[bjames]

Hi everyone,

For some reason I cannot for the life of me record video to hard drive using Media Recorder. I have the Octane Digital video option installed and trying to capture video from input 1 from a Sony DSR-1800 playing a DVcam tape. The video displays inside Media recorder just fine and on my NTSC monitor. No matter what setting I use (uncompressed, RGB, SGI video, Movie etc) 5 seconds after I click on the red record button I get this error "MVR VL error for Source 1" and fails to create a file. I have tried recording to my RAID drives, default tmp directory, home directory and even logged into the other accounts (including root) and still get the exact same error. In the system conole, I get the following message;

WARNING: srvVgiIDMA: FAILED to get lowmen pool 13
WARNING: dms lomem: DMA page 0x29e97 too high - set SRV_LOMEM_POOL

The system console returned the message above every time I try to capture video after the error message appears in Media recorder.

Any help would be great.. thanks

---

OK.. after some research I found there may be an issue with too much RAM ( greater than 1.5gigs). So to test the theory, I removed 3gigs from memory to olny leave 1 gig of ram (2 x 512mb sticks in bank 1). What do you know, all the erros went away and I can record directly to my RAID drive uncompressed without a single dropper frame.

So now my question is, am I stuck with only using 1 gig ram or is there an environment setting that will allow me to have Media record only 'see' 1 gig RAM out of my 4 gigs?

[chulofiasco]

Hi everyone,

For some reason I cannot for the life of me record video to hard drive using Media Recorder.  I have the Octane Digital video option installed and trying to capture video from input 1 from a Sony DSR-1800 playing a DVcam tape.  The video displays inside Media recorder just fine and on my NTSC monitor.  No matter what setting I use (uncompressed, RGB, SGI video, Movie etc) 5 seconds after I click on the red record button I get this error "MVR VL error for Source 1" and fails to create a file.  I have tried recording to my RAID drives, default tmp directory, home directory and even logged into the other accounts (including root) and still get the exact same error.  In the system conole, I get the following message;

WARNING: srvVgiIDMA: FAILED to get lowmen pool 13
WARNING: dms lomem: DMA page 0x29e97 too high - set SRV_LOMEM_POOL

The system console returned the message above every time I try to capture video after the error message appears in Media recorder.

Any help would be great.. thanks

---

OK..  after some research I found there may be an issue with too much RAM ( greater than 1.5gigs).  So to test the theory, I removed 3gigs from memory to olny leave 1 gig of ram (2 x 512mb sticks in bank 1).  What do you know, all the erros went away and I can record directly to my RAID drive uncompressed without a single dropper frame.

So now my question is, am I stuck with only using 1 gig ram or is there an environment setting that will allow me to have Media record only 'see' 1 gig RAM out of my 4 gigs?

how did you make out with this?

[gijoe77]

who knew having too much RAM was a bad thing...

[indigofan]

There is some documentation for systune related to Video capture and playback, will see if I can find it... Shouldn't have to remove RAM to capture video.

[chulofiasco]

There is some documentation for systune related to Video capture and playback, will see if I can find it... Shouldn't have to remove RAM to capture video.

Oh please, do see if you can find it.  What a trippy experience.  

I pulled a 512 mem kit and replaced with 256 to see if 1.75 gigs of ram would fly, and the computer said no ma'am.

It is very strange.

[weblacky]

Because this might affect me someday as well I'd like to know the answer. Looking around I only found one old post that talks about changes required to prevent something like this and lower memory being the immediate fix to it. But these are source code changes to the capture application (https://lost-contact.mit.edu/afs/pdc.kth...cord.dmic/). And I don't know whether the source is in all the distributions that you can simply hit rebuild or how this came about. So someone does need to do a little bit of extra work but this is what I found:

http://archive.irixnet.org/apocrypha/nek...rob/1.html

Post:

rooprob
Who joined Sept. 1, 2010, 6:20 a.m.
and authored 37 notes
Wrote on the subject of O2 dmrecord error at Feb. 16, 2012, 4:17 p.m...
May as well complete this thread.

I found the reason in the source for dmrecord.dmic (/usr/share/src/dmedia), which it seems to share behaviour of the stock utility /usr/sbin/dmrecord.

capture.c
DC(dmBufferSetPoolDefaults(p,options.outbufs,video.xferbytes,DM_TRUE,DM_TRUE));

The establishment of the buffer for the video -> compressor sets both cacheable and mapped to TRUE. This is actually a cache coherency overhead when passing data between the two pools which seems to drive the system into a problem state. When I set to FALSE, I observe consistent realtime compression without any issues over and over.

That was a bit of a rat hole. Learned about the dmedia libraries, written three different versions of dmrecord (forking and event driven), a par parser, some graphviz transformations of the call trace data. Pretty pictures too

Educational, but it was finding jrecvid.c on ftp.sgi.com (amazingly still functional) and on a hunch that led me to the apparent fix. Learning is fun.

I'm transitioning jobs at the moment which is why I have a stupid amount of free time right now to expend on this puzzle. (I'm not allowed to touch much at my current job, just hang around to answer questions until the end of the month). I may consider learning motif to make a little UI for my doings.

[gijoe77]

 written three different versions of dmrecord (forking and event driven), a par parser, some graphviz transformations of the call trace data. Pretty pictures too

this all sounds amazing!  Tell us more about this par parser?  I've been playing with par to troubleshoot various issues, and frankly I get a little overwhelmed with all the info it spits out - your custom tools sound fantastic!  

Also, do you mind uploading the fixed dmrecord you compiled? (or any of the other custom stuff you mentioned really!)

[weblacky]

 written three different versions of dmrecord (forking and event driven), a par parser, some graphviz transformations of the call trace data. Pretty pictures too

this all sounds amazing!  Tell us more about this par parser?  I've been playing with par to troubleshoot various issues, and frankly I get a little overwhelmed with all the info it spits out - your custom tools sound fantastic!  

Also, do you mind uploading the fixed dmrecord you compiled? (or any of the other custom stuff you mentioned really!)

You're misreading the post, I posted the ORIGINAL post (which is why it says "post:" and gives dates and such).  I'm not the original poster...I reposted and linked to a post thread archive that talks about recompiling for possibly this issue.

I have nothing else to input at this time.

[gijoe77]

ah ok, sorry about that

[digitalIRIX]

This may or may not work but try setting 'set SRV_LOMEM_POOL 1'