mac-on-linux-devel — Development discussion

[Mac-on-linux-devel] [PATCH 1/3] Compile fix for snprintf

[Alexander G. <ag...@su...>]

While building the linker couldn't find snprintf in the keyboard map
application. So let's just use sprintf - should work out too.

Signed-off-by: Alexander Graf <ag...@su...>
---
 src/molelf/keyremap.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/src/molelf/keyremap.c b/src/molelf/keyremap.c
index 2db3b95..325de9d 100644
--- a/src/molelf/keyremap.c
+++ b/src/molelf/keyremap.c
@@ -507,7 +507,7 @@ print_menu_( mtab_t *t, int offs, int active )
 				modified = 1;
 		}
 		tc.h = offs;
-		snprintf( buf, 200, "%s  %-20s <0>%s \n", 
+		sprintf( buf, "%s  %-20s <0>%s \n",
 			 (t->cur == i)? active ? "<5>" : "<4>"  : "<3>", m->text, 
 			 modified ? "<0>  Modified" : "" );
 		draw_str( buf );
-- 
1.6.0.2

[Mac-on-linux-devel] [PATCH 2/3] Make CPU virtualization mode selectable

[Alexander G. <ag...@su...>]

We want to give the user the choice between KVM and MOL kernel modules,
so let's make the config option selectable and fix all references to
it to actually reference to the config option.

Signed-off-by: Alexander Graf <ag...@su...>
---
 config/Kconfig-common |   11 ++++++++---
 src/Makefile          |    2 +-
 src/cpu/Makefile      |    4 ++--
 3 files changed, 11 insertions(+), 6 deletions(-)

diff --git a/config/Kconfig-common b/config/Kconfig-common
index 8944dba..0993b1e 100644
--- a/config/Kconfig-common
+++ b/config/Kconfig-common
@@ -1,11 +1,16 @@
 mainmenu "Mac-on-Linux Configuration"
 
 menu "Machine Specific Build Targets"
-config PPC
-	bool
-	default y
+
+choice
+	prompt "Virtualization type"
+	default PPC
+
+config MOL
+	bool "MOL kernel module"
 	help
 	  No help yet.
+endchoice
 
 #config MPC107
 #	bool "MPC107 board (Crescendo)"
diff --git a/src/Makefile b/src/Makefile
index c5bad18..2733edb 100644
--- a/src/Makefile
+++ b/src/Makefile
@@ -1,7 +1,7 @@
 
 include			../config/Makefile.top
 
-BUILD_MODS		:= y
+BUILD_MODS		:= $(if $(CONFIG_MOL),y)
 NETMODS			:= $(CONFIG_TUN)$(CONFIG_SHEEP)
 NETDRIVER		:= $(if $(LINUX),$(if $(NETMODS),$(BUILD_MODS)))
 
diff --git a/src/cpu/Makefile b/src/cpu/Makefile
index a4bc0dd..3251973 100644
--- a/src/cpu/Makefile
+++ b/src/cpu/Makefile
@@ -2,7 +2,7 @@
 include			../../config/Makefile.top
 
 SUBDIRS-$(MPC107)	= mpc107
-SUBDIRS-$(PPC)		= ppc
+SUBDIRS-$(CONFIG_MOL)	= ppc
 SUBDIRS-$(X86)		= i386
 
 XTARGETS		= cpu
@@ -10,7 +10,7 @@ XTARGETS		= cpu
 
 cpu-OBJS		= mainloop.o molcpu.o $(obj-y)
 obj-$(CONFIG_DEBUGGER)	+= breakpoints.o
-obj-$(PPC)		+= ppc/libppc.a
+obj-$(CONFIG_MOL)	+= ppc/libppc.a
 obj-$(X86)		+= i386/libi386.a
 obj-$(MPC107)		+= ppc/libmpc107.a
 
-- 
1.6.0.2

Re: [Mac-on-linux-devel] mol and linux 2.6.31

[didier <di...@gm...>]

On Sun, Nov 15, 2009 at 5:01 AM, didier <di...@gm...> wrote:
> Hi
>
> As anyone MOL running *well* with 2.6.31 kernel?
>
Ok replying to myself.
I updated svn HEAD with a fix...

