public inbox for libabigail@sourceware.org
 help / color / mirror / Atom feed
* Where are we with libabigail 2.1.1
@ 2022-10-14 21:53 Ben Woodard
  0 siblings, 0 replies; only message in thread
From: Ben Woodard @ 2022-10-14 21:53 UTC (permalink / raw)
  To: libabigail

[-- Attachment #1: Type: text/plain, Size: 5287 bytes --]

Dodji and I both want to do a quick turn around on libabigail 2.1.1 to 
fix some remaining bugs.

So I went through one of my big distro wide tests. It is considerably 
slower on Fedora 37 than it was on 36. It has been running for 3.2 days 
when the last run on Fedora 36 took 23 hours Maybe something can be done 
with this bug for 2.2 but I don't think it is a 2.1.1 issue 
https://sourceware.org/bugzilla/show_bug.cgi?id=29303 . It is still not 
done but I think that I have enough data to allow me to clean up some 
old bugs and see what the current problems are.

There are 2 fairly bad self comparison bugs that both seem to bite C++ a 
handful of other 1-off kinds of bugs that I attached to: 
https://sourceware.org/bugzilla/show_bug.cgi?id=29413

  * https://sourceware.org/bugzilla/show_bug.cgi?id=29407 - the most
    important problem
  * https://sourceware.org/bugzilla/show_bug.cgi?id=29673 - less
    important but still very common
  * https://sourceware.org/bugzilla/show_bug.cgi?id=29683
  * https://sourceware.org/bugzilla/show_bug.cgi?id=29684
  * https://sourceware.org/bugzilla/show_bug.cgi?id=29685 - after wading
    through this a bit more it may be a duplicate of 29407. I'll recheck
    it when dodji closes 29407.

The first two are fairly important but the last three do not seem as 
critical.

Then there is a weird one associated with blas that causes 
std::out_of_range - 
https://sourceware.org/bugzilla/show_bug.cgi?id=29690 and another weird 
one is https://sourceware.org/bugzilla/show_bug.cgi?id=29692 where 
kdelibs3 exits with 13 an odd exit code.

Then there are a bunch of asserts some of which we knew about before and 
others are new. I think the most common of these can be converted to an 
error string. I attached all the assert bugs to this: 
https://sourceware.org/bugzilla/show_bug.cgi?id=29412

  * ../../../libabigail/src/abg-comparison.cc:4543
    https://sourceware.org/bugzilla/show_bug.cgi?id=29686
      o haproxy
  * ../../../libabigail/src/abg-corpus.cc:649
    https://sourceware.org/bugzilla/show_bug.cgi?id=29687
      o coq-core
  * ../../../libabigail/src/abg-dwarf-reader.cc:17334
    https://sourceware.org/bugzilla/show_bug.cgi?id=29339
      o julia-devel
      o julia
  * ../../../libabigail/src/abg-ir.cc:14680
    https://sourceware.org/bugzilla/show_bug.cgi?id=29688
      o firebird
      o gcc-gnat
      o llvm7.0
  * ../../../libabigail/src/abg-ir.cc:17342
    https://sourceware.org/bugzilla/show_bug.cgi?id=29333
      o aws
      o ghdl
      o gnatcoll-core
      o gprbuild
      o matreshka-amf
      o templates_parser
      o Ada related ones
        https://sourceware.org/bugzilla/show_bug.cgi?id=29360
          + GtkAda3
          + GtkAda
          + xmlada
  * /usr/include/c++/12/bits/stl_vector.h:1230
    https://sourceware.org/bugzilla/show_bug.cgi?id=29689
      o ogre
  * ../../../libabigail/src/abg-ir.cc:26139
    https://sourceware.org/bugzilla/show_bug.cgi?id=29691
      o mingw32-gcc-gfortran
      o xorg-x11-server-Xephyr
  * ../../../libabigail/src/abg-ir.cc:8712
    https://sourceware.org/bugzilla/show_bug.cgi?id=29340
      o matreshka-fastcgi
      o matreshka
  * ../../../libabigail/src/abg-ir-priv.h:687
    https://sourceware.org/bugzilla/show_bug.cgi?id=29672
      o arm-none-eabi-gcc-cs-c++
      o arm-none-eabi-gcc-cs
      o gcc
      o kdelibs
      o llvm13
      o llvm14
      o llvm
  * ../../../libabigail/src/abg-reader.cc:3400 There are a lot of ocaml
    ones that trip this assert. The ones below are not obviously ocaml
    but they hit the same assert. I tend to think that they may be a
    different problem.
      o alt-ergo-gui
      o coccinelle
      o coq-core
      o frama-c
      o guestfs-tools
      o haxe
      o hevea
      o opam-installer
      o opam
      o planets
      o supermin
      o virt-v2v
      o why3
      o zenon
      o ocaml related https://sourceware.org/bugzilla/show_bug.cgi?id=29610
          + ocaml-atd
          + ocaml-biniou-devel
          + ocaml-biniou
          + ocaml-bisect-ppx
          + ocaml-camlp5-devel
          + ocaml-camlp5
          + ocaml-cil-cilly
          + ocaml-cinaps
          + ocaml-cppo
          + ocaml-cudf
          + ocaml-dune
          + ocaml-gettext-camomile
          + ocaml-gettext-devel
          + ocaml-gettext
          + ocaml-lablgtk3
          + ocaml-lwt-ppx
          + ocaml-mdx
          + ocaml-menhir
          + ocaml-merlin
          + ocamlmod
          + ocaml-ocamlbuild
          + ocaml-ocamldoc
          + ocaml-ocamlgraph-tools
          + ocaml-ocamlnet
          + ocaml-ocp-indent
          + ocaml-octavius
          + ocaml-odoc
          + ocaml-omake
          + ocaml
          + ocaml-ppx-assert
          + ocaml-ppx-base
          + ocaml-ppx-cold
          + ocaml-ppx-custom-printf
          + ocaml-ppx-deriving
          + ocaml-ppx-expect
          + ocaml-ppx-here
          + ocaml-ppx-import
          + ocaml-ppx-inline-test
          + ocaml-ppx-js-style
          + ocaml-ppx-let
          + ocaml-ppxlib
          + ocaml-psmt2-frontend
          + ocaml-qtest
          + ocaml-sedlex
          + ocaml-tplib-tools
          + ocaml-tyxml-jsx
          + ocaml-tyxml-ppx
          + ocaml-xmlm
          + ocaml-yojson


^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2022-10-14 21:53 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-10-14 21:53 Where are we with libabigail 2.1.1 Ben Woodard

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).