From: Ronald G. Minnich On Mon, 7 Mar 2005, Dmitriy Budko wrote:
Does anybody needs LinuxBIOS for VMware virtual machines? If you want it please describe why do you want it.
it would sure make it easier to do full-up sims of linuxbios-equipped systems.
So, you don't need it for "production" systems, only for development, do you?
How difficult is it to port Intel 440BX chipset support from V1 to V2?
It seems to me that supporting a new SIO chips (the emulated NS PC97338/PC87338) should be easy.
Dmitriy Budko, VMware
"Dmitriy Budko" dbudko@vmware.com writes:
From: Ronald G. Minnich On Mon, 7 Mar 2005, Dmitriy Budko wrote:
Does anybody needs LinuxBIOS for VMware virtual machines? If you want it please describe why do you want it.
it would sure make it easier to do full-up sims of linuxbios-equipped systems.
So, you don't need it for "production" systems, only for development, do you?
How difficult is it to port Intel 440BX chipset support from V1 to V2?
Not terribly hard. It is more of a time/desire thing. Does vmware accurately simulate what is required to bring memory up or do we get to short cut that. If the latter the porting could be done very quickly. Memory initialization is generally the hard part, of a LinuxBIOS port and would likely be trivial under vmware :)
It seems to me that supporting a new SIO chips (the emulated NS PC97338/PC87338) should be easy.
Right it should not be a problem there.
The case that was discussed recently with vmware was to do things the other way around. The observation was made that for server clusters dedicated to running vmware, running LinuxBIOS underneath could help boot time and cluster management.
Eric
How difficult is it to port Intel 440BX chipset support from V1 to V2?
Not terribly hard. It is more of a time/desire thing. Does vmware accurately simulate what is required to bring memory up or do we
Yep. not to bad. I've got the beginnings of the port already done. I got hung up on getting my dump_spd routine to return somthing else besides zero. Its really wierd. I can actually see the data on the SMbus happening but I don't ever seem to get a result back.
Like Eric says if you don't have to to the RAM init then it will probally move really quickly.
I haven't had chance to work on it for a while. If your are interested I can whip up a patch vs current cvs and send it to you. It's not in the V2 tree.
On Mon, 7 Mar 2005, Richard Smith wrote:
Yep. not to bad. I've got the beginnings of the port already done. I got hung up on getting my dump_spd routine to return somthing else besides zero. Its really wierd. I can actually see the data on the SMbus happening but I don't ever seem to get a result back.
ah, richard, it just occurred to me. A lot of the spd code starts the op, and waits for the 'active' bit to go to 0 in the status. If the CPU is too fast, it will sample the 'active' bit before the smbus hardware starts to do anything. Then you read back bad data.
As I found on some chipsets, the CPU can be too fast, and it should: 1. start op 2. wait for 'smbus active' indicator to go to 1 3. wait for 'smbus active' indicator to go to 0
is this by any chance your problem?
ron
As I found on some chipsets, the CPU can be too fast, and it should:
- start op
- wait for 'smbus active' indicator to go to 1
- wait for 'smbus active' indicator to go to 0
is this by any chance your problem?
The code is basiclly a port of the working V1 assembly code converted to C. I seem to remember that the V1 code did exactly what you are talking about. Perhaps I goofed up a flag polarity or some other core piece of info while doing it.
I'll whip up a patch today when I get to work and post it up for anyone interested to look at. Perhaps a fresh set of eyes will catch the issue.
I'll need a place to upload to. Or I guess you can commit to CVS. Didn't someone mention that they were going to provide space for stuff like this?
On Mon, 2005-03-07 at 22:05, Richard Smith wrote:
How difficult is it to port Intel 440BX chipset support from V1 to V2?
Not terribly hard. It is more of a time/desire thing. Does vmware accurately simulate what is required to bring memory up or do we
Yep. not to bad. I've got the beginnings of the port already done. I got hung up on getting my dump_spd routine to return somthing else besides zero. Its really wierd. I can actually see the data on the SMbus happening but I don't ever seem to get a result back.
There are two user space program you can try first in the V1 tree. One is the util/scanspd, the other is util/sis/spd_dump.
Ollie
On Mon, 7 Mar 2005, Dmitriy Budko wrote:
So, you don't need it for "production" systems, only for development, do you?
I would say so.
How difficult is it to port Intel 440BX chipset support from V1 to V2?
easy. Been done. Ask Richard Smith.
It seems to me that supporting a new SIO chips (the emulated NS PC97338/PC87338) should be easy.
yes.
ron
-
Dmitriy Budko -
ebiederman＠lnxi.com -
Li-Ta Lo -
Richard Smith -
Ronald G. Minnich