Messages in this thread | | | Date | Wed, 25 Feb 2009 16:57:15 -0800 | From | "H. Peter Anvin" <> | Subject | Re: [PATCH 14/15] bitops: Change the bitmap index from int to unsigned long [frv] |
| |
David Howells wrote: > >> A casual grep finds no less than 3524 instances of "static inline <type>" -- >> presumably with a function name following -- and only 447 instances of >> "static inline" without a type in the kernel. > > But how does it break down between "static inline type\nfunction_name" and > "static inline\ntype function_name"? That's more to the point. >
I believe that is the breakdown is roughly what you see above, i.e. over 8:1; the pattern I used was looking for "^static inline[^;(]*$", and a visual examination of the results shows that even if my line count is slighly off the lopsidedness is still dramatic.
-hpa
| |