Okay, if the core vfs layer doesn't handle over 2 GB, how does mke2fs make
a filesystem on a > 2GB partition? (or does this not work?). For that
matter, couldn't fdisk have potential problems with extended partitions
that are beyond 2 GB? (it has to write the extended partition map out
there).
Block device access works with the llseek() command. But that largely
bypasses most of the VFS layer. The really big kicker is the VM code,
which doesn't support > 32-bit file accesses.
It's possible to modify the ext2 code to bypass the VM layer for > 2GB
files. This would mean that read() and write() would work, but not
mmap(). So yes, it will eventually happen.
(If someone wants to make it happen sooner, and has some kernel
programming experience, they should contact me; I'd be happy to point
them at what needs to happen.)
- Ted
-
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/