lost interrupt hell - Plea for Help

Hi all,

*bang*

*bang*

Hear that? Its my head, against the wall. ;)

I am trying to convert my audio cds to mp3, which involves first ripping the track as a wav.

I have two drives, one Creative DVD 5x, one LG CDRW drive. My chipet is VIA K7 266 pro, 1.2ghz duron.

I have tried every combination of master / slave between the two drives, the drives on their own, scsi emulation through ide-scsi, purely as IDE drives, ommitting ide cdrom support from teh kernel completely and only using ide-scsi... every time I try to get a track ripped, dmesg fills up with hdX: lost interrupt.

If I try to rip from the DVD drive, the system hangs and its reset button time.

I am using 2.4.18.

Can anyone tell me where I'm going wrong? Is there anything from my system you need to see to help me?

Many thanks,

mikeH

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi again. I've thought I've already given up but just to be sure and not forget about anything (yes, some people - like me - don't know when to stop and go home):

1. Have you tried another ide cables?
2. Do they stick ok?
3. How about unplugging CD completely from the motherboard?
4. Perhaps you need to compile via-specific options into your kernel.

In my case, every time I've seen lost interrupt it was something from the above.

bye
T.

- --

** A C programmer asked whether computer had Buddha's nature.      **
** As the answer, master did "rm -rif" on the programmer's home    **
** directory. And then the C programmer became enlightened...      **
**                                                                 **
** Tomasz Rola          mailto:[email protected]             **

-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 5.0i for non-commercial use
Charset: noconv

iQA/AwUBPNpe5RETUsyL9vbiEQITcgCgkbVDKfJBOgysvqSueVhfkpFHPWYAoJcN uVjucDNFdWbqhQ8XBse+TxNl
=9MG2
-----END PGP SIGNATURE-----

But none of the above justifies kernel hang. It's a bug to fix. There is IDE maintainer for 2.4:

Andre Hedrick <[email protected]> [09 apr 2002]
ATA/ATAPI Storage Architect [2.0,2.2,2.4] HBA interface developer
Serial ATA Architect [future release] Voting NCITS member AT-Attachment Committee

Ask him what details he needs and maybe he will be able to help. --
vda

I have same problem (see some other emails from me, I haven't posted
about anything else), via motherboard, audio cd ripping doesn't work (I get lost interrupt messages).

I am pretty sure it's kernel problem (or motherboard design issue,
which is sort of kernel problem as well), I tried (none of it makes any difference):

- different cd-rom drivers (same situation),
- dma on/off
- 32 bit access on/off

everything else except of audio ripping works (burning, data cd
reading, HD connected instead of cdrom drive etc.)

PCI IDE card works (in same conmputer, same cdrom drive, same IDE
cables).

the IRQ works ok, as long as the activity is something else than audio
ripping - the numbers in /proc/interrupts go up, there doesn't seem to be any interrupt conflict (it's assigned by BIOS, IRQ 15, nothing else seems to be using 15, at least nothing that would be visible on bios bootup screen (it prints irq assignments) or linux (/proc/interupts doesn't list 15 at all if the cdrom drive is not there, it lists hdc if the drive is in).

there are other people who have the same problem, it looks like all
are via chipsets related (I just got email from another guy who cannot find a solution either).

this is with stable kernel (I tried 2.4.17 and 2.4.18), via config
option compiled in (I haven't tried without via support in kernel yet, I'll try that).

any ideas? TIA

erik

You can try compiling without VIA chipset support, but it makes no difference.
Now, with the latest prepatches, -ac patches and ide patches, I am getting spurious "8259A interrupt: IRQ7." all over the place too. Seems like the linux kernel does not play well with AMD Cpus + VIA chipsets, which is a real shame as thats what all my machines are :(

mikeH

It's not only with VIA chipsets, I have an Athlon system with a SIS chipset and there I get the spurious "8259A interrupt: IRQ7." as well... luckily the message is only displayed once, but it always appears in the first 15 min after startup.

