timeouts on ATAPI cd-burner

[Peter Wullinger]

On Thu, Feb 26, 2004 at 06:20:36PM +0100, Thomas Quinot wrote:
> * Peter Wullinger, 2004-02-19 :
> 

> > I have upgraded to 5.2-CURRENT #5 from yesterday, to
> > test if the following commit fixes my problem,
> > beeing to able to write to an PHILIPS CDRW48P P1.6
> > ATAPI burner using cdrecord: 

I've upgraded to a new -CURRENT in the meanwhile,
but as far a I can tell, the behaviour has not changed.

The output below is from a FreeBSD 5.2.1-RELEASE-p1 kernel,
(cvs'd from yesterday):

I you need more debugging output or other information,
please contact me.

> 
> Does burning CDs with burncd work on your setup?
> 

% burncd -f /dev/acd1  -d -e -t -v data image.iso
adding type 0x08 file image.iso size 659200 KB 329600 blocks
next writeable LBA -150
CUE sheet:
41 00 00 14 00 00 00 00
41 01 01 10 00 00 02 00
41 aa 01 14 00 49 10 32
addr = 0 size = 675020800 blocks = 329600
writing from file image.iso size 659200 KB

only wrote -1 of 32768 bytes: Input/output error

Hangs for a while, but is interruptible by ^C.
Trying it once again (after the camcontrol sequence you
suggested below):

% ps waxl | grep burncd
    0   814   724   0  76  0  1324  556 ATA re D+    v1    0:00.00 burncd -f /dev/acd1 -d -e -t -v data image.iso

> Also, when cdrecord hangs, you can try to at least unlock the drive with
> the following commands:
> 
> # FLUSH CACHE (abort CD write sequence)
> camcontrol cmd 2:1:0 -c 0x35 -v &&
> # PREVENT/ALLOW MEDIUM REMOVAL
> camcontrol cmd 2:1:0 -c 0x1e -v

That one hangs, too:

% cdrecord -v -dao -dummy image.iso
[...]
Waiting for reader process to fill input buffer ... input buffer ready.
BURN-Free is OFF.
Sending CUE sheet...
Writing pregap for track 1 at -150

% ps waxl | grep camcontrol
    0   755   749   0  -8  0  1364  752 cgticb D+    v2    0:00.00 camcontrol cmd 2:0:0 -c 0x35 -v

Cheers,
	Peter