System hang or extreme slow down - DMA wait failed

Greetings,

I too am intermittantly experiencing this error while recording from my PVR-250 every few days. Typical recordings are 2 or 3 half-hour or occasional 2.5 hour shows each day usually via svideo input. Though my system doesn't hang, from the point of the message, recordings are truncated/empty. The worst part is that I cannot count on my mythtv box to record programs reliably any more, and that just pleases the heck out of the girlfriend. :-( Reboots are the only thing that fixes it (haven't gotten to the point of unloading all the modules and dependancies). And with the SCCA Runoffs being broadcast soon on SPEED (look for me in the FM race on 12/4/2005!), I would be very sad to miss those recordings.

Unfortunately, many changes have occurred since it was reliable, including hardware and OS. Original setup was a Dell 2400 2.4GHz HT SMP-enabled running FC3 with ivtv 0.3.x and several kernel revisions (latest was 2.6.13-rc6). Recently upgraded to a faster box for smoother decoding of HDTV. In its previous life, the "new" box was reliable (FC3 64bit, WinXP gaming, video editing). It's a homebuilt with Asus A8V, Athlon64 3400+, FC4 2.6.13-1.1532_FC4 (non-SMP, regular 32-bit x86), variety of ivtv's attempted (now at 0.4.0), variety of PVR-250 firmwares attempted (now at 0x02050032). Got a slow-speed fan blowing on the cards (with a HD3000) so I don't think it is a heat issue (didn't have the fan at all in the Dell).

By noting the duration of the last short recording, I can pinpoint the /var/log/messages that occurred when the actual recording stopped. Example:

Nov 10 21:49:45 k8 kernel: ivtv0 warning: ENC: REG_DMAXFER 2 wait failed

Since 2.6.14-1.1637_FC4 kernel just came out and tomorrow's a holiday, I think I'll try getting the new kernel going just to try to eliminate that.

Given that it appears to be a DMA issue, any suggestions? Any chance it's a timing issue with the CONFIG_HZ=250 interrupt frequency?

Interrupts are reasonably spread out:

$ cat /proc/interrupts
           CPU0
  0:   21612381          XT-PIC  timer
  1:      32223          XT-PIC  i8042
  2:          0          XT-PIC  cascade
  4:          4          XT-PIC  lirc_serial
  5:    1349185          XT-PIC  VIA8237, ehci_hcd:usb1, ivtv0
  7:    2500673          XT-PIC  cx88[0], cx88[0]
  8:          1          XT-PIC  rtc
  9:          0          XT-PIC  acpi
 10:      21029          XT-PIC  libata, SysKonnect SK-98xx, eth1, uhci_hcd:usb4, uhci_hcd:usb5
 11:    2227364          XT-PIC  uhci_hcd:usb2, uhci_hcd:usb3, ohci1394, nvidia
 14:     408039          XT-PIC  ide0
 15:    4124993          XT-PIC  ide1
NMI:          0
ERR:          0

I'd be happy to work with the ivtv developers in any possible way.

Thanks! Chris Schanzle

I too have the same thing in my system, but it does not correspond to playback or recording. My system will be stable for about a day, then I will notice that it is "almost locked" I say almost because it is so slow that it takes 5 minutes to login. When I get in the CPU is idle and the disk access is minimal to none.

I'm running Knoppmyth R5A22 (Debian), I too just changed originally I ran fine with no issues but it was an FC3 install, same hardware.

