ide_dmaproc

ide_dmaproc

[Henning Loeser]

Hi
encouraged by Simon success with his Promise IDE card I went for
an other try with my AEC6260M ide card.
I patched a standart 2.2.15 kernel source with Andre's ide-patches (they
wouldn't install on Pauls rsynced stable tree) and after that applied
Michels pci-patches (the new ones), took a deep breath and rebooted, but
the error I allready told about (the AEC6260M message) prevailed :-(
I switched the cards around to find out whether it's the pci-slot ... no
success.
I keep getting errormassages stateing:
>ide_dmaproc: chipset supported ide_dma_lostirq func only: 11

the card and driver where detected and installed during boot :
>ide0 at 0xf2000840-0xf2000847,0x000 on irq 23

Any ideas on where I should start looking for the error?
Anyone some spare time to help me with this ??? :-(

Henning
--------------------------------------------------------------------------
Henning Loeser					      Fachbereich Physik
email: henning.loeser@physik.uni-marburg.de	      AG Oberflaechen
						      Philipps-Uni Marburg


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: ide_dmaproc

[Michel Lanners]

Hi there,

On  25 May, this message from Henning Loeser echoed through cyberspace:
> encouraged by Simon success with his Promise IDE card I went for
> an other try with my AEC6260M ide card.

Welcome to the club ;-)

> I patched a standart 2.2.15 kernel source with Andre's ide-patches (they
> wouldn't install on Pauls rsynced stable tree)

Yeah, I haven't bothered to rsync the stable tree, as kernel.org has
always worked fine for me....

> I keep getting errormassages stateing:
>>ide_dmaproc: chipset supported ide_dma_lostirq func only: 11

Could you provide the complete detection messages from dmesg, and a full
line of error meessage? You might also want to contact Andre about
this...

Michel

-------------------------------------------------------------------------
Michel Lanners                 |  " Read Philosophy.  Study Art.
23, Rue Paul Henkes            |    Ask Questions.  Make Mistakes.
L-1710 Luxembourg              |
email   mlan@cpu.lu            |
http://www.cpu.lu/~mlan        |                     Learn Always. "


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: ide_dmaproc

[Henning Loeser]

Hi,

On Thu, 25 May 2000, Michel Lanners wrote:
> Welcome to the club ;-)
Thanks ... It's always nice to be welcome :-)

>
> > I patched a standart 2.2.15 kernel source with Andre's ide-patches (they
> > wouldn't install on Pauls rsynced stable tree)
>
> Yeah, I haven't bothered to rsync the stable tree, as kernel.org has
> always worked fine for me....

Oh I didn't even try yours, Andre's allready wouldn't work ...

> Could you provide the complete detection messages from dmesg, and a full
> line of error meessage? You might also want to contact Andre about
> this...
I allready contacted Andre, the driver works swell on x86 Boxes ... :-(,
he said he couldn't help me any futher in the lack of a PPC to check
what's wrong.

The detection message from dmesg is:

Uniform Multi-Platform E-IDE driver Revision: 6.30
ide: Assuming 33MHz system bus speed for PIO modes; override with
idebus=xx
AEC6260R: IDE controller on PCI bus 00 dev 68
AEC6260R: not 100% native mode: will probe irqs later
AEC6260R: ROM enabled at 0x80810000
AEC6260: reg49h=0x00 ATA-66 Cable Port0
    ide0: BM-DMA at 0xf2000800-0xf2000807, BIOS settings: hda:pio, hdb:pio
AEC6260: reg49h=0x00 ATA-66 Cable Port1
    ide1: BM-DMA at 0xf2000808-0xf200080f, BIOS settings: hdc:pio, hdd:pio
hda: IBM-DJNA-352030, ATA DISK drive
ide0 at 0xf2000840-0xf2000847,0x000 on irq 23
hda: IBM-DJNA-352030, 19470MB w/1966kB Cache, CHS=39560/16/63, UDMA(66)

The first ide0 and ide1 detection look a little strange to me, as there is
only one drive connected to ide0 and the adress looks wrong aswell. The
second ide0 statement appears to be correct with 0xf2000840-847 ...

