From mboxrd@z Thu Jan 1 00:00:00 1970 From: Holger Schurig Subject: Re: [PATCH v2 00/12] can: rx-offload: add implmentation and switch flexcan driver to use it Date: Wed, 07 Sep 2016 08:33:57 +0200 Message-ID: <878tv48bhm.fsf@gmail.com> References: <1467657137-18891-1-git-send-email-mkl@pengutronix.de> Mime-Version: 1.0 Content-Type: text/plain Return-path: Received: from mail-wm0-f67.google.com ([74.125.82.67]:33213 "EHLO mail-wm0-f67.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754746AbcIGGeA (ORCPT ); Wed, 7 Sep 2016 02:34:00 -0400 Received: by mail-wm0-f67.google.com with SMTP id w207so1295032wmw.0 for ; Tue, 06 Sep 2016 23:34:00 -0700 (PDT) In-Reply-To: <1467657137-18891-1-git-send-email-mkl@pengutronix.de> Sender: linux-can-owner@vger.kernel.org List-ID: To: Marc Kleine-Budde , linux-can@vger.kernel.org Marc Kleine-Budde writes: Late answer :-) This driver was tested with Linux 4.7.2 an on i.MX6Q device. Not by me, I don't even have their exact test software. I just get the reports ... Using some external tool (maybe Kvaser?) they generated 80% busload at 500 kbps, the driver survived for 3 days without any data loss. So, compared with the current in-kernel driver, this is a definitive improvement! Then they did same CAN ping-ping test with reboots and wrote: "Out of about 16,000 reboots we had one failure. This failed both CAN interfaces and the logs show that the PTX-C received all CAN messages and returned them, but the test machine for some reason did not receive any CAN messages." I don't have any idea why Linux thought it was sending but the receiver didn't receive anything. And I'm no CAN expert, not at all. Wouldn't I get an error condition on the sending side? How would a program get this error, is it a return value to sendto() Holger