From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?UTF-8?Q?Andreas_F=c3=a4rber?= Date: Wed, 13 Apr 2016 20:17:31 +0200 Subject: [U-Boot] [PATCH] jetson-tk1: Set fdtfile environment variable In-Reply-To: <570E88BD.1050006@wwwdotorg.org> References: <1460551699-30796-1-git-send-email-afaerber@suse.de> <570E41C0.1010904@suse.de> <570E665F.405@wwwdotorg.org> <570E6B19.1090402@suse.de> <570E7B17.9030701@wwwdotorg.org> <570E84F3.5080806@suse.de> <570E88BD.1050006@wwwdotorg.org> Message-ID: <570E8D3B.40601@suse.de> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de Am 13.04.2016 um 19:58 schrieb Stephen Warren: > On 04/13/2016 11:42 AM, Andreas F?rber wrote: >> Am 13.04.2016 um 19:00 schrieb Stephen Warren: >>> Anyway, nothing in your benefits-of-EFI statement implies that relying >>> on $fdtfile being set is correct. That's a new requirement that didn't >>> exist before. Either the requirement needs to be removed (e.g. using a >>> default FDT filename such as "${soc}-${board}${boardver}.dts") or only >>> enabling this functionality on boards that do set $fdtfile, since it >>> relies on that. >> >> $fdtfile needs to be the Linux filename. It does not always follow the >> same pattern as the U-Boot variables you suggest here. >> CONFIG_DEFAULT_DEVICE_TREE ".dtb" might work better, and that was my >> question to you. > > That pattern is a good default that at least historically applied to all > the systems where the distro bootcmds were enabled. Perhaps the set of > systems using the distro bootcmds has increased now so the default isn't > always applicable. Boards can set $fdtfile /if/ needed because of that, > but I don't think should be forced to in all cases where the default > makes sense. I really don't care whether we set fdtfile and use $fdtfile or whether we insert the filename string directly into the appropriate command variable... My point is U-Boot via its jetson-tk1_defconfig / .config knows this (or should know) better than any user. And it seemed to me that variables were not exactly used sparingly in the distro mechanism so far, so I don't see why not to populate that variable _if_ we know what its value needs to be. Do you have any real reason for being against populating fdtfile at whatever level turns out to be suitable? I believe there is no argument that this patch will not be applied. However I am strongly rejecting your attitude that everything is there already with variables and that nothing new is needed. Something needs to be done somewhere - and we need to figure out what exactly and where for minimum impact to the release. Thanks, Andreas -- SUSE Linux GmbH, Maxfeldstr. 5, 90409 N?rnberg, Germany GF: Felix Imend?rffer, Jane Smithard, Graham Norton HRB 21284 (AG N?rnberg)