Nov 14 14:36:04 mythnew kernel: ivtv0 warning: DMA_TO: REG_DMAXFER 2 wait failed Nov 14 14:37:04 mythnew kernel: ivtv0 warning: DMA_TO: REG_DMAXFER 2 wait failed Nov 14 14:38:07 mythnew kernel: ivtv0 warning: DMA_TO: REG_DMAXFER 2 wait failed Nov 14 14:39:06 mythnew kernel: ivtv0 warning: DMA_TO: REG_DMAXFER 2 wait failed Nov 14 14:40:09 mythnew kernel: ivtv0 warning: DMA_TO: REG_DMAXFER 2 wait failed

I would concur, but with different logs:

Nov 14 07:29:04 pthome kernel: ivtv0 warning: DEC: Sched Buffer end reached 0x02010000
Nov 14 07:29:04 pthome kernel: ivtv0 warning:  DEC: Mailbox 10: 0x00000000 0x02000000 0x02000000 0x02000000

and then, ivtv-dec takes 100% of CPU and the PVR (350) refuses to do anything

Just a follow-up on my previous comment #2 above. I moved the PVR-250 back to the original system (Dell Dimension 4600 P4 2.4GHz HT/SMP) and *copied* (not reinstall, but file-by-file copy to ensure both systems were running the same software/configs) the current FC4 (arch=i386, AMD64 3400+) install. PVR-250 works beautifully. So I'm quite certain the new box, an Asus A8V Rev. 2.0 board, is corrupting or just failing to pass data reliably on the PCI bus. In retrospect, it did seem to error more frequently when there was more disk activity as opposed to being idle, especially when the HD-3000 and the PVR-250 were recording simultaneously. I have mythtv doing commercial flagging concurrent with recording, which just adds to the disk activity while recording.

Haven't had the downtime window to upgrade the BIOS (I think I need Windows), but for a slightly different board ("A8V Deluxe") BIOS vers 1014 2005/07/25 includes "Fix WINTV-PVR-500 TV Card issue". Sounds like we're in the ballpark, if not still way out in left field.

Perhaps you guys are too struggling with (VIA?) chipset reliability issues?

Good luck!

Hi, all. Well, with the PVR-250 working perfectly in the Dell for the last month or so, I got around to updating the BIOS on the A8V Deluxe from 1008 to 1017 and moved the PVR-250 back. No joy. Further googling showed the BIOS fix solved a POST issue with the PVR-500.

So I bailed on the VIA chipset and bought an spiffy new nForce4 A8N-SLI Premium (yes, overkill, but I *like* the idea of the heatpipe on the northbridge and reviews say "everything works" with Linux). Sad to report I still have video and audio corruption issues. I'm not getting any huge, fatal errors (yet, but that could be just because I haven't run it long enough).

And to add insult to injury, the A8N supports only PCI Express, which means I needed to snag an NVIDIA 6600 (non-GT, just plain, big heatsink w/o fan). Not too far after which I realize I now don't have an 'xv' device any more, and I get "tearing" during panning of HDTV material. Enabling OpenGL sync slows playback to the point of pauses (with an AMD64 3400 taking 99%-ish CPU?).

I copied off/restored my 700GB of recordings from reiserfs to jfs. With the Reiserfs filesystem, deletions of 7GB recordings (typical for 1 hr of HDTV) caused IOBOUND errors. Interestingly, during my file restore or recording, the jfs filesystem makes per-second updates to the filesystem, whereas XFS would make 5-second 'burst' flushes. Now that I hear seeking every second during a recording, given the chance I think I might go XFS next time (thank goodness for gigabit ethernet and jumbo frames to make this happen < day).

I updated mythtv from svn 7567 to 8622.

Off to update ivtv from 0.4.0 to latest (0.4.2) "just because." And double-check my firmware releases...currently is reported as 0x02050032.

The saga continues...

i changed my system from K8VSE to A8V amd4600 X2

same trouble here.

but the system hangs up quite often (every 1/2-3h)

say: did the bios checking do any good?

i will update mine now.

