From owner-freebsd-x11@freebsd.org Fri Mar 18 00:54:23 2016 Return-Path: Delivered-To: freebsd-x11@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 2F098AD43EE for ; Fri, 18 Mar 2016 00:54:23 +0000 (UTC) (envelope-from am_dxer@fastmail.fm) Received: from out2-smtp.messagingengine.com (out2-smtp.messagingengine.com [66.111.4.26]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 03C66C6B for ; Fri, 18 Mar 2016 00:54:22 +0000 (UTC) (envelope-from am_dxer@fastmail.fm) Received: from compute4.internal (compute4.nyi.internal [10.202.2.44]) by mailout.nyi.internal (Postfix) with ESMTP id 98F4020D9A for ; Thu, 17 Mar 2016 20:54:16 -0400 (EDT) Received: from web1 ([10.202.2.211]) by compute4.internal (MEProxy); Thu, 17 Mar 2016 20:54:16 -0400 DKIM-Signature: v=1; a=rsa-sha1; c=relaxed/relaxed; d=fastmail.fm; h=cc :content-transfer-encoding:content-type:date:from:in-reply-to :message-id:mime-version:references:subject:to:x-sasl-enc :x-sasl-enc; s=mesmtp; bh=Cdn9Oh8SjorjFZuMkmv75tyPvFo=; b=e6evFD OxJcOY2vLGhHVNoupE7/Z0eQnckzlFFYU8ao06Z829/J1MssWmG9fT7TLlZL4IDj 9E3F/ENpSgMe2TOKzXUY4H2OiiJVaGJXqh8fO/RVvq4h1PzMIk9hTNVQ9FeWPp4P 7M3BbGApd9ZFLGBhULQBRivCRpOy6PbtTa28s= DKIM-Signature: v=1; a=rsa-sha1; c=relaxed/relaxed; d= messagingengine.com; h=cc:content-transfer-encoding:content-type :date:from:in-reply-to:message-id:mime-version:references :subject:to:x-sasl-enc:x-sasl-enc; s=smtpout; bh=Cdn9Oh8SjorjFZu Mkmv75tyPvFo=; b=tTmgICS2KbS50AoI2iUirVrADlP6IAq4+TDY7I3rjLKHsey zFyCRHe6OIPkQmeWSI6VVWFOvNIE8h1hHCyqDhIsOfm5y4Qpm7yCBhuMmRdSO8cO LsGr0SeWcFCBAoC57ZtfXImmq6l0+oX+2CLBLfW6e/mTn1NDBtrfLd8NC5F0= Received: by web1.nyi.internal (Postfix, from userid 99) id 63989AED468; Thu, 17 Mar 2016 20:54:16 -0400 (EDT) Message-Id: <1458262456.542489.552574226.316C2F67@webmail.messagingengine.com> X-Sasl-Enc: jsq/ppCSuLKlHTBxzMXrfClu6QmvN1vXyLzlmCWjk1yj 1458262456 From: am_dxer@fastmail.fm To: Ed Maste Cc: "freebsd-x11@freebsd.org" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Type: text/plain X-Mailer: MessagingEngine.com Webmail Interface - html In-Reply-To: References: <1457974114.3576662.548762770.0E1628BF@webmail.messagingengine.com> <20160315171920.GZ1741@kib.kiev.ua> <1458086366.4068324.550308746.4DBF0576@webmail.messagingengine.com> Subject: Re: i915 update to Linux 3.8 ready! Date: Thu, 17 Mar 2016 20:54:16 -0400 X-BeenThere: freebsd-x11@freebsd.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: X11 on FreeBSD -- maintaining and support List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 18 Mar 2016 00:54:23 -0000 On Thu, Mar 17, 2016, at 02:16 PM, Ed Maste wrote: > On 17 March 2016 at 11:20, Ed Maste wrote: > > On 15 March 2016 at 19:59, wrote: > >> > >> Thank you both for the suggestions. After suspending and resuming the > >> machine, I find that the power used by the GPU is significantly less. > > > > Thank you for testing, and I see the same behaviour on my Thinkpad > > X220: after loading i915kms.ko only RC6 is nonzero and RC6+ and RC6++ > > remain at zero, while after > > boot RC6++ is much higher. > > I tried reverting the Sandy Bridge part of the change that was > identified as the source of the similar problem in Linux, and it fixes > the power consumption issue for me. > > My partial revert is just removes the write to the GEN6_MBCTL register > in gen6_init_clock_gating(). Of course there's some reason it was > added in Linux, but the driver was functional on Sandy Bridge and > earlier hardware for years prior to the recent FreeBSD i915 update, > without that register write. Thus I think it's reasonable to disable > it until we either fully understand the problem or (more likely) just > update to a later Linux i915 version and verify normal Sandy Bridge > power consumption. > > The patch: https://people.freebsd.org/~emaste/D5665.diff > > Code review: https://reviews.freebsd.org/D5665 I applied your patch today and compiled a new kernel. The power consumption in the EliteBook is also back to normal after loading the i915kms driver. Suspending and resuming the machine is no longer required. Thank you for the help and suggestions and for solving this issue.