Didier

[Mac-on-linux-devel] mol and linux 2.6.31

[didier <di...@gm...>]

Hi

As anyone MOL running *well* with 2.6.31 kernel?

Here sometime it enters in an unkillable loop in kernel mode.
hide details 2:42 AM (2 hours ago)
	
Hi

As anyone MOL running *well* with 2.6.31 kernel?

Here sometime it enters in an unkillable loop in kernel mode.

Re: [Mac-on-linux-devel] Patch for MOL compilation on PowerPC system with kernels >= 2.6.26

[Marco B. <mar...@sl...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Thomas Juerges wrote:
> attached is a patch for vanilla kernels >= 2.6.26 which allows MOL to
> compile on PowerPC systems with those kernels installed.
THANKS!
As soon I'll fix my powerpc machine I'll probably add it to slackintosh.

- --
Marco Bonetti
BT3 EeePC 70x enhancing module: http://sid77.slackware.it/bt3/
Slackintosh Linux Project Developer: http://workaround.ch/
Linux-live for powerpc: http://workaround.ch/pub/rsync/mb/linux-live/
My webstuff: http://sidbox.homelinux.org/

My GnuPG key id: 0x86A91047
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFJaNFnyPKw+YapEEcRAoO9AJ4mIvvHs9Xp3ZPhfmtBZb42KasLagCbBR3J
+32GrCdphWcg/Lb62DwHKjw=
=9eCc
-----END PGP SIGNATURE-----

[Mac-on-linux-devel] Patch for MOL compilation on PowerPC system with kernels >= 2.6.26

[Thomas J. <th...@so...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

attached is a patch for vanilla kernels >= 2.6.26 which allows MOL to
compile on PowerPC systems with those kernels installed.  Before it did
not compile because the kernel maintainers decided to rename the
arch/ppc/include/ directory to arch/powerpc/include.

Question: any chances that MOL gets rid of the handle_mm_fault problem
without patching the kernel?

Cheers,
	Thomas
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.9 (GNU/Linux)

iEYEARECAAYFAklmHNQACgkQMiyRSD9s+zVcXACfa6cU/JSFes3ydZI9VfdggHdA
B8QAn2N2KMz2NgHL2PXJb3C3a387Vk5D
=wmoD
-----END PGP SIGNATURE-----

Re: [Mac-on-linux-devel] [PATCH] add partial backing store support to qcow drive format.

[Joseph J. <jo...@ge...>]

didier wrote:
> Hi,
>
> Caveats:
> - only absolute pathname for the backing store file.
> - only 1023 pathname length (I don't remember why this one).
>
> Didier
>   
Hey,

Thanks for all of the patches!  I haven't been very active on MOL
development lately due to my laptop being broken, but that's resolved
now, so I have no more excuses. ;)  I'll review the patches and get the
repo updated this weekend.

If anyone is interested in helping to maintain and develop MOL, I'd be
happy to add you as a developer to the project.

Thanks,
-Joe

[Mac-on-linux-devel] [PATCH] add partial backing store support to qcow drive format.

[didier <di...@gm...>]

Hi,

Caveats:
- only absolute pathname for the backing store file.
- only 1023 pathname length (I don't remember why this one).

Didier

[Mac-on-linux-devel] [PATCH] replace BIT with MOL_BIT.

[didier <di...@gm...>]

Hi,
There's a conflicting define for BIT in recent kernels and mol.

This patch extends the patch from sc...@su... (cf:
http://bb.zhang.free.fr/suse-mol-patch-for-2.6.25/suse-ppc32-mol-BIT)
and replaces all BIT occurrences in SVN code with MOL_BIT.

With this patch mol SVN compiles and works with  Linux 2.6.28, at least for me.

Didier

[Mac-on-linux-devel] [PATCH] add the session ID to the MAC tun address.

[didier <di...@gm...>]

Hi,
Some OSX code doesn't like if there's two Macs with the same ethernet
address on the network (like UUID generation for AFP 3.2).

This patch add  g_session_id to the MAC.


Didier

[Mac-on-linux-devel] [PATCH] missing a %s in a src/main/linux/misc.c printf call.

