Compare commits

..
1 Commits
Author SHA1 Message Date
nobody 08cc744ea3 This commit was manufactured by cvs2svn to create tag 'MONO_0_4'.
svn path=/tags/MONO_0_4/mono/; revision=175
2001-07-16 02:06:13 +00:00
11569 changed files with 21623 additions and 2029704 deletions

No files matched your search

+8 -20
View File
@@ -1,23 +1,11 @@
aclocal.m4
config.cache
config.guess
config.h
config.h.in
config.log
config.status
config.sub
configure
libtool
ltmain.sh
ltconfig
Makefile
Makefile.in
mono-*.tar.gz
stamp-h
config.cache
stamp-h.in
mono.pc
mono.spec
autom4te.cache
mint.pc
stamp-h1
config.h.in
config.status
aclocal.m4
configure
stamp-h
libtool
Makefile.in
config.log
-1
View File
@@ -1,3 +1,2 @@
Miguel de Icaza ([email protected])
Paolo Molaro ([email protected])
Dietmar Maurer ([email protected])
-485
View File
@@ -1,485 +0,0 @@
The Mono runtime is licensed under the terms of the GNU
Library General Public License, version 2.
GNU LIBRARY GENERAL PUBLIC LICENSE
Version 2, June 1991
Copyright (C) 1991 Free Software Foundation, Inc.
675 Mass Ave, Cambridge, MA 02139, USA
Everyone is permitted to copy and distribute verbatim copies
of this license document, but changing it is not allowed.
[This is the first released version of the library GPL. It is
numbered 2 because it goes with version 2 of the ordinary GPL.]
Preamble
The licenses for most software are designed to take away your
freedom to share and change it. By contrast, the GNU General Public
Licenses are intended to guarantee your freedom to share and change
free software--to make sure the software is free for all its users.
This license, the Library General Public License, applies to some
specially designated Free Software Foundation software, and to any
other libraries whose authors decide to use it. You can use it for
your libraries, too.
When we speak of free software, we are referring to freedom, not
price. Our General Public Licenses are designed to make sure that you
have the freedom to distribute copies of free software (and charge for
this service if you wish), that you receive source code or can get it
if you want it, that you can change the software or use pieces of it
in new free programs; and that you know you can do these things.
To protect your rights, we need to make restrictions that forbid
anyone to deny you these rights or to ask you to surrender the rights.
These restrictions translate to certain responsibilities for you if
you distribute copies of the library, or if you modify it.
For example, if you distribute copies of the library, whether gratis
or for a fee, you must give the recipients all the rights that we gave
you. You must make sure that they, too, receive or can get the source
code. If you link a program with the library, you must provide
complete object files to the recipients so that they can relink them
with the library, after making changes to the library and recompiling
it. And you must show them these terms so they know their rights.
Our method of protecting your rights has two steps: (1) copyright
the library, and (2) offer you this license which gives you legal
permission to copy, distribute and/or modify the library.
Also, for each distributor's protection, we want to make certain
that everyone understands that there is no warranty for this free
library. If the library is modified by someone else and passed on, we
want its recipients to know that what they have is not the original
version, so that any problems introduced by others will not reflect on
the original authors' reputations.
Finally, any free program is threatened constantly by software
patents. We wish to avoid the danger that companies distributing free
software will individually obtain patent licenses, thus in effect
transforming the program into proprietary software. To prevent this,
we have made it clear that any patent must be licensed for everyone's
free use or not licensed at all.
Most GNU software, including some libraries, is covered by the ordinary
GNU General Public License, which was designed for utility programs. This
license, the GNU Library General Public License, applies to certain
designated libraries. This license is quite different from the ordinary
one; be sure to read it in full, and don't assume that anything in it is
the same as in the ordinary license.
The reason we have a separate public license for some libraries is that
they blur the distinction we usually make between modifying or adding to a
program and simply using it. Linking a program with a library, without
changing the library, is in some sense simply using the library, and is
analogous to running a utility program or application program. However, in
a textual and legal sense, the linked executable is a combined work, a
derivative of the original library, and the ordinary General Public License
treats it as such.
Because of this blurred distinction, using the ordinary General
Public License for libraries did not effectively promote software
sharing, because most developers did not use the libraries. We
concluded that weaker conditions might promote sharing better.
However, unrestricted linking of non-free programs would deprive the
users of those programs of all benefit from the free status of the
libraries themselves. This Library General Public License is intended to
permit developers of non-free programs to use free libraries, while
preserving your freedom as a user of such programs to change the free
libraries that are incorporated in them. (We have not seen how to achieve
this as regards changes in header files, but we have achieved it as regards
changes in the actual functions of the Library.) The hope is that this
will lead to faster development of free libraries.
The precise terms and conditions for copying, distribution and
modification follow. Pay close attention to the difference between a
"work based on the library" and a "work that uses the library". The
former contains code derived from the library, while the latter only
works together with the library.
Note that it is possible for a library to be covered by the ordinary
General Public License rather than by this special one.
GNU LIBRARY GENERAL PUBLIC LICENSE
TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
0. This License Agreement applies to any software library which
contains a notice placed by the copyright holder or other authorized
party saying it may be distributed under the terms of this Library
General Public License (also called "this License"). Each licensee is
addressed as "you".
A "library" means a collection of software functions and/or data
prepared so as to be conveniently linked with application programs
(which use some of those functions and data) to form executables.
The "Library", below, refers to any such software library or work
which has been distributed under these terms. A "work based on the
Library" means either the Library or any derivative work under
copyright law: that is to say, a work containing the Library or a
portion of it, either verbatim or with modifications and/or translated
straightforwardly into another language. (Hereinafter, translation is
included without limitation in the term "modification".)
"Source code" for a work means the preferred form of the work for
making modifications to it. For a library, complete source code means
all the source code for all modules it contains, plus any associated
interface definition files, plus the scripts used to control compilation
and installation of the library.
Activities other than copying, distribution and modification are not
covered by this License; they are outside its scope. The act of
running a program using the Library is not restricted, and output from
such a program is covered only if its contents constitute a work based
on the Library (independent of the use of the Library in a tool for
writing it). Whether that is true depends on what the Library does
and what the program that uses the Library does.
1. You may copy and distribute verbatim copies of the Library's
complete source code as you receive it, in any medium, provided that
you conspicuously and appropriately publish on each copy an
appropriate copyright notice and disclaimer of warranty; keep intact
all the notices that refer to this License and to the absence of any
warranty; and distribute a copy of this License along with the
Library.
