From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1O92ro-0006Rx-MT for qemu-devel@nongnu.org; Mon, 03 May 2010 17:07:52 -0400 Received: from [140.186.70.92] (port=59797 helo=eggs.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1O92rm-0006Rb-Dv for qemu-devel@nongnu.org; Mon, 03 May 2010 17:07:51 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.69) (envelope-from ) id 1O92rk-00043C-RN for qemu-devel@nongnu.org; Mon, 03 May 2010 17:07:50 -0400 Received: from smtp-out.google.com ([74.125.121.35]:50789) by eggs.gnu.org with esmtp (Exim 4.69) (envelope-from ) id 1O92rk-00042j-Jl for qemu-devel@nongnu.org; Mon, 03 May 2010 17:07:48 -0400 Received: from kpbe20.cbf.corp.google.com (kpbe20.cbf.corp.google.com [172.25.105.84]) by smtp-out.google.com with ESMTP id o43L7iD9017298 for ; Mon, 3 May 2010 14:07:45 -0700 Received: from bwz28 (bwz28.prod.google.com [10.188.26.28]) by kpbe20.cbf.corp.google.com with ESMTP id o43L7g8p014791 for ; Mon, 3 May 2010 14:07:43 -0700 Received: by bwz28 with SMTP id 28so1605585bwz.14 for ; Mon, 03 May 2010 14:07:42 -0700 (PDT) MIME-Version: 1.0 Date: Mon, 3 May 2010 14:07:42 -0700 Message-ID: From: Frank Berreth Content-Type: multipart/alternative; boundary=0016e6da9c81d19d460485b6fb12 Subject: [Qemu-devel] How to map PCI memory into the VM without trapping List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org --0016e6da9c81d19d460485b6fb12 Content-Type: text/plain; charset=ISO-8859-1 Hello, I am using a modified patch for the ivshmem device that was posted here recently and discovered some performance issues with it. Essentially I measure write performance into the PCI memory that is mapped into the VM as really slow (about 60mb/sec). I assume that is because there is trapping on each memory access (there are no handlers installed for handling read/write access to that memory though). The device shows up in the guest as a UIO device but I also tried accessing other PCI memory like the vga memory (16mb from vga-pci.c) and it has similar performance (through /sys/device/.../resource...). My question is how I can map that memory into the VM so it doesn't get trapped on in the normal use case. The mappings (registering PCI bar etc) is done as in the patch submitted recently). Thanks, Frank. --0016e6da9c81d19d460485b6fb12 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Hello,

I am using a modified patch for the ivshmem devic= e that was posted here recently and discovered some performance issues with= it. Essentially I measure write performance into the PCI memory that is ma= pped into the VM as really slow (about 60mb/sec). I assume that is because = there is trapping on each memory access (there are no handlers installed fo= r handling read/write access to that memory though). The device shows up in= the guest as a UIO device but I also tried accessing other PCI memory like= the vga memory (16mb from vga-pci.c) and it has similar performance (throu= gh /sys/device/.../resource...).

My question is how I can map that memory into the VM so= it doesn't get trapped on in the normal use case. The mappings (regist= ering PCI bar etc) is done as in the patch submitted recently).

Thanks,
Frank.

--0016e6da9c81d19d460485b6fb12--