From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: <51C213E4.5050905@grandegger.com> Date: Wed, 19 Jun 2013 22:26:12 +0200 From: Wolfgang Grandegger MIME-Version: 1.0 References: <50CB364C.70706@wanadoo.fr> <50CB561D.5070706@grandegger.com> <51B0353B.7060206@grandegger.com> In-Reply-To: <51B0353B.7060206@grandegger.com> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Subject: Re: [Xenomai] Init issue with rtcan_flexcan List-Id: Discussions about the Xenomai project List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Henri Roosen Cc: Thierry Bultel , xenomai@xenomai.org Hi Henri, On 06/06/2013 09:07 AM, Wolfgang Grandegger wrote: > Hi Henri, > > On 06/05/2013 09:15 AM, Henri Roosen wrote: >> On Fri, Dec 14, 2012 at 5:38 PM, Wolfgang Grandegger wrote: >> >>> On 12/14/2012 03:23 PM, Thierry Bultel wrote: >>>> Hi, >>>> some days ago I have posted a patch for the rtcan_flexcan to be >>>> correctly probed on the iMX6q >>>> >>>> It is functional, but I have not tested it enough, I am sorry about that. >>>> It only works if the linux flexcan driver module has been used once >>> before. >>>> >>>> Else I have got IRQs but no packet go to the upper layer, I think it is >>>> stuck in passive state, I am investigating on it. >>> >>> It would be nice if you could port the missing patches from the mainline >>> Linux kernel to rtcan_flexcan. >>> >> >> I have ported code from the mainline linux flexcan driver that fixes the >> behavior described by Thierry. Please find attached: >> 0001-make-rtcan_flexcan-driver-compatible-with-non-OpenFi.patch >> 0002-Fix-rtcan_flexcan-for-mx6-for-OpenFirmware-platforms.patch >> >> The attached patches can be applied to the current mainline xenomai git >> (anywhere after commit-3ddc61cda095a9ae51277608d7fd4e9c2c481625). They make >> Thierry's patch obsolete. >> >> I have tested the rtcan_flexcan driver in the non-CONFIG_OF configuration >> only. On a wandboard with iMX6Solo using the freescale 3.0 kernel. I only >> compile tested the CONFIG_OF configuration for the current ipipe master >> (3.8) because wandboard solo is not yet supported by OpenFirmware. It would >> be great if someone could test the OF version. > > Thanks for your contribution. Looks good at a first glance. Just need to > check back- (and forward-) compatibility. I will also try to get access > to boards using Flexcan. Hope to find some time over the WE. So far I didn't get a chance to test you driver modifications. Anyway, the patch looks good. Gilles, could you please add my "Acked-by: Wolfgang Grandegger " and apply the patch, thanks. Thanks for your contribution, Wolfgang.