[00/11] Remove remaining parts of congestion tracking code.

Here is another refresh of my "remove congestion tracking" series.

I removed the small tweaks to read-ahead code because they raised
questions and are actually required for this series.

Everything else is the same.

Andrew: please drop the version of this that is currently in your tree
even if you don't take this.  The changes to fuse/nfs/ceph are not
appropriate and I wouldn't want them to land by mistake.

Thanks,
NeilBrown

---

NeilBrown (11):
      DOC: convert 'subsection' to 'section' in gfp.h
      MM: document and polish read-ahead code.
      MM: improve cleanup when ->readpages doesn't process all pages.
      fuse: remove reliance on bdi congestion
      nfs: remove reliance on bdi congestion
      ceph: remove reliance on bdi congestion
      Remove inode_congested()
      Remove bdi_congested() and wb_congested() and related functions
      f2fs: replace congestion_wait() calls with io_schedule_timeout()
      block/bfq-iosched.c: use "false" rather than "BLK_RW_ASYNC"
      Remove congestion tracking framework.


 Documentation/core-api/mm-api.rst |  19 ++++-
 Documentation/filesystems/vfs.rst |  16 ++--
 block/bfq-iosched.c               |   2 +-
 drivers/block/drbd/drbd_int.h     |   3 -
 drivers/block/drbd/drbd_req.c     |   3 +-
 fs/ceph/addr.c                    |  22 +++---
 fs/ceph/super.c                   |   1 +
 fs/ceph/super.h                   |   1 +
 fs/ext2/ialloc.c                  |   5 --
 fs/f2fs/compress.c                |   4 +-
 fs/f2fs/data.c                    |   3 +-
 fs/f2fs/f2fs.h                    |   6 ++
 fs/f2fs/segment.c                 |   8 +-
 fs/f2fs/super.c                   |   6 +-
 fs/fs-writeback.c                 |  37 ---------
 fs/fuse/control.c                 |  17 -----
 fs/fuse/dev.c                     |   8 --
 fs/fuse/file.c                    |  17 +++++
 fs/nfs/write.c                    |  14 +++-
 fs/nilfs2/segbuf.c                |  15 ----
 fs/xfs/xfs_buf.c                  |   3 -
 include/linux/backing-dev-defs.h  |   8 --
 include/linux/backing-dev.h       |  50 ------------
 include/linux/fs.h                |   9 ++-
 include/linux/nfs_fs_sb.h         |   1 +
 include/trace/events/writeback.h  |  28 -------
 mm/backing-dev.c                  |  57 --------------
 mm/fadvise.c                      |   5 +-
 mm/readahead.c                    | 122 ++++++++++++++++++++++++++++--
 mm/vmscan.c                       |  21 +----
 30 files changed, 212 insertions(+), 299 deletions(-)

--
Signature

On Tue, 2022-02-22 at 14:17 +1100, NeilBrown wrote:
> The bdi congestion tracking in not widely used and will be removed.
> 
> CEPHfs is one of a small number of filesystems that uses it, setting
> just the async (write) congestion flags at what it determines are
> appropriate times.
> 
> The only remaining effect of the async flag is to cause (some)
> WB_SYNC_NONE writes to be skipped.
> 
> So instead of setting the flag, set an internal flag and change:
>  - .writepages to do nothing if WB_SYNC_NONE and the flag is set
>  - .writepage to return AOP_WRITEPAGE_ACTIVATE if WB_SYNC_NONE
>     and the flag is set.
> 
> The writepages change causes a behavioural change in that pageout() can
> now return PAGE_ACTIVATE instead of PAGE_KEEP, so SetPageActive() will
> be called on the page which (I think) wil further delay the next attempt
> at writeout.  This might be a good thing.
> 
> Signed-off-by: NeilBrown <neilb@suse.de>

Maybe. I have to wonder whether all of this is really useful.

When things are congested we'll avoid trying to issue new writeback
requests. Note that we don't prevent new pages from being dirtied here -
- only their being written back.

This also doesn't do anything in the DIO or sync_write cases, so if we
lose caps or are doing DIO, we'll just keep churning out "unlimited"
writes in those cases anyway.

With ceph too, we're not likely to be dealing with a single server as
well. One OSD could be struggling to keep up but others are OK. Do we
really want to throttle writeback to the ones that are fine?

