Hi Anssi,

Anssi Saari wrote on Sat, Oct 12 2002:

>How do you explain then that I for example can't write audio CDs any
>faster than 14x in Linux and the system is hopelessly bogged down due
>to this light IO task already at 8x writing?

I am not Joerg, but I experienced the same thing during early trials.
In my case it was simply a non optimal kernel-setup. The whole thing
seems to be highly dependant on the chipset of the motherboard
(meaning the kernel driver support for the chipset). My linux box has
an elderly AMD-K6-III-400 CPU and an ALI 15x3 chipset.

Here's my .config for Kernel 2.4.18 (only relevant parts quoted):

#
# ATA/IDE/MFM/RLL support
#
CONFIG_IDE=y

#
# IDE, ATA and ATAPI Block devices
#
CONFIG_BLK_DEV_IDE=y
CONFIG_BLK_DEV_IDESCSI=y
CONFIG_BLK_DEV_IDEPCI=y
CONFIG_BLK_DEV_IDEDMA_PCI=y
CONFIG_BLK_DEV_ADMA=y
CONFIG_IDEDMA_PCI_AUTO=y
CONFIG_BLK_DEV_IDEDMA=y
CONFIG_BLK_DEV_ALI15X3=y
CONFIG_IDEDMA_AUTO=y
CONFIG_BLK_DEV_IDE_MODES=y

I do *not* have IDE-CD driver in the kernel, have kernel boot-parm
"ide0=ata66 hda=ide-scsi" in /etc/lilo.conf and issue 
"hdparm -d1 -X66 -c3  /dev/hda" during system-initialization.

I found out, that it is important to activate the chipset-drivers for
my ALI15x3 chipset (see above).

>And in FreeBSD and Windows this is not a problem?

I have a network-connected Win2K box with a PX-W4824A and here is
absolutely no problem whatsoever. Just right-click the IDE-controller
in "Device Manager", select Properties and switch the drop-down box to
"DMA when available", reboot.

>On the other hand, this slowdown happens only if the writer is
>connected to VIA 686b.

This feeds my suspicions about driver-support of the chipset.

>If it's connected to a
>CMD 649, Promise pdc20265 or Intel PIIX3 there's no problem. 

I could not get this Promise controller to work with a cd-writer.

>I think Joerg put it very well in calling it a kernel design bug, as
>that's exactly what it is.

We probably all agree on that :-/ the IDE drivers have been orphaned
for quite a while now and the current development (almost no scsi
cd-writers available) has put further tension on the situation.

>The only good thing about it is that it doesn't affect many IDE
>controllers, the VIA 686b is the only one I know of.

It really seems to be highly hardware-dependant.

Kind regards   Frederick


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Reply via email to