not
Update:
I ran boot -c with verbose on, and here are the last entries:
various probing failed messages (doesn't look like any problems), then:
ioapic0: conflicting map entries for pin 0
pctr: 686-class user-level performance counters enabled
mtrr: Pentium Pro MTRR Support
hard hang -- no further
Paul Taulborg wrote:
Kind of bummer, as I will be losing 64 bit support by use i386. This is
an Intel Xeon, which should be compatible with the amd64 branch.
I am not expert to say yes or no here. May be someone else will confirm
or deny. For now I would assume wrongly may be, but I wouldn't
Paul Taulborg wrote:
Update:
I ran boot -c with verbose on, and here are the last entries:
various probing failed messages (doesn't look like any problems), then:
ioapic0: conflicting map entries for pin 0
pctr: 686-class user-level performance counters enabled
mtrr: Pentium Pro MTRR Support
i386 GENERIC works, and boots up normally.
i386 SMP (bsd.mp) hangs at the line below (and additional lines in my
other message).
I have to hard power the machine down at this hang (i386 SMP kernel).
This same exact thing occurs with both 4.1 and current (just downloaded
the current today).
Paul Taulborg wrote:
Kind of bummer, as I will be losing 64 bit support by use i386. This is
an Intel Xeon, which should be compatible with the amd64 branch.
To answer your question, I guess it depend on the version of your processor.
http://marc.info/?l=openbsd-miscm=117112049507303w=2
I
Paul,
If you want to try the AMD64 mp kernel with the patch I point out to you
on tech of a few days ago and see if that help you or not, I can make
the kernel I built that night for you to download and try if you trust
it. I would say to built your own, but if you want to do a quick test
Paul Taulborg wrote:
I went through every option in the BIOS, and there is nothing at all
related to ACPI. :(
Your BIOS is version 35, and there is a very long list of BIOS upgrades
from Intel. The latest one for this board, if I am not mistaken is 44
and you have 35.
bios0: vendor Intel
David B. wrote:
Hi,
are there any plans to release a bsd.mp version for sparc64? My box
currently can only use cpu0; I have 4 processors, and it seems a shame
to waste all of that power.
thanks
There is, of course, desire to support SMP on mvme88k, sparc, sparc64,
Alpha, and just about
21 matches
Mail list logo