[Mono-dev] Cross-compile Mono 2.2 for Windows on Fedora 10 using mingw32

2009-01-26 Thread Mikkel Kruse Johnsen
Hi

I'm trying to cross-compile mono for windows, using Linux (Fedora 10,
cross-compile environment mingw32-*)

But I get an error running genmdesc.pl

---
Making all in mini
make[3]: Entering directory `/root/build/mono-2.2/mono/mini'
perl ./genmdesc.pl __i386__ . ./cpu-x86.md cpu-x86.h x86_desc
Running: i686-pc-mingw32-cpp -undef  -U__i386__ -U__x86_64__ -U__ppc__
-U__powerpc__ -U__arm__ -U__sparc__ -Usparc -U__s390__ -Us390 -U__ia64__
-U__alpha__ -U__mips__ -D__i386__ ./mini-ops.h|
Invalid opcode addps at line 412
make[3]: *** [cpu-x86.h] Error 9
make[3]: Leaving directory `/root/build/mono-2.2/mono/mini'



Could someone please help me out ?



-bash-3.2# i686-pc-mingw32-cpp --version
i686-pc-mingw32-cpp (GCC) 4.3.2
Copyright (C) 2008 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is
NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR
PURPOSE.


-bash-3.2# cat /etc/profile.d/mingw32.sh 
# Environment variables for MinGW.

alias mingw32-env='eval `rpm --eval %{_mingw32_env}`'

PREFIX=/usr/i686-pc-mingw32/sys-root
TARGET=mingw

export CC=i686-pc-mingw32-gcc
export CPP=i686-pc-mingw32-cpp
export CXXCPP=/lib/i686-pc-mingw32-cpp
export CXX=i686-pc-mingw32-g++
export AR=i686-pc-mingw32-ar
export RANLIB=i686-pc-mingw32-ranlib
export DLLTOOL=i686-pc-mingw32-dlltool
export OBJDUMP=i686-pc-mingw32-objdump

export CFLAGS=-O2 -march=i586 -mms-bitfields
-I/usr/lib/gcc/i686-pc-mingw32/4.3.2/include/
export CXXFLAGS=-O2 -march=i586 -mms-bitfields
export PKG_CONFIG_PATH=$PREFIX/$TARGET/lib/pkgconfig
export PATH=$PREFIX/bin:
$PREFIX/$TARGET/bin:/usr/libexec/gcc/i686-pc-mingw32/4.3.2:/bin:/usr/bin
export LD_LIBRARY_PATH=$PREFIX/$TARGET/lib
export LDFLAGS=-L$PREFIX/$TARGET/lib



-bash-3.2# rpm -qa | grep mingw
mingw32-runtime-3.15.1-10.fc10.noarch
mingw32-iconv-1.12-7.fc10.noarch
mingw32-termcap-1.3.1-5.fc10.noarch
mingw32-libjpeg-6b-6.fc10.noarch
mingw32-pango-1.22.1-1.fc10.noarch
mingw32-openssl-0.9.8g-1.fc10.noarch
mingw32-filesystem-40-3.fc10.noarch
mingw32-freetype-2.3.8-1.fc10.noarch
mingw32-sqlite-3.6.6.2-1.fc10.noarch
mingw32-binutils-2.18.50_20080109_2-10.fc10.i386
mingw32-gcc-c++-4.3.2-12.fc10.i386
mingw32-fontconfig-2.6.0-5.fc10.noarch
mingw32-atk-1.24.0-2.fc10.noarch
mingw32-jasper-1.900.1-5.fc10.noarch
mingw32-w32api-3.12-8.fc10.noarch
mingw32-gcc-4.3.2-12.fc10.i386
mingw32-expat-2.0.1-1.fc10.noarch
mingw32-libidn-1.9-1.fc10.noarch
mingw32-curl-7.18.2-2.fc10.noarch
mingw32-dbus-1.2.4-0.2.20081031svn.fc10.noarch
mingw32-pixman-0.13.2-2.fc10.noarch
mingw32-gettext-0.17-7.fc10.noarch
mingw32-libxml2-2.7.2-4.fc10.noarch
mingw32-cpp-4.3.2-12.fc10.i386
mingw32-zlib-1.2.3-11.fc10.noarch
mingw32-glib2-2.18.1-2.fc10.noarch
mingw32-gtk2-2.14.4-3.fc10.noarch
mingw32-libpng-1.2.34-2.fc10.noarch
mingw32-cairo-1.8.0-3.fc10.noarch
mingw32-libssh2-0.18-1.fc10.noarch
mingw32-libglade2-2.6.3-2.fc10.noarch


 Med Venlig Hilsen / Kind Regards


 
 
Mikkel Kruse
Johnsen
Adm.Dir.

