From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from mms1.broadcom.com ([216.31.210.17]:3209 "EHLO mms1.broadcom.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753710Ab3IFQuX convert rfc822-to-8bit (ORCPT ); Fri, 6 Sep 2013 12:50:23 -0400 Message-ID: <522A07C2.5040906@broadcom.com> (sfid-20130906_185026_622904_67F4E3CF) Date: Fri, 6 Sep 2013 18:50:10 +0200 From: "Arend van Spriel" MIME-Version: 1.0 To: "Hauke Mehrtens" cc: =?UTF-8?B?UmFmYcWCIE1pxYJlY2tp?= , linux-wireless Subject: Re: bcma problem on x86_64 References: <522879D0.2030107@broadcom.com> <52299AC3.5010809@broadcom.com> <5229F3D1.9000200@broadcom.com> <522A04B4.3050708@hauke-m.de> In-Reply-To: <522A04B4.3050708@hauke-m.de> Content-Type: text/plain; charset=utf-8; format=flowed Sender: linux-wireless-owner@vger.kernel.org List-ID: On 09/06/2013 06:37 PM, Hauke Mehrtens wrote: > On 09/06/2013 05:25 PM, Arend van Spriel wrote: >> On 09/06/2013 11:05 AM, Arend van Spriel wrote: >>> On 09/06/2013 10:05 AM, Rafał Miłecki wrote: >>>> Hi, >>>> >>>> 2013/9/5 Arend van Spriel : >>>>> Since 3.11-rc4 I am seeing a problem with bcma on x64 (see attached >>>>> log). I >>>>> thought I misconfigured my setup, but just upgraded to 3.11 and I am >>>>> still >>>>> seeing the same issue. Did you have any reports like this? >>>> >>>> Unfortunately I wasn't testing final 3.11 with x86_64, I'll give it a >>>> try over the weekend. >>>> >>> >>> I am bisecting. Will let you know when I find something. >> >> Bisect points to: >> >> fd4edf197544bae1c77d84bad354aa7ce1d08ce1 is the first bad commit >> commit fd4edf197544bae1c77d84bad354aa7ce1d08ce1 >> Author: Hauke Mehrtens >> Date: Mon Jul 15 13:15:08 2013 +0200 >> >> bcma: fix handling of big addrl >> >> The return value of bcma_erom_get_addr_desc() is a unsigned value >> and it >> could wrap around in the two complement writing. This happens for one >> core in the BCM4708 SoC. >> >> Signed-off-by: Hauke Mehrtens >> Signed-off-by: John W. Linville >> >> It is probably caused by using IS_ERR_VALUE() macro which does a >> unsigned long cast, which gives different results on 64-bit platform. >> >> This patch was submitted upstream yesterday by Dave for 3.12-rc1. >> >> Regards, >> Arend >> > Hi Arend, > > Thanks for spotting this. This commit is not in final 3.11, otherwise > I would have suspected it before. Yeah. It will need to be fixed for 3.12 after the merge window. > Could you please try the attached patch. Just tried my own patch, which essentially does the same (not using a macro). So you may add Acked-by: or Tested-by: Arend van Spriel Whatever you think is most appropriate. Regards, Arend