OSS cannot find audio card.

[Iggo]

Hi, I have an intel High definition Audio controller (on a pavillion
dv2268ea) which cannot be seen by oss driver.
First, lspci give mes 0x8086 0x27d8 (also in linux) but the audio card
seems to be a conexant...
Why? (I expected to be an intel one, reading 0x8068).
Looking herre and there in OSS source seems that Intel HDA should be
supported... What I'm missing?

Are there some documents about the OSS port? (I would like to be able
to get some parts (drivers) of new oss releases and merge them on
minix3 oss...)

Best regads Iggo.

[pieter]

Hi Iggo,

On Nov 29, 1:10 am, Iggo <gepdi...@gmail.com> wrote:
> Hi, I have an intel  High definition Audio controller (on a pavillion
> dv2268ea) which cannot be seen by oss driver.
> First, lspci give mes 0x8086 0x27d8 (also in linux) but the audio card
> seems to be a conexant...
> Why? (I expected to be an intel one, reading 0x8068).
> Looking herre and there in OSS source seems that Intel HDA should be
> supported... What I'm missing?

I don't really know. I didn't have the chance to test an hdaudio card.
I think hdaudio cards give a lot of problems on OSS, but I'm not sure.
Reading from the source of the oss_hdaudio module I see that 0x8086,
0x27d8 is indeed supported. In the file hdaudio_codecids.h on line 900
there are some codec vendor/device ids for Conexant. Perhaps this
messes up the initialization of your card. I think the only way to
find out for sure is to add some debugging prints to the source file
or so.
I could imagine that something as this happens:
oss_hdaudio_attach() // oss_hdaudio.c
init_HDA() // oss_hdaudio.c
hdaudio_mixer_create() // hdaudio_codec.c
attach_codec() // hdaudio_codec.c
// no right codec vendor/device id
// ncodecs == 0, returns NULL
// returns NULL
// returns 0
// returns 0, which is an error, nothing attached.

> Are there some documents about the OSS port? (I would like to be able
> to get some parts (drivers) of new oss releases and merge them on
> minix3 oss...)

Yes there are. You can find it at www.few.vu.nl/~hphijma . I don't
think drivers of new releases would work on this version of OSS. I
ported both OSS 4.0 and OSS 4.1 and I think there were huge
differences in for example the vmix part of OSS. Porting a new release
might just be easier. But perhaps I'm wrong.
If things in the document need clearification, please just let me
know.

Kind regards,

Pieter

[pieter]

Hi Iggo,

On Nov 29, 1:10 am, Iggo <gepdi...@gmail.com> wrote:

> Hi, I have an intel  High definition Audio controller (on a pavillion
> dv2268ea) which cannot be seen by oss driver.
> First, lspci give mes 0x8086 0x27d8 (also in linux) but the audio card
> seems to be a conexant...
> Why? (I expected to be an intel one, reading 0x8068).
> Looking herre and there in OSS source seems that Intel HDA should be
> supported... What I'm missing?

I don't really know. I didn't have the chance to test an hdaudio card.
I think hdaudio cards give a lot of problems on OSS, but I'm not sure.
Reading from the source of the oss_hdaudio module I see that 0x8086,
0x27d8 is indeed supported. In the file hdaudio_codecids.h on line 900
there are some codec vendor/device ids for Conexant. Perhaps this
messes up the initialization of your card. I think the only way to
find out for sure is to add some debugging prints to the source file
or so.
I could imagine that something as this happens:
oss_hdaudio_attach() // oss_hdaudio.c
init_HDA() // oss_hdaudio.c
hdaudio_mixer_create() // hdaudio_codec.c
attach_codec() // hdaudio_codec.c
// no right codec vendor/device id
// ncodecs == 0, returns NULL
// returns NULL
// returns 0
// returns 0, which is an error, nothing attached.

> Are there some documents about the OSS port? (I would like to be able
> to get some parts (drivers) of new oss releases and merge them on
> minix3 oss...)

[Iggo]

Thank you very much! I'll read it!

Maybe you could answer these questions.

If I add the vid and did in drivers.conf, OSS founds my card!
Even if it complains:
"pci_reserve3: BAR 1 is incomplete" - but it should be ok right?
the big problem is:
"VM: restricted call VM_MAP_PHYS from 109696 instead of 0x28"
"VM: pagefault SIGSEGV 109696 bad addr 0x7 err 0x4 nopage read"
where the 109696 number increments by 1 every restart..

With X qemu mplayer compiler utils, browser and (hopefully audio) I
could try minix as my main machine...
It could be very interesting..

Best regards, Iggo.

[pieter]

On Nov 30, 6:18 pm, Iggo <gepdi...@gmail.com> wrote:
> Maybe you could answer these questions.
>
> If I add the vid and did in drivers.conf, OSS founds my card!

Hm, this is strange. There should be the following in the
drivers.conf:

pci class 4/1 # Multimedia / Audio device

This was enough for all the cards I tested.

> Even if it complains:
> "pci_reserve3: BAR 1 is incomplete" - but it should be ok right?

I see this all the time. It doesn't give any issues as far as I know.

> the big problem is:
> "VM: restricted call VM_MAP_PHYS from 109696 instead of 0x28"
> "VM: pagefault SIGSEGV 109696 bad addr 0x7 err 0x4 nopage read"
> where the 109696 number increments by 1 every restart..

Ai, that is a big problem. I had many problems but never this kind...
Is there any way you could give me more information? For example, does
it happen in the oss_hdaudio module or in the minix modules? setup/
Minix/oss/build/osscore.c or kernel/OS/Minix/os_minix.c

Kind regards,

Pieter

[Ben Gras]

> > the big problem is:
> > "VM: restricted call VM_MAP_PHYS from 109696 instead of 0x28"
> > "VM: pagefault SIGSEGV 109696 bad addr 0x7 err 0x4 nopage read"
> > where the 109696 number increments by 1 every restart..
>
> Ai, that is a big problem. I had many problems but never this kind...
> Is there any way you could give me more information? For example, does
> it happen in the oss_hdaudio module or in the minix modules? setup/
> Minix/oss/build/osscore.c   or   kernel/OS/Minix/os_minix.c

That seems to be an error message from an "old" VM, in which only TTY
and MEMORY are allowed to map in physical memory. Did MAP_PHYS ever
work for any of the drivers you tested pieter? If so I wonder how :)

If the memory the driver wants to have mapped in is listed in the pci
configuration space, it should have permission to map it in without
further ado in the latest version (svn HEAD).

