From owner-freebsd-arm@FreeBSD.ORG Fri Sep 26 20:40:41 2014 Return-Path: Delivered-To: freebsd-arm@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [8.8.178.115]) (using TLSv1 with cipher ADH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTPS id B7DCB9C0 for ; Fri, 26 Sep 2014 20:40:41 +0000 (UTC) Received: from mailgate-02.zdv.uni-mainz.de (mailgate-02.zdv.Uni-Mainz.DE [IPv6:2001:4c80:40:62d:203:ffff:fe5d:b2f6]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (Client CN "IronPort Appliance Demo Certificate", Issuer "IronPort Appliance Demo Certificate" (not verified)) by mx1.freebsd.org (Postfix) with ESMTPS id B199D2E6 for ; Fri, 26 Sep 2014 20:40:39 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=uni-mainz.de; i=@uni-mainz.de; q=dns/txt; s=ironport; t=1411764040; x=1443300040; h=from:to:subject:date:message-id:references:in-reply-to: content-transfer-encoding:mime-version; bh=CtevmpFjceZsHLwpT++5Zav2WxCz2gDwNmb/zurNBcQ=; b=gjbebiE6cQEAiIGaXd0l/Bz3MdUNC6I52NAvKtrqf8rOSv5mudKdsKfA 53677bDFCPn0tosNni1STeeQOPEa2oVJRBLhbkBT0GFvRDE/epbYvHO/r WIUa30rZWFHXgHP5Ejoo0S+z/2P2EORnOOvPIYKNai/rD84dthBDtZHWs A=; X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AqsEALLOJVQKXgZY/2dsb2JhbABfDoJddlcEyjwKhnpUAoEjAXuEAwEBAQQBAQFLIBcEAgEIDgMEAQEBJwcnAQkBFAkIAgQBBwIFBAEHEgECBIgdAQy/HQEXiiqFIwEBVgaCMVOBQQWLE4sKhAaEZYVqjhCDI0BsB4EIOYECAQEB X-IPAS-Result: AqsEALLOJVQKXgZY/2dsb2JhbABfDoJddlcEyjwKhnpUAoEjAXuEAwEBAQQBAQFLIBcEAgEIDgMEAQEBJwcnAQkBFAkIAgQBBwIFBAEHEgECBIgdAQy/HQEXiiqFIwEBVgaCMVOBQQWLE4sKhAaEZYVqjhCDI0BsB4EIOYECAQEB Received: from e14hub-02.zdv.uni-mainz.de ([10.94.6.88]) by mailgate-02.zdv.uni-mainz.de with ESMTP/TLS/AES128-SHA; 26 Sep 2014 22:40:36 +0200 Received: from e15be-03.zdv.Uni-Mainz.DE (2001:4c80:40:606:92e2:baff:fe19:9239) by E14HUB-02.zdv.Uni-Mainz.DE (2001:4c80:40:606:21d:d8ff:feb7:1c60) with Microsoft SMTP Server (TLS) id 14.3.210.2; Fri, 26 Sep 2014 22:40:36 +0200 Received: from e15be-02.zdv.Uni-Mainz.DE (2001:4c80:40:606:92e2:baff:fe19:8fb0) by e15be-03.zdv.Uni-Mainz.DE (2001:4c80:40:606:92e2:baff:fe19:9239) with Microsoft SMTP Server (TLS) id 15.0.995.29; Fri, 26 Sep 2014 22:39:53 +0200 Received: from e15be-02.zdv.Uni-Mainz.DE ([fe80::92e2:baff:fe19:8fb0]) by e15be-02.zdv.Uni-Mainz.DE ([fe80::92e2:baff:fe19:8fb0%15]) with mapi id 15.00.0995.028; Fri, 26 Sep 2014 22:39:53 +0200 From: =?iso-8859-1?Q?Wei=DF=2C__Dr=2E_J=FCrgen?= To: 'David Rayson' , "freebsd-arm@freebsd.org" Subject: RE: Jetson TK1 board support Thread-Topic: Jetson TK1 board support Thread-Index: AQHP18iUBDzcFmGHMkWO9pZPd3gkEJwT3s2w Date: Fri, 26 Sep 2014 20:39:52 +0000 Message-ID: <2c451765bffb43e8b9dab56927bb351a@e15be-02.zdv.Uni-Mainz.DE> References: <542271AE.6070807@andrew.cmu.edu> In-Reply-To: <542271AE.6070807@andrew.cmu.edu> Accept-Language: de-DE, en-US Content-Language: de-DE X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [134.93.178.81] Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-BeenThere: freebsd-arm@freebsd.org X-Mailman-Version: 2.1.18-1 Precedence: list List-Id: "Porting FreeBSD to ARM processors." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 26 Sep 2014 20:40:41 -0000 Hi, sorry, I did not have any time during the week. I just sent a mail to the list with a link to my changes. Only serial, USB2 and PCIe/Ethernet hardware is working - so no SATA. The drivers rely on u-boot to initialize the hardware. While this is ok for pinmux, other initializations should be done by the drivers. The interrupt handling for PCIe is rather ad hoc. The interrupt routing should honor the FDT description. The Tegra platform has a GIC with extensions for interrupt routing. I just made a copy of the GIC code end extended it in a few cases. There should probably be a mechanism to do this without duplicating code. I changed some non tegra files to get FreeBSD running on the hardware. There should be better solutions, which can be merged back to the FreeBSD source tree. For example the problem with cache coherency due to aggressive L2 prefetch awaits a real solution. There is no code to change the cpu clock yet. There is no support for SDHCI or EMMC. So I would consider this a first step, which allows to do native development on the platform.