https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98950
Bug ID: 98950 Summary: jump threading memory leak Product: gcc Version: 11.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: tree-optimization Assignee: unassigned at gcc dot gnu.org Reporter: rguenth at gcc dot gnu.org Target Milestone: --- Tried to find my way around the code but I'm not too familiar with it. This shows up (at least) when building 521.wrf_r with -Ofast -flto ==19644== 832 bytes in 52 blocks are definitely lost in loss record 7,748 of 9,681 ==19644== at 0x4C2E94F: operator new(unsigned long) (in /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so) ==19644== by 0xC750AE: thread_jumps::convert_and_register_current_path(edge_def*) (tree-ssa-threadbackward.c:472) ==19644== by 0xC75452: thread_jumps::register_jump_thread_path_if_profitable(tree_node*, tree_node*, basic_block_def*) [clone .part.37] (tree-ssa-threadbackward.c:561) ==19644== by 0xC76487: register_jump_thread_path_if_profitable (tree-ssa-threadbackward.c:553) ==19644== by 0xC76487: thread_jumps::handle_phi(gphi*, tree_node*, basic_block_def*) (tree-ssa-threadbackward.c:600) ==19644== by 0xC763B4: thread_jumps::fsm_find_control_statement_thread_paths(tree_node*) (tree-ssa-threadbackward.c:730) ==19644== by 0xC764D4: thread_jumps::handle_phi(gphi*, tree_node*, basic_block_def*) (tree-ssa-threadbackward.c:594) ==19644== by 0xC763B4: thread_jumps::fsm_find_control_statement_thread_paths(tree_node*) (tree-ssa-threadbackward.c:730) ==19644== by 0xC76B8E: (anonymous namespace)::pass_thread_jumps::execute(function*) (tree-ssa-threadbackward.c:828) ==19644== by 0x9E58A3: execute_one_pass(opt_pass*) (passes.c:2567) ==19644== by 0x9E6100: execute_pass_list_1(opt_pass*) (passes.c:2656) ==19644== by 0x9E6112: execute_pass_list_1(opt_pass*) (passes.c:2657) ==19644== by 0x9E6154: execute_pass_list(function*, opt_pass*) (passes.c:2667)