[Bug tree-optimization/126865] [17 Regression] ICE at -O2 -march=rv64gcv on riscv64 during GIMPLE pass: slp
"cvs-commit at gcc dot gnu.org via Gcc-bugs" <[email protected]>
| Newsgroups | gmane.comp.gcc.bugs |
|---|---|
| Message-ID | <[email protected]/bugzilla/> |
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=126865 --- Comment #2 from GCC Commits <cvs-commit at gcc dot gnu.org> --- The master branch has been updated by Richard Biener <[email protected]>: https://gcc.gnu.org/g:c8545a3143330983e51d913f0a01187258b14510 commit r17-3287-gc8545a3143330983e51d913f0a01187258b14510 Author: Richard Biener <[email protected]> Date: Fri Aug 14 12:16:41 2026 +0200 tree-optimization/126865 - SLP node with trapping stmt in wrong BB The following fixes a SLP discovery issue where we fail to properly ensure that when there's a trapping stmt all stmts of a node belong to the same BB. PR tree-optimization/126865 * tree-vect-slp.cc (vect_build_slp_tree_1): Track common BB and whether there's any trapping stmt in the matching set and verify the same BB constraint after we've seen all stmts. * gcc.dg/vect/bb-slp-pr126865.c: New testcase. * gcc.dg/vect/bb-slp-pr100778-1.c: Adjust.