The Error messages look like:
 hda:hda: lost interrupt
hda: lost interrupt
ide_dmaproc: chipset supported ide_dma_lostirq func only: 11
hda: lost interrupt
ide_dmaproc: chipset supported ide_dma_lostirq func only: 11
hda: lost interrupt
 hda1ide_dmaproc: chipset supported ide_dma_lostirq func only: 11
[snip]
 hda6 hda7ide_dmaproc: chipset supported ide_dma_lostirq func only: 11
hda: lost interrupt
 hda8 hda9

For all nine Partitions on the drive. If I start pdisk /dev/hda I can read
the partition map (after a couple of seconds) and my /var/log/messages
file reads:
May 26 10:17:28 ma1304 kernel: ide_dmaproc: chipset supported
ide_dma_lostirq func only: 11
May 26 10:17:28 ma1304 kernel: hda: lost interrupt
May 26 10:17:38 ma1304 kernel: ide_dmaproc: chipset supported
ide_dma_lostirq func only: 11
May 26 10:17:38 ma1304 kernel: hda: lost interrupt

Just for completeness, the lspci -vvx reads:

00:0b.0 Host bridge: Apple Computer Inc. Bandit PowerPC host bridge (rev
03)

00:0d.0 SCSI storage controller: Artop Electronic Corp: Unknown device
0007 (rev 01)
        Subsystem: Unknown device 1191:0007
        Control: I/O+ Mem- BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr- Stepping- SERR- FastB2B-
        Status: Cap+ 66Mhz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
        Latency: 6 min, 6 max, 128 set
        Interrupt: pin A routed to IRQ 23
        Region 0: I/O ports at f2000840
        Region 1: I/O ports at f2000830
        Region 2: I/O ports at f2000820
        Region 3: I/O ports at f2000810
        Region 4: I/O ports at f2000800
        Capabilities: [58] Power Management version 1
                Flags: PMEClk- AuxPwr- DSI- D1- D2- PME-
                Status: D0 PME-Enable- DSel=0 DScale=0 PME-
00: 91 11 07 00 05 00 90 02 01 00 00 01 00 80 00 00
10: 41 08 00 00 31 08 00 00 21 08 00 00 11 08 00 00
20: 01 08 00 00 00 00 00 00 00 00 00 00 91 11 07 00
30: 01 00 81 80 58 00 00 00 00 00 00 00 00 01 06 06

00:0f.0 VGA compatible controller: 3Dfx Interactive, Inc.: Unknown device
0005 (rev 01)

00:10.0 Class ff00: Apple Computer Inc. Grand Central I/O (rev 02)
01:0b.0 Non-VGA unclassified device: Apple Computer Inc. Control Video
01:0d.0 Class ff00: Apple Computer Inc. PlanB Video-In (rev 01)

Do you know what this is supposed to mean? Everything seems to work,
untill the kernel actually tries to access the drive. I tried ide0=reset
in the kernel options aswell as ide0=autotune. Didn't help.

Henning

--------------------------------------------------------------------------
Henning Loeser					      Fachbereich Physik
email: henning.loeser@physik.uni-marburg.de	      AG Oberflaechen
						      Philipps-Uni Marburg


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: ide_dmaproc

[Michel Lanners]

Hi there,

On  26 May, this message from Henning Loeser echoed through cyberspace:
>> Could you provide the complete detection messages from dmesg, and a full
>> line of error meessage? You might also want to contact Andre about
>> this...
> I allready contacted Andre, the driver works swell on x86 Boxes ... :-(,
> he said he couldn't help me any futher in the lack of a PPC to check
> what's wrong.

Well, he might not be able to check himself, but might well be able to
tell you under what circumstances those error messages do appear...

