On 14:04 11/09, Dave Chinner wrote:
> On Tue, Sep 10, 2019 at 03:33:27PM -0700, Andres Freund wrote:
> > Hi,
> >
> > Especially with buffered io it's fairly easy to hit contention on the
> > inode lock, during writes. With something like io_uring, it's even
> > easier, because it currently (but see [1]) farms out buffered writes to
> > workers, which then can easily contend on the inode lock, even if only
> > one process submits writes. But I've seen it in plenty other cases too.
> >
> > Looking at the code I noticed that several parts of the "nowait aio
> > support" (cf 728fbc0e10b7f3) series introduced code like:
> >
> > static ssize_t
> > ext4_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
> > {
> > ...
> > if (!inode_trylock(inode)) {
> > if (iocb->ki_flags & IOCB_NOWAIT)
> > return -EAGAIN;
> > inode_lock(inode);
> > }
>
> The ext4 code is just buggy here - we don't support RWF_NOWAIT on
> buffered writes. Buffered reads, and dio/dax reads and writes, yes,
> but not buffered writes because they are almost guaranteed to block
> somewhere. See xfs_file_buffered_aio_write():
>
> if (iocb->ki_flags & IOCB_NOWAIT)
> return -EOPNOTSUPP;
>
> generic_write_checks() will also reject IOCB_NOWAIT on buffered
> writes, so that code in ext4 is likely in the wrong place...
Yes, but inode_trylock is checking if we can get inode sem immidiately,
and if not bail, instead of waiting for the sem, as opposed to rejecting
bufferd I/O nowait writes.
generic_write_checks() has the checks which disallows nowait without direct
writes, so we can do away with those checks in ext4_file_write_iter().
However, the return code in generic_write_checks() is -EINVAL and
-ENOTSUPP in ext4_file_write_iter(). Removing the check in write_iter()
will change the error code to -EINVAL from -EOPNOTSUPP.
--
Goldwyn
