On Tue, Jun 25, 2013 at 06:49:12PM +0100, Rupesh Gujare wrote: > On 25/06/13 18:02, Greg KH wrote: > >On Tue, Jun 25, 2013 at 05:30:02PM +0100, Rupesh Gujare wrote: > >>convert all debug messages from printk to dev_dbg() & add kernel config to > >>enable/disable these messages during compilation. > >No, just use the built-in dynamic debug code in the kernel, no need to > >provide any new macros or functions or most importantly, no new Kconfig > >options. > > > > New macro (oz_trace) is being used as pointer to "struct device *" > is not available in all functions for dev_dbg() function. > Please let me know if there is better way to handle this, I will be > happy to rework on this. Then use pr_debug() for the places you don't have a struct device *, everywhere else should use dev_dbg(). > As well new Kconfig option was added to pass CFLAGS to compiler, so > that dev_dbg will get compiled on system where DYNAMIC_DEBUG is not > defined. No, just rely on the overall config option, do you really think the kernel should have an individual Kconfig debug option for every individual driver? We have been removing these for years, don't move backwards in this area please. > I was assuming that it is a standard practice, as I can find similar > Kconfig option for other drivers. Or am I wrong in my understanding > ? Those options for those drivers should be removed, they were added before we had the dynamic debug option. New drivers shouldn't have this at all. > Main idea here is to replace printk with dev_dbg(), which gives us > option to enable individual log during runtime if DYNAMIC_DEBUG is > enabled & provide mechanism to compile this when DYNAMIC_DEBUG is > not enabled. > > Again many of these logs will be removed in future, as currently > there are too many logs which are not required, hence idea is to > only change macro & then remove remaining logs in future. No, remove the unneeded calls now, why wait? thanks, greg k-h _______________________________________________ devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxx http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