Linet
Ørholmgade 6 st tv
Copenhagen N 2200
Denmark

Work:+45
21287793
Mobile: +45
21287793
Email:
mik...@linet.dk
IM:
mik...@linet.dk
(MSN)
 Professional
Profile
Healthcare 


Network
Consultant 
___
Mono-devel-list mailing list
Mono-devel-list@lists.ximian.com
http://lists.ximian.com/mailman/listinfo/mono-devel-list


[Mono-dev] Is Preview 1 of release 2.4 still coming out today?

2009-01-26 Thread John M

Hi,

I was wondering if the first preview of release 2.4 was due today and if it
includes the fixes that didn't make it into release 2.2?

Many Thanks,

John
-- 
View this message in context: 
http://www.nabble.com/Is-Preview-1-of-release-2.4-still-coming-out-today--tp21664229p21664229.html
Sent from the Mono - Dev mailing list archive at Nabble.com.

___
Mono-devel-list mailing list
Mono-devel-list@lists.ximian.com
http://lists.ximian.com/mailman/listinfo/mono-devel-list


[Mono-dev] Mono porting and splitted tests

2009-01-26 Thread Ian Dichkovsky
Hi,

We are now doing mono port to MIPS n32  64 architecture. Yes, we
aware of Mark Mason's efforts with the MIPS port :-)

During porting we came to the conclusion that single big binary test
sets (basic.exe, basic-long.exe, etc.) are inconvinient. On the
initial stage of porting, mono binary could crash in the middle of
testing and the rest of functions just hadn't a chance to run. These
crashes doesn't give us clear picture, which functions passed and
which failed. We splitted tests to different files, so each test
function has its own source and binary.

We think that it would be a good idea to add these splitted tests to
SVN. It will simplify future mono ports to the new platforms. Sources
available upon request.

Regards,
Ian
___
Mono-devel-list mailing list
Mono-devel-list@lists.ximian.com
http://lists.ximian.com/mailman/listinfo/mono-devel-list


Re: [Mono-dev] Mono porting and splitted tests

2009-01-26 Thread Ian Dichkovsky
Hi,

Yes, I got your point.
In our case it was useful to split several crashed tests among several
developers.
Also, it was easy to get number of tests passed/failed to know how
much was done and how much was left.

Regards,
Ian

2009/1/26 Zoltan Varga var...@gmail.com:
 Hi,

  I had done mono ports to many platforms, and I didn't find this to
 be a problem. You
 can run mono using
 mono -v --regression basic.exe
 and it will print out each method as it is JIT compiled, so you can
 see which one is causing
 a crash.

 Zoltan

 On Mon, Jan 26, 2009 at 1:23 PM, Ian Dichkovsky ian...@gmail.com wrote:
 Hi,

 We are now doing mono port to MIPS n32  64 architecture. Yes, we
 aware of Mark Mason's efforts with the MIPS port :-)

 During porting we came to the conclusion that single big binary test
 sets (basic.exe, basic-long.exe, etc.) are inconvinient. On the
 initial stage of porting, mono binary could crash in the middle of
 testing and the rest of functions just hadn't a chance to run. These
 crashes doesn't give us clear picture, which functions passed and
 which failed. We splitted tests to different files, so each test
 function has its own source and binary.

 We think that it would be a good idea to add these splitted tests to
 SVN. It will simplify future mono ports to the new platforms. Sources
 available upon request.

 Regards,
 Ian
 ___
 Mono-devel-list mailing list
 Mono-devel-list@lists.ximian.com
 http://lists.ximian.com/mailman/listinfo/mono-devel-list


___
Mono-devel-list mailing list
Mono-devel-list@lists.ximian.com
http://lists.ximian.com/mailman/listinfo/mono-devel-list


Re: [Mono-dev] Online Documentation Broken

2009-01-26 Thread Bojan Rajkovic
Miguel De Icaza wrote:
 Hello,

 This happens when we upgrade the system live without restarting 
 HTTPd.   Can you retry?

Works now, thanks much!
___
Mono-devel-list mailing list
Mono-devel-list@lists.ximian.com
http://lists.ximian.com/mailman/listinfo/mono-devel-list


[Mono-dev] [PATCH] Fix for Ping when pinging localhost

2009-01-26 Thread Christian Prochnow
Hi,

attached is a patch for
https://bugzilla.novell.com/show_bug.cgi?id=334453

When pinging localhost the Ping class also receives the ICMP_ECHO packet sent 
by itself. The patch modifies Ping to discard IcmpMessage's with Type==8.