[Iggo]

As soon as I've little spare time (tomorrow night, I hope), I'll do
every test you need me to do.
There is some guide I've to follow?
I explain better. I will use the 3.1.5 cdrom image,(pci_debug=1),
install from the net OSS package,
modify the driver.conf launch soundon, and after that I'll try to know
(with some printf addictions) if osscore.c or os_minix.c generate the
problem.

That's right?

If there are more things I can do to help you, please, feel free to
ask.

To ben: I'm using kernel image of 3.1.5 cdrom and the rest compiled
from svn of about 1 weak ago.. unluckily new kernel prevents X to
recognize my Nvidia go7200 chipset :-(

Thank you. regards, Iggo.

[pieter]

> On Dec 2, 6:21 pm, Ben Gras <b...@few.vu.nl> wrote:
>
> > > > the big problem is:
> > > > "VM: restricted call VM_MAP_PHYS from 109696 instead of 0x28"
> > > > "VM: pagefault SIGSEGV 109696 bad addr 0x7 err 0x4 nopage read"
> > > > where the 109696 number increments by 1 every restart..
>
> > > Ai, that is a big problem. I had many problems but never this kind...
> > > Is there any way you could give me more information? For example, does
> > > it happen in the oss_hdaudio module or in the minix modules? setup/
> > > Minix/oss/build/osscore.c   or   kernel/OS/Minix/os_minix.c
>
> > That seems to be an error message from an "old" VM, in which only TTY
> > and MEMORY are allowed to map in physical memory. Did MAP_PHYS ever
> > work for any of the drivers you tested pieter? If so I wonder how :)

It did once... Me too :)

> > If the memory the driver wants to have mapped in is listed in the pci
> > configuration space, it should have permission to map it in without
> > further ado in the latest version (svn HEAD).

On Dec 2, 6:34 pm, Iggo <gepdi...@gmail.com> wrote:
> As soon as I've little spare time (tomorrow night, I hope), I'll do
> every test you need me to do.
> There is some guide I've to follow?
> I explain better. I will use the 3.1.5 cdrom image,(pci_debug=1),
> install from the net OSS package,
> modify the driver.conf launch soundon, and after that I'll try to know
> (with some printf addictions) if osscore.c or os_minix.c generate the
> problem.

I tested a memory mapped card and could reproduce your errors. I
switched to svn 5712 and got a new error message:
VM: unauthorized mapping of...
The file osscore.c turned out to have two bugs that are fixed now. Ben
promised to make a new package for OSS today.
With the patch my memory mapped card runs without a problem.

So, if you could test the new OSS on svn >= 5712 with your hda_intel
card, that would be great.

Kind regards,

Pieter

[Ben Gras]

Pieter's fix is applied and a new OSS package is available on the
server (i.e. through packman).

[Iggo]

Surely you are fast...! :-)

This evening I'll try and let you know!

Thank you very much!

[Iggo]

Ok, I did svn checkout, make world, installed OSS from packman from
Network. Rebooted with image=/boot/image/ and:

OSS: Bad CORB size
OSS: Cannot attach device
OSS: Detecting service 0x8086, 0x27d8 failed.

Oh, I should start with the no_dma (or similar..) option for the
at_wini.

(please keep in mind I'm a newbye so if you need some tests to be
done, the more you specify the better).

Thank you very much.

[pieter]

On Dec 3, 8:20 pm, Iggo <gepdi...@gmail.com> wrote:
> Ok,  I did svn checkout, make world, installed OSS from packman from
> Network. Rebooted with image=/boot/image/ and:
>
> OSS: Bad CORB size
> OSS: Cannot attach device
> OSS: Detecting service 0x8086, 0x27d8 failed.

Hi Iggo,

Ok, this is better, but it could still mean some problem with memory
mapping.
I'm interested in the following lines in kernel/drv/oss_hdaudio/
oss_hdaudio.c (from line 1816) :

pci_read_config_dword (osdev, PCI_MEM_BASE_ADDRESS_0, &devc-
>membar_addr);

devc->membar_addr &= ~7;

/* get virtual address */
devc->azbar =
(void *) MAP_PCI_MEM (devc->osdev, 0, devc->membar_addr, 16 *
1024);

What are the values of devc->membar_addr at any moment? Is this the
same as what is reported by pci_debug=1 at boot? What happens if you
comment devc->membar_addr &= ~7; ? What is the value of devc->azbar?

So please add for example this:

pci_read_config_dword (osdev, PCI_MEM_BASE_ADDRESS_0, &devc-
>membar_addr);
printf("devc->membar_addr (1): 0x%x\n", devc->membar_addr);

devc->membar_addr &= ~7;

printf("devc->membar_addr (2): 0x%x\n", devc->membar_addr);

/* get virtual address */
devc->azbar =
(void *) MAP_PCI_MEM (devc->osdev, 0, devc->membar_addr, 16 *
1024);
printf("devc->azbar: 0x%x\n", devc->azbar);

Also I'm interested in the value of tmp at line 1102 and 1120. And I'm
interested in where it goes wrong. Is it for the CORB values or the
RIRB values. The "Bad CORB size\n" string on line 1134 should probably
be: "Bad RIRB size\n". So if it fails on the CORB size, does it also
fail on the RIRB size? You can figure it out by commenting out the
CORB size part.

You can compile oss by just running sh build.minix in the source
directory.

Kind regards,

Pieter

[Iggo]

Tomorrow night I should have a little time... (sorry but I'm very busy
at moment... :-( )

I'll do any check that could help.

BTW, have you ever joined the minix chat? I read it often so it could
be a useful way to communicate in a more "interactive way"
I eventually could do tests for you in "real time", if you think this
could help more.

Thank you very much.

I really apreciate the detailed instructions, so there is less room
for mistakes by my side.

Best regards Iggo.

[Iggo]

Sorry for being late...

membar_addr 1:0x4
membar_addr 2:0x0
azbar:0xffffffff

After that VM:pagefault
I cannot see the two printf for tmp, maybe the pagefault comes before?

Sorry but very hard day at work,
tomorrow I'll do the tests with a little more "brain",
I'll also do some checks for CORB|RIRB...
and I'll post'em.

Thank you for the patience.

Best regards Iggo.

[pieter]

On Dec 4, 11:10 pm, Iggo <gepdi...@gmail.com> wrote:
> Sorry for being late...
>
> membar_addr 1:0x4
> membar_addr 2:0x0
> azbar:0xffffffff
>
> After that VM:pagefault
> I cannot see the two printf for tmp, maybe the pagefault comes before?
>
> Sorry but very hard day at work,
> tomorrow I'll do the tests with a little more "brain",
> I'll also do some checks for CORB|RIRB...
> and I'll post'em.
>
> Thank you for the patience.
>
> Best regards Iggo.