Howdy. Just slapped together a Debian MythTV box. Custom-built 2.6.15.1 kernel w/ PVR-250 on an MSI KT400 Ultra (VIA chipset) mainboard. RAID-5 w/ 5 drives; one per channel, distributed among a SiI680 chipset, an HPT370 chipset, and the on-board VIA controller (VP_IDE chipset). ivtv version 0.4.2, tveeprom in 2.6.15.1 kernel. I just got this problem last night:

Feb 11 21:40:32 localhost kernel: ivtv0 warning: ENC: (0) DMA Error 0x0000000b

Feb 11 21:40:37 localhost kernel: ivtv0 warning: ENC: REG_DMAXFER 2 wait failed

Then every attempt to access the card (w/ no data):

Feb 12 03:40:46 localhost kernel: ivtv0 warning: ENC: DMA still Pending while stopping capture!

Feb 12 03:42:08 localhost kernel: ivtv0 warning: ENC: DMA still Pending while stopping capture!

I have other DMA issues too (this is after a reboot); I'm not sure how related they are:

Feb 12 08:43:42 localhost kernel: ivtv0 warning: ENC: (0) DMA Error 0x0000000b

Feb 12 08:43:54 localhost kernel: hdg: 0 bytes in FIFO

Feb 12 08:43:54 localhost kernel: hdg: timeout waiting for DMA

Feb 12 08:49:28 localhost kernel: hde: 0 bytes in FIFO

Feb 12 08:49:28 localhost kernel: hde: timeout waiting for DMA

Feb 12 08:49:42 localhost kernel: hdg: 0 bytes in FIFO

Feb 12 08:49:42 localhost kernel: hdg: timeout waiting for DMA

I think that hde and hdg are on the same controller card (still need to check that out). No filesystem corruption that I can detect. Note that the PVR-250 still works. If I get the REG_DMAXFER 2 wait failed message, then I am in la-la land (so far it's just happened once, but the system's only been up about two days now).

Note that I am using the tveeprom.ko module that's in the 2.6.15.1 kernel, since I also have an HD3000 (I haven't set it up and started using it yet).

Also note that I am a kernel hacker. I wrote the BSD Secure Levels LSM (in mainline) and the eCryptfs cryptographic filesystem (hopefully in mainline soon :-). Let me know if you would like me to do any debugging; I'll see what I can do in my spare time.

Thanks,

Mike

mike@halcrow.us

I reverted back to the AMD 3400+/A8V/AGP (xv-capable) system (the tearing and cpu requirements drove me nuts). Just the other day removed just about everything mythtv related and reinstalled the latest atrpms mythtv (0.19-124), new ivtv (0.4.3). Still corrupts. Posted again to the ivtv-driver list and Hans sent a first-cut patch to gather more info and hopefully handle the dma error case better. (Thanks, Hans!). No improvement just yet, but it seems like the dma retransmit error handling is the cause.

Clearly the frequency of the errors increases as the system is more "busy". I.e., if ivtv is the only thing running, there are few errors. With mythcommflag started when recording starts (mythtvsetup, general, 7th screen), play a prerecorded hdtv program, errors are logged just about every minute.

Meanwhile, I'm running ivtv with PIO mode (no dma) when I need to be sure the recordings are good. Edit ivtv-driver.h and change '#define IVTV_ENC_PIO' from 0 to 1. Note recording CPU utilization will go to about 45% (~100-fold increase) and you probably won't be able to view hdtv recordings.

Well I just lost a hard drive an upgraded to Knoppmyth R5B7, I was hoping that the failing drive mught be my problem, or something in my build but, I have hangs/freezes again.

These don't kill the box, they do recover in a few minutes, I haven't had to reboot. What's weird is the box was sitting idle, just serving up NFS to a frontend watching a video. right when I exited the video on the frontend the backend froze. Took about 3 minutes and now it's responsive again for a bit then the freezing continues.

Will have to bookmark this page and see where it goes. If anyone wants some particulars about my hardware/settings/software I will try to contribute what I've got.

Unfortunately reading about other system swaps doesn't make me thik I have many options for fixing this soon.

Did anyone replace memory?

Yep this system is now utter junk. Using Knoppmyth R5B7 that seems to use IVTV 0.4.3 and the hangs are worse than ever sometimes the systems staying up for only a few minutes.

