Re: toolchain/46827: Cannot build the kernel, even Generic.

[Rifat Mahmud <rftmhmd%gmail.com@localhost>]

The following reply was made to PR toolchain/46827; it has been noted by GNATS.

From: Rifat Mahmud <rftmhmd%gmail.com@localhost>
To: gnats-bugs%netbsd.org@localhost
Cc: 
Subject: Re: toolchain/46827: Cannot build the kernel, even Generic.
Date: Fri, 24 Aug 2012 14:18:05 +0600

 --047d7b603c129107ec04c7fe9dc6
 Content-Type: text/plain; charset=ISO-8859-1
 
 > Does it fail if you rerun the command outside of the build system?
 Yes. That's how I debugged it. I gave the the backtrace result in the 7th
 mail. Here it is again.
 
 On Fri, Aug 24, 2012 at 1:20 PM, David Laight <david%l8s.co.uk@localhost> 
wrote:
 
 > The following reply was made to PR toolchain/46827; it has been noted by
 > GNATS.
 >
 > From: David Laight <david%l8s.co.uk@localhost>
 > To: gnats-bugs%NetBSD.org@localhost
 > Cc:
 > Subject: Re: toolchain/46827: Cannot build the kernel, even Generic.
 > Date: Fri, 24 Aug 2012 08:08:55 +0100
 >
 >  On Thu, Aug 23, 2012 at 11:50:01AM +0000, rftmhmd%gmail.com@localhost wrote:
 >  > >Synopsis:       Cannot build the kernel, even Generic.
 >  ...
 >  > Then run '/build.sh -m i386 kernel=GENERIC'
 >  >
 >  > It'll not end successfully and gives this error(involving libcompat):
 > http://pastebin.com/9hsmR1qd
 >  >
 >  > I reproduced this around 15 time.
 >
 >  Does it fail if you rerun the command outside of the build system?
 >  (The full command line is in the debug info.)
 >  Debugging will be easier if it does.
 >  Try reducing the number of input files on the command line - the
 >  bug might be caused by one specific file, or by the number of files.
 >
 >  Run 'ulimit -c unlimited' in the shell before running the command
 >  so that the 'abort' generates a core dump.
 >
 >  run 'gdb full_path_to_ar core_file' then 'bt' to get a back trace.
 >  The arguments to sprintf() are likely to be useful - the format
 >  string might identify the faulting call, and the associated
 >  arguments why it goes wrong.
 >
 >         David
 >
 >  --
 >  David Laight: david%l8s.co.uk@localhost
 >
 >
 
 --047d7b603c129107ec04c7fe9dc6
 Content-Type: text/html; charset=ISO-8859-1
 Content-Transfer-Encoding: quoted-printable
 
 <span style=3D"color:rgb(34,34,34);font-family:arial,sans-serif;font-size:1=
 3px;background-color:rgb(255,255,255)">&gt; Does it fail if you rerun the c=
 ommand outside of the build system?</span><br>Yes. That&#39;s how I debugge=
 d it. I gave the the backtrace result in the 7th mail. Here it is again.<di=
 v>
 <br><div class=3D"gmail_quote">On Fri, Aug 24, 2012 at 1:20 PM, David Laigh=
 t <span dir=3D"ltr">&lt;<a href=3D"mailto:david%l8s.co.uk@localhost"; 
target=3D"_blank=
 ">david%l8s.co.uk@localhost</a>&gt;</span> wrote:<br><blockquote 
class=3D"gmail_quote=
 " style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
 <div class=3D"im">The following reply was made to PR toolchain/46827; it ha=
 s been noted by GNATS.<br>
 <br>
 </div>From: David Laight &lt;<a 
href=3D"mailto:david%l8s.co.uk@localhost";>david@l8s.c=
 o.uk</a>&gt;<br>
 To: gnats-bugs%NetBSD.org@localhost<br>
 <div class=3D"im">Cc:<br>
 Subject: Re: toolchain/46827: Cannot build the kernel, even Generic.<br>
 </div>Date: Fri, 24 Aug 2012 08:08:55 +0100<br>
 <br>
 =A0On Thu, Aug 23, 2012 at 11:50:01AM +0000, <a href=3D"mailto:rftmhmd@gmai=
 l.com">rftmhmd%gmail.com@localhost</a> wrote:<br>
 =A0&gt; &gt;Synopsis: =A0 =A0 =A0 Cannot build the kernel, even Generic.<br=
 >
 =A0...<br>
 <div class=3D"im">=A0&gt; Then run &#39;/build.sh -m i386 kernel=3DGENERIC&=
 #39;<br>
 =A0&gt;<br>
 </div>=A0&gt; It&#39;ll not end successfully and gives this error(involving=
  libcompat):<a href=3D"http://pastebin.com/9hsmR1qd"; target=3D"_blank">http=
 ://pastebin.com/9hsmR1qd</a><br>
 =A0&gt;<br>
 =A0&gt; I reproduced this around 15 time.<br>
 <br>
 =A0Does it fail if you rerun the command outside of the build system?<br>
 =A0(The full command line is in the debug info.)<br>
 =A0Debugging will be easier if it does.<br>
 =A0Try reducing the number of input files on the command line - the<br>
 =A0bug might be caused by one specific file, or by the number of files.<br>
 <br>
 =A0Run &#39;ulimit -c unlimited&#39; in the shell before running the comman=
 d<br>
 =A0so that the &#39;abort&#39; generates a core dump.<br>
 <br>
 =A0run &#39;gdb full_path_to_ar core_file&#39; then &#39;bt&#39; to get a b=
 ack trace.<br>
 =A0The arguments to sprintf() are likely to be useful - the format<br>
 =A0string might identify the faulting call, and the associated<br>
 =A0arguments why it goes wrong.<br>
 <br>
 =A0 =A0 =A0 =A0 David<br>
 <br>
 =A0--<br>
 =A0David Laight: <a 
href=3D"mailto:david%l8s.co.uk@localhost";>david%l8s.co.uk@localhost</a><br>
 <br>
 </blockquote></div><br></div>
 
 --047d7b603c129107ec04c7fe9dc6--