=20 Besides that, the kernel seems to be quite stable - at least with the compiles I did. Regards Juergen Juergen Weiss |Universitaet Mainz, Zentrum fuer Datenverarbeitung, weiss@uni-mainz.de |55099 Mainz, Tel: +49(6131)39-26361, FAX: +49(6131)39-2= 6407 > -----Original Message----- > From: owner-freebsd-arm@freebsd.org [mailto:owner-freebsd-arm@freebsd.org= ] On Behalf Of > David Rayson > Sent: Wednesday, September 24, 2014 9:25 AM > To: freebsd-arm@freebsd.org > Subject: Re: Jetson TK1 board support >=20 > Hi, >=20 > What other work would be useful to get this port working well? I might > be interested in working on improving it, but first I want to make sure > I have a clear sense of what's been done so far (and how stable/not it > is) and what still remains to be done. >=20 > --David >=20 > > Hi, > > > > I have a rather rough port of FreeBSD current on arm to Jetson TK1. I > > used Stephen Warren's tegra u-boot sources, which initialize and config= ure > > USB and PCIe. > > > > So SMP, USB and the onboard PCIe Ethernet adapter work. > > > > After Ian's changes to busdma_machdep-v6 (r269212) I had problems with > > cache coherency with the Ethernet adapter. Seems this is due to the agg= ressive > > L2 prefetcher of Cortex A15. Disabling L2 prefetch does help, as well a= s > > invalidating the cache a second time after the dma transfer. I'm not > > sure what the correct solution to this problem is. I wonder how > > other Cortex A15 platforms (exynos5) handle this. > > > > I will probably be able to do some cleanups and put patches on the web > > within a week. > > > > Regards > > > > Juergen > > > > Juergen Weiss |Universitaet Mainz, Zentrum fuer Datenverarbeitung, > > weiss at uni-mainz.de |55099 > Mainz, Tel: +49(6131)39-26361, FAX: +49(6131)39-26407 > > > > >/ -----Original Message----- > > />/ From:owner-freebsd-arm at freebsd.org > [mailto:owner-fr= eebsd-arm at > freebsd.org ] On = Behalf Of > > />/ Ian Lepore > > />/ Sent: Sunday, September 21, 2014 3:44 PM > > />/ To: Lundberg, Johannes > > />/ Cc:freebsd-arm at freebsd.org arm> > > />/ Subject: Re: Jetson TK1 board support > > />/ > > />/ On Sun, 2014-09-21 at 16:45 +0900, Lundberg, Johannes wrote: > > />/ > Great! > > />/ > > > />/ > What I've done so far is > > />/ > > > />/ > - build and patch (enable API) u-boot-nvidia on freebsd (i think= i got it > > />/ > fromgit://nv-tegra.nvidia.com/3rdparty/u-boot.git, the normal u= -boot > > />/ > wouldn't work...) > > />/ > - flash u-boot-dtb-tegra.img onto the board's mmc using nvidia's= flash tool > > />/ > on ubuntu > > />/ > - build an image using crochet and dd to sd card (so far I copie= d the > > />/ > beaglebone setup, just to get a ubldr and a kernel file) > > />/ > > > />/ > > > />/ > From u-boot I can see all devices. I load ubldr with > > />/ > fatload mmc 1:1 0x80200000 ubldr > > />/ > bootelf 0x80200000 > > />/ > > > />/ > ubldr load fine but, from ubldr I can only see the mmc 0 and net= devices. > > />/ > There's no sd card (mmc 1), and no ufs partition.. > > />/ > > > />/ > > > />/ > > > />/ > > > />/ > -- > > />/ > Johannes Lundberg > > />/ > BRILLIANTSERVICE CO., LTD. > > />/ > > > />/ > On Fri, Sep 19, 2014 at 8:25 PM, John Howie > wrote: > > />/ > > > />/ > > Hi all, > > />/ > > > > />/ > > I am up for testing and supporting this board. I ordered and r= eceived > > />/ > > mine, but have not really had a chance to use it due to work t= o-date. The > > />/ > > good news is the next few months I will have bandwidth. > > />/ > > > > />/ > > Regards, > > />/ > > > > />/ > > John > > />/ > > > > />/ > > > > />/ > > On 9/19/14, 12:15 PM, "Lundberg, Johannes" > > />/ > > > wrote: > > />/ > > > > />/ > > >Hi > > />/ > > > > > />/ > > >I started working on adding the Jetson TK1 board to Crochet. = Is there any > > />/ > > >work in progress on this? > > />/ > > >I guess there is quite a lot of work that has to been done to= get full > > />/ > > >support for it in the kernel as well.. > > />/ > > > > > />/ > > >Best regards > > />/ > > >-- > > />/ > > >Johannes Lundberg > > />/ > > > > > />/ > > />/ You may have to change some u-boot options to support multiple mmc= /sd > > />/ interfaces. Look in the config header for CONFIG_SYS_MMC_MAX_DEVI= CE; if > > />/ it's not there you may need to add it. For wandboard I also had t= o add > > />/ a freescale-specific one, CONFIG_SYS_FSL_USDHC_NUM, so there may b= e > > />/ something like that you need to find as well. > > />/ > > />/ -- Ian > > />/ > > />/ > > />/ _______________________________________________ > > />/ freebsd-arm at freebsd.org > mailing list > > />/ http://lists.freebsd.org/mailman/listinfo/freebsd-arm > > />/ To unsubscribe, send any mail to "freebsd-arm-unsubscribe at freeb= sd.org > "/ > _______________________________________________ > freebsd-arm@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-arm > To unsubscribe, send any mail to "freebsd-arm-unsubscribe@freebsd.org"