Re: [Sbcl-devel] [Sbcl-commits] CVS: sbcl/tests compiler-2.impure-cload.lisp, NONE, 1.1

[Nikodemus S. <nik...@ra...>]

Nikodemus Siivola <nik...@ra...> writes:

> Christophe Rhodes <cs...@ca...> writes:
>
>> Nikodemus Siivola <de...@us...> writes:
>
>>> (assert (equal (test) '(0 0 0)))
>>> (assert (equal (test) '(13 13 13))) ; sanity check
>>
>> I'm not entirely sure what this is testing, I'm afraid, but it looks
>> quite like undefined behaviour to me, given that it's processed by
>
> You are referring to the second case? Looking at it now, I don't
> understand anynmore why I thought that was necessary.

Took a few minutes for memory to page. The second assert checks that
the fdefinitions have actually been altered. If they weren't
for some reason, the first case could work even in the even of
a regression.

>> issue right now, but consider putting this test into a file that isn't
>> file-compiled.
>
> Point.

Will do.

Cheers,

  -- Nikodemus              Schemer: "Buddha is small, clean, and serious."
                   Lispnik: "Buddha is big, has hairy armpits, and laughs."