Hi, Alan Stern <stern@xxxxxxxxxxxxxxxxxxx> writes: >> >>> The following series was compile-tested only (so far, at least). I >> >>> wanted to get some comments from folks to see what you guys think >> >>> about this before running tests. >> >>> >> >>> I don't have any device available which would support PTM_STATUS so I >> >>> guess I'd have to implement it on g_zero, if at all possible. >> >>> >> >>> Best Regards >> >>> >> >>> Felipe Balbi (4): >> >>> usb: core: add Status Type definitions >> >>> usb: core: rename usb_get_status() 'type' argument to 'recip' >> >>> usb: core: add a 'type' parameter to usb_get_status() >> >>> usb: core: add two usb get status helpers >> >> >> >> You should switch the order of patches 3 and 4. That is, replace >> >> usb_get_status() with usb_get_std_status() first, and then add the type >> >> parameter and the other helper routine. That way you won't have to >> >> update all the callers twice. >> > >> > heh, good point :-) >> >> Actually, without patch 3 where it is, I won't have the new 'type' >> parameter to write the helpers. What I could do, is in a single patch >> add the helpers (and the new argument) while also converting all current >> users to usb_get_std_status(). That would basically be a combination of >> patches 3 and 4. > > Make patch 3 add the first helper (usb_get_std_status) and convert all > the callers. The helper can simply call the existing routine, with no > extra argument. > > Then patch 4 will add the extra argument and the new helper. It won't > have to update the callers at all, just change the first helper. fair enough. I just did that. Will wait until next week to resend. -- balbi
Attachment:
signature.asc
Description: PGP signature
