I've just reread Linus's notes about IPC filesystem and to me idea about
metafiles "above" usual root looks great: //proc for /proc, //dev for devfs
and so on. No need to mount in every chroot environment, not need to manual
mount anything at all, easy change of root: just umount //root and mount
other thing there. And POSIX agrees that //root or //proc CAN be special
so all is correct. So just kernel changes are needed. Right ? Not quite:
-- cut --
[khim@localhost khim]$ strace rpm -qa 2>&1 | grep //
open("//var/state/rpm/packages.rpm", O_RDONLY) = 3
stat("//var/state/rpm/nameindex.rpm", {st_mode=0, st_size=0, ...}) = 0
open("//var/state/rpm/nameindex.rpm", O_RDONLY) = 4
stat("//var/state/rpm/fileindex.rpm", {st_mode=0, st_size=0, ...}) = 0
open("//var/state/rpm/fileindex.rpm", O_RDONLY) = 5
stat("//var/state/rpm/providesindex.rpm", {st_mode=0, st_size=0, ...}) = 0
open("//var/state/rpm/providesindex.rpm", O_RDONLY) = 6
stat("//var/state/rpm/requiredby.rpm", {st_mode=0, st_size=0, ...}) = 0
open("//var/state/rpm/requiredby.rpm", O_RDONLY) = 7
stat("//var/state/rpm/conflictsindex.rpm", {st_mode=0, st_size=0, ...}) = 0
open("//var/state/rpm/conflictsindex.rpm", O_RDONLY) = 8
stat("//var/state/rpm/groupindex.rpm", {st_mode=0, st_size=0, ...}) = 0
open("//var/state/rpm/groupindex.rpm", O_RDONLY) = 9
stat("//var/state/rpm/triggerindex.rpm", {st_mode=0, st_size=0, ...}) = 0
open("//var/state/rpm/triggerindex.rpm", O_RDONLY) = 10
[khim@localhost khim]$
-- cut --
That is: in LOTS of current Linux systems some vital programs rely on fact
that currently "//"=="/". If we want EVER move to make "//" special IMHO
it's better to add kernel warning NOW. Perhaps configurable so it can be
disabled (not in compile time, but in run-time). This way at least EACH AND
EVERY Linux vendor will be aware about problem. What do you think ?
-
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/
This archive was generated by hypermail 2b29 : Tue Feb 29 2000 - 21:00:23 EST