asyncread problem with jit.qt.movie

Jul 28, 2007 at 3:49pm

asyncread problem with jit.qt.movie

I have two separate jit.qt.movie instances and I feed them asyncread messages wth paths to videos. Every once in awhile when loading movies, Max locks up then crashes. Looking at the crash log, I see the following thread (and function) crash:

Exception: EXC_BAD_ACCESS (0×0001)
Codes: KERN_INVALID_ADDRESS (0×0001) at 0x376af000

Thread 0 Crashed:
0 com.cycling74.JitterAPI 0x16369c8c jit_gworld_clear + 204
1 com.cycling74.JitterAPI 0x163b4021 jit_qt_movie_clear_offscreen + 65
2 com.cycling74.JitterAPI 0x163b40ef jit_qt_movie_rebuild_gworld + 121
3 com.cycling74.JitterAPI 0x163b4149 jit_qt_movie_mreset + 28
4 com.cycling74.JitterAPI 0x163b51ff jit_qt_movie_fromgworld + 1580
5 com.cycling74.JitterAPI 0x163b533e jit_qt_movie_matrix_calc + 117
6 com.cycling74.MaxMSP46 0x0011efe7 object_method + 1431 (obex.c:801)
7 com.cycling74.MaxAPI 0x01815d78 object_method + 155
8 com.cycling74.JitterAPI 0x16367ff2 jit_object_method + 136
9 com.cycling74.jit.qt.movie 0x316bb34b max_jit_qt_movie_outputmatrix + 181
10 com.cycling74.JitterAPI 0x1638620d max_jit_usurp_qfn + 94
11 com.cycling74.MaxMSP46 0x000598c5 sched_dequeue + 129 (sched.c:351)
12 com.cycling74.MaxMSP46 0x00026bff max_doeventtimerproc + 101 (main.c:677)
13 com.cycling74.MaxMSP46 0x00026ccb max_eventtimerproc + 35 (main.c:724)
14 com.apple.HIToolbox 0x92e29baa TimerVector + 31
15 com.apple.CoreFoundation 0x9082d7e2 CFRunLoopRunSpecific + 3341
16 com.apple.CoreFoundation 0x9082cace CFRunLoopRunInMode + 61
17 com.apple.HIToolbox 0x92df19b8 RunCurrentEventLoopInMode + 285
18 com.apple.HIToolbox 0x92df10c2 ReceiveNextEventCommon + 385
19 com.apple.HIToolbox 0x92e39b94 _AcquireNextEvent + 58
20 com.apple.HIToolbox 0x92e399dc RunApplicationEventLoop + 150
21 com.cycling74.MaxMSP46 0×00027854 app_run + 52 (main.c:1519)
22 com.cycling74.MaxMSP46 0x00027afe main + 680 (main.c:416)
23 com.cycling74.MaxMSP46 0x00002ba2 _start + 216
24 com.cycling74.MaxMSP46 0x00002ac9 start + 41

I am running on a MacBookPro Core 2 Duo 2.4 Ghz. Has anyone else seen this problem? I am running the latest version of both Jitter & Max/MSP.

Rick

#33075
Jul 28, 2007 at 5:04pm

Ive seen this as well.

What versions are you running? When I updated to 1.6.3 final I saw
this almost completely vanish. I think. I havent been patching /
performing much at all lately, but the little ive done has been
pretty stable with 10.4.10/QT 7.1.6, Max 4.6.3 and Jitter 1.6.3.

On Jul 28, 2007, at 11:49 AM, Rick Burnett wrote:

>
> I have two separate jit.qt.movie instances and I feed them
> asyncread messages wth paths to videos. Every once in awhile when
> loading movies, Max locks up then crashes. Looking at the crash
> log, I see the following thread (and function) crash:
>
> Exception: EXC_BAD_ACCESS (0×0001)
> Codes: KERN_INVALID_ADDRESS (0×0001) at 0x376af000
>
> Thread 0 Crashed:
> 0 com.cycling74.JitterAPI 0x16369c8c jit_gworld_clear + 204
> 1 com.cycling74.JitterAPI 0x163b4021
> jit_qt_movie_clear_offscreen + 65
> 2 com.cycling74.JitterAPI 0x163b40ef
> jit_qt_movie_rebuild_gworld + 121
> 3 com.cycling74.JitterAPI 0x163b4149 jit_qt_movie_mreset
> + 28
> 4 com.cycling74.JitterAPI 0x163b51ff
> jit_qt_movie_fromgworld + 1580
> 5 com.cycling74.JitterAPI 0x163b533e
> jit_qt_movie_matrix_calc + 117
> 6 com.cycling74.MaxMSP46 0x0011efe7 object_method + 1431
> (obex.c:801)
> 7 com.cycling74.MaxAPI 0x01815d78 object_method + 155
> 8 com.cycling74.JitterAPI 0x16367ff2 jit_object_method + 136
> 9 com.cycling74.jit.qt.movie 0x316bb34b
> max_jit_qt_movie_outputmatrix + 181
> 10 com.cycling74.JitterAPI 0x1638620d max_jit_usurp_qfn + 94
> 11 com.cycling74.MaxMSP46 0x000598c5 sched_dequeue + 129
> (sched.c:351)
> 12 com.cycling74.MaxMSP46 0x00026bff max_doeventtimerproc
> + 101 (main.c:677)
> 13 com.cycling74.MaxMSP46 0x00026ccb max_eventtimerproc +
> 35 (main.c:724)
> 14 com.apple.HIToolbox 0x92e29baa TimerVector + 31
> 15 com.apple.CoreFoundation 0x9082d7e2 CFRunLoopRunSpecific
> + 3341
> 16 com.apple.CoreFoundation 0x9082cace CFRunLoopRunInMode + 61
> 17 com.apple.HIToolbox 0x92df19b8
> RunCurrentEventLoopInMode + 285
> 18 com.apple.HIToolbox 0x92df10c2
> ReceiveNextEventCommon + 385
> 19 com.apple.HIToolbox 0x92e39b94 _AcquireNextEvent + 58
> 20 com.apple.HIToolbox 0x92e399dc
> RunApplicationEventLoop + 150
> 21 com.cycling74.MaxMSP46 0×00027854 app_run + 52 (main.c:
> 1519)
> 22 com.cycling74.MaxMSP46 0x00027afe main + 680 (main.c:416)
> 23 com.cycling74.MaxMSP46 0x00002ba2 _start + 216
> 24 com.cycling74.MaxMSP46 0x00002ac9 start + 41
>
>
> I am running on a MacBookPro Core 2 Duo 2.4 Ghz. Has anyone else
> seen this problem? I am running the latest version of both Jitter
> & Max/MSP.
>
> Rick

v a d e //

http://www.vade.info
abstrakt.vade.info

#109634
Jul 28, 2007 at 5:40pm

I have a fresh 10.4.10 with latest quicktime, and a fresh install of the 4.6.3 and 1.6.3. Interesting enough, I don’t see it happening on my MacBook with all the same things. Just on my MacBookPro.

Now, I switched the asyncread to use read instead and it stopped. Now I am TERRIFIED of the show I am doing next Friday because a crash would just be HORRIBLE.

*sigh*

#109635
Jul 29, 2007 at 6:28pm

Hey, so… did I speak to soon or…. ??

on async read.

fuck. I thought my new patch somehow worked around this.

