From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:44787) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Yp1Sc-00049x-Ur for qemu-devel@nongnu.org; Sun, 03 May 2015 17:30:03 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1Yp1SZ-0000dt-OR for qemu-devel@nongnu.org; Sun, 03 May 2015 17:30:02 -0400 Received: from cantor2.suse.de ([195.135.220.15]:43264 helo=mx2.suse.de) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Yp1SZ-0000do-Hs for qemu-devel@nongnu.org; Sun, 03 May 2015 17:29:59 -0400 Message-ID: <55469352.9000707@suse.de> Date: Sun, 03 May 2015 23:29:54 +0200 From: Alexander Graf MIME-Version: 1.0 References: <1426705700-2564-1-git-send-email-drjones@redhat.com> <20150318191819.GB7077@hawk.usersys.redhat.com> In-Reply-To: <20150318191819.GB7077@hawk.usersys.redhat.com> Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] the arm cache coherency cluster "v2" List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Andrew Jones , kvmarm@lists.cs.columbia.edu, qemu-devel@nongnu.org, ard.biesheuvel@linaro.org, christoffer.dall@linaro.org, marc.zyngier@arm.com, peter.maydell@linaro.org, pbonzini@redhat.com Cc: catalin.marinas@arm.com, lersek@redhat.com, m.smarduch@samsung.com On 18.03.15 20:18, Andrew Jones wrote: > On Wed, Mar 18, 2015 at 03:08:20PM -0400, Andrew Jones wrote: >> In reply to this message I'll send two series' one for KVM and >> one for QEMU. The two series' are their respective component >> complements, and attempt to implement cache coherency for arm >> guests using emulated devices, where the emulator (qemu) uses >> cached memory for the device memory, but the guest uses >> uncached - as device memory is generally used. Right now I've >> just focused on VGA vram. >> >> This approach is the "MADV_UNCACHED" type that Paolo suggested. >> This type of approach could also be described as "make userspace's >> memory access type match the expected access type of the guest", >> and Mario has suggested using a memory driver, which could have >> the same result. >> >> The coming series' is inspired by both Paolo's and Mario's >> suggestions, but it uses a kvm memslot flag, rather than an >> madvise flag, and thus for the memory driver, it's just KVM. >> >> See the thread >> >> https://lists.gnu.org/archive/html/qemu-devel/2015-03/msg01254.html >> >> for some more background. >> >> Thanks in advance for comments. >> >> drew > > I forgot to mention that I've done some light testing with this. > It seems to work, and without (to eye) noticeable performance > degradation. Just for the record, I couldn't get it to work :). But I'm looking forward to the next version with MMU notifiers! Alex