In article <[email protected]>, Izaak <[email protected]> wrote: >On Sat, Feb 14, 2015 at 10:25:59AM -0500, Christos Zoulas wrote: >> On Feb 14, 4:05pm, [email protected] (matthew green) wrote: >> -- Subject: re: Dealing with debugging macros for printing as part of aprint >> | changing from DPRINTF() to aprint_debug*() will likely change >> | the semantics. these debug messages will now trigger with >> | "boot -x", instead of a DEBUG kernel, won't they? >> | >> | that might be OK -- autoconf messages shouldn't happen normally, >> | unlike other debug messages.. >> >> I meant change the printf's inside DPRINTF() to aprint_debug*() but >> leave the DPRINTF(). Sorry I was not clear. >Sorry, I did not completely realize the implication of my suggestion -- I >am a beginner. If I do it your way though, then I run into the original >problem that I presented -- i.e. in a few places there is shared printf >through DPRINTF being used in both autoconfiguration and elsewhere in the >driver, which should be split into aprint* and log*.
Not a problem, things are complicated enough so we all get them wrong sometimes. Another way would be to introduce ADPRINTF, but this sounds like we are going overboard with complexity and combinations. >So, if I find instances where DPRINTF is only used in autoconfiguration >and nowhere else, then I will do as you say and change the printf inside >the DPRINTF, otherwise I will just leave it alone for now. Still, this >issue will come up again in the second part of the project because all of >those other printfs inside macros will need to be converted. Perhaps fixing the DPRINTF's that deal with autoconf to aprint_debug is the way to go after all... But that introduces code bloat? christos
