On Wed, Dec 06, 2006 at 09:05:50PM -0500, Ralph Siemsen wrote: > dann frazier wrote: > > > The manpage for set_therm says: > > "The program requires a char device node /dev/temperature with > > major=10, minor=131 to interface with the ds1620 device driver" > > > > However, on my freshly-installed-with-etch-rc1 system, I have a > > /dev/temp, not a /dev/temperature. /dev/temp has the right major/minor, > > so it seems like /dev/temp should be used instead. > > At one point in time, there was objection to using the standard /dev > entry because the netwinder's output format when you read from this > device does not match the standard (celsius vs farenheit, if memory > serves). This may be a moot point, as /proc/therm gives it in > human-readable form anyways.
I really don't know enough about it to have an opinion on /proc interface vs. /dev interface. I'm just going on what I read in the man page which details using /dev/temperature and not /dev/temp - since they have the same major/minor, their behavior should be identical. > Out of curiosity, on your newly-minted etch system, do you have some of > the other "netwinder special" devices, like /dev/nwflash for example? The only other thing that look interesting to me under /dev is nwbutton. nwflash doesn't appear by default, but is automatically created by udev when I modprobe nwflash. [EMAIL PROTECTED]:~$ ls -l /dev | grep -v -e pty -e tty -e ram -e loop -e hda total 0 lrwxrwxrwx 1 root root 13 Dec 6 15:51 MAKEDEV -> /sbin/MAKEDEV crw------- 1 root root 5, 1 Dec 6 15:51 console lrwxrwxrwx 1 root root 11 Dec 6 15:51 core -> /proc/kcore drwxr-xr-x 5 root root 100 Dec 6 15:51 disk crw-rw---- 1 root video 29, 0 Dec 6 15:51 fb0 lrwxrwxrwx 1 root root 13 Dec 6 15:51 fd -> /proc/self/fd crw-rw-rw- 1 root root 1, 7 Dec 6 15:51 full prw------- 1 root root 0 Dec 6 15:51 initctl drwxr-xr-x 2 root root 60 Dec 6 15:51 input crw-r----- 1 root kmem 1, 2 Dec 6 15:51 kmem crw-rw---- 1 root root 1, 11 Dec 6 15:51 kmsg srw-rw-rw- 1 root root 0 Dec 6 15:51 log drwxr-xr-x 2 root root 60 Dec 6 15:51 mapper crw-r----- 1 root kmem 1, 1 Dec 6 15:51 mem crw-rw---- 1 root root 90, 0 Dec 6 15:51 mtd0 crw-rw---- 1 root root 90, 1 Dec 6 15:51 mtd0ro brw-rw---- 1 root disk 31, 0 Dec 6 15:51 mtdblock0 drwxr-xr-x 2 root root 60 Dec 6 15:51 net crw-rw-rw- 1 root root 1, 3 Dec 6 15:51 null crw-rw---- 1 root root 10, 158 Dec 6 15:51 nwbutton crw-r----- 1 root kmem 1, 4 Dec 6 15:51 port crw------- 1 root root 108, 0 Dec 6 15:51 ppp srw-rw---- 1 root lp 0 Dec 6 15:51 printer crw-rw---- 1 root root 10, 1 Dec 6 15:51 psaux crw-rw-rw- 1 root root 5, 2 Dec 6 23:39 ptmx drwxr-xr-x 2 root root 0 Dec 6 15:50 pts crw-rw-rw- 1 root root 1, 8 Dec 6 15:51 random drwxrwxrwt 2 root root 40 Dec 6 15:51 shm lrwxrwxrwx 1 root root 24 Dec 6 15:51 sndstat -> /proc/asound/oss/sndstat lrwxrwxrwx 1 root root 15 Dec 6 15:51 stderr -> /proc After 'modprobe nwflash': $ ls -l /dev/nwflash crw-rw---- 1 root root 10, 160 Dec 6 23:46 /dev/nwflash > As the upstream maintainer for nwutils, I am happy to make the change if > that's what folks want... I don't read the LSB stuff in too much detail > so I have no idea what the "official" behaviours are supposed to be. I don't really think the LSB comes into play here as this isn't a standardized interface. However, I think a reasonable method is to attempt to read/use /dev/temp and if that fails try /dev/temperature. -- dann frazier | HP Open Source and Linux Organization -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
