Re: [PATCH v4 2/4] mm: kick writeback flusher for IOCB_DONTCACHE with targeted dirty tracking

[Jeff Layton]

On Sun, 2026-05-03 at 16:45 +0200, Jan Kara wrote:
> On Fri 01-05-26 10:49:36, Jeff Layton wrote:
> > The IOCB_DONTCACHE writeback path in generic_write_sync() calls
> > filemap_flush_range() on every write, submitting writeback inline in
> > the writer's context.  Perf lock contention profiling shows the
> > performance problem is not lock contention but the writeback submission
> > work itself — walking the page tree and submitting I/O blocks the writer
> > for milliseconds, inflating p99.9 latency from 23ms (buffered) to 93ms
> > (dontcache).
> > 
> > Replace the inline filemap_flush_range() call with a flusher kick that
> > drains dirty pages in the background.  This moves writeback submission
> > completely off the writer's hot path.
> > 
> > To avoid flushing unrelated buffered dirty data, add a dedicated
> > WB_start_dontcache bit and wb_check_start_dontcache() handler that uses
> > the per-wb WB_DONTCACHE_DIRTY counter to determine how many pages to
> > write back.  The flusher writes back that many pages from the oldest dirty
> > inodes (not restricted to dontcache-specific inodes). This helps
> > preserve I/O batching while limiting the scope of expedited writeback.
> > 
> > Like WB_start_all, the WB_start_dontcache bit coalesces multiple
> > DONTCACHE writes into a single flusher wakeup without per-write
> > allocations.
> > 
> > Also add WB_REASON_DONTCACHE as a new writeback reason for tracing
> > visibility, and target the correct cgroup writeback domain via
> > unlocked_inode_to_wb_begin().
> > 
> > dontcache-bench results (same host, T6F_SKL_1920GBF, 251 GiB RAM,
> > xfs on NVMe, fio io_uring):
> > 
> > Buffered and direct I/O paths are unaffected by this patchset. All
> > improvements are confined to the dontcache path:
> > 
> > Single-stream throughput (MB/s):
> >                         Before    After    Change
> >   seq-write/dontcache      298      897    +201%
> >   rand-write/dontcache     131      236     +80%
> > 
> > Tail latency improvements (seq-write/dontcache):
> >   p99:    135,266 us  ->  23,986 us   (-82%)
> >   p99.9: 8,925,479 us ->  28,443 us   (-99.7%)
> > 
> > Multi-writer (4 jobs, sequential write):
> >                                 Before    After    Change
> >   dontcache aggregate (MB/s)     2,529    4,532     +79%
> >   dontcache p99 (us)             8,553    1,002     -88%
> >   dontcache p99.9 (us)         109,314    1,057     -99%
> > 
> >   Dontcache multi-writer throughput now matches buffered (4,532 vs
> >   4,616 MB/s).
> > 
> > 32-file write (Axboe test):
> >                                 Before    After    Change
> >   dontcache aggregate (MB/s)     1,548    3,499    +126%
> >   dontcache p99 (us)            10,170      602     -94%
> >   Peak dirty pages (MB)          1,837      213     -88%
> > 
> >   Dontcache now reaches 81% of buffered throughput (was 35%).
> > 
> > Competing writers (dontcache vs buffered, separate files):
> >                                 Before    After
> >   buffered writer                  868      433 MB/s
> >   dontcache writer                 415      433 MB/s
> >   Aggregate                      1,284      866 MB/s
> > 
> >   Previously the buffered writer starved the dontcache writer 2:1.
> >   With per-bdi_writeback tracking, both writers now receive equal
> >   bandwidth. The aggregate matches the buffered-vs-buffered baseline
> >   (863 MB/s), indicating fair sharing regardless of I/O mode.
> > 
> >   The dontcache writer's p99.9 latency collapsed from 119 ms to
> >   33 ms (-73%), eliminating the severe periodic stalls seen in the
> >   baseline. Both writers now share identical latency profiles,
> >   matching the buffered-vs-buffered pattern.
> > 
> > The per-bdi_writeback dirty tracking dramatically reduces peak dirty
> > pages in dontcache workloads, with the 32-file test dropping from
> > 1.8 GB to 213 MB. Dontcache sequential write throughput triples and
> > multi-writer throughput reaches parity with buffered I/O, with tail
> > latencies collapsing by 1-2 orders of magnitude.
> > 
> > Assisted-by: Claude:claude-opus-4-6
> > Signed-off-by: Jeff Layton <jlayton@xxxxxxxxxx>
> 
> Nice and looks good to me now. Feel free to add:
> 
> Reviewed-by: Jan Kara <jack@xxxxxxx>
> 
> One nit below:
> 
> > +/**
> > + * filemap_dontcache_kick_writeback - kick flusher for IOCB_DONTCACHE writes
> > + * @mapping:	address_space that was just written to
> > + *
> > + * Kick the writeback flusher thread to expedite writeback of dontcache
> > + * dirty pages.  Uses a dedicated WB_start_dontcache bit so that only
> > + * pages tracked by WB_DONTCACHE_DIRTY are written back, rather than
> > + * flushing the entire BDI's dirty pages.
> 
> This comment is a bit confusing as in fact we write arbitrary dirty pages.
> It is only the amount of pages that is influenced by WB_DONTCACHE_DIRTY. So
> I'd rephrase the last sentence like: We queue writeback for the inode's wb
> for as many pages as there are dontcache pages but we don't restrict
> writeback to dontcache pages only. This significantly improves performance
> over either writing all wb's pages or writing only dontcache pages.
> Although it doesn't guarantee quick writeback and reclaim of dontcache
> pages it keeps the amount of dirty pages in check and over longer term
> dontcache pages get written and reclaimed by background writeback even with
> this rough heuristic.
> 
> 								Honza
> 

I'll add that. Thanks for the suggestion and review!
-- 
Jeff Layton <jlayton@xxxxxxxxxx>