[Bug binutils/3237] New: readelf fails to dump complete DWARF 3 info
When "readelf -w" is used on client.o in http://sources.redhat.com/bugzilla/show_bug.cgi?id=3191 it fails to dump .gnu.linkonce.wi.client.cpp.7682538d section since it assumes there is only one set of debug sections. I don't know what the best way to handle it is. One possibility is to combine the same kind of sections first before dumping. Otherwise, we have to track which .debug_line a recond is referring to. However, we will need to clearly mark begin and end of subsections. -- Summary: readelf fails to dump complete DWARF 3 info Product: binutils Version: 2.18 (HEAD) Status: NEW Severity: normal Priority: P2 Component: binutils AssignedTo: unassigned at sources dot redhat dot com ReportedBy: hjl at lucon dot org CC: bug-binutils at gnu dot org http://sourceware.org/bugzilla/show_bug.cgi?id=3237 --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. ___ bug-binutils mailing list bug-binutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug ld/3191] Dwarf 2 reader in linker doesn't suppor DW_FORM_ref_addr
--- Additional Comments From hjl at lucon dot org 2006-09-20 22:22 --- Here is a new patch http://sourceware.org/ml/binutils/2006-09/msg00205.html -- http://sourceware.org/bugzilla/show_bug.cgi?id=3191 --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. ___ bug-binutils mailing list bug-binutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug ld/3191] Dwarf 2 reader in linker doesn't suppor DW_FORM_ref_addr
--- Additional Comments From hjl at lucon dot org 2006-09-20 20:57 --- My patch is wrong. It just hides the real problem. dwarf2.c doesn't really support more than one debug_xxx sections properly. It won't be easy to fix. Since dwarf2.c is mainly used for error reporting, it isn't a serious problem. BTW, readelf also assumes there is one set of DWARF debug sections. It fails to dump client.o properly. -- http://sourceware.org/bugzilla/show_bug.cgi?id=3191 --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. ___ bug-binutils mailing list bug-binutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug binutils/3235] New: objdump on AMD64 - wrong output for prefixed offset operands.
Problem:
Opcodes 0xA0-0xA3 (mov instruction) each contain a memory offset encoded within
the instruction. The offsets are address sized, however, on x86_64 the offset
size defaults to 64-bits (see AMD64 Architecture Programmer's Manual, Volume 3,
page 192). Because the offset is address sized, the address override prefix of
0x67 toggles between 64-bit/32-bit addresses when in 64-bit mode. However,
objdump always displays the offset as a 64-bit address.
Testcase:
-
memoffset.c:
-- cut
char tstStr[] = "hello world";
int main()
{
printf("Address of tstStr is 0x%lx\n", tstStr);
asm("mov $'z', %al\n\t"
"nop\n\t"
"nop\n\t"
"nop\n\t"
"nop\n\t"
"nop\n\t"
"nop\n\t"
"nop\n\t"
"nop\n\t"
"nop\n\t"
"nop\n\t");
printf("tstStr is \"%s\"\n", tstStr);
return 0;
}
-- cut
I don't know how to force gas to generate the 0xA2 mov instruction (it
generates a 0x88 mov), so I have to do this by hand. When I run the program,
the address printed is 0x500908, so using a hex editor I replace the sequence
of nops with "0x67 0xa2 0x08 0x09 0x50 0x00". After the hexedit, the program
is rerun, and the output displayed is "zello world" instead of "hello world"
(the mov instruction was successful). objdump however still displays the
address as 64-bits:
400500: b0 7a mov$0x7a,%al
400502: 67 a2 08 09 50 00 90addr32 mov %al,0xbe90909000500908
400509: 90 90 be
which is impossible since this is an invalid address (composition of tstStr's
address and the next 3 opcodes). Instead, the output should be:
400500: b0 7a mov$0x7a,%al
400502: 67 a2 08 09 50 00 90addr32 mov %al,0x500908
400509: 90 nop
40050A: 90 nop
...
