[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <Y4Z7caUUaS+nV+bk@x1n>
Date: Tue, 29 Nov 2022 16:36:49 -0500
From: Peter Xu <peterx@...hat.com>
To: Andrew Morton <akpm@...ux-foundation.org>
Cc: linux-mm@...ck.org, linux-kernel@...r.kernel.org,
James Houghton <jthoughton@...gle.com>,
Jann Horn <jannh@...gle.com>,
Andrea Arcangeli <aarcange@...hat.com>,
Rik van Riel <riel@...riel.com>,
Nadav Amit <nadav.amit@...il.com>,
Miaohe Lin <linmiaohe@...wei.com>,
Muchun Song <songmuchun@...edance.com>,
Mike Kravetz <mike.kravetz@...cle.com>,
David Hildenbrand <david@...hat.com>
Subject: Re: [PATCH 00/10] mm/hugetlb: Make huge_pte_offset() thread-safe for
pmd unshare
On Tue, Nov 29, 2022 at 12:51:17PM -0800, Andrew Morton wrote:
> On Tue, 29 Nov 2022 14:35:16 -0500 Peter Xu <peterx@...hat.com> wrote:
>
> > [ 17.975943] Oops: 0000 [#1] PREEMPT SMP NOPTI
>
> Do we know which kernel versions are affected here?
Since lockless walk of huge_pte_offset() existed since the initial git
commit, it should be the time when we introduced pmd sharing for hugetlb,
which means any kernel after commit 39dde65c9940 ("[PATCH] shared page
table for hugetlb page", 2006-12-07) should be prone to the issue at least
for x86 (as pmd sharing was proposed initially only for x86).
--
Peter Xu
Powered by blists - more mailing lists