Relax :) I think we both want the same thing. I've got all the time
you need.

You don't have the new version of OSS. Apparently it's not yet in the
network repository. Please get the new version from the svn by
executing the following:
(first get svn from packman)

cd /usr/bigsrc
rm -rf build oss-v4.1-build1051-src-bsd
svn --username anonymous co https://gforge.cs.vu.nl/svn/minix/trunk/bigports/oss-v4.1-build1051-src-bsd/

Then compile it by:

cd oss-v4.1-build1051-src-bsd
sh build.minix

Hopefully it now works...

Kind regards,

Pieter

[Iggo]

Ooops... sorry.

The name of the package was the same... also the build 1051.

Today I'll follow your guide and try again with this new realease.

Good day.

Only for your information... the notebook with this audio chipset is
an hp (who likes "strange customized hw" in my opinion)
In this notebook there are also keys for enabling/disabling audio or
change volume, and in some forums seems also this peculiar model
soffers of problems with hearphones or digital output.
So, even if audio won't work for me it's not a big issue, and so I'm
doing this only for sake of knowledge.
In xubuntu karmic it works.

> svn --username anonymous cohttps://gforge.cs.vu.nl/svn/minix/trunk/bigports/oss-v4.1-build1051-s...

[Iggo]

Ok,

membar_addr 1:0xd6400004
membar_addr 2:0xd6400000
azbar:0x80000000

VM: pagefault SIGSEGV 840930 bad addr 0x0 err 0x6 nopage WRITE

Note: 840930 grows rapidly after every rs restart...

But again, no tmp(1) tmp(2)...

P.S. If you think could be helpful.. I could open a ssh on the
notebook...
but we have to find a way to communicate more interactively ...

Best regards Iggo.

[Colin Fowler]

AHHHA! I *think* I know what this is! It might be a bug in the PCI
driver that I discovered yesterday. pc_debug=1. How many BARs does
Minix tell you the card has? Is it more than 1? Are both non-zero? If
so...

Basically you have a 64bit BAR that Minix is treating as 32bit BARs.
The 04 at the end of the BAR address is the giveaway. That means that
the card is mapped into a 64bit address space, not a 32bit one. You
map the lower 32 bits thinking it's the BAR and you get rubbish when
trying to read and write to the card.

You might have a situation like this:

BAR0 0xe0000004
BAR1 0xd4000000

That could be 1 64 bit BAR being treated as two 32bit.

What you really need is this

BAR0 0xe0000000
BAR1 0x00000000

Both 64 bit and 32 bit addresses match and the first nibble of BAR0
is 0000 not 0100 (4). 2nd two bits of that nibble tell the card where
in the address space it can map. (Page 203 of the PCI Spec)

please post the pci_debug=1 output

Colin

> --
>
> You received this message because you are subscribed to the Google Groups "minix3" group.
> To post to this group, send email to min...@googlegroups.com.
> To unsubscribe from this group, send email to minix3+un...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/minix3?hl=en.
>
>
>

[Iggo]

Ehm... I pretend I undestood... :-)

I tried the boot with pci_debug... but the only "bar" I saw was about
103c 30b2 device :"bar_0 0x100 bytes at 0xd6102400 memory" (something
like that)
but the output runs fast...

After soundon the output was the same as the one I've posted before..
only with a PCI: do_acl... and a Pci_reserve_3 for IRQ 10..

But I fear it's not what you're expecting...

Did I miss something?

If I didn't misunderstand, should I force BAR value (membar_addr) to
something to go some steps further?
Do you want me to try?

Thank you, best regards Iggo.

On Dec 5, 4:27 pm, Colin Fowler <elethio...@gmail.com> wrote:
> AHHHA! I *think* I know what this is! It might be a bug in the PCI
> driver that I discovered yesterday. pc_debug=1. How many BARs does
> Minix tell you the card has? Is it more than 1? Are both non-zero? If
> so...
>
> Basically you have a 64bit BAR that Minix is treating as 32bit BARs.
> The 04 at the end of the BAR address is the giveaway. That means that
> the card is mapped into a 64bit address space, not a 32bit one. You
> map the lower 32 bits thinking it's the BAR and you get rubbish when
> trying to read and write to the card.
>
> You might have a situation like this:
>
> BAR0 0xe0000004
> BAR1 0xd4000000
>
> That could be 1 64 bit BAR being treated as two 32bit.
>
> What you really need is this
>
> BAR0 0xe0000000
> BAR1 0x00000000
>
> Both 64 bit and 32 bit addresses match and the first nibble of  BAR0
> is 0000 not 0100 (4). 2nd two bits of that nibble tell the card where
> in the address space it can map. (Page 203 of the PCI Spec)
>
> please post the pci_debug=1 output
>
> Colin
>

[Colin Fowler]

Can you post the contents of /var/log/messages after booting with
pci_debug=1 on? Just the lines from the last startup, don't post the
whole thing.

Colin

[Iggo]

It's embarassing..
(Yesterday night I could swear there wasn't this log on messages..
today I delete and recreate messages... and now here it is. Maybe I
was drunk. ;-)

Dec 6 09:38:59 Host-001 kernel: 0.27.0: Unknown device, vendor 8086
(Intel), device 27D8
Dec 6 09:38:59 Host-001 kernel: Device index: 2
Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x103c, did 0x30b2
Dec 6 09:38:59 Host-001 kernel: class Multimedia device (4/3/0)
Dec 6 09:38:59 Host-001 kernel: IRQ 10 for INTA
Dec 6 09:38:59 Host-001 kernel: bar_0: 0x4000 bytes at 0xd6400000
memory
Dec 6 09:38:59 Host-001 kernel: type = 0x4
Dec 6 09:38:59 Host-001 kernel: bar_1: 0x10 bytes at 0x0 memory
Dec 6 09:38:59 Host-001 kernel: @0x50 (0xc8426001): capability type
0x1: PCI Power Management
Dec 6 09:38:59 Host-001 kernel: @0x60 (0x00807005): capability type
0x5: Message Signaled Interrupts
Dec 6 09:38:59 Host-001 kernel: @0x70 (0x00910010): capability type
0x10: (unknown type)
Dec 6 09:38:59 Host-001 kernel: 0.28.0: Unknown device, vendor 8086
(Intel), device 27D0
Dec 6 09:38:59 Host-001 kernel: Device index: 3

I hope this is what you are looking for!

Best regards Iggo.