[didier <di...@gm...>]

Hi,
trivial patch attached

Didier

[Mac-on-linux-devel] init.S assembly macros

[Leslie V. <les...@gm...>]

Hello!

I see in another thread which I cannot reply to, Joe speaks of cleaning up
the assembly code
(
http://sourceforge.net/mailarchive/forum.php?thread_name=48BAF655.3040905%40sokrates.homelinux.net&forum_name=mac-on-linux-devel
).

Playing around with it, I can see that the problem is the multi-line macros
that the GNU assembler
does not appear to support. Since there's not much assembly I was thinking
of just unrolling all the
macros into normal code.

But first:

1. Is there some way that GAS can support multi-line macros?
2. Does m4 interpret and expand the #define's?
3. If so, can some trick be done in asm.m4 to support them?

If those tricks won't work, would I accomplish anything by
unrolling the macros? I don't know enough about PPC assembly
to contribute much more.

Les

Re: [Mac-on-linux-devel] Status of mol-kernel-merge

[Joseph J. <jo...@ge...>]

Thomas Juerges wrote:
> Hi folks,
>
> what is the current status of the mol-kernel-merge?  Does anybody work
> on this?
>
> I checked the trunk version out and gave it a try.  Compilation of the
> C-files is ok, but assembler files (as said in the Makefile) fail -
> after I got the Makefile going.  One suggestion:  how about collecting
> all the macros in one file and not have them cluttered all over the place?
>
>
> Cheers,
>     Thomas

Hi Thomas!

As you can tell, there hasn't been a lot of work done on MOL lately, 
mostly due to my hectic schedule.  Cleaning up of the assembly is my 
current task, but I haven't finished much more than what is in the SVN.  
If you'd like to help clean it up, I'd be happy to add you to the project.

Thanks,
-Joe

[Mac-on-linux-devel] Status of mol-kernel-merge

[Thomas J. <th...@so...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi folks,

what is the current status of the mol-kernel-merge?  Does anybody work
on this?

I checked the trunk version out and gave it a try.  Compilation of the
C-files is ok, but assembler files (as said in the Makefile) fail -
after I got the Makefile going.  One suggestion:  how about collecting
all the macros in one file and not have them cluttered all over the place?


Cheers,
	Thomas
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.6 (GNU/Linux)

iD8DBQFIuvZVMiyRSD9s+zURAhF3AJ45UfYK9OP0RuPUY/vv9vTvzU4f5wCfXSme
VL9wfbIef7ViNc0KUkGxMFE=
=yEaW
-----END PGP SIGNATURE-----

Re: [Mac-on-linux-devel] Taking a look at the svn: how is latest kernel support?

[Bin Z. <yan...@gm...>]

On Jan 10, 2008 9:39 PM, Joseph Jezak <jo...@ge...> wrote:
> Marco Bonetti wrote:
> > hi guys,
> > I'm browsing sf.net svn repositories, reading the changelog at
> > http://mac-on-linux.svn.sourceforge.net/viewvc/mac-on-linux/trunk/CHANGELOG?revision=142&view=markup
> > it says "* Build fixes for newer kernels".
> >
> > Is it working?
> >
> > if so I'm going to add a svn version of mol to slackintosh-current (which
> > is running on 2.6.23.12) otherwise I'll have to remove mol from the mac/
> > set (but I'll probably put 0.9.72.1 in extra/ with an older kernel, is
> > just a too important package to be left out).
> > Thanks for any input and for your hard work.
> >
> > ciao
> >
> >
> No, it's still not working.  You can get it to build against 2.6.23+
> with the patches in SVN, along with this patch to the kernel (which I
> should really send upstream):
> http://dev.gentoo.org/~josejx/mol-mm.patch
>
> However, it locks up or segfaults on me with this configuration and I'm
> not sure what changes in 2.6.23 broke it.  I haven't had time to do a
> proper bisection, hopefully in the next few weeks.
>

