From mboxrd@z Thu Jan 1 00:00:00 1970 From: Xiangfu Liu Subject: Re: SDIO wifi card works with 2.6.36, not with 3.2 Date: Sat, 26 Nov 2011 09:02:55 +0800 Message-ID: <4ED03ABF.5070604@gmail.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from mail-bw0-f46.google.com ([209.85.214.46]:55001 "EHLO mail-bw0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754254Ab1KZBDO (ORCPT ); Fri, 25 Nov 2011 20:03:14 -0500 Received: by bke11 with SMTP id 11so5063708bke.19 for ; Fri, 25 Nov 2011 17:03:13 -0800 (PST) In-Reply-To: Sender: linux-mmc-owner@vger.kernel.org List-Id: linux-mmc@vger.kernel.org To: =?UTF-8?B?VGhvbWFzIExhbmfDpXM=?= Cc: linux-mmc@vger.kernel.org Hi I am not sure if this is the root cause. by revert this commit.[1] our SDIO ks7010 wifi card work again in linux 3.0. check out this emails, under mailing list: SDIO single IRQ optimization breaks libertas http://permalink.gmane.org/gmane.linux.kernel.mmc/8338 [1] Revert "mmc: sdio: optimized SDIO IRQ handling for single irq" This reverts commit 06e8935febe687e2a561707d4c7ca4245d261dbe. On 11/26/2011 06:26 AM, Thomas Lang=C3=A5s wrote: > I have a slightly weird problem; I have an Samsung Exynos 4210 with a= n BCM4329 > wifi-chip connected through the MMC-bus using SDIO. > > This setup works fine with 2.6.35.7 Android-kernel, but not with > 3.2-rc1 (non-Android). > > CMD52 read/write works correctly, but CMD53 only works with reads. Al= l CMD53 > writes fails. > > Here's an example output of the errorlog: > [ 267.175000] brcmf_sdioh_request_word: rw=3D1, func=3D1, addr=3D0x0= bf98, nbytes=3D4 > [ 267.175000] mmc1: starting CMD53 arg 957f3004 flags 000001b5 > [ 267.175000] mmc1: blksz 4 blocks 1 flags 00000100 tsac 1000 ms= nsac 0 > [ 267.175000] sdhci [sdhci_irq()]: *** mmc1 got interrupt: 0x0020800= 3 > [ 267.175000] mmc1: req done (CMD53): 0: 00001000 00000000 00000000 = 00000000 > [ 267.175000] mmc1: 0 bytes transferred: -84 > > I've checked the setup registers (pad-controlls to see that all pads > are set to mmc > instead of gpio, and mmc registers). All are identical on Android an= d > non-Android, > so it doesn'tseem like the problem is there at least. > > Any ideas on how to proceed with locating the problem? >