What I'm trying to explain is that kernel style is minimalist. Focus on writing the simplest, most readable code you can. Kernel coders are 10 times more likely to use "int i" than they are to use "unsigned int i". $ git grep " int i;" | wc -l 13956 $ git grep " unsigned int i;" | wc -l 1508 "int" is the default number in C and "unsigned int" has a modifier and so it's a tiny bit special. When you use "unsigned int" you're drawing attention to your variable because it is not just a normal number. Someone reviewing that code immediately thinks, "How high is NVEC_POOL_SIZE exactly?" Certainly that was my thought when I saw it declared as a size_t. But in fact the for loop is a boring ordinary for loop. You've tricked the reviewers, and you've distracted them from the 5 second sleep elsewhere in your code. Obviously, I'm exaggerating a bit and in this case it doesn't matter one way or the other. Go ahead and declare it as unsigned so long as you understand about kernel style. Don't be clever, fancy or special. Don't optimize. Don't stand out. And don't do more work than you have to. regards, dan carpenter _______________________________________________ devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxx http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