Tried moving the card on the PCI bus etc. Since others have noted problems after changing motherboards, RAM etc I'm not sure what to think. Maybe it's a bad card? Worked for a couple years flawlessly.

Don't know if it matters that I am also using the TVOUT features.

Apr  4 18:51:23 mythtv kernel: ivtv:  ==================== START INIT IVTV ====================
Apr  4 18:51:23 mythtv kernel: ivtv:  version 0.4.3 (tagged release) loading
Apr  4 18:51:23 mythtv kernel: ivtv:  Linux version: 2.6.15-chw-2 SMP preempt 586 gcc-3.3
Apr  4 18:51:23 mythtv kernel: ivtv:  In case of problems please include the debug info between
Apr  4 18:51:23 mythtv kernel: ivtv:  the START INIT IVTV and END INIT IVTV lines, along with
Apr  4 18:51:23 mythtv kernel: ivtv:  any module options, when mailing the ivtv-users mailinglist.
Apr  4 18:51:23 mythtv kernel: ivtv0: Autodetected WinTV PVR 350 card (cx23415 based)
Apr  4 18:51:23 mythtv kernel: ACPI: PCI Interrupt 0000:02:0a.0[A] -> GSI 22 (level, low) -> IRQ 21
Apr  4 18:51:23 mythtv kernel: ivtv0: Unreasonably low latency timer, setting to 64 (was 32)
Apr  4 18:51:23 mythtv kernel: ivtv0: i2c attach to card #0 ok [client=tveeprom, addr=50]
Apr  4 18:51:23 mythtv kernel: tuner 1-0061: chip found @ 0xc2 (ivtv i2c driver #0)
Apr  4 18:51:23 mythtv kernel: ivtv0: i2c attach to card #0 ok [client=(tuner unset), addr=61]
Apr  4 18:51:23 mythtv kernel: saa7115 1-0021: saa7115 found @ 0x42 (ivtv i2c driver #0)
Apr  4 18:51:23 mythtv kernel: ivtv0: i2c attach to card #0 ok [client=saa7115, addr=21]
Apr  4 18:51:23 mythtv kernel: saa7127 1-0044: saa7127 found @ 0x88 (ivtv i2c driver #0)
Apr  4 18:51:23 mythtv kernel: ivtv0: i2c attach to card #0 ok [client=saa7127, addr=44]
Apr  4 18:51:23 mythtv kernel: msp3400 1-0040: chip=MSP4448G-A2 +nicam +simple +simpler +radio mode=simpler
Apr  4 18:51:23 mythtv kernel: msp3400 1-0040: msp34xxg daemon started
Apr  4 18:51:23 mythtv kernel: ivtv0: i2c attach to card #0 ok [client=MSP4448G-A2, addr=40]
Apr  4 18:51:23 mythtv kernel: tveeprom 1-0050: Hauppauge model 48132, rev J323, serial# 7065187
Apr  4 18:51:23 mythtv kernel: tveeprom 1-0050: tuner model is Philips FM1236 (idx 23, type 2)
Apr  4 18:51:23 mythtv kernel: tveeprom 1-0050: TV standards NTSC(M) (eeprom 0x08)
Apr  4 18:51:23 mythtv kernel: tveeprom 1-0050: audio processor is MSP4448 (idx 27)
Apr  4 18:51:23 mythtv kernel: tveeprom 1-0050: decoder processor is SAA7115 (idx 19)
Apr  4 18:51:23 mythtv kernel: tveeprom 1-0050: has radio, has IR remote
Apr  4 18:51:24 mythtv kernel: ivtv0: loaded v4l-cx2341x-enc.fw firmware (262144 bytes)
