Lots of people do. For instance me 8)
> simple DMA solves all latency problems, and the performance
> difference between DMA and UDMA on a CDROM drive is almost
> zero. (since the CDROM is the bottleneck).
Try a 72x cdrom sometime. Yes they do exist.
> > > But are you sure that every ATAPI CDROM drive on the planet
> > > has this hardware flaw.
> > Its the controller not the CDROM at fault
> are you sured that the BX chipset is flawed ?
> It seems to work well.
I never said the Intel 440bx chipset is flawed. I said the VIA
VP3/MVP3/MVP4 is flawed. 440bx is fine AFAIK.
> > > I stressed my CDROM drives quite a bit, and was unable to lock up the box.
> > What controller and what drives
> Mainboard Gigabyte GA 6BXD (dual) with BX chipset.
Then you should be ok
> > > How is this handled my other OSes ?
> > They dont run DMA.
> Wrong: try to dig on deja.com searching articles about
> DVD software playback on windoze:
> for most people it is almost impossible to get skipfree video
> without DMA.
This is irrelevant if the controller cant run DMA (eg promise u66).
> > > > Promise Ultra66. No DMA on CDROM/DVDs. It just wont work. At all.
> > > PS: why should recent ATAPI CDROM let you set the DMA mode, when they
> > > know that it could lead to problems ?
> > Its the controller not the CDROM at fault
> I was wondering how much % of the controllers are faulty in this respect.
Lots. All VIA VP3/MVP3/MVP4 boards for example. This is a large percentage
of boards now shipping, especially asia
-Dan
-
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/