Messages in this thread | | | From | "Stephen C. Tweedie" <> | Date | Thu, 10 Feb 2000 14:31:39 +0000 (GMT) | Subject | Re: 2.4 Features |
| |
Hi,
On Wed, 9 Feb 2000 21:15:14 +0000 (GMT), Riley Williams <rhw@MemAlpha.CX> said:
>> No. Ext3 is currently fully backwards compatible with ext2.
> From experience, no it isn't. However, I think what you meant to > say was that the ext3 driver is fully capable of reading ext2 file > systems, in which case you are of course right.
That's what backwards compatibility _means_.
>> Once you unmount an ext3 filesystem cleanly, the journal bit is >> cleared, and a standard ext2 filesystem code and standard ext2 >> utilities will work just fine with the filesystem.
> First, your use of the word "cleanly" invalidates your claim that it's > backwards compatible since the latter claim is meaningless unless it's > true with unclean unmounts.
No --- ext2 can't arbitrarily read ext3 filesystems. That would imply *forwards* compatibility, which has never been claimed. ext3 can leave a filesystem in a state ext2 can deal with if you unmount cleanly, allowing you to migrate between ext2 and ext3 easily, but ext2 simply cannot be allowed to read and write uncleanly-unmounted ext3 filesystems without first forcing the ext3 journal to be recovered (that would just invite completely unnecessary filesystem corruption).
> My memory may be wrong here, but it seems to remember SCT stating that > he chose to rename to ext3 since the journalling facility was likely > to make the driver not be backwards compatible with the current code.
> SCT: Am I wrong here?
Yes! It was chosen to keep the new codebase separate from the old, trusted ext2 code.
--Stephen
- 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/
| |