Mailing List Archive: ivtv: devel

pci_find_slot failure patch

Index | Next | Previous | View Threaded

lewing at novell

Jul 26, 2005, 11:45 AM

Post #1 of 11 (1726 views)
Permalink

pci_find_slot failure patch

I stuck my pvr 250 in my dual opteron box and the ivtv driver was unable
to properly initialize. A little digging and a small change fixed
things for me. I have no idea if it is correct but I thought I'd pass
it along, the patch is against 0.3.6r but the change appears to still be
needed in 0.3.7a.

--Larry

Attachments:

ivtv.patch (0.46 KB)

hydrateordie at earthlink

Jul 26, 2005, 12:20 PM

Post #2 of 11 (1705 views)
Permalink

Re: pci_find_slot failure patch [In reply to]

I'm having trouble getting my PVR-150 to work with ivtv... All I get
is black static, and I have no sound when I run
# /usr/lib/ivtv/ptune-ui.pl &
# mplayer -vo xv /dev/video0

I've been following the Fedora Core Myth guide at http://wilsonet.com/
mythtv/fcmyth.php

I'm on NTSC and I did cp HcwMakoA.ROM /lib/modules to install the
audio firmware from the Hauppage cd.

The /etc/modprob.conf details are:
# ivtv (PVR-150)
alias char-major-81 videodev
alias char-major-81-0 ivtv
alias tveeprom tveeprom-ivtv
alias tuner tuner-ivtv
alias msp3400 msp3400-ivtv

and here's the output from

tac /var/log/messages |
sed -n '/=\ \ END INIT IVTV\ \ =/,/= START INIT IVTV =/p;
/= START INIT IVTV =/q' |
tac

Jul 26 12:02:51 localhost kernel: ivtv: ==================== START
INIT IVTV ====================
Jul 26 12:02:51 localhost kernel: ivtv: version 0.3.7 (a) loading
Jul 26 12:02:51 localhost kernel: ivtv: Linux version:
2.6.12-1.1398_FC4 686 REGPARM 4KSTACKS gcc-4.0
Jul 26 12:02:51 localhost kernel: ivtv: In case of problems please
include the debug info
Jul 26 12:02:51 localhost kernel: ivtv: between the START INIT IVTV
and END INIT IVTV lines when
Jul 26 12:02:51 localhost kernel: ivtv: mailing the ivtv-devel
mailinglist.
Jul 26 12:02:51 localhost kernel: ivtv: Autodetected WinTV PVR 150 card
Jul 26 12:02:51 localhost kernel: ivtv: Found an iTVC16 based chip
Jul 26 12:02:51 localhost kernel: ACPI: PCI Interrupt 0000:00:08.0[A]
-> Link [LNKA] -> GSI 11 (level, low) -> IRQ 11
Jul 26 12:02:51 localhost kernel: ivtv: Unreasonably low latency
timer, setting to 64 (was 32)
Jul 26 12:02:51 localhost kernel: ivtv: VIA PCI device: 0x3205
vendor: 0x1106
Jul 26 12:02:51 localhost kernel: ivtv: Failed to load module tveeprom
Jul 26 12:02:51 localhost kernel: ivtv: i2c client addr: 0x50 not
found for command 0x0!
Jul 26 12:02:51 localhost kernel: ivtv: Error -19 reading Hauppauge
eeprom.
Jul 26 12:02:51 localhost kernel: ivtv: Possible causes: the tveeprom
module was not loaded, or
Jul 26 12:02:51 localhost kernel: ivtv: the eeprom kernel module was
loaded before the tveeprom module.
Jul 26 12:02:51 localhost kernel: ivtv: Failed to load module tuner
Jul 26 12:02:52 localhost kernel: cx25840: requesting /lib/modules/
HcwMakoA.ROM
Jul 26 12:02:53 localhost kernel: cx25840: firmware loaded
Jul 26 12:02:53 localhost kernel: ivtv: i2c attach [client=cx25840
[50],ok]
Jul 26 12:02:53 localhost kernel: ivtv: i2c attach [client=wm8775
[50],ok]
Jul 26 12:02:53 localhost kernel: ivtv: Could not detect tuner
standard, defaulting to NTSC.
Jul 26 12:02:54 localhost kernel: ivtv: Encoder revision: 0x02050032
Jul 26 12:02:54 localhost kernel: ivtv: Encoder Firmware may be
buggy, use version 0x02040011
Jul 26 12:02:54 localhost kernel: ivtv: Configuring WinTV PVR 150
card with 4 streams
Jul 26 12:02:54 localhost kernel: ivtv: Registered v4l2 device for
encoder MPEG minor 0
Jul 26 12:02:54 localhost kernel: ivtv: Create DMA encoder MPEG
stream: 0 x 32768 buffers (0KB total)
Jul 26 12:02:54 localhost kernel: ivtv: Registered v4l2 device for
encoder YUV minor 32
Jul 26 12:02:54 localhost kernel: ivtv: Create DMA encoder YUV
stream: 0 x 10800 buffers (0KB total)
Jul 26 12:02:54 localhost kernel: ivtv: Registered v4l2 device for
encoder VBI minor 224
Jul 26 12:02:54 localhost kernel: ivtv: Create DMA encoder VBI
stream: 0 x 8736 buffers (0KB total)
Jul 26 12:02:54 localhost kernel: ivtv: Registered v4l2 device for
encoder PCM audio minor 24
Jul 26 12:02:54 localhost kernel: ivtv: Create DMA encoder PCM audio
stream: 0 x 4608 buffers (0KB total)
Jul 26 12:02:54 localhost kernel: ivtv: i2c client id: 0x02 not found!
Jul 26 12:02:54 localhost kernel: ivtv: Switching standard to NTSC.
Jul 26 12:02:54 localhost kernel: ivtv: i2c client id: 0x02 not found!
Jul 26 12:02:54 localhost kernel: ivtv: i2c client addr: 0x43 not
found for command 0x40307603!
Jul 26 12:02:54 localhost kernel: ivtv: Initialized WinTV PVR 150,
card #0
Jul 26 12:02:54 localhost kernel: ivtv: ==================== END
INIT IVTV ====================

