From mboxrd@z Thu Jan 1 00:00:00 1970 From: Marek Vasut Date: Wed, 25 Jan 2012 18:53:20 +0100 Subject: [U-Boot] non SPL boot on beagleboard-xm In-Reply-To: <4F203A6F.40305@ti.com> References: <4F203A6F.40305@ti.com> Message-ID: <201201251853.20895.marek.vasut@gmail.com> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de > On 01/25/2012 08:43 AM, Peter Meerwald wrote: > > Hello Tom, > > > >>> MLO/SPL + current u-boot breaks the USB ethernet patches (i.e. booting > >>> via tftp) > >> > >> Even with 'dc off' before you start usb? > > > > I was not aware of the u-boot command 'dc off' -- yes, it solve my USB > > ethernet problem > > > > here is my story (from bisecting): > > > > v2011.06 + Koen et al. patches (git://github.com/koenkooi/u-boot.git) > > works > > > > shortly afterwards, dcache was enabled (commit > > c2dd0d45540397704de9b13287417d21049d34c6) and USB ethernet fails > > > > v2011.09 works if using 'dc off' before 'usb start' (I tried it as a > > recent non-SPL u-boot version) > > > > u-boot current (137703b811502dfea364650fb3e17f20b4c21333) works as well > > with 'dc off' before 'usb start' > > So what's going on is that the dcache has been enabled now, but the USB > stack is not safe, so we have to issue 'dcache off' (or 'dc off') before > starting USB (or in other cases, networking). Why would it not be safe? Because we don't have the unified cache ops in place yet? :-( M