On Sep 27 at 21:35, Manuel Bouyer spoke:

> Well, if it only downgrades to  Ultra-DMA mode 1, it's still using DMA.

Hm. I cut off too mutch:

Sep 27 23:03:02 snoopy /netbsd: pciide1:0:0: lost interrupt
Sep 27 23:03:02 snoopy /netbsd:         type: atapi tc_bcount: 32 tc_skip: 0
Sep 27 23:03:02 snoopy /netbsd: pciide1:0:0: bus-master DMA error: missing interrupt, status=0x21
Sep 27 23:03:02 snoopy /netbsd: cd0: transfer error, downgrading to Ultra-DMA mode 1
Sep 27 23:03:02 snoopy /netbsd: cd0(pciide1:0:0): using PIO mode 4, Ultra-DMA mode 1 (using DMA data transfers)
Sep 27 23:03:03 snoopy /netbsd: pciide1:0:0: lost interrupt
Sep 27 23:03:03 snoopy /netbsd:         type: atapi tc_bcount: 32 tc_skip: 0
Sep 27 23:03:03 snoopy /netbsd: pciide1:0:0: bus-master DMA error: missing interrupt, status=0x21
Sep 27 23:03:03 snoopy /netbsd: cd0: transfer error, downgrading to DMA mode 2
Sep 27 23:03:03 snoopy /netbsd: cd0(pciide1:0:0): using PIO mode 4, DMA mode 2 (using DMA data transfers)
Sep 27 23:03:04 snoopy /netbsd: pciide1:0:0: lost interrupt
Sep 27 23:03:04 snoopy /netbsd:         type: atapi tc_bcount: 32 tc_skip: 0
Sep 27 23:03:04 snoopy /netbsd: pciide1:0:0: bus-master DMA error: missing interrupt, status=0x21
Sep 27 23:03:04 snoopy /netbsd: cd0: transfer error, downgrading to PIO mode 4
Sep 27 23:03:04 snoopy /netbsd: cd0(pciide1:0:0): using PIO mode 4


Interestingly DMA mode 1 is skipped. So I tried:

cd*	at atapibus? drive ? flags 0x0f90	# ATAPI CD-ROM drives

But this and some other combinations except 0x0ff0 resulted in tons
of such messages while trying cdrecord -scanbus:


Sep 27 23:16:43 snoopy /netbsd: pciide2:0:0: lost interrupt
Sep 27 23:16:43 snoopy /netbsd:         type: atapi tc_bcount: 32 tc_skip: 0
Sep 27 23:16:43 snoopy /netbsd: pciide2:0:0: bus-master DMA error: missing interrupt, status=0x21
Sep 27 23:16:44 snoopy /netbsd: pciide2:1:0: lost interrupt
Sep 27 23:16:44 snoopy /netbsd:         type: atapi tc_bcount: 32 tc_skip: 0
Sep 27 23:16:44 snoopy /netbsd: pciide2:1:0: bus-master DMA error: missing interrupt, status=0x1


The ATA controllers are Promise ATA 100 TX2 and ATA 133 TX2. Is
there a known problem with these?

> I found that 80 pin cables could improve the situation on some occasions,
> even for Ultra-DMA 2 (where it's not required). If you have one around,
> it's worth a try.

I tried one. But the result is the same.

-Hanspeter