https://gcc.gnu.org/bugzilla/show_bug.cgi?id=73650
--- Comment #7 from Segher Boessenkool ---
[ i have a patch, it is bootstrapping+checking ]
[ the revision in #c4 is a red herring ]
Yeah, single_set holds for that, and LRA then treats it as a reg-reg
move (because that is what the single set
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=73650
--- Comment #6 from Alan Modra ---
It's a bit interesting that the problem insn (from ira dump):
(insn 70 66 76 3 (parallel [
(set (reg:CC 198)
(compare:CC (reg/v:SI 188 [ n ])
(const_int 0 [0])))
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=73650
Alan Modra changed:
What|Removed |Added
Attachment #39160|0 |1
is obsolete|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=73650
--- Comment #4 from Segher Boessenkool ---
First bad revision is r237376. Fails with -m32 -mcpu=e5500 -mlra.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=73650
--- Comment #3 from Segher Boessenkool ---
it fails with -misel.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=73650
--- Comment #2 from Segher Boessenkool ---
It fails with -mtune={e55,e500mc,e5500,e6500} and works without it.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=73650
Segher Boessenkool changed:
What|Removed |Added
Status|UNCONFIRMED |NEW
Last reconfirmed|
