From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:43575) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1aimOA-0002a6-Ut for qemu-devel@nongnu.org; Wed, 23 Mar 2016 13:16:14 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1aimO7-0005jC-Jt for qemu-devel@nongnu.org; Wed, 23 Mar 2016 13:16:10 -0400 Received: from relay1.mentorg.com ([192.94.38.131]:44622) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1aimO7-0005j6-Dj for qemu-devel@nongnu.org; Wed, 23 Mar 2016 13:16:07 -0400 Message-ID: <1458753364.2947.22.camel@mentor.com> From: Hollis Blanchard Date: Wed, 23 Mar 2016 10:16:04 -0700 In-Reply-To: <56F2CA0F.9030307@redhat.com> References: <1455744555-22101-1-git-send-email-hollis_blanchard@mentor.com> <1455744555-22101-2-git-send-email-hollis_blanchard@mentor.com> <1458751625.2947.19.camel@mentor.com> <56F2CA0F.9030307@redhat.com> Content-Type: text/plain; charset="UTF-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Subject: Re: [Qemu-devel] [PATCH 2/2] trace: separate MMIO tracepoints from TB-access tracepoints List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Paolo Bonzini Cc: qemu-devel@nongnu.org, stefanha@redhat.com On Wed, 2016-03-23 at 17:53 +0100, Paolo Bonzini wrote: > > On 23/03/2016 17:47, Hollis Blanchard wrote: > > > > Paolo, is it true that only TB-invalidating writes go through the > > io_mem_notdirty path? I'm looking at the live migration code now, > > and it > > seems like every memory write will go through that path when global > > dirty memory logging is enabled. > When live migration is enabled, writes to clean memory (almost all of > them) will go through that path indeed.  Some writes to the > framebuffer > will go through that path too. > > It depends on > >       cpu_physical_memory_is_clean( >                         memory_region_get_ram_addr(section->mr) + > xlat)) > > in tlb_set_page_with_attrs. Would "memory_region_notdirty_read/write" be a better tracepoint name than "memory_region_tb_read/write"? --  Hollis Blanchard Mentor Graphics Emulation Division