gdb on x86

gdb on x86

[Tom Tromey]

I'm running Red Hat x86 Linux 6.2.

I'm using gdb from sourceware.  My latest gdb/ChangeLog entry starts:

    2000-08-04  Kevin Buettner  <kevinb@redhat.com>

	    * minsyms.c (build_minimal_symbol_hash_tables): New function.
	    (compact_minimal_symbols): Don't construct hash tables here.

I did a clean build (deleted old build tree) after checking this out.

This gdb crashes immediately when I try to debug my Java program.
Stack trace appended.

Tom

#0  0x401844a7 in memcpy (dstpp=0x8484980, srcpp=0x836f21c, len=2147483648)
    at ../sysdeps/generic/memcpy.c:55
#1  0x817239a in string_appendn (p=0xbfffef3c, s=0x836f21c "", n=-2147483648)
    at ../../src/libiberty/cplus-dem.c:4633
#2  0x816f8d6 in demangle_arm_hp_template (work=0xbffff0d0, 
    mangled=0xbffff0c0, n=-2147483648, declp=0xbfffef3c)
    at ../../src/libiberty/cplus-dem.c:2272
#3  0x816f930 in demangle_class_name (work=0xbffff0d0, mangled=0xbffff0c0, 
    declp=0xbfffef3c) at ../../src/libiberty/cplus-dem.c:2295
#4  0x816f971 in demangle_class (work=0xbffff0d0, mangled=0xbffff0c0, 
    declp=0xbffff0a8) at ../../src/libiberty/cplus-dem.c:2350
#5  0x816e4b2 in demangle_signature (work=0xbffff0d0, mangled=0xbffff0c0, 
    declp=0xbffff0a8) at ../../src/libiberty/cplus-dem.c:1253
#6  0x816fb26 in iterate_demangle_function (work=0xbffff0d0, 
    mangled=0xbffff0c0, declp=0xbffff0a8, 
    scan=0x836f202 "__1_2147483647__2147483648")
    at ../../src/libiberty/cplus-dem.c:2432
#7  0x816ff13 in demangle_prefix (work=0xbffff0d0, mangled=0xbffff0c0, 
    declp=0xbffff0a8) at ../../src/libiberty/cplus-dem.c:2661
#8  0x816df4d in internal_cplus_demangle (work=0xbffff0d0, 
    mangled=0x836f21c "") at ../../src/libiberty/cplus-dem.c:949
#9  0x816deaa in cplus_demangle (
    mangled=0x836f1f8 "_Utf58_0_1__1_2147483647__2147483648", options=3)
    at ../../src/libiberty/cplus-dem.c:902
#10 0x80ca3ac in install_minimal_symbols (objfile=0x833d4c8)
    at ../../src/gdb/minsyms.c:959
#11 0x810589a in elfstab_build_psymtabs (objfile=0x833d4c8, mainline=0, 
    staboffset=855456, stabsize=294432, stabstroffset=1149888, 
    stabstrsize=356281) at ../../src/gdb/dbxread.c:2590
#12 0x8108858 in elf_symfile_read (objfile=0x833d4c8, mainline=0)
    at ../../src/gdb/elfread.c:645
#13 0x80995d0 in syms_from_objfile (objfile=0x833d4c8, addrs=0xbffff268, 
    mainline=1, verbo=0) at ../../src/gdb/symfile.c:733
#14 0x8099781 in symbol_file_add (name=0x8337120 "./SimpleTestHarness", 
    from_tty=0, addrs=0x0, mainline=1, flags=32) at ../../src/gdb/symfile.c:862
#15 0x8099957 in symbol_file_command (args=0xbffff974 "./SimpleTestHarness", 
    from_tty=0) at ../../src/gdb/symfile.c:974
#16 0x80d9a2c in do_captured_command (data=0xbffff588)
    at ../../src/gdb/top.c:671
#17 0x80d99a3 in catch_errors (func=0x80d9a1c <do_captured_command>, 
    args=0xbffff588, errstring=0x8272f80 "", mask=6) at ../../src/gdb/top.c:607
#18 0x80d9a6a in catch_command_errors (
    command=0x809983c <symbol_file_command>, 
    arg=0xbffff974 "./SimpleTestHarness", from_tty=0, mask=6)
    at ../../src/gdb/top.c:691
#19 0x8081c41 in captured_main (data=0xbffff7b0) at ../../src/gdb/main.c:603
#20 0x80d99a3 in catch_errors (func=0x8081434 <captured_main>, 
    args=0xbffff7b0, errstring=0x824dc02 "", mask=6) at ../../src/gdb/top.c:607
#21 0x8081e93 in main (argc=3, argv=0xbffff804) at ../../src/gdb/main.c:761
From tromey@cygnus.com Sat Aug 05 16:58:00 2000
From: Tom Tromey <tromey@cygnus.com>
To: tromey@cygnus.com
Cc: gdb@sources.redhat.com
Subject: Re: gdb on x86
Date: Sat, 05 Aug 2000 16:58:00 -0000
Message-id: <87og37qmqu.fsf@creche.i-did-not-set--mail-host-address--so-shoot-me>
References: <87snsjqo5m.fsf@creche.i-did-not-set--mail-host-address--so-shoot-me>
X-SW-Source: 2000-08/msg00040.html
Content-length: 189

Following up my own post...

Tom> This gdb crashes immediately when I try to debug my Java program.
Tom> Stack trace appended.

The fix is to bring over the latest demangler from gcc.

Tom
From dberlin@redhat.com Sat Aug 05 17:48:00 2000
From: Daniel Berlin <dberlin@redhat.com>
To: tromey@cygnus.com
Cc: gdb@sources.redhat.com
Subject: Re: gdb on x86
Date: Sat, 05 Aug 2000 17:48:00 -0000
Message-id: <m3d7jni59y.fsf@dan2.cygnus.com>
References: <87snsjqo5m.fsf@creche.i-did-not-set--mail-host-address--so-shoot-me> <87og37qmqu.fsf@creche.i-did-not-set--mail-host-address--so-shoot-me>
X-SW-Source: 2000-08/msg00041.html
Content-length: 414

Tom Tromey <tromey@cygnus.com> writes:

Yes, yes, but you need to merge it with the changes i made to the
demangler, as they still aren't apparently actually in the gcc
sources, yet have been approved.
--Dan
> Following up my own post...
> 
> Tom> This gdb crashes immediately when I try to debug my Java program.
> Tom> Stack trace appended.
> 
> The fix is to bring over the latest demangler from gcc.
> 
> Tom