Hi Sergei, On Thu, Jun 17, 2010 at 19:42:11, Sergei Shtylyov wrote: > Christophe Aeschlimann wrote: > > >>> I was surprised to see that the PINMUX registers configuration (in > >>> da850.c) is SOC dependant (part of common initialisation). > > >> Frankly speaking, I don't see where the pin lists are a part of common > >> initialization. Unless you mean da850_pins[] -- that is SoC specific of > >> course as the PinMux registers are situated in the SoC. > > > Yes I realised that after Caglar's answer. I first thought the > > da850_pins structure in da850.c was used to define the functionality of > > every PINMUX where it just defines a possible configuration. > > >>> I think the PINMUX configuration is definitely a board specific setting. > > >> The pin lists there are just the default ones, readily overridable by > >> the board code. With the annoyng exception of da850_nand_pins[] and > >> da850_nor_pins[] which should have been in board-da850-evm.c really... > > >>> Do you agree with me ? Did I miss something and the PINMUX > >>> configuration can already be easily overloaded in the board specific > >>> configuration ? > > >> Yes. > > > I was just surprised I had to add my own PINMUX possible configuration > > inside the da850_pins struct in da850.c > > No wonder -- that table is just incomplete. TI was too lazy to enter > every PinMux field there.
Why should every pin on the SoC be entered here? Linux code doesn't care about many pins on the SoC. Entering each and every pin here would be code bloat. Adding pin definitions as and when required is a more efficient approach. Thanks, Sekhar _______________________________________________ Davinci-linux-open-source mailing list [email protected] http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
