On Thu, 11 Aug 2005, Sid Boyce wrote:
This driver is not mentioned directly on Intel's pages about the D945GNT motherboard (they're afraid of Microsoft to say it loud, I suppose), but google will eventually dig it out.
In my reply I mentioned I'd seen reference to it on the kernel mailing list last week, a post from someone at Intel.
Sid, I'm sorry I missed your reply. As far as I can tell, it didn't show up on this list. Can you please resend it, or at least send me a copy of it? Thank you.
1. It won't boot from a harddrive unless one of the partitions is marked as bootable (Windows style) in the partition table. The BIOS (tested the newest one - 1788) is quirky alltogether and many of the features are twisted or don't work as described in the documentation. Why in the world would somebody make the BIOS read the partition table?
I always thought a partition had to be made bootable in order to be able to boot from it. I've always done that since the days when we used Minix bootlace and shoelace to boot Linux. You'd also note that on any Linux install of any flavour, partitioning the drive takes care of that.
My point here is that its not the job of the BIOS to read the partition table and decide if a partition is bootable or not. This is the job of the boot loader. There are so many useful things that only the BIOS can do and it doesn't, so why waste EPROM space for stuff like that and impose some stupid un-necesarry (and un-documented) rules. Thank you, Florin