On Dec 6, 12:16 am, Colin Fowler <elethio...@gmail.com> wrote:
> Can you post the contents of /var/log/messages after booting with
> pci_debug=1  on? Just the lines from the last startup, don't post the
> whole thing.
>
> Colin
>

[Iggo]

Just to be sure I'll post all the last startup.... :-)

Dec 6 09:38:59 Host-001 syslogd: restart

Dec 6 09:38:59 Host-001 kernel:

Dec 6 09:38:59 Host-001 kernel: MINIX 3.1.5. (R3.1.5-r5612)
Dec 6 09:38:59 Host-001 kernel: Copyright 2009, Vrije Universiteit,
Amsterdam, The Netherlands
Dec 6 09:38:59 Host-001 kernel: MINIX is open source software, see
http://www.minix3.org
Dec 6 09:38:59 Host-001 kernel: pci_intel_init: unknown device
(8086/27A0)
Dec 6 09:38:59 Host-001 kernel: 0.0.0: Unknown device, vendor 8086
(Intel), device 27A0
Dec 6 09:38:59 Host-001 kernel: Device index: 0

Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x103c, did 0x30b2

Dec 6 09:38:59 Host-001 kernel: class Host bridge (6/0/0)
Dec 6 09:38:59 Host-001 kernel: @0xe0 (0x51090009): capability type
0x9: (unknown type)
Dec 6 09:38:59 Host-001 kernel: 0.1.0: Unknown device, vendor 8086
(Intel), device 27A1
Dec 6 09:38:59 Host-001 kernel: Device index: 1
Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x0, did 0x0
Dec 6 09:38:59 Host-001 kernel: class PCI-to-PCI bridge (6/4/0)

Dec 6 09:38:59 Host-001 kernel: IRQ 10 for INTA

Dec 6 09:38:59 Host-001 kernel: I/O window: base 0xf000, limit
0xfff, size -57344
Dec 6 09:38:59 Host-001 kernel: Memory window: base 0xd4000000,
limit 0xd5ffffff, size 0x2000000
Dec 6 09:38:59 Host-001 kernel: Prefetchable memory window: base
0xc0000000, limit 0xcfffffff, size 0x10000000
Dec 6 09:38:59 Host-001 kernel: @0x88 (0x0000800d): capability type
0xd: (unknown type)
Dec 6 09:38:59 Host-001 kernel: @0x80 (0xc8029001): capability type
0x1: PCI Power Management
Dec 6 09:38:59 Host-001 kernel: @0x90 (0x0000a005): capability type
0x5: Message Signaled Interrupts
Dec 6 09:38:59 Host-001 kernel: @0xa0 (0x01410010): capability type

0x10: (unknown type)
Dec 6 09:38:59 Host-001 kernel: 0.27.0: Unknown device, vendor 8086
(Intel), device 27D8
Dec 6 09:38:59 Host-001 kernel: Device index: 2
Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x103c, did 0x30b2
Dec 6 09:38:59 Host-001 kernel: class Multimedia device (4/3/0)
Dec 6 09:38:59 Host-001 kernel: IRQ 10 for INTA
Dec 6 09:38:59 Host-001 kernel: bar_0: 0x4000 bytes at 0xd6400000
memory
Dec 6 09:38:59 Host-001 kernel: type = 0x4
Dec 6 09:38:59 Host-001 kernel: bar_1: 0x10 bytes at 0x0 memory
Dec 6 09:38:59 Host-001 kernel: @0x50 (0xc8426001): capability type
0x1: PCI Power Management
Dec 6 09:38:59 Host-001 kernel: @0x60 (0x00807005): capability type
0x5: Message Signaled Interrupts
Dec 6 09:38:59 Host-001 kernel: @0x70 (0x00910010): capability type
0x10: (unknown type)
Dec 6 09:38:59 Host-001 kernel: 0.28.0: Unknown device, vendor 8086
(Intel), device 27D0
Dec 6 09:38:59 Host-001 kernel: Device index: 3

Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x0, did 0x0
Dec 6 09:38:59 Host-001 kernel: class PCI-to-PCI bridge (6/4/0)
Dec 6 09:38:59 Host-001 kernel: IRQ 11 for INTA
Dec 6 09:38:59 Host-001 kernel: I/O window: base 0x2000, limit
0x2fff, size 4096
Dec 6 09:38:59 Host-001 kernel: Memory window: base 0xd2000000,
limit 0xd3ffffff, size 0x2000000
Dec 6 09:38:59 Host-001 kernel: Prefetchable memory window: base
0xd0000000, limit 0xd1ffffff, size 0x2000000
Dec 6 09:38:59 Host-001 kernel: @0x40 (0x01418010): capability type
0x10: (unknown type)
Dec 6 09:38:59 Host-001 kernel: @0x80 (0x00009005): capability type
0x5: Message Signaled Interrupts
Dec 6 09:38:59 Host-001 kernel: @0x90 (0x0000a00d): capability type
0xd: (unknown type)
Dec 6 09:38:59 Host-001 kernel: @0xa0 (0xc8020001): capability type
0x1: PCI Power Management
Dec 6 09:38:59 Host-001 kernel: 0.28.2: Unknown device, vendor 8086
(Intel), device 27D4
Dec 6 09:38:59 Host-001 kernel: Device index: 4
Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x0, did 0x0
Dec 6 09:38:59 Host-001 kernel: class PCI-to-PCI bridge (6/4/0)
Dec 6 09:38:59 Host-001 kernel: IRQ 11 for INTC
Dec 6 09:38:59 Host-001 kernel: I/O window: base 0xf000, limit
0xfff, size -57344
Dec 6 09:38:59 Host-001 kernel: Memory window: base 0xfff00000,
limit 0xfffff, size 0x200000
Dec 6 09:38:59 Host-001 kernel: Prefetchable memory window: base
0xfff00000, limit 0xfffff, size 0x200000
Dec 6 09:38:59 Host-001 kernel: @0x40 (0x01418010): capability type
0x10: (unknown type)
Dec 6 09:38:59 Host-001 kernel: @0x80 (0x00009005): capability type
0x5: Message Signaled Interrupts
Dec 6 09:38:59 Host-001 kernel: @0x90 (0x0000a00d): capability type
0xd: (unknown type)
Dec 6 09:38:59 Host-001 kernel: @0xa0 (0xc8020001): capability type
0x1: PCI Power Management
Dec 6 09:38:59 Host-001 kernel: 0.28.3: Unknown device, vendor 8086
(Intel), device 27D6
Dec 6 09:38:59 Host-001 kernel: Device index: 5
Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x0, did 0x0
Dec 6 09:38:59 Host-001 kernel: class PCI-to-PCI bridge (6/4/0)
Dec 6 09:38:59 Host-001 kernel: IRQ 10 for INTD
Dec 6 09:38:59 Host-001 kernel: I/O window: base 0xf000, limit
0xfff, size -57344
Dec 6 09:38:59 Host-001 kernel: Memory window: base 0xd6000000,
limit 0xd60fffff, size 0x100000
Dec 6 09:38:59 Host-001 kernel: Prefetchable memory window: base
0xfff00000, limit 0xfffff, size 0x200000
Dec 6 09:38:59 Host-001 kernel: @0x40 (0x01418010): capability type
0x10: (unknown type)
Dec 6 09:38:59 Host-001 kernel: @0x80 (0x00009005): capability type
0x5: Message Signaled Interrupts
Dec 6 09:38:59 Host-001 kernel: @0x90 (0x0000a00d): capability type
0xd: (unknown type)
Dec 6 09:38:59 Host-001 kernel: @0xa0 (0xc8020001): capability type
0x1: PCI Power Management
Dec 6 09:38:59 Host-001 kernel: 0.29.0: Unknown device, vendor 8086
(Intel), device 27C8
Dec 6 09:38:59 Host-001 kernel: Device index: 6

Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x103c, did 0x30b2

Dec 6 09:38:59 Host-001 kernel: class USB (with UHC) (C/3/0)
Dec 6 09:38:59 Host-001 kernel: IRQ 11 for INTA
Dec 6 09:38:59 Host-001 kernel: bar_4: 32 bytes at 0x1800 I/O
Dec 6 09:38:59 Host-001 kernel: 0.29.1: Unknown device, vendor 8086
(Intel), device 27C9
Dec 6 09:38:59 Host-001 kernel: Device index: 7

Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x103c, did 0x30b2

Dec 6 09:38:59 Host-001 kernel: class USB (with UHC) (C/3/0)
Dec 6 09:38:59 Host-001 kernel: IRQ 10 for INTB
Dec 6 09:38:59 Host-001 kernel: bar_4: 32 bytes at 0x1820 I/O
Dec 6 09:38:59 Host-001 kernel: 0.29.2: Unknown device, vendor 8086
(Intel), device 27CA
Dec 6 09:38:59 Host-001 kernel: Device index: 8

Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x103c, did 0x30b2

Dec 6 09:38:59 Host-001 kernel: class USB (with UHC) (C/3/0)
Dec 6 09:38:59 Host-001 kernel: IRQ 11 for INTC
Dec 6 09:38:59 Host-001 kernel: bar_4: 32 bytes at 0x1840 I/O
Dec 6 09:38:59 Host-001 kernel: 0.29.3: Unknown device, vendor 8086
(Intel), device 27CB
Dec 6 09:38:59 Host-001 kernel: Device index: 9

Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x103c, did 0x30b2

Dec 6 09:38:59 Host-001 kernel: class USB (with UHC) (C/3/0)
Dec 6 09:38:59 Host-001 kernel: IRQ 10 for INTD
Dec 6 09:38:59 Host-001 kernel: bar_4: 32 bytes at 0x1860 I/O
Dec 6 09:38:59 Host-001 kernel: 0.29.7: Unknown device, vendor 8086
(Intel), device 27CC
Dec 6 09:38:59 Host-001 kernel: Device index: 10

Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x103c, did 0x30b2

Dec 6 09:38:59 Host-001 kernel: class Serial bus controller (C/3/20)
Dec 6 09:38:59 Host-001 kernel: IRQ 11 for INTA
Dec 6 09:38:59 Host-001 kernel: bar_0: 0x400 bytes at 0xd6404000
memory
Dec 6 09:38:59 Host-001 kernel: @0x50 (0xc9c25801): capability type
0x1: PCI Power Management
Dec 6 09:38:59 Host-001 kernel: @0x58 (0x20a0000a): capability type
0xa: (unknown type)
Dec 6 09:38:59 Host-001 kernel: 0.30.0: Unknown device, vendor 8086
(Intel), device 2448
Dec 6 09:38:59 Host-001 kernel: Device index: 11
Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x0, did 0x0
Dec 6 09:38:59 Host-001 kernel: class Subtractive decode PCI-to-PCI
bridge (6/4/1)
Dec 6 09:38:59 Host-001 kernel: I/O window: base 0x3000, limit
0x3fff, size 4096
Dec 6 09:38:59 Host-001 kernel: Memory window: base 0xd6100000,
limit 0xd61fffff, size 0x100000
Dec 6 09:38:59 Host-001 kernel: Prefetchable memory window: base
0xfff00000, limit 0xfffff, size 0x200000
Dec 6 09:38:59 Host-001 kernel: @0x50 (0x0000000d): capability type
0xd: (unknown type)
Dec 6 09:38:59 Host-001 kernel: 0.31.0: Unknown device, vendor 8086
(Intel), device 27B9
Dec 6 09:38:59 Host-001 kernel: Device index: 12

Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x103c, did 0x30b2

Dec 6 09:38:59 Host-001 kernel: class ISA bridge (6/1/0)
Dec 6 09:38:59 Host-001 kernel: @0xe0 (0x100c0009): capability type
0x9: (unknown type)
Dec 6 09:38:59 Host-001 kernel: 0.31.1: Unknown device, vendor 8086
(Intel), device 27DF
Dec 6 09:38:59 Host-001 kernel: Device index: 13

Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x103c, did 0x30b2

Dec 6 09:38:59 Host-001 kernel: class IDE controller (1/1/8A)
Dec 6 09:38:59 Host-001 kernel: PCI: device 0.31.1 uses INTA but is
not assigned any IRQ
Dec 6 09:38:59 Host-001 kernel: bar_0: 8 bytes at 0x0 I/O
Dec 6 09:38:59 Host-001 kernel: bar_1: 4 bytes at 0x0 I/O
Dec 6 09:38:59 Host-001 kernel: bar_2: 8 bytes at 0x0 I/O
Dec 6 09:38:59 Host-001 kernel: ba
Dec 6 09:38:59 Host-001 kernel: r_3: 4 bytes at 0x0 I/O
Dec 6 09:38:59 Host-001 kernel: bar_4: 16 bytes at 0x1880 I/O
Dec 6 09:38:59 Host-001 kernel: primary channel is not in native
mode, clearing BARs 0 and 1
Dec 6 09:38:59 Host-001 kernel: secondary channel is not in native
mode, clearing BARs 2 and 3
Dec 6 09:38:59 Host-001 kernel: skipping bar 0
Dec 6 09:38:59 Host-001 kernel: skipping bar 1
Dec 6 09:38:59 Host-001 kernel: skipping bar 2
Dec 6 09:38:59 Host-001 kernel: skipping bar 3
Dec 6 09:38:59 Host-001 kernel: 0.31.2: Unknown device, vendor 8086
(Intel), device 27C4
Dec 6 09:38:59 Host-001 kernel: Device index: 14

Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x103c, did 0x30b2

Dec 6 09:38:59 Host-001 kernel: class IDE controller (1/1/8F)
Dec 6 09:38:59 Host-001 kernel: IRQ 10 for INTB
Dec 6 09:38:59 Host-001 kernel: bar_0: 8 bytes at 0x18c8 I/O
Dec 6 09:38:59 Host-001 kernel: bar_1: 4 bytes at 0x18ac I/O
Dec 6 09:38:59 Host-001 kernel: bar_2: 8 bytes at 0x18c0 I/O
Dec 6 09:38:59 Host-001 kernel: bar_3: 4 bytes at 0x18a8 I/O
Dec 6 09:38:59 Host-001 kernel: bar_4: 16 bytes at 0x18b0 I/O
Dec 6 09:38:59 Host-001 kernel: bar_5: 0x400 bytes at 0xd6404400
memory
Dec 6 09:38:59 Host-001 kernel: @0x70 (0x40020001): capability type
0x1: PCI Power Management
Dec 6 09:38:59 Host-001 kernel: 0.31.3: Unknown device, vendor 8086
(Intel), device 27DA
Dec 6 09:38:59 Host-001 kernel: Device index: 15

Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x103c, did 0x30b2

Dec 6 09:38:59 Host-001 kernel: class SMBus (C/5/0)
Dec 6 09:38:59 Host-001 kernel: IRQ 10 for INTB
Dec 6 09:38:59 Host-001 kernel: bar_4: 32 bytes at 0x18e0 I/O
Dec 6 09:38:59 Host-001 kernel: (warning) unsupported ISA bridge
8086/27B9 for bus 0
Dec 6 09:38:59 Host-001 kernel: 0.1.0: PCI-to-PCI bridge: 8086/27A1
Dec 6 09:38:59 Host-001 kernel: bus(table) = 1, bus(sec) = 1, bus
(subord) = 1
Dec 6 09:38:59 Host-001 kernel: 1.0.0: Unknown device, vendor 10DE
(nVidia Corporation), device 01D6
Dec 6 09:38:59 Host-001 kernel: Device index: 16

Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x103c, did 0x30b2

Dec 6 09:38:59 Host-001 kernel: class VGA-compatible controller
(3/0/0)

Dec 6 09:38:59 Host-001 kernel: IRQ 10 for INTA

Dec 6 09:38:59 Host-001 kernel: bar_0: 0x1000000 bytes at 0xd5000000
memory
Dec 6 09:38:59 Host-001 kernel: bar_1: 0x8000000 bytes at 0xc0000000
prefetchable memory

Dec 6 09:38:59 Host-001 kernel: type = 0x4

Dec 6 09:38:59 Host-001 kernel: bar_2: 0x10 bytes at 0x0 memory
Dec 6 09:38:59 Host-001 kernel: bar_3: 0x1000000 bytes at 0xd4000000

memory
Dec 6 09:38:59 Host-001 kernel: type = 0x4

Dec 6 09:38:59 Host-001 kernel: bar_4: 0x10 bytes at 0x0 memory
Dec 6 09:38:59 Host-001 kernel: @0x60 (0x00026801): capability type
0x1: PCI Power Management
Dec 6 09:38:59 Host-001 kernel: @0x68 (0x00807805): capability type
0x5: Message Signaled Interrupts
Dec 6 09:38:59 Host-001 kernel: @0x78 (0x00010010): capability type
0x10: (unknown type)
Dec 6 09:38:59 Host-001 kernel: 0.28.0: PCI-to-PCI bridge: 8086/27D0
Dec 6 09:38:59 Host-001 kernel: bus(table) = 2, bus(sec) = 2, bus
(subord) = 3
Dec 6 09:38:59 Host-001 kernel: 0.28.2: PCI-to-PCI bridge: 8086/27D4
Dec 6 09:38:59 Host-001 kernel: bus(table) = 3, bus(sec) = 4, bus
(subord) = 4
Dec 6 09:38:59 Host-001 kernel: 0.28.3: PCI-to-PCI bridge: 8086/27D6
Dec 6 09:38:59 Host-001 kernel: bus(table) = 4, bus(sec) = 5, bus
(subord) = 5
Dec 6 09:38:59 Host-001 kernel: 4.0.0: Unknown device, vendor 8086
(Intel), device 4222
Dec 6 09:38:59 Host-001 kernel: Device index: 17
Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x103c, did 0x135c
Dec 6 09:38:59 Host-001 kernel: class Other network controller
(2/80/0)

Dec 6 09:38:59 Host-001 kernel: IRQ 10 for INTA

Dec 6 09:38:59 Host-001 kernel: bar_0: 0x1000 bytes at 0xd6000000
memory
Dec 6 09:38:59 Host-001 kernel: @0xc8 (0xc822d001): capability type
0x1: PCI Power Management
Dec 6 09:38:59 Host-001 kernel: @0xd0 (0x0080e005): capability type
0x5: Message Signaled Interrupts
Dec 6 09:38:59 Host-001 kernel: @0xe0 (0x00110010): capability type
0x10: (unknown type)
Dec 6 09:38:59 Host-001 kernel: 0.30.0: PCI-to-PCI bridge: 8086/2448
Dec 6 09:38:59 Host-001 kernel: bus(table) = 5, bus(sec) = 8, bus
(subord) = 8
Dec 6 09:38:59 Host-001 kernel: 5.8.0: Unknown device, vendor 8086
(Intel), device 1092
Dec 6 09:38:59 Host-001 kernel: Device index: 18

Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x103c, did 0x30b2

Dec 6 09:38:59 Host-001 kernel: class Ethernet controller (2/0/0)

Dec 6 09:38:59 Host-001 kernel: IRQ 10 for INTA

Dec 6 09:38:59 Host-001 kernel: bar_0: 0x1000 bytes at 0xd6100000
memory
Dec 6 09:38:59 Host-001 kernel: bar_1: 64 bytes at 0x3000 I/O
Dec 6 09:38:59 Host-001 kernel: @0xdc (0xfe220001): capability type
0x1: PCI Power Management
Dec 6 09:38:59 Host-001 kernel: 5.9.0: Unknown device, vendor 1180
(unknown), device 0832
Dec 6 09:38:59 Host-001 kernel: Device index: 19

Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x103c, did 0x30b2

Dec 6 09:38:59 Host-001 kernel: class IEEE 1394 (OpenHCI) (C/0/10)

