From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tom Rini Date: Thu, 14 Jul 2016 17:33:50 -0400 Subject: [U-Boot] [PATCH] arm: start.S: Fix startup of dragonboard410c In-Reply-To: <1cab9e50-da06-8b5d-419a-6881c94d3116@arm.com> References: <1468354509-27476-1-git-send-email-mateusz.kulikowski@gmail.com> <1cab9e50-da06-8b5d-419a-6881c94d3116@arm.com> Message-ID: <20160714213350.GH7260@bill-the-cat> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de On Tue, Jul 12, 2016 at 11:10:28PM +0100, Andr? Przywara wrote: > On 12/07/16 21:15, Mateusz Kulikowski wrote: > > Hi Mateusz, > > > Commit d73718f3 breaks devices where: > > - There is EL2/EL3 firmware and > > - U-Boot starts in NS EL1 and > > - EL2/EL3 firmware didn't unlocked write-access to CPUECTLR_EL1. > > > > This patch makes that change opt-out configuration option, > > and disables it for dragonboard410c. > > thanks for bringing this up! > You are totally right, as this register isn't architecturally defined, > we should never set it without checking for a particular core. > Actually even that is not enough, as it's also access restricted - even > defaulting to RO - as you mentioned. > > So I am very much for a board specific _opt-in_ solution, also as ARM > Trusted Firmware takes care of the bit already, for instance. > > Mingkai Hu, can you say for which board/SoC/family the original patch > was needed for? Layerscape? OK, I'm going to revert the original patch here (and would recommend that distros that are picking up v2016.07 do the same). Thanks all! -- Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: Digital signature URL: