From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:37417) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1V2fBv-0007af-Pl for qemu-devel@nongnu.org; Fri, 26 Jul 2013 06:24:09 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1V2fBr-0005p2-R6 for qemu-devel@nongnu.org; Fri, 26 Jul 2013 06:24:07 -0400 Received: from mx1.redhat.com ([209.132.183.28]:10255) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1V2fBr-0005oy-Hb for qemu-devel@nongnu.org; Fri, 26 Jul 2013 06:24:03 -0400 Message-ID: <51F24E3F.3080102@redhat.com> Date: Fri, 26 Jul 2013 12:23:59 +0200 From: Gerd Hoffmann MIME-Version: 1.0 References: <1374829812-5201-1-git-send-email-aik@ozlabs.ru> <51F2484E.4020308@redhat.com> <1374833448.6142.130.camel@pasglop> In-Reply-To: <1374833448.6142.130.camel@pasglop> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH v3] hcd-ohci: add dma error handling List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Benjamin Herrenschmidt Cc: Alexey Kardashevskiy , Anthony Liguori , qemu-devel@nongnu.org On 07/26/13 12:10, Benjamin Herrenschmidt wrote: > On Fri, 2013-07-26 at 11:58 +0200, Gerd Hoffmann wrote: >> On 07/26/13 11:10, Alexey Kardashevskiy wrote: >>> Current hcd-ohci does not handle DMA errors. However they may happen >>> so here we introduce simple error handling. >>> >>> On such errors, a typical OHCI will stop operating, signal the guest >>> about the error by sending "UnrecoverableError Event", set itself into >>> error state and set "Detected Parity Error" in its PCI config space >>> to signal that it got an error and so does the patch. >>> >>> This also adds ohci_stop() call to ohci_bus_start() to handle possible >>> failure of qemu_new_timer_ns(). >>> >>> Signed-off-by: Alexey Kardashevskiy >> >> Looks sane to me now. When I get an ack from Benjamin I'll go put it >> into the usb patch queue. > > From me ? Heh, I barely remember what the code looks like in there, if > you are both happy with it I don't think you need me :-) Well, ppc emulation uses ohci so I through an ack from a ppc person which also commented on previous revisions would be nice to have ... > The only possibly comment is that I would have called ohci_stop() something > a bit clearer such as ohci_die() or ohci_buserror() ... IE. That function > does more than just "stop", it also signals an error, and that attribute > should probably be described by the function name. xhci has an xhci_die() function for simliar purposes so I think ohci_die() would be a good pick. cheers, Gerd