[ Spinoff from PR29721. ] Consider the following test-case: ... $ cat test.c void foo (const char *s) { } int main (void) { foo ("foo"); return 0; } ... Compiled without debug info: ... $ gcc test.c ... Disassembly of main: ... 0000000000400578 <main>: 400578: a9bf7bfd stp x29, x30, [sp, #-16]! 40057c: 910003fd mov x29, sp 400580: 90000000 adrp x0, 400000 <_init-0x408> 400584: 91192000 add x0, x0, #0x648 400588: 97fffff7 bl 400564 <foo> 40058c: 52800000 mov w0, #0x0 // #0 400590: a8c17bfd ldp x29, x30, [sp], #16 400594: d65f03c0 ret ... Now, let's do "info frame" at each insn in main (so, using ni to skip over foo), as well as printing $sp and $fp: ... $ gdb a.out Reading symbols from a.out... (gdb) b *main Breakpoint 1 at 0x400578 (gdb) r Starting program: a.out Breakpoint 1, 0x0000000000400578 in main () (gdb) display /x $sp 1: /x $sp = 0xfffffffff3b0 (gdb) display /x $fp 2: /x $fp = 0xfffffffff3b0 (gdb) info frame Stack level 0, frame at 0xfffffffff3b0: pc = 0x400578 in main; saved pc = 0xfffff7e5bfa0 Arglist at 0xfffffffff3b0, args: Locals at 0xfffffffff3b0, Previous frame's sp is 0xfffffffff3b0 (gdb) ni 0x000000000040057c in main () 1: /x $sp = 0xfffffffff3a0 2: /x $fp = 0xfffffffff3a0 (gdb) info frame Stack level 0, frame at 0xfffffffff3b0: pc = 0x40057c in main; saved pc = 0xfffff7e5bfa0 Arglist at 0xfffffffff3a0, args: Locals at 0xfffffffff3a0, Previous frame's sp is 0xfffffffff3b0 Saved registers: x29 at 0xfffffffff3a0, x30 at 0xfffffffff3a8 (gdb) ni 0x0000000000400580 in main () 1: /x $sp = 0xfffffffff3a0 2: /x $fp = 0xfffffffff3a0 (gdb) info frame Stack level 0, frame at 0xfffffffff3b0: pc = 0x400580 in main; saved pc = 0xfffff7e5bfa0 Arglist at 0xfffffffff3a0, args: Locals at 0xfffffffff3a0, Previous frame's sp is 0xfffffffff3b0 Saved registers: x29 at 0xfffffffff3a0, x30 at 0xfffffffff3a8 (gdb) ni 0x0000000000400584 in main () 1: /x $sp = 0xfffffffff3a0 2: /x $fp = 0xfffffffff3a0 (gdb) info frame Stack level 0, frame at 0xfffffffff3b0: pc = 0x400584 in main; saved pc = 0xfffff7e5bfa0 Arglist at 0xfffffffff3a0, args: Locals at 0xfffffffff3a0, Previous frame's sp is 0xfffffffff3b0 Saved registers: x29 at 0xfffffffff3a0, x30 at 0xfffffffff3a8 (gdb) ni 0x0000000000400588 in main () 1: /x $sp = 0xfffffffff3a0 2: /x $fp = 0xfffffffff3a0 (gdb) info frame Stack level 0, frame at 0xfffffffff3b0: pc = 0x400588 in main; saved pc = 0xfffff7e5bfa0 Arglist at 0xfffffffff3a0, args: Locals at 0xfffffffff3a0, Previous frame's sp is 0xfffffffff3b0 Saved registers: x29 at 0xfffffffff3a0, x30 at 0xfffffffff3a8 (gdb) ni 0x000000000040058c in main () 1: /x $sp = 0xfffffffff3a0 2: /x $fp = 0xfffffffff3a0 (gdb) info frame Stack level 0, frame at 0xfffffffff3b0: pc = 0x40058c in main; saved pc = 0xfffff7e5bfa0 Arglist at 0xfffffffff3a0, args: Locals at 0xfffffffff3a0, Previous frame's sp is 0xfffffffff3b0 Saved registers: x29 at 0xfffffffff3a0, x30 at 0xfffffffff3a8 (gdb) ni 0x0000000000400590 in main () 1: /x $sp = 0xfffffffff3a0 2: /x $fp = 0xfffffffff3a0 (gdb) info frame Stack level 0, frame at 0xfffffffff3b0: pc = 0x400590 in main; saved pc = 0xfffff7e5bfa0 Arglist at 0xfffffffff3a0, args: Locals at 0xfffffffff3a0, Previous frame's sp is 0xfffffffff3b0 Saved registers: x29 at 0xfffffffff3a0, x30 at 0xfffffffff3a8 (gdb) ni 0x0000000000400594 in main () 1: /x $sp = 0xfffffffff3b0 2: /x $fp = 0xfffffffff3b0 (gdb) info frame Stack level 0, frame at 0xfffffffff3c0: pc = 0x400594 in main; saved pc = 0x4004a4 Arglist at 0xfffffffff3b0, args: Locals at 0xfffffffff3b0, Previous frame's sp is 0xfffffffff3c0 Saved registers: x29 at 0xfffffffff3b0, x30 at 0xfffffffff3b8 (gdb) ni 0x0000fffff7e5bfa0 in __libc_start_main () from /lib64/libc.so.6 1: /x $sp = 0xfffffffff3b0 2: /x $fp = 0xfffffffff3b0 ... The "frame at" bit lists 0xfffffffff3b0 except at the last insn. The frame address is calculated here in aarch64_make_prologue_cache_1: ... unwound_fp = get_frame_register_unsigned (this_frame, cache->framereg); if (unwound_fp == 0) return; cache->prev_sp = unwound_fp + cache->framesize; ... For insns after the prologue, we have cache->framereg == fp and cache->framesize == 16, so unwound_fp + cache->framesize gives the wrong answer once fp has been restored to entry value by the before-last insn.