On my ibook G4,  it locks up only for 2.6.24-rc6.
mol 0.9.72.1 works for kernel 2.6.23 (with
EXPORT_SYMBOL_GPL(handle_mm_fault);) with this patch

diff -Nur mol.orig/kmod/mol-ioctl.h mol/kmod/mol-ioctl.h
--- mol.orig/kmod/mol-ioctl.h	2007-08-07 16:34:50.000000000 +0200
+++ mol/kmod/mol-ioctl.h	2007-12-20 21:18:06.000000000 +0100
@@ -17,6 +17,8 @@
 #ifndef _H_MOL_IOCTL
 #define _H_MOL_IOCTL

+#include <asm/ioctl.h>
+
 #ifndef __ASSEMBLY__
 #include "mmutypes.h"


Best regards,
Bin

> -Joe
>
>
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
> _______________________________________________
> Mac-on-linux-devel mailing list
> Mac...@li...
> https://lists.sourceforge.net/lists/listinfo/mac-on-linux-devel
>

Re: [Mac-on-linux-devel] Taking a look at the svn: how is latest kernel support?

[Joseph J. <jo...@ge...>]

Marco Bonetti wrote:
> hi guys,
> I'm browsing sf.net svn repositories, reading the changelog at
> http://mac-on-linux.svn.sourceforge.net/viewvc/mac-on-linux/trunk/CHANGELOG?revision=142&view=markup
> it says "* Build fixes for newer kernels".
>
> Is it working?
>
> if so I'm going to add a svn version of mol to slackintosh-current (which
> is running on 2.6.23.12) otherwise I'll have to remove mol from the mac/
> set (but I'll probably put 0.9.72.1 in extra/ with an older kernel, is
> just a too important package to be left out).
> Thanks for any input and for your hard work.
>
> ciao
>
>   
No, it's still not working.  You can get it to build against 2.6.23+
with the patches in SVN, along with this patch to the kernel (which I
should really send upstream):
http://dev.gentoo.org/~josejx/mol-mm.patch

However, it locks up or segfaults on me with this configuration and I'm
not sure what changes in 2.6.23 broke it.  I haven't had time to do a
proper bisection, hopefully in the next few weeks.

-Joe

[Mac-on-linux-devel] Taking a look at the svn: how is latest kernel support?

[Marco B. <mar...@sl...>]

hi guys,
I'm browsing sf.net svn repositories, reading the changelog at
http://mac-on-linux.svn.sourceforge.net/viewvc/mac-on-linux/trunk/CHANGELOG?revision=142&view=markup
it says "* Build fixes for newer kernels".

Is it working?

if so I'm going to add a svn version of mol to slackintosh-current (which
is running on 2.6.23.12) otherwise I'll have to remove mol from the mac/
set (but I'll probably put 0.9.72.1 in extra/ with an older kernel, is
just a too important package to be left out).
Thanks for any input and for your hard work.

ciao

-- 
Marco Bonetti
Slackintosh Linux Project Developer: http://workaround.ch/
Linux-live for powerpc: http://workaround.ch/pub/rsync/mb/linux-live
My webstuff: http://sidbox.homelinux.org

My GnuPG key id: 0x86A91047

Re: [Mac-on-linux-devel] MOL on MacIntel?

[Chadwick <gm...@gm...>]

On Dec 23, 2007 8:32 PM, Nathan Smith <nda...@gm...> wrote:
> On Sat, 2007-12-22 at 18:35 -0800, John Wright wrote:
> > Is there any possibility that MOL will be ported to
> > Intel Macs?  I had to go MacIntel when my PPC Mac
> > Mini died;  I like Fedora 8 on the Intel mini well
> > enough, but I sure miss being able to run the
> > occasional Mac app without re-booting!
> >
> > Is there anything that I am not aware of that makes
> > this difficult?  Like, Apple munging with the Intel
> > MacOS to prevent its running on non-Apple hardware,
> > that might make it difficult to boot from Linux?
> >
> >         -jmw-
> >
> >  Dr. John M. Wright, Dept. of Chemistry & Biochemistry, M/S 0314, UCSD,
> >  La Jolla, CA 92093-0314; email: jwright(at)ucsd(dot)edu; phone: 858-534-3049
> >
> > ()  ascii ribbon campaign - against html mail
> > /\                        - against microsoft attachments
> >
>
> There are no currently plans to port MOL to x86 due to the scope of the
> task and the limited resources (namely time) of the MOL development
> team.  As I understand it, MOL relies on some ppc-specific features, so
> a port to x86 would require a major effort.
>
> Though perhaps the increasing demand for such a product will eventually
> lead to a port (or Mac OS X virtualization options from other
> virtualization software suites).  However, there may be legal problems
> such software, since Apple's license does not allow for OS X to be run
> on non-Apple hardware, which a port of MOL would facilitate (though I
> read that 10.5 server may allow virtualization on non-Apple hardware).
>
> --
> Nathan Smith
> nda...@gm...
>

