https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114206
--- Comment #9 from Andrew Pinski <pinskia at gcc dot gnu.org> --- Created attachment 60885 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=60885&action=edit slightly more complex testcase Still fails with `-Os -fno-tree-ccp -fno-tree-copy-prop -fno-tree-forwprop -fno-tree-fre -fno-tree-vrp` . But it shows that simply looking at arguments to see if what passes is &b will also fail. I wonder if the life range analysis that is in cfgexpand should be shared here.
