From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from [140.186.70.92] (port=33005 helo=eggs.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1Powho-0005HM-Bl for qemu-devel@nongnu.org; Mon, 14 Feb 2011 06:35:01 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1Powhn-0005M2-4L for qemu-devel@nongnu.org; Mon, 14 Feb 2011 06:35:00 -0500 Received: from cantor2.suse.de ([195.135.220.15]:46451 helo=mx2.suse.de) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Powhm-0005Lt-QQ for qemu-devel@nongnu.org; Mon, 14 Feb 2011 06:34:59 -0500 Message-ID: <4D591361.5030407@suse.de> Date: Mon, 14 Feb 2011 12:34:57 +0100 From: Alexander Graf MIME-Version: 1.0 References: <1297121191-31245-1-git-send-email-agraf@suse.de> <4D591085.9020906@redhat.com> In-Reply-To: <4D591085.9020906@redhat.com> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Subject: [Qemu-devel] Re: [RFC] prep: enable irq sharing on ide again List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Gerd Hoffmann Cc: Kevin Wolf , =?UTF-8?B?QW5kcmVhcyBGw6RyYmU=?= =?UTF-8?B?cg==?= , QEMU Developers , Aurelien Jarno Gerd Hoffmann wrote: > On 02/08/11 00:26, Alexander Graf wrote: >> The new ISA infrastructure checks for potential irq sharing bugs on >> interrupt lines, because usually irq lines on isa can't be shared. >> >> The PREP spec however mandates that the irq lines for both IDE ports >> are shared and according to Aurelien this also used to work just fine. >> >> So let's add a way to enable this sharing again, so we don't introduce >> unnecessary regressions over older versions of Qemu. > > Had a patch for that, got shoot down for reasons I don't remember, > attached for reference. It basically allows IRQ sharing in case the > two devices sharing the IRQ are of the same kind. In that case you > usually have a single guest driver handling both devices and IRQ > sharing works most of the time. > > I don't mind much which approach we take ... I'm very indifferent myself :). This is not an issue we should waste too much time/effort on. Alex