On 08/13/2012 11:04 PM, Theodore Ts'o wrote: > I had an idea which I think would make inline directories more > efficient, although it wouldn't be a backwards compatible change with > the existing patch sets to date. that should be fine since we don't have much machines deployed by now. > > We don't actually need to store the entry for '.', since we know what > that should be, and we could just store the parent directory in a > 4-byte entry. By not storing the full directory entries for "." and > "..", we would save 20 bytes, which for a 256 byte inode where only > 120 bytes or so are available for the inline directory, is pretty > significant. (We can just synthesize them for the benefit of readdir > and lookup --- and I'm pretty sure the VFS is doing its own synthesis > for "." and possibly ".." already as far as lookup is concerned.) As andreas said in another e-mail, that would make the codes a little complex. So in general, in my previous version, I just tried to refactor most of the codes so that they can work for both inline inode and dir blocks. But if these changes are merged, we have to re-write all of the codes that are currently shared, such as ext4_find_entry, empty_dir, ext4_add_entry etc. Having said that, maybe it is worth the codes effort if we can see a great improvement for inline dir. Since Andreas and you are both OK with this new solution, I can try to re-work the inline data patch to support it. Thanks Tao -- To unsubscribe from this list: send the line "unsubscribe linux-ext4" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
