AGESA is in assembly code now, and it is only can be released to IBV under NDA. IBV use the source code in their framework.
Current LinuxBIOS support code for rev F are LinuxBIOS native code.
YH
-----Original Message----- From: linuxbios-bounces@linuxbios.org [mailto:linuxbios-bounces@linuxbios.org] On Behalf Of Ronald G Minnich Sent: Thursday, August 10, 2006 9:17 AM To: jf simon Cc: linuxbios@linuxbios.org Subject: Re: [LinuxBIOS] AMD Turion - AGESA
jf simon wrote:
Ronald G Minnich wrote:
looks like another nice piece of software to support binary BIOSes?
Is
the source available? ron
No. AMD told us they give it to bios vendors only. -jf simon
my suspicion is we don't need it.
YH? Any word on what this is?
ron
Lu, Yinghai wrote:
AGESA is in assembly code now, and it is only can be released to IBV under NDA. IBV use the source code in their framework.
Current LinuxBIOS support code for rev F are LinuxBIOS native code.
so we don't need agesa.
ron
Hi Ron,
Ronald G Minnich wrote:
Lu, Yinghai wrote:
AGESA is in assembly code now, and it is only can be released to IBV under NDA. IBV use the source code in their framework.
Current LinuxBIOS support code for rev F are LinuxBIOS native code.
so we don't need agesa.
ron
I have read on the mailing list that Intel didn't want to see linuxbios running on their CPUs (certainly a shame). But it was my understanding that AMD was willing to share information so that linubios runs on AMD CPUs. Is this true? The agesa thing seems to contradict this since AMD doesn't provide the code of it. Thanks -jf simon
p5.vert.ukl.yahoo.com uncompressed Fri Aug 11 12:27:00 GMT 2006 ___________________________________________________________________________ Découvrez un nouveau moyen de poser toutes vos questions quelque soit le sujet ! Yahoo! Questions/Réponses pour partager vos connaissances, vos opinions et vos expériences. http://fr.answers.yahoo.com
jf simon jfaslist@yahoo.fr writes:
I have read on the mailing list that Intel didn't want to see linuxbios running on their CPUs (certainly a shame). But it was my understanding that AMD was willing to share information so that linubios runs on AMD CPUs. Is this true? The agesa thing seems to contradict this since AMD doesn't provide the code of it.
Neither is true. As this thread clarified even the latest AMD work does not require AGESA.
In addition while it sucks to work with Intel you can work with them, if you have an appropriate business case. We coming close to getting having a port to the latest Intel chipset completed.
There are camps within Intel that are strongly opposed to the idea of LinuxBIOS in general. But those camps are not the ones making the decision.
AMD has a general policy of being friendly and enabling inovators (or something like that), making AMD much easier to deal with.
Eric
On Fri, Aug 11, 2006 at 02:32:33PM +0200, jf simon wrote:
But it was my understanding that AMD was willing to share information so that linubios runs on AMD CPUs. Is this true?
It is certainly true. There is at least one person active on this mailing list working for AMD who contributes a lot of information and code.
The agesa thing seems to contradict this since AMD doesn't provide the code of it.
On Fri, Aug 11, 2006 at 06:44:55AM -0600, Eric W. Biederman wrote:
As this thread clarified even the latest AMD work does not require AGESA.
To re-iterate, AGESA is not useful in LinuxBIOS - from what I understand it's a bunch of assembly that "plugs in" to traditional BIOS codebases in order to quickly assemble a new BIOS for a particular AMD CPU.
Traditional BIOS codebases are generally all assembly and not very well-designed. I would argue that LinuxBIOS is the opposite.
LinuxBIOS has excellent support for several AMD CPUs, lately a lot of work is going into the GX2 platform since that's what OLPC is using. Earlier lots of effort was put into making Opteron based servers work perfectly.
Look around in the source tree. It has a little bit of a learning curve (which I haven't climbed to the top) but you should be able find your way around pretty quickly.
I'd suggest looking around in the src/cpu/amd/model_fxx and src/northbridge/amd/amdk8 directories for interesting code.
E.g. src/cpu/amd/model_fxx/model_fxx_init.c at 14kb might be what you're asking for.
//Peter
LinuxBIOS has excellent support for several AMD CPUs, lately a lot of work is going into the GX2 platform since that's what OLPC is using. Earlier lots of effort was put into making Opteron based
Indeed and its just about ready to come alive and kicking. Just a few more kinks to work out. Most of which aren't really even LinuxBIOS kinks. Look for to show up on a in a test board RSN.
Lu, Yinghai wrote:
AGESA is in assembly code now, and it is only can be released to IBV under NDA. IBV use the source code in their framework.
Current LinuxBIOS support code for rev F are LinuxBIOS native code.
YH
I am new to linuxbios and I would appreciate if you could point me to the place where linuxbios is doing what AGESA does (that is when the CPU is set from its min P-state to max P-State)? Thanks a lot -jf simon
p4.vert.ukl.yahoo.com uncompressed Thu Aug 10 16:27:00 GMT 2006 ___________________________________________________________________________ Découvrez un nouveau moyen de poser toutes vos questions quelque soit le sujet ! Yahoo! Questions/Réponses pour partager vos connaissances, vos opinions et vos expériences. http://fr.answers.yahoo.com
Lu, Yinghai wrote:
AGESA is in assembly code now, and it is only can be released to IBV under NDA. IBV use the source code in their framework.
Current LinuxBIOS support code for rev F are LinuxBIOS native code.
Is all the source to support the NPT Family 10h DDR2 processors now in the current tree?
-Bari
-
Bari Ari -
ebiederman＠lnxi.com -
jf simon -
Lu, Yinghai -
Peter Stuge -
Richard Smith -
Ronald G Minnich