Version info:
-
[EMAIL PROTECTED] ~/memOffset $ objdump -v
GNU objdump 2.16.1
[EMAIL PROTECTED] ~/memOffset $ uname -ar
Linux collateral 2.6.16-gentoo-r9 #1 SMP PREEMPT Fri Sep 8 11:52:36 EDT 2006
x86_64 AMD Opteron(tm) Processor 246 GNU/Linux
--
Summary: objdump on AMD64 - wrong output for prefixed offset
operands.
Product: binutils
Version: unspecified
Status: NEW
Severity: normal
Priority: P2
Component: binutils
AssignedTo: unassigned at sources dot redhat dot com
ReportedBy: fruffell at cs dot uwaterloo dot ca
CC: bug-binutils at gnu dot org
http://sourceware.org/bugzilla/show_bug.cgi?id=3235
--- You are receiving this mail because: ---
You are on the CC list for the bug, or are watching someone who is.
___
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-binutils
Re: [Bug ld/3191] Dwarf 2 reader in linker doesn't suppor DW_FORM_ref_addr
On Wed, 2006-09-20 at 19:02 +, hjl at lucon dot org wrote: > When reporting linker error, dwarf2 reader doesn't support more than one > .debug_info section. You can only have more than one debug_info section when -feliminate-dwarf2-dups is used. This isn't a commonly used option, and there may be a number of problems with it. This is allowed by DWARF3, but not by DWARF2. I see how your patch might help now. The debug info is 32-bits as I mentioned earlier. The problem here is that pointers are 64-bits. read_address uses addr_size, and hence is reading a 64-bit address when it should be reading a 32-bit address, thus getting a number that is too large. Your fix adding 64-bit debug support helps because in the 32-bit case you replaced the read_address call with a read_4_bytes call, which correctly reads the 32-bit value we want. However, this is speculation, as I don't know how to reproduce the failure. I don't understand why this hasn't come up before. Also, it isn't clear what this has to do with the further information you have provided about multiple debug_info sections. -- Jim Wilson, GNU Tools Support, http://www.specifix.com ___ bug-binutils mailing list bug-binutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug ld/3191] Dwarf 2 reader in linker doesn't suppor DW_FORM_ref_addr
--- Additional Comments From wilson at specifix dot com 2006-09-20 19:31 --- Subject: Re: Dwarf 2 reader in linker doesn't suppor DW_FORM_ref_addr On Wed, 2006-09-20 at 19:02 +, hjl at lucon dot org wrote: > When reporting linker error, dwarf2 reader doesn't support more than one > .debug_info section. You can only have more than one debug_info section when -feliminate-dwarf2-dups is used. This isn't a commonly used option, and there may be a number of problems with it. This is allowed by DWARF3, but not by DWARF2. I see how your patch might help now. The debug info is 32-bits as I mentioned earlier. The problem here is that pointers are 64-bits. read_address uses addr_size, and hence is reading a 64-bit address when it should be reading a 32-bit address, thus getting a number that is too large. Your fix adding 64-bit debug support helps because in the 32-bit case you replaced the read_address call with a read_4_bytes call, which correctly reads the 32-bit value we want. However, this is speculation, as I don't know how to reproduce the failure. I don't understand why this hasn't come up before. Also, it isn't clear what this has to do with the further information you have provided about multiple debug_info sections. -- http://sourceware.org/bugzilla/show_bug.cgi?id=3191 --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. ___ bug-binutils mailing list bug-binutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug ld/3191] Dwarf 2 reader in linker doesn't suppor DW_FORM_ref_addr