I've hit a wall, and I'm pretty much stuck as to what I should try.
Thank you all very much!

-KG

tatrafford at comcast

Jul 26, 2005, 12:48 PM

Post #3 of 11 (1699 views)
Permalink

Re: tveeprom not starting [In reply to]

hydrateordie [at] earthlink wrote:

> I'm having trouble getting my PVR-150 to work with ivtv... All I get
> is black static, and I have no sound when I run
<snip>
> The /etc/modprob.conf details are:
> # ivtv (PVR-150)
> alias char-major-81 videodev
> alias char-major-81-0 ivtv
> alias tveeprom tveeprom-ivtv
> alias tuner tuner-ivtv
> alias msp3400 msp3400-ivtv

Did you install from source, or from ATrpms? If you installed from
source, get rid of those last three lines, rerun depmod -a, and reload
ivtv.
--
Tyler Trafford

-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
_______________________________________________
ivtv-devel mailing list
ivtv-devel [at] lists
https://lists.sourceforge.net/lists/listinfo/ivtv-devel

lewing at novell

Jul 26, 2005, 6:32 PM

Post #4 of 11 (1671 views)
Permalink

Re: pci_find_slot failure patch [In reply to]

On Tue, 2005-07-26 at 13:45 -0500, Larry Ewing wrote:
> I stuck my pvr 250 in my dual opteron box and the ivtv driver was unable
> to properly initialize. A little digging and a small change fixed
> things for me. I have no idea if it is correct but I thought I'd pass
> it along, the patch is against 0.3.6r but the change appears to still be
> needed in 0.3.7a.
>

Sigh, after all these years you'd think I could get a diff right,
clearly the patch I sent before was reversed, here is the correct patch.

--Larry

Attachments:

ivtv.patch (0.50 KB)

bmayland at leoninedev

Jul 27, 2005, 4:39 AM

Post #5 of 11 (1703 views)
Permalink

Re: pci_find_slot failure patch [In reply to]

Larry Ewing wrote:

>Sigh, after all these years you'd think I could get a diff right,
>clearly the patch I sent before was reversed, here is the correct patch.
>
>
hehe yeah I noticed that. The code isn't right though, but I didn't
realize what it was trying to do either the first time I saw it. What
we're looking for there is the host bus the card is on, not the card
itself. The idea is to determine of the PCI bus itself is considered
buggy and adjust parameters on it accordingly. Do you not have a
0:00:00.0 device in your lspci?

You're right, the code that is there is wrong though.

