Linux OEM edition on Red Hat / CentOS

[Gulganei]

I'm trying to get the Linux OEM edition of Sage running on my primary server which is running CentOS 4 (Red Hat EL 4). I first installed the Gentoo distribution that came on the ISO onto a spare hard drive on my server box, activated Sage and confirmed that Sage was working correctly with my MediaMVP extender. I then proceeded to move the relevant directories in /opt to my server's main hard drive to try and get it running under CentOS. I was able to get the ivtv 0.4.2 module built and recognizing the PVR-150 in the machine. I then ran Sage which seems to work properly until I tried to set up the tuner card which gives me the following error:

ERROR (528,0x16): There was a problem setting up software audio compression.

I also noticed these messages in dmesg:

kernel: ioctl32(java:4457): Unknown cmd fd(32) cmd(80685600){00} arg(08389dc8) on /dev/video0
kernel: ioctl32(java:4457): Unknown cmd fd(31) cmd(40085618){00} arg(e360fb08) on /dev/video0
kernel: ioctl32(java:4457): Unknown cmd fd(33) cmd(80685600){00} arg(083266d0) on /dev/video0
kernel: ioctl32(java:4457): Unknown cmd fd(31) cmd(40085618){00} arg(e360faf4) on /dev/video0

Could these issues be caused by the version of the ivtv module I'm using (0.4.2 vs. 0.5.1 that the Gentoo distro uses)?

Any help or suggestions would be appreciated!

->Dan

[SHS]

try del the Sage.properties file and reboot and start over

[_Demo_]

You will need to install ivtv 0.5.1

_Demo_

[Gulganei]

So, ivtv 0.5.1 requires a newer kernel than comes with Centos 4 in order to compile. I upgraded my kernel to 2.6.15 and got ivtv 0.5.1 to compile. Unfortunately, it seems that this version of ivtv has issues with 64-bit flavors of Linux (my server is an Athlon 64). I was getting the dreaded "Error -12" that other ivtv users have been reporting in various forums. The problem is fixed in later versions, and I was able to get ivtv 0.7.0 working correctly and did a successful test capture by running "cat /dev/video0 > foo.mpg". However, Sage still complains about software audio compression when run with this driver version.

Oh, and I tried SHS's suggestion to no avail.

Any other suggestions?

->Dan

[_Demo_]

Could you attach a debug log.

Thanks

_Demo_

[Gulganei]

Sure, here is a log file after deleting Wiz.* and clearing out Sage.properties to include only a debug_logging statement.

->Dan

[_Demo_]

It isn't able to identify your card, probably because you are not using the same ivtv version...

_Demo_

[Gulganei]

This is what is in the system log when I install the IVTV module:

ivtv: ==================== START INIT IVTV ====================
ivtv: version 0.7.0 (development revision 3143) loading
ivtv: Linux version: 2.6.9-22.0.2.EL gcc-3.4
ivtv: In case of problems please include the debug info between
ivtv: the START INIT IVTV and END INIT IVTV lines, along with
ivtv: any module options, when mailing the ivtv-users mailinglist.
ivtv0: Autodetected Hauppauge WinTV PVR-150 card (cx23416 based)
ACPI: PCI interrupt 0000:00:07.0[A] -> GSI 18 (level, low) -> IRQ 169
ivtv0: Unreasonably low latency timer, setting to 64 (was 32)
tveeprom 0-0050: Hauppauge model 26032, rev C199, serial# 8163262
tveeprom 0-0050: tuner model is TCL 2002N 5H (idx 99, type 50)
tveeprom 0-0050: TV standards NTSC(M) (eeprom 0x08)
tveeprom 0-0050: audio processor is CX25841 (idx 35)
tveeprom 0-0050: decoder processor is CX25841 (idx 28)
tveeprom 0-0050: has no radio, has IR remote
tuner 0-0061: chip found @ 0xc2 (ivtv i2c driver #0)
cx25840 0-0044: cx25841-23 found @ 0x88 (ivtv i2c driver #0)
cx25840 0-0044: loaded v4l-cx25840.fw firmware (14264 bytes)
wm8775 0-001b: chip found @ 0x36 (ivtv i2c driver #0)
ivtv0: loaded v4l-cx2341x-enc.fw firmware (262144 bytes)
ivtv0: Encoder revision: 0x02050032
ivtv0: Allocate DMA encoder MPEG stream: 128 x 32768 buffers (4096KB total)
ivtv0: Allocate DMA encoder YUV stream: 194 x 10800 buffers (2048KB total)
ivtv0: Allocate DMA encoder VBI stream: 120 x 17472 buffers (2048KB total)
ivtv0: Allocate DMA encoder PCM audio stream: 455 x 4608 buffers (2048KB total)
tuner 0-0061: type set to 50 (TCL 2002N)
ivtv0: Initialized Hauppauge WinTV PVR-150, card #0
ivtv: ==================== END INIT IVTV ====================

[AboveUnrefined]

Unfortunately, it seems that this version of ivtv has issues with 64-bit flavors of Linux

Are you running in pure 64 bit mode? I'm curious to know if you need 64 bit drivers to use the devices you're trying to use or not... If you're not in 64 bit mode (it doesn't sound like you are...) I don't see why ivtv 0.5.1 would give you problems...

