From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([140.186.70.92]:45493) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1QusVs-0006ag-Sd for qemu-devel@nongnu.org; Sat, 20 Aug 2011 16:51:29 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1QusVr-0000Ri-II for qemu-devel@nongnu.org; Sat, 20 Aug 2011 16:51:28 -0400 Received: from mail-fx0-f45.google.com ([209.85.161.45]:62981) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1QusVr-0000RS-8r for qemu-devel@nongnu.org; Sat, 20 Aug 2011 16:51:27 -0400 Received: by fxbb27 with SMTP id b27so2927552fxb.4 for ; Sat, 20 Aug 2011 13:51:26 -0700 (PDT) Date: Sat, 20 Aug 2011 22:51:22 +0200 From: "Edgar E. Iglesias" Message-ID: <20110820205122.GA29545@zapo> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Subject: [Qemu-devel] TCG slowdown List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Cc: jan.kiszka@siemens.com Lately, my TCG emulated machines have been running terribly slow. Like a factor 3 or 4 slower, hardly usable. I bisected it to this: commit d5ab9713d2d4037fd56b0adddd26c8d4dc11cf09 Author: Jan Kiszka Date: Tue Aug 2 16:10:21 2011 +0200 Avoid allocating TCG resources in non-TCG mode Do not allocate TCG-only resources like the translation buffer when running over KVM or XEN. Saves a "few" bytes in the qemu address space and is also conceptually cleaner. Signed-off-by: Jan Kiszka Signed-off-by: Anthony Liguori It turns out that the init of the tb cache is now moved to before ram_size gets it's default value. And because the default tb cache size, if no -tb-size option is given, is based on the ram_size it is now zero.. Cheers