FWIW, the original patch that added this stuff was this:

commit 2baba25019ec564cd247af74013873d69a0b8190
Author: Yehuda Sadeh <yehuda@hq.newdream.net>
Date:   Fri Dec 18 13:51:57 2009 -0800

    ceph: writeback congestion control
    
    Set bdi congestion bit when amount of write data in flight exceeds adjustable
    threshold.
    
    Signed-off-by: Yehuda Sadeh <yehuda@hq.newdream.net>
    Signed-off-by: Sage Weil <sage@newdream.net>

...but it's pretty scant on details.

The only reason I can see to throttle writeback like this is to prevent
you from having too much memory tied up in writeback requests, but we
aren't limiting other requests in the same way.

Maybe we would do better to just rip this stuff out?

> ---
>  fs/ceph/addr.c  |   22 +++++++++++++---------
>  fs/ceph/super.c |    1 +
>  fs/ceph/super.h |    1 +
>  3 files changed, 15 insertions(+), 9 deletions(-)
> 
> diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c
> index c98e5238a1b6..dc7af34640dd 100644
> --- a/fs/ceph/addr.c
> +++ b/fs/ceph/addr.c
> @@ -563,7 +563,7 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
>  
>  	if (atomic_long_inc_return(&fsc->writeback_count) >
>  	    CONGESTION_ON_THRESH(fsc->mount_options->congestion_kb))
> -		set_bdi_congested(inode_to_bdi(inode), BLK_RW_ASYNC);
> +		fsc->write_congested = true;
>  
>  	req = ceph_osdc_new_request(osdc, &ci->i_layout, ceph_vino(inode), page_off, &len, 0, 1,
>  				    CEPH_OSD_OP_WRITE, CEPH_OSD_FLAG_WRITE, snapc,
> @@ -623,7 +623,7 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
>  
>  	if (atomic_long_dec_return(&fsc->writeback_count) <
>  	    CONGESTION_OFF_THRESH(fsc->mount_options->congestion_kb))
> -		clear_bdi_congested(inode_to_bdi(inode), BLK_RW_ASYNC);
> +		fsc->write_congested = false;
>  
>  	return err;
>  }
> @@ -635,6 +635,10 @@ static int ceph_writepage(struct page *page, struct writeback_control *wbc)
>  	BUG_ON(!inode);
>  	ihold(inode);
>  
> +	if (wbc->sync_mode == WB_SYNC_NONE &&
> +	    ceph_inode_to_client(inode)->write_congested)
> +		return AOP_WRITEPAGE_ACTIVATE;
> +
>  	wait_on_page_fscache(page);
>  
>  	err = writepage_nounlock(page, wbc);
> @@ -707,8 +711,7 @@ static void writepages_finish(struct ceph_osd_request *req)
>  			if (atomic_long_dec_return(&fsc->writeback_count) <
>  			     CONGESTION_OFF_THRESH(
>  					fsc->mount_options->congestion_kb))
> -				clear_bdi_congested(inode_to_bdi(inode),
> -						    BLK_RW_ASYNC);
> +				fsc->write_congested = false;
>  
>  			ceph_put_snap_context(detach_page_private(page));
>  			end_page_writeback(page);
> @@ -760,6 +763,10 @@ static int ceph_writepages_start(struct address_space *mapping,
>  	bool done = false;
>  	bool caching = ceph_is_cache_enabled(inode);
>  
> +	if (wbc->sync_mode == WB_SYNC_NONE &&
> +	    fsc->write_congested)
> +		return 0;
> +
>  	dout("writepages_start %p (mode=%s)\n", inode,
>  	     wbc->sync_mode == WB_SYNC_NONE ? "NONE" :
>  	     (wbc->sync_mode == WB_SYNC_ALL ? "ALL" : "HOLD"));
> @@ -954,11 +961,8 @@ static int ceph_writepages_start(struct address_space *mapping,
>  
>  			if (atomic_long_inc_return(&fsc->writeback_count) >
>  			    CONGESTION_ON_THRESH(
> -				    fsc->mount_options->congestion_kb)) {
> -				set_bdi_congested(inode_to_bdi(inode),
> -						  BLK_RW_ASYNC);
> -			}
> -
> +				    fsc->mount_options->congestion_kb))
> +				fsc->write_congested = true;
>  
>  			pages[locked_pages++] = page;
>  			pvec.pages[i] = NULL;
> diff --git a/fs/ceph/super.c b/fs/ceph/super.c
> index bf79f369aec6..4a3b77d049c7 100644
> --- a/fs/ceph/super.c
> +++ b/fs/ceph/super.c
> @@ -802,6 +802,7 @@ static struct ceph_fs_client *create_fs_client(struct ceph_mount_options *fsopt,
>  	fsc->have_copy_from2 = true;
>  
>  	atomic_long_set(&fsc->writeback_count, 0);
> +	fsc->write_congested = false;
>  
>  	err = -ENOMEM;
>  	/*
> diff --git a/fs/ceph/super.h b/fs/ceph/super.h
> index 67f145e1ae7a..0bd97aea2319 100644
> --- a/fs/ceph/super.h
> +++ b/fs/ceph/super.h
> @@ -121,6 +121,7 @@ struct ceph_fs_client {
>  	struct ceph_mds_client *mdsc;
>  
>  	atomic_long_t writeback_count;
> +	bool write_congested;
>  
>  	struct workqueue_struct *inode_wq;
>  	struct workqueue_struct *cap_wq;
> 
>

