From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S965763AbXAYKkX (ORCPT ); Thu, 25 Jan 2007 05:40:23 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S965765AbXAYKkX (ORCPT ); Thu, 25 Jan 2007 05:40:23 -0500 Received: from mx1-2.mail.ru ([194.67.23.121]:25307 "EHLO mx1.mail.ru" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S965763AbXAYKkU (ORCPT ); Thu, 25 Jan 2007 05:40:20 -0500 Date: Thu, 25 Jan 2007 13:45:38 +0300 From: Evgeniy Dushistov To: Andrew Morton Cc: linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org Subject: Re: [PATCH 3/3] ufs: rellocation fix Message-ID: <20070125104538.GA4037@rain> Mail-Followup-To: Andrew Morton , linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org References: <20070122150751.GA11129@rain> <20070124202107.4788d802.akpm@osdl.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20070124202107.4788d802.akpm@osdl.org> User-Agent: Mutt/1.5.13 (2006-08-11) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Jan 24, 2007 at 08:21:07PM -0800, Andrew Morton wrote: > On Mon, 22 Jan 2007 18:07:51 +0300 > Evgeniy Dushistov wrote: > > > - for (i = 0; i < count; i += blk_per_page) { > > + for (i = 0; i < count; i = (i | mask) + 1) { > > This is a funny looking thing. As far as I can tell, this is a more > complicated way of doing the same thing as the old code did. > > Am I missing something? > No, you right, thanks, work should be done with baseblk+i, not with i. (It is imposible to catch on my test enviroment: size of page 4096 and baseblk always point to begin of data block which size of >=4096 and power of 2). Here is patch for patch, it also contains typo fixing in previous patch which lead to compilation failure if ufs debug turn on. May be it is possible to integrate this patch into previous one, or resend new version of last patch? --- Index: linux-2.6.20-rc5/fs/ufs/balloc.c =================================================================== --- linux-2.6.20-rc5.orig/fs/ufs/balloc.c +++ linux-2.6.20-rc5/fs/ufs/balloc.c @@ -227,14 +227,14 @@ failed: * We can come here from ufs_writepage or ufs_prepare_write, * locked_page is argument of these functions, so we already lock it. */ -static void ufs_change_blocknr(struct inode *inode, unsigned int baseblk, +static void ufs_change_blocknr(struct inode *inode, unsigned int beg, unsigned int count, unsigned int oldb, unsigned int newb, struct page *locked_page) { const unsigned mask = (1 << (PAGE_CACHE_SHIFT - inode->i_blkbits)) - 1; struct address_space * const mapping = inode->i_mapping; pgoff_t index, cur_index; - unsigned i, pos, j; + unsigned end, pos, j; struct page *page; struct buffer_head *head, *bh; @@ -246,8 +246,8 @@ static void ufs_change_blocknr(struct in cur_index = locked_page->index; - for (i = 0; i < count; i = (i | mask) + 1) { - index = (baseblk+i) >> (PAGE_CACHE_SHIFT - inode->i_blkbits); + for (end = count + beg; beg < end; beg = (beg | mask) + 1) { + index = beg >> (PAGE_CACHE_SHIFT - inode->i_blkbits); if (likely(cur_index != index)) { page = ufs_get_locked_page(mapping, index); @@ -258,7 +258,7 @@ static void ufs_change_blocknr(struct in head = page_buffers(page); bh = head; - pos = i & mask; + pos = beg & mask; for (j = 0; j < pos; ++j) bh = bh->b_this_page; j = 0; @@ -267,7 +267,7 @@ static void ufs_change_blocknr(struct in pos = bh->b_blocknr - oldb; if (pos < count) { UFSD(" change from %llu to %llu\n", - (unsigned long long)pos + odlb, + (unsigned long long)pos + oldb, (unsigned long long)pos + newb); bh->b_blocknr = newb + pos; unmap_underlying_metadata(bh->b_bdev, -- /Evgeniy