https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122955
--- Comment #10 from Dhruv Chawla <dhruvc at gcc dot gnu.org> --- I was able to get a longer stack trace, and FWIW it seems like the cycle is starting from gt_ggc_mx_ipa_return_value_summary: #890333 0x0000000000fd84a4 in gt_ggc_mx_lang_tree_node (x_p=0xfffff62c52a0) at ./gtype-lto.h:376 #890334 0x0000000000fd8390 in gt_ggc_mx_lang_tree_node (x_p=0xfffff62c5690) at ./gtype-lto.h:359 #890335 0x0000000000fd75f4 in gt_ggc_mx_lang_tree_node (x_p=0xfffff62cc000) at ./gtype-lto.h:203 #890336 0x0000000000fd854c in gt_ggc_mx_lang_tree_node (x_p=0xfffff62c5738) at ./gtype-lto.h:382 #890337 0x0000000000fd75f4 in gt_ggc_mx_lang_tree_node (x_p=0xfffff62d8780) at ./gtype-lto.h:203 #890338 0x0000000000fd854c in gt_ggc_mx_lang_tree_node (x_p=0xfffff62d52a0) at ./gtype-lto.h:382 #890339 0x0000000000fd8390 in gt_ggc_mx_lang_tree_node (x_p=0xfffff62d57e0) at ./gtype-lto.h:359 #890340 0x00000000014c0054 in gt_ggc_mx_ipa_vr (x_p=0xffffdcf1acd0) at gtype-desc.cc:2958 #890341 0x00000000015bf8fc in gt_ggc_mx_ipa_return_value_summary (x_p=0xffffeee9fa38) at ./gt-ipa-prop.h:44 #890342 0x00000000015bf9e4 in gt_ggc_mx (x=@0xffffd7600cb8: 0xffffeee9fa38) at ./gt-ipa-prop.h:62 #890343 0x00000000015cbca0 in hash_map<int_hash<int, 0, -1>, ipa_return_value_summary*, simple_hashmap_traits<default_hash_traits<int_hash<int, 0, -1> >, ipa_return_value_summary*> >::hash_entry::ggc_mx (e=...) at ../../src/gcc/hash-map.h:76 #890344 0x00000000015cb394 in hash_map<int_hash<int, 0, -1>, ipa_return_value_summary*, simple_hashmap_traits<default_hash_traits<int_hash<int, 0, -1> >, ipa_return_value_summary*> >::hash_entry::ggc_maybe_mx (e=...) at ../../src/gcc/hash-map.h:82 #890345 0x00000000015c9c68 in gt_ggc_mx<hash_map<int_hash<int, 0, -1>, ipa_return_value_summary*, simple_hashmap_traits<default_hash_traits<int_hash<int, 0, -1> >, ipa_return_value_summary*> >::hash_entry> (h=0xffffde198c80) at ../../src/gcc/hash-table.h:1255 #890346 0x00000000015c67a0 in gt_ggc_mx<int_hash<int, 0, -1>, ipa_return_value_summary*, simple_hashmap_traits<default_hash_traits<int_hash<int, 0, -1> >, ipa_return_value_summary*> > (h=0xffffde198c80) at ../../src/gcc/hash-map.h:321 #890347 0x00000000015c2e1c in gt_ggc_mx<ipa_return_value_summary> (summary=@0xffffffffede0: 0xffffde198bb0) at ../../src/gcc/symbol-summary.h:322 #890348 0x00000000015bf98c in gt_ggc_mx_function_summary_ipa_return_value_summary__ (x_p=0xffffde198bb0) at ./gt-ipa-prop.h:54 #890349 0x000000000138bb24 in ggc_mark_root_tab (rt=0x3b6c830 <gt_ggc_r_gt_ipa_prop_h>) at ../../src/gcc/ggc-common.cc:75 #890350 0x000000000138bcb0 in ggc_mark_roots () at ../../src/gcc/ggc-common.cc:104 #890351 0x000000000103c364 in ggc_collect (mode=GGC_COLLECT_HEURISTIC) at ../../src/gcc/ggc-page.cc:2332 #890352 0x00000000017c74fc in execute_one_pass (pass=0x4abe3f0) at ../../src/gcc/passes.cc:2740 #890353 0x00000000017c84c8 in execute_ipa_pass_list (pass=0x4abe3f0) at ../../src/gcc/passes.cc:3101 #890354 0x0000000000fe1774 in do_whole_program_analysis () at ../../src/gcc/lto/lto.cc:528 #890355 0x0000000000fe1bd4 in lto_main () at ../../src/gcc/lto/lto.cc:675 #890356 0x000000000197e2e0 in compile_file () at ../../src/gcc/toplev.cc:455 #890357 0x0000000001982a30 in do_compile () at ../../src/gcc/toplev.cc:2224 #890358 0x0000000001983004 in toplev::main (this=0xfffffffff0b0, argc=769, argv=0x4a75070) at ../../src/gcc/toplev.cc:2389 #890359 0x0000000003886cc4 in main (argc=31, argv=0xfffffffff258) at ../../src/gcc/main.cc:39 But given that this is a stack overflow I don't know how trustworthy the backtrace is.
