From mboxrd@z Thu Jan 1 00:00:00 1970 From: Michal Simek Date: Mon, 14 Aug 2017 13:50:56 +0000 Subject: Re: [PATCH 2/2] video: fbdev: Enable Xilinx FB for ZynqMP Message-Id: <994f5b9f-9ba7-ad80-84c4-de1edd9e56d0@xilinx.com> List-Id: References: <757d86d1-cca5-3a99-a371-a4c7e1c76f9b@xilinx.com> In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Daniel Vetter , Michal Simek Cc: Linus Walleij , "linux-kernel@vger.kernel.org" , "monstr@monstr.eu" , Benjamin Gaignard , Tomi Valkeinen , Bartlomiej Zolnierkiewicz , Arnd Bergmann , "linux-fbdev@vger.kernel.org" , Krzysztof Kozlowski , Paul Burton On 14.8.2017 15:49, Daniel Vetter wrote: > On Mon, Aug 14, 2017 at 3:46 PM, Michal Simek wrote: >> On 14.8.2017 15:42, Linus Walleij wrote: >>> On Tue, Aug 8, 2017 at 10:10 AM, Michal Simek wrote: >>> >>>> Enable this driver for Xilinx ZynqMP. >>>> >>>> Signed-off-by: Michal Simek >>> >>> Not a comment on the patch per se, but I would advice you to look into >>> migrating the Xilinx frame buffer to DRI/DRM/KMS. >>> >>> The recent CMA helpers has made it very simple and focused to write >>> DRM drivers. See for example drivers/gpu/drm/pl111 >> >> There is a lot of work done by xilinx in this area gpu/drm/kms. I am >> trying to upstream what xilinx has in soc tree and this change is there. >> Maybe in this case it is useless and none is using it but it is there >> and I don't want to break user. There shouldn't be any technical problem >> to run this old fb driver in pl on arm64. > > A proper drm kms driver gives you backwards compat for fbdev for free. > And we're taking in patches to fill feature gaps in our fbdev > emulation as people hit them. There no reason nowadays at all to merge > new fbdev native drivers, your users will be taken care of by a drm > kms driver fully. You can look at xilinx work in this area. https://github.com/Xilinx/linux-xlnx Unfortunately they haven't started to upstream it. In this particular case I am not aware about anybody who is going to write new driver for this soft IP. Thanks, Michal