Maybe another interesting point:  it is difficult to find working
replacements in the PPC realm of Linux.. but usually there is a
``universal'' or separate PPC binary for osx.  Many things not
perfectly available in PPC Linux seem to be available for x86 Linux..
so I wonder if macbookpro counts.. it should, right?

For me, I have to run mol on my pb g4 for Finale, File Maker, First
Class, and good working flash with sound (homestarrunner!)  Since mol
does not help me with Logic or ProTools, I have to reboot to use
those.. which I would have to do with a macbookpro, too.  I just
wouldn't have to reboot for FM or First class.. or flash.  And I'm
sure I can be happy with Rosegarden (or denemo, or noteedit, or
learning musixtex) instead of Finale.

-- 
Chadwick

Re: [Mac-on-linux-devel] MOL on MacIntel?

[Nathan S. <nda...@gm...>]

On Sat, 2007-12-22 at 18:35 -0800, John Wright wrote: 
> Is there any possibility that MOL will be ported to
> Intel Macs?  I had to go MacIntel when my PPC Mac
> Mini died;  I like Fedora 8 on the Intel mini well
> enough, but I sure miss being able to run the
> occasional Mac app without re-booting!
> 
> Is there anything that I am not aware of that makes
> this difficult?  Like, Apple munging with the Intel
> MacOS to prevent its running on non-Apple hardware,
> that might make it difficult to boot from Linux?
> 
>         -jmw-
> 
>  Dr. John M. Wright, Dept. of Chemistry & Biochemistry, M/S 0314, UCSD,
>  La Jolla, CA 92093-0314; email: jwright(at)ucsd(dot)edu; phone: 858-534-3049
> 
> ()  ascii ribbon campaign - against html mail
> /\                        - against microsoft attachments
> 

There are no currently plans to port MOL to x86 due to the scope of the
task and the limited resources (namely time) of the MOL development
team.  As I understand it, MOL relies on some ppc-specific features, so
a port to x86 would require a major effort.

Though perhaps the increasing demand for such a product will eventually
lead to a port (or Mac OS X virtualization options from other
virtualization software suites).  However, there may be legal problems
such software, since Apple's license does not allow for OS X to be run
on non-Apple hardware, which a port of MOL would facilitate (though I
read that 10.5 server may allow virtualization on non-Apple hardware).

-- 
Nathan Smith
nda...@gm...

[Mac-on-linux-devel] MOL on MacIntel?

[John W. <jwr...@sa...>]

Is there any possibility that MOL will be ported to
Intel Macs?  I had to go MacIntel when my PPC Mac
Mini died;  I like Fedora 8 on the Intel mini well
enough, but I sure miss being able to run the
occasional Mac app without re-booting!

Is there anything that I am not aware of that makes
this difficult?  Like, Apple munging with the Intel
MacOS to prevent its running on non-Apple hardware,
that might make it difficult to boot from Linux?

        -jmw-

 Dr. John M. Wright, Dept. of Chemistry & Biochemistry, M/S 0314, UCSD,
 La Jolla, CA 92093-0314; email: jwright(at)ucsd(dot)edu; phone: 858-534-3049

()  ascii ribbon campaign - against html mail
/\                        - against microsoft attachments

Re: [Mac-on-linux-devel] Build system help

[Joseph J. <jo...@ge...>]

>>> Ist MOL's BootX related to Mac OS X's / Darwin's BootX? Darwin's
>>> BootX builds just fine ...
>>
>> Yes and no, it's certainly based on OSX's and Darwin's BootX, but
>> it's modified a bit.  The issue with MOL's version of BootX is that
>> it needs to build a static ELF version.
> 
> A Mac's firmware can't load mach-o either. Darwin's BootX features an
> mach-o to XCOFF converter. Would this be helpful?

