This can't have been build-tested.
On Sunday 15 March 2009, [email protected] wrote: > static struct edma_soc_info dm355_edma_info = { DM355 ... > .n_channel = 64, > .n_region = 4, > .n_slot = 128, > .n_tc = 2, > .noevent = dma_chan_dm355_no_event, > + .queue_tc_mapping = dm646x_queue_tc_mapping, > + .queue_priority_mapping = dm646x_queue_priority_mapping, .... vs DM646x. Those mapping tables should probably use signed 8-bit integers; no point in wasting memory. Please preserve field alignment, as you did for at least the dm646x. And I think I should have looked more carefully at the EDMA documentation before defining those n_* fields, since the EDMA_CCCFG register provides them. It'd be best to get rid of those n_* fields. - Dave > }; > _______________________________________________ Davinci-linux-open-source mailing list [email protected] http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
