Received-SPF: pass (google.com: domain of linux-kernel+bounces-18287-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.48.161 as permitted sender) client-ip=147.75.48.161;
Message-ID: <b074426bb17af70b9db98e7de7023180822a8c40.camel@surriel.com>
Subject: Re: [PATCH 0/1] selftest/mm/hugetlb: SIGBUS on stolen page
From: Rik van Riel <riel@surriel.com>
To: Andrew Morton <akpm@linux-foundation.org>, Breno Leitao
 <leitao@debian.org>
Cc: linux-mm@kvack.org, vegard.nossum@oracle.com, rppt@kernel.org, 
 songmuchun@bytedance.com, shy828301@gmail.com,
 linux-kselftest@vger.kernel.org,  linux-kernel@vger.kernel.org
Date: Fri, 05 Jan 2024 14:00:58 -0500
In-Reply-To: <20240105084238.306269c3f3a63cb0fd130baa@linux-foundation.org>
References: <20240105155419.1939484-1-leitao@debian.org>
	 <20240105084238.306269c3f3a63cb0fd130baa@linux-foundation.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
User-Agent: Evolution 3.46.4 (3.46.4-1.fc37) 
Precedence: bulk
MIME-Version: 1.0
Sender: riel@surriel.com

On Fri, 2024-01-05 at 08:42 -0800, Andrew Morton wrote:
> On Fri,=C2=A0 5 Jan 2024 07:54:18 -0800 Breno Leitao <leitao@debian.org>
> wrote:
>=20
> > This test case triggers a race between madvise(MADV_DONTNEED) and
> > mmap() in a single huge page, which got stolen (while reserved).
> >=20
> > Once the only page is stolen, the memory previously mmaped (and
> > madvise(MADV_DONTNEED) got a SIGBUS when accessed.
> >=20
> > I am not adding this test to the un_vmtests.sh scripts, since this
> > test
> > fails at upstream.
>=20
> Oh.=C2=A0 Is a fix for this in the pipeline?=C2=A0 If so, I assume that o=
nce
> the
> fix is merged, we enable this test in run_vmtests?
>=20
I've got some ideas on how to fix it, and hope to get
a fix to you and Mike by next week.

I'll ask Mike if I run into any unexpected complications.

--=20
All Rights Reversed.