--- Additional Comments From hjl at lucon dot org 2006-09-20 19:02 --- Gcc may generate: .section.gnu.linkonce.wi.client.cpp.7682538d,"",@progbits ... DW.client.cpp.7682538d.2b: .uleb128 0xd .long 0x16a .byte 0x1 .long .LASF10 .byte 0x1 .byte 0x25 .byte 0x1 .uleb128 0xe .long 0x114 .byte 0x1 .byte 0x0 .section.debug_info ... .uleb128 0x8d .long 0x1c3f .quad DW.client.cpp.7682538d.2b .byte 0x0 When reporting linker error, dwarf2 reader doesn't support more than one .debug_info section. -- What|Removed |Added Summary|Linker doesn't support |Dwarf 2 reader in linker |DWARF3 DW_FORM_ref_addr |doesn't suppor ||DW_FORM_ref_addr http://sourceware.org/bugzilla/show_bug.cgi?id=3191 --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. ___ bug-binutils mailing list bug-binutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug ld/3191] Linker doesn't support DWARF3 DW_FORM_ref_addr
-- What|Removed |Added Attachment #1306 is|0 |1 obsolete|| http://sourceware.org/bugzilla/show_bug.cgi?id=3191 --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. ___ bug-binutils mailing list bug-binutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug ld/3191] Linker doesn't support DWARF3 DW_FORM_ref_addr
--- Additional Comments From pluto at agmk dot net 2006-09-20 09:26 --- Created an attachment (id=1307) --> (http://sourceware.org/bugzilla/attachment.cgi?id=1307&action=view) preprocessed sources and objects from gcc-4.1/4.2 -- http://sourceware.org/bugzilla/show_bug.cgi?id=3191 --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. ___ bug-binutils mailing list bug-binutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug ld/3191] Linker doesn't support DWARF3 DW_FORM_ref_addr
--- Additional Comments From pluto at agmk dot net 2006-09-20 09:24 --- (In reply to comment #11) > (In reply to comment #8) > > It looks like a gcc bug. See > > > > http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29146 > > No it does not. > What version of GCC did you produce this object file? with gcc 4.1.2-20060817 (without patch from #7) i get: $ g++ -O2 -fno-strict-aliasing -fwrapv -march=x86-64 \ -gdwarf-2 -g2 -feliminate-dwarf2-dups client-gcc41.ii -c $ g++ -o /dev/null client-gcc41.o (...) /usr/bin/ld: Dwarf Error: DW_FORM_strp offset (660090) greater than or equal to .debug_str size (27111). /usr/bin/ld: Dwarf Error: DW_FORM_strp offset (660090) greater than or equal to .debug_str size (27111). (...) with gcc 4.1.2-20060817 (with patch from #7) it works fine. with gcc-4.2 (without patch from #7) it works fine. with gcc-4.2 (with patch from #7) not tested (probably works too). > And how did you build GCC while you are at that? $ gcc -v Reading specs from /usr/lib64/gcc/x86_64-pld-linux/4.1.2/specs Target: x86_64-pld-linux Configured with: ../configure --prefix=/usr --with-local-prefix=/usr/local --libdir=/usr/lib64 --libexecdir=/usr/lib64 --infodir=/usr/share/info --mandir=/usr/share/man --x-libraries=/usr/lib64 --enable-shared --enable-threads=posix --enable-languages=c,c++,fortran,objc,obj-c++,ada,java --enable-c99 --enable-long-long --enable-multilib --enable-nls --disable-werror --with-gnu-as --with-gnu-ld --with-demangler-in-ld --with-system-zlib --with-slibdir=/lib64 --without-system-libunwind --enable-cmath --with-long-double-128 --with-gxx-include-dir=/usr/include/c++/4.1.2 --disable-libstdcxx-pch --enable-__cxa_atexit --enable-libstdcxx-allocator=new --with-qt4dir=/usr/lib64/qt4 --disable-libjava-multilib --enable-libgcj --enable-libgcj-multifile --enable-libgcj-database --enable-gtk-cairo --enable-java-awt=qt,gtk,xlib --enable-jni --enable-xmlj --enable-alsa --enable-dssi x86_64-pld-linux Thread model: posix gcc version 4.1.2 20060817 (prerelease) (PLD-Linux) -- http://sourceware.org/bugzilla/show_bug.cgi?id=3191 --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. ___ bug-binutils mailing list bug-binutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-binutils
