From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tomi Valkeinen Date: Wed, 20 Mar 2013 11:44:47 +0000 Subject: Re: [PATCH 03/14] OMAPDSS: DSI: simplify dsi configuration Message-Id: <5149A12F.2050101@ti.com> MIME-Version: 1 Content-Type: multipart/mixed; boundary="------------enig7EC2ECED6B419F2C6A283B20" List-Id: References: <1362743569-10289-1-git-send-email-tomi.valkeinen@ti.com> <1362743569-10289-4-git-send-email-tomi.valkeinen@ti.com> <51499C73.3060802@ti.com> In-Reply-To: <51499C73.3060802@ti.com> To: Archit Taneja Cc: linux-omap@vger.kernel.org, linux-fbdev@vger.kernel.org --------------enig7EC2ECED6B419F2C6A283B20 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable On 2013-03-20 13:24, Archit Taneja wrote: > On Friday 08 March 2013 05:22 PM, Tomi Valkeinen wrote: >> We have a bunch of dsi functions that are used to do the basic >> configuration for DSI. To simplify things, and to make sure we have al= l >> the necessary information, create a single dsi config function, which >> does the basic configuration. >=20 > I had split these funcs in the manner so that they could be converted > into generic output ops or something equivalent to what we anticipated > CDF to represent encoders. Hence, we may have to split this into smalle= r > funcs again later :p Well, it was from the CDF discussions that this change arose. Everybody wanted a simpler way than n+1 functions. And I think it makes sense. It makes it possible to manage the configuration as one "whole", instead of small bits that may have interdependencies. E.g the size of the output affects video mode calculations, so one had to make the calls in certain order. Now we have all the needed information in one piece. We could, perhaps, have common parts between different video busses, but I'm not sure if configuration is one of those common parts. > Also, set_size and set_timings were 2 different ops for command and > video mode panels respectively. omapdss_dsi_set_size() also came in use= > when we supported rotation in Taal. We have an equivalent func for rfbi= =2E Yep. I felt it's a bit confusing, so I just combined them. Even for command mode some timing information is good (well, pixel clock), to calculate proper DSI bus speed. I think this also works in case of panel rotation. From DSS's point of view (and that's what we're talking about when setting the timings) there's no rotation. It's the panel's internal thing. Tomi --------------enig7EC2ECED6B419F2C6A283B20 Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) Comment: Using GnuPG with undefined - http://www.enigmail.net/ iQIcBAEBAgAGBQJRSaEvAAoJEPo9qoy8lh718c4P/2vpBNVX4bLiUCKUauphB8Fq O3kBIsQNjbATjVWrAqb8OOcQTqDvYMTacDxXJmaanO59jt/nyJZUQUG7gM0Zn+VK p4Fx+0rF7dRYlNNkESIlPPPLdf+1jioO1xIm4BsdwE/9zLvrh5pxDrNeDdQp7zMr p67Qs/i8B0YhcDZTezVdGTN8ehpwP4KJccycGlbk7Iuro9WIK751rRzGZL8UBuWi 8wpoMgEkvjsLzqSSLDYglAlBU8Gf9oqKoRKG6QPmirZoxuZHsgymtIPZTPOmgrYJ aSNlEa/VZVoeyF9zfq5PBK+c+I3JQHmOq0ttsVYbUKPuxbrHp22ZhZjy+ZgGp4z3 tUbwUgFKXLyZ+0erjtCO/MOnZKysVZNA/vUB4MSLLHqxuC2qQy/KXOqosvenEo6k 83/dedQaFv68TA7Y1/gpTJ9N+vSxURjqW+8swvCPpyfv5COuSb8FZ7vSevDmJNAC lBgxOOWHCHkyVEpFNSmVSN5rKO9BE6RjnbW9W2Dm4pQbrQXfvqQcbwWOZqDLSFSa 72NS7NUDb/VuzE9REuFHRcfpGrpzhKhNyyC4mZMI4LnhXBlw6kffTSrOutntK57I ijHmR3YBgD/1vbKHyAusisghFW7XkQWBDvblHyz4urAJGET2R7z2he0uDlFJdSqD HKeyLeYOR8SBb6l898kj =dlDW -----END PGP SIGNATURE----- --------------enig7EC2ECED6B419F2C6A283B20--