Of course, but we're not using the firmware to load Mach-O, we're
using MOL's ELF loader.  We either have to find a way to build ELF
of OSX, add another loader to MOL, or simply just not build BootX on
OSX and bundle a precompiled copy for OSX users.

-Joe

Re: [Mac-on-linux-devel] Build system help

[Markus H. <ma...@ju...>]

Am 29.10.2007 um 00:19 schrieb Joseph Jezak:

> Markus Hitter wrote:
>
>> How would I make the full command line visible which is used to
>> compile molaccel.S?
>
> Type make in the directory containing molaccel.S.

Surprise surprise, this works indeed, thanks.

>> Ist MOL's BootX related to Mac OS X's / Darwin's BootX? Darwin's
>> BootX builds just fine ...
>
> Yes and no, it's certainly based on OSX's and Darwin's BootX, but
> it's modified a bit.  The issue with MOL's version of BootX is that
> it needs to build a static ELF version.

A Mac's firmware can't load mach-o either. Darwin's BootX features an  
mach-o to XCOFF converter. Would this be helpful?


Markus

- - - - - - - - - - - - - - - - - - -
Dipl. Ing. Markus Hitter
http://www.jump-ing.de/

Re: [Mac-on-linux-devel] Build system help

[Joseph J. <jo...@ge...>]

Markus Hitter wrote:
> Hello all,
> 
> a few hours ago I fetched mol from SVN to build the thing on a  
> PowerMac G4 running Mac OS X 10.4.10 and Xcode 2.4.1. Now, as  
> expected, I get errors somewhere in bootx.
> 
> I'd like to fix them, but can't find out how to get more verbose  
> compiler messages.
> 
> So, my questions are:
> 
> How would I make the full command line visible which is used to  
> compile molaccel.S?

Type make in the directory containing molaccel.S.

> and

> Ist MOL's BootX related to Mac OS X's / Darwin's BootX? Darwin's  
> BootX builds just fine ...

Yes and no, it's certainly based on OSX's and Darwin's BootX, but
it's modified a bit.  The issue with MOL's version of BootX is that
it needs to build a static ELF version.  If you know how to do that
in Darwin/OSX, I'd appreciate it, otherwise we'll probably have to
just ship a pre-compiled version with OSX.  Another option would be
to build a MachO loader for MOL.  Unfortunately, I haven't spent any
time on getting things working in OSX since I last committed OSX
fixes. :|

If you need more details, please let me know!
-Joe

[Mac-on-linux-devel] Build system help

[Markus H. <ma...@ju...>]

Hello all,

a few hours ago I fetched mol from SVN to build the thing on a  
PowerMac G4 running Mac OS X 10.4.10 and Xcode 2.4.1. Now, as  
expected, I get errors somewhere in bootx.

I'd like to fix them, but can't find out how to get more verbose  
compiler messages.

So, my questions are:

How would I make the full command line visible which is used to  
compile molaccel.S?

and

Ist MOL's BootX related to Mac OS X's / Darwin's BootX? Darwin's  
BootX builds just fine ...


Thanks,
Markus

- - - - - - - - - - - - - - - - - - -
Dipl. Ing. Markus Hitter
http://www.jump-ing.de/

Re: [Mac-on-linux-devel] [Mac-on-linux-users] mol 0.7.22.1 and Kernel 2.6.23

[Christoph H. <hc...@ls...>]

On Sun, Jul 29, 2007 at 07:21:29PM -0400, Joseph Jezak wrote:
> Okay.  It's kind of fundamental for MOL.  MOL causes faults by 
> executing a bogus instruction to setup the mmu for guest execution. 
>  The handler is used if the requested page is not present. 
> Christoph, since you were the one that submitted the patch for 
> removal, would you mind if we re-added the export for handle_mm_fault?

Obviosuly I won't add the export back.  Having modules calling into
page fault handlers is not exactly what we want, and not exporting it
heps catching crap like that.