Re: [Sbcl-devel] SBCL compiler barfs on CFFI declarations

[Andy H. <ah...@gm...>]

I observed the same error using SBCL's ffi (noted at the end of my
June 5 mail to sbcl-devel), which I reduced down to the following:

(defun unused-alien ()
 (SB-ALIEN:WITH-ALIEN ((ALIEN1923 (ARRAY (SB-ALIEN:UNSIGNED 8) 72)))
   (values)))

..with the observation that adding code that uses the 'array' variable
will cause it to compile, and that non-array types don't seem to
trigger it. This is not necessarily the same circumstance as in the
code you attached (which, frankly, looks odd - passing a stat pointer
as a boolean flag?), unless perhaps if the compiler optimizes away the
lone occurance of 'follow-symlink' inside COND based on its type,
which would make it again the case of an unused alien array.