From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([208.118.235.92]:45851) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1UDXjm-0004fe-UQ for qemu-devel@nongnu.org; Thu, 07 Mar 2013 05:07:49 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1UDXjk-0006yx-G3 for qemu-devel@nongnu.org; Thu, 07 Mar 2013 05:07:46 -0500 Received: from mx1.redhat.com ([209.132.183.28]:18008) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1UDXjk-0006yr-8O for qemu-devel@nongnu.org; Thu, 07 Mar 2013 05:07:44 -0500 Received: from int-mx01.intmail.prod.int.phx2.redhat.com (int-mx01.intmail.prod.int.phx2.redhat.com [10.5.11.11]) by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id r27A7hKo003937 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK) for ; Thu, 7 Mar 2013 05:07:43 -0500 Date: Thu, 7 Mar 2013 11:07:42 +0100 From: Stefan Hajnoczi Message-ID: <20130307100742.GB14726@stefanha-thinkpad.redhat.com> References: <1362584735-30911-1-git-send-email-stefanha@redhat.com> <1362584735-30911-6-git-send-email-stefanha@redhat.com> <5137703D.8080401@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <5137703D.8080401@redhat.com> Subject: Re: [Qemu-devel] [PATCH 5/5] threadpool: drop global thread pool List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Paolo Bonzini Cc: Kevin Wolf , qemu-devel@nongnu.org On Wed, Mar 06, 2013 at 05:35:09PM +0100, Paolo Bonzini wrote: > Il 06/03/2013 16:45, Stefan Hajnoczi ha scritto: > > Now that each AioContext has a ThreadPool and the main loop AioContext > > can be fetched with qemu_get_aio_context(), we can eliminate the concept > > of a global thread pool from thread-pool.c. > > > > The submit functions must take a ThreadPool* argument. > > This is certainly ok for thread-pool.c. For raw-posix and raw-win32, > what about adding already a bdrv_get_aio_context() function and using > that in paio_submit? Is it putting the cart before the horse? Thanks, we might as well do that. Then I won't have to go back and modify block/raw-posix.c and block/raw-win32.c in the next patch series which lets BlockDriverState bind to an AioContext. Stefan