Return-Path: linux-nfs-owner@vger.kernel.org Received: from mail-ig0-f180.google.com ([209.85.213.180]:39926 "EHLO mail-ig0-f180.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755037AbaHNVjl (ORCPT ); Thu, 14 Aug 2014 17:39:41 -0400 Received: by mail-ig0-f180.google.com with SMTP id l13so221038iga.1 for ; Thu, 14 Aug 2014 14:39:41 -0700 (PDT) From: Weston Andros Adamson To: linux-nfs@vger.kernel.org Cc: Weston Andros Adamson Subject: [PATCH 2/2] nfs: can_coalesce_requests must enforce contiguity Date: Thu, 14 Aug 2014 17:39:33 -0400 Message-Id: <1408052373-4306-3-git-send-email-dros@primarydata.com> In-Reply-To: <1408052373-4306-1-git-send-email-dros@primarydata.com> References: <1408052373-4306-1-git-send-email-dros@primarydata.com> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Sender: linux-nfs-owner@vger.kernel.org List-ID: Commit 6094f83864c1d1296566a282cba05ba613f151ee "nfs: allow coalescing of subpage requests" got rid of the requirement that requests cover whole pages, but it made some incorrect assumptions. It turns out that callers of this interface can map adjacent requests (by file position as seen by req_offset + req->wb_bytes) to different pages, even when they could share a page. An example is the direct I/O interface - iov_iter_get_pages_alloc may return one segment with a partial page filled and the next segment (which is adjacent in the file position) starts with a new page. Reported-by: Toralf Förster Signed-off-by: Weston Andros Adamson --- fs/nfs/pagelist.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 73476df..d512887 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -836,6 +836,14 @@ static bool nfs_can_coalesce_requests(struct nfs_page *prev, return false; if (req_offset(req) != req_offset(prev) + prev->wb_bytes) return false; + if (req->wb_page == prev->wb_page) { + if (req->wb_pgbase != prev->wb_pgbase + prev->wb_bytes) + return false; + } else { + if (req->wb_pgbase != 0 || + prev->wb_pgbase + prev->wb_bytes != PAGE_CACHE_SIZE) + return false; + } } size = pgio->pg_ops->pg_test(pgio, prev, req); WARN_ON_ONCE(size > req->wb_bytes); -- 1.8.5.2 (Apple Git-48)