You may charge a fee for the physical act of transferring a copy,
and you may at your option offer warranty protection in exchange for a
fee.
2. You may modify your copy or copies of the Library or any portion
of it, thus forming a work based on the Library, and copy and
distribute such modifications or work under the terms of Section 1
above, provided that you also meet all of these conditions:
a) The modified work must itself be a software library.
b) You must cause the files modified to carry prominent notices
stating that you changed the files and the date of any change.
c) You must cause the whole of the work to be licensed at no
charge to all third parties under the terms of this License.
d) If a facility in the modified Library refers to a function or a
table of data to be supplied by an application program that uses
the facility, other than as an argument passed when the facility
is invoked, then you must make a good faith effort to ensure that,
in the event an application does not supply such function or
table, the facility still operates, and performs whatever part of
its purpose remains meaningful.
(For example, a function in a library to compute square roots has
a purpose that is entirely well-defined independent of the
application. Therefore, Subsection 2d requires that any
application-supplied function or table used by this function must
be optional: if the application does not supply it, the square
root function must still compute square roots.)
These requirements apply to the modified work as a whole. If
identifiable sections of that work are not derived from the Library,
and can be reasonably considered independent and separate works in
themselves, then this License, and its terms, do not apply to those
sections when you distribute them as separate works. But when you
distribute the same sections as part of a whole which is a work based
on the Library, the distribution of the whole must be on the terms of
this License, whose permissions for other licensees extend to the
entire whole, and thus to each and every part regardless of who wrote
it.
Thus, it is not the intent of this section to claim rights or contest
your rights to work written entirely by you; rather, the intent is to
exercise the right to control the distribution of derivative or
collective works based on the Library.
In addition, mere aggregation of another work not based on the Library
with the Library (or with a work based on the Library) on a volume of
a storage or distribution medium does not bring the other work under
the scope of this License.
3. You may opt to apply the terms of the ordinary GNU General Public
License instead of this License to a given copy of the Library. To do
this, you must alter all the notices that refer to this License, so
that they refer to the ordinary GNU General Public License, version 2,
instead of to this License. (If a newer version than version 2 of the
ordinary GNU General Public License has appeared, then you can specify
that version instead if you wish.) Do not make any other change in
these notices.
Once this change is made in a given copy, it is irreversible for
that copy, so the ordinary GNU General Public License applies to all
subsequent copies and derivative works made from that copy.
This option is useful when you wish to copy part of the code of
the Library into a program that is not a library.
4. You may copy and distribute the Library (or a portion or
derivative of it, under Section 2) in object code or executable form
under the terms of Sections 1 and 2 above provided that you accompany
it with the complete corresponding machine-readable source code, which
must be distributed under the terms of Sections 1 and 2 above on a
medium customarily used for software interchange.
If distribution of object code is made by offering access to copy
from a designated place, then offering equivalent access to copy the
source code from the same place satisfies the requirement to
distribute the source code, even though third parties are not
compelled to copy the source along with the object code.
5. A program that contains no derivative of any portion of the
Library, but is designed to work with the Library by being compiled or
linked with it, is called a "work that uses the Library". Such a
work, in isolation, is not a derivative work of the Library, and
therefore falls outside the scope of this License.
However, linking a "work that uses the Library" with the Library
creates an executable that is a derivative of the Library (because it
contains portions of the Library), rather than a "work that uses the
library". The executable is therefore covered by this License.
Section 6 states terms for distribution of such executables.
When a "work that uses the Library" uses material from a header file
that is part of the Library, the object code for the work may be a
derivative work of the Library even though the source code is not.
Whether this is true is especially significant if the work can be
linked without the Library, or if the work is itself a library. The
threshold for this to be true is not precisely defined by law.
If such an object file uses only numerical parameters, data
structure layouts and accessors, and small macros and small inline
functions (ten lines or less in length), then the use of the object
file is unrestricted, regardless of whether it is legally a derivative
work. (Executables containing this object code plus portions of the
Library will still fall under Section 6.)
Otherwise, if the work is a derivative of the Library, you may
distribute the object code for the work under the terms of Section 6.
Any executables containing that work also fall under Section 6,
whether or not they are linked directly with the Library itself.
6. As an exception to the Sections above, you may also compile or
link a "work that uses the Library" with the Library to produce a
work containing portions of the Library, and distribute that work
under terms of your choice, provided that the terms permit
modification of the work for the customer's own use and reverse
engineering for debugging such modifications.
You must give prominent notice with each copy of the work that the
Library is used in it and that the Library and its use are covered by
this License. You must supply a copy of this License. If the work
during execution displays copyright notices, you must include the
copyright notice for the Library among them, as well as a reference
directing the user to the copy of this License. Also, you must do one
of these things:
a) Accompany the work with the complete corresponding
machine-readable source code for the Library including whatever
changes were used in the work (which must be distributed under
Sections 1 and 2 above); and, if the work is an executable linked
with the Library, with the complete machine-readable "work that
uses the Library", as object code and/or source code, so that the
user can modify the Library and then relink to produce a modified
executable containing the modified Library. (It is understood
that the user who changes the contents of definitions files in the
Library will not necessarily be able to recompile the application
to use the modified definitions.)
b) Accompany the work with a written offer, valid for at
least three years, to give the same user the materials
specified in Subsection 6a, above, for a charge no more
than the cost of performing this distribution.
c) If distribution of the work is made by offering access to copy
from a designated place, offer equivalent access to copy the above
specified materials from the same place.
d) Verify that the user has already received a copy of these
materials or that you have already sent this user a copy.
For an executable, the required form of the "work that uses the
Library" must include any data and utility programs needed for
reproducing the executable from it. However, as a special exception,
the source code distributed need not include anything that is normally
distributed (in either source or binary form) with the major
components (compiler, kernel, and so on) of the operating system on
which the executable runs, unless that component itself accompanies
the executable.
It may happen that this requirement contradicts the license
restrictions of other proprietary libraries that do not normally
accompany the operating system. Such a contradiction means you cannot
use both them and the Library together in an executable that you
distribute.