> The detection message from dmesg is:
>
> Uniform Multi-Platform E-IDE driver Revision: 6.30
> ide: Assuming 33MHz system bus speed for PIO modes; override with
> idebus=xx
> AEC6260R: IDE controller on PCI bus 00 dev 68
> AEC6260R: not 100% native mode: will probe irqs later
> AEC6260R: ROM enabled at 0x80810000
> AEC6260: reg49h=0x00 ATA-66 Cable Port0
>     ide0: BM-DMA at 0xf2000800-0xf2000807, BIOS settings: hda:pio, hdb:pio
> AEC6260: reg49h=0x00 ATA-66 Cable Port1
>     ide1: BM-DMA at 0xf2000808-0xf200080f, BIOS settings: hdc:pio, hdd:pio
> hda: IBM-DJNA-352030, ATA DISK drive
> ide0 at 0xf2000840-0xf2000847,0x000 on irq 23
> hda: IBM-DJNA-352030, 19470MB w/1966kB Cache, CHS=39560/16/63, UDMA(66)
>
> The first ide0 and ide1 detection look a little strange to me, as there is
> only one drive connected to ide0 and the adress looks wrong aswell. The
> second ide0 statement appears to be correct with 0xf2000840-847 ...

No, that's actually OK. Those are separate memory regions. I'm not an
IDE expert, so I can't explain what they are for...

> The Error messages look like:
>  hda:hda: lost interrupt
> hda: lost interrupt

Ohhhh, the lost interrupt problem... That was reported quite often
recently. Anyone know what the fix is/was? It never ocured on my system,
so I didn't bother... You might want to look through the list archives
for the solution. Although, I wonder whether it wasn't specifically on
the G4 systems when booted with the wrong booter, i.e. not relevant
here...

> ide_dmaproc: chipset supported ide_dma_lostirq func only: 11
> hda: lost interrupt
> ide_dmaproc: chipset supported ide_dma_lostirq func only: 11
> hda: lost interrupt
[snip]
> Just for completeness, the lspci -vvx reads:
>
> 00:0b.0 Host bridge: Apple Computer Inc. Bandit PowerPC host bridge (rev
> 03)
>
> 00:0d.0 SCSI storage controller: Artop Electronic Corp: Unknown device
> 0007 (rev 01)

SCSI controller? I suppose they masquerade so the regular PCI BIOS and
M$ OS'es leave the controller alone..

>         Subsystem: Unknown device 1191:0007
>         Control: I/O+ Mem- BusMaster+ SpecCycle- MemWINV- VGASnoop-
> ParErr- Stepping- SERR- FastB2B-
>         Status: Cap+ 66Mhz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
> <TAbort- <MAbort- >SERR- <PERR-
>         Latency: 6 min, 6 max, 128 set
>         Interrupt: pin A routed to IRQ 23
>         Region 0: I/O ports at f2000840
>         Region 1: I/O ports at f2000830
>         Region 2: I/O ports at f2000820
>         Region 3: I/O ports at f2000810
>         Region 4: I/O ports at f2000800
>         Capabilities: [58] Power Management version 1
>                 Flags: PMEClk- AuxPwr- DSI- D1- D2- PME-
>                 Status: D0 PME-Enable- DSel=0 DScale=0 PME-
> 00: 91 11 07 00 05 00 90 02 01 00 00 01 00 80 00 00
> 10: 41 08 00 00 31 08 00 00 21 08 00 00 11 08 00 00
> 20: 01 08 00 00 00 00 00 00 00 00 00 00 91 11 07 00
> 30: 01 00 81 80 58 00 00 00 00 00 00 00 00 01 06 06

Except the funky latency (not really important, I think), everything
looks ok....

