Re: Crash of Frama-C IDE with OCaml 4.02

yoann padioleau <[email protected]> Tue, 16 Sep 2014 09:22:51 -0700
Newsgroups gmane.comp.lang.ocaml.lib.gtk
Message-ID <CAFksq2_Smfd---xR2AhcARfpF74CTwx0gjSq2Pu1SWdbjhKrww@mail.gmail.com>
--===============5081138321052491565==
Content-Type: multipart/alternative; boundary=e89a8ff1ca5cb5286d0503312969

--e89a8ff1ca5cb5286d0503312969
Content-Type: text/plain; charset=UTF-8

Does the new release also fix this error on macos 10.9:

(codemap:72639): GLib-CRITICAL **: Source ID 17 was not found when
attempting to remove it

I get hundreds of those errors with codemap, and only with recent macos
versions.

https://github.com/facebook/pfff/wiki/CodeMap

)

On Mon, Sep 15, 2014 at 11:34 PM, Jacques Garrigue <
[email protected]> wrote:

> On 2014/09/16 02:36, Adrien Nader wrote:
>
> > On Mon, Sep 15, 2014, Boris Yakobowski wrote:
> >> Hi Adrien,
> >>
> >> Thanks for you answer.
> >>
> >> We also noticed those commits, and surmised that they were probably
> >> related. What I do not understand is, what change in OCaml 4.02
> >> transforms the bug from a silent one into a crash ?
> >
> > I can't really tell. As was mentionned on the caml-list, 4.02 has a
> > number of optimizations and different behaviours that might trigger bugs
> > more easily.
> >
> > I wasn't able to look at the backtrace since it requires a log in but I
> > can also imagine that repeated API abuse through the use of floats while
> > believing doubles are at use could trigger a crash.
> >
> > As for a release, I don't see anything against having one. Not many
> > changes but almost one year has passed since the last one. It's
> > typically Jacques how's doing them so it's better to leave the last word
> > to him probably. Nice thing is that with few changes it's easier to
> > test.
>
> Since the crash in Frama-C is a serious problem, I can do a release this
> week.
> Does anyone have some changes waiting?
>
>         Jacques
> _______________________________________________
> Lablgtk-list mailing list
> [email protected]
> https://lists.ocamlcore.org/cgi-bin/listinfo/lablgtk-list
>

--e89a8ff1ca5cb5286d0503312969
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>Does the new release also fix this error on macos 10.=
9:</div><div><br></div><div>(codemap:72639): GLib-CRITICAL **: Source ID 17=
 was not found when attempting to remove it</div><div><br></div><div>I get =
hundreds of those errors with codemap, and only with recent macos versions.=
</div><div><br></div><div><a href=3D"https://github.com/facebook/pfff/wiki/=
CodeMap">https://github.com/facebook/pfff/wiki/CodeMap</a><br></div><div><b=
r></div><div>)</div></div><div class=3D"gmail_extra"><br><div class=3D"gmai=
l_quote">On Mon, Sep 15, 2014 at 11:34 PM, Jacques Garrigue <span dir=3D"lt=
r">&lt;<a href=3D"mailto:[email protected]" target=3D"_blank">ga=
[email protected]</a>&gt;</span> wrote:<br><blockquote class=3D"gm=
ail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-le=
ft:1ex"><span class=3D"">On 2014/09/16 02:36, Adrien Nader wrote:<br>
<br>
&gt; On Mon, Sep 15, 2014, Boris Yakobowski wrote:<br>
&gt;&gt; Hi Adrien,<br>
&gt;&gt;<br>
&gt;&gt; Thanks for you answer.<br>
&gt;&gt;<br>
&gt;&gt; We also noticed those commits, and surmised that they were probabl=
y<br>
&gt;&gt; related. What I do not understand is, what change in OCaml 4.02<br=
>
&gt;&gt; transforms the bug from a silent one into a crash ?<br>
&gt;<br>
&gt; I can&#39;t really tell. As was mentionned on the caml-list, 4.02 has =
a<br>
&gt; number of optimizations and different behaviours that might trigger bu=
gs<br>
&gt; more easily.<br>
&gt;<br>
&gt; I wasn&#39;t able to look at the backtrace since it requires a log in =
but I<br>
&gt; can also imagine that repeated API abuse through the use of floats whi=
le<br>
&gt; believing doubles are at use could trigger a crash.<br>
&gt;<br>
&gt; As for a release, I don&#39;t see anything against having one. Not man=
y<br>
&gt; changes but almost one year has passed since the last one. It&#39;s<br=
>
&gt; typically Jacques how&#39;s doing them so it&#39;s better to leave the=
 last word<br>
&gt; to him probably. Nice thing is that with few changes it&#39;s easier t=
o<br>
&gt; test.<br>
<br>
</span>Since the crash in Frama-C is a serious problem, I can do a release =
this week.<br>
Does anyone have some changes waiting?<br>
<span class=3D"HOEnZb"><font color=3D"#888888"><br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 Jacques<br>
</font></span><div class=3D"HOEnZb"><div class=3D"h5">_____________________=
__________________________<br>
Lablgtk-list mailing list<br>
<a href=3D"mailto:[email protected]">Lablgtk-list@list=
s.forge.ocamlcore.org</a><br>
<a href=3D"https://lists.ocamlcore.org/cgi-bin/listinfo/lablgtk-list" targe=
t=3D"_blank">https://lists.ocamlcore.org/cgi-bin/listinfo/lablgtk-list</a><=
br>
</div></div></blockquote></div><br></div>

--e89a8ff1ca5cb5286d0503312969--

--===============5081138321052491565==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
Lablgtk-list mailing list
[email protected]
https://lists.ocamlcore.org/cgi-bin/listinfo/lablgtk-list

--===============5081138321052491565==--