On Thu, 2022-02-24 at 16:41 +1100, NeilBrown wrote:
> On Thu, 24 Feb 2022, Jeff Layton wrote:
> > On Tue, 2022-02-22 at 14:17 +1100, NeilBrown wrote:
> > > The bdi congestion tracking in not widely used and will be removed.
> > > 
> > > CEPHfs is one of a small number of filesystems that uses it, setting
> > > just the async (write) congestion flags at what it determines are
> > > appropriate times.
> > > 
> > > The only remaining effect of the async flag is to cause (some)
> > > WB_SYNC_NONE writes to be skipped.
> > > 
> > > So instead of setting the flag, set an internal flag and change:
> > >  - .writepages to do nothing if WB_SYNC_NONE and the flag is set
> > >  - .writepage to return AOP_WRITEPAGE_ACTIVATE if WB_SYNC_NONE
> > >     and the flag is set.
> > > 
> > > The writepages change causes a behavioural change in that pageout() can
> > > now return PAGE_ACTIVATE instead of PAGE_KEEP, so SetPageActive() will
> > > be called on the page which (I think) wil further delay the next attempt
> > > at writeout.  This might be a good thing.
> > > 
> > > Signed-off-by: NeilBrown <neilb@suse.de>
> > 
> > Maybe. I have to wonder whether all of this is really useful.
> > 
> > When things are congested we'll avoid trying to issue new writeback
> > requests. Note that we don't prevent new pages from being dirtied here -
> > - only their being written back.
> > 
> > This also doesn't do anything in the DIO or sync_write cases, so if we
> > lose caps or are doing DIO, we'll just keep churning out "unlimited"
> > writes in those cases anyway.
> 
> I think the point of congestion tracking is to differentiate between
> sync and async IO.  Or maybe "required" and "optional".
> Eventually the "optional" IO will become required, but if we can delay
> it until a time when there is less "required" io, then maybe we can
> improve perceived latency.
> 
> "optional" IO here is write-back and read-ahead.  If the load of
> "required" IO is bursty, and if we can shuffle that optional stuff into
> the quiet periods, we might win.
> 

In that case, maybe we should be counting in-flight reads too and deny
readahead when the count crosses some threshold? It seems a bit silly to
only look at writes when it comes to "congestion".

> Whether this is a real need is an important question that I don't have an
> answer for.  And whether it is better to leave delayed requests in the
> page cache, or in the low-level queue with sync requests able to
> over-take them - I don't know.  If you have multiple low-level queue as
> you say you can with ceph, then lower might be better.
> 
> The block layer has REQ_RAHEAD ..  maybe those request get should get a
> lower priority ... though I don't think they do.
> NFS has a 3 level priority queue, with write-back going at a lower
> priority ... I think... for NFSv3 at least.
> 
> Sometimes I suspect that as all our transports have become faster, we
> have been able to ignore the extra latency caused by poor scheduling of
> optional requests.  But at other times when my recently upgraded desktop
> is struggling to view a web page while compiling a kernel ...  I wonder
> if maybe we don't have the balance right any more.
> 
> So maybe you are right - maybe we can rip all this stuff out.
> 

I lean more toward just removing it. The existing implementation seems a
bit half-baked with the gaps in what's being counted. Granted, the
default congestion threshold is pretty high with modern memory sizes, so
it probably doesn't come into play much in practice, but removing it
would reduce some complexity in the client.