Apr  4 18:51:24 mythtv kernel: ivtv0: loaded v4l-cx2341x-dec.fw firmware (262144 bytes)
Apr  4 18:51:24 mythtv kernel: ivtv0: Encoder revision: 0x02050032
Apr  4 18:51:24 mythtv kernel: ivtv0: Decoder revision: 0x02020023
Apr  4 18:51:24 mythtv kernel: ivtv0: Allocate DMA encoder MPEG stream: 128 x 32768 buffers (4096KB total)
Apr  4 18:51:24 mythtv kernel: ivtv0: Allocate DMA encoder YUV stream: 194 x 10800 buffers (2048KB total)
Apr  4 18:51:24 mythtv kernel: ivtv0: Allocate DMA encoder VBI stream: 120 x 17472 buffers (2048KB total)
Apr  4 18:51:24 mythtv kernel: ivtv0: Allocate DMA encoder PCM audio stream: 455 x 4608 buffers (2048KB total)
Apr  4 18:51:24 mythtv kernel: ivtv0: Create encoder radio stream
Apr  4 18:51:24 mythtv kernel: ivtv0: Allocate DMA decoder MPEG stream: 16 x 65536 buffers (1024KB total)
Apr  4 18:51:24 mythtv kernel: ivtv0: Allocate DMA decoder VBI stream: 512 x 2048 buffers (1024KB total)
Apr  4 18:51:24 mythtv kernel: ivtv0: Create decoder VOUT stream
Apr  4 18:51:24 mythtv kernel: ivtv0: Allocate DMA decoder YUV stream: 24 x 43200 buffers (1024KB total)
Apr  4 18:51:25 mythtv kernel: ivtv0: loaded v4l-cx2341x-init.mpg firmware (155648 bytes)
Apr  4 18:51:25 mythtv kernel: tuner 1-0061: type set to 2 (Philips NTSC (FI1236,FM1236 and compatibles))
Apr  4 18:51:25 mythtv kernel: ivtv0: Initialized WinTV PVR 350, card #0
Apr  4 18:51:25 mythtv kernel: ivtv:  ====================  END INIT IVTV  ====================
Apr  4 18:51:25 mythtv kernel: ivtv0-osd: Framebuffer module loaded (attached to ivtv card id 0)
Apr  4 18:51:25 mythtv kernel: ivtv0-osd: screen coords: [0 0] -> [720 480]
Apr  4 18:51:25 mythtv kernel: ivtv0-osd: original global alpha = 208
Apr  4 18:51:25 mythtv kernel: ivtv0-osd: current OSD state = 39
Apr  4 18:51:25 mythtv kernel: ivtv0-osd: new global alpha = 208 (1 255 0)
Apr  4 18:51:25 mythtv kernel: ivtv0-osd: framebuffer at 0xe1510000, mapped to 0xe1d10000, size 1665k
Apr  4 18:51:25 mythtv kernel: ivtv0-osd: mode is 720x480x32, linelength=2880
Apr  4 18:51:25 mythtv kernel: ivtv0-osd warning: ivtvfb_set_par
Apr  4 18:51:25 mythtv kernel: Console: switching to colour frame buffer device 90x30
Apr  4 18:51:25 mythtv kernel: ivtv0-osd: fb0: cx23415 TV out frame buffer device
Apr  4 18:51:25 mythtv kernel: ivtv0: Allocate DMA decoder OSD stream: 16 x 65536 buffers (1024KB total)
Apr  4 18:51:25 mythtv kernel: cx2388x v4l2 driver version 0.0.5 loaded
Apr  4 18:51:25 mythtv kernel: lirc_i2c: chip found @ 0x18 (Hauppauge IR)
Apr  4 18:51:25 mythtv kernel: ivtv0: i2c attach to card #0 ok [client=Hauppauge IR, addr=18]
Apr  4 18:51:25 mythtv kernel: lirc_dev: lirc_register_plugin: sample_rate: 10

