[Bug middle-end/44081] Incorrect nonnull assumed in code generation

[jakub at gcc dot gnu dot org]

------- Comment #7 from jakub at gcc dot gnu dot org  2010-05-12 06:25 -------
I can't reproduce the problem either, neither in 4.6 nor in 4.4.  4.5.0 (rather
than current 4.5 snapshots) had a bug with nonnull attribute if a function is
IPA optimized and some arguments are optimized out, but I don't think that's
the case here.

The nonnull arguments are 1 based, sorry for not reading the testcase carefully
when replying in #c3.

memcpy etc. with NULL argument is invalid, the standard say that those
functions
copy from object pointed by the argument (or to object pointed by the
argument).
NULL doesn't point to any object (the standard say that it compares unequal to
pointer to any object).


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44081