Re: PROBLEM: CD burning at 12x uses excessive CPU, although DMA is enabled

[Anssi Saari <as@sci.fi>]

On Fri, Aug 30, 2002 at 08:14:53AM +0200, Markus Plail wrote:

> >What is strange is that problem dont appears with ISO images
> >(cdrecord-burned). There is any already-know solution for that problem
> >with VIA686b IDE?

> The problem isn't really the southbridge, the problem is, that the
> kernel doesn't seem to use (I still didn't get a definitive answer)
> DMA when doing things with bigger blocksizes (grabbing audio CDs,
> writing DAO CDs). 

I doubt that's the problem. As I've said before, I don't have this huge
slowdown problem if I plug the writer to a Promise pdc20265 or CMD649,
neither of which supports DMA for ATAPI devices. These controllers
however abort CD writing randomly so they are not a workaround... I
also don't have your DAO vs. TAO problem.

> It seems that older chipsets just deal way better with PIO
> modes (the PIIX3 doesn't have DMA, does it?).

Yes it does.

> For me it helped at least a bit to do the following:
> - Disable interrupt sharing for IDE devices
> - hdparm -d1 -c3 -u0 -X34/66 /dev/burner

I've tried disabling interrupt sharing for IDE that but it didn't
change anything.  hdparm -c3 I haven't tried yet, I'll have to see if
it would help.