From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-bk0-f46.google.com ([209.85.214.46]:43515 "EHLO mail-bk0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751475Ab2I0G2O (ORCPT ); Thu, 27 Sep 2012 02:28:14 -0400 MIME-Version: 1.0 In-Reply-To: <20120927035425.GN15236@dastard> References: <1348404995-14372-1-git-send-email-zwu.kernel@gmail.com> <1348404995-14372-7-git-send-email-zwu.kernel@gmail.com> <20120927035425.GN15236@dastard> Date: Thu, 27 Sep 2012 14:28:12 +0800 Message-ID: Subject: Re: [RFC v2 06/10] vfs: enable hot data tracking From: Zhi Yong Wu To: Dave Chinner Cc: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, linux-btrfs@vger.kernel.org, linux-ext4@vger.kernel.org, linuxram@linux.vnet.ibm.com, viro@zeniv.linux.org.uk, cmm@us.ibm.com, tytso@mit.edu, marco.stornelli@gmail.com, stroetmann@ontolinux.com, diegocg@gmail.com, chris@csamuel.org, Zhi Yong Wu Content-Type: text/plain; charset=ISO-8859-1 Sender: linux-btrfs-owner@vger.kernel.org List-ID: On Thu, Sep 27, 2012 at 11:54 AM, Dave Chinner wrote: > On Sun, Sep 23, 2012 at 08:56:31PM +0800, zwu.kernel@gmail.com wrote: >> From: Zhi Yong Wu >> >> Miscellaneous features that implement hot data tracking >> and generally make the hot data functions a bit more friendly. >> >> Signed-off-by: Zhi Yong Wu >> --- >> fs/direct-io.c | 10 ++++++++++ >> include/linux/hot_tracking.h | 11 +++++++++++ >> mm/filemap.c | 8 ++++++++ >> mm/page-writeback.c | 21 +++++++++++++++++++++ >> mm/readahead.c | 9 +++++++++ >> 5 files changed, 59 insertions(+), 0 deletions(-) >> >> diff --git a/fs/direct-io.c b/fs/direct-io.c >> index f86c720..3773f44 100644 >> --- a/fs/direct-io.c >> +++ b/fs/direct-io.c >> @@ -37,6 +37,7 @@ >> #include >> #include >> #include >> +#include "hot_tracking.h" >> >> /* >> * How many user pages to map in one call to get_user_pages(). This determines >> @@ -1297,6 +1298,15 @@ __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, >> prefetch(bdev->bd_queue); >> prefetch((char *)bdev->bd_queue + SMP_CACHE_BYTES); >> >> + /* Hot data tracking */ >> + if (TRACK_THIS_INODE(iocb->ki_filp->f_mapping->host) >> + && iov_length(iov, nr_segs) > 0) { >> + hot_rb_update_freqs(iocb->ki_filp->f_mapping->host, >> + (u64)offset, >> + (u64)iov_length(iov, nr_segs), >> + rw & WRITE); >> + } > > That's a bit messy. I'd prefer a static inline function that hides > all this. e.g. Do you think of moving the condition into hot_inode_udate_freqs(), not adding another new function? > > track_hot_inode_ranges(inode, offset, length, rw) > { > if (inode->i_sb->s_flags & MS_HOT_TRACKING) > hot_inode_freq_update(inode, offset, length, rw); > } > >> diff --git a/mm/page-writeback.c b/mm/page-writeback.c >> index 5ad5ce2..552c861 100644 >> --- a/mm/page-writeback.c >> +++ b/mm/page-writeback.c >> @@ -35,6 +35,7 @@ >> #include /* __set_page_dirty_buffers */ >> #include >> #include >> +#include >> #include >> >> /* >> @@ -1895,13 +1896,33 @@ EXPORT_SYMBOL(generic_writepages); >> int do_writepages(struct address_space *mapping, struct writeback_control *wbc) >> { >> int ret; >> + pgoff_t start = 0; >> + u64 prev_count = 0, count = 0; >> >> if (wbc->nr_to_write <= 0) >> return 0; >> + >> + /* Hot data tracking */ >> + if (TRACK_THIS_INODE(mapping->host) >> + && wbc->range_cyclic) { >> + start = mapping->writeback_index << PAGE_CACHE_SHIFT; >> + prev_count = (u64)wbc->nr_to_write; >> + } > > Why only wbc->range_cyclic? This won't record things like > synchronous writes or fsync-triggered writes, are are far more > likely to be to hot ranges in a file... sorry, i don't undersand what wbc->range_cyclic means. OK, i will fix it in next version. > > Cheers, > > Dave. > -- > Dave Chinner > david@fromorbit.com -- Regards, Zhi Yong Wu