[Bug tdep/30021] [gdb/tdep, powerpc64le] previous frame inner to this frame (corrupt stack?)

["vries at gcc dot gnu.org" <sourceware-bugzilla@sourceware.org>]

https://sourceware.org/bugzilla/show_bug.cgi?id=30021

--- Comment #3 from Tom de Vries <vries at gcc dot gnu.org> ---
I analyzed this down to rs6000_epilogue_frame_prev_register giving the wrong
value for r31 at that insn.

This fix (or workaround?) make the test-case pass:
...
diff --git a/gdb/rs6000-tdep.c b/gdb/rs6000-tdep.c
index 592b447..45b4454 100644
--- a/gdb/rs6000-tdep.c
+++ b/gdb/rs6000-tdep.c
@@ -3852,6 +3852,20 @@ struct rs6000_frame_cache
 {
   struct rs6000_frame_cache *info =
     rs6000_epilogue_frame_cache (this_frame, this_cache);
+  if (regnum == 31)
+    /* In foo, we get rs6000_in_function_epilogue_frame_p == 1 starting
+       the insn at 0x100006ec.
+         00000000100006dc <foo>:
+        100006dc:   f8 ff e1 fb     std     r31,-8(r1)
+        100006e0:   d1 ff 21 f8     stdu    r1,-48(r1)
+        100006e4:   78 0b 3f 7c     mr      r31,r1
+        100006e8:   30 00 3f 38     addi    r1,r31,48
+        100006ec:   f8 ff e1 eb     ld      r31,-8(r1)
+        100006f0:   20 00 80 4e     blr
+       However, r31 is only restored after executing that insn, so asking for
+       r31 at 0x100006ec will just read the reg and return the wrong value.
+       Workaround this by getting r1 instead.  */
+    regnum = 1;
   return trad_frame_get_prev_register (this_frame, info->saved_regs, regnum);
 }

...

Using the patch, we have instead:
...
+b *foo + 12
Breakpoint 1 at 0x100006e8
+display /x $r1
+display /x $r31
+run

Breakpoint 1, 0x00000000100006e8 in foo ()
1: /x $r1 = 0x7fffffffec80
2: /x $r31 = 0x7fffffffec80
+info frame
Stack level 0, frame at 0x7fffffffecb0:
 pc = 0x100006e8 in foo; saved pc = 0x100006b0
 called by frame at 0x7fffffffed20
 Arglist at 0x7fffffffec80, args: 
 Locals at 0x7fffffffec80, Previous frame's sp is 0x7fffffffecb0
 Saved registers:
  r31 at 0x7fffffffeca8
+si
0x00000000100006ec in foo ()
1: /x $r1 = 0x7fffffffecb0
2: /x $r31 = 0x7fffffffec80
+info frame
Stack level 0, frame at 0x7fffffffecb0:
 pc = 0x100006ec in foo; saved pc = 0x100006b0
 called by frame at 0x7fffffffed20
 Arglist at 0x7fffffffec80, args: 
 Locals at 0x7fffffffec80, Previous frame's sp in r1
+si
0x00000000100006f0 in foo ()
1: /x $r1 = 0x7fffffffecb0
2: /x $r31 = 0x7fffffffecb0
+info frame
Stack level 0, frame at 0x7fffffffecb0:
 pc = 0x100006f0 in foo; saved pc = 0x100006b0
 called by frame at 0x7fffffffed20
 Arglist at 0x7fffffffecb0, args: 
 Locals at 0x7fffffffecb0, Previous frame's sp in r1
...

-- 
You are receiving this mail because:
You are on the CC list for the bug.