From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:33700) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1X2vls-0002Au-5w for qemu-devel@nongnu.org; Fri, 04 Jul 2014 01:10:59 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1X2vlh-0003U0-Gr for qemu-devel@nongnu.org; Fri, 04 Jul 2014 01:10:52 -0400 Received: from youngberry.canonical.com ([91.189.89.112]:59897) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1X2vlh-0003Tq-7e for qemu-devel@nongnu.org; Fri, 04 Jul 2014 01:10:41 -0400 Received: from mail-vc0-f181.google.com ([209.85.220.181]) by youngberry.canonical.com with esmtpsa (TLS1.0:RSA_ARCFOUR_SHA1:16) (Exim 4.71) (envelope-from ) id 1X2vlg-0000PB-FB for qemu-devel@nongnu.org; Fri, 04 Jul 2014 05:10:40 +0000 Received: by mail-vc0-f181.google.com with SMTP id il7so1151647vcb.12 for ; Thu, 03 Jul 2014 22:10:39 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: <1404395521-11158-1-git-send-email-stefanha@redhat.com> References: <1404395521-11158-1-git-send-email-stefanha@redhat.com> Date: Fri, 4 Jul 2014 13:10:39 +0800 Message-ID: From: Ming Lei Content-Type: text/plain; charset=UTF-8 Subject: Re: [Qemu-devel] [PATCH 0/2] coroutine: dynamically scale pool size List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Stefan Hajnoczi Cc: Kevin Wolf , Paolo Bonzini , qemu-devel On Thu, Jul 3, 2014 at 9:51 PM, Stefan Hajnoczi wrote: > The coroutine pool reuses exited coroutines to make qemu_coroutine_create() > cheap. The size of the pool is capped to prevent it from hogging memory after > a period of high coroutine activity. Previously the max size was hardcoded to > 64 but this doesn't scale with guest size. > > A guest with lots of disks can do more parallel I/O and therefore requires a > larger coroutine pool size. This series tries to solve the problem by scaling > pool size according to the number of drives. > > Ming: Please let me know if this eliminates the rt_sigprocmask system calls you > are seeing. It should solve part of the performance regression you have seen > in qemu.git/master virtio-blk dataplane. With both the two coroutine patches and the block plug&unplug patches, performance of qemu.git/master virtio-blk dataplane can recover to level of QEMU 2.0. > > Stefan Hajnoczi (2): > coroutine: make pool size dynamic > block: bump coroutine pool size for drives > > block.c | 4 ++++ > include/block/coroutine.h | 11 +++++++++++ > qemu-coroutine.c | 25 +++++++++++++++++++------ > 3 files changed, 34 insertions(+), 6 deletions(-) > > -- > 1.9.3 > >