[Bug middle-end/126775] [17 Regression] ICE: tree check: expected real_cst, have vector_cst in generic_simplify_MINUS_EXPR, at generic-match-6.cc:7470 with -ffinite-math-only since r17-3158

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=126775

--- Comment #4 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Roger Sayle <[email protected]>:

https://gcc.gnu.org/g:579349b0f12681ffcbfd14eafe6efdc8da67a1f2

commit r17-3466-g579349b0f12681ffcbfd14eafe6efdc8da67a1f2
Author: Roger Sayle <[email protected]>
Date:   Thu Aug 20 09:31:48 2026 +0100

    PR middle-end/126775: ICE optimizing (T)0.0 - x with -ffinite-math-only

    This patch fixes PR middle-end/126775, an ICE caused by my recent change
    to match.pd around tweaking the conditions under which 0.0 - x can
    safely be transformed into -x.  Unfortunately, that change assumed
    that real_zerop@0 implies that TREE_CODE(@0) == REAL_CST.  Alas things
    aren't that simple, so this fix introduces a new real_negzerop predicate
    that in addition to REAL_CST also handles VECTOR_CST and COMPLEX_CST,
    and most importantly fails gracefully on TREE_CODEs that it isn't
    expecting.

    2026-08-20  Roger Sayle  <[email protected]>

    gcc/ChangeLog
            PR middle-end/126775
            * match.pd (0.0 - x -> -x): Use new real_negzerop function.
            * tree.cc (real_negzerop): New predicate function to test if
            a tree expression is -0.0 or equivalent (like real_zerop).
            * tree.h (real_negzerop): Prototype here.

    gcc/testsuite/ChangeLog
            PR middle-end/126775
            * gcc.dg/pr126775.c: New test case.