Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:18 as permitted sender) client-ip=2620:137:e000::1:18;
From:   Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To:     linux-kernel@vger.kernel.org
Cc:     Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        stable@vger.kernel.org, Andreas Gruenbacher <agruenba@redhat.com>,
        Sasha Levin <sashal@kernel.org>
Subject: [PATCH 5.17 026/158] gfs2: Disable page faults during lockless buffered reads
Date:   Mon, 23 May 2022 19:03:03 +0200
Message-Id: <20220523165834.860951526@linuxfoundation.org>
In-Reply-To: <20220523165830.581652127@linuxfoundation.org>
References: <20220523165830.581652127@linuxfoundation.org>
User-Agent: quilt/0.66
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Precedence: bulk

From: Andreas Gruenbacher <agruenba@redhat.com>

[ Upstream commit 52f3f033a5dbd023307520af1ff551cadfd7f037 ]

During lockless buffered reads, filemap_read() holds page cache page
references while trying to copy data to the user-space buffer.  The
calling process isn't holding the inode glock, but the page references
it holds prevent those pages from being removed from the page cache, and
that prevents the underlying inode glock from being moved to another
node.  Thus, we can end up in the same kinds of distributed deadlock
situations as with normal (non-lockless) buffered reads.

Fix that by disabling page faults during lockless reads as well.

Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/gfs2/file.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c
index fa071d738c78..c781c19303db 100644
--- a/fs/gfs2/file.c
+++ b/fs/gfs2/file.c
@@ -956,14 +956,16 @@ static ssize_t gfs2_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
 			return ret;
 		iocb->ki_flags &= ~IOCB_DIRECT;
 	}
+	pagefault_disable();
 	iocb->ki_flags |= IOCB_NOIO;
 	ret = generic_file_read_iter(iocb, to);
 	iocb->ki_flags &= ~IOCB_NOIO;
+	pagefault_enable();
 	if (ret >= 0) {
 		if (!iov_iter_count(to))
 			return ret;
 		written = ret;
-	} else {
+	} else if (ret != -EFAULT) {
 		if (ret != -EAGAIN)
 			return ret;
 		if (iocb->ki_flags & IOCB_NOWAIT)
-- 
2.35.1