Re: TritonII IDE interface not PCI compliant?

Alan Cox (alan@lxorguk.ukuu.org.uk)
Tue, 17 Jun 1997 19:33:13 +0100 (BST)


> I've now concluded that by doing something like a memcpy, the CPU
> can starve PCI busmasters to death.

That should be settable by the BIOS into the chipset.

There are worse problems with some chipsets it seems. Trying to run a BT848 based
video grabber flat out with a clone chipset I've seen stale cache lines and memory
corruption