From owner-freebsd-current@freebsd.org Fri Sep 18 09:03:17 2015 Return-Path: Delivered-To: freebsd-current@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 37B399CF257 for ; Fri, 18 Sep 2015 09:03:17 +0000 (UTC) (envelope-from grembo@freebsd.org) Received: from mail.grem.de (outcast.grem.de [213.239.217.27]) by mx1.freebsd.org (Postfix) with SMTP id 9F33A1C3E for ; Fri, 18 Sep 2015 09:03:16 +0000 (UTC) (envelope-from grembo@freebsd.org) Received: (qmail 40801 invoked by uid 89); 18 Sep 2015 08:56:34 -0000 Received: from unknown (HELO ?192.168.250.192?) (mg@grem.de@88.217.180.156) by mail.grem.de with ESMTPA; 18 Sep 2015 08:56:34 -0000 Content-Type: text/plain; charset=utf-8 Mime-Version: 1.0 (1.0) Subject: Re: Intel Haswell support - Any updates? From: Michael Gmelin X-Mailer: iPhone Mail (13A344) In-Reply-To: <20150918053445.GA1929@c720-r276659> Date: Fri, 18 Sep 2015 10:56:34 +0200 Cc: Nikola Pajkovsky , Stefan Wendler , freebsd-current@freebsd.org Content-Transfer-Encoding: quoted-printable Message-Id: <39E6D124-98A1-4289-AD3D-CFAA372C3074@freebsd.org> References: <55FA7324.5010603@icloud.com> <55FADB38.7000406@pcbsd.org> <87twqt2gs9.fsf@gooddata.com> <11271727.FrCOrXGY1l@hactar> <87a8sk205y.fsf@gooddata.com> <20150918053445.GA1929@c720-r276659> To: Matthias Apitz X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.20 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 18 Sep 2015 09:03:17 -0000 > On 18 Sep 2015, at 07:34, Matthias Apitz wrote: >=20 >> El d=C3=ADa Thursday, September 17, 2015 a las 11:42:01PM +0200, Nikola P= ajkovsky escribi=C3=B3: >>=20 >> Do not run those commands in X, otherwise it hangs. >>=20 >> $ vidcontrol -i mode | grep 1920 >> 383 (0x17f) 0x0000000f G 1920x1080x32 D 8x16 0xa0000 64k 64k 0xe0000000= 8100k >=20 > On my C720 the above cmd does not give any usefull information: >=20 >=20 > mode# flags type size font window linear buffe= r > --------------------------------------------------------------------------= ---- > 0 (0x000) 0x00000000 T 0x0 0x402653478 0x280996d6 131362k 0k= 0x00000000 0k > 1 (0x001) 0x00000000 T 0x0 0x402653478 0x280996d6 131362k 0k= 0x00000000 0k > 2 (0x002) 0x00000000 T 0x0 0x402653478 0x280996d6 131362k 0k= 0x00000000 0k > 3 (0x003) 0x00000000 T 0x0 0x402653478 0x280996d6 131362k 0k= 0x00000000 0k > 4 (0x004) 0x00000000 T 0x0 0x402653478 0x280996d6 131362k 0k= 0x00000000 0k > 5 ... >=20 >=20 > 507 (0x1fb) 0x00000000 T 0x0 0x402653478 0x280996d6 131362k 0= k 0x00000000 0k > 508 (0x1fc) 0x00000000 T 0x0 0x402653478 0x280996d6 131362k 0= k 0x00000000 0k > 509 (0x1fd) 0x00000000 T 0x0 0x402653478 0x280996d6 131362k 0= k 0x00000000 0k > 510 (0x1fe) 0x00000000 T 0x0 0x402653478 0x280996d6 131362k 0= k 0x00000000 0k > 511 (0x1ff) 0x00000000 T 0x0 0x402653478 0x280996d6 131362k 0= k 0x00000000 0k >=20 Last time I checked on the c720, the correct memory areas were marked as wri= te-combine by default (see "memcontrol list"), so you can't accomplish any f= urther performance gains by tuning it manually. - m