[Bug gas/10150] New: GAS doesn't allow ljmp foo:bar syntax when using .intel_syntax mode

2009-05-13 Thread jacob at whotookspaz dot org
GAS currently doesn't allow the following syntax for ljmp when using
.intel_syntax noprefix: ljmp 0x8:gdt_update_flush

The equavilent AT&T syntax is: ljmp $0x8, $gdt_update_flush.

The error GAS spits out is foo.S:bar: Error: ambiguous operand size or operands
invalid for `ljmp'.

The current workaround I'm using is turning on .att_syntax then using the
equavilent AT&T syntax. However, this is obviously suboptimal.

-- 
   Summary: GAS doesn't allow ljmp foo:bar syntax when using
.intel_syntax mode
   Product: binutils
   Version: unspecified
Status: NEW
  Severity: normal
  Priority: P2
 Component: gas
AssignedTo: unassigned at sources dot redhat dot com
ReportedBy: jacob at whotookspaz dot org
CC: bug-binutils at gnu dot org
 GCC build triplet: elf-i386
  GCC host triplet: linux-elf-i386
GCC target triplet: linux-elf-i386


http://sourceware.org/bugzilla/show_bug.cgi?id=10150

--- 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/10151] New: libbfd: adr2errline resolution does not work for 32bit programs on x86_64 plattforms

2009-05-13 Thread estellnb at gmail dot com
Using the addr2errline sample program from
http://linuxgazette.net/151/misc/melinte/addr2line.c
I get the following results:

> gcc addr2errline.c -o addr2errline -lbfd -liberty -lz
> ./addr2errline getlineno 0x804b335 0x804b2aa
main [getlineno.c:61]
ThisLine [getlineno.c:49]

> gcc -m32 addr2errline.c -o addr2errline -lbfd -liberty -lz
> ./addr2errline getlineno 0x804b335 0x804b2aa
>  ** no result, nothing found, 0 lines of output **

It does not depend on the library version. If I link against old 32bit bfd &
liberty librays from x86_32 days of OpenSuse10.1 it does not work either.
Perhaps this is some general incompatibility of 32/64bit plattforms.

Debugging addr2errline.c I found that the error turns out at the following
point:
slurp_symtab(bfd *abfd)
{
...
if ((bfd_get_file_flags(abfd) & HAS_SYMS) == 0)
return -1;

binutils-2.19-9.3

-- 
   Summary: libbfd: adr2errline resolution does not work for 32bit
programs on x86_64 plattforms
   Product: binutils
   Version: 2.19
Status: NEW
  Severity: normal
  Priority: P2
 Component: binutils
AssignedTo: unassigned at sources dot redhat dot com
ReportedBy: estellnb at gmail dot com
CC: bug-binutils at gnu dot org
 GCC build triplet: ELF32, Intel 80386
  GCC host triplet: x86_64
GCC target triplet: i686-pc-linux-gnu


http://sourceware.org/bugzilla/show_bug.cgi?id=10151

--- 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


libbfd: adr2errline resolution does not work for 32bit programs on x86_64 plattforms

2009-05-13 Thread Elmar Stellnberger
Using the addr2errline sample program from
http://linuxgazette.net/151/misc/melinte/addr2line.c
I get the following results:

> gcc addr2errline.c -o addr2errline -lbfd -liberty -lz
> ./addr2errline getlineno 0x804b335 0x804b2aa
main [getlineno.c:61]
ThisLine [getlineno.c:49]

> gcc -m32 addr2errline.c -o addr2errline -lbfd -liberty -lz
> ./addr2errline getlineno 0x804b335 0x804b2aa
>  ** no result, nothing found, 0 lines of output **

It does not depend on the library version. If I link against old 32bit bfd &
liberty librays from x86_32 days of OpenSuse10.1 it does not work either.
Perhaps this is some general incompatibility of 32/64bit plattforms.