7. You may place library facilities that are a work based on the
Library side-by-side in a single library together with other library
facilities not covered by this License, and distribute such a combined
library, provided that the separate distribution of the work based on
the Library and of the other library facilities is otherwise
permitted, and provided that you do these two things:
a) Accompany the combined library with a copy of the same work
based on the Library, uncombined with any other library
facilities. This must be distributed under the terms of the
Sections above.
b) Give prominent notice with the combined library of the fact
that part of it is a work based on the Library, and explaining
where to find the accompanying uncombined form of the same work.
8. You may not copy, modify, sublicense, link with, or distribute
the Library except as expressly provided under this License. Any
attempt otherwise to copy, modify, sublicense, link with, or
distribute the Library is void, and will automatically terminate your
rights under this License. However, parties who have received copies,
or rights, from you under this License will not have their licenses
terminated so long as such parties remain in full compliance.
9. You are not required to accept this License, since you have not
signed it. However, nothing else grants you permission to modify or
distribute the Library or its derivative works. These actions are
prohibited by law if you do not accept this License. Therefore, by
modifying or distributing the Library (or any work based on the
Library), you indicate your acceptance of this License to do so, and
all its terms and conditions for copying, distributing or modifying
the Library or works based on it.
10. Each time you redistribute the Library (or any work based on the
Library), the recipient automatically receives a license from the
original licensor to copy, distribute, link with or modify the Library
subject to these terms and conditions. You may not impose any further
restrictions on the recipients' exercise of the rights granted herein.
You are not responsible for enforcing compliance by third parties to
this License.
11. If, as a consequence of a court judgment or allegation of patent
infringement or for any other reason (not limited to patent issues),
conditions are imposed on you (whether by court order, agreement or
otherwise) that contradict the conditions of this License, they do not
excuse you from the conditions of this License. If you cannot
distribute so as to satisfy simultaneously your obligations under this
License and any other pertinent obligations, then as a consequence you
may not distribute the Library at all. For example, if a patent
license would not permit royalty-free redistribution of the Library by
all those who receive copies directly or indirectly through you, then
the only way you could satisfy both it and this License would be to
refrain entirely from distribution of the Library.
If any portion of this section is held invalid or unenforceable under any
particular circumstance, the balance of the section is intended to apply,
and the section as a whole is intended to apply in other circumstances.
It is not the purpose of this section to induce you to infringe any
patents or other property right claims or to contest validity of any
such claims; this section has the sole purpose of protecting the
integrity of the free software distribution system which is
implemented by public license practices. Many people have made
generous contributions to the wide range of software distributed
through that system in reliance on consistent application of that
system; it is up to the author/donor to decide if he or she is willing
to distribute software through any other system and a licensee cannot
impose that choice.
This section is intended to make thoroughly clear what is believed to
be a consequence of the rest of this License.
12. If the distribution and/or use of the Library is restricted in
certain countries either by patents or by copyrighted interfaces, the
original copyright holder who places the Library under this License may add
an explicit geographical distribution limitation excluding those countries,
so that distribution is permitted only in or among countries not thus
excluded. In such case, this License incorporates the limitation as if
written in the body of this License.
13. The Free Software Foundation may publish revised and/or new
versions of the Library General Public License from time to time.
Such new versions will be similar in spirit to the present version,
but may differ in detail to address new problems or concerns.
Each version is given a distinguishing version number. If the Library
specifies a version number of this License which applies to it and
"any later version", you have the option of following the terms and
conditions either of that version or of any later version published by
the Free Software Foundation. If the Library does not specify a
license version number, you may choose any version ever published by
the Free Software Foundation.
14. If you wish to incorporate parts of the Library into other free
programs whose distribution conditions are incompatible with these,
write to the author to ask for permission. For software which is
copyrighted by the Free Software Foundation, write to the Free
Software Foundation; we sometimes make exceptions for this. Our
decision will be guided by the two goals of preserving the free status
of all derivatives of our free software and of promoting the sharing
and reuse of software generally.
NO WARRANTY
15. BECAUSE THE LIBRARY IS LICENSED FREE OF CHARGE, THERE IS NO
WARRANTY FOR THE LIBRARY, TO THE EXTENT PERMITTED BY APPLICABLE LAW.
EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR
OTHER PARTIES PROVIDE THE LIBRARY "AS IS" WITHOUT WARRANTY OF ANY
KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE
LIBRARY IS WITH YOU. SHOULD THE LIBRARY PROVE DEFECTIVE, YOU ASSUME
THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION.
16. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN
WRITING WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY
AND/OR REDISTRIBUTE THE LIBRARY AS PERMITTED ABOVE, BE LIABLE TO YOU
FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR
CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE
LIBRARY (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING
RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A
FAILURE OF THE LIBRARY TO OPERATE WITH ANY OTHER SOFTWARE), EVEN IF
SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH
DAMAGES.
END OF TERMS AND CONDITIONS
Appendix: How to Apply These Terms to Your New Libraries
If you develop a new library, and you want it to be of the greatest
possible use to the public, we recommend making it free software that
everyone can redistribute and change. You can do so by permitting
redistribution under these terms (or, alternatively, under the terms of the
ordinary General Public License).
To apply these terms, attach the following notices to the library. It is
safest to attach them to the start of each source file to most effectively
convey the exclusion of warranty; and each file should have at least the
"copyright" line and a pointer to where the full notice is found.
<one line to give the library's name and a brief idea of what it does.>
Copyright (C) <year> <name of author>
This library is free software; you can redistribute it and/or
modify it under the terms of the GNU Library General Public
License as published by the Free Software Foundation; either
version 2 of the License, or (at your option) any later version.
This library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Library General Public License for more details.
You should have received a copy of the GNU Library General Public
License along with this library; if not, write to the Free
Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
Also add information on how to contact you by electronic and paper mail.
You should also get your employer (if you work as a programmer) or your
school, if any, to sign a "copyright disclaimer" for the library, if
necessary. Here is a sample; alter the names:
Yoyodyne, Inc., hereby disclaims all copyright interest in the
library `Frob' (a library for tweaking knobs) written by James Random Hacker.
<signature of Ty Coon>, 1 April 1990
Ty Coon, President of Vice
That's all there is to it!
-2005
View File
File diff suppressed because it is too large. Load diff
+1 -221
View File
@@ -1,221 +1 @@
AUTOMAKE_OPTIONS = foreign
ACLOCAL_AMFLAGS = -I .
