On Tuesday 14 April 2009, Sergei Shtylyov wrote: > > Texas Instruments DA830/OMAP-L137 glue layer for the MUSBMHRDC driver. Please hold off on these patches -- except maybe as an RFC -- until the DA830 stuff merges to mainline. A noted before, it can't build yet, so it's nowhere near suitable for merging... That said, a quick look over this code suggests that it's basically OK. Though: > + /* Workaround: setup IRQs through both register sets. */ Workaround *what* problem? Please reference an erratum or something, so that if a better workaround (perhaps more sensitive to chip revisions, like "not needed on rev C") becomes available, this can be properly removed. > + musb->clock = clk_get(NULL, "USB20CLK"); You should provide the device. I tihnk with the latest rework of the DaVinci clock framework, the NULL is best passed for the clock name instead. Also: - that OTG timer is still messy, and probably at least partially wrong ... likely it'll get cleaned up on "normal" DaVinci chips first though. - Is this still using the 1.3 RTL instead of something newer? - Didn't you fix all the "ret != IRQ_HANDLED" cases for the IRQ handler? - Handling of the PHY is no better here than in the DaVinci support ... sigh. -- 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
