From mboxrd@z Thu Jan 1 00:00:00 1970 From: Chao Yu Subject: Re: [f2fs-dev] [PATCH] f2fs: fix unnecessary periodic wakeup of discard thread when dev is busy Date: Sun, 2 Sep 2018 16:52:40 +0800 Message-ID: References: <1535708366-11318-1-git-send-email-stummala@codeaurora.org> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1535708366-11318-1-git-send-email-stummala@codeaurora.org> Content-Language: en-US Sender: linux-kernel-owner@vger.kernel.org To: Sahitya Tummala , Jaegeuk Kim , Chao Yu , linux-f2fs-devel@lists.sourceforge.net Cc: linux-kernel@vger.kernel.org List-Id: linux-f2fs-devel.lists.sourceforge.net On 2018/8/31 17:39, Sahitya Tummala wrote: > When dev is busy, discard thread wake up timeout can be aligned with the > exact time that it needs to wait for dev to come out of busy. This helps > to avoid unnecessary periodic wakeups and thus save some power. > > Signed-off-by: Sahitya Tummala > --- > fs/f2fs/segment.c | 8 +++++++- > 1 file changed, 7 insertions(+), 1 deletion(-) > > diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c > index 8bcbb50..df14030 100644 > --- a/fs/f2fs/segment.c > +++ b/fs/f2fs/segment.c > @@ -1379,6 +1379,8 @@ static int issue_discard_thread(void *data) > struct discard_policy dpolicy; > unsigned int wait_ms = DEF_MIN_DISCARD_ISSUE_TIME; > int issued; > + unsigned long interval = sbi->interval_time[REQ_TIME] * HZ; > + long delta; > > set_freezable(); > > @@ -1410,7 +1412,11 @@ static int issue_discard_thread(void *data) > __wait_all_discard_cmd(sbi, &dpolicy); > wait_ms = dpolicy.min_interval; > } else if (issued == -1){ > - wait_ms = dpolicy.mid_interval; > + delta = (sbi->last_time[REQ_TIME] + interval) - jiffies; I agree that we need to consider power consumption. One more consideration is that discard thread may need different submission frequency comparing to garbage collection thread, maybe a little fast, would it be better to split sbi->interval_time[REQ_TIME] according to gc/discard type. How do you think? Thanks, > + if (delta > 0) > + wait_ms = jiffies_to_msecs(delta); > + else > + wait_ms = dpolicy.mid_interval; > } else { > wait_ms = dpolicy.max_interval; > } >