Regression: graceful unwind termination when we'd need unavailable/uncollect memory or registers to unwind further

[Jan Kratochvil <jan.kratochvil@redhat.com>]

Hi Pedro,

on a testfile from the Fedora patchset there is a regression, IMO a valid one:

e67dca4a095f4db1022fd0753ef0bda3873dc1bc is the first bad commit
commit e67dca4a095f4db1022fd0753ef0bda3873dc1bc
Author: Pedro Alves <pedro@codesourcery.com>
Date:   Fri Mar 18 18:52:28 2011 +0000

    	gdb/
    	* frame.c (frame_unwind_register): Throw an error if unwinding the
    	register failed.
[...]
:040000 040000 6bdd3071cc91073d64ffd2a1030c44ce2b1956b9 a4487e736a1087aee5d5b5207cee5a56ae936f42 M	gdb
bisect run success

http://pkgs.fedoraproject.org/gitweb/?p=gdb.git;a=blob_plain;f=gdb-6.8-bz457187-largefile-test.patch;hb=f15/master
 = This file is included in this mail.

(gdb) x/i 0x400078
   0x400078:    hlt    
PASS: gdb.arch/i386-biarch-core.exp: .text is readable
->
(gdb) x/i 0x400078
PC not available
FAIL: gdb.arch/i386-biarch-core.exp: .text is readable

The core file does not have accessible registers.  It was not a goal of the
testcase, it tests something else.  But it has found this unrelated
regression.  I believe one should be able to evaluate PC-indepenent
expressions even if PC is not available.


Even the first part is regressing (although it stays PASS->PASS):
(gdb) core-file gdb.arch/i386-biarch-core.core
[New LWP 6901]
warning: Couldn't recognize general-purpose registers in core file.
Core was generated by `./bad'.
Program terminated with signal 11, Segmentation fault.
warning: Couldn't recognize general-purpose registers in core file.
#0  0x00000000 in ?? ()
->
(gdb) core-file gdb.arch/i386-biarch-core.core
[New LWP 6901]
warning: Couldn't recognize general-purpose registers in core file.
PC register is not available

The core file reason is not displayed while it could be, I have tried the
following patch does not help in this case:
	[patch 3/3] Display core reasons even during thread error
	http://sourceware.org/ml/gdb-patches/2011-02/msg00675.html


Thanks,
Jan


--- /dev/null	2009-04-19 14:49:00.974648389 +0200
+++ gdb-6.8/gdb/testsuite/gdb.arch/i386-biarch-core.exp	2009-04-19 16:30:12.000000000 +0200
@@ -0,0 +1,61 @@
+# This testcase is part of GDB, the GNU debugger.
+
+# Copyright 2009 Free Software Foundation, Inc.
+
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program; if not, write to the Free Software
+# Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.  
+
+# Test ability to load an elf64-i386 core file.  The provided core file was
+# elf64-x8664 one but it got binary patched to i386:
+# Elf32_Ehdr.e_machine @0x12..0x13
+# Elf64_Ehdr.e_machine @0x12..0x13
+# #define EM_386           3              /* Intel 80386 */
+# #define EM_X86_64       62              /* AMD x86-64 architecture */
+# patch @0x12: 0x3E -> 0x03
+
+if { ![istarget "i?86-*-*"] && ![istarget "x86_64-*-*"] } then {
+    verbose "Skipping i386-biarch-core test."
+    return
+}
+
+set testfile "i386-biarch-core"
+set corebz2uufile ${srcdir}/${subdir}/${testfile}.core.bz2.uu
+set corefile ${objdir}/${subdir}/${testfile}.core
+# Entry point of the original executable.
+set address 0x400078
+
+if {[catch "system \"uudecode -o - ${corebz2uufile} | bzip2 -dc >${corefile}\""] != 0} {
+    untested "failed uudecode or bzip2"
+    return -1
+}
+file stat ${corefile} corestat
+if {$corestat(size) != 102400} {
+    untested "uudecode or bzip2 produce invalid result"
+    return -1
+}
+
+gdb_exit
+gdb_start
+gdb_reinitialize_dir $srcdir/$subdir
+
+# Wrongly built GDB complains by:
+# "..." is not a core dump: File format not recognized
+# As the provided test core has 64bit PRSTATUS i386 built GDB cannot parse it.
+# This is just a problem of the test care, real-world elf64-i386 file will have
+# 32bit PRSTATUS.  One cannot prepare elf64-i386 core file from elf32-i386 by
+# objcopy as it corrupts the core file beyond all recognition.
+# "\r\nCore was generated by `\[^\r\n\]*'\\.\r\nProgram terminated with signal 11, Segmentation fault\\.\r\n.*"
+gdb_test "core-file ${corefile}"
+
+gdb_test "x/i $address" "\r\n\[ \t\]*$address:\[ \t\]*hlt\[ \t\]*" ".text is readable"
--- /dev/null	2009-04-19 14:49:00.974648389 +0200
+++ gdb-6.8/gdb/testsuite/gdb.arch/i386-biarch-core.S	2009-04-19 14:52:28.000000000 +0200
@@ -0,0 +1,22 @@
+/* This testcase is part of GDB, the GNU debugger.
+
+   Copyright 2009 Free Software Foundation, Inc.
+  
+   This program is free software; you can redistribute it and/or modify
+   it under the terms of the GNU General Public License as published by
+   the Free Software Foundation; either version 2 of the License, or
+   (at your option) any later version.
+  
+   This program is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+   GNU General Public License for more details.
+  
+   You should have received a copy of the GNU General Public License
+   along with this program; if not, write to the Free Software
+   Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ */
+
+	.globl _start
+_start:
+	hlt
--- /dev/null	2009-04-19 14:49:00.974648389 +0200
+++ gdb-6.8/gdb/testsuite/gdb.arch/i386-biarch-core.core.bz2.uu	2009-04-19 15:02:12.000000000 +0200
@@ -0,0 +1,13 @@
+begin 600 i386-biarch-core.core.bz2
+M0EIH.3%!629361`P1\P`!)?_____\9'@"8Q)6P380'9@'&#`0D@``"``%(``
+M@`#`"!<(L`%F"(:$GH13::F-)M&D&U,AD:`--#)M0&FT0XR9--,)D9`P(Q-&
+M",(-&F``02)%38HT]0T`&AH```'H@``T^>9T*(,("&)SE`>`9@+GP=[,N)KB
+M'I8BL(L]N5TCY\%V]/?DB.BN*UZ'U@]TN7-]UJ5\_%0QTT<*086#%MHT7XVJ
+M9D"+C!"2*L:8D1XPD!`--M@*XT1H5RFYN&)(!0P0#:`I:;2;$5M&\*9"0@%:
+MK@X[T()M)9N7`D$VA!^63)%,;@8LT`(7\K&[7G;U:"B6'!GG+46ALOZF.2F-
+M!@>C*%86X$-]C2`KE;HG)UL(913VR2G]0BD:J=Z_`G@S,`W%.8RMS-#5P:J0
+MAJ2\8&X?@DE;UF68QHM<,D`('::J65/S:PAG*R-09["8DBI)'V]Y.[(/AM*L
+M"X_O^V;%FY.S6Q]FM=D37>5F,%4-F1ZF#,CFJVU;H*^IT<(%<V`.32$`JU["
+/G`68?\7<D4X4)`0,$?,`
+`
+end