[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <20230504233858.103068-1-mike.kravetz@oracle.com>
Date: Thu, 4 May 2023 16:38:57 -0700
From: Mike Kravetz <mike.kravetz@...cle.com>
To: linux-kernel@...r.kernel.org, linux-mm@...ck.org,
linux-fsdevel@...r.kernel.org
Cc: Matthew Wilcox <willy@...radead.org>,
Ackerley Tng <ackerleytng@...gle.com>,
Sidhartha Kumar <sidhartha.kumar@...cle.com>,
Muchun Song <songmuchun@...edance.com>, vannapurve@...gle.com,
erdemaktas@...gle.com, Andrew Morton <akpm@...ux-foundation.org>,
Mike Kravetz <mike.kravetz@...cle.com>
Subject: [PATCH 0/1] fix page_cache_next/prev_miss off by one error
A cover letter is not necessary for this single patch as all information
is present in the commit message. However, I am not 100% comfortable in
this change and would REALLY like to get comments from Matthew.
When reporting this issue, Ackerley Tng suggested a solution by creating
a new filemap_has_folio() function[1]. I believe that would be an
acceptable way to proceed although we would also need to change the
other use of page_cache_next_miss in hugetlb.c.
When looking more closely, it looks like page_cache_next/prev_miss do
not work exactly as described. The result is the following patch.
IIUC, prior to hugetlb use of page_cache_next/prev_miss, it was only
used by readahead code. My patch does change the return value and has
potential to impact the readahead users. That is why I am not 100%
comfortable with this.
In any case, this is broken in v6.3 so we need a fix.
[1] https://lore.kernel.org/linux-mm/cover.1683069252.git.ackerleytng@google.com/
Mike Kravetz (1):
page cache: fix page_cache_next/prev_miss off by one
mm/filemap.c | 26 ++++++++++++++++----------
1 file changed, 16 insertions(+), 10 deletions(-)
--
2.40.0
Powered by blists - more mailing lists