From: gregory.clement@free-electrons.com (Gregory CLEMENT) Date: Wed, 30 Sep 2015 11:28:07 +0200 Subject: [PATCH v4 1/2] ARM: dma-mapping: Don't use outer_flush_range when the L2C is coherent In-Reply-To: <20150929171054.GN21513@n2100.arm.linux.org.uk> (Russell King's message of "Tue, 29 Sep 2015 18:10:54 +0100") References: <1443545458-14807-1-git-send-email-gregory.clement@free-electrons.com> <1443545458-14807-2-git-send-email-gregory.clement@free-electrons.com> <20150929171054.GN21513@n2100.arm.linux.org.uk> Message-ID: <87fv1wqms8.fsf@free-electrons.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org Hi Russell, On mar., sept. 29 2015, Russell King - ARM Linux wrote: > On Tue, Sep 29, 2015 at 06:50:57PM +0200, Gregory CLEMENT wrote: >> When a L2 cache controller is used in a system that provides hardware > > You're talking about L2 cache here, but you're also masking out the L1 > cache maintanence (dmac_*) too. It's my understanding that we don't > yet support coherency to L1 yet. Do you suggest to not masking the dmac_* operation ? Initially I tought that L1 and L2 cache maintanence operation were more or less linked but I might mix up. Thanks, Gregory -- Gregory Clement, Free Electrons Kernel, drivers, real-time and embedded Linux development, consulting, training and support. http://free-electrons.com