Michel

-------------------------------------------------------------------------
Michel Lanners                 |  " Read Philosophy.  Study Art.
23, Rue Paul Henkes            |    Ask Questions.  Make Mistakes.
L-1710 Luxembourg              |
email   mlan@cpu.lu            |
http://www.cpu.lu/~mlan        |                     Learn Always. "


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: ide_dmaproc - YaBoot

[Henning Loeser]

On Fri, 26 May 2000, Michel Lanners wrote:
> Ohhhh, the lost interrupt problem... That was reported quite often
> recently. Anyone know what the fix is/was? It never ocured on my system,
> so I didn't bother... You might want to look through the list archives
> for the solution. Although, I wonder whether it wasn't specifically on
> the G4 systems when booted with the wrong booter, i.e. not relevant
> here...
What I found seems to have been related to not using yaboot. As I have a
'old 7500' with MacOS8.1 I thought BootX would be just fine. But maybe
not. Can I use yaboot on my machine? The WebPages are not very clear on
that part. BenH's site states that yaboot is onbly for 'NewWorld' macs,
but Eric's FAQ states that yaboot is for 'NewWorld' macs, but users of
earlier machines should consider switching if they encounter odd problems
...

Well I do encounter strange Problems ...

So should I go through the hassel of installing yaboot?

Henning

--------------------------------------------------------------------------
Henning Loeser					      Fachbereich Physik
email: henning.loeser@physik.uni-marburg.de	      AG Oberflaechen
						      Philipps-Uni Marburg


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: ide_dmaproc - YaBoot

[Benjamin Herrenschmidt]

On Sun, May 28, 2000, Henning Loeser <Henning.Loeser@Physik.Uni-
Marburg.DE> wrote:

>What I found seems to have been related to not using yaboot. As I have a
>'old 7500' with MacOS8.1 I thought BootX would be just fine. But maybe
>not. Can I use yaboot on my machine? The WebPages are not very clear on
>that part. BenH's site states that yaboot is onbly for 'NewWorld' macs,
>but Eric's FAQ states that yaboot is for 'NewWorld' macs, but users of
>earlier machines should consider switching if they encounter odd problems
>...
>
>Well I do encounter strange Problems ...
>
>So should I go through the hassel of installing yaboot?

Currently, yaboot will only work on "newworld" type machines. You won't
be able to use it on a 7500. Can you veryfy that bus master is correctly
enabled for your card in the PCI command register ?)

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: ide_dmaproc - YaBoot

[Ethan Benson]

[-- Attachment #1: Type: text/plain, Size: 2072 bytes --]

On Sun, May 28, 2000 at 03:20:30PM +0200, Henning Loeser wrote:

> What I found seems to have been related to not using yaboot. As I have a
> 'old 7500' with MacOS8.1 I thought BootX would be just fine. But maybe
> not. Can I use yaboot on my machine? The WebPages are not very clear on
> that part. BenH's site states that yaboot is onbly for 'NewWorld' macs,
> but Eric's FAQ states that yaboot is for 'NewWorld' macs, but users of
> earlier machines should consider switching if they encounter odd problems
> ...

yaboot will not work on oldworld macs, you need to use quik instead.

> Well I do encounter strange Problems ...

seems typical of bootx lately...

> So should I go through the hassel of installing yaboot?

no because it would be a waste of time since it won't work ;-)  you
should go to the hassle of installing quik instead. (hint use debian
quik it actually works)

before anyone starts bitching about how quik is so horrible to setup,
let me say this, yaboot would be no different, they are both OF boot
loaders the only difference is yaboot is essentially a second stage
loader only, quik has a first stage loader which eliminates the need
for a bootstrap partition.  The problems people have with quik is the
following:

most [oldworld] macs don't have OF video drivers, this makes
interfacing with OF someone annoying.

oldworld macs have `the BrokenFirmware from hell' not even MS has made
anything as broken as this firmware ;-)

both of these problems exist just as well with yaboot as they do with
quik.   the only reason yaboot is less painful on newworld is due to
the one-quarter-way usable OF, and the OF video driver. 

it might be possible to make yaboot work on an oldworld, but why? all
you would end up with is quik second stage. (which is basically all
yaboot is, quik second stage with bits of silo and other assorted
stuff to work on newworlds)

anyway my point is if you want OF booting on an oldworld you want
quik, not yaboot. 

-- 
Ethan Benson
http://www.alaska.net/~erbenson/

[-- Attachment #2: Type: application/pgp-signature, Size: 240 bytes --]

Re: ide_dmaproc - YaBoot

[Henning Loeser]

On Sun, 28 May 2000, Benjamin Herrenschmidt wrote:
> Currently, yaboot will only work on "newworld" type machines. You won't
> be able to use it on a 7500. Can you veryfy that bus master is correctly
> enabled for your card in the PCI command register ?)
How do I do that?
lspci -vvx says :
[snip]
Control: I/O+ Mem- BusMaster+ [snip]
Is that it? If I can believe that I would say bus master is enabled, but
is it enabeled correctly ?????

--
Hi! I'm a .signature virus! Copy me into your ~/.signature to help me
spread!
--------------------------------------------------------------------------
Henning Loeser					      Fachbereich Physik
email: henning.loeser@physik.uni-marburg.de	      AG Oberflaechen
						      Philipps-Uni Marburg


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: ide_dmaproc - YaBoot

[Henning Loeser]

On Sun, 28 May 2000, Ethan Benson wrote:
> yaboot will not work on oldworld macs, you need to use quik instead.
Ok
>
> > Well I do encounter strange Problems ...
>
> seems typical of bootx lately...
>
> > So should I go through the hassel of installing yaboot?
>
> no because it would be a waste of time since it won't work ;-)  you
> should go to the hassle of installing quik instead. (hint use debian
> quik it actually works)

Well beeing lazy I would only do that if it does promise some progress in
getting the card to work. I'll see what I can find out about quik, but
maybe I have to give it a try to find out ... :-(

Something else seems to be fishy aswell maybe the driver (???) or
whatever. I went ahead and tried a new 2.4.0test1-ac1 kernel and all of
the sudden the driver doesn't even see the drive any more ... :
AEC6260R: IDE controller on PCI bus 00 dev 68
AEC6260R: chipset revision 1
AEC6260R: not 100% native mode: will probe irqs later
AEC6260R: ROM enabled at 0x80810000
AEC6260: reg49h=0x00 ATA-66 Cable Port0
    ide0: BM-DMA at 0x0800-0x0807, BIOS settings: hda:pio, hdb:pio
AEC6260: reg49h=0x00 ATA-66 Cable Port1
    ide1: BM-DMA at 0x0808-0x080f, BIOS settings: hdc:pio, hdd:pio
hda: ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ, ATA DISK drive
ide0 at 0x840-0x847,0x000 on irq 23
hda: 187353000 sectors (95925 MB) w/11565KiB Cache, CHS=23130/90/90
hda: INVALID GEOMETRY: 90 PHYSICAL HEADS?

and then doesn't even enable the drive. I emailed that to Andre Hedrick to
see whether he knows whats going on.

Henning

--------------------------------------------------------------------------
Henning Loeser					      Fachbereich Physik
email: henning.loeser@physik.uni-marburg.de	      AG Oberflaechen
						      Philipps-Uni Marburg


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: ide_dmaproc - YaBoot

[Ethan Benson]

[-- Attachment #1: Type: text/plain, Size: 2326 bytes --]

On Sun, May 28, 2000 at 04:35:13PM +0200, Henning Loeser wrote:
> 
> Well beeing lazy I would only do that if it does promise some progress in
> getting the card to work. I'll see what I can find out about quik, but
> maybe I have to give it a try to find out ... :-(

quik is annoyingly not well documented, the only docs you can find are
very old and unmaintained...  

the old docs are still somewhat accurate, only things i ignore is the
thing about putting kernel arguments and such in the boot-file OF
variable, just write a correct quik.conf and you don't need to fiddle
with that.  i helped someone over irc setup quik successfully on a
7200 (though it won't boot 2.2.15 only 2.2.14, still haven't figured
that out...) but someone else who tried the same thing on another 7200
(with same quik sources) just gets DEFAULT CATCH! 

i can't say i have ever setup quik personally (just through irc ;-)  i
don't have any oldworld macs (except a powerbook 540c that is quite
opposed to me putting any sort of *nix on it) 

> Something else seems to be fishy aswell maybe the driver (???) or
> whatever. I went ahead and tried a new 2.4.0test1-ac1 kernel and all of
> the sudden the driver doesn't even see the drive any more ... :
> AEC6260R: IDE controller on PCI bus 00 dev 68
> AEC6260R: chipset revision 1
> AEC6260R: not 100% native mode: will probe irqs later
> AEC6260R: ROM enabled at 0x80810000
> AEC6260: reg49h=0x00 ATA-66 Cable Port0
>     ide0: BM-DMA at 0x0800-0x0807, BIOS settings: hda:pio, hdb:pio
> AEC6260: reg49h=0x00 ATA-66 Cable Port1
>     ide1: BM-DMA at 0x0808-0x080f, BIOS settings: hdc:pio, hdd:pio
> hda: ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ, ATA DISK drive
  
what the devil is all that ZZZZZZZZ crap???

> ide0 at 0x840-0x847,0x000 on irq 23
> hda: 187353000 sectors (95925 MB) w/11565KiB Cache, CHS=23130/90/90
> hda: INVALID GEOMETRY: 90 PHYSICAL HEADS?

hmm 95925MB, wow thats a nice disk you got there ;-)  

there is obviously something seriously bogus going on there, its
coming up with total crap for specs on the drive...

> and then doesn't even enable the drive. I emailed that to Andre Hedrick to
> see whether he knows whats going on.

ill bet it don't enable the drive ;-)  

-- 
Ethan Benson
http://www.alaska.net/~erbenson/

[-- Attachment #2: Type: application/pgp-signature, Size: 240 bytes --]

Re: ide_dmaproc - YaBoot

[Michel Lanners]

Hi there,

On  28 May, this message from Henning Loeser echoed through cyberspace:
> Something else seems to be fishy aswell maybe the driver (???) or
> whatever. I went ahead and tried a new 2.4.0test1-ac1 kernel and all of
> the sudden the driver doesn't even see the drive any more ... :
> AEC6260R: IDE controller on PCI bus 00 dev 68
> AEC6260R: chipset revision 1
> AEC6260R: not 100% native mode: will probe irqs later
> AEC6260R: ROM enabled at 0x80810000
> AEC6260: reg49h=0x00 ATA-66 Cable Port0
>     ide0: BM-DMA at 0x0800-0x0807, BIOS settings: hda:pio, hdb:pio
> AEC6260: reg49h=0x00 ATA-66 Cable Port1
>     ide1: BM-DMA at 0x0808-0x080f, BIOS settings: hdc:pio, hdd:pio
> hda: ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ, ATA DISK drive
> ide0 at 0x840-0x847,0x000 on irq 23
          ^^^^^^^^^^^
You're not using my PCI patches. I'm amazed it finds anything at all on
your card; but anyway, get my latest 2.3.99 patches (2.3.99-pre9, that
is), and apply them. From the above, it seems only part of the
communication of the card with the driver works ok, whereas the rest
reads schrott.

Michel

-------------------------------------------------------------------------
Michel Lanners                 |  " Read Philosophy.  Study Art.
23, Rue Paul Henkes            |    Ask Questions.  Make Mistakes.
L-1710 Luxembourg              |
email   mlan@cpu.lu            |
http://www.cpu.lu/~mlan        |                     Learn Always. "


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: ide_dmaproc - YaBoot

[Henning Loeser]