Is there anyway you can maybe plug in the device you're using into a script somewhere instead of autodetecting it?

[Gulganei]

I installed the x64 version of the kernel and all RPMs, so I assume I'm in full 64-bit mode. The 0.5.1 drivers compile but simply don't work correctly -- it gives me an error about loading firmware at the end of its startup (even though the previous log messages all indicate that the firmware was loaded successfully) and ends with an error -12.

[Dekard]

I'm watching this thread since I'm considering doing a similiar thing. I'm interested in setting up Sagetv on a Clark Connect server I'm running. If anyone here has tried that I'd be interested in hearing about your results.

[MrD]

I am seeing the exact same issues using both version 0.4.1 and 0.6.2 of the ivtv driver.

This is with kernels 2.6.11 and 2.6.16 on fedora core 4 using an AMD64 3200+ in x86_64 mode.

I am wondering if it is 64bit related.

ioctl32(java:2963): Unknown cmd fd(13) cmd(40085618){00} arg(f0dd9d54) on /dev/video0
ioctl32(java:3255): Unknown cmd fd(25) cmd(80685600){00} arg(08372f58) on /dev/video0
ioctl32(java:3255): Unknown cmd fd(13) cmd(40085618){00} arg(f0dd9bd4) on /dev/video0

ivtv:  ==================== START INIT IVTV ====================
ivtv:  version 0.6.2 (development snapshot compiled on Mon May 22 01:20:12 2006) loading
ivtv:  Linux version: 2.6.16-1.2111_FC4 gcc-4.0
ivtv:  In case of problems please include the debug info between
ivtv:  the START INIT IVTV and END INIT IVTV lines, along with
ivtv:  any module options, when mailing the ivtv-users mailinglist.
ivtv0: Autodetected Hauppauge WinTV PVR-150 card (cx23416 based)
GSI 18 sharing vector 0xB9 and IRQ 18
ACPI: PCI Interrupt 0000:00:08.0[A] -> GSI 16 (level, low) -> IRQ 18
ivtv0: Unreasonably low latency timer, setting to 64 (was 32)
tveeprom 1-0050: Hauppauge model 26132, rev F0B2, serial# 9398641
tveeprom 1-0050: tuner model is TCL M2523_5N_E (idx 112, type 4)
tveeprom 1-0050: TV standards NTSC(M) (eeprom 0x08)
tveeprom 1-0050: audio processor is CX25841 (idx 35)
tveeprom 1-0050: decoder processor is CX25841 (idx 28)
tveeprom 1-0050: has no radio, has IR remote
tuner 1-0061: chip found @ 0xc2 (ivtv i2c driver #0)
cx25840 1-0044: cx25841-23 found @ 0x88 (ivtv i2c driver #0)
cx25840 1-0044: loaded v4l-cx25840.fw firmware (16382 bytes)
wm8775 1-001b: chip found @ 0x36 (ivtv i2c driver #0)
ivtv0: loaded v4l-cx2341x-enc.fw firmware (262144 bytes)
ivtv0: Encoder revision: 0x02050032
ivtv0: Allocate DMA encoder MPEG stream: 128 x 32768 buffers (4096KB total)
ivtv0: Allocate DMA encoder YUV stream: 194 x 10800 buffers (2048KB total)
ivtv0: Allocate DMA encoder VBI stream: 120 x 17472 buffers (2048KB total)
ivtv0: Allocate DMA encoder PCM audio stream: 455 x 4608 buffers (2048KB total)
tuner 1-0061: tuner type not set
tuner 1-0061: tuner type not set
ivtv0: Initialized Hauppauge WinTV PVR-150, card #0
ivtv:  ====================  END INIT IVTV  ====================

Fri 5/26 16:31:30.482 Loaded
Fri 5/26 16:31:30.487 Schedule skipping encoder video0because it failed to load: sage.EncodingException: ERROR (528,0x16): There was a problem setting up software audio compression.

-MrD

[_Demo_]

This error means it has failed setting the video standard. We have not done tests on 64 bits kernels yet.

_Demo_

[MrD]

Originally Posted by _Demo_

This error means it has failed setting the video standard. We have not done tests on 64 bits kernels yet.

_Demo_

Yep. Using fc5 32bit card setup works.

Is there a roadmap for x86_64 functionality?

Thanks!

