I spent a bit of time looking at this today.
On Fri, Jun 12, 2009 at 10:33 AM, Theodore Tso<tytso@xxxxxxx> wrote:
> On Fri, Jun 12, 2009 at 06:01:12PM +0530, Aneesh Kumar K.V wrote:
>> Hi,
>>
>> I noticed yesterday that a write to fallocate
>> space via directIO results in fallback to buffer_IO. ie the userspace
>> pages get copied to the page cache and then call a sync.
>>
>> I guess this defeat the purpose of using directIO. May be we should
>> consider this a high priority bug.
My simple experiment -- without a journal -- shows that you're
observation is correct. *Except* if FALLOC_FL_KEEP_SIZE is used in
the fallocate() call, in which case the page cache is *not* used.
Pseudo-code example:
open(O_DIRECT)
fallocate(mode, 512MB)
while (! written 100MB)
write(64K)
close()
If mode == FALLOC_FL_KEEP_SIZE, then no page cache is used.
Otherwise, we *do* go through the page cache.
It comes down to the fact that, since the i_size is not updated with
KEEP_SIZE, then ext4_get_block() is called with create = 1, since the
block that's needed is "beyond" the file end.
>
> I agree that many of users of fallocate() feature (i.e. databases) are
> going to consider this to be a major misfeature.
>
> There's going to be a major performance hit though --- O_DIRECT is
> supposed to be synchronous if all of the alignment requirements are
> met, which means that by the time the write(2) system call returns,
> the data is guaranteed to be on disk. But if we need to manipulate
> the extent tree to indicate that the block is now in use (so the data
> is actually accessible), do we force a synchronous journal commit or
> not? If we don't, then a crash right after an O_DIRECT right into an
> uninitialized region will cause the data to be "lost" (or at least,
> unavailable via the read/write system call). If we do, then the first
> write into uninitialized block will cause a synchronous journal commit
> that will be Slow And Painful, and it might destroy most of the
> performance benefits that might tempt an enterprise database client to
> use fallocate() in the first place.
>
> I wonder how XFS deals with this case? It's a problem that is going
> to hit any journalled filesystem that wants to support fallocate() and
> direct I/O.
>
> One thing I can think of potentially doing is to check to see if the
> extent tree block has already been journalled, and if it is not
> currently involved the current transaction or the previous committing
> transaction, *and* if there is space in the extent tree to mark the
> current unitialized block as initialized (i.e., if the extent needs to
> be split, there is sufficient space so we don't have to allocate a new
> leaf block for the extent tree), we could update the leaf block in
> place and then synchronously write it out, and thus avoid needing to
> do a synchronous journal commit.
In my example above, when KEEP_SIZE is used, it appears that
converting the uninit extent to initialized never failed. I haven't
waded through ext4_ext_convert_to_initialized() to see how it might
fail, and tried to get it to do so.
It would be interesting to see if making this work -- having the
blocks allocated and the buffer mapped -- for O_DIRECT writes in the
absence of a journal, at least, would be feasible. It would certainly
be useful, to us at least.
Thanks,
Curt
>
> In any case, adding this support is going to be non-trivial. If
> someone has time to work on it in the next 2-3 weeks or so, I can push
> it to Linus as a bug fix --- but I'm concerned the fixing this may be
> tricky enough (and the patch invasive enough) that it might be
> challenging to get this fixed in time for 2.6.31.
>
> - Ted
> --
> 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
>
--
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
