From owner-freebsd-arm@freebsd.org Fri Aug 17 15:47:51 2018 Return-Path: Delivered-To: freebsd-arm@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 1FA2510730A0 for ; Fri, 17 Aug 2018 15:47:51 +0000 (UTC) (envelope-from greg@unrelenting.technology) Received: from hraggstad.unrelenting.technology (hraggstad.unrelenting.technology [71.19.146.151]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client CN "hraggstad.unrelenting.technology", Issuer "Let's Encrypt Authority X3" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id A75DA8459D for ; Fri, 17 Aug 2018 15:47:50 +0000 (UTC) (envelope-from greg@unrelenting.technology) DKIM-Signature: v=1; a=rsa-sha256; c=simple/simple; d=unrelenting.technology; h=date:from:subject:to:message-id; s=default; bh=D+t0+TTq6wKQWZWk2Gn69aR+vscwlMREnlEvhewPbws=; b=mY01Iv4mny9q1t/rbM72s3pVbopcRLTehatThBXB9h7RMGbYTxpUlqzs9BzIrpDdVhPJh39YOm2F+xHFH/oXbSVAt3EapeYFhgFWZKqVcnL0/yYNnCc4YZ0pUBXbxwOz6HPSJzTZjvfh51kSwvKHJGMbFD+PPxGiZ9n9M8suwi8= Received: by hraggstad.unrelenting.technology (OpenSMTPD) with ESMTPSA id 2b9bf57b TLS version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NO; Fri, 17 Aug 2018 15:47:46 +0000 (UTC) Date: Fri, 17 Aug 2018 18:47:40 +0300 From: Greg V Subject: Re: Rockchip RK3399 (ROCKPro64) boots to multiuser To: Emmanuel Vadot Cc: freebsd-arm Message-Id: <1534520860.4036.0@hraggstad.unrelenting.technology> In-Reply-To: <1534366621.3897.2@hraggstad.unrelenting.technology> References: <1533577708.4175.0@hraggstad.unrelenting.technology> <1534253037.1656.0@hraggstad.unrelenting.technology> <20180815105602.b106e1f55a3f839880b1b60e@bidouilliste.com> <1534362095.3897.1@hraggstad.unrelenting.technology> <20180815224449.98b920836c2c7f8610449835@bidouilliste.com> <1534366621.3897.2@hraggstad.unrelenting.technology> X-Mailer: geary/0.12.2 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: quoted-printable X-BeenThere: freebsd-arm@freebsd.org X-Mailman-Version: 2.1.27 Precedence: list List-Id: "Porting FreeBSD to ARM processors." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 17 Aug 2018 15:47:51 -0000 On Wed, Aug 15, 2018 at 11:57 PM, Greg V =20 wrote: >=20 >=20 > On Wed, Aug 15, 2018 at 11:44 PM, Emmanuel Vadot=20 > wrote: >> On Wed, 15 Aug 2018 22:41:34 +0300 >> Greg V wrote: >>> Alright everyone, good news ? I managed to reclock the CPU!!! >>>=20 >>> The patch is now at https://reviews.freebsd.org/D16732 >>=20 >> Thanks a lot !! >> I'll have a deeper look when I'm back from BSDCam. >>=20 >>> (and I think the style is more correct now. Though it's really=20 >>> =7F=7Ffscking >>> silly that the style doesn't like making "table-like" structures=20 >>> =7F=7Flook >>> like tables, i.e. with one-line "rows".) >>>=20 >>> Plus the hack you need to reclock the CPU right now at >>> https://gist.github.com/myfreeweb/88cb9340652f56498f4be770c77b9d61 >>>=20 >>> (the hack allows cpufreq_dt to deal with clock only, no voltage ? >>> since we don't have all the drivers for voltage.) >>=20 >> Are you able to switch to any frequency with that ? >> I would expect the cpu to hang if the voltage is too low or too=20 >> high. >> (I encounter that on RK3328) >=20 > Yeah =E2=80=94 I maxed the clocks for both big and LITTLE cores and got=20 > pretty great performance. >=20 > e.g. unixbench dhrystone index with cpuset to a big core: 804 =E2=80=94=20 > which is more than the 737 I got on Scaleway's ThunderX VPS! > ThunderX is still way better on unixbench's other tests though. > Not that unixbench is a great test=E2=80=A6 >=20 > Compiling neovim also took *way* less time than on RPi/ROCK64. >=20 > So, I think the big cores' voltage regulator (silergy,syr827) might=20 > just default to the highest voltage. > The chip gets rather warm when just idling in FreeBSD=E2=80=A6 Update: tried porting the fanpwr driver from OpenBSD: https://gist.github.com/myfreeweb/584de9b746a328e10c904395afe8a48f Reports 1.0V on boot. For some reason, cpufreq doesn't see the=20 regulator though =E2=80=94 any idea why could that be?? (cpufreq_dt shouldn't require the controller and regulator to be=20 separate nodes, right? There are other drivers like sy8106a where it's=20 all one node=E2=80=A6) Also, overclocked to 2.184GHz, still works great (benchmark score went=20 up again.) I guess either the syr827 is not actually running 1.0 V, or the=20 provided table is waaaay overvolted, or I won the silicon lottery and=20 my chip is just that good. Maybe I should write an efuse driver to look at the leakage=20 measurements=E2=80=A6 =