Thread 0 Crashed:
0 com.cycling74.JitterAPI 0x16669c74 jit_gworld_clear + 204
1 com.cycling74.JitterAPI 0x166b4009
jit_qt_movie_clear_offscreen + 65
2 com.cycling74.JitterAPI 0x166b40d7
jit_qt_movie_rebuild_gworld + 121
3 com.cycling74.JitterAPI 0x166b4131 jit_qt_movie_mreset + 28
4 com.cycling74.JitterAPI 0x166b4e99
jit_qt_movie_fromgworld + 734
5 com.cycling74.JitterAPI 0x166b5326
jit_qt_movie_matrix_calc + 117
6 com.cycling74.MaxMSP46 0x0011efe7 object_method + 1431
(obex.c:801)
7 com.cycling74.MaxAPI 0x01815d78 object_method + 155
8 com.cycling74.JitterAPI 0x16667fda jit_object_method + 136
9 com.cycling74.jit.qt.movie 0x3072034b
max_jit_qt_movie_outputmatrix + 181
10 com.cycling74.JitterAPI 0x166861f5 max_jit_usurp_qfn + 94
11 com.cycling74.JitterAPI 0x166893bc max_jit_usurp_method +
156
12 com.cycling74.MaxMSP46 0×00030755 typedmess_fun + 1487
(message.c:631)
13 com.cycling74.MaxMSP46 0x00030c96 typedmess + 80
(message.c:399)
14 com.cycling74.MaxAPI 0x0180b9f4 typedmess + 59
15 com.cycling74.JitterAPI 0x1669a2ab max_jit_mop_bang + 51
16 com.cycling74.MaxMSP46 0x000e7b38 outlet_bang + 352
(inletoutlet.c:670)
17 com.cycling74.MaxMSP46 0x000e6718 inlet_bang + 100
(inletoutlet.c:171)
18 com.cycling74.MaxMSP46 0x000e7b38 outlet_bang + 352
(inletoutlet.c:670)
19 com.cycling74.MaxMSP46 0x000e6718 inlet_bang + 100
(inletoutlet.c:171)
20 com.cycling74.MaxMSP46 0x000e7b38 outlet_bang + 352
(inletoutlet.c:670)
21 com.cycling74.MaxMSP46 0x0009c904 route_anything(route*,
symbol*, short, atom*) + 162 (route.c:212)
22 com.cycling74.MaxMSP46 0x0009ca7c route_bang(route*) +
44 (route.c:156)
23 com.cycling74.MaxMSP46 0x000e7b38 outlet_bang + 352
(inletoutlet.c:670)
24 com.cycling74.MaxMSP46 0x000e6718 inlet_bang + 100
(inletoutlet.c:171)
25 com.cycling74.MaxMSP46 0x000e7b38 outlet_bang + 352
(inletoutlet.c:670)
26 com.cycling74.MaxMSP46 0x000e6718 inlet_bang + 100
(inletoutlet.c:171)
27 com.cycling74.MaxMSP46 0x000e7b38 outlet_bang + 352
(inletoutlet.c:670)
28 com.cycling74.MaxMSP46 0x000e6718 inlet_bang + 100
(inletoutlet.c:171)
29 com.cycling74.MaxMSP46 0x000e7af3 outlet_bang + 283
(inletoutlet.c:657)
30 com.cycling74.MaxMSP46 0x000b2fdb trigger_iterate
(unpack*, long, double, symbol*, symbol*, short, atom*, short) + 843
(trigger.c:30)
31 com.cycling74.MaxMSP46 0x000b32a1 trigger_bang(unpack*)
+ 81 (trigger.c:146)
32 com.cycling74.MaxMSP46 0x000e7b38 outlet_bang + 352
(inletoutlet.c:670)
33 com.cycling74.MaxMSP46 0x000e7b38 outlet_bang + 352
(inletoutlet.c:670)
34 com.cycling74.MaxMSP46 0x0009e647 send_bang(send*) + 23
(send.c:32)
35 com.cycling74.MaxMSP46 0x000e7af3 outlet_bang + 283
(inletoutlet.c:657)
36 com.cycling74.MaxMSP46 0x000b2fdb trigger_iterate
(unpack*, long, double, symbol*, symbol*, short, atom*, short) + 843
(trigger.c:30)
37 com.cycling74.MaxMSP46 0x000b32a1 trigger_bang(unpack*)
+ 81 (trigger.c:146)
38 com.cycling74.MaxMSP46 0x000e7af3 outlet_bang + 283
(inletoutlet.c:657)
39 com.cycling74.MaxMSP46 0x000b2fdb trigger_iterate
(unpack*, long, double, symbol*, symbol*, short, atom*, short) + 843
(trigger.c:30)
40 com.cycling74.MaxMSP46 0x000b32a1 trigger_bang(unpack*)
+ 81 (trigger.c:146)
41 com.cycling74.MaxMSP46 0x000e7af3 outlet_bang + 283
(inletoutlet.c:657)
42 com.cycling74.MaxAPI 0x0180a469 outlet_bang + 30
43 com.cycling74.qmetro 0x3084dc02 qmetro_qfn + 20
44 com.cycling74.MaxMSP46 0x000598c5 sched_dequeue + 129
(sched.c:351)
45 com.cycling74.MaxMSP46 0x00026bff max_doeventtimerproc +
101 (main.c:677)
46 com.cycling74.MaxMSP46 0x00026ccb max_eventtimerproc +
35 (main.c:724)
47 com.apple.HIToolbox 0x92e24a8a TimerVector + 31
48 com.apple.CoreFoundation 0x9082d7e2 CFRunLoopRunSpecific +
3341
49 com.apple.CoreFoundation 0x9082cace CFRunLoopRunInMode + 61
50 com.apple.HIToolbox 0x92dec8d8
RunCurrentEventLoopInMode + 285
51 com.apple.HIToolbox 0x92debfe2 ReceiveNextEventCommon
+ 385
52 com.apple.HIToolbox 0x92e34a74 _AcquireNextEvent + 58
53 com.apple.HIToolbox 0x92e348bc
RunApplicationEventLoop + 150
54 com.cycling74.MaxMSP46 0×00027854 app_run + 52 (main.c:
1519)
55 com.cycling74.MaxMSP46 0x00027afe main + 680 (main.c:416)
56 com.cycling74.MaxMSP46 0x00002ba2 _start + 216
57 com.cycling74.MaxMSP46 0x00002ac9 start + 41

