> rm -f conftest.err[]m4_ifval([$1], [ conftest.$ac_ext])
> ])# _AC_PREPROC_IFELSE
> 
> Likewise for the other uses.

Ok.

> AC_COMPILE_IFELSE([feature a],
>   [AC_COMPILE_IFELSE([feature b], [], [])])
> 
> Is there ever a chance that files left behind by compiling feature a will 
> interfere with correct detection of whether feature b is supported?

I think so, indeed I preferred the previous way it's done.  But I guess
we have to live with backwards compatibility.

Thinking more about it, I suppose this is the reason why
_AC_COMPILE_IFELSE_BODY does *at the beginning*

  rm -f conftest.$ac_objext

And this could be replaced by rm -f $ac_cleanfiles.  (And a macro
AC_CLEAN_OUTPUT could add to ac_cleanfiles, ensuring that the file is
deleted before the next test).  But maybe it's overkill, and this patch
correctly preserves <=2.63 behavior.

Thanks for the review.

Paolo


Reply via email to