Eric Miao <eric.y.miao@xxxxxxxxx> writes: > Robert, > > I seem to remember PXA27x has 24 EPs, and why NR_PXA_ENDPOINTS > is 16 here? You're correct, there are 24 endpoints. The trick is we make a mapping between : - USB endpoints (ie. enpoint number, config, alternate setting, interface) - PXA endpoints As things go, we don't have 24 mappings, but far less. That's why the static table of endpoints is smaller than 24 (as we have 16 mappings). And I think the compiler is right. On buggy hardware, if an interrupt is generated on endpoint epW for example, an access out of bounds occurs. I didn't ack the patch yet because I hoped to find a "prettier" solution (was thinking of a while loop instead of the for, and reuse what we discovered while optimizing the dma irq handler loop). I need few days to grasp a bit of time. Cheers. -- Robert -- 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
