From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tejun Heo Subject: Re: [PATCH 02/10] freezer: skip waking up tasks with PF_FREEZER_SKIP set Date: Mon, 29 Apr 2013 15:08:31 -0700 Message-ID: <20130429220831.GC2395@htj.dyndns.org> References: <1367271946-7239-1-git-send-email-ccross@android.com> <1367271946-7239-3-git-send-email-ccross@android.com> <20130429215157.GA2395@htj.dyndns.org> <20130429215745.GB2395@htj.dyndns.org> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Content-Disposition: inline In-Reply-To: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: containers-bounces-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org Errors-To: containers-bounces-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org To: Colin Cross Cc: Len Brown , Linux PM list , containers-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org, lkml , "Rafael J. Wysocki" , Arve =?iso-8859-1?B?SGr4bm5lduVn?= , Pavel Machek , cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: linux-pm@vger.kernel.org Hey, On Mon, Apr 29, 2013 at 03:02:19PM -0700, Colin Cross wrote: > See the first patch in the series (which isn't available in the > archive yet, so I can't link to it). The short version is that It didn't arrive in my lkml folder either. Maybe vger is taking some time distributing emails. > Android goes through suspend/resume very often (every few seconds when > on a busy wifi network with the screen off), and a significant portion > of the energy used to go in and out of suspend is spent in the > freezer. This patch series takes the most common userspace sleep > points and converts them to PF_FREEZER_SKIP, which reduces the number > of context switches for every suspend or resume event on a > freshly-booted Android device from 1000 to 25, and reduces the time Ah, okay, so you're spreading PF_FREEZER_SKIP. When you post patches which touch the freezer can you please cc me and Oleg Nesterov (I'll ping him this time)? Freezer has been very subtly broken in various ways and many kthread users are still broken, so let's tread carefully. > spent freezing by a factor of 5. It will have a similar effect on a > non-Android system, although those generally don't care about > suspend/resume optimization. Yeah, if it's something which makes actual difference rather than "this seems to be a good idea" thing, sure, let's find a way. Thanks. -- tejun