Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Fri, 21 Apr 2023 13:25:04 +0100
From:   Mel Gorman <mgorman@techsingularity.net>
To:     Johannes Weiner <hannes@cmpxchg.org>
Cc:     linux-mm@kvack.org, Kaiyang Zhao <kaiyang2@cs.cmu.edu>,
        Vlastimil Babka <vbabka@suse.cz>,
        David Rientjes <rientjes@google.com>,
        linux-kernel@vger.kernel.org, kernel-team@fb.com
Subject: Re: [RFC PATCH 01/26] block: bdev: blockdev page cache is movable
Message-ID: <20230421122504.pq76pj6hka2xdbpu@techsingularity.net>
References: <20230418191313.268131-1-hannes@cmpxchg.org>
 <20230418191313.268131-2-hannes@cmpxchg.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-15
Content-Disposition: inline
In-Reply-To: <20230418191313.268131-2-hannes@cmpxchg.org>
Precedence: bulk

On Tue, Apr 18, 2023 at 03:12:48PM -0400, Johannes Weiner wrote:
> While inspecting page blocks for the type of pages in them, I noticed
> a large number of blockdev cache in unmovable blocks. However, these
> pages are actually on the LRU, and their mapping has a .migrate_folio
> callback; they can be reclaimed and compacted as necessary.
> 
> Put them into movable blocks, so they don't cause pollution, and
> subsequent proliferation, of unmovable blocks.
> 
> Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>

Deserves to be split out and needs careful treatment. I don't recall exactly
but I believe I encountered a problem with these allocations in the very
distant past. Some of the allocations were movable and others weren't
or had very long lifetimes. Something like being superblock-related or
directory inodes being pinned for very long lengths of time. This is a
*long* time ago and the picture almost certainly has changed but it's
very possible that a high percentage, but not 100% of these allocations
are not movable in practice.

-- 
Mel Gorman
SUSE Labs