--- Comment #9 from pinskia at gcc dot gnu dot org 2005-11-04 15:25 ---
Closed as invalid.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Stat
--- Comment #8 from pinskia at gcc dot gnu dot org 2005-11-04 15:24 ---
Woops, that should have been ...
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #7 from pinskia at gcc dot gnu dot org 2005-11-04 15:24 ---
Woops, that should have been ...
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #6 from pinskia at gcc dot gnu dot org 2005-11-04 14:05 ---
Removing volatile will change the semantics of the program and therefor not a
bug. GCC docs mention with (I cannot find right now as I am at a Kiosk and it
is hard to find anything) longjmp, volatile is required.
--- Comment #5 from worm at arrakis dot es 2005-11-04 13:35 ---
Sorry for my stupidity -- the jet-lag after a long trip
seems to affect me --.
Remove the "volatile" statement from the arguments list
of L1foo.
Then the assignments to the variable V1 are moved after
the call to cl_throw()
--- Comment #4 from pinskia at gcc dot gnu dot org 2005-11-04 13:25 ---
I still cannot reproduce it.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=24635
--- Comment #3 from worm at arrakis dot es 2005-11-04 12:13 ---
Created an attachment (id=10143)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=10143&action=view)
This one triggers the bug
The new file definitely triggers the problem with statement reordering if
compiled with (-O2)
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
CC||pinskia at gcc dot gnu dot
|
