[Bug default/19098] new function pointer code loses data about function pointers

[Bug default/19098] new function pointer code loses data about function pointers

[woodard at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=19098

Ben Woodard <woodard at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|---                         |FIXED

--- Comment #2 from Ben Woodard <woodard at redhat dot com> ---
This one looks like it is fine with master after
09de4435ce4c3d9df2aeda6c0fa615faa2a63bb5 

hype356@ben:~/bin/abidw --abidiff
/collab/usr/global/tools/order/spack/opt/chaos_5_x86_64_ib/gcc@4.4.7/libtiff@4.0.3-3bd949f3/lib/libtiff.so.5.2.0
hype356@ben:echo $?
0

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug default/19098] new function pointer code loses data about function pointers

[woodard at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=19098

Ben Woodard <woodard at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|new function pointer code   |new function pointer code
                   |doesn't seem to accept      |loses data about function
                   |generic function pointers   |pointers

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug default/19098] New: new function pointer code doesn't seem to accept generic function pointers

[woodard at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=19098

            Bug ID: 19098
           Summary: new function pointer code doesn't seem to accept
                    generic function pointers
           Product: libabigail
           Version: unspecified
            Status: NEW
          Severity: normal
          Priority: P2
         Component: default
          Assignee: dodji at redhat dot com
          Reporter: woodard at redhat dot com
                CC: libabigail at sourceware dot org
  Target Milestone: ---

Created attachment 8693
  --> https://sourceware.org/bugzilla/attachment.cgi?id=8693&action=edit
reproducing ELF file.

I think that this one is related to some of the new function pointer code.
This was tested with code from the dodji/fixes branch

It looks like the function pointer is not saving enough type information to
test correctly against itself.

hype260@ben:~/bin/abidw --abidiff
/collab/usr/global/tools/order/spack/opt/chaos_5_x86_64_ib/gcc@4.4.7/libtiff@4.0.3-3bd949f3/lib/libtiff.so.5.2.0 
Functions changes summary: 0 Removed, 1 Changed, 0 Added function
Variables changes summary: 0 Removed, 0 Changed, 0 Added variable

1 function with some indirect sub-type change:

  [C]'function TIFFExtendProc TIFFSetTagExtender(TIFFExtendProc)' has some
indirect sub-type changes:
    return type changed:
      underlying type 'void ()*' changed:
        in pointed to type 'function type void (TIFF*)':
    parameter 1 of type 'typedef TIFFExtendProc' changed:
      underlying type 'void ()*' changed:
        in pointed to type 'function type void (TIFF*)':

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug default/19098] new function pointer code doesn't seem to accept generic function pointers

[woodard at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=19098

--- Comment #1 from Ben Woodard <woodard at redhat dot com> ---
Another version of what I think is the same problem is:

hype260@ben:cat
collab/usr/global/tools/order/spack/opt/chaos_5_x86_64_ib/gcc@4.4.7/lcms@2.6-cc59eee2/lib/liblcms2.so.2.0.6/stderr
Functions changes summary: 0 Removed, 1 Changed, 0 Added function
Variables changes summary: 0 Removed, 1 Changed, 0 Added variable

1 function with some indirect sub-type change:

  [C]'function void _cmsInstallAllocFunctions(cmsPluginMemHandler*,
_cmsMemPluginChunkType*)' has some indirect sub-type changes:
    parameter 1 of type 'cmsPluginMemHandler*' has sub-type changes:
      in pointed to type 'typedef cmsPluginMemHandler':
        underlying type 'struct __anonymous_struct__' changed:
          2 data member changes:
           type of '_cmsMallocFnPtrType __anonymous_struct__::MallocPtr'
changed:
             underlying type 'void* ()*' changed:
               in pointed to type 'function type void* (typedef cmsContext,
typedef cmsUInt32Number)':

           type of '_cmsMalloZerocFnPtrType
__anonymous_struct__::MallocZeroPtr' changed:
             underlying type 'void* ()*' changed:
               in pointed to type 'function type void* (typedef cmsContext,
typedef cmsUInt32Number)':

    parameter 2 of type '_cmsMemPluginChunkType*' has sub-type changes:
      in pointed to type 'typedef _cmsMemPluginChunkType':
        underlying type 'struct __anonymous_struct__' changed:
          2 data member changes:
           type of '_cmsMallocFnPtrType __anonymous_struct__::MallocPtr'
changed:
             underlying type 'void* ()*' changed:
               in pointed to type 'function type void* (typedef cmsContext,
typedef cmsUInt32Number)':

           type of '_cmsMalloZerocFnPtrType
__anonymous_struct__::MallocZeroPtr' changed:
             underlying type 'void* ()*' changed:
               in pointed to type 'function type void* (typedef cmsContext,
typedef cmsUInt32Number)':



1 Changed variable:

  [C]'_cmsMemPluginChunkType _cmsMemPluginChunk' was changed:
    type of variable changed:
     underlying type 'struct __anonymous_struct__' changed, as reported earlier

When this bug is fixed for libtiff I'll go back and retest with liblcms2.

-- 
You are receiving this mail because:
You are on the CC list for the bug.