RE: Recording with SAA7134 card


] -----Original Message-----
] From: video4linux-list-admin@xxxxxxxxxx
] [mailto:video4linux-list-admin@xxxxxxxxxx] On Behalf Of Wan Tat Chee
] Sent: Tuesday, December 17, 2002 4:56 PM
] To: video4linux-list@xxxxxxxxxx
] Subject: Re: Recording with SAA7134 card
]
]
] Gerd,
]
] On 16 Dec 2002, Gerd Knorr wrote:
]
] > That is a bit tricky (i should probably add a cmd line
] switch ...): Go
] > to the plugin directory and rename the v4l2 plugin into
] something not
] > ending in .so.  xawtv will ignore it then ...
]
]
] Ok.
] Here's my test results for v4l1 support:
]
] 2.4.20pre11 (0.1.x driver), NVidia driver
] xawtv has no problems with v4l1 driver whether with xv or without xv.
]
] 2.4.20 (0.2.2 driver), NVidia driver
] xawtv complains about invalid norm (although the config. menu
] shows PAL, NTSC, SEACAM, AUTO, with PAL as default), they
] can't be selected. No video output at all, in any mode (xv,
] noxv, grabdisplay).
]
] I see some error about "pal" not being a valid norm (and then
] lists PAL, NTSC, SEACAM, etc.) It might be a string matching problem?
]
] Mplayer complains about "Unable to set Norm" when accessing
] v4l as well (it uses v4l1). In mplayer's case, it displays TV
] frames alternating with a green output screen.

I get green frames too.

Configuration: FlyVideo3000, Stock RedHat 8.0, Kernel 2.4.20,
v4l2-patch-2.4.20 , saa7134-0.2.2

When using xawtv the video from the card is perfect ('cept for being
able to select norm).

When using vic (http://www.openmash.org) the captured video appears to
have periodic pure green
frames. It looks like it's not a direct 1:1 interleave, but several
normal frames, several green
frames, several normal frames ... (~50% duty cycle)

In addition, occasionally when using vic a kernel error is generated:

Dec 18 07:16:10 test kernel: Unable to handle kernel paging request at
virtual address 09ce3020
Dec 18 07:16:10 test kernel:  printing eip:
Dec 18 07:16:10 test kernel: c013d034
Dec 18 07:16:10 test kernel: *pde = 00000000
Dec 18 07:16:10 test kernel: Oops: 0000
Dec 18 07:16:10 test kernel: CPU:    0
Dec 18 07:16:10 test kernel: EIP:    0010:[<c013d034>]    Not tainted
Dec 18 07:16:10 test kernel: EFLAGS: 00010202
Dec 18 07:16:10 test kernel: eax: 00001600   ebx: 09ce3000   ecx:
00001600   edx: 00001600
Dec 18 07:16:10 test kernel: esi: 00000297   edi: 09ce3000   ebp:
00000000   esp: c3a15e68
Dec 18 07:16:10 test kernel: ds: 0018   es: 0018   ss: 0018
Dec 18 07:16:10 test kernel: Process magicdev (pid: 1218,
stackpage=c3a15000)
Dec 18 07:16:10 test kernel: Stack: c0141ac3 00001600 00000000 00000000
cff55220 cff55220 00000000 00000000
Dec 18 07:16:10 test kernel:        00001600 c013d13e cff55220 00000000
00000000 00001600 c01509a6 00001600
Dec 18 07:16:10 test kernel:        00000000 00001600 c027ea68 00001600
c0141fae 00001600 00000000 ceb0a6e8
Dec 18 07:16:10 test kernel: Call Trace:    [<c0141ac3>] [<c013d13e>]
[<c01509a6>] [<c0141fae>] [<d09b043d>]
Dec 18 07:16:10 test kernel:   [<c0148b73>] [<d09bdfc1>] [<c019f0b2>]
[<c0142173>] [<c0142268>] [<c013a723>]
Dec 18 07:16:10 test kernel:   [<c013a648>] [<c013a9e3>] [<c01090ef>]
Dec 18 07:16:10 test kernel:
Dec 18 07:16:10 test kernel: Code: 8b 7b 20 66 39 53 0c 74 33 4e 85 f6
89 fb 7f ec 45 83 fd 02

Could it be related?

   -Greg

--