With Myth it's getting the vbi insertion stuff enabled
which is asking for more memory which is failing. With
just dd'ing i doubt it is doing that. 
As for the 2nd crash i have no idea but i may look
later tonight.

Can you send me a complete log from ivtv_init to the
first crash and i'll take a look through the buffer
allocation. 
Since i don't actually need any buffers in the Xv case
for YUV output i will probably try to stop it
allocating it for that case.

John
--- Lucas Meijer <[EMAIL PROTECTED]> wrote:
> Heya.
> 
> With 3.5f I can switch between mpeg decoding, yuv
> decoding as much as I 
> like. Whenever I throw in myth's livetv however,
> things go bad.
> Either I crash when starting up livetv, or the next
> recording I watch 
> after having been in livetv. I've tried a lot of
> times, and have seen 
> two different looking types of problems in dmesg.
> Both of them pasted 
> below:
> 
> ivtv: Allocate DMA stream 0 using 128 16384 byte
> buffers  2097152 kbytes 
> total
> ivtv: Allocate DMA stream 2 using 41 50540 byte
> buffers  2097152 kbytes 
> total
> ivtv: Setup VBI API header 0x0000bd06 pkts 1 buffs
> 48 ln 34 sz 64
> ivtv: Setup VBI start 0x0018ea04 frames 48 fpi 1
> lines 0x000000f1
> saa7115: decoder disable output
> saa7115: decoder enable output
> ivtv: VBI insertion started
> ivtv: Allocate DMA stream 5 using 16 65536 byte
> buffers  1048576 kbytes 
> total
> mythfrontend: page allocation failure. order:4,
> mode:0xd0
>   [<c0142d4e>] __alloc_pages+0x2ae/0x440
>   [<c0145b9f>] kmem_getpages+0x2f/0xa0
>   [<c0146708>] alloc_slabmgmt+0x58/0x70
>   [<c01468ab>] cache_grow+0xab/0x150
>   [<c0146aca>] cache_alloc_refill+0x17a/0x250
>   [<c0146e6e>] __kmalloc+0x7e/0x90
>   [<d0911b7b>] ivtv_init_buffer+0x3b/0x150 [ivtv]
>   [<d0911cc0>] ivtv_init_queue_nolock+0x30/0x80
> [ivtv]
>   [<d0911d5e>] ivtv_init_queues_nolock+0x4e/0x80
> [ivtv]
>   [<d09134f4>] ivtv_stream_alloc+0xf4/0x220 [ivtv]
>   [<d0915ed1>] ivtv_claim_stream+0x51/0x1d0 [ivtv]
>   [<d0917c7e>] ivtv_v4l2_write+0x8e/0x150 [ivtv]
>   [<c015ceb7>] vfs_write+0xa7/0x160
>   [<c015d041>] sys_write+0x51/0x80
>   [<c01032df>] syscall_call+0x7/0xb
> ivtv: No memory on buffer alloc!
> ivtv: Buffer alloc failed!
> ivtv: Couldn't allocate buffers, streamtype 5
> 
> 
> and the other one:
> 
> Unable to handle kernel NULL pointer dereference at
> virtual address 00000030
>   printing eip:
> d091830d
> *pde = 00000000
> Oops: 0000 [#1]
> PREEMPT SMP
> Modules linked in: snd_pcm_oss snd_mixer_oss
> snd_seq_oss 
> snd_seq_midi_event snd_
> seq snd_seq_device 8250_pnp 8250 serial_core
> snd_intel8x0 snd_ac97_codec 
> snd_pcm
>   snd_timer snd soundcore snd_page_alloc w83627hf
> i2c_sensor i2c_isa 
> i2c_nforce2
> i2c_dev ivtv_fb lirc_i2c lirc_dev msp3400 saa7127
> saa7115 tuner tveeprom 
> ivtv i2
> c_algo_bit i2c_core videodev forcedeth
> CPU:    0
> EIP:    0060:[<d091830d>]    Not tainted VLI
> EFLAGS: 00010246   (2.6.11-gentoo-r8)
> EIP is at ivtv_v4l2_open+0xad/0x360 [ivtv]
> eax: 00000000   ebx: d093b7e4   ecx: 00000000   edx:
> 00000002
> esi: ce9cf200   edi: d093b7e0   ebp: 00000010   esp:
> cfa4dea0
> ds: 007b   es: 007b   ss: 0068
> Process mythfrontend (pid: 9028, threadinfo=cfa4d000
> task=cc2f2080)
> Stack: cfc13660 0000001b 000021b0 00000000 00000000
> 00000000 00000002 
> 00000000
>         d093b480 cf38abc0 d08cd7c0 cf8b4e20 d08cc24f
> ce8c5d84 cf8b4e20 
> cfc9e0f8
>         00000001 00000000 cfc9e0e0 d08cc110 00000000
> ce8c5d84 c0166371 
> ce8c5d84
> Call Trace:
>   [<d08cc24f>] video_open+0x13f/0x290 [videodev]
>   [<d08cc110>] video_open+0x0/0x290 [videodev]
>   [<c0166371>] chrdev_open+0xf1/0x1c0
>   [<c015bfa4>] dentry_open+0x164/0x260
>   [<c015be32>] filp_open+0x62/0x70
>   [<c03a006d>] _spin_unlock+0xd/0x30
>   [<c015c255>] sys_open+0x45/0xf0
>   [<c01032df>] syscall_call+0x7/0xb
> Code: 93 d0 81 c7 e0 b7 93 d0 8d b6 00 00 00 00 8b
> 74 24 18 c1 e6 08 03 
> b3 90 a8
>   01 00 f6 05 94 9f 93 d0 02 0f 85 80 02 00 00 8b 46
> 3c <39> 68 30 0f 84 
> 62 02 00
>   00 ff 44 24 18 8b 54 24 18 3b 97 90 a8
>   <6>note: mythfrontend[9028] exited with
> preempt_count 1
> ivtv: close stopping embedded VBI capture
> 
> 
>
-------------------------------------------------------
> This SF.Net email is sponsored by Oracle Space
> Sweepstakes
> Want to be the first software developer in space?
> Enter now for the Oracle Space Sweepstakes!
>
http://ads.osdn.com/?ad_id=7412&alloc_id=16344&op=click
> _______________________________________________
> ivtv-devel mailing list
> [email protected]
>
https://lists.sourceforge.net/lists/listinfo/ivtv-devel
> 


-------------------------------------------------------
This SF.Net email is sponsored by Oracle Space Sweepstakes
Want to be the first software developer in space?
Enter now for the Oracle Space Sweepstakes!
http://ads.osdn.com/?ad_id=7412&alloc_id=16344&op=click
_______________________________________________
ivtv-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ivtv-devel

Reply via email to