* Sergei Shtylyov | 2013-08-14 22:39:49 [+0400]: >Hello. Hello Sergei, > Perhaps it's worth documenting the "mg" prefix in >Documentation/devicetree/bindings/vendor-prefixes.txt if it's >official? Okay, I sent a separate patch for this. >>>>+- power : Should be "250". This signifies the controller can supply >>>>up to >>>>+ 500mA when operating in host mode. > >>> This one too. > >>Hmm. I'm not too happy with this beeing as defined by the USB >>specification. Would you mind if I change it to mA instead? > > I don't care much. I just wanted it prefixed. okay. >>>>+DMA >>>>+~~~ >>>>+- compatible: ti,am3359-cppi41 > >>> "ti,cppi41" should follow I think. > >>I would go for ti,am3359-cppi41 because it has been tested on this >>device. I wouldn't mind flipping 9 by an x. > > No, wildcards are not allowed. Then it stays as it is. >< The problem is that it is >>not generic cppi41 because it needs that glue thingy > > Glue thingy doesn't control or affect CPPI 4.1 controller in any >way I suspect. > >>and it is only >>able to do USB transfers. > > Who told you this? CPPI 4.1 is a generic packet oriented DMA controller. >The USB DMA registers do not belong to CPPI 4.1, and they control >only how CPPI FIFOs form the packets for CPPI 4.1 (it concatenates >USB packets in generic RNDIS mode) Yes, this part is understood. However I need the glue layer which is am335x specific. The glue layer covers interrupt handling. Everything is seems cppi41 specific. We had this discussion once I think. I need to pass this information somehow and passing it the binding is a good solution imho. >>So adding a trailing USB might be good > > No, it's not good. > >>but the am335x has only one cppi41 engine which does USB (the network one >>is cppi3 something). >>What you think? > > I didn't get the last passage, frankly speaking. Yes, I know CPPI >4.1 only serves USB and is a part of a large, dual controller USB >peripheral in AM335x but at least it's not per MUSB controller... So I would say we leave it as it. No wildcard, no trailing USB. This driver as it can now handle the cppi41 on am3359. If it pops up on another SoC it will use the same binding the glue layer (for interrupt handling) is compatible or it will be extended. >WBR, Sergei Sebastian -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