Best regards,

Christian Prochnow


-- 
Christian Prochnow
Geschäftsführer

SecuLogiX Systems GmbH
Mohriner Allee 28
12347 Berlin

http://www.seculogix.de

mobile:  +49 (0)177 313 02 57
   fon:  +49 (0)700 SECULOGIX

Geschäftsführer: Christian Prochnow
Handelsregister: B 96491, Amtsgericht Charlottenburg

Index: class/System/System.Net.NetworkInformation/ChangeLog
===
--- class/System/System.Net.NetworkInformation/ChangeLog	(revision 124499)
+++ class/System/System.Net.NetworkInformation/ChangeLog	(working copy)
@@ -1,3 +1,8 @@
+2009-01-26  Christian Prochnow cpr...@seculogix.de
+
+	* Ping.cs: ignore ICMP echo request packet in case we ping
+	  localhost.
+
 2009-01-11  Atsushi Enomoto  atsu...@ximian.com
 
 	* NetworkInformationPermission.cs,
Index: class/System/System.Net.NetworkInformation/Ping.cs
===
--- class/System/System.Net.NetworkInformation/Ping.cs	(revision 124499)
+++ class/System/System.Net.NetworkInformation/Ping.cs	(working copy)
@@ -233,9 +233,11 @@
 		continue;
 
 	IcmpMessage recv = new IcmpMessage (bytes, headerLength, bodyLength);
-	if (recv.Identifier != identifier)
-		continue; // ping reply to different request. discard it.
 
+	/* discard ping reply to different request or echo requests if running on same host. */
+	if (recv.Identifier != identifier || recv.Type == 8)
+		continue; 
+
 	return new PingReply (address, recv.Data, options, rtt, recv.IPStatus);
 } while (true);
 			}
@@ -461,6 +463,8 @@
 		return IPStatus.ParameterProblem;
 	case 4:
 		return IPStatus.SourceQuench;
+	case 8:
+		return IPStatus.Success;
 	}
 	return IPStatus.Unknown;
 	//throw new NotSupportedException (String.Format (Unexpected pair of ICMP message type and code: type is {0} and code is {1}, Type, Code));
___
Mono-devel-list mailing list
Mono-devel-list@lists.ximian.com
http://lists.ximian.com/mailman/listinfo/mono-devel-list


[Mono-dev] Announcing Mono 2.4 Preview 1...

2009-01-26 Thread Thomas Wiest
Hey Everyone,

We've just released Mono 2.4 Preview 1 today!

Please help us out by giving it a try with your applications.

As always, you can get the preview/RC releases here:
http://mono.ximian.com/monobuild/preview/download-preview/

Please report any bugs that you may find using our Bugs page, AND reply to this 
thread with the bug numbers so we can track them:
http://www.mono-project.com/Bugs

You can see the bugs we're tracking for Mono 2.4 here:
https://bugzilla.novell.com/buglist.cgi?query_format=advancedclassification=Monotarget_milestone=2.4.xorder=bugs.bug_status%2Cbugs.bug_severity

The earlier you file the bugs and reply to this message, the more likely your 
bugs will get fixed.

Special attention is given to regressions, so if you can tell us a version of 
Mono where the bug worked and you add [Regression] to the beginning of the 
summary of the bug, then it is much more likely your bug will get fixed.

Please help the Mono team to make 2.4 the best ever.

Thanks!

Mono QA

___
Mono-devel-list mailing list
Mono-devel-list@lists.ximian.com
http://lists.ximian.com/mailman/listinfo/mono-devel-list


Re: [Mono-dev] Class status pages are empty

2009-01-26 Thread Miguel de Icaza
Hello,

   Do you mind filing a bug report, so we can get around updating these
files?

 Hi,
 
 Microsoft .NET Framework 2.0 Service Pack 2 finally was released as a 
 standalone update:
 http://www.microsoft.com/downloads/details.aspx?FamilyID=5B2C0358-915B-4EB5-9B1D-10E506DA9D0Fdisplaylang=en
 
 As a result the following version should be used for comparison:
 1.1 SP1
 2.0 SP2
 3.5 SP1
 
 Updating to 2.0 SP2 would also render the but not for detecting if 
 there are extra APIs (we will have them, as we are now tracking 3.5) 
 clause obsolete.
 
 Also note that the actual status pages at http://go-mono.com/status are 
 empty for me using Internet Explorer but work with Firefox.
 
