On 8/30/06, Attila Kinali <[EMAIL PROTECTED]> wrote:
On Wed, 30 Aug 2006 13:42:49 -0700 "Stephen Pollei" <[EMAIL PROTECTED]> wrote:
> Of course I think Linux is still encoding major/minor number into a > single 32bit number using 12 bits for major, and 20 bits for minor. > Thats 4096 possibilities for major number and 1,048,576 for minor. > They most likely would want you to use your own major number therefore > you can have far more minor numbers than you could use at this point. Actualy, you only get a range of minor numbers these days. The days where a major number had to be assigned per device are over.
Really? http://www.lanana.org/docs/device-list/devices-2.6+.txt http://www.lanana.org/docs/device-list/devices.txt I looked through both these lists.. and I don't see a hint of that being true. Largest assigned major was 257 .. So it's not like they have shortage of major numbers. I did read somewhere else that 2.4 and earlier has 8bit major and 8 bit minor. That's tighter but they still have a few open. http://www.lanana.org/docs/device-list/instructions.txt says "To have a major number allocated, or a minor number in situations where that applies (e.g. busmice) ..."... OG isn't a bus mouse(10), or usb(180), or low density serial port(204-205)... Those are about only majors which do seem to mix different types of devices together. Closest thing to OG case seems to be : Matrox mga_vid(83), 3dfx voodoo(107), and nvidia(195). They all got their own major.. And there seems to be some generic majors: video for linux(81), linuxtv.org DVB(212), linVideoStreams(194), Universal frame buffer(29), and Virtual console capture devices(7). They also all got their own major. Also there are plenty of majors that can be locally assigned. 60 to 63, 120 to 127, and 240 to 254. So that handles the embedded 2.4 case that you mentioned. Any one running a embedded kernel can be assumed to be able to locally assign a major number. -- http://dmoz.org/profiles/pollei.html http://sourceforge.net/users/stephen_pollei/ http://www.orkut.com/Profile.aspx?uid=2455954990164098214 http://stephen_pollei.home.comcast.net/ _______________________________________________ Open-graphics mailing list [email protected] http://lists.duskglow.com/mailman/listinfo/open-graphics List service provided by Duskglow Consulting, LLC (www.duskglow.com)
