https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80635
--- Comment #35 from Manuel López-Ibáñez <manu at gcc dot gnu.org> --- In any case, looking at the uninit dump with -fdump-tree-all-all-lineno it appears that GCC knows the block where the warning is triggered is never executed: ;; basic block 13, loop depth 0, count 0 (precise), probably never executed ;; prev block 12, next block 14, flags: (NEW, REACHABLE, VISITED) ;; pred: 12 [never] count:0 (precise) (EH,EXECUTABLE) ;; starting at line -1 <L6>: [LP 2] # DEBUG c1$16$iD.2602 => c1$16$i_9 # DEBUG c1D.2601 => 1 # DEBUG thisD.2599 => &c1D.2412 [./example.cpp:20:23] # DEBUG D#2ptD.0 => [./example.cpp:20:23] &[./example.cpp:20:18] [./example.cpp:20:18] c1D.2412.D.2424.tD.2426 [./example.cpp:20:23] # DEBUG thisD.2600 => D#2ptD.0 [./example.cpp:9:11] # DEBUG BEGIN_STMT [./example.cpp:9:13] # .MEM_27 = VDEF <.MEM_15> xD.2332 = c1$16$i_9; [./example.cpp:9:18] goto <bb 15>; [0.00%] ;; succ: 15 [never] count:0 (precise) (FALLTHRU,EXECUTABLE) I remember seeing other bugs where the block info could be used to avoid warning.