My latest hangs don't mention any IVTV errors like in the past. Is everyone using Knoppmyth? maybe it's them....

I am experiencing the same issue but only when I attempt use live TV in MythTV (SVN 19.fixes). IVTV 0.4.4, Linux 2.6.15-ck7... I just tried watching some live TV while nothing was recording on the PVR-500 and I noticed the occasional prebuffer pause at first and then everything was fine. However, when I stopped watching live TV it locked up for a couple of minutes:

ivtv0 warning: ENC: REG_DMAXFER 2 wait failed ivtv1 warning: ENC: REG_DMAXFER 2 wait failed ivtv1 warning: ENC: REG_DMAXFER 2 wait failed ivtv0 warning: ENC: REG_DMAXFER 2 wait failed ivtv0 warning: ENC: REG_DMAXFER 2 wait failed ivtv1 warning: ENC: REG_DMAXFER 2 wait failed ivtv0 warning: ENC: REG_DMAXFER 2 wait failed ivtv1 warning: ENC: REG_DMAXFER 2 wait failed ivtv1 warning: ENC: REG_DMAXFER 2 wait failed ivtv0 warning: ENC: REG_DMAXFER 2 wait failed ivtv0 warning: ENC: REG_DMAXFER 2 wait failed ivtv0 warning: ENC: REG_DMAXFER 2 wait failed ivtv1 warning: ENC: REG_DMAXFER 2 wait failed ivtv0 warning: ENC: REG_DMAXFER 2 wait failed ivtv0 warning: ENC: REG_DMAXFER 2 wait failed ivtv1 warning: ENC: REG_DMAXFER 2 wait failed ivtv0: All encoder MPEG stream buffers are full. Dropping data. ivtv0: Cause: the application is not reading fast enough. ivtv1: All encoder MPEG stream buffers are full. Dropping data. ivtv1: Cause: the application is not reading fast enough. ivtv1: All encoder MPEG stream buffers are full. Dropping data. ivtv1: Cause: the application is not reading fast enough. ivtv1 warning: ENC: REG_DMAXFER 2 wait failed ivtv1 warning: ENC: REG_DMAXFER 2 wait failed ivtv1 warning: ENC: REG_DMAXFER 2 wait failed ivtv1 warning: ENC: REG_DMAXFER 2 wait failed ivtv1 warning: ENC: REG_DMAXFER 2 wait failed ivtv1 warning: ENC: REG_DMAXFER 2 wait failed ivtv1 warning: ENC: REG_DMAXFER 2 wait failed ivtv1 warning: ENC: REG_DMAXFER 2 wait failed ivtv1 warning: ENC: REG_DMAXFER 2 wait failed ivtv1 warning: ENC: REG_DMAXFER 2 wait failed ivtv1 warning: ENC: REG_DMAXFER 2 wait failed ivtv1 warning: ENC: REG_DMAXFER 2 wait failed ivtv1 warning: ENC: REG_DMAXFER 2 wait failed ivtv1 warning: ENC: REG_DMAXFER 2 wait failed ivtv1 warning: ENC: REG_DMAXFER 2 wait failed ivtv1 warning: ENC: REG_DMAXFER 2 wait failed ivtv1 warning: ENC: REG_DMAXFER 2 wait failed

I'm surprised both ivtv0 and ivtv1 are complaining when I should theoretically only have one in use (both channels were idle).

My problem is now gone. I got fed up and switched motherborads and CPU. Now running on an ASUS nforce board with an AMD.

System hasn't frozen or slowed in 4 days

i'm getting the following error after a period of using mythtv. Apr 25 20:41:56 junior kernel: ivtv0 warning: ENC: (0) DMA Error 0x0000000b

shortly thereafter the system does the hang/slowdown. only rebooting seems to fix it. i do not see any other ivtv related errors in the log between that one and the reboot.

this seems to be happening only relatively recently. ?perhaps related to the 15 april firmware rpms?

fc4. amd 2200+ one hauppage 250.

