[Bug middle-end/36124] conditional loop becomes infinite loop in -O2 (gcc 4.2 -> 4.3 regression)

--- Comment #11 from scottwood at freescale dot com 2008-05-13 21:13 --- (In reply to comment #4) > If your code invokes undefined behavior, how is gcc going > to read your mind? If GCC can tell that it is undefined behavior, then warning the user is more useful than silently producing

[Bug middle-end/36124] conditional loop becomes infinite loop in -O2 (gcc 4.2 -> 4.3 regression)

--- Comment #10 from pinskia at gcc dot gnu dot org 2008-05-13 20:02 --- *** Bug 36232 has been marked as a duplicate of this bug. *** -- pinskia at gcc dot gnu dot org changed: What|Removed |Added -

[Bug middle-end/36124] conditional loop becomes infinite loop in -O2 (gcc 4.2 -> 4.3 regression)

--- Comment #9 from rguenth at gcc dot gnu dot org 2008-05-07 19:42 --- decrementing a NULL pointer invokes undefined behavior, incrementing not. -- rguenth at gcc dot gnu dot org changed: What|Removed |Added ---

[Bug middle-end/36124] conditional loop becomes infinite loop in -O2 (gcc 4.2 -> 4.3 regression)

--- Comment #8 from cyprien+gccbug at cypou dot net 2008-05-04 20:13 --- On some embedded machines, the SDRAM lays on 0x address. So it is not so meaningless to increment or decrement from/to NULL pointer. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36124

[Bug middle-end/36124] conditional loop becomes infinite loop in -O2 (gcc 4.2 -> 4.3 regression)

--- Comment #7 from cyprien+gccbug at cypou dot net 2008-05-04 17:31 --- it's right, using --foo unstead of foo-- gives a better result. But it does not make me happy (about the possibility of a bug) -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36124

[Bug middle-end/36124] conditional loop becomes infinite loop in -O2 (gcc 4.2 -> 4.3 regression)

--- Comment #6 from pinskia at gcc dot gnu dot org 2008-05-04 17:09 --- >Now, this code should not rely on undefined behaviour: It does because incrementing or decrementing to a NULL Pointer is undefined. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36124

[Bug middle-end/36124] conditional loop becomes infinite loop in -O2 (gcc 4.2 -> 4.3 regression)

--- Comment #5 from cyprien+gccbug at cypou dot net 2008-05-04 17:03 --- Now, this code should not rely on undefined behaviour: extern void func(int,void*); void test() { register long *foo = (long*) (4*sizeof(*foo)) - 1; register int index; for(index=0;index<4;index++)

[Bug middle-end/36124] conditional loop becomes infinite loop in -O2 (gcc 4.2 -> 4.3 regression)

--- Comment #4 from kargl at gcc dot gnu dot org 2008-05-04 17:01 --- (In reply to comment #2) > shouldn't gcc report a warning in this case ? > > because silently entering an infinite loop is not very kind... > If your code invokes undefined behavior, how is gcc going to read your mi

[Bug middle-end/36124] conditional loop becomes infinite loop in -O2 (gcc 4.2 -> 4.3 regression)

--- Comment #3 from pinskia at gcc dot gnu dot org 2008-05-04 16:45 --- (In reply to comment #2) > shouldn't gcc report a warning in this case ? Maybe, but really depending on undefined behavior is bad too. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36124

[Bug middle-end/36124] conditional loop becomes infinite loop in -O2 (gcc 4.2 -> 4.3 regression)

--- Comment #2 from cyprien+gccbug at cypou dot net 2008-05-04 16:41 --- shouldn't gcc report a warning in this case ? because silently entering an infinite loop is not very kind... -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36124

[Bug middle-end/36124] conditional loop becomes infinite loop in -O2 (gcc 4.2 -> 4.3 regression)

--- Comment #1 from pinskia at gcc dot gnu dot org 2008-05-04 16:34 --- Pointers types overflow is undefined which is what you are seeing. -- pinskia at gcc dot gnu dot org changed: What|Removed |Added --