On Thu, 31 Jan 2002, Geoffrey wrote: > Douglas Gilbert wrote:
> > Chuck, > > It is difficult to help when you give so little information. > > Lets assume you have an IDE ATAPI cdwriter connected to /dev/hdd > > In this case it may be worth turning off (or reducing) the DMA > > with either one of these 2 commands: > > hdparm -d0 -c1 /dev/hdd > > hdparm -d 1 -X 34 /dev/hdd > > > I attempted this same solution upon your request before, but since the > IDE drive is seen as a scsi drive , the comand returns: > > /dev/hdd: No such device or address > > Suggestions? Are you running a current kernel with "use DMA when you can" enabled in the "generic IDE" config? Is your drive on hdd at all? The no such device is a clue, and if you didn't change it to the actual connection point of your device it obviously won't work. I missed the X34 the first time, never had the thought to try that, and won't until I take a backup, but enabling DMS on generic chipsets was what it took to improve things on my systems. <ExpertMode> You can also find help by using some of the new scheduling patches, preempt or low latency, the new J6 or later scheduler, etc. Be aware that this stuff is experimental, although it has been stable for me even when mixed with rmap and a few other things. Currently typing on: 2.4.17-jl15-ll-ngpt and about to hack in child run first this weekend. </ExpertMode> -- -bill davidsen ([EMAIL PROTECTED]) "The secret to procrastination is to put things off until the last possible moment - but no longer" -me -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
