[Bug target/91453] ICE in elimination_costs_in_insn at reload1.c:3547

[ubizjak at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91453

Uroš Bizjak  changed:

   What|Removed |Added

 Status|UNCONFIRMED |RESOLVED
 Resolution|--- |DUPLICATE

--- Comment #3 from Uroš Bizjak  ---
Fixed by r274535.

*** This bug has been marked as a duplicate of bug 91454 ***

[Bug target/91453] ICE in elimination_costs_in_insn at reload1.c:3547

[ubizjak at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91453

--- Comment #2 from Uroš Bizjak  ---
Latest SVN creates the following RTX in _.ira dump:

(insn 144 12 145 2 (set (subreg:V2DI (reg:DI 161) 0)
(vec_merge:V2DI (vec_duplicate:V2DI (reg:DI 111 [ ubound.0 ]))
(const_vector:V2DI [
(const_int 0 [0]) repeated x2
])
(const_int 1 [0x1]))) "deferred_character_23.f90":27:0 -1
 (nil))

This is suspiciously close to PR91454. "-1" hints at unknown insn.

[Bug target/91453] ICE in elimination_costs_in_insn at reload1.c:3547

[ubizjak at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91453

--- Comment #1 from Uroš Bizjak  ---
Are you sure? It compiles OK with gcc-9 for me.