[Bug rtl-optimization/91860] [10 Regression] ICE: in decompose, at rtl.h:2279 with -Og -fipa-cp -g --param=max-combine-insns=3

[wilson at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91860

--- Comment #10 from Jim Wilson <wilson at gcc dot gnu.org> ---
Author: wilson
Date: Fri Oct 11 18:41:35 2019
New Revision: 276901

URL: https://gcc.gnu.org/viewcvs?rev=276901&root=gcc&view=rev
Log:
Extend subst to simplify CONST_INT inside SIGN_EXTEND.

This addresses PR 91860 which has four testcases triggering internal errors.
The problem here is that in combine when handling debug insns, we are trying
to substitute
(sign_extend:DI (const_int 8160 [0x1fe0]))
as the value for
(reg:DI 78 [ _9 ])
in the debug insn
(debug_insn 29 28 30 2 (var_location:QI d (subreg:QI (reg:DI 78 [ _9 ]) 0))
"tmp4.c":11:5 -1
     (nil))
This eventually triggers an abort because 8160 is not a sign-extended
QImode value.

In subst there is already code check for a CONST_INT inside a ZERO_EXTEND and
simplify it.  This needs to be extended to also handle a SIGN_EXTEND the same
way.

        gcc/
        PR rtl-optimization/91860
        * combine.c (subst): If new_rtx is a constant, also check for
        SIGN_EXTEND when deciding whether to call simplify_unary_operation.

        gcc/testsuite/
        PR rtl-optimization/91860
        * gcc.dg/pr91860-1.c: New testcase.
        * gcc.dg/pr91860-2.c: New testcase.
        * gcc.dg/pr91860-3.c: New testcase.
        * gcc.dg/pr91860-4.c: New testcase.

Added:
    trunk/gcc/testsuite/gcc.dg/pr91860-1.c
    trunk/gcc/testsuite/gcc.dg/pr91860-2.c
    trunk/gcc/testsuite/gcc.dg/pr91860-3.c
    trunk/gcc/testsuite/gcc.dg/pr91860-4.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/combine.c
    trunk/gcc/testsuite/ChangeLog