Debugging addr2errline.c I found that the error turns out at the following
point:
slurp_symtab(bfd *abfd)
{
...
if ((bfd_get_file_flags(abfd) & HAS_SYMS) == 0)
return -1;

binutils-2.19-9.3
posted as: Bug 10151


___
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-binutils


[Bug ld/10152] New: [2.20 regression] ld segfault linking the kernel

2009-05-13 Thread doko at debian dot org
http://sourceware.org/ml/binutils/2009-05/msg00220.html

Paul Brook schrieb:
> The patch below implements linker processing of ARM unwinding tables 
> (SHT_ARM_EXIDX).
> 
> 2009-05-05  Paul Brook  
> 
>   bfd/
>   * bfd-in.h (elf32_arm_fix_exidx_coverage): Add prototype.
>   * bfd-in2.h: Regenerate.
>   * elf32-arm.c (arm_unwind_edit_type, arm_unwind_table_edit): Define.
>   (_arm_elf_section_data): Add text and exidx fields.
>   (add_unwind_table_edit, get_arm_elf_section_data, adjust_exidx_size,
>   insert_cantunwind_after, elf32_arm_fix_exidx_coverage, offset_prel31,
>   copy_exidx_entry): New functions.
>   (elf32_arm_write_section): Fixup .ARM.exidx contents.
> 
>   ld/
>   * emultempl/armelf.em (compare_output_sec_vma): New function.
>   (gld${EMULATION_NAME}_finish): Add exidx munging code.

This breaks kernel builds on arm-linux-gnueabi. Checked by reverting this
change. command line used and object files at https://launchpad.net/bugs/375991

  Matthias

Program received signal SIGSEGV, Segmentation fault.
bfd_set_section_size (abfd=0x0, ptr=0x7e5f8, val=8) at ../../bfd/section.c:1297
1297  if (abfd->output_has_begun)
(gdb) bt
#0  bfd_set_section_size (abfd=0x0, ptr=0x7e5f8, val=8) at 
../../bfd/section.c:1297
#1  0x4005c60c in elf32_arm_fix_exidx_coverage (text_section_order=, num_text_sections=4,
info=) at ../../bfd/elf32-arm.c:8425
#2  0x00026980 in gldarmelf_linux_eabi_finish () at earmelf_linux_eabi.c:420
#3  0x00021ea8 in ldemul_finish () at ../../ld/ldemul.c:90
#4  0x0001ac08 in lang_process () at ../../ld/ldlang.c:6282
#5  0x0001f154 in main (argc=1, argv=0x12) at ../../ld/ldmain.c:479

-- 
   Summary: [2.20 regression] ld segfault linking the kernel
   Product: binutils
   Version: 2.20 (HEAD)
Status: NEW
  Severity: normal
  Priority: P2
 Component: ld
AssignedTo: unassigned at sources dot redhat dot com
ReportedBy: doko at debian dot org
CC: bug-binutils at gnu dot org
GCC target triplet: arm-linux-gnueabi


http://sourceware.org/bugzilla/show_bug.cgi?id=10152

--- 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/10152] [2.20 regression] ld segfault linking the kernel

2009-05-13 Thread doko at debian dot org


-- 
   What|Removed |Added

 CC||paul at codesourcery dot com


http://sourceware.org/bugzilla/show_bug.cgi?id=10152

--- 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/6714] WindRes 2.18 can't concatenate literal strings anymore?

2009-05-13 Thread stu at nosnilmot dot com

--- Additional Comments From stu at nosnilmot dot com  2009-05-13 19:34 
---
Here's a test case:

$ cat stringconcattest.rc 
#define PIXMAPSDIR "pixmaps/"

TRAY_ICON_TEST  ICON PIXMAPSDIR "testicon.ico"

$ ls -l pixmaps/testicon.ico 
-rw-rw-r--. 1 stu stu 318 2009-05-13 20:30 pixmaps/testicon.ico

$ i686-pc-mingw32-windres -i stringconcattest.rc -o stringconcattest.o
i686-pc-mingw32-windres: icon file `pixmaps/' does not contain icon data

$ i686-pc-mingw32-windres --version
GNU windres (GNU Binutils) 2.19.1


-- 
   What|Removed |Added

 CC||stu at nosnilmot dot com


http://sourceware.org/bugzilla/show_bug.cgi?id=6714

--- 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