From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id E9A43182BF for ; Mon, 6 Nov 2023 11:28:31 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.b="Uax+WBUG" Received: by smtp.kernel.org (Postfix) with ESMTPSA id E9DC8C433CC; Mon, 6 Nov 2023 11:28:30 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linuxfoundation.org; s=korg; t=1699270111; bh=KHYMSuYzfCZFn7iKlZaFmc8xH2hKsNURN3exYOYIYF4=; h=Date:From:To:Cc:Subject:References:In-Reply-To:From; b=Uax+WBUGC+X1enzmvLmjGabeE12V0LXoYD77+J300r07tfDIRIziiqnd99eIcc2xe 4WBQZoJPBW/Tw3TswVDR6L5ZKf3bXfKbPHf+cpL8YlhGv0hKXLed8EYJxBi6OzKfPL PKVUyjH1W4qz5PvcblJoPRSRhpNDG2rlwbmaLTo4= Date: Mon, 6 Nov 2023 12:28:28 +0100 From: Greg KH To: Matthias Schiffer Cc: Sasha Levin , stable@vger.kernel.org, Faiz Abbas , =?iso-8859-1?Q?Beno=EEt?= Cousson , Tony Lindgren , Rob Herring , Mark Rutland , linux-omap@vger.kernel.org, devicetree@vger.kernel.org, linux-kernel@vger.kernel.org, linux@ew.tq-group.com Subject: Re: [PATCH 5.4] Revert "ARM: dts: Move am33xx and am43xx mmc nodes to sdhci-omap driver" Message-ID: <2023110613-fringe-ended-5390@gregkh> References: <20231106095048.8396-1-matthias.schiffer@ew.tq-group.com> Precedence: bulk X-Mailing-List: devicetree@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20231106095048.8396-1-matthias.schiffer@ew.tq-group.com> On Mon, Nov 06, 2023 at 10:50:48AM +0100, Matthias Schiffer wrote: > This reverts commit d0c69c722ff16ce2481a5e0932c6d5b172109f21. > > The reverted commit completely breaks MMC on the AM33xx/AM437x for > multiple reasons: > > - The changed compatible strings ti,am335-sdhci and ti,am437-sdhci > aren't supported on Linux 5.4 at all, so no driver is found > - Even when additionally backporting the support for these compatible > strings in the sdhci-omap driver, I could not the the MMC interfaces > to work on our TQMa335x SoM - the interface would time out during card > initialization for both an eMMC and an SD card. > > I did not investigate the cause of the timeouts further, and instead > just reverted the commit - switching to a different MMC driver in a stable > kernel seems like a rather risky change unless it's thoroughly tested, > which has obviously not happened in this case. > > The reverted commit is also given as a Stable-dep-of commit 2eb502f496f7 > ("ARM: dts: am33xx: Fix MMCHS0 dma properties"), however the conflict > resulting when only the one commit is reverted is trivial to resolve, > which leads to working MMC controllers again. > > Signed-off-by: Matthias Schiffer > --- > > I have not checked if other stable kernels exhibit the same breakage; it > might be a good idea to revert the change for all stable branches unless > it is proven that the sdhci-omap driver actually works. This was only backported to 5.4.y, thanks for catching this, now queued up. greg k-h