From mboxrd@z Thu Jan 1 00:00:00 1970 From: Michael Ellerman Subject: Re: [RFC PATCH] drm/ttm: force cached mappings for system RAM on ARM Date: Tue, 15 Jan 2019 22:31:29 +1100 Message-ID: <87a7k2yx66.fsf@concordia.ellerman.id.au> References: <20190110072841.3283-1-ard.biesheuvel@linaro.org> <5d8135de-80fe-9c0e-2206-ecb809f64cdb@daenzer.net> <55facfb9-92af-86b8-40e9-d63b887b5592@amd.com> <9f956898-7973-98ee-6bf1-e1d445e9d365@amd.com> <20190114191350.GA29600@fuggles.cambridge.arm.com> <20190114193548.GB29600@fuggles.cambridge.arm.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20190114193548.GB29600@fuggles.cambridge.arm.com> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: "linux-arm-kernel" Errors-To: linux-arm-kernel-bounces+linux-arm-kernel=m.gmane.org@lists.infradead.org To: Will Deacon , "Koenig, Christian" Cc: Carsten Haitzler , Ard Biesheuvel , David Airlie , benh@kernel.crashing.org, Michel =?utf-8?Q?D=C3=A4nz?= =?utf-8?Q?er?= , Linux Kernel Mailing List , dri-devel , "Huang, Ray" , "Zhang, Jerry" , linux-arm-kernel , Bernhard =?utf-8?Q?Rosenkr=C3=A4n?= =?utf-8?Q?zer?= List-Id: dri-devel@lists.freedesktop.org Hi Will, Will Deacon writes: > [+ BenH and MPE] > > On Mon, Jan 14, 2019 at 07:21:08PM +0000, Koenig, Christian wrote: >> Am 14.01.19 um 20:13 schrieb Will Deacon: ... > >> > The Arm architecture (and others including Power afaiu) doesn't >> > guarantee coherency when memory is accessed using mismatched cacheability >> > attributes. ... > >> As far as I know Power doesn't really supports un-cached memory at all, >> except for a very very old and odd configuration with AGP. > > Hopefully Michael/Ben can elaborate here, but I was under the (possibly > mistaken) impression that mismatched attributes could cause a machine-check > on Power. That's what I've always been told, but I can't actually find where it's documented, I'll keep searching. But you're right that mixing cached / uncached is not really supported, and probably results in a machine check or worse. cheers