On Sun, 28 May 2000, Michel Lanners wrote:
> Hi there,
>
> On  28 May, this message from Henning Loeser echoed through cyberspace:
> > Something else seems to be fishy aswell maybe the driver (???) or
> > whatever. I went ahead and tried a new 2.4.0test1-ac1 kernel and all of
> > the sudden the driver doesn't even see the drive any more ... :
> > AEC6260R: IDE controller on PCI bus 00 dev 68
> > AEC6260R: chipset revision 1
> > AEC6260R: not 100% native mode: will probe irqs later
> > AEC6260R: ROM enabled at 0x80810000
> > AEC6260: reg49h=0x00 ATA-66 Cable Port0
> >     ide0: BM-DMA at 0x0800-0x0807, BIOS settings: hda:pio, hdb:pio
> > AEC6260: reg49h=0x00 ATA-66 Cable Port1
> >     ide1: BM-DMA at 0x0808-0x080f, BIOS settings: hdc:pio, hdd:pio
> > hda: ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ, ATA DISK drive
> > ide0 at 0x840-0x847,0x000 on irq 23
>           ^^^^^^^^^^^
> You're not using my PCI patches. I'm amazed it finds anything at all on
> your card; but anyway, get my latest 2.3.99 patches (2.3.99-pre9, that
> is), and apply them. From the above, it seems only part of the
> communication of the card with the driver works ok, whereas the rest
> reads schrott.

Well this is with a 2.4.0test1 kernel, I wasn't shure whether your patches
would install on that. I'll try to apply your 2.3.99-pre9 patches onto
that source. But that has to wait untill tomorrow ;-)

Henning
--------------------------------------------------------------------------
Henning Loeser					      Fachbereich Physik
email: henning.loeser@physik.uni-marburg.de	      AG Oberflaechen
						      Philipps-Uni Marburg


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: ide_dmaproc - YaBoot

[Henning Loeser]

Hi,
ok I applied your 2.3.99-pre9 patches to 2.4.0.test1-ac4 sources and at
least the drive is recognized again. dmesg reads:
ide: Assuming 33MHz system bus speed for PIO modes; override with
idebus=xx
AEC6260R: IDE controller on PCI bus 00 dev 68
AEC6260R: chipset revision 1
AEC6260R: not 100% native mode: will probe irqs later
AEC6260R: ROM enabled at 0x80810000
AEC6260: reg49h=0x00 ATA-66 Cable Port0
    ide0: BM-DMA at 0xf2000800-0xf2000807, BIOS settings: hda:pio, hdb:pio
AEC6260: reg49h=0x00 ATA-66 Cable Port1
    ide1: BM-DMA at 0xf2000808-0xf200080f, BIOS settings: hdc:pio, hdd:pio
hda: IBM-DJNA-352030, ATA DISK drive
ide0 at 0xf2000840-0xf2000847,0x000 on irq 23
hda: 39876480 sectors (20417 MB) w/1966KiB Cache, CHS=39560/16/63,
UDMA(66)

but I'm back at the lost interrupt :-(

Partition check:
 hda:hda: lost interrupt
hda: lost interrupt
ide_dmaproc: chipset supported ide_dma_lostirq func only: 13
hda: lost interrupt
ide_dmaproc: chipset supported ide_dma_lostirq func only: 13
hda: lost interrupt
 hda1ide_dmaproc: chipset supported ide_dma_lostirq func only: 13
hda: lost interrupt
 hda2 hda3ide_dmaproc: chipset supported ide_dma_lostirq func only: 13
hda: lost interrupt
 hda4 hda5ide_dmaproc: chipset supported ide_dma_lostirq func only: 13
hda: lost interrupt
 hda6 hda7ide_dmaproc: chipset supported ide_dma_lostirq func only: 13
hda: lost interrupt
 hda8 hda9

I was wondering after Ben's commend if OF sets the bummaster correct for
the card. Or if I should do something like:
      pci_read_config_word( pdev, PCI_COMMAND, &cmd );
      cmd |= (PCI_COMMAND_MEMORY | PCI_COMMAND_IO);
      pci_write_config_word( pdev, PCI_COMMAND, cmd );

How do I verify the busmaster operation is set up cerrectly ?

Henning

--------------------------------------------------------------------------
Henning Loeser					      Fachbereich Physik
email: henning.loeser@physik.uni-marburg.de	      AG Oberflaechen
						      Philipps-Uni Marburg


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/