-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
_______________________________________________
ivtv-devel mailing list
ivtv-devel [at] lists
https://lists.sourceforge.net/lists/listinfo/ivtv-devel

lewing at novell

Jul 27, 2005, 8:50 AM

Post #6 of 11 (1680 views)
Permalink

Re: pci_find_slot failure patch [In reply to]

On Wed, 2005-07-27 at 07:39 -0400, Bryan Mayland wrote:
> Larry Ewing wrote:
>
> >Sigh, after all these years you'd think I could get a diff right,
> >clearly the patch I sent before was reversed, here is the correct patch.
> >
> >
> hehe yeah I noticed that. The code isn't right though, but I didn't
> realize what it was trying to do either the first time I saw it. What
> we're looking for there is the host bus the card is on, not the card
> itself. The idea is to determine of the PCI bus itself is considered
> buggy and adjust parameters on it accordingly. Do you not have a
> 0:00:00.0 device in your lspci?
>

I don't appear to. I've attached the lspci output from that machine.

> You're right, the code that is there is wrong though.
>

--Larry

Attachments:

lspci.text (2.21 KB)

markknecht at gmail

Jul 27, 2005, 11:59 AM

Post #7 of 11 (1704 views)
Permalink

Re: pci_find_slot failure patch [In reply to]

On 7/27/05, Larry Ewing <lewing [at] novell> wrote:
> On Wed, 2005-07-27 at 07:39 -0400, Bryan Mayland wrote:
> > Larry Ewing wrote:
> >
> > >Sigh, after all these years you'd think I could get a diff right,
> > >clearly the patch I sent before was reversed, here is the correct patch.
> > >
> > >
> > hehe yeah I noticed that. The code isn't right though, but I didn't
> > realize what it was trying to do either the first time I saw it. What
> > we're looking for there is the host bus the card is on, not the card
> > itself. The idea is to determine of the PCI bus itself is considered
> > buggy and adjust parameters on it accordingly. Do you not have a
> > 0:00:00.0 device in your lspci?
> >
>
> I don't appear to. I've attached the lspci output from that machine.
>
> > You're right, the code that is there is wrong though.
> >
>
> --Larry

"Shades of PCI Bridge troubles Batman, it looks like the Riddles has returned!"

Strange listing:

a) PCI Bus #1 is missing.
b) Bus 0 - devices 0-5 are missing

It might be good to post back the output of pci -vv but trim it to
show on'y the PCI-PCI bridges. For instance here's one bridge in my
problematic machine:

0000:00:1e.0 PCI bridge: Intel Corporation 82801 PCI Bridge (rev c2)
(prog-if 00 [Normal decode])
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr- Stepping- SERR+ FastB2B-
Status: Cap- 66Mhz- UDF- FastB2B+ ParErr- DEVSEL=fast >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
Latency: 0
Bus: primary=00, secondary=01, subordinate=01, sec-latency=32
I/O behind bridge: 00009000-00009fff
Memory behind bridge: f8000000-f9ffffff
Prefetchable memory behind bridge: f0000000-f7ffffff
BridgeCtl: Parity- SERR+ NoISA+ VGA- MAbort- >Reset- FastB2B-

The 'Bus:' section let's us see which bridges are creating the network
on the way to the video device.

- Mark

-------------------------------------------------------
SF.Net email is Sponsored by the Better Software Conference & EXPO September
19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
_______________________________________________
ivtv-devel mailing list
ivtv-devel [at] lists
https://lists.sourceforge.net/lists/listinfo/ivtv-devel

bmayland at leoninedev

Jul 27, 2005, 12:18 PM

Post #8 of 11 (1669 views)
Permalink

Re: pci_find_slot failure patch [In reply to]

Mark Knecht wrote:

>"Shades of PCI Bridge troubles Batman, it looks like the Riddles has returned!"
>
>
It's worse than I thought, Chum, we don't even use the code that
this was orignally added to enable! I'd suggest removing the whole
thing, from the pci_bus = down to the end of the function, and removing
the via_fix member from the ivtv struct altogether. I'd make a patch,
but I've got my code in a broken place right now. Would you mind doing
that Larry?

