From mboxrd@z Thu Jan 1 00:00:00 1970 From: Kevin Hilman Date: Fri, 31 May 2013 19:31:43 +0000 Subject: Re: [PATCH 05/32] OMAPDSS: fix dss_get_ctx_loss_count for DT Message-Id: <871u8nymhs.fsf@linaro.org> List-Id: References: <1369906493-27538-1-git-send-email-tomi.valkeinen@ti.com> <1369906493-27538-6-git-send-email-tomi.valkeinen@ti.com> <874ndk5sse.fsf@linaro.org> <51A8BF17.4050209@ti.com> In-Reply-To: <51A8BF17.4050209@ti.com> (Grygorii Strashko's message of "Fri, 31 May 2013 18:17:43 +0300") MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Grygorii Strashko Cc: Tomi Valkeinen , linux-fbdev@vger.kernel.org, linux-omap@vger.kernel.org, Archit Taneja Hi Grygorii, Grygorii Strashko writes: [...] > I've a question: > Why don't add API in "Kernel/Base PM" framework for context lost detection? > Like pm_was_ctx_lost(). I'm not aware of any proposals to do this in a generic way. Feel free to propose one if you're so moved. One complexity is that such a call is not needed on platforms that use generic power domains (not OMAP). With gen_pd, the runtime PM callbacks do not get called unless the power domains actually lose power. So for gen_pd, you assume context has been lost when your runtime PM callbacks are called. This is a bit different than how we've done it on OMAP, and there's been some work in exploring how to adapt OMAP to gen_pd, but there's not anyone actively working on it at the moment. That being said, I'm still leaning towards simply removing the context loss logic from all the OMAP drivers. This feature adds a relatively minor optimization, and a more configurable optimization could be done using runtime PM autosuspend timeouts (configurable from userspace.) Combined with the fact that adoption of gen_pd means we don't need it, I'm included to drop it all together. Kevin From mboxrd@z Thu Jan 1 00:00:00 1970 From: Kevin Hilman Subject: Re: [PATCH 05/32] OMAPDSS: fix dss_get_ctx_loss_count for DT Date: Fri, 31 May 2013 12:31:43 -0700 Message-ID: <871u8nymhs.fsf@linaro.org> References: <1369906493-27538-1-git-send-email-tomi.valkeinen@ti.com> <1369906493-27538-6-git-send-email-tomi.valkeinen@ti.com> <874ndk5sse.fsf@linaro.org> <51A8BF17.4050209@ti.com> Mime-Version: 1.0 Content-Type: text/plain Return-path: Received: from mail-pb0-f43.google.com ([209.85.160.43]:62487 "EHLO mail-pb0-f43.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756928Ab3EaTbq (ORCPT ); Fri, 31 May 2013 15:31:46 -0400 Received: by mail-pb0-f43.google.com with SMTP id ma3so2724977pbc.2 for ; Fri, 31 May 2013 12:31:46 -0700 (PDT) In-Reply-To: <51A8BF17.4050209@ti.com> (Grygorii Strashko's message of "Fri, 31 May 2013 18:17:43 +0300") Sender: linux-omap-owner@vger.kernel.org List-Id: linux-omap@vger.kernel.org To: Grygorii Strashko Cc: Tomi Valkeinen , linux-fbdev@vger.kernel.org, linux-omap@vger.kernel.org, Archit Taneja Hi Grygorii, Grygorii Strashko writes: [...] > I've a question: > Why don't add API in "Kernel/Base PM" framework for context lost detection? > Like pm_was_ctx_lost(). I'm not aware of any proposals to do this in a generic way. Feel free to propose one if you're so moved. One complexity is that such a call is not needed on platforms that use generic power domains (not OMAP). With gen_pd, the runtime PM callbacks do not get called unless the power domains actually lose power. So for gen_pd, you assume context has been lost when your runtime PM callbacks are called. This is a bit different than how we've done it on OMAP, and there's been some work in exploring how to adapt OMAP to gen_pd, but there's not anyone actively working on it at the moment. That being said, I'm still leaning towards simply removing the context loss logic from all the OMAP drivers. This feature adds a relatively minor optimization, and a more configurable optimization could be done using runtime PM autosuspend timeouts (configurable from userspace.) Combined with the fact that adoption of gen_pd means we don't need it, I'm included to drop it all together. Kevin