Re: cdrecord dev=ATA cannont scanbus as non-root

["J.A. Magallon" <jamagallon@able.es>]

On 2004.11.30, Jan Engelhardt wrote:
> >Err, as it tries to open a device and it does not exist.
> >I tries sequentially
> >hda, hdb, hdc.. up to 256 until it finds something to open.
> >If it exists, but has not permissions, it keeps trying on the next.
> >But if it is not present, cdrecord gives up.
> 
> Reasonable procedure. Albeit, leaky:
> Imagine you use devfs... open()ing something like hd* would probably always
> create a node. Although there is a max on possible, meaningful, nodes,
> someone could use this to fill up /dev with
> ridiculuous amounts of nodes, all of which are to my knowledge in kernel space.
> "Well then, goodbye".
> 
> Luckily, even the very default config does not create arbitrarily nodes, and
> cdrecord doesnot probe arbitr. devices.
> 
> So I guess, yes, cdrecord should probe harder. Preferably by looking into /sys
> when using a 2.6 system.
> 

Patch for cdrecord:

--- cdrtools-2.01/libscg/scsi-linux-ata.c.orig	2004-11-30 19:02:37.929176615 +0100
+++ cdrtools-2.01/libscg/scsi-linux-ata.c	2004-11-30 19:06:11.316213702 +0100
@@ -385,8 +385,6 @@
 						device, f, errno);
 					}
 					return (-2);
-				} else if (errno == ENOENT || errno == ENODEV) {
-					break;
 				}
 			} else {
 				/* ugly hack, make better, when you can. Alex */

--
J.A. Magallon <jamagallon()able!es>     \               Software is like sex:
werewolf!able!es                         \         It's better when it's free
Mandrakelinux release 10.2 (Cooker) for i586
Linux 2.6.10-rc2-jam3 (gcc 3.4.1 (Mandrakelinux 10.1 3.4.1-4mdk)) #1