On Mon, Jan 13, 2003 at 12:35:50AM +0100, Ronald Bultje wrote:
> On Sat, 2003-01-11 at 15:41, [EMAIL PROTECTED] wrote:
> > -> Every time when I start recording for the first time since the modules
> > have been loaded, I get an "error syncing on a buffer: timer expired"
> > after about 2 seconds (= first write to disk).
>
> Urm...? On a 440BX mainboard?!?
It's a Compaq, Pentium III 450. They have a reputation of slightly altering
hardware sometimes :-s
But it is definitely a 440BX chipset.
Output of lspci:
00:00.0 Host bridge: Intel Corp. 440BX/ZX/DX - 82443BX/ZX/DX Host bridge (rev 03)
00:01.0 PCI bridge: Intel Corp. 440BX/ZX/DX - 82443BX/ZX/DX AGP bridge (rev 03)
00:0d.0 Multimedia video controller: Brooktree Corporation Bt878 Video Capture (rev 11)
00:0d.1 Multimedia controller: Brooktree Corporation Bt878 Audio Capture (rev 11)
00:0e.0 Ethernet controller: Intel Corp. 82557/8/9 [Ethernet Pro 100] (rev 08)
00:10.0 Multimedia video controller: Zoran Corporation ZR36057PQC Video cutting
chipset (rev 02)
00:14.0 ISA bridge: Intel Corp. 82371AB/EB/MB PIIX4 ISA (rev 02)
00:14.1 IDE interface: Intel Corp. 82371AB/EB/MB PIIX4 IDE (rev 01)
00:14.2 USB Controller: Intel Corp. 82371AB/EB/MB PIIX4 USB (rev 01)
00:14.3 Bridge: Intel Corp. 82371AB/EB/MB PIIX4 ACPI (rev 02)
01:00.0 VGA compatible controller: ATI Technologies Inc 3D Rage Pro AGP 1X/2X (rev 5c)
/proc/interrupts:
CPU0
0: 24116 XT-PIC timer
1: 357 XT-PIC keyboard
2: 0 XT-PIC cascade
5: 0 XT-PIC SoundBlaster
8: 3 XT-PIC rtc
9: 50 XT-PIC DC10plus[0]
11: 110 XT-PIC eth0, bttv
14: 2648 XT-PIC ide0
15: 16 XT-PIC ide1
NMI: 0
LOC: 24079
ERR: 0
> Last resort solution: --file-flush=0. ;-). If it doesn't help, even if
> the card has its own IRQ, please report it again and I'll have a closer
> look. Even first-time records should work (doh!).
Tried this before without success. Although right now, I do notice
something else re: file-flush=0.
I have a FAT filesystem on the disk where I record, because I used to
have a windows installation where I wanted to be able to access the
recorded material.
Now, I have the timer expired problem both when I record to an ext2 with
some free space, or the usual FAT fs (without special --file-flush setting).
But with --file-flush=0, recording to an ext2 fs seems to be stable
(although I can't record more than a few minutes on this ext2 because of
disk space right now, but so far it has worked every time).
What I mean by "stable":
It seems: when running lavrec interactively, all the fields in the
bottom status line are continuously being updated; but everytime on a
sync() (or with file-flush=0: when the disks writes a chunk of data),
there is a short pause: the values (time, int, ...) stay the same, and after
the data has been written, the "int" field is close to 0 for a short time
(instead of a normal value of 39-40 which I approximately get in between
these sync's). So, while writing to disk, no frames are being fetched from
the video device, and after writing, all pending frames are quickly
retrieved at once - must be that we have been in kernel mode for that while
I guess? I just say this because I'm not sure whether that's what is
supposed to happen or not.
Now, when I run lavrec on the ext2 fs and with --file-flush=0, lavrec
doesn't hang anymore on every write to disk (it still does hang for a
while sometimes, but only once every 20 seconds or so).
I'll free some more space, or I'll convert the FAT to ext2 one of these
days, and report further.
Thanks for the hint.
Thanks to all you guys working on the driver and the other software!
Kind regards,
Jonas
-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
_______________________________________________
Mjpeg-users mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/mjpeg-users
