--- Comment #5 from rguenth at gcc dot gnu dot org 2009-07-29 18:12 ---
long lo_b_2268X;
long lo_a_2267X;
long lo_c_2271X;
...
lo_a_2267X = 65535 & a_1962X;
lo_b_2268X = 65535 & b_2266X;
...
lo_c_2271X = ((lo_a_2267X) * (lo_b_2268X));
...
if ((536870911 < lo_c_2271X)) {
...
--- Comment #4 from sperber at deinprogramm dot de 2009-07-29 17:42 ---
(From update of attachment 18271)
In the preprocessed file, the test that gets elided is in line 20307.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40907
--- Comment #3 from pinskia at gcc dot gnu dot org 2009-07-29 17:41 ---
>an overflow check in the internal arithmetic gets elided.
Does -Wno-strict-overflow fix the issue? Does -fwrapv fix the issue?
If so you are depending on overflow being defined as wrapping and the code is
undefin
--- Comment #2 from pinskia at gcc dot gnu dot org 2009-07-29 17:40 ---
../s48/c/scheme48vm-32.c: In function 'current_code_vector':
../s48/c/scheme48vm-32.c:1426:8: warning: 'v_231X' may be used uninitialized in
this function
../s48/c/scheme48vm-32.c: In function 'HtopD12136':
../s48/c/
--- Comment #1 from sperber at deinprogramm dot de 2009-07-29 17:37 ---
Created an attachment (id=18271)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=18271&action=view)
preprocessed source file
gcc -v output:
kramer[420] gcc -v -save-temps -c -DHAVE_CONFIG_H
-D__COMPILING_SCHEM
