This is the mail archive of the binutils@sourceware.org mailing list for the binutils project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [binutils-gdb] Fix the linker so that it will not silently generate ELF binaries with invalid program headers. Fix


Hi Nick,

On Wed, 23 Nov 2016, Nick Clifton wrote:

> commit 1a9ccd70f9a75dc6b48d340059f28ef3550c107b
> Author: Nick Clifton <nickc@redhat.com>
> Date:   Wed Nov 23 11:10:39 2016 +0000
> 
>     Fix the linker so that it will not silently generate ELF binaries with invalid program headers.  Fix readelf to report such invalid binaries.

 This change has caused a `mips-vxworks' and `mipsel-vxworks' target 
regression:

FAIL: VxWorks executable test 2 (dynamic)

and I cannot figure out either from the commit description or the lengthy 
Bugzilla discussion whether it is an actual functional VxWorks regression 
or whether we can just adjust the regexps in the test case and consider it 
handled.

 Your commit changed output from this test case from:

Elf file type is EXEC (Executable file)
Entry point 0x80400
There are 5 program headers, starting at offset 52

Program Headers:
  Type           Offset   VirtAddr   PhysAddr   FileSiz MemSiz  Flg Align
  PHDR           0x000034 0x00001034 0x00000000 0x000a0 0x000a0 R E 0x4
  INTERP         0x001000 0x00080000 0x00080000 0x00013 0x00013 R   0x1
      [Requesting program interpreter: /usr/lib/libc.so.1]
  LOAD           0x001000 0x00080000 0x00080000 0x00408 0x00408 R E 0x1000
  LOAD           0x002000 0x00081000 0x00081000 0x00804 0x00c00 RW  0x1000
  DYNAMIC        0x002000 0x00081000 0x00081000 0x00078 0x00078 RW  0x4

 Section to Segment mapping:
  Segment Sections...
   00
   01     .interp
   02     .interp .hash .dynsym .dynstr .text
   03     .dynamic .got .data .rld_map
   04     .dynamic

to:

Elf file type is EXEC (Executable file)
Entry point 0x80400
There are 5 program headers, starting at offset 52

Program Headers:
  Type           Offset   VirtAddr   PhysAddr   FileSiz MemSiz  Flg Align
  PHDR           0x000034 0x0007f034 0x0007f034 0x000a0 0x000a0 R E 0x4
  INTERP         0x001000 0x00080000 0x00080000 0x00013 0x00013 R   0x1
      [Requesting program interpreter: /usr/lib/libc.so.1]
  LOAD           0x000000 0x0007f000 0x0007f000 0x01408 0x01408 R E 0x1000
  LOAD           0x002000 0x00081000 0x00081000 0x00804 0x00c00 RW  0x1000
  DYNAMIC        0x002000 0x00081000 0x00081000 0x00078 0x00078 RW  0x4

 Section to Segment mapping:
  Segment Sections...
   00
   01     .interp
   02     .interp .hash .dynsym .dynstr .text
   03     .dynamic .got .data .rld_map
   04     .dynamic

This places the first LOAD segment below the location set with the linker 
script, which I fear may affect run-time interpretation.  The test case 
has the addresses expected spelled out in full for a reason I suppose.

 Please help with resolving this problem.

  Maciej


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]