-------------------------------------------------------
SF.Net email is Sponsored by the Better Software Conference & EXPO September
19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
_______________________________________________
ivtv-devel mailing list
ivtv-devel [at] lists
https://lists.sourceforge.net/lists/listinfo/ivtv-devel

bhoar at speakeasy

Jul 27, 2005, 12:18 PM

Post #9 of 11 (1688 views)
Permalink

Re: pci_find_slot failure patch [In reply to]

On Wed, 27 Jul 2005, Mark Knecht wrote:
> On 7/27/05, Larry Ewing <lewing [at] novell> wrote:
> > On Wed, 2005-07-27 at 07:39 -0400, Bryan Mayland wrote:
> > > hehe yeah I noticed that. The code isn't right though, but I didn't
> > > realize what it was trying to do either the first time I saw it. What
> > > we're looking for there is the host bus the card is on, not the card
> > > itself. The idea is to determine of the PCI bus itself is considered
> > > buggy and adjust parameters on it accordingly. Do you not have a
> > > 0:00:00.0 device in your lspci?
> >
> > I don't appear to. I've attached the lspci output from that machine.
> >
> > > You're right, the code that is there is wrong though.
> >
> > --Larry
>
> "Shades of PCI Bridge troubles Batman, it looks like the Riddles has returned!"
>
> Strange listing:
>
> a) PCI Bus #1 is missing.
> b) Bus 0 - devices 0-5 are missing

One thing I noticed as well is the mention of PCI-X bridges. Could
there be an issue related to that?

-brendan

-------------------------------------------------------
SF.Net email is Sponsored by the Better Software Conference & EXPO September
19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
_______________________________________________
ivtv-devel mailing list
ivtv-devel [at] lists
https://lists.sourceforge.net/lists/listinfo/ivtv-devel

markknecht at gmail

Jul 27, 2005, 1:06 PM

Post #10 of 11 (1694 views)
Permalink

Re: pci_find_slot failure patch [In reply to]

On 7/27/05, Brendan Hoar <bhoar [at] speakeasy> wrote:
> On Wed, 27 Jul 2005, Mark Knecht wrote:
> > On 7/27/05, Larry Ewing <lewing [at] novell> wrote:
> > > On Wed, 2005-07-27 at 07:39 -0400, Bryan Mayland wrote:
> > > > hehe yeah I noticed that. The code isn't right though, but I didn't
> > > > realize what it was trying to do either the first time I saw it. What
> > > > we're looking for there is the host bus the card is on, not the card
> > > > itself. The idea is to determine of the PCI bus itself is considered
> > > > buggy and adjust parameters on it accordingly. Do you not have a
> > > > 0:00:00.0 device in your lspci?
> > >
> > > I don't appear to. I've attached the lspci output from that machine.
> > >
> > > > You're right, the code that is there is wrong though.
> > >
> > > --Larry
> >
> > "Shades of PCI Bridge troubles Batman, it looks like the Riddles has returned!"
> >
> > Strange listing:
> >
> > a) PCI Bus #1 is missing.
> > b) Bus 0 - devices 0-5 are missing
>
> One thing I noticed as well is the mention of PCI-X bridges. Could
> there be an issue related to that?
>
> -brendan
>

Absolutely. PCI-X is pretty new. (I served on the original committee
to write the specs) It's out there in more and more machines now. I
don't remember all the details but IIRC there were some small
differences in the original specs about doing config cycles across
them. This may or may not have survived in the final versions of the
specs.

- Mark

-------------------------------------------------------
SF.Net email is Sponsored by the Better Software Conference & EXPO September
19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
_______________________________________________
ivtv-devel mailing list
ivtv-devel [at] lists
https://lists.sourceforge.net/lists/listinfo/ivtv-devel

lewing at novell

Jul 29, 2005, 1:47 PM

Post #11 of 11 (1684 views)
Permalink

Re: pci_find_slot failure patch [In reply to]

Here you go, this just removes the via_fix variable and the code that
sets it.

--Larry

On Wed, 2005-07-27 at 15:18 -0400, Bryan Mayland wrote:
> Mark Knecht wrote:
>
> >"Shades of PCI Bridge troubles Batman, it looks like the Riddles has returned!"
> >
> >
> It's worse than I thought, Chum, we don't even use the code that
> this was orignally added to enable! I'd suggest removing the whole
> thing, from the pci_bus = down to the end of the function, and removing
> the via_fix member from the ivtv struct altogether. I'd make a patch,
> but I've got my code in a broken place right now. Would you mind doing
> that Larry?
>
>
> -------------------------------------------------------
> SF.Net email is Sponsored by the Better Software Conference & EXPO September
> 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
> Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
> Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
> _______________________________________________
> ivtv-devel mailing list
> ivtv-devel [at] lists
> https://lists.sourceforge.net/lists/listinfo/ivtv-devel

Attachments:

ivtv-less.patch (1.54 KB)

Index | Next | Previous | View Threaded

Interested in having your list archived? Contact Gossamer Threads