Ronald G Minnich <[EMAIL PROTECTED]> writes:
> This causes PIII/800 systems to hang. I'm beginning to think that for
> these more tricky optimizations I'd like to do it when Linux hits runlevel
> 3. This will ensure that we can always get to runlevel 1, so that if there
> is a problem we can recover in linux.
>
> More detail:
>
> For every chipset, I would like the chipset to have 'guaranteed' timing.
> This timing should stay that way until Linux runlevel 1. Then, as part of
> the transition to Runlevel 3, a script or a smart program could set bits
> that improve performance.
>
> There are a lot of advantages to this idea.
>
> 1) Even if you get the performance bits wrong, you can always get the
> machine up to Linux. It's much easier to debug this way.
> 2) You can test new enhancements, and not worry that the machine will
> simply go comatose. We've seen this with regular BIOS, where the only
> option once you set something wrong is to open the case and clear CMOS.
> Imagine doing this to 1024 machines.
>
> Comments?
This is a good point.
Ron do you envision upgrading the flash on multiple machines in production?
Machines are usually put together and setup in a seperate step from
actuall use. And then once the machines work reliably the machines
are put into production.
Usually there is manual attention given to each node before it is put
into production so having to fix the machines is not fatal at that
point. Also probably only one or two are set up as test cases before
the rest are similiarly optimized.
Then there is the case of SDRAM timings. Which you really can't mess
with once you are up and running.
I think a better option is likely to have a linuxBIOS booter than
reads a flag and decides which copy of linuxBIOS it will boot. And
then put in some information so that if a node fails to boot the
booter won't boot the new linuxBIOS image.
Eric
