On Fri, Apr 22, 2022 at 07:10:34AM +0200, Greg KH wrote: > On Thu, Apr 21, 2022 at 08:30:25PM -0400, Ian Cowan wrote: > > On Thu, Apr 21, 2022 at 06:52:04PM +0300, Dan Carpenter wrote: > > > On Thu, Apr 21, 2022 at 11:22:00AM -0400, Ian Cowan wrote: > > > > > > > > For using the dev_dbg() macro, do you define this in the header file > > > > (i.e. for this it would be videocodec.h), or where should this be > > > > included from? > > > > > > dev_dbg() is defined in include/linux/dev_printk.h. Look around at how > > > it's used. pr_debug() might be an option, but I don't know if we will > > > accept that, we prefer dev_dbg(). > > > > > > regards, > > > dan carpenter > > > > > > > I'm about to submit the modified patch, but I went and looked and we > > cannot use dev_dbg() because these specific drivers do not have any > > association with a device struct. > > Then please fix that issue, as there is a real struct device that they > are using somewhere. That is the correct solution as drivers should > never use pr_* calls directly. > > thanks, > > greg k-h > I just resubmitted and this time used dev_dbg() (I split the resubmitted patch into 4 parts that will work if they are applied in order). There were a few places that pr_debug() had to be used (or it can be changed back to dprintk()) because they are cleanup functions and do not have any devices available at the cleanup. If there is a better way to handle that, I will go back and make a modification to that. Other than those 3 spots, dev_dbg() is being used. --- Ian Cowan
