Messages in this thread |  | | | Subject | Re: [patch] x86: phase out forced inlining | | From | Harvey Harrison <> | | Date | Tue, 04 Mar 2008 10:14:13 -0800 |
| |
On Tue, 2008-03-04 at 10:09 -0800, H. Peter Anvin wrote: > Harvey Harrison wrote: > > > >> __inline__ : 1380 > > > > Lots of them in include/asm-*...not sure if there is a reason for this. > > > > Preferred form for code that's exported to userspace (since gcc > complains with -ansi -pedantic otherwise.) >
Figured it would be something like that. Would it be reasonable to move towards eliminating __inline?
Also, since the exported headers already go through unifdef, could we move to using inline everywhere in the kernel and add a processing step to make it __inline__ in the exported headers?
Harvey
|  |