From: Erwin Authried (eauth@softsys.co.at)
Date: Tue Sep 03 2002 - 11:23:10 EDT
Am Die, 2002-09-03 um 15.54 schrieb Bernhard Kuhn: > David McCullough schrieb: > > http://www.uclinux.org/~bkuhn/Platforms/Coldfire/tarifa/20020903/patch-linux-2.4.19-uc0-m5272c3-mtd+ppcboot > > > > This one changes the MAJOR device for the blkmem driver. I don't think to > > many people would like that ;-) > > Depends: this change allows to use the MTD-block device and blkmem > device > in parallel. You can certainly leave out this modification and anybody > who whises to use an old-style kernel-attached romfs in parallel with > the MTD block device can manualy modify blkmem.c ... > I believe that using both drivers on the same flash device isn't a good idea. The drivers don't know from each other, and when you start using the MTD driver to write to the flash, you will run into big trouble. > > > If you are only using blkmem for the RAM based filesystem access you > > can remove the need for it completely without too much trouble. It's > > much cleaner to have MTD without blkmem :-) > > That's correct, but i didn't yet had the time to write an mtdram driver > that replaces the blkmem driver :-) Or does there yet exist a mtdram > driver that directly maps a kernel attached filesystem? > > > > Have a look at the nettel-uc.c map driver for MTD. It provides the RAM > > based partition for the in memory filesystem and also the partitions needed > > read and write flash using MTD. Just include the driver for RAM devices I > > think. Ignore the erase config stuff in there ;-) > > I will have a look at it - but most mtdram drivers i know are using a > fixed > portion of memory while i realy like the idea of having a kernel > attached romfs. Wasn't there a patch from Greg for this purpose at the beginning of this month? Regards, Erwin This message resent by the uclinux-dev@uclinux.org list server http://www.uClinux.org/
This archive was generated by hypermail 2.1.4 : Thu Sep 19 2002 - 13:21:48 EDT