Rudmer

Sorry for the delay...

I build a 2.4.19-pre8-ac4 kernel without local apic and now after running for 4 hours I still did NOT get the message "spurious 8259A interrupt:
IRQ7." while with apic enabled the message appears after ~3 min... so this is local apic related.

This is an athlon on a SIS-chip mobo.

I also found this in dmesg:
PCI: PCI BIOS revision 2.10 entry at 0xfdb01, last bus=1
PCI: Using configuration type 1
PCI: Probing PCI hardware
Unknown bridge resource 0: assuming transparent
PCI: Using IRQ router SIS [1039/0008] at 00:02.0

and this from lspci:
00:00.0 Host bridge: Silicon Integrated Systems [SiS]: Unknown device 0735 (rev 01)

also related??

Rudmer

Yeah, the spurious interrupt does seem to be an AMD apic problem, but the lost interrupt (on ripping audio) seems to be a VIA chipset problem, as people with KT266 chipsets are having boot problems / audio rip problems regardless of the processor type. Lucky me, I get both ;)

mike

doesn't anybody know what the problem is? while it was reported quite
a few times there was no helpful response to the problem (positive (fix)or negative (it's unlikely to be fixed).

I mean I would expect that somebody at least suspects what the problem
is and can say whether there is any chance it will be fixed or otherwise shed some light on the problem. anything...

I am talking about the via and audio ripping problem. I have tried
2.4.17, 2.4.18 and 2.4.18 with ide patches from linux-ide.org (2.4.19-pr7 didn't compile)

?

TIA

erik

I have just switched motherboards in my file server, which previously had no problems ripping audio from my Creative 52X drives. The new motherboard has the KT266A chipset, but the CD drives are _not_ connected to that chipset's IDE ports. I am getting "lost interrupt" messages when I try to rip audio from the drives, or even mount ISO9660 discs (which do eventually succeed, they just take over a minute to mount). So far I have done the following:

- turned off "dma" and "unmaskirq" for the CD drives
- tried ide-scsi/sg instead of ide-cd
- tried booting with "noapic"
- tried 2.4.19-pre8 and 2.4.19-pre8-ac4

Nothing has helped. The machine configuration is an MSI KT7266-Pro2RU motherboard, KT266A chipset with on-board Promise PDC20265R FastTrak "lite"). There is also a Promise PDC20262 (Ultra66TX2) in a PCI slot, and that is where the CD drives are connected. Each CD drive is the master on its channel, and one of them also has a Iomega ZIP 250 ATAPI drive as its slave. Interestingly, the ZIP drive works perfectly, no "lost interrupt" messages at all.

lspci output is below:

00:00.0 Host bridge: VIA Technologies, Inc. VT8367 [KT266] 00:01.0 PCI bridge: VIA Technologies, Inc. VT8367 [KT266 AGP] 00:05.0 SCSI storage controller: Adaptec AHA-7850 (rev 03) 00:06.0 Ethernet controller: National Semiconductor Corporation DP83815 (MacPhyter) Ethernet Controller
00:07.0 Ethernet controller: Lite-On Communications Inc LNE100TX (rev 20) 00:08.0 Ethernet controller: National Semiconductor Corporation DP83820 10/100/1000 Ethernet Controller
00:09.0 Unknown mass storage controller: Promise Technology, Inc. 20262 (rev
01)
00:0b.0 USB Controller: NEC Corporation USB (rev 41)
00:0b.1 USB Controller: NEC Corporation USB (rev 41)
00:0b.2 USB Controller: NEC Corporation USB 2.0 (rev 02)
00:0c.0 RAID bus controller: Promise Technology, Inc. 20265 (rev 02) 00:11.0 ISA bridge: VIA Technologies, Inc. VT8233 PCI to ISA Bridge 00:11.1 IDE interface: VIA Technologies, Inc. Bus Master IDE (rev 06) 01:00.0 VGA compatible controller: ATI Technologies Inc Rage 128 RF/SG AGP

relevant sections of .config:

CONFIG_X86=y
CONFIG_ISA=y
CONFIG_UID16=y
CONFIG_EXPERIMENTAL=y
CONFIG_MODULES=y
CONFIG_KMOD=y
CONFIG_MK7=y
CONFIG_X86_WP_WORKS_OK=y
CONFIG_X86_INVLPG=y
CONFIG_X86_CMPXCHG=y
CONFIG_X86_XADD=y
CONFIG_X86_BSWAP=y
CONFIG_X86_POPAD_OK=y
CONFIG_RWSEM_XCHGADD_ALGORITHM=y
CONFIG_X86_TSC=y
CONFIG_X86_GOOD_APIC=y
CONFIG_X86_USE_3DNOW=y
CONFIG_X86_PGE=y
CONFIG_X86_USE_PPRO_CHECKSUM=y
CONFIG_X86_MCE=y
CONFIG_NOHIGHMEM=y
CONFIG_MTRR=y
CONFIG_X86_UP_APIC=y
CONFIG_X86_UP_IOAPIC=y
CONFIG_X86_LOCAL_APIC=y
CONFIG_X86_IO_APIC=y
CONFIG_PCI=y
CONFIG_PCI_GODIRECT=y
CONFIG_PCI_DIRECT=y
CONFIG_PCI_NAMES=y
CONFIG_BLK_DEV_IDE=y
CONFIG_BLK_DEV_IDEDISK=y
CONFIG_IDEDISK_MULTI_MODE=y
CONFIG_BLK_DEV_IDECD=m
CONFIG_BLK_DEV_IDEFLOPPY=y
CONFIG_BLK_DEV_IDESCSI=m
CONFIG_BLK_DEV_IDEPCI=y
CONFIG_IDEPCI_SHARE_IRQ=y
CONFIG_BLK_DEV_IDEDMA_PCI=y
CONFIG_IDEDMA_PCI_AUTO=y
CONFIG_BLK_DEV_IDEDMA=y
CONFIG_BLK_DEV_PDC202XX=y
CONFIG_PDC202XX_BURST=y
CONFIG_PDC202XX_FORCE=y
CONFIG_BLK_DEV_VIA82CXXX=y
CONFIG_IDEDMA_AUTO=y
CONFIG_BLK_DEV_IDE_MODES=y

----- Original Message -----
From: "Erik Steffl" <[email protected]>
To: <[email protected]>
Sent: Sunday, May 19, 2002 12:14 AM
Subject: Re: lost interrupt hell - Plea for Help

the lost

people

regardless

Can you rip audio tracks to wav files ? If not, this might be a problem with the apic on amd chips rather than an IDE problem.

Never tried to rip audio tracks... will try it after the weekend

I do have local APIC enabled, so it could be that the problem is there... will build a kernel without apic (after the weekend)

Rudmer

FWIW I have intel processor. It seems like the 'lost interrupt' while
ripping audio CDs is specific to VIA based motherboards.

erik

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

And - last but not least:

5. Interrupts - perhaps try to remove everything from your computer, except gfx, cpu, ram, and ide drives (hdds and dvd, not cd). But I think of this as a last resort.

6. Related to 5 - I would also check DVD with different hardware or/and with different OS - just to be sure. If it's not working then perhaps it is faulty. Or your motherboard, maybe.

bye
T.

- --

** A C programmer asked whether computer had Buddha's nature.      **
** As the answer, master did "rm -rif" on the programmer's home    **
** directory. And then the C programmer became enlightened...      **
**                                                                 **
** Tomasz Rola          mailto:[email protected]             **

-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 5.0i for non-commercial use
Charset: noconv

iQA/AwUBPNpgyRETUsyL9vbiEQLsFACgvpY505l1K9b/DhqICryM1+z3GL8AnjTM YaGzPdfhz99Yw1qdTa45vHdx
=/rqc
-----END PGP SIGNATURE-----