Dec 6 09:38:59 Host-001 kernel: IRQ 10 for INTA

Dec 6 09:38:59 Host-001 kernel: bar_0: 0x800 bytes at 0xd6101000
memory
Dec 6 09:38:59 Host-001 kernel: @0xdc (0xfe020001): capability type
0x1: PCI Power Management
Dec 6 09:38:59 Host-001 kernel: 5.9.1: Unknown device, vendor 1180
(unknown), device 0822
Dec 6 09:38:59 Host-001 kernel: Device index: 20

Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x103c, did 0x30b2

Dec 6 09:38:59 Host-001 kernel: class Base system peripheral (8/5/0)
Dec 6 09:38:59 Host-001 kernel: IRQ 11 for INTB
Dec 6 09:38:59 Host-001 kernel: bar_0: 0x100 bytes at 0xd6101800
memory
Dec 6 09:38:59 Host-001 kernel: @0x80 (0xfe020001): capability type
0x1: PCI Power Management
Dec 6 09:38:59 Host-001 kernel: 5.9.2: Unknown device, vendor 1180
(unknown), device 0843
Dec 6 09:38:59 Host-001 kernel: Device index: 21

Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x103c, did 0x30b2

Dec 6 09:38:59 Host-001 kernel: class Base system peripheral
(8/80/0)
Dec 6 09:38:59 Host-001 kernel: IRQ 11 for INTB
Dec 6 09:38:59 Host-001 kernel: bar_0: 0x100 bytes at 0xd6101c00
memory
Dec 6 09:38:59 Host-001 kernel: @0x80 (0xfe020001): capability type
0x1: PCI Power Management
Dec 6 09:38:59 Host-001 kernel: 5.9.3: Unknown device, vendor 1180
(unknown), device 0592
Dec 6 09:38:59 Host-001 kernel: Device index: 22

Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x103c, did 0x30b2

Dec 6 09:38:59 Host-001 kernel: class Base system peripheral
(8/80/0)
Dec 6 09:38:59 Host-001 kernel: IRQ 11 for INTB
Dec 6 09:38:59 Host-001 kernel: bar_0: 0x100 bytes at 0xd6102000
memory
Dec 6 09:38:59 Host-001 kernel: @0x80 (0xfe020001): capability type
0x1: PCI Power Management
Dec 6 09:38:59 Host-001 kernel: 5.9.4: Unknown device, vendor 1180
(unknown), device 0852
Dec 6 09:38:59 Host-001 kernel: Device index: 23

Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x103c, did 0x30b2

Dec 6 09:38:59 Host-001 kernel: class Base system peripheral
(8/80/0)
Dec 6 09:38:59 Host-001 kernel: IRQ 11 for INTB
Dec 6 09:38:59 Host-001 kernel: bar_0: 0x100 bytes at 0xd6102400
memory
Dec 6 09:38:59 Host-001 kernel: @0x80 (0xfe020001): capability type
0x1: PCI Power Management
Dec 6 09:38:59 Host-001 kernel: PCI: do_acl: setting ACL for 73127
('floppy') at entry 0
Dec 6 09:38:59 Host-001 kernel: PCI: do_acl: setting ACL for 73129
('at_wini_0') at entry 1
Dec 6 09:38:59 Host-001 kernel: pci_reserve3: for proc 73129, adding
I/O range [0x1880..0x188f]
Dec 6 09:38:59 Host-001 kernel: pci_reserve3: for proc 73129, adding
I/O range [0x18c8..0x18cf]
Dec 6 09:38:59 Host-001 kernel: pci_reserve3: for proc 73129, adding
I/O range [0x18ac..0x18af]
Dec 6 09:38:59 Host-001 kernel: pci_reserve3: for proc 73129, adding
I/O range [0x18c0..0x18c7]
Dec 6 09:38:59 Host-001 kernel: pci_reserve3: for proc 73129, adding
I/O range [0x18a8..0x18ab]
Dec 6 09:38:59 Host-001 kernel: pci_reserve3: for proc 73129, adding
I/O range [0x18b0..0x18bf]
Dec 6 09:38:59 Host-001 kernel: pci_reserve3: for proc 73129, should
add memory range [0xd6404400..0xd64047ff]
Dec 6 09:38:59 Host-001 kernel: pci_reserve3: adding IRQ 10
Dec 6 09:38:59 Host-001 kernel: at_wini0 (compat): pci_reserve 14
failed!
Dec 6 09:38:59 Host-001 kernel: PCI: do_acl: setting ACL for 73130
('at_wini_1') at entry 2
Dec 6 09:38:59 Host-001 kernel: PCI: do_acl: setting ACL for 36582
('fs_c0d4p3s0') at entry 3
Dec 6 09:38:59 Host-001 kernel: PCI: do_acl: setting ACL for 36585
('is') at entry 4
Dec 6 09:38:59 Host-001 kernel: PCI: do_acl: setting ACL for 36588
('readclock.drv') at entry 5
Dec 6 09:38:59 Host-001 kernel: Sun Dec 6 09:38:49 GMT 2009
Dec 6 09:38:59 Host-001 kernel: PCI: do_acl: setting ACL for 36597
('fs_c0d4p3s2') at entry 5
Dec 6 09:38:59 Host-001 kernel: PCI: do_acl: setting ACL for 36603
('fs_c0d4p3s1') at entry 6
Dec 6 09:38:59 Host-001 kernel: PCI: do_acl: setting ACL for 36610
('random') at entry 7
Dec 6 09:38:59 Host-001 kernel: PCI: do_acl: setting ACL for 36618
('fxp') at entry 8
Dec 6 09:38:59 Host-001 kernel: PCI: do_acl: setting ACL for 366
Dec 6 09:38:59 Host-001 kernel: 24 ('inet') at entry 9
Dec 6 09:38:59 Host-001 kernel: pci_reserve3: for proc 36618, should
add memory range [0xd6100000..0xd6100fff]
Dec 6 09:38:59 Host-001 kernel: pci_reserve3: for proc 36618, adding
I/O range [0x3000..0x303f]
Dec 6 09:38:59 Host-001 kernel: pci_reserve3: adding IRQ 10
Dec 6 09:38:59 Host-001 kernel: fxp#0: device revision: 82557B
Dec 6 09:38:59 Host-001 kernel: PCI: do_acl: setting ACL for 36627
('printer') at entry 10
Dec 6 09:38:59 Host-001 kernel: PCI: do_acl: setting ACL for 36630
('ipc') at entry 11
Dec 6 09:38:59 Host-001 sshd[101]: error: setsockopt SO_REUSEADDR:
Function not implemented
Dec 6 09:38:59 Host-001 sshd[101]: Server listening on 0.0.0.0 port
22.
Dec 6 09:43:09 Host-001 sshd[116]: Invalid user viruz from
192.168.1.212
Dec 6 09:43:09 Host-001 sshd[116]: input_userauth_request: invalid
user viruz
Dec 6 09:43:09 Host-001 sshd[116]: Failed none for invalid user viruz
from 192.168.1.212 port 53632 ssh2
Dec 6 09:43:09 Host-001 sshd[116]: Failed keyboard-interactive for
invalid user viruz from 192.168.1.212 port 53632 ssh2
Dec 6 09:43:10 Host-001 sshd[116]: Failed password for invalid user
viruz from 192.168.1.212 port 53632 ssh2
Dec 6 09:43:12 Host-001 sshd[116]: Connection closed by 192.168.1.212
Dec 6 09:43:20 Host-001 sshd[118]: Accepted password for root from
192.168.1.212 port 53641 ssh2
Dec 6 09:43:20 Host-001 sshd[118]: subsystem request for sftp

