On 01/14/2014 07:29 AM, Rafał Miłecki wrote: > 2014/1/13 Arend van Spriel <arend@xxxxxxxxxxxx>: >> The IOCTL register in the agent/wrapper contains additional bits >> that are core specific and use in the core reset sequence. > > I'm not sure about this. I don't think we want to keep device-specific > bits in a commond (bcma's) code. True. We can put these in brcm80211 includes so both brcmfmac and brcmsmac can use it. That said the bcma_core_disable() and bcma_core_enable() do need to be corrected to be reliable. Patch #2 in this series show what needs to be changed. Regards, Arend > For example, I can't see PCI defines (like linux/pci_regs.h) having > any device specific bits in this shared code. > > Our case with bcma is a bit more tricky, because we have a single > register with common and device-specific bits at the same time. So > maybe I'll find another example. > > What about USB? Let's say linux/usb/ch9.h. There is a struct > usb_ctrlrequest that has bRequestType and bRequest. > bRequestType accepts some common values like USB_DIR_IN, > USB_RECIP_INTERFACE, etc. They are defined in the same file. > bRequest values are device-specific and you don't have them defined in > the above file. > That makes pretty much sense for me. > -- > To unsubscribe from this list: send the line "unsubscribe linux-wireless" in > the body of a message to majordomo@xxxxxxxxxxxxxxx > More majordomo info at http://vger.kernel.org/majordomo-info.html > -- To unsubscribe from this list: send the line "unsubscribe linux-wireless" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
