* Alexandre Bailon <abailon@xxxxxxxxxxxx> [170118 09:51]: > On 01/18/2017 06:43 PM, Tony Lindgren wrote: > > * Alexandre Bailon <abailon@xxxxxxxxxxxx> [170118 06:54]: > >> On 01/18/2017 12:29 AM, Tony Lindgren wrote: > >>> * Alexandre Bailon <abailon@xxxxxxxxxxxx> [170117 05:16]: > >>>> Currently, the CPPI 4.1 driver is not completely generic and > >>>> only work on dsps. > >>>> Add two functions which will help to move platform code from CPPI 4.1 to > >>>> MUSB glue driver: > >>>> - cppi41_register_dma_callback() to register a platform dma_callback > >>>> (e.g to acknowledge interrupt on DMA completion). > >>>> - cppi41_dma_controller_to_musb() to convert a dma_controller pointer to > >>>> a musb pointer. This is required to get a musb pointer from DMA callbacks. > >>> > >>> Can't you just add a new function to struct dma_controller in musb_dma.h? > >> Yes, I can add the dma_callback function to dma_controller and then > >> remove cppi41_register_dma_callback(). > > > > OK > > > >> But I still to define cppi41_dma_controller_to_musb() except if I add a > >> musb pointer to dma_controller. > > > > Yeah makes sense to me. Or get it from the hw_ep that already has it? > Yes hw_ep has it but it only available in cppi41_dma_callback(). > I also need to get the musb pointer from dma_exit() callback > (defined in musb_platform_ops). > dma_exit only have one argument: struct dma_controller. > Having musb in dma_controller make thing more easier. > The other solution is to add a new argument to dma_exit(). Seems like adding to struct dma_controller is less changes. Bin might have some preference here though. Regards, Tony -- To unsubscribe from this list: send the line "unsubscribe dmaengine" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
