Return-Path: linux-nfs-owner@vger.kernel.org Received: from mail-ig0-f178.google.com ([209.85.213.178]:54882 "EHLO mail-ig0-f178.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757973AbaDVV3Y (ORCPT ); Tue, 22 Apr 2014 17:29:24 -0400 Received: by mail-ig0-f178.google.com with SMTP id hn18so217399igb.11 for ; Tue, 22 Apr 2014 14:29:23 -0700 (PDT) From: Weston Andros Adamson To: trond.myklebust@primarydata.com Cc: linux-nfs@vger.kernel.org, Weston Andros Adamson Subject: [PATCH 17/17] nfs: support page groups in nfs_read_completion Date: Tue, 22 Apr 2014 17:29:25 -0400 Message-Id: <1398202165-78897-18-git-send-email-dros@primarydata.com> In-Reply-To: <1398202165-78897-1-git-send-email-dros@primarydata.com> References: <1398202165-78897-1-git-send-email-dros@primarydata.com> Sender: linux-nfs-owner@vger.kernel.org List-ID: nfs_read_completion relied on the fact that there was a 1:1 mapping of page to nfs_request, but this has now changed. Regions not covered by a request have already been zeroed elsewhere. Signed-off-by: Weston Andros Adamson --- fs/nfs/read.c | 24 +++++++++++++++++------- 1 file changed, 17 insertions(+), 7 deletions(-) diff --git a/fs/nfs/read.c b/fs/nfs/read.c index c6b7dd0..e3613e2 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -182,7 +182,6 @@ static void nfs_page_group_set_uptodate(struct nfs_page *req) SetPageUptodate(req->wb_page); } -/* Note io was page aligned */ static void nfs_read_completion(struct nfs_pgio_header *hdr) { unsigned long bytes = 0; @@ -192,14 +191,25 @@ static void nfs_read_completion(struct nfs_pgio_header *hdr) while (!list_empty(&hdr->pages)) { struct nfs_page *req = nfs_list_entry(hdr->pages.next); struct page *page = req->wb_page; + unsigned long start = req->wb_pgbase; + unsigned long end = req->wb_pgbase + req->wb_bytes; if (test_bit(NFS_IOHDR_EOF, &hdr->flags)) { - if (bytes > hdr->good_bytes) - zero_user(page, 0, PAGE_SIZE); - else if (hdr->good_bytes - bytes < PAGE_SIZE) - zero_user_segment(page, - hdr->good_bytes & ~PAGE_MASK, - PAGE_SIZE); + /* note: regions of the page not covered by a + * request are zeroed in nfs_readpage_async / + * readpage_async_filler */ + if (bytes > hdr->good_bytes) { + /* nothing in this request was good, so zero + * the full extent of the request */ + zero_user_segment(page, start, end); + + } else if (hdr->good_bytes - bytes < req->wb_bytes) { + /* part of this request has good bytes, but + * not all. zero the bad bytes */ + start += hdr->good_bytes - bytes; + WARN_ON(start < req->wb_pgbase); + zero_user_segment(page, start, end); + } } bytes += req->wb_bytes; if (test_bit(NFS_IOHDR_ERROR, &hdr->flags)) { -- 1.8.5.2 (Apple Git-48)