------- Comment #6 from kurt at garloff dot de  2009-02-17 16:09 -------
Some more info:
HEAD seems to be much faster here (10m30s).
The author uses -fno-tree-pre in gmic (which has a lot of the same code),
not sure whether it helps significantly.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39210

Reply via email to