SUBDIRS = $(libgc_dir) mono $(ikvm_jni_dir) docs data runtime scripts man samples support web
# Keep in sync with SUBDIRS
## 'tools' is not normally built
DIST_SUBDIRS = libgc mono ikvm-jni docs data runtime scripts man samples support web tools
EXTRA_DIST= mono.pc.in mono.spec.in mint.pc.in
pkgconfigdir = $(libdir)/pkgconfig
if JIT_SUPPORTED
pkgconfig_DATA= mono.pc mint.pc
else
pkgconfig_DATA= mint.pc
endif
DISTCLEANFILES= mono.pc mint.pc
#
# The following targets are used to compile and install
# mono and mcs without running into the out-of-sync
# problems by doing the compilation in the right order.
#
mcs_topdir=$(top_srcdir)/../mcs
.PHONY: bootstrap faststrap fasterstrap bootstrap-clean bootstrap-check get-monolite-latest monolite-bootstrap
.PHONY: tmpinst-dir tmpinst-mcs-basic-build tmpinst-mcs-short-build tmpinst-mcs-full-build tmpinst-monolite
.PHONY: mono-do-runtime-only mcs-do-basic-build mcs-do-short-build mcs-do-full-build mcs-do-clean-profiles mcs-do-run-test
.PHONY: populate-runtime-subdir
if USE_JIT
mono_runtime = mono/mini/mono
else
mono_runtime = mono/interpreter/mint
endif
# Minimal set of directories to build, and files used to run the 'mcs' compiler
MCS_BASIC_DIRS = jay mcs class/corlib class/System
MCS_BASIC_FILES = \
$(mcs_topdir)/mcs/mcs.exe \
$(mcs_topdir)/class/lib/default/mscorlib.dll \
$(mcs_topdir)/class/lib/default/System.dll
# Set of directories and files the 'mcs' compiler can use
## We build class/System twice to resolve the cyclic dependency of class/System.XML
MCS_DIRS = $(MCS_BASIC_DIRS) class/System.XML class/System class/Mono.CSharp.Debugger
MCS_FILES = $(MCS_BASIC_FILES) \
$(mcs_topdir)/class/lib/default/System.Xml.dll \
$(mcs_topdir)/class/lib/default/Mono.CSharp.Debugger.dll
tmpinst = _tmpinst
if PLATFORM_WIN32
populate_profiles = default:net_1_1
# bootstraps using the .NET C# compiler
bootstrap: tmpinst-dir mono-do-runtime-only
$(MAKE) fasterstrap
else
populate_profiles = default:net_1_1 net_2_0:net_2_0
bootstrap: tmpinst-mcs-basic-build mono-do-runtime-only
$(MAKE) tmpinst-mcs-short-build
$(MAKE) fasterstrap
endif
faststrap: tmpinst-mcs-short-build
$(MAKE) mono-do-runtime-only
$(MAKE) fasterstrap
fasterstrap: tmpinst-mcs-full-build
$(MAKE) populate-runtime-subdir
$(MAKE)
bootstrap-clean: clean mcs-do-clean-profiles
bootstrap-check: check mcs-do-run-test
# building with monolite
get-monolite-latest:
wget -O $(srcdir)/../monolite-latest.tar.gz http://www.go-mono.com/daily/monolite-latest.tar.gz
monolite-bootstrap: tmpinst-monolite mono-do-runtime-only
rm -f $(MCS_FILES)
$(MAKE) tmpinst-mcs-short-build
$(MAKE) fasterstrap
# internal targets
tmpinst-mcs-basic-build: mcs-do-basic-build tmpinst-dir
mv -f $(MCS_BASIC_FILES) $(tmpinst)/lib
tmpinst-mcs-short-build: mcs-do-short-build tmpinst-dir
mv -f $(MCS_FILES) $(tmpinst)/lib
tmpinst-mcs-full-build: mcs-do-full-build tmpinst-dir
cp -f $(MCS_FILES) $(tmpinst)/lib
tmpinst-monolite: tmpinst-dir
-rm -f $(tmpinst)/lib/mcs.exe $(tmpinst)/lib/*.dll
-rm -fr $(tmpinst)/monolite-*
srcdir=`cd $(srcdir) && pwd` && cd $(tmpinst) && ( gzip -d -c $$srcdir/../monolite-latest.tar.gz | tar xf - )
mv -f $(tmpinst)/monolite-*/mcs.exe $(tmpinst)/monolite-*/*.dll $(tmpinst)/lib
rm -fr $(tmpinst)/monolite-*
# Build enough of 'mono' to run the stage1 mcs.exe
mono-do-runtime-only: $(CONFIG_HEADER)
for dir in $(libgc_dir) mono data; do \
(cd $$dir && $(MAKE)) || exit 1 ; \
done
mcs-do-basic-build:
rm -f $(MCS_FILES)
for dir in $(MCS_BASIC_DIRS); do \
(cd $(mcs_topdir)/$$dir && $(MAKE) PROFILE=default USE_BOOT_COMPILE=yes BOOTSTRAP_MCS="mcs -d:BOOTSTRAP_WITH_OLDLIB") || exit 1 ; \
done
mcs-do-short-build:
tmpinst=`cd $(tmpinst) && pwd` ; \
PATH=$$tmpinst/bin:$$PATH; export PATH ; \
for dir in $(MCS_DIRS); do \
(cd $(mcs_topdir)/$$dir && $(MAKE) PROFILE=default) || exit 1 ; \
done
mcs-do-full-build:
tmpinst=`cd $(tmpinst) && pwd` ; \
PATH=$$tmpinst/bin:$$PATH ; export PATH ; \
cd $(mcs_topdir) && $(MAKE) all-profiles
mcs-do-clean-profiles:
cd $(mcs_topdir) && $(MAKE) clean-profiles
mcs-do-run-test: tmpinst-dir
tmpinst=`cd $(tmpinst) && pwd` ; \
PATH=$$tmpinst/bin:$$PATH ; export PATH ; \
cd $(mcs_topdir) && $(MAKE) PROFILE=default run-test
populate-runtime-subdir:
test -n '$(populate_profiles)'
@srcdir=`cd $(srcdir) && pwd`; tmpinst=`pwd`/$(tmpinst) ; \
PATH=$$tmpinst/bin:$$PATH ; export PATH ; \
for profile in $(populate_profiles); do \
( set fnord `echo $$profile | sed 's,:, ,g'` && \