 Kornél
 
 Atsushi Eno wrote:
  They are now at http://go-mono.com/status
  
  Atsushi Eno
  
  Leszek Ciesielski wrote:
  On Sun, Sep 21, 2008 at 7:30 PM, Miguel de Icaza mig...@novell.com wrote:
  Hello,
 
  Perhaps some members of the Mono team use gui-compare, but I think 
  end-users
  / third-party developers appreciate an easy way to consult the API
  compatibility level of certain types/members.
 
  The only painful part of the online class status pages (that I see) is 
  that
  the masterinfo files are not automatically updated (after changes to
  mono-api-info). This could be scheduled as part of the win32 buildbot
  though.
  Well, the masterinfo files are really static.   Every time there is a
  new release, someone goes scavenging for all the files, and runs
  mono-api-info on the given assemblies and then we version the result.
 
  The actual masterinfos that we have are all listed in
  mono-tools/gui-compare/InfoManager.cs.  Just grep for Uri there.
 
  So its merely an issue of fixing the build, and running more
  mono-api-diffs and generate more web pages for each of the profiles we
  want to compare.   I think those should be:
 
 1.1
 2.0
 3.5 SP1
 
  On a side note: the generation of the masterinfo files is less that
  stellar.  We have used a mix of the Reflection-based tools, and the
  Cecil tools to extract the information, and they are *slightly*
  different.
 
  In the last batch (done for 3.5SP1) I used the Cecil tools for half the
  assemblies (those that lived under Windows/) and reflection ones for the
  ones that live under Reference Assemblies because the Cecil ones
  required an assembly resolver that did not work.
 
  So the result is mixed.
 
  Does gui-compare use the same masterinfo input files (for .NET) as
  corcompare?
  Yes.
 
  Gert
 
  -Original Message-
  From: mono-devel-list-boun...@lists.ximian.com
  [mailto:mono-devel-list-boun...@lists.ximian.com] On Behalf Of Miguel de
  Icaza
  Sent: zondag 21 september 2008 19:32
  To: Gert Driesen
  Cc: 'mono-devel-list'
  Subject: Re: [Mono-dev] Class status pages are empty
 
  Hello,
 
  What was the reason for discontinuing the (online) class status pages?
  We never really discontinued it, at least in my case, I stopped caring.
 
  They were just difficult to maintain, and doing quick changes were not
  reflected quickly (we had to wait a few hours for either the process to
  run, or setup a local system which was never well documented and was
  messy).
 
  Chris Toshok wrote gui-compare, which was very useful to quickly iterate
  over an API and have since then barely used the Web UI.  I am not sure
  if others have done it, but I think some of us just stopped using it.
 
  I agree that a web version has a lot of value, and would love to keep it
  around, but in its current form its painful to deal with.   The other
  bit that is important to point out is that it has not been upgraded to
  deal with the various new profiles, that is something that would have to
  be done as well.
 
  Sorry for resurrecting an old discussion, but a lot of pages contain
  links to the Mono Class Status page. Which - if if understood this
  discussion - is no longer maintained. Could someone put a warning
  there that the page no longer works, and what should be used instead?
  ___
  Mono-devel-list mailing list
  Mono-devel-list@lists.ximian.com
  http://lists.ximian.com/mailman/listinfo/mono-devel-list
 
  
  ___
  Mono-devel-list mailing list
  Mono-devel-list@lists.ximian.com
  http://lists.ximian.com/mailman/listinfo/mono-devel-list
  
 ___
 Mono-devel-list mailing list
 Mono-devel-list@lists.ximian.com
 http://lists.ximian.com/mailman/listinfo/mono-devel-list

___
Mono-devel-list mailing list
Mono-devel-list@lists.ximian.com
http://lists.ximian.com/mailman/listinfo/mono-devel-list


Re: [Mono-dev] Runtime.MonoIline

2009-01-26 Thread Rodrigo Kumpera
Yes, something like that. But no need for the GList.


