NetBSD-Bugs archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
Re: bin/38667: cpuctl does not properly deal with AMD Phenom
The following reply was made to PR bin/38667; it has been noted by GNATS.
From: Paul Goyette <paul%whooppee.com@localhost>
To: Chris Gilbert <chris%dokein.co.uk@localhost>
Cc: gnats-bugs%NetBSD.org@localhost, gnats-admin%netbsd.org@localhost,
netbsd-bugs%netbsd.org@localhost
Subject: Re: bin/38667: cpuctl does not properly deal with AMD Phenom
Date: Sat, 17 May 2008 05:45:52 -0700 (PDT)
On Sat, 17 May 2008, Chris Gilbert wrote:
>> + case 0x01:
>> + switch (model) {
>> + case 0x02:
>> + ret = "Phenom";
>
> I think this would have to actually be "Family 10h", as Opteron's also
> report the same cpuid. See page 5 of:
> http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/41322.pdf
>
> If you're ok with that I'll check in the change.
Hmmm, when all this was still in-kernel we were able to differentiate
(or at least, unambiguously identify my quad-core Phenom).
Here's the dmesg output from previous 4.99.60 kernel:
cpu0 at mainbus0 apid 0: (boot processor)
cpu0: AMD Phenom(tm) 9600 Quad-Core Processor, 2310.80 MHz
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
So it seems to me we lost something when moving this all to userland.
> I guess we should also pick-up and fix errata 298 if necessary.
> Although that is more complex as I believe the registers have to be
> updated on every core.
Not sure what errata 298 is - guess I'll have to research that.
----------------------------------------------------------------------
| Paul Goyette | PGP DSS Key fingerprint: | E-mail addresses: |
| Customer Service | FA29 0E3B 35AF E8AE 6651 | paul%whooppee.com@localhost
|
| Network Engineer | 0786 F758 55DE 53BA 7731 | pgoyette%juniper.net@localhost
|
----------------------------------------------------------------------
Home |
Main Index |
Thread Index |
Old Index