echo "cd $(mcs_topdir) && $(MAKE) PROGRAM_INSTALL_DIR=$$srcdir/runtime PROFILE=$$2 LIBRARY_INSTALL_DIR=$$srcdir/runtime/$$3 prefix=$$tmpinst/prefix install" && \
cd $(mcs_topdir) && \
$(MAKE) PROGRAM_INSTALL_DIR=$$srcdir/runtime PROFILE=$$2 LIBRARY_INSTALL_DIR=$$srcdir/runtime/$$3 prefix=$$tmpinst/prefix install ) || exit 1 ; \
done
find $(tmpinst)/prefix -type f -path '*/jay*' -exec rm -f '{}' ';'
test -z "`find $(tmpinst)/prefix -type f -print`"
rm -rf $(tmpinst)/prefix
tmpinst-dir:
mkdir -p $(tmpinst) $(tmpinst)/bin $(tmpinst)/lib $(tmpinst)/etc/mono
$(MAKE) tmpinst-dir-contents
tmpinst-dir-contents: $(tmpinst)/bin/mono $(tmpinst)/bin/mcs $(tmpinst)/etc/mono/machine.config
if PLATFORM_WIN32
$(tmpinst)/bin/mono: $(srcdir)/Makefile.am
echo '#! /bin/sh' > $@
echo 'builddir="'`pwd`'"; win_builddir="'`cygpath -w -a .`'"' >> $@
echo 'tmpinst="$$win_builddir\\$(tmpinst)"' >> $@
echo 'if test -z "$$MONO_PATH" ; then MONO_PATH="$$tmpinst\\lib"; else MONO_PATH="$${MONO_PATH};$$tmpinst\\lib" ; fi' >> $@
echo 'MONO_CFG_DIR="$$tmpinst\\etc"' >> $@
echo 'export MONO_PATH MONO_CFG_DIR' >> $@
echo 'exec "$$builddir/libtool" --mode=execute "$$builddir/mono.exe" "$$@"' >> $@
chmod +x $@
else
$(tmpinst)/bin/mono: $(srcdir)/Makefile.am
echo '#! /bin/sh' > $@
echo 'builddir="'`pwd`'"; tmpinst=$$builddir/$(tmpinst)' >> $@
echo 'if test -z "$$MONO_PATH" ; then MONO_PATH=$$tmpinst/lib; else MONO_PATH=$${MONO_PATH}:$$tmpinst/lib ; fi' >> $@
echo 'MONO_CFG_DIR=$$tmpinst/etc' >> $@
echo 'export MONO_PATH MONO_CFG_DIR' >> $@
echo 'exec "$$builddir/libtool" --mode=execute "$$builddir/$(mono_runtime)" --config "$$builddir/data/config" "$$@"' >> $@
chmod +x $@
endif
$(tmpinst)/bin/mcs: $(srcdir)/Makefile.am
(t=`cd $(tmpinst) && pwd`; echo '#! /bin/sh'; echo 'exec "'"$$t/bin/mono"'" "'"$$t/lib/mcs.exe"'" "$$@"') > $@
chmod +x $@
$(tmpinst)/etc/mono/machine.config: $(srcdir)/data/machine.config
rm -f $@
srcdir=`cd $(srcdir) && pwd`; cd $(tmpinst)/etc/mono && $(LN_S) $$srcdir/data/machine.config machine.config
# Obsolete target -- don't use.
.PHONY: fullbuild
fullbuild: bootstrap
$(MAKE) install
echo "*** 'fullbuild' is an obsolete target. Use 'make bootstrap && make install'."
exit 1
win32getdeps:
wget http://www.go-mono.com/archive/pkgconfig-0.11-20020310.zip
wget http://www.go-mono.com/archive/glib-2.0.4-20020703.zip
wget http://www.go-mono.com/archive/glib-dev-2.0.4-20020703.zip
wget http://www.go-mono.com/archive/libiconv-1.7.zip
wget http://www.go-mono.com/archive/libiconv-dev-1.7.zip
wget http://www.go-mono.com/archive/libintl-0.10.40-20020101.zip
wget http://www.jroith.de/nsis4cygwin.zip
unzip -n -d / pkgconfig-0.11-20020310.zip
unzip -n -d / glib-2.0.4-20020703.zip
unzip -n -d / glib-dev-2.0.4-20020703.zip
unzip -n -d / libiconv-1.7.zip
unzip -n -d / libiconv-dev-1.7.zip
unzip -n -d / libintl-0.10.40-20020101.zip
unzip -n -d / nsis4cygwin.zip
win32setup:
makensis /DMILESTONE=$(VERSION) /DSOURCE_INSTALL_DIR=$(SOURCE_INSTALL_DIR) /DBUILDNUM=$(BUILDNUM) monowiz.win32.nsi
SUBDIRS = mono doc
-1578
View File
File diff suppressed because it is too large. Load diff
+4 -252
View File
@@ -1,256 +1,8 @@
This is MonoNet.
This is Mono.
Ximian's portable implementation of .NET
1. Installation
2. Using Mono
3. Directory Roadmap
1. Compilation and Installation
===============================
a. Build Requirements
---------------------
To build Mono, you will need the following components:
* pkg-config
Available from: http://www.freedesktop.org/Software/pkgconfig
* glib 2.0
Available from: http://www.gtk.org/
Optional dependencies:
* ICU library
http://oss.software.ibm.com/icu/index.html
You will need this one to get complete support for the international
features of the .NET Framework.
* Cairo and libgdiplus
If you want to get support for System.Drawing, you will need to get
both Libgdiplus and Cairo.
b. Building the Software
------------------------
If you obtained this package as an officially released tarball,
this is very simple, use configure and make:
./configure --prefix=/usr/local
make
make install
Mono supports a JIT engine on x86, SPARC and PowerPC systems.
The various commands that ship with Mono default to the JIT engine
on x86 and SPARC, to turn it on for PPC systems, use the --with-jit=yes
command line option to configure.
MacOS X Users: you will need to download the latest Boehm GC
Alpha release for garbage collection to work properly.
