On Thu, 08 Jun 2000, you wrote:
> Can someone help me here?
>
> VT 82C597 Apollo VP3
>
> hda: Maxtor 84000A6, ATA DISK drive
> hdb: WDC WD181AA, ATA DISK drive
if you reverse the order of these drives, you could destroy them. PLEASE see
http://kt.linuxcare.com/kernel-traffic/kt20000214_54.epl (second item)
before something happens.
Put one with your CD or something, but get them off the same channel.
and hda is faster because it is a Maxtor and can use real DMA, and it has a
better timing setup and it is the MASTER--never expect the slave to be faster.
>
> hda: 7825932 sectors (4007 MB) w/256KiB Cache, CHS=970/128/63, DMA
> hdb: 35419104 sectors (18135 MB) w/2048KiB Cache, CHS=2204/255/63,
> UDMA(33)
>
> So why is hda faster?
>
> DMA off:
>
> /dev/hda:
> Timing buffered disk reads: 64 MB in 10.19 seconds = 6.28 MB/sec
> /dev/hdb:
> Timing buffered disk reads: 64 MB in 12.60 seconds = 5.08 MB/sec
>
> DMA on:
>
> /dev/hda:
> Timing buffered disk reads: 64 MB in 7.68 seconds = 8.33 MB/sec
> /dev/hdb:
> Timing buffered disk reads: hdb: timeout waiting for DMA
> ide_dmaproc: chipset supported ide_dma_timeout func only: 12
> hdb: irq timeout: status=0x58 { DriveReady SeekComplete DataRequest }
Fortunate you are that that is all that is happening. Kernel Panic "lost
interupt" is another possibility as is no error message (but corrupted data)
>
> OK, so what is wrong here?
Read the article--it is a signal reflection under advanced pentium code.
Civileme
