Em Tue, 22 Jul 2014 04:05:05 +0300 Antti Palosaari <crope@xxxxxx> escreveu: > On 07/22/2014 03:51 AM, Mauro Carvalho Chehab wrote: > > Em Tue, 22 Jul 2014 03:08:19 +0300 > > Antti Palosaari <crope@xxxxxx> escreveu: > > > >> So what. Those were mostly WARNING only and all but long lines were some > >> new checks added to checkpatch recently. chekcpatch gets all the time > >> new and new checks, these were added after I have made that driver. I > >> will surely clean those later when I do some new changes to driver and > >> my checkpatch updates. > > > > Antti, > > > > I think you didn't read my comments in the middle of the checkpatch stuff. > > Please read my email again. I'm not requiring you to fix the newer checkpatch > > warning (Missing a blank line after declarations), and not even about the > > 80-cols warning. The thing is that there are two issues there: > > > > 1) you're adding API bits at msi2500 driver, instead of moving them > > to videodev2.h (or reusing the fourcc types you already added there); > > If you look inside driver code, you will see those defines are not used > - but commented out. It is simply dead definition compiler optimizes > away. It is code I used on my tests, but finally decided to comment out > to leave some time add those later to API. I later moved 2 of those to > API, that is done in same patch serie. > > No issue here. > > > 2) you're handling jiffies wrong inside the driver. > > > > As you may know, adding a driver at staging is easier than at the main > > tree, as we don't care much about checkpatch issues (and not even about > > some more serious issues). However, when moving stuff out of staging, > > we review the entire driver again, to be sure that it is ok. > > That jiffie check is also rather new and didn't exists time drive was > done. Jiffie is used to calculate debug sample rate. There is multiple > times very similar code piece, which could be optimized to one. My plan > merge all those ~5 functions to one and use jiffies using macros as > checkpatch now likes. I don't see meaningful fix it now as you are going > to rewrite that stuff in near future in any case. Ok, I'll apply the remaining patches. > Silencing all those checkpatch things is not very hard job though. If > you merge that stuff to media/master I can do it right away (I am > running older kernel and older checkpatch currently). FYI, I always use the checkpatch available on our tree, no matter what Kernel I'm running. My scripts just call ./scripts/checkpatch.pl. Regards, Mauro -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
