On Tue, 29 Feb 2000, WANG,YIDING (HP-SanJose,ex1) wrote:
> Marc,
>
> Thanks for the valuable suggestion! However, passing parameters to the
> driver will not do the work for me. Basically what I am trying to do is to
> set up persistent binding environment for Linux. This happens to Fibre
> Channel world when new device plugged into the loop or Fibric, the existing
> device address could change which leads to target id changes. The persisten
> binding is to preserve all existing device target id unchanged. This
> requires to put existing target id and the coresponding device world wide
> name pair somewhere in a file (or ROM), and be a refernece for the system
> next boot.
>
> If Linux has some system calls to allow driver access from a file during the
> loading process, the solution will be obvious.
>
If this only has to work on an ix86 machine, you could put a few longwords
of data in CMOS. Every recent machine I've seen has plenty of room. The
only problem is finding out how the BIOS checksums the current contents.
You only have to do this once, but it has to be done. Some checksum the
entire data area, some only checksum from the top of the 'clock' to the
end of their data. You have to find out what works because you have
to checksum using the BIOS protocol or else the next boot will fail due
to a checksum error. Because this has to be compatible amongst many
BIOSes, it has to do the initial 'protocol' check at runtime.
Cheers,
Dick Johnson
Penguin : Linux version 2.3.41 on an i686 machine (800.63 BogoMips).
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Tue Feb 29 2000 - 21:00:23 EST