On Wed, 1 Nov 2006, Daniel Jacobowitz wrote: > On Wed, Nov 01, 2006 at 06:59:42PM +0000, Joseph S. Myers wrote: > > I'd rather simply fix glibc to work with C99 inline semantics. For the > > above you might use > > > > #define tolower(c) __tolower_inline(c) > > static __inline __attribute__((__always_inline__)) int tolower_inline ... > > > > and #undef tolower when defining the out of line function. That would > > apply to all cases where a user-visible function in a header is currently > > inlined directly; for a purely internal function in the header, we can > > arrange for all declarations to use "inline" without "extern" when in C99 > > mode. > > Isn't the whole point that the current extern inline isn't > __always_inline__, but leaves it to the compiler's judgement?
In most cases where glibc isn't using always_inline that's a mistake because the glibc code predates always_inline. (For example, the inlines in sys/stat.h clearly should be always_inline.) If the extern inline function shouldn't be always_inline and only exists as an inline, glibc already arranges code to export the inline versions from the header into the library. If it shouldn't be always_inline and has an exported version, then the __foo_inline could be made an alias for the real function. -- Joseph S. Myers [EMAIL PROTECTED]
