From owner-freebsd-arm@freebsd.org Wed Jun 13 08:01:36 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 2349A1010C27 for ; Wed, 13 Jun 2018 08:01:36 +0000 (UTC) (envelope-from manu@bidouilliste.com) Received: from mailman.ysv.freebsd.org (mailman.ysv.freebsd.org [IPv6:2001:1900:2254:206a::50:5]) by mx1.freebsd.org (Postfix) with ESMTP id A08F472DEF for ; Wed, 13 Jun 2018 08:01:35 +0000 (UTC) (envelope-from manu@bidouilliste.com) Received: by mailman.ysv.freebsd.org (Postfix) id 609DE1010C1D; Wed, 13 Jun 2018 08:01:35 +0000 (UTC) Delivered-To: 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 225531010C19 for ; Wed, 13 Jun 2018 08:01:35 +0000 (UTC) (envelope-from manu@bidouilliste.com) Received: from mail.blih.net (mail.blih.net [212.83.177.182]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client CN "mail.blih.net", Issuer "mail.blih.net" (not verified)) by mx1.freebsd.org (Postfix) with ESMTPS id 80C4C72DEA; Wed, 13 Jun 2018 08:01:34 +0000 (UTC) (envelope-from manu@bidouilliste.com) Received: from mail.blih.net (mail.blih.net [212.83.177.182]) by mail.blih.net (OpenSMTPD) with ESMTP id 7479aa05; Wed, 13 Jun 2018 10:01:33 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha1; c=relaxed; d=bidouilliste.com; h=date :from:to:cc:subject:message-id:in-reply-to:references :mime-version:content-type:content-transfer-encoding; s=mail; bh=gk5EP7JLkDRH3asq2At+JRmJhUk=; b=VkrlLUstzPczqZJ2e7NBeyUwbQsw FO8NIXks8qC5C22sJGU4outIzsBipQJO78FppYl8otwT7Es6i41xna4ruFVVvJWE qN4vBJoQhpiyVMJNknDDCs4CWvS/qxT7Dud1OpjIG6u2weXzhMgY0gj6ZzqsDPV2 1fT9h9QugzZ8zVA= DomainKey-Signature: a=rsa-sha1; c=nofws; d=bidouilliste.com; h=date :from:to:cc:subject:message-id:in-reply-to:references :mime-version:content-type:content-transfer-encoding; q=dns; s= mail; b=i33x+1Qf7hEOrHjD2kBo5H5IlcW2R98q4dEW++v4vJtY+1mneJeQg3Gq vro/wKWXHcWf28qFf4eCOznwEeTtHrPtquGAOIrkqeyBC9zS7rlZVA2QOc5IkCII cJdCHClxXec0/aJc6dwBCzk4xkdG+uT328lOn0scvRydKqtPPsc= Received: from skull.home.blih.net (ip-9.net-89-3-105.rev.numericable.fr [89.3.105.9]) by mail.blih.net (OpenSMTPD) with ESMTPSA id a37d8573 TLS version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NO; Wed, 13 Jun 2018 10:01:32 +0200 (CEST) Date: Wed, 13 Jun 2018 10:01:32 +0200 From: Emmanuel Vadot To: Ganbold Tsagaankhuu Cc: Kyle Evans , Rogiel Sulzbach , "freebsd-arm@freebsd.org" Subject: Re: De-orbit Allwinner A10/A20/A31 for 12.0 Message-Id: <20180613100132.af9a6fa4e4c9d627ffd28a5a@bidouilliste.com> In-Reply-To: References: <20180612223248.f95d9ce3961187576e220614@bidouilliste.com> <1270d2d0-7f10-4454-af30-d5d62ec5bd73@Spark> X-Mailer: Sylpheed 3.7.0 (GTK+ 2.24.32; amd64-portbld-freebsd12.0) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-BeenThere: freebsd-arm@freebsd.org X-Mailman-Version: 2.1.26 Precedence: list List-Id: "Porting FreeBSD to ARM processors." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 13 Jun 2018 08:01:36 -0000 On Wed, 13 Jun 2018 11:32:08 +0800 Ganbold Tsagaankhuu wrote: > Hi, > > On Wed, Jun 13, 2018 at 7:47 AM, Kyle Evans wrote: > > > On Tue, Jun 12, 2018 at 3:51 PM, Rogiel Sulzbach > > wrote: > > > On Jun 12, 2018, 5:33 PM -0300, Emmanuel Vadot , > > wrote: > > >> > > >> Hello arm@ > > >> > > >> I don't like to send this email but ... > > >> > > >> I want to remove A10/A20/A31 support for FreeBSD 12, here is the > > >> reason why : > > >> > > >> - I don't have time to maintain thoses SoC > > >> - For A10/A20 we currently don't boot on it since it requires new > > >> clocks binding (handled by aw_ccung) and I have no time to do the > > >> switch. > > >> - H3 (which is 32bits too) is way better in term of performance and > > >> price, you can buy a "good" $10 board while getting an A20 will cost > > >> you more for some reason. > > >> - My A31 board (BananapiM1) just died today after a long time of being > > >> my arm32 reference board. > > >> - I don't