If you obtained this as a snapshot, you will need an existing
Mono installation. To upgrade your installation, unpack both
mono and mcs:
tar xzf mcs-XXXX.tar.gz
tar xzf mono-XXXX.tar.gz
mv mono-XXX mono
mv mcs-XXX mcs
cd mono
./autogen.sh --prefix=/usr/local
make bootstrap
c. Building the software from CVS
---------------------------------
If you are building the software from CVS, make sure that you
have up-to-date mcs and mono sources:
cvs co mono mcs
Then, go into the mono directory, and configure:
cd mono
./autogen.sh --prefix=/usr/local
Depending on whether you have an existing Mono installation or
not, you can try the following:
i. If you have an existing Mono installation
-----------------------------------------
First verify that you have a working installation:
echo 'class X { static void Main () { System.Console.Write("OK");}}' > x.cs
Compile:
mcs x.cs
And run:
mono x.exe
If you get the output `OK' and no errors, and you have mono
version 0.31 or later, continue. Otherwise, you can try option
(ii) below.
You are ready to start your CVS upgrade. Compile with
make bootstrap
This will automatically go into the mcs/ tree and build the
binaries there, and copy them into the appropriate
sub-directories of mono/runtime/.
Now, go to step (iii) below.
ii. If you don't have a working Mono installation
---------------------------------------------
If you don't have a working Mono installation, an obvious choice
is to install the latest released packages of 'mono' for your
distribution and go back to step (i).
You can also try a slightly more risky approach that should work
almost all the time.
This works by first getting the latest version of the 'monolite'
distribution, which contains just enough to run the 'mcs'
compiler. You do this by
make get-monolite-latest
This should place a monolite-latest.tar.gz in the parent
directory of the 'mono' source tree. You can then run:
make monolite-bootstrap
This will automatically gunzip and untar the tarball, place the
files appropriately, and then complete the bootstrap.
iii. Testing and Installation
------------------------
You can run the mono and mcs testsuites with the command:
make -k bootstrap-check
Expect to find several testsuite failures, especially in the
mcs/ tree. As a sanity check, you can compare the failures you
got with
http://go-mono.com/tests/displayTestResults.php
You can now install it:
make install
Failure to follow these steps will result in a broken installation.
iv. Other useful "bootstrap"-like facilities
----------------------------------------
If you have a CVS snapshot that you keep updating periodically,
and/or do your development in, you may try using a couple of
specialty make targets that may be slightly faster than a
"make bootstrap".
You can try a two-stage bootstrap with:
make faststrap
This assumes that you have already run a "make bootstrap" on the
tree before.
If you want to avoid waiting even for that, you can try the
riskier one-stage build:
make fasterstrap
This should be attempted only if you're sure that the sources of
the mcs compiler itself and the sources of the libraries used by
it have not been changed.
2. Using Mono
=============
Once you have installed the software, you can run a few programs:
* runtime engine
mono program.exe
or
mint program.exe
* C# compiler
mcs program.cs
* CIL Disassembler
monodis program.exe
See the man pages for mono(1), mint(1), monodis(1) and mcs(2)
for further details.
3. Directory Roadmap
====================
doc/
Contains the web site contents.
docs/
Technical documents about the Mono runtime.
data/
Configuration files installed as part of the Mono runtime.
mono/
The core of the Mono Runtime.
metadata/
The object system and metadata reader.
jit/
The Just in Time Compiler.
dis/
CIL executable Disassembler
cli/
Common code for the JIT and the interpreter.
io-layer/
The I/O layer and system abstraction for
emulating the .NET IO model.
cil/
Common Intermediate Representation, XML
definition of the CIL bytecodes.
interp/
Interpreter for CLI executables.
arch/
Architecture specific portions.
man/
Manual pages for the various Mono commands and programs.
scripts/
Scripts used to invoke Mono and the corresponding program.
runtime/
A directory holding a pre-compiled version of the Mono
runtime.
----------
To edit the go-mono website, see doc/README.
-37
View File
@@ -1,37 +0,0 @@
#undef HAVE_INET_PTON
#undef HAVE_INET_ATON
#undef HAVE_MSG_NOSIGNAL
#undef HAVE_SOL_IP
#undef HAVE_SOL_TCP
#undef HAVE_SOL_IPV6
#undef HAVE_IP_PKTINFO
#undef HAVE_STRUCT_IP_MREQN
#undef HAVE_STRUCT_IP_MREQ
#undef HAVE_PTHREAD_MUTEX_TIMEDLOCK
#undef USE_MONO_MUTEX
#undef HAVE_LARGE_FILE_SUPPORT
#undef HAVE_NEW_ICONV
#undef NAME_DEV_RANDOM
#undef HAVE_CRYPT_RNG
#undef HAVE_BOEHM_GC
#undef USE_INCLUDED_LIBGC
#undef DISABLE_SHARED_HANDLES
#undef HAVE_TIMEZONE
#undef HAVE_TM_GMTOFF
#undef MONO_SIZEOF_SUNPATH
#undef MONO_ZERO_ARRAY_LENGTH
#undef PTHREAD_POINTER_ID
#undef HAVE_GC_H
#undef HAVE_GC_GCJ_MALLOC
#undef HAVE_GC_ENABLE
#undef WITH_BUNDLE
#undef HAVE_GETHOSTBYNAME2_R
#undef HAVE_ICU
#undef HAVE_KW_THREAD
#undef HAVE_SIGNBIT
#undef USE_MACH_SEMA
#undef HAVE_SOCKLEN_T
#undef HAVE_TRUNCL
#undef HAVE_AINTL
#undef HAVE_SCANDIR
#undef HAVE_WORKING_SIGALTSTACK
+6 -142
View File
@@ -1,142 +1,6 @@
#!/bin/sh
# Run this to generate all the initial makefiles, etc.
# Ripped off from GNOME macros version
DIE=0
srcdir=`dirname $0`
test -z "$srcdir" && srcdir=.
if [ -n "$MONO_PATH" ]; then
# from -> /mono/lib:/another/mono/lib
# to -> /mono /another/mono
for i in `echo ${MONO_PATH} | tr ":" " "`; do
i=`dirname ${i}`
if [ -n "{i}" -a -d "${i}/share/aclocal" ]; then
ACLOCAL_FLAGS="-I ${i}/share/aclocal $ACLOCAL_FLAGS"
fi
if [ -n "{i}" -a -d "${i}/bin" ]; then
PATH="${i}/bin:$PATH"
fi
done
export PATH
fi
(autoconf --version) < /dev/null > /dev/null 2>&1 || {
echo
echo "**Error**: You must have \`autoconf' installed to compile Mono."