On Jul 28, 2007, at 1:40 PM, Rick Burnett wrote:

>
> I have a fresh 10.4.10 with latest quicktime, and a fresh install
> of the 4.6.3 and 1.6.3. Interesting enough, I don’t see it
> happening on my MacBook with all the same things. Just on my
> MacBookPro.
>
> Now, I switched the asyncread to use read instead and it stopped.
> Now I am TERRIFIED of the show I am doing next Friday because a
> crash would just be HORRIBLE.
>
> *sigh*

v a d e //

http://www.vade.info
abstrakt.vade.info

#109636
Jul 30, 2007 at 10:34am

Please try the following:

In your patch, make a message box “; jit qtoptimize 0″ and click it.
Try your patch. Does the crashing go away?

jb

#109637
Jul 30, 2007 at 2:02pm

Sorry, that was a wild goose-chase, and isn’t the issue. I’ll report
back when I have more information.

jb

Am 30.07.2007 um 12:34 schrieb Jeremy Bernstein:

> Please try the following:
>
> In your patch, make a message box “; jit qtoptimize 0″ and click
> it. Try your patch. Does the crashing go away?

#109638
Jul 30, 2007 at 7:42pm

Sounds good. One thing I noticed is that it doesn’t happen as easily on my previous MacBook, although I could get it to do it once in awhile. The MacBookPro it is a LOT more common.

I am still testing as well to see what I can figure out.

I am not sure who is or is not part of Cycling74 so I did contact support, they said you were working on it, so I will not file a bug report.

Is it generally a good idea to just talk to support through the message board? I want to make sure I follow the recommended guidelines for this sort of thing.

Thanks,
rick

#109639
Jul 30, 2007 at 8:04pm

In general, we prefer that you go through support, but since I
happened to catch this one and have reproduced it (and, I think,
solved it), you don’t need to file a support request (although if you
do, you’ll get a shiny new bug reference number which I can use
internally).

The solution isn’t simple, and can’t be worked around in Max, so it’s
going to require some testing before it’s ripe enough to release.
I’ll keep you posted.

jb

Am 30.07.2007 um 21:42 schrieb Rick Burnett:

>
> Sounds good. One thing I noticed is that it doesn’t happen as
> easily on my previous MacBook, although I could get it to do it
> once in awhile. The MacBookPro it is a LOT more common.
>
> I am still testing as well to see what I can figure out.
>
> I am not sure who is or is not part of Cycling74 so I did contact
> support, they said you were working on it, so I will not file a bug
> report.
>
> Is it generally a good idea to just talk to support through the
> message board? I want to make sure I follow the recommended
> guidelines for this sort of thing.
>
> Thanks,
> rick

#109640
Jul 30, 2007 at 8:13pm

Thanks, and I *really* appreciate the level of support I get. I very much enjoy working in max/msp/jitter because it takes things from inception to working software in a fraction of the time as compared to custome coding. I can work in OSX and have made a few cocoa apps (I do love Objective-C) but for my playing around, that extra horsepower I could get just isn’t worth it.

I kind of compare max/MSP to ‘perl’, which I use a lot of. You trade raw processing power for usability and rapid development. Plus, working on a dual 2.4Ghz laptop is sure nice :)

For now, I am not going to file a bug report and just keep on this through this thread. For future issues I will ask first around here just to make sure it isn’t something silly I am doing, then go with the bug report path.

Thanks!

Rick

#109641
Jul 30, 2007 at 11:04pm

Thanks as well from me. This one was a major headache for some time.
Will there be a new jitlib or a Jitter 1.6.4 ? Idly curious, not
being pushy.

On Jul 30, 2007, at 4:04 PM, Jeremy Bernstein wrote:
v a d e //

http://www.vade.info
abstrakt.vade.info

#109642

You must be logged in to reply to this topic.