ivtv rpms and install dates: ivtv-kmdl-2.6.16-1.2096_FC4-0.6.1-109.rhfc4.at 2006 Apr 24 ivtv-firmware-dec-2.02.023-9.at 2006 Apr 15 ivtv-firmware-enc-2.04.024-9.at 2006 Apr 15 ivtv-0.6.1-109.rhfc4.at 2006 Mar 30 ivtv-kernheaders-0.6.1-109.rhfc4.at 2006 Mar 30 ivtv-kmdl-2.6.15-1.1833_FC4-0.4.4-107.rhfc4.at 2006 Mar 30 ivtv-kmdl-2.6.16-1.2069_FC4-0.6.1-109.rhfc4.at 2006 Mar 30 ivtv-firmware-audio-0.0.1-5.at 2006 Mar 11 perl-Video-ivtv-0.13-7.rhfc4.at 2006 Mar 07

I'm also getting the hange and have to reboot. Same versions as above. Seems to happen most often when I change channels or a recording ends.

ivtv0 warning: ENC: (0) DMA Error 0x0000000b ivtv0 warning: ENC: REG_DMAXFER 2 wait failed ivtv0 warning: ENC: DMA still Pending while stopping capture!

amd x2 4400 PVR350.

The system was stable on FC4 as of 2-3 weeks ago, although I probably didn't have the latest packages at that time. I reinstalled to FC5 about 2 weeks ago and the problems have started since.

I am having the same problem as everyone else. Asus A8n-SLI. Gentoo system running MythTV 0.19, IVTV 0.6.2, kernel 2.6.16, 2 PVR-150s and an HD3000.

Had the problem with previous version of MythTV, and with IVTV 0.4.2.

It seems like it is related to out motherboards. I will be monitoring this page.

Information that may help the developers... --- Linux video capture interface: v1.00 ivtv: ==================== START INIT IVTV ==================== ivtv: version 0.6.2 (tagged release) loading ivtv: Linux version: 2.6.16-gentoo-r7 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 Link [APC1] enabled at IRQ 16 GSI 21 sharing vector 0xD9 and IRQ 21 ACPI: PCI Interrupt 0000:05:06.0[A] -> Link [APC1] -> GSI 16 (level, low) -> IRQ 21 ivtv0: Unreasonably low latency timer, setting to 64 (was 32) tveeprom 2-0050: Hauppauge model 26052, rev C199, serial# 8165771 tveeprom 2-0050: tuner model is TCL 2002N 5H (idx 99, type 50) tveeprom 2-0050: TV standards NTSC(M) (eeprom 0x08) tveeprom 2-0050: audio processor is CX25843 (idx 37) tveeprom 2-0050: decoder processor is CX25843 (idx 30) tveeprom 2-0050: has no radio, has IR remote tuner 2-0061: chip found @ 0xc2 (ivtv i2c driver #0) cx25840 2-0044: cx25843-23 found @ 0x88 (ivtv i2c driver #0) cx25840 2-0044: loaded v4l-cx25840.fw firmware (14264 bytes) wm8775 2-001b: 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 2-0061: type set to 50 (TCL 2002N) ivtv0: Initialized Hauppauge WinTV PVR-150, card #0 ivtv: ====================== NEXT CARD ====================== ivtv1: Autodetected Hauppauge WinTV PVR-150 card (cx23416 based) ACPI: PCI Interrupt 0000:05:08.0[A] -> Link [APC3] -> GSI 18 (level, low) -> IRQ 20 ivtv1: Unreasonably low latency timer, setting to 64 (was 32) tuner 3-0061: chip found @ 0xc2 (ivtv i2c driver #1) cx25840 3-0044: cx25841-23 found @ 0x88 (ivtv i2c driver #1) cx25840 3-0044: loaded v4l-cx25840.fw firmware (14264 bytes) wm8775 3-001b: chip found @ 0x36 (ivtv i2c driver #1) tveeprom 3-0050: Hauppauge model 26032, rev C199, serial# 7841418 tveeprom 3-0050: tuner model is TCL 2002N 5H (idx 99, type 50) tveeprom 3-0050: TV standards NTSC(M) (eeprom 0x08) tveeprom 3-0050: audio processor is CX25841 (idx 35) tveeprom 3-0050: decoder processor is CX25841 (idx 28) tveeprom 3-0050: has no radio, has IR remote ivtv1: loaded v4l-cx2341x-enc.fw firmware (262144 bytes) ivtv1: Encoder revision: 0x02050032 ivtv1: Allocate DMA encoder MPEG stream: 128 x 32768 buffers (4096KB total) ivtv1: Allocate DMA encoder YUV stream: 194 x 10800 buffers (2048KB total) ivtv1: Allocate DMA encoder VBI stream: 120 x 17472 buffers (2048KB total) ivtv1: Allocate DMA encoder PCM audio stream: 455 x 4608 buffers (2048KB total) tuner 3-0061: type set to 50 (TCL 2002N) ivtv1: Initialized Hauppauge WinTV PVR-150, card #1 ivtv: ==================== END INIT IVTV ==================== ---

