Re: [PATCH 1/1] mm: vmscan: Reduce throttling due to a failure to make progress

[Mel Gorman]

On Tue, Nov 30, 2021 at 06:59:58PM +0100, Mike Galbraith wrote:
> On Tue, 2021-11-30 at 17:27 +0000, Mel Gorman wrote:
> >
> > Obviously a fairly different experience and most likely due to the
> > underlying storage.
> 
> I bet a virtual nickle this is the sore spot.
> 

You win a virtual nickle!

Using an older laptop with slower storage and less memory a frequency
analysis of the stall reasons and source showed the top triggering event
was

Event count:                6210
 mm_vmscan_throttled: nid=0 usec_timeout=100000 usect_delayed=xxx reason=VMSCAN_THROTTLE_WRITEBACK

 => trace_event_raw_event_mm_vmscan_throttled <ffffffff9987224a>
 => reclaim_throttle <ffffffff99873df2>
 => shrink_node <ffffffff99875bd5>
 => do_try_to_free_pages <ffffffff99875cf8>
 => try_to_free_pages <ffffffff998772e3>
 => __alloc_pages_slowpath.constprop.114 <ffffffff998c2ad9>
 => __alloc_pages <ffffffff998c366b>
 => folio_alloc <ffffffff998e4107>
 => page_cache_ra_unbounded <ffffffff99868fab>
 => filemap_fault <ffffffff9985cb13>
 => __do_fault <ffffffff99899361>
 => __handle_mm_fault <ffffffff998a1470>
 => handle_mm_fault <ffffffff998a19ba>
 => do_user_addr_fault <ffffffff99688734>
 => exc_page_fault <ffffffff9a07bcd7>
 => asm_exc_page_fault <ffffffff9a200ace>

-- 
Mel Gorman
SUSE Labs