-MrD

[_Demo_]

There isn't a roadmap since we don't currently have a 64 bit system to try things on but if you can point us to how to fix that issue we could try to implement it.

_Demo_

[jreichen]

Originally Posted by _Demo_

There isn't a roadmap since I don't currently have a 64 bit system to try things on but if you can point us to how to fix that issue I could try to implement it.

_Demo_

I installed ivtv 0.6.2 from the ivtv web site and I pretty much have the same dmesg as MrD with everything looking OK. Here's the info from /opt/sagetv/server/sagetv_0.txt

Sat 5/27 8:01:16.692 Schedule skipping encoder video3because it failed to load: sage.EncodingException: ERROR (143,0x13): There was a problem setting up software audio compression.
Sat 5/27 8:01:16.707 Loaded
Sat 5/27 8:01:16.712 Schedule skipping encoder video2because it failed to load: sage.EncodingException: ERROR (528,0x16): There was a problem setting up software audio compression.
Sat 5/27 8:01:16.712 Schedule skipping encoder video0because it failed to load: sage.EncodingException: ERROR (528,0x16): There was a problem setting up software audio compression.
Sat 5/27 8:01:17.099 Starting webserver with args: [-p, 8080, -root, webserver, -t, throttles.properties, -s, servlets.properties, -r, realms.properties, -a, aliases.properties, -nohup]
Sat 5/27 8:01:17.148 Acme httpd blackbird:8080 listening.
Sat 5/27 8:01:40.862 Error in MediaServerConnection of :java.io.EOFException
java.io.EOFException
Sat 5/27 8:01:40.862    at sage.MediaServer$a.a(Unknown Source)
Sat 5/27 8:01:40.862    at sage.MediaServer$a.run(Unknown Source)
Sat 5/27 8:01:40.862    at java.lang.Thread.run(Unknown Source)
Sat 5/27 8:01:44.068 locale = en
Sat 5/27 8:01:49.938 Schedule skipping encoder video3because it failed to load: sage.EncodingException: ERROR (143,0x13): There was a problem setting up software audio compression.
Sat 5/27 8:02:31.566 Schedule skipping encoder video3because it failed to load: sage.EncodingException: ERROR (143,0x13): There was a problem setting up software audio compression.
Sat 5/27 8:05:05.361 Closing down MiniPlayer
java.lang.NullPointerException
java.lang.NullPointerException
Sat 5/27 8:05:05.362    at sage.bq.byte(Unknown Source)
Sat 5/27 8:05:05.362    at sage.bq.free(Unknown Source)
Sat 5/27 8:05:05.362    at sage.VideoFrame.try(Unknown Source)
Sat 5/27 8:05:05.362    at sage.VideoFrame.a(Unknown Source)
Sat 5/27 8:05:05.362    at sage.VideoFrame.run(Unknown Source)
Sat 5/27 8:05:05.362    at java.lang.Thread.run(Unknown Source)
Sat 5/27 8:05:05.385 Schedule skipping encoder video3because it failed to load: sage.EncodingException: ERROR (143,0x13): There was a problem setting up software audio compression.
Sat 5/27 8:05:12.982 Schedule skipping encoder video3because it failed to load: sage.EncodingException: ERROR (143,0x13): There was a problem setting up software audio compression.
Sat 5/27 8:05:56.701 Schedule skipping encoder video3because it failed to load: sage.EncodingException: ERROR (143,0x13): There was a problem setting up software audio compression.
Sat 5/27 8:10:01.274 Schedule skipping encoder video3because it failed to load: sage.EncodingException: ERROR (143,0x13): There was a problem setting up software audio compression.
java.io.EOFException
Sat 5/27 8:10:34.620    at sage.bc$a.run(Unknown Source)

[MrD]

Originally Posted by jreichen

I installed ivtv 0.6.2 from the ivtv web site and I pretty much have the same dmesg as MrD with everything looking OK. Here's the info from /opt/sagetv/server/sagetv_0.txt

That's interesting that you get a stack trace... what JDK are you using am64 or x86?

-MrD

[MrD]

Originally Posted by _Demo_

There isn't a roadmap since we don't currently have a 64 bit system to try things on but if you can point us to how to fix that issue we could try to implement it.

_Demo_

Cool. The only way I can think to help (without having the code) would be to add logging statements at each transition, (i.e. java, JNI library, prior to ioctl code), to make sure the parameters generated in the Java layer match watch is getting passed to the hardware layer.

It would also be interested to see if the ioctl hardware queries are returning the same information back to sage as the 32bit version (i.e. more logging).

I would be willing to run troubleshooting on both systems given the additional logging.

-MrD

[jreichen]

Originally Posted by MrD

That's interesting that you get a stack trace... what JDK are you using am64 or x86?

-MrD

x86

[Gulganei]

Have the problems with 64-bit still not been resolved? I tried the latest SageTV 6 and still get the same errors.