echo "Download the appropriate package for your distribution,"
echo "or get the source tarball at ftp://ftp.gnu.org/pub/gnu/"
DIE=1
}
if [ -z "$LIBTOOL" ]; then
LIBTOOL=`which glibtool 2>/dev/null`
if [ ! -x "$LIBTOOL" ]; then
LIBTOOL=`which libtool`
fi
fi
(grep "^AM_PROG_LIBTOOL" $srcdir/configure.in >/dev/null) && {
($LIBTOOL --version) < /dev/null > /dev/null 2>&1 || {
echo
echo "**Error**: You must have \`libtool' installed to compile Mono."
echo "Get ftp://ftp.gnu.org/pub/gnu/libtool-1.2d.tar.gz"
echo "(or a newer version if it is available)"
DIE=1
}
}
grep "^AM_GNU_GETTEXT" $srcdir/configure.in >/dev/null && {
grep "sed.*POTFILES" $srcdir/configure.in >/dev/null || \
(gettext --version) < /dev/null > /dev/null 2>&1 || {
echo
echo "**Error**: You must have \`gettext' installed to compile Mono."
echo "Get ftp://alpha.gnu.org/gnu/gettext-0.10.35.tar.gz"
echo "(or a newer version if it is available)"
DIE=1
}
}
(automake --version) < /dev/null > /dev/null 2>&1 || {
echo
echo "**Error**: You must have \`automake' installed to compile Mono."
echo "Get ftp://ftp.gnu.org/pub/gnu/automake-1.3.tar.gz"
echo "(or a newer version if it is available)"
DIE=1
NO_AUTOMAKE=yes
}
# if no automake, don't bother testing for aclocal
test -n "$NO_AUTOMAKE" || (aclocal --version) < /dev/null > /dev/null 2>&1 || {
echo
echo "**Error**: Missing \`aclocal'. The version of \`automake'"
echo "installed doesn't appear recent enough."
echo "Get ftp://ftp.gnu.org/pub/gnu/automake-1.3.tar.gz"
echo "(or a newer version if it is available)"
DIE=1
}
if test "$DIE" -eq 1; then
exit 1
fi
if test -z "$*"; then
echo "**Warning**: I am going to run \`configure' with no arguments."
echo "If you wish to pass any to it, please specify them on the"
echo \`$0\'" command line."
echo
fi
case $CC in
xlc )
am_opt=--include-deps;;
esac
if grep "^AM_PROG_LIBTOOL" configure.in >/dev/null; then
if test -z "$NO_LIBTOOLIZE" ; then
echo "Running libtoolize..."
${LIBTOOL}ize --force --copy
fi
fi
echo "Running aclocal $ACLOCAL_FLAGS ..."
aclocal $ACLOCAL_FLAGS || {
echo
echo "**Error**: aclocal failed. This may mean that you have not"
echo "installed all of the packages you need, or you may need to"
echo "set ACLOCAL_FLAGS to include \"-I \$prefix/share/aclocal\""
echo "for the prefix where you installed the packages whose"
echo "macros were not found"
exit 1
}
if grep "^AM_CONFIG_HEADER" configure.in >/dev/null; then
echo "Running autoheader..."
autoheader || { echo "**Error**: autoheader failed."; exit 1; }
fi
echo "Running automake --gnu $am_opt ..."
automake --add-missing --gnu $am_opt ||
{ echo "**Error**: automake failed."; exit 1; }
echo "Running autoconf ..."
autoconf || { echo "**Error**: autoconf failed."; exit 1; }
if test -d $srcdir/libgc; then
echo Running libgc/autogen.sh ...
(cd $srcdir/libgc ; NOCONFIGURE=1 ./autogen.sh "$@")
echo Done running libgc/autogen.sh ...
fi
conf_flags="--enable-maintainer-mode --enable-compile-warnings" #--enable-iso-c
if test x$NOCONFIGURE = x; then
echo Running $srcdir/configure $conf_flags "$@" ...
$srcdir/configure $conf_flags "$@" \
&& echo Now type \`make\' to compile $PKG_NAME || exit 1
else
echo Skipping configure process.
fi
libtoolize --automake
automake -a
autoheader
aclocal $ACLOCAL_FLAGS
autoconf
./configure $*
+10 -1280
View File
File diff suppressed because it is too large. Load diff
-4
View File
@@ -1,4 +0,0 @@
Makefile.in
Makefile
config
.cvsignore
File diff suppressed because it is too large. Load diff
-12
View File
@@ -1,12 +0,0 @@
monodir = $(sysconfdir)/mono
EXTRA_DIST = config.in \
machine.config \
DefaultWsdlHelpGenerator.aspx \
browscap.ini
mono_DATA = config \
machine.config \
DefaultWsdlHelpGenerator.aspx \
browscap.ini
-13082
View File
File diff suppressed because it is too large. Load diff
-10
View File
@@ -1,10 +0,0 @@
<configuration>
<dllmap dll="cygwin1.dll" target="@LIBC@" />
<dllmap dll="libc" target="@LIBC@" />
<dllmap dll="libintl" target="@INTL@" />
<dllmap dll="libxslt.dll" target="libxslt@libsuffix@" />
<dllmap dll="libmySQL.dll" target="libmysqlclient@libsuffix@" />
<dllmap dll="odbc32.dll" target="libodbc@libsuffix@" />
<dllmap dll="oci" target="clntsh" />
<dllmap dll="db2cli" target="libdb2_36@libsuffix@"/>
</configuration>
-216
View File
@@ -1,216 +0,0 @@
<?xml version="1.0" encoding="utf-8"?>
<configuration>
<configSections>
<section name="appSettings" type="System.Configuration.NameValueFileSectionHandler, System, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
<section name="system.diagnostics" type="System.Diagnostics.DiagnosticsConfigurationHandler, System, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
<section name="system.runtime.remoting" type="System.Configuration.IgnoreSectionHandler, System, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" allowLocation="false"/>
<sectionGroup name="system.web">
<section name="httpHandlers"
type="System.Web.Configuration.HttpHandlersSectionHandler, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<section name="httpModules"
type="System.Web.Configuration.HttpModulesConfigurationHandler, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<section name="machineKey"
type="System.Web.Configuration.MachineKeyConfigHandler, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<section name="authentication"
type="System.Web.Configuration.AuthenticationConfigHandler, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<section name="authorization"
type="System.Web.Configuration.AuthorizationConfigHandler, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<section name="globalization"
type="System.Web.Configuration.GlobalizationConfigurationHandler, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<section name="sessionState"
type="System.Web.SessionState.SessionStateSectionHandler, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<section name="trace"
type="System.Web.Configuration.TraceConfigurationHandler, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<section name="webServices"
type="System.Web.Services.Configuration.WebServicesConfigurationSectionHandler, System.Web.Services, Version=1.0.5000.0 Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<section name="webControls"
type="System.Web.Configuration.WebControlsSectionHandler, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<section name="compilation"
type="System.Web.Configuration.CompilationConfigurationHandler, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<section name="httpRuntime"
