2.4.23-pre4 ide-scsi irq timeout

2.4.23-pre4 ide-scsi irq timeout

[Norbert Preining]

Hi!

I have been using 2.4.23-pre3 without any problems, but when I booted
2.4.23-pre4 ide-scsi (bound to hdc) could not initialize the cdrom
drive. 

Here is the relevant part of the boot.msg for the bad start:
Uniform Multi-Platform E-IDE driver Revision: 7.00beta4-2.4
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
ICH3M: IDE controller at PCI slot 00:1f.1
PCI: Enabling device 00:1f.1 (0005 -> 0007)
ICH3M: chipset revision 2
ICH3M: not 100% native mode: will probe irqs later
    ide0: BM-DMA at 0x1860-0x1867, BIOS settings: hda:DMA, hdb:pio
    ide1: BM-DMA at 0x1868-0x186f, BIOS settings: hdc:DMA, hdd:pio
hda: TOSHIBA MK4021GAS, ATA DISK drive
blk: queue c039b260, I/O limit 4095Mb (mask 0xffffffff)
hdc: QSI CD-RW/DVD-ROM SBW-242, ATAPI CD/DVD-ROM drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
ide1 at 0x170-0x177,0x376 on irq 15
hda: attached ide-disk driver.
hda: host protected area => 1
hda: 78140160 sectors (40008 MB), CHS=4864/255/63, UDMA(100)
hdc: attached ide-scsi driver.
Partition check:
 hda: hda1 hda2 hda3 hda4
SCSI subsystem driver Revision: 1.00
scsi0 : SCSI host adapter emulation for IDE ATAPI devices
scsi : aborting command due to timeout : pid 0, scsi0, channel 0, id 0, lun 0 0x12 00 00 00 ff 00 
hdc: irq timeout: status=0xd0 { Busy }
hdc: DMA disabled
hdc: ATAPI reset complete
hdc: irq timeout: status=0xc0 { Busy }
hdc: ATAPI reset complete
hdc: irq timeout: status=0xc0 { Busy }
scsi0 channel 0 : resetting for second half of retries.
SCSI bus is being reset for host 0 channel 0.
scsi : aborting command due to timeout : pid 1, scsi0, channel 0, id 0, lun 0 0x12 00 00 00 ff 00 
SCSI host 0 abort (pid 1) timed out - resetting
SCSI bus is being reset for host 0 channel 0.
SCSI host 0 channel 0 reset (pid 1) timed out - trying harder
SCSI bus is being reset for host 0 channel 0.
hdc: irq timeout: status=0xd0 { Busy }
hdc: ATAPI reset complete
hdc: irq timeout: status=0xc0 { Busy }
hdc: ATAPI reset complete
hdc: irq timeout: status=0xc0 { Busy }
scsi : aborting command due to timeout : pid 2, scsi0, channel 0, id 0, lun 0 0x12 00 00 00 ff 00 
SCSI host 0 abort (pid 2) timed out - resetting
SCSI bus is being reset for host 0 channel 0.
SCSI host 0 channel 0 reset (pid 2) timed out - trying harder
SCSI bus is being reset for host 0 channel 0.
hdc: irq timeout: status=0xd0 { Busy }
hdc: ATAPI reset complete
hdc: irq timeout: status=0xc0 { Busy }
hdc: ATAPI reset complete
hdc: irq timeout: status=0xc0 { Busy }


while for a good boot with 2.4.23-pre3 we have
Uniform Multi-Platform E-IDE driver Revision: 7.00beta4-2.4
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
ICH3M: IDE controller at PCI slot 00:1f.1
PCI: Enabling device 00:1f.1 (0005 -> 0007)
ICH3M: chipset revision 2
ICH3M: not 100% native mode: will probe irqs later
    ide0: BM-DMA at 0x1860-0x1867, BIOS settings: hda:DMA, hdb:pio
    ide1: BM-DMA at 0x1868-0x186f, BIOS settings: hdc:DMA, hdd:pio
hda: TOSHIBA MK4021GAS, ATA DISK drive
blk: queue c039b280, I/O limit 4095Mb (mask 0xffffffff)
hdc: QSI CD-RW/DVD-ROM SBW-242, ATAPI CD/DVD-ROM drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
ide1 at 0x170-0x177,0x376 on irq 15
hda: attached ide-disk driver.
hda: host protected area => 1
hda: 78140160 sectors (40008 MB), CHS=4864/255/63, UDMA(100)
hdc: attached ide-scsi driver.
Partition check:
 hda: hda1 hda2 hda3 hda4
SCSI subsystem driver Revision: 1.00
scsi0 : SCSI host adapter emulation for IDE ATAPI devices
  Vendor: QSI       Model: CDRW/DVD SBW-242  Rev: UX08
  Type:   CD-ROM                             ANSI SCSI revision: 02
Attached scsi CD-ROM sr0 at scsi0, channel 0, id 0, lun 0
sr0: scsi3-mmc drive: 4x/24x writer cd/rw xa/form2 cdda tray
Uniform CD-ROM driver Revision: 3.12


I only have patched in cpufreq, nothing else, and I am running
debian/sid.

