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:   Tue, 28 Jun 2022 12:27:40 +0100
From:   Matthew Wilcox <willy@infradead.org>
To:     Dave Chinner <david@fromorbit.com>
Cc:     "Darrick J. Wong" <djwong@kernel.org>, linux-xfs@vger.kernel.org,
        linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
        Christoph Hellwig <hch@lst.de>, linux-mm@kvack.org
Subject: Re: Multi-page folio issues in 5.19-rc4 (was [PATCH v3 25/25] xfs:
 Support large folios)
Message-ID: <YrrlrMK/7pyZwZj2@casper.infradead.org>
References: <20211216210715.3801857-1-willy@infradead.org>
 <20211216210715.3801857-26-willy@infradead.org>
 <YrO243DkbckLTfP7@magnolia>
 <Yrku31ws6OCxRGSQ@magnolia>
 <Yrm6YM2uS+qOoPcn@casper.infradead.org>
 <YrosM1+yvMYliw2l@magnolia>
 <20220628073120.GI227878@dread.disaster.area>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20220628073120.GI227878@dread.disaster.area>
Precedence: bulk

On Tue, Jun 28, 2022 at 05:31:20PM +1000, Dave Chinner wrote:
> So using this technique, I've discovered that there's a dirty page
> accounting leak that eventually results in fsx hanging in
> balance_dirty_pages().

Alas, I think this is only an accounting error, and not related to
the problem(s) that Darrick & Zorro are seeing.  I think what you're
seeing is dirty pages being dropped at truncation without the
appropriate accounting.  ie this should be the fix:

+++ b/mm/huge_memory.c
@@ -2443,6 +2443,8 @@ static void __split_huge_page(struct page *page, struct list_head *list,
                        __delete_from_page_cache(head + i, NULL);
                        if (shmem_mapping(head->mapping))
                                shmem_uncharge(head->mapping->host, 1);
+                       else
+                               folio_account_cleaned(page_folio(head + i));
                        put_page(head + i);
                } else if (!PageAnon(page)) {
                        __xa_store(&head->mapping->i_pages, head[i].index,