[KJ][PATCH] kj-devel.pl

[Vicente Feito <vicente.feito@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 1689 bytes --]

This patch goes against the original file here: 
http://www.kerneljanitors.org/scripts/kj-devel.pl

Randy: The comment you've made about spinlocks, you can easily let them out of 
the checking by using --nospinlocks, but, I don't understand the comment on 
using the nr line at the begining instead of using it at the end, I mean, any 
special reason? (Also, I'm not a native english speaker so I really appreciate 
the corrections, since I've learned english from HBO, not kidding :).

I've added checks for the use of memset with backward parameters, because 
ultimately a lot of these appeared.

I was about to rewrite some part of the code last night to make it more user 
friendly in order to allow other people add their regexes easily, but I don't 
know if someone is even using this, so I've avoided that by now, if I get 
some feedback on new things to check (besides the ones I already got to add) 
I'll rewrite that part, I think it would help.

I'm still wondering wether or not to check for the use of task queues, I think 
people still use them, so suggestions are more than welcome about this (and 
anything else).

Now that ldd3 is out, people can even send more suggestions in order to check 
for other things, but that's up to the janitors team members who hang in the 
channel 24/7 hardly speaking (how is this possible?they're always on profound 
meditation states, shhh ;)

kj-devel.pl net/*/*.c - it's plagued of return ESOMETHING (it's plagued of 
everything) instead of return -ESOMETHING;  (of course there are some false 
positives because of the existance of i.e. return ETH_HLEN which is correct) 
but some are wrong.

Vicente.

[-- Attachment #2: kj-devel.patch --]
[-- Type: text/x-diff, Size: 1443 bytes --]

256,257c256,257
< 					print "Unlocking a lock that hasn't been aquired ".
< 						"yet(could be a false positive) - Line:$linenr\n\n";
---
> 					print "Unlocking a lock that hasn't been acquired ".
> 						"yet (could be a false positive) - Line:$linenr\n\n";
272,273c272,273
< 					print "Unlocking a lock that hasn't been aquired ".
< 						"yet(could be a false positive) - Line:$linenr\n\n";
---
> 					print "Unlocking a lock that hasn't been acquired ".
> 						"yet (could be a false positive) - Line:$linenr\n\n";
287,288c287,288
< 					print "Unlocking a lock that hasn't been aquired ".
< 						"yet(could be a false positive) - Line:$linenr\n\n";
---
> 					print "Unlocking a lock that hasn't been acquired ".
> 						"yet (could be a false positive) - Line:$linenr\n\n";
302c302
< 						"yet(could be a false positive) - Line:$linenr\n\n";
---
> 						"yet (could be a false positive) - Line:$linenr\n\n";
347c347,351
< 				print "Using foo[] it's recommended over *foo: saves memory references & code\n";
---
> 				print "Using foo[] is recommended over *foo: saves memory references & code\n";
> 				print "$filename:$linenr:$line\n\n";
> 			}
> 			if ($line=~/memset\((.*),( *sizeof.*),( *\d+)\)/) {
> 				print "Your memset parameters appears to be backwards\n";
359c363
< 			print "If you're building a module, be concious that you must use the". 
---
> 			print "If you're building a module, note that you must use the". 

[-- Attachment #3: Type: text/plain, Size: 167 bytes --]

_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.osdl.org
http://lists.osdl.org/mailman/listinfo/kernel-janitors