On Sat, Jan 24, 2009 at 7:43 PM, Jerry Maine crashfou...@gmail.com wrote:

 -BEGIN PGP SIGNED MESSAGE-
 Hash: SHA1

 You mean something like this? (Released under MIT/X11 license)


 gboolean*
 mono_index_custom_attr (MonoImage *image, guint32 idx, const char
 *attr_assembly, const char *attr_name_space, const char *attr)
 {
guint32 mtoken, i, len;
guint32 cols [MONO_CUSTOM_ATTR_SIZE];
MonoTableInfo *ca;
//MonoCustomAttrInfo *ainfo;
GList *tmp, *list = NULL;
MonoMethod* attr_ctor;

ca = image-tables [MONO_TABLE_CUSTOMATTRIBUTE];

if ((attr_assembly == NULL) || (attr_name_space == NULL) || (attr
 == NULL))
return FALSE;

if (!mono_image_has_assembly_ref(assembly-image, attr_assembly)
return FALSE;

 i = mono_metadata_custom_attrs_from_index (image, idx);
if (!i)
return NULL;
i --;
while (i  ca-rows) {
if (mono_metadata_decode_row_col (ca, i,
 MONO_CUSTOM_ATTR_PARENT) != idx)
break;
list = g_list_prepend (list, GUINT_TO_POINTER (i));
++i;
}
len = g_list_length (list);
if (!len)
return NULL;

for (i = 0, tmp = list; i  len; ++i, tmp = tmp-next) {
mono_metadata_decode_row (ca, GPOINTER_TO_UINT (tmp-data),
 cols, MONO_CUSTOM_ATTR_SIZE);
mtoken = cols [MONO_CUSTOM_ATTR_TYPE] 
 MONO_CUSTOM_ATTR_TYPE_BITS;
switch (cols [MONO_CUSTOM_ATTR_TYPE] 
 MONO_CUSTOM_ATTR_TYPE_MASK) {
case MONO_CUSTOM_ATTR_TYPE_METHODDEF:
mtoken |= MONO_TOKEN_METHOD_DEF;
break;
case MONO_CUSTOM_ATTR_TYPE_MEMBERREF:
mtoken |= MONO_TOKEN_MEMBER_REF;
break;
default:
g_error (Unknown table for custom attr type %08x, cols
 [MONO_CUSTOM_ATTR_TYPE]);
break;
}
attr_ctor = mono_get_method (image, mtoken, NULL);
if (!attr_ctor) {
g_warning (Can't find custom attr constructor image: %s
 mtoken: 0x%08x, image-name, mtoken);
g_list_free (list);
//g_free (ainfo);
return FALSE;
} else {
if (strcmp (attr_ctor-klass-image-assembly_name,
 attr_assembly) )
continue;

if (strcmp (attr_ctor-klass-name_space, attr_name_space) )
continue;

if (!strcmp (attr_ctor-klass-name, attr_name) ) {
g_list_free (list);
return TRUE;
}
}
}
g_list_free (list);

return FALSE;
 }
 -BEGIN PGP SIGNATURE-
 Version: GnuPG v1.4.9 (GNU/Linux)
 Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

 iEYEARECAAYFAkl7i3sACgkQ1jvea6V8vHJO5ACfRi0EdEaVLnURhKs2gkguhILV
 32oAnAh3013UYegJZ+kbdNxG5oehjfnZ
 =uOXp
 -END PGP SIGNATURE-

 ___
 Mono-devel-list mailing list
 Mono-devel-list@lists.ximian.com
 http://lists.ximian.com/mailman/listinfo/mono-devel-list

___
Mono-devel-list mailing list
Mono-devel-list@lists.ximian.com
http://lists.ximian.com/mailman/listinfo/mono-devel-list


Re: [Mono-dev] Is Preview 1 of release 2.4 still coming out today?

2009-01-26 Thread Thomas Wiest
John M wrote:
 Hi,

 I was wondering if the first preview of release 2.4 was due today and if it
 includes the fixes that didn't make it into release 2.2?
   
Yes, it was released today (see my e-mail).

As far as the fixes, some have been fixed, some haven't. Check the
specific bugs for details on if they've been included in 2.4p1 or not.

Thanks,
Thomas
___
Mono-devel-list mailing list
Mono-devel-list@lists.ximian.com
http://lists.ximian.com/mailman/listinfo/mono-devel-list


Re: [Mono-dev] [PATCH] Fix for Ping when pinging localhost

2009-01-26 Thread Gonzalo Paniagua Javier
On Mon, 2009-01-26 at 16:48 +0100, Christian Prochnow wrote:
 Hi,
 
 attached is a patch for
 https://bugzilla.novell.com/show_bug.cgi?id=334453
 
 When pinging localhost the Ping class also receives the ICMP_ECHO packet sent 
 by itself. The patch modifies Ping to discard IcmpMessage's with Type==8.

This is now in HEAD and the 2-4 branch.
Thanks!

-Gonzalo


___
Mono-devel-list mailing list
Mono-devel-list@lists.ximian.com
http://lists.ximian.com/mailman/listinfo/mono-devel-list