On reverse-next Core was generated by `gdb --args /home/ak/hle/obj-perf/perf report --branch-history --stdio'. Program terminated with signal SIGSEGV, Segmentation fault. #0 get_frame_type (frame=0x0) at frame.c:2458 (gdb) p frame $1 = (struct frame_info *) 0x0 (gdb) l 2453 } 2454 2455 enum frame_type 2456 get_frame_type (struct frame_info *frame) 2457 { 2458 if (frame->unwind == NULL) 2459 /* Initialize the frame's unwinder because that's what 2460 provides the frame's type. */ 2461 frame_unwind_find_by_frame (frame, &frame->prologue_cache); 2462 return frame->unwind->type; (gdb) bt #0 get_frame_type (frame=0x0) at frame.c:2458 #1 skip_artificial_frames (frame=0x0) at frame.c:434 #2 0x00000000006b12f1 in frame_unwind_caller_id (next_frame=<optimized out>) at frame.c:501 #3 0x00000000005c5676 in process_event_stop_test (ecs=ecs@entry=0x7ffc8bdca820) at infrun.c:6620 #4 0x00000000005c6d16 in handle_inferior_event_1 (ecs=0x7ffc8bdca820) at infrun.c:4939 #5 handle_inferior_event (ecs=ecs@entry=0x7ffc8bdca820) at infrun.c:5362 #6 0x00000000005c9043 in fetch_inferior_event (client_data=<optimized out>) at infrun.c:3922 #7 0x00000000005e0fc8 in inferior_event_handler (event_type=<optimized out>, client_data=<optimized out>) at inf-loop.c:44 #8 0x00000000005df3cd in gdb_wait_for_event (block=block@entry=0) at event-loop.c:834 #9 0x00000000005df5e0 in gdb_do_one_event () at event-loop.c:298 #10 0x00000000005df68e in start_event_loop () at event-loop.c:347 #11 0x00000000005d92e3 in captured_command_loop (data=data@entry=0x0) at main.c:318
This was fixed on master by 33b4777ca1b7b456af8201b98eda27d1b272cbab. I'm backporting the fix to the 7.11 branch.
The gdb-7.11-branch branch has been updated by Markus Metzger <mmetzger@sourceware.org>: https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=2ef34d11f61d79dcb152713aa059051d8cd3295d commit 2ef34d11f61d79dcb152713aa059051d8cd3295d Author: Markus Metzger <markus.t.metzger@intel.com> Date: Fri Feb 5 09:32:53 2016 +0100 btrace: fix PR gdb/19829 This is a backport of 33b4777ca1b7 btrace, frame: fix crash in get_frame_type a038fa3e14a4 stack: check frame_unwind_caller_id 2f3ef606b912 frame: add skip_tailcall_frames In skip_artificial_frames we repeatedly call get_prev_frame_always until we get a non-inline and non-tailcall frame assuming that there must be such a frame eventually. For record targets, however, we may have a frame chain that consists only of artificial frames. This leads to a crash in get_frame_type when dereferencing a NULL frame pointer. Change skip_artificial_frames and skip_tailcall_frames to return NULL in such a case and modify each caller to cope with a NULL return. In frame_unwind_caller_pc and frame_unwind_caller_arch, we simply assert that the returned value is not NULL. Their caller was supposed to check frame_unwind_caller_id before calling those functions. In other cases, we thrown an error. In infcmd further move the skip_tailcall_frames call to the forward-stepping case since we don't need a frame for reverse execution and we don't want to fail because of that. Reverse-finish does make sense for a tailcall frame. gdb/ * frame.h (skip_tailcall_frames): New. * infcmd.c (finish_command): Call skip_tailcall_frames. * frame.c (skip_artificial_frames): Return NULL if only artificial frames are found. Update comment. (frame_pop): Call skip_tailcall_frames. (frame_unwind_caller_id): Handle NULL return. (frame_unwind_caller_pc, frame_unwind_caller_arch): Assert that skip_artificial_frames does not return NULL. (frame_pop): Add an error if only tailcall frames are found. * infcmd.c (finish_command): Move skip_tailcall_frames call into forward- execution case. Add an error if only tailcall frames are found. * stack.c (frame_info): Check frame_unwind_caller_id. testsuite/ * gdb.btrace/tailcall-only.exp: New. * gdb.btrace/tailcall-only.c: New. * gdb.btrace/x86_64-tailcall-only.S: New. * gdb.btrace/i686-tailcall-only.S: New.
Fixed by 2ef34d11f61d btrace: fix PR gdb/19829.