public inbox for libabigail@sourceware.org
 help / color / mirror / Atom feed
* [Bug default/18941] New: Add a warning option to libabigail tools  to report possible DWARF issues
@ 2015-01-01  0:00 dodji at redhat dot com
  2015-01-01  0:00 ` [Bug default/18941] " dodji at redhat dot com
                   ` (3 more replies)
  0 siblings, 4 replies; 5+ messages in thread
From: dodji at redhat dot com @ 2015-01-01  0:00 UTC (permalink / raw)
  To: libabigail

https://sourceware.org/bugzilla/show_bug.cgi?id=18941

            Bug ID: 18941
           Summary: Add a warning option to libabigail tools  to report
                    possible DWARF issues
           Product: libabigail
           Version: unspecified
            Status: NEW
          Severity: normal
          Priority: P2
         Component: default
          Assignee: dodji at redhat dot com
          Reporter: dodji at redhat dot com
                CC: libabigail at sourceware dot org
  Target Milestone: ---

Sometimes it's interesting to raise awareness about some noticeable constructs
found in the debug information of the binary we are interested in.

For instance, there can be situations where a library does not have any debug
information for a class because the compiler knows that it will be emitted
along with the vtable in another library.  In that case, only a simple forward
declaration is found for the class in the current library.

This is fine for debugging, but interferes with ABI verification.

In this case, it would be useful that abidiff, abidw or other related tools
emit a clear diagnostic about the situation.  The library maintainer could deal
with the issue by using e.g the -femit-class-debug-always option of the G++
where she sees fit.

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

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [Bug default/18941] Add a warning option to libabigail tools  to report possible DWARF issues
  2015-01-01  0:00 [Bug default/18941] New: Add a warning option to libabigail tools to report possible DWARF issues dodji at redhat dot com
@ 2015-01-01  0:00 ` dodji at redhat dot com
  2015-01-01  0:00 ` dodji at redhat dot com
                   ` (2 subsequent siblings)
  3 siblings, 0 replies; 5+ messages in thread
From: dodji at redhat dot com @ 2015-01-01  0:00 UTC (permalink / raw)
  To: libabigail

https://sourceware.org/bugzilla/show_bug.cgi?id=18941

dodji at redhat dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|normal                      |enhancement

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

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [Bug default/18941] Add a warning option to libabigail tools  to report possible DWARF issues
  2015-01-01  0:00 [Bug default/18941] New: Add a warning option to libabigail tools to report possible DWARF issues dodji at redhat dot com
  2015-01-01  0:00 ` [Bug default/18941] " dodji at redhat dot com
@ 2015-01-01  0:00 ` dodji at redhat dot com
  2016-01-01  0:00 ` dodji at redhat dot com
  2016-01-01  0:00 ` woodard at redhat dot com
  3 siblings, 0 replies; 5+ messages in thread
From: dodji at redhat dot com @ 2015-01-01  0:00 UTC (permalink / raw)
  To: libabigail

https://sourceware.org/bugzilla/show_bug.cgi?id=18941

--- Comment #1 from dodji at redhat dot com ---
We'd also like to detect possible ODR violations, now that libabigail knows
about ODR (https://en.wikipedia.org/wiki/One_Definition_Rule).

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

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [Bug default/18941] Add a warning option to libabigail tools  to report possible DWARF issues
  2015-01-01  0:00 [Bug default/18941] New: Add a warning option to libabigail tools to report possible DWARF issues dodji at redhat dot com
                   ` (2 preceding siblings ...)
  2016-01-01  0:00 ` dodji at redhat dot com
@ 2016-01-01  0:00 ` woodard at redhat dot com
  3 siblings, 0 replies; 5+ messages in thread
From: woodard at redhat dot com @ 2016-01-01  0:00 UTC (permalink / raw)
  To: libabigail

https://sourceware.org/bugzilla/show_bug.cgi?id=18941

--- Comment #2 from Ben Woodard <woodard at redhat dot com> ---
Many of the cases which I think need warnings are more prosaic.

If it comes along a case where the DWARF has been stripped or was never
generated, then most if not all of the ABI verification is not being done. This
should be a higher priority warning. This could also remind people to point the
libabigail utilities toward an external debuginfo file. I would say that this
is a high enough priority that it should be the default unless some --quiet
option is passed on the command line.

A lower priority warning would be a exported function entry point for a
non-static function that exists in ELF but doesn't exist in the available
DWARF. This is important because we need to know the things that we are not yet
examining. I thought the option should be something like -Wunexamined-functions

A similar situation is the one you mention where you have a forward declaration
but no information about the class. -Wunexamined-classes

I think that -Wunions probably should be there too. This would trip if
libabigail found any unions in the DWARF and emit the union's name for manual
verification. I do not recall seeing this in the release notes. We should
probably have a note saying that unions are not examined for ABI compatibility
in the release notes.

And finally because the number of these warnings will likely grow over time we
should probably have something like:
-W which turns on common problems
-Wall which turns on every single warning.

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

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [Bug default/18941] Add a warning option to libabigail tools  to report possible DWARF issues
  2015-01-01  0:00 [Bug default/18941] New: Add a warning option to libabigail tools to report possible DWARF issues dodji at redhat dot com
  2015-01-01  0:00 ` [Bug default/18941] " dodji at redhat dot com
  2015-01-01  0:00 ` dodji at redhat dot com
@ 2016-01-01  0:00 ` dodji at redhat dot com
  2016-01-01  0:00 ` woodard at redhat dot com
  3 siblings, 0 replies; 5+ messages in thread
From: dodji at redhat dot com @ 2016-01-01  0:00 UTC (permalink / raw)
  To: libabigail

https://sourceware.org/bugzilla/show_bug.cgi?id=18941

dodji at redhat dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Depends on|                            |20937

--- Comment #3 from dodji at redhat dot com ---
(In reply to dodji from comment #1)
> We'd also like to detect possible ODR violations, now that libabigail knows
> about ODR (https://en.wikipedia.org/wiki/One_Definition_Rule).

I have filled enhancement request #20937 to track this specifically as it's a
request that has come up on several occasions.


Referenced Bugs:

https://sourceware.org/bugzilla/show_bug.cgi?id=20937
[Bug 20937] Detect ODR violations
-- 
You are receiving this mail because:
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2016-12-06 12:43 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2015-01-01  0:00 [Bug default/18941] New: Add a warning option to libabigail tools to report possible DWARF issues dodji at redhat dot com
2015-01-01  0:00 ` [Bug default/18941] " dodji at redhat dot com
2015-01-01  0:00 ` dodji at redhat dot com
2016-01-01  0:00 ` dodji at redhat dot com
2016-01-01  0:00 ` woodard at redhat dot com

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).