type="System.Web.Configuration.HttpRuntimeConfigurationHandler, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<section name="customErrors"
type="System.Web.Configuration.CustomErrorsConfigHandler, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<section name="pages"
type="System.Web.Configuration.PagesConfigurationHandler, System.Web, Version=1.0.5000.0, Culture=neutral" />
<section name="clientTarget"
type="System.Web.Configuration.ClientTargetSectionHandler, System.Web, Version=1.0.5000.0, Culture=neutral" />
</sectionGroup>
<sectionGroup name="system.net">
<section name="settings"
type="System.Net.Configuration.NetConfigurationHandler, System, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
<section name="connectionManagement"
type="System.Net.Configuration.ConnectionManagementHandler, System, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
<section name="authenticationModules"
type="System.Net.Configuration.NetAuthenticationModuleHandler, System, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
<section name="defaultProxy"
type="System.Net.Configuration.DefaultProxyHandler, System, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
<section name="webRequestModules"
type="System.Net.Configuration.WebRequestModuleHandler, System, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
</sectionGroup>
<section name="system.drawing" type="System.Configuration.NameValueSectionHandler, System, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
</configSections>
<system.net>
<connectionManagement>
<add address="*" maxconnection="2" />
</connectionManagement>
<authenticationModules>
<add type="System.Net.BasicClient, System, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
<add type="System.Net.DigestClient, System, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
<add type="System.Net.NtlmClient, System, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
</authenticationModules>
<defaultProxy>
<proxy />
<!-- bypassonlocal and proxyaddress are ok -->
<!-- usessystemdefault is not supported -->
</defaultProxy>
<webRequestModules>
<add prefix="http" type="System.Net.HttpRequestCreator, System, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
<add prefix="https" type="System.Net.HttpRequestCreator, System, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
<add prefix="file" type="System.Net.FileWebRequestCreator, System, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
</webRequestModules>
<settings>
<ipv6 enabled="false"/>
</settings>
</system.net>
<system.web>
<httpHandlers>
<add verb="*" path="Trace.axd" type="System.Web.Handlers.TraceHandler, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<add verb="*" path="*.aspx" type="System.Web.UI.PageHandlerFactory, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<add verb="*" path="*.asmx" validate="false" type="System.Web.Services.Protocols.WebServiceHandlerFactory, System.Web.Services, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<add verb="*" path="*.ashx" type="System.Web.UI.SimpleHandlerFactory, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<add verb="GET" path="WebResource.axd" type="System.Web.Handlers.AssemblyResourceLoader, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<add verb="*" path="*.asax" type="System.Web.HttpForbiddenHandler, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<add verb="*" path="*.ascx" type="System.Web.HttpForbiddenHandler, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<add verb="*" path="*.config" type="System.Web.HttpForbiddenHandler, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<add verb="*" path="*.cs" type="System.Web.HttpForbiddenHandler, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<add verb="*" path="*.rem" type="System.Runtime.Remoting.Channels.Http.HttpRemotingHandlerFactory, System.Runtime.Remoting, Culture=neutral, PublicKeyToken=b77a5c561934e089" validate="false" />
<add verb="*" path="*.soap" type="System.Runtime.Remoting.Channels.Http.HttpRemotingHandlerFactory, System.Runtime.Remoting, Culture=neutral, PublicKeyToken=b77a5c561934e089" validate="false" />
<add verb="GET,HEAD" path="*" type="System.Web.StaticFileHandler, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<add verb="*" path="*" type="System.Web.HttpMethodNotAllowedHandler, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
</httpHandlers>
<httpModules>
<add name="FormsAuthentication" type="System.Web.Security.FormsAuthenticationModule, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<add name="UrlAuthorization" type="System.Web.Security.UrlAuthorizationModule, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<add name="Session" type="System.Web.SessionState.SessionStateModule, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
<add name="OutputCache" type="System.Web.Caching.OutputCacheModule, System.Web, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a" />
</httpModules>
<authentication mode="Forms">
<forms name=".MONOAUTH" loginUrl="login.aspx" protection="All" timeout="30" path="/">
<credentials passwordFormat="Clear">
<!--<user name="gonzalo" password="gonz"/>-->
</credentials>
</forms>
</authentication>
<machineKey validationKey="AutoGenerate" decryptionKey="AutoGenerate" validation="SHA1" />
<globalization requestEncoding="utf-8"
responseEncoding="utf-8"
fileEncoding="utf-8"/>
<!--
culture="en-US"
uiculture="en-US" />
-->
<sessionState mode="InProc" />
<webServices>
<protocols>
<add name="HttpSoap"/>
<add name="HttpPost"/>
<add name="HttpGet"/>
<add name="Documentation"/>
</protocols>
<wsdlHelpGenerator href="DefaultWsdlHelpGenerator.aspx" />
</webServices>
<webControls clientScriptsLocation="/web_scripts" />
<compilation debug="false" defaultLanguage="c#">
<compilers>
<compiler language="cs;c#;csharp" extension=".cs" warningLevel="1" compilerOptions=""
type="Microsoft.CSharp.CSharpCodeProvider, System, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
<compiler language="vb;vbs;visualbasic;vbscript" extension=".vb"
warningLevel="1" compilerOptions=""
type="Microsoft.VisualBasic.VBCodeProvider, System, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
</compilers>
<assemblies>
<!--<add assembly="mscorlib" /> -->
<add assembly="System" />
<add assembly="System.Xml" />
<add assembly="System.Data" />
<add assembly="System.Web" />
<add assembly="System.Web.Services" />
<add assembly="System.Drawing" />
<add assembly="*" /> <!-- Add assemblies in bin directory -->
</assemblies>
</compilation>
<httpRuntime executionTimeout="90"
maxRequestLength="4096"
useFullyQualifiedRedirectUrl="false"
minFreeThreads="8"
minLocalRequestