On 1/4/24 02:34, Jakub Jelinek wrote:
Hi!
The following testcase ICEs when rslt is SSA_NAME_OCCURS_IN_ABNORMAL_PHI
and we call replace_uses_by with a INTEGER_CST def, where it ICEs on:
if (e->flags & EDGE_ABNORMAL
&& !SSA_NAME_OCCURS_IN_ABNORMAL_PHI (val))
because val is not an SSA_NAME. One way would be to add
&& TREE_CODE (val) == SSA_NAME
check in between the above 2 lines in replace_uses_by.
And/or the following patch just punts propagating constants to
SSA_NAME_OCCURS_IN_ABNORMAL_PHI rslt uses.
Or we could punt somewhere earlier in final value replacement (but dunno
where).
Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk?
Or do you want the replace_uses_by change (or both that and the following,
something else)?
2024-01-04 Jakub Jelinek <ja...@redhat.com>
PR tree-optimization/113201
* tree-scalar-evolution.cc (final_value_replacement_loop): Don't call
replace_uses_by on SSA_NAME_OCCURS_IN_ABNORMAL_PHI rslt.
* gcc.c-torture/compile/pr113201.c: New test.
I think we have to punt somewhere -- we can't insert the constant
initialization on an abnormal edge. So OK for trunk.
jeff
