From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1FPRZk-000360-NU for qemu-devel@nongnu.org; Fri, 31 Mar 2006 16:54:36 -0500 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1FPRZj-00035V-9u for qemu-devel@nongnu.org; Fri, 31 Mar 2006 16:54:36 -0500 Received: from [199.232.76.173] (helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1FPRZj-00035P-69 for qemu-devel@nongnu.org; Fri, 31 Mar 2006 16:54:35 -0500 Received: from [64.233.162.197] (helo=zproxy.gmail.com) by monty-python.gnu.org with esmtp (Exim 4.52) id 1FPRcC-0004yq-KP for qemu-devel@nongnu.org; Fri, 31 Mar 2006 16:57:08 -0500 Received: by zproxy.gmail.com with SMTP id n29so658915nzf for ; Fri, 31 Mar 2006 13:54:33 -0800 (PST) Message-ID: <2ad73a0603311354v63cea0d6g702fba34855bffad@mail.gmail.com> Date: Fri, 31 Mar 2006 18:54:33 -0300 From: "=?ISO-8859-1?Q?Andr=E9_Braga?=" Subject: Re: [Qemu-devel] [PATCH] Improving timer/clock for win32 In-Reply-To: <000401c65486$d2ca4520$0464a8c0@athlon> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline References: <000401c65486$d2ca4520$0464a8c0@athlon> Reply-To: qemu-devel@nongnu.org List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Kazu, have you tested whether this avoids the hang (and 100% CPU time usage by QEMU, and being unable to kill the process by absolutely no means - Process Explorer's 'kill' included - other than rebooting) on WinXP host + kqemu and WinXP guest when opening the guests' clock application by double-clicking it on the taskbar? Thanks, A. On 3/31/06, Kazu wrote: > Hi, > > This patch improves timer/clock for win32. Here is the patch and a binary= . > > http://www.h7.dion.ne.jp/~qemu-win/download/qemu-20060330-timer.patch > http://www.h6.dion.ne.jp/~kazuw/qemu-win/qemu-20060330-timer.zip > > It supports 1ms interval interrupt. > It also improves the delay when it is used with Kqemu/Qvm86. > > A main reason of delay of time is that when main_loop_wait is slept and a= n > interrupt timer is expired, the timer interrupt doesn't occured. I used a= n > event object to wake up the main loop. > > I used QueryPerformanceCounter/QueryPerformanceFrequency to measure time.= It > uses ACPI PM Timer on my system. Some system use RDTSC value. It depends = on > your system. > > It might be possible to run on SMP(HT or Multi-core) system. I haven't > tested it, though. > > Regards, > Kazu