From mboxrd@z Thu Jan 1 00:00:00 1970 From: Wu Fengguang Subject: Re: [PATCH 4/5] writeback: fix dirtied pages accounting on redirty Date: Tue, 22 Nov 2011 21:59:19 +0800 Message-ID: <20111122135919.GB10545@localhost> References: <20111121130342.211953629@intel.com> <20111121131216.032489128@intel.com> <20111121135146.0fc7e3a7.akpm@linux-foundation.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: "linux-fsdevel@vger.kernel.org" , Jan Kara , Peter Zijlstra , Christoph Hellwig , LKML , Chris Mason To: Andrew Morton Return-path: Received: from mga14.intel.com ([143.182.124.37]:11025 "EHLO mga14.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755352Ab1KVN7Z (ORCPT ); Tue, 22 Nov 2011 08:59:25 -0500 Content-Disposition: inline In-Reply-To: <20111121135146.0fc7e3a7.akpm@linux-foundation.org> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On Tue, Nov 22, 2011 at 05:51:46AM +0800, Andrew Morton wrote: > On Mon, 21 Nov 2011 21:03:46 +0800 > Wu Fengguang wrote: > > > De-account the accumulative dirty counters on page redirty. > > > > Page redirties (very common in ext4) will introduce mismatch between > > counters (a) and (b) > > > > a) NR_DIRTIED, BDI_DIRTIED, tsk->nr_dirtied > > b) NR_WRITTEN, BDI_WRITTEN > > > > This will introduce systematic errors in balanced_rate and result in > > dirty page position errors (ie. the dirty pages are no longer balanced > > around the global/bdi setpoints). > > > > ... > > > > --- linux-next.orig/mm/page-writeback.c 2011-11-17 20:57:15.000000000 +0800 > > +++ linux-next/mm/page-writeback.c 2011-11-17 20:57:16.000000000 +0800 > > @@ -1792,6 +1792,17 @@ int __set_page_dirty_nobuffers(struct pa > > } > > EXPORT_SYMBOL(__set_page_dirty_nobuffers); > > > > +void account_page_redirty(struct page *page) > > +{ > > + struct address_space *mapping = page->mapping; > > + if (mapping && mapping_cap_account_dirty(mapping)) { > > + current->nr_dirtied--; > > + dec_zone_page_state(page, NR_DIRTIED); > > + dec_bdi_stat(mapping->backing_dev_info, BDI_DIRTIED); > > + } > > +} > > Again, writeback doesn't seem to be the best place to be adding > undocumented code. Sorry, hope this comment can explain when to call the function and why. /* * Call this whenever redirtying a page, to de-account the dirty counters * (NR_DIRTIED, BDI_DIRTIED, tsk->nr_dirtied), so that they match the written * counters (NR_WRITTEN, BDI_WRITTEN) in long term. The mismatches will lead to * systematic errors in balanced_dirty_ratelimit and the dirty pages position * control. */ > > +EXPORT_SYMBOL(account_page_redirty); > > The export is unneeded? btrfs will need to call into it. Here is the patch. I'll include it in the v2 series. Thanks, Fengguang --- Subject: btrfs: fix dirtied pages accounting on sub-page writes Date: Mon Aug 08 15:19:47 CST 2011 When doing 1KB sequential writes to the same page, balance_dirty_pages_ratelimited_nr() should be called once instead of 4 times, the latter makes the dirtier tasks be throttled much too heavy. Fix it with proper de-accounting on clear_page_dirty_for_io(). CC: Chris Mason Signed-off-by: Wu Fengguang --- fs/btrfs/file.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) --- linux-next.orig/fs/btrfs/file.c 2011-11-17 20:13:47.000000000 +0800 +++ linux-next/fs/btrfs/file.c 2011-11-17 20:18:51.000000000 +0800 @@ -1136,7 +1136,8 @@ again: GFP_NOFS); } for (i = 0; i < num_pages; i++) { - clear_page_dirty_for_io(pages[i]); + if (clear_page_dirty_for_io(pages[i])) + account_page_redirty(pages[i]); set_page_extent_mapped(pages[i]); WARN_ON(!PageLocked(pages[i])); }