Yesterday evening I thought "enough is enough - time to fix RAMDISK!" and
decided to play with it a bit.
I thought there are two ways to proceed:
1. Maintain the original (Linus') idea of having a very subtle driver
that depends too tight on the current (and future?) changes and state
of Buffer and Page Cache subsystems.
2. Abandon the above idea and implement a very simple driver that however
is rock-solid to withstand any changes outside itself (provided
kernel interfaces stay the same, of course).
So, I chose 2. as it was obviously easier :)
I don't call this "going back to static RAMDISK" as it is not truly static
- it vmalloc's the area for itself only on the first open so if you don't
use any ramdisks you don't waste any space.
Please review it:
http://www.ocston.org/~tigran/patches/ramdisk-2.3.35-pre2.patch
Tested only plain ramdisk part (not initrd, i.e.)
Regards,
------
Tigran A. Aivazian | http://www.sco.com
Escalations Research Group | tel: +44-(0)1923-813796
Santa Cruz Operation Ltd | http://www.ocston.org/~tigran
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/