Re: io_uring force_nonblock vs POSIX_FADV_WILLNEED

[Andres Freund <andres@anarazel.de>]

Hi,

On 2020-02-01 09:22:45 -0700, Jens Axboe wrote:
> On 2/1/20 2:43 AM, Andres Freund wrote:
> > Seems like either WILLNEED would have to always be deferred, or
> > force_page_cache_readahead, __do_page_cache_readahead would etc need to
> > be wired up to know not to block. Including returning EAGAIN, despite
> > force_page_cache_readahead and generic_readahead() intentially ignoring
> > return values / errors.
> > 
> > I guess it's also possible to just add a separate precheck that looks
> > whether there's any IO needing to be done for the range. That could
> > potentially also be used to make DONTNEED nonblocking in case everything
> > is clean already, which seems like it could be nice. But that seems
> > weird modularity wise.
> 
> Good point, we can block on the read-ahead. Which is counter intuitive,
> but true.

> I'll queue up the below for now, better safe than sorry.
> 
> 
> diff --git a/fs/io_uring.c b/fs/io_uring.c
> index fb5c5b3e23f4..1464e4c9b04c 100644
> --- a/fs/io_uring.c
> +++ b/fs/io_uring.c
> @@ -2728,8 +2728,7 @@ static int io_fadvise(struct io_kiocb *req, struct io_kiocb **nxt,
>  	struct io_fadvise *fa = &req->fadvise;
>  	int ret;
>  
> -	/* DONTNEED may block, others _should_ not */
> -	if (fa->advice == POSIX_FADV_DONTNEED && force_nonblock)
> +	if (force_nonblock)
>  		return -EAGAIN;
>  
>  	ret = vfs_fadvise(req->file, fa->offset, fa->len, fa->advice);

Hm, that seems a bit broad. It seems fairly safe to leave
POSIX_FADV_{NORMAL,RANDOM,SEQUENTIAL} as sync. I guess there's there's
the argument that that's not something one does frequently enough to
care, but it's not hard to imagine wanting to change to RANDOM for a few
reads and then back to NORMAL.

Greetings,

Andres Freund