From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751565AbcDPD34 (ORCPT ); Fri, 15 Apr 2016 23:29:56 -0400 Received: from linuxhacker.ru ([217.76.32.60]:49116 "EHLO fiona.linuxhacker.ru" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751069AbcDPD3z (ORCPT ); Fri, 15 Apr 2016 23:29:55 -0400 From: green@linuxhacker.ru To: Jens Axboe , "Kirill A. Shutemov" Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org, Tejun Heo , Michal Hocko , Oleg Drokin Subject: [PATCH] mm: Do not discard partial pages with POSIX_FADV_DONTNEED Date: Fri, 15 Apr 2016 23:28:54 -0400 Message-Id: <1460777334-3484107-1-git-send-email-green@linuxhacker.ru> X-Mailer: git-send-email 2.1.0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Oleg Drokin I noticed that the logic in fadvise64_64 syscall is incorrect for partial pages. While first page of the region is correctly skipped if it is partial, the last page of the region is mistakenly discarded. This leads to problems for applications that read data in non-page-aligned chunks discarding already processed data between the reads. Signed-off-by: Oleg Drokin --- mm/fadvise.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/mm/fadvise.c b/mm/fadvise.c index b8024fa..6c707bf 100644 --- a/mm/fadvise.c +++ b/mm/fadvise.c @@ -126,6 +126,17 @@ SYSCALL_DEFINE4(fadvise64_64, int, fd, loff_t, offset, loff_t, len, int, advice) */ start_index = (offset+(PAGE_SIZE-1)) >> PAGE_SHIFT; end_index = (endbyte >> PAGE_SHIFT); + if ((endbyte & ~PAGE_MASK) != ~PAGE_MASK) { + /* First page is tricky as 0 - 1 = -1, but pgoff_t + * is unsigned, so the end_index >= start_index + * check below would be true and we'll discard the whole + * file cache which is not what was asked. + */ + if (end_index == 0) + break; + + end_index--; + } if (end_index >= start_index) { unsigned long count = invalidate_mapping_pages(mapping, -- 2.1.0