Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Sun, 24 Jan 2021 14:54:47 -0800 (PST)
From:   David Rientjes <rientjes@google.com>
To:     Vlastimil Babka <vbabka@suse.cz>
cc:     Charan Teja Reddy <charante@codeaurora.org>,
        akpm@linux-foundation.org, mhocko@suse.com, khalid.aziz@oracle.com,
        ngupta@nitingupta.dev, vinmenon@codeaurora.org, linux-mm@kvack.org,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH V3] mm/compaction: correct deferral logic for proactive
 compaction
In-Reply-To: <80a1a433-c520-4c73-61ce-55cf33739fc5@suse.cz>
Message-ID: <627a82ec-94ef-a233-4637-28bc82a886e9@google.com>
References: <1610989938-31374-1-git-send-email-charante@codeaurora.org> <af22a056-5c27-256f-74d-63de8fd33084@google.com> <80a1a433-c520-4c73-61ce-55cf33739fc5@suse.cz>
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Precedence: bulk

On Wed, 20 Jan 2021, Vlastimil Babka wrote:

> On 1/19/21 8:26 PM, David Rientjes wrote:
> > On Mon, 18 Jan 2021, Charan Teja Reddy wrote:
> > 
> >> should_proactive_compact_node() returns true when sum of the
> >> weighted fragmentation score of all the zones in the node is greater
> >> than the wmark_high of compaction, which then triggers the proactive
> >> compaction that operates on the individual zones of the node. But
> >> proactive compaction runs on the zone only when its weighted
> >> fragmentation score is greater than wmark_low(=wmark_high - 10).
> >> 
> >> This means that the sum of the weighted fragmentation scores of all the
> >> zones can exceed the wmark_high but individual weighted fragmentation
> >> zone scores can still be less than wmark_low which makes the unnecessary
> >> trigger of the proactive compaction only to return doing nothing.
> >> 
> >> Issue with the return of proactive compaction with out even trying is
> >> its deferral. It is simply deferred for 1 << COMPACT_MAX_DEFER_SHIFT if
> >> the scores across the proactive compaction is same, thinking that
> >> compaction didn't make any progress but in reality it didn't even try.
> > 
> > Isn't this an issue in deferred compaction as well?  It seems like 
> > deferred compaction should check that work was actually performed before 
> > deferring subsequent calls to compaction.
> 
> Direct compaction does, proactive not.
> 
> > In other words, I don't believe deferred compaction is intended to avoid 
> > checks to determine if compaction is worth it; it should only defer 
> > *additional* work that was not productive.
> 
> Yeah, that should be more optimal.
> 

Charan, is this something you'd like to follow up on, or should I take a 
look instead?

Thanks!