want the code to stay like the old 32 bits rockchip or > > >> amlogic code that can't even boot nowadays. > > >> > > >> If someone want to step up and convert A10/A20 to ccung I'll be > > >> glad to help/review but if no one step up it's unreasonable for me to > > >> support those right now and I prefer to remove completely the support. > > >> > > >> The plan is to remove support in one month (July 12th) if there is no > > >> volunteer to update the current code. > > >> > > >> Cheers, > > >> > > >> -- > > >> Emmanuel Vadot > > > > > > Hello Emmanuel, > > > > > > we have recently considered to switching from i.MX6 to a A20 for cost > > reasons. > > > > > > I have just bought a Banana Pi with an A20 to test performance and > > compare it to the Wandboard. The problem with H3 (and all other better > > socs) is the lack of a proper SATA controller. A real controller is only > > available in A10/A20 SoCs, all others go through a horrible SATA-to-USB > > adapter that is horribly slow. > > > > > > Anyway, in case we decide to switch, I can keep the A20 running > > -CURRENT. I might need some help getting started since I do not have a lot > > of experience in the kernel, other than the AHCI driver for the imx. > > > > Hi Rogiel, > > > > FWIW- I'm going to go ahead and write the ccung bits for A10/A20 so > > that I can make use of my Lamobo R1 when the switch driver for that is > > finished. > > > > This is probably not a bad time to pare down the DTS+U-Boot we build > > for these SoC to what people are actually willing to claim they use > > with FreeBSD, though. > > > > > First of all I would like to thank manu@ and kevans@ and other committers > who improved the support of Allwinner SoC related codes and maintaining > them up to now. > I basically agree with manu@, but there could be few points we may have to > consider: > - It is nice to have the support of a board which has SATA controller, as > mentioned Rogiel Yes that's for me the only interesting thing on those SoC nowadays. > - According to https://wiki.freebsd.org/FreeBSD/arm/Allwinner most of > drivers work on A20 boards (if we add ccung support) and it could be a > reference code for attracting new developers, users etc. Correct the only missing thing is ccung (and test/maintainance of course) > - If we nuke A10/A20 bits, then we maybe need to obtain some other boards > with H3 SoC for gjb@ or anyone who builds official images, I believe those > images are tested on those boards. I plan to deorbit some images and make a generic image, now sure how for now that's all :) > - I know it is time consuming, but I think a developer has to make sure > his/her code changes work across boards/SoC family, or at least ask others > to test or apply similar changes to their boards/SoCs just to make sure > everything works. Correct me if I'm wrong here. Sure, that's why I always test on 32bit allwinner (A31/H3) and 64bit (A64/H5) > - I know I should be blamed for not updating Rockchip RK3188 and Amlogic > codes, but I hope some of the bits can be used/referenced later on for > boards with SoC like RK3328, RK3399 etc, since I do have the latter one. Nothing to be blamed for, we all have a life :) I had a quick glance at the rockchip code before starting working on RK3328 and I think that now everything supported before is supported for 64bits SoC (with a few extras) > Kyle, > > Adding ccung support to A10/A20 is very much appreciated. > I will wait for your switch driver, that would be really awesome. > I think I have same board too :) > > thanks a lot for understanding, > > Ganbold > > > > > > > Thanks, > > > > Kyle Evans > > _______________________________________________ > > freebsd-arm@freebsd.org mailing list > > https://lists.freebsd.org/mailman/listinfo/freebsd-arm > > To unsubscribe, send any mail to "freebsd-arm-unsubscribe@freebsd.org" > > -- Emmanuel Vadot