On Mon, 21 Mar 2005 10:10:24 +0100 (CET), "Jean Delvare" <khali at linux-fr.org> wrote: > >Hi Grant, > >The macro *is* in fact used, although the (ab)use of macros all over the >place make it hard to see it the first place (grep can't help here). Oops. Sorry. > >Look at these two lines: > >set_bits(DIV, fan_div1, fan_div[0], GL518_REG_MISC, 0xc0, 6); >set_bits(DIV, fan_div2, fan_div[1], GL518_REG_MISC, 0x30, 4); > >Climb up the preprocessing chain, and you'll see that the DIV_TO_REG >macro is actually needed. You could have noticed by simply >compile-testing your patch. I invite you to always do. Which I forgot to do 'cos no see match. :( Are there tools for this, I started writing something to get at preprocessor output but it seems to want a perl script to parse output of gcc -E. Hmm, so if I'm not sure of something, comment it out and ask the compiler, get a line number or string to look for? >You just hit one of the various reasons why I prefer inline functions to >macros when both are possible. At least the compiler keeps track of >function users and will let you know the unused ones. Thanks for feedback. Perhaps I'll just go shooting macros to make the code a bit more obvious? Cheers, Grant.