Best wishes

Norbert

-------------------------------------------------------------------------------
Norbert Preining <preining AT logic DOT at>         Technische Universität Wien
gpg DSA: 0x09C5B094      fp: 14DF 2E6C 0307 BE6D AD76  A9C0 D2BF 4AA3 09C5 B094
-------------------------------------------------------------------------------
QUABBS (pl.n.)
The substances which emerge when you squeeze a blackhead.
			--- Douglas Adams, The Meaning of Liff

Re: 2.4.23-pre4 ide-scsi irq timeout

[Trond Myklebust]

>>>>> " " == Norbert Preining <preining@logic.at> writes:

     > I only have patched in cpufreq, nothing else, and I am running
     > debian/sid.

I saw the same thing, and have narrowed it down to a possible compiler
bug. Just drop gcc-3.3, and all will be well.

Cheers,
  Trond

Re: 2.4.23-pre4 ide-scsi irq timeout

[Norbert Preining]

On Son, 14 Sep 2003, Trond Myklebust wrote:
>      > I only have patched in cpufreq, nothing else, and I am running
>      > debian/sid.
> 
> I saw the same thing, and have narrowed it down to a possible compiler
> bug. Just drop gcc-3.3, and all will be well.

Hmmm, that sounds very strange, since I used the same gcc for the
previous kernels (pre3 and before), too!?

Best wishes

Norbert

-------------------------------------------------------------------------------
Norbert Preining <preining AT logic DOT at>         Technische Universität Wien
gpg DSA: 0x09C5B094      fp: 14DF 2E6C 0307 BE6D AD76  A9C0 D2BF 4AA3 09C5 B094
-------------------------------------------------------------------------------
FARNHAM (n.)
The feeling you get about four o'clock in the afternoon when you
haven't got enough done.
			--- Douglas Adams, The Meaning of Liff

Re: 2.4.23-pre4 ide-scsi irq timeout

[Trond Myklebust]

>>>>> " " == Norbert Preining <preining@logic.at> writes:

     > Hmmm, that sounds very strange, since I used the same gcc for
     > the previous kernels (pre3 and before), too!?

If it's gcc-3.3.2-0pre3 (the current most recent gcc in sid) then I
spent most of Saturday compiling various kernels (including a 2.4.22
that used to work). Every kernel I compiled with that particular
revision showed the ide-scsi irq timeout bug on bootup. Furthermore,
/dev/hdc was unreadable...

Dropping back to gcc-3.2 and the same kernel, same config,
same... works fine (including 2.4.23-pre4).

Now it may indeed be that the problem is a missing 'barrier()' or
something like that in the ide code, but I haven't got the time to try
to track it down...

Cheers,
  Trond

Re: 2.4.23-pre4 ide-scsi irq timeout

[Norbert Preining]

On Mon, 15 Sep 2003, Trond Myklebust wrote:
>      > Hmmm, that sounds very strange, since I used the same gcc for
>      > the previous kernels (pre3 and before), too!?
> 
> If it's gcc-3.3.2-0pre3 (the current most recent gcc in sid) then I

I have here 
[~] gcc --version
gcc (GCC) 3.3.2 20030908 (Debian prerelease)

and my currently running kernel has been compiled with:
Linux version 2.4.23-pre3 (root@gandalf) (gcc-Version 3.3.2 20030908 (Debian prerelease)) #3 Mit Sep 10 08:55:40 CEST 2003

so this *cannot* be the problem! It is the SAME compiler I am running at
the moment.

I am definitely not the opinion that this is the problem, sorry.

Best wishes

Norbert

-------------------------------------------------------------------------------
Norbert Preining <preining AT logic DOT at>         Technische Universität Wien
gpg DSA: 0x09C5B094      fp: 14DF 2E6C 0307 BE6D AD76  A9C0 D2BF 4AA3 09C5 B094
-------------------------------------------------------------------------------
SPROSTON GREEN (n.)
The violent colour of one of Nigel Rees's jackets, worn when he thinks
he's being elegant.
			--- Douglas Adams, The Meaning of Liff

Re: 2.4.23-pre4 ide-scsi irq timeout

[Trond Myklebust]

>>>>> " " == Norbert Preining <preining@logic.at> writes:

     > I am definitely not the opinion that this is the problem,
     > sorry.

Then I can only suggest that you run through the bitkeeper repository
(see http://linux.bkbits.net/linux-2.4) and do a binary search
applying the patches between 2.4.23-pre3 and 2.4.23-pre4 until you
find the guilty party.

Should be fairly easy, as there only appears to be one SCSI related
changeset (which only adds one line to a blacklist) and no IDE or
IRQ-related changes at all...

Cheers,
  Trond

Re: 2.4.23-pre4 ide-scsi irq timeout

[Edward Macfarlane Smith]

On Saturday 13 September 2003 23:01, Norbert Preining wrote:
> Hi!
>
> I have been using 2.4.23-pre3 without any problems, but when I booted
> 2.4.23-pre4 ide-scsi (bound to hdc) could not initialize the cdrom
> drive.
>
> Best wishes
>
> Norbert
>

I saw something very similar in 2.4.23-pre2, couldn't swear it was exactly the 
same, but definitely very similar. The very odd thing was that it only 
occured the first time I booted the kernel (at which pointed I rebooted to 
the old one and rebuilt it, but then forgot to copy the new image into /boot 
and rebooted). The next time it booted fine and since then (20+ boots?) it 
seems to have worked fine every time. I've got ide-scsi on hdc as well and 
had applied the cpufreq patch too, wonder if thats important. Not had a 
chance to try 2.4.23-pre4 yet, I've been away. I'm using gcc version 3.3 
20030226 (prerelease) (SuSE Linux).
Regards,
Edward