[Colin Fowler]

Looks fine. The second BAR is 0x0 so it's not the bug I'm looking at.

Colin

[pieter]

Hi Iggo,

The addresses in what you posted earlier:

> > >> > membar_addr 1:0xd6400004
> > >> > membar_addr 2:0xd6400000
> > >> > azbar:0x80000000

> > >> > VM: pagefault SIGSEGV 840930 bad addr 0x0 err 0x6 nopage WRITE

> > >> > Note: 840930 grows rapidly after every rs restart...

> > >> > But again, no tmp(1) tmp(2)...

In combination with this:

On Dec 6, 10:51 am, Iggo <gepdi...@gmail.com> wrote:
> Dec 6 09:38:59 Host-001 kernel: 0.27.0: Unknown device, vendor 8086
> (Intel), device 27D8
> Dec 6 09:38:59 Host-001 kernel: Device index: 2
> Dec 6 09:38:59 Host-001 kernel: Subsystem: Vid 0x103c, did 0x30b2
> Dec 6 09:38:59 Host-001 kernel: class Multimedia device (4/3/0)
> Dec 6 09:38:59 Host-001 kernel: IRQ 10 for INTA
> Dec 6 09:38:59 Host-001 kernel: bar_0: 0x4000 bytes at 0xd6400000
> memory
> Dec 6 09:38:59 Host-001 kernel: type = 0x4
> Dec 6 09:38:59 Host-001 kernel: bar_1: 0x10 bytes at 0x0 memory
> Dec 6 09:38:59 Host-001 kernel: @0x50 (0xc8426001): capability type
> 0x1: PCI Power Management
> Dec 6 09:38:59 Host-001 kernel: @0x60 (0x00807005): capability type
> 0x5: Message Signaled Interrupts
> Dec 6 09:38:59 Host-001 kernel: @0x70 (0x00910010): capability type
> 0x10: (unknown type)
> Dec 6 09:38:59 Host-001 kernel: 0.28.0: Unknown device, vendor 8086
> (Intel), device 27D0
> Dec 6 09:38:59 Host-001 kernel: Device index: 3

look fine to me. So, I don't understand why it doesn't work.

> > >> > P.S. If you think could be helpful.. I could open a ssh on the
> > >> > notebook...
> > >> > but we have to find a way to communicate more interactively ...

Perhaps, this is a good idea. I'm very curious now, so perhaps google
chat or something?

Cheers,

Pieter

[Iggo]

I thing could be nice to use minix chat...?
(only things, along with skype I've ever used for chatting... It
requires no software installation nor registration so it is very
"speedy").

"http://webchat.freenode.net/?channels=minix"

Otherwise tell me..

Only tell me a couple of things:

Do you want me to reainstall a "clear 3.1.5 installation in the
meanwhile?"

Telnet or ssh? (ssh is already set up)

Best regards, Iggo.

[Iggo]

Sorry... but seems we have not met by some minutes...

Tonight I'm there, with the exception of some moment to eat, to
smoke.. or something else...

See you.

[Iggo]

Beg your pardon...

Thank you for all the work you did, I really apreciate it.

Having a "superficial" look at intel docs for ich7 and hdaudio, I had
the impression it should be "retrocompatible" with ac97 standard...

Am I mistaking?

If it is so, should there be a way to use the audio card in such a
compatible way?

Thank you very much!

Best regards Iggo.

[Greg King]

On Dec 2 2009, 8:49 am, pieter wrote:
> On Nov 30, 6:18 pm, Iggo wrote:
>
> > If I add the VID and DID in "drivers.conf", OSS finds my card!
>
> Hmm, that is strange. There should be the following in the

> "drivers.conf":
>
> pci class 4/1 # Multimedia / Audio device
>

> That was enough for all the cards I tested.

On Dec 6 2009, 4:51 am, Iggo wrote:
> It's embarassing.
> Yesterday night, I could swear there wasn't this log on "messages".
> Today, I deleted and recreated "messages". And now, here it is. Maybe I

> was drunk. ;-)
>
> Dec  6 09:38:59 Host-001 kernel: 0.27.0: Unknown device, vendor 8086
> (Intel), device 27D8
> Dec  6 09:38:59 Host-001 kernel: Device index: 2
> Dec  6 09:38:59 Host-001 kernel: Subsystem: Vid 0x103c, did 0x30b2
> Dec  6 09:38:59 Host-001 kernel:        class Multimedia device (4/3/0)
> Dec  6 09:38:59 Host-001 kernel:        IRQ 10 for INTA
> Dec  6 09:38:59 Host-001 kernel:        bar_0: 0x4000 bytes at 0xd6400000
> memory
> Dec  6 09:38:59 Host-001 kernel: type = 0x4
> Dec  6 09:38:59 Host-001 kernel:        bar_1: 0x10 bytes at 0x0 memory
> Dec  6 09:38:59 Host-001 kernel:  @0x50 (0xc8426001): capability type
> 0x1: PCI Power Management
> Dec  6 09:38:59 Host-001 kernel:  @0x60 (0x00807005): capability type
> 0x5: Message Signaled Interrupts
> Dec  6 09:38:59 Host-001 kernel:  @0x70 (0x00910010): capability type
> 0x10: (unknown type)
>

> I hope this is what you are looking for!

> Dec 6 09:38:59 Host-001 kernel: class Multimedia device (4/3/0)

Can that be why OSS didn't recognize Iggo's card -- Intel used a
different PCI sub-class number? If that is so, then Pieter needs to
add another "pci class" line to the "osscore" section of
"system.conf".