Proposed patch to trigger this PR in test-case gdb.base/unwind-on-each-insn.exp: https://sourceware.org/pipermail/gdb-patches/2023-January/195923.html
The master branch has been updated by Tom de Vries <vries@sourceware.org>: https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=29e09a42f1d1af3d97652e54b6b3cac68cf61d15 commit 29e09a42f1d1af3d97652e54b6b3cac68cf61d15 Author: Tom de Vries <tdevries@suse.de> Date: Mon Jan 23 16:49:36 2023 +0100 [gdb/tdep, aarch64] Fix frame address of last insn Consider the test-case test.c, compiled without debug info: ... void foo (const char *s) { } int main (void) { foo ("foo"); return 0; } ... Disassembly of foo: ... 0000000000400564 <foo>: 400564: d10043ff sub sp, sp, #0x10 400568: f90007e0 str x0, [sp, #8] 40056c: d503201f nop 400570: 910043ff add sp, sp, #0x10 400574: d65f03c0 ret ... Now, let's do "info frame" at each insn in foo, as well as printing $sp and $x29 (and strip the output of info frame to the first line, for brevity): ... $ gdb -q a.out Reading symbols from a.out... (gdb) b *foo Breakpoint 1 at 0x400564 (gdb) r Starting program: a.out Breakpoint 1, 0x0000000000400564 in foo () (gdb) display /x $sp 1: /x $sp = 0xfffffffff3a0 (gdb) display /x $x29 2: /x $x29 = 0xfffffffff3a0 (gdb) info frame Stack level 0, frame at 0xfffffffff3a0: (gdb) si 0x0000000000400568 in foo () 1: /x $sp = 0xfffffffff390 2: /x $x29 = 0xfffffffff3a0 (gdb) info frame Stack level 0, frame at 0xfffffffff3a0: (gdb) si 0x000000000040056c in foo () 1: /x $sp = 0xfffffffff390 2: /x $x29 = 0xfffffffff3a0 (gdb) info frame Stack level 0, frame at 0xfffffffff3a0: (gdb) si 0x0000000000400570 in foo () 1: /x $sp = 0xfffffffff390 2: /x $x29 = 0xfffffffff3a0 (gdb) info frame Stack level 0, frame at 0xfffffffff3a0: (gdb) si 0x0000000000400574 in foo () 1: /x $sp = 0xfffffffff3a0 2: /x $x29 = 0xfffffffff3a0 (gdb) info frame Stack level 0, frame at 0xfffffffff3b0: pc = 0x400574 in foo; saved pc = 0x40058c (gdb) si 0x000000000040058c in main () 1: /x $sp = 0xfffffffff3a0 2: /x $x29 = 0xfffffffff3a0 ... The "frame at" bit lists 0xfffffffff3a0 except at the last insn, where it lists 0xfffffffff3b0. The frame address is calculated here in aarch64_make_prologue_cache_1: ... unwound_fp = get_frame_register_unsigned (this_frame, cache->framereg); if (unwound_fp == 0) return; cache->prev_sp = unwound_fp + cache->framesize; ... For insns after the prologue, we have cache->framereg == sp and cache->framesize == 16, so unwound_fp + cache->framesize gives the wrong answer once sp has been restored to entry value by the before-last insn. Fix this by detecting the situation that the sp has been restored. This fixes PRs tdep/30010 and tdep/30011. This also fixes the aarch64 FAILs in gdb.reverse/solib-precsave.exp and gdb.reverse/solib-reverse.exp I reported in PR gdb/PR29721. Tested on aarch64-linux. PR tdep/30010 PR tdep/30011 Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=30010 Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=30011
Fixed by commit.