Errors: i2c_adapter i2c-5: SMBus Quick command not supported, can't probe for chips i2c_adapter i2c-6: SMBus Quick command not supported, can't probe for chips i2c_adapter i2c-7: SMBus Quick command not supported, can't probe for chips Losing some ticks... checking if CPU frequency changed. /var/tmp/portage/lirc-0.8.0/work/lirc-0.8.0/drivers/lirc_imon/lirc_imon.c: IR port opened ivtv0 warning: ENC: (0) DMA Error 0x0000000b ivtv0 warning: ENC: (0) DMA Error 0x0000000b ivtv0 warning: ENC: (0) DMA Error 0x0000000b ivtv0 warning: ENC: (0) DMA Error 0x0000000b ivtv0 warning: ENC: (0) DMA Error 0x0000000b ivtv0 warning: ENC: (0) DMA Error 0x0000000b ivtv0 warning: ENC: (0) DMA Error 0x0000000b ivtv0 warning: ENC: (0) DMA Error 0x0000000b ivtv0 warning: ENC: (0) DMA Error 0x0000000b ivtv0 warning: ENC: (0) DMA Error 0x0000000b ivtv0 warning: ENC: (0) DMA Error 0x0000000b ivtv0 warning: ENC: (0) DMA Error 0x0000000b ivtv0 warning: ENC: (0) DMA Error 0x0000000b ivtv0 warning: ENC: (0) DMA Error 0x0000000b ivtv0 warning: ENC: (0) DMA Error 0x0000000b ivtv0 warning: ENC: (0) DMA Error 0x0000000b ivtv0 warning: ENC: (0) DMA Error 0x0000000b ivtv0 warning: ENC: (0) DMA Error 0x0000000b ivtv0 warning: ENC: (0) DMA Error 0x0000000b ivtv0 warning: ENC: REG_DMAXFER 2 wait failed ivtv0 warning: ENC: DMA still Pending while stopping capture! ivtv0 warning: ENC: DMA still Pending while stopping capture! ivtv0 warning: ENC: DMA still Pending while stopping capture!

I am the one who posted the comment above. My drives are SATA. I have enabled AHCI and things seem to be getting a little better.

The errors are still coming up, but it seems like they are more infrequent. Only time will tell if this works.

I put up a little bit about this on my blog so that anyone from here can contact me if there's anything I should try. I will also be checking back here from time to time to see if there are any new developments.

http://www.morph3ous.net/2006/05/28/ivtv-driver-error/

Never mind, I just got an:

ivtv0 warning: ENC: REG_DMAXFER 2 wait failed

The video froze...

System is still responsive. (I've never had a problem with system responsiveness when this problem crops up.)

Milestone Release 0.4.2 deleted

Milestone Release 0.10.0 deleted

Turns out that the DMA engine of the cx23415/6 is buggy and can crash, depending on your hardware/software environment. There isn't much that can be done about it.

Some guidelines:

- don't use RAID - disable any CPU frequency changer deamons - Intel chipsets seem to perform better than nVidia or ATI chipsets.