On Saturday 27 January 2007 20:43, Rony wrote:
Hello All,
Yesterday when I was searching for core2 info on the net, I came across complaints that the Intel 965 board PATa (parallel IDE port) is not detected by the Linux distros and the alternative is to use a SATA CD ROM or usb install. Even Dinesh J. had mentioned it.
My query is that why does a distro need to understand or talk to the IDE or PATA port controller chips when the port itself is a universal port with pins dedicated to power, read, write, address and data?
If only things were that simple. Or rather if chipset mfg would keep things that simple.
Shouldn't the BIOS be the interface that takes communication _internally_ between the port and its controller chip?
Because bios are very buggy and slow (aprticularly for disk access). To overcome that you copy bios to ram and then overwrite it with your own. Or avoid it all together - you are ofcourse presuming that the bios has left the system in a sane state when it jumped to the boot sector.
Another point is that why is the old XP SP2 2002 edition able to talk to the _new_ PATA controllers on 965 while the new linux distros cannot? Is there a windows specific command structure secretly available between M$ and the BIOS manufacturers?
You got it. Sign an NDA with the chipset mfg and the bios vendor. Why cause the chipset has bugs which the bios corrects and u would not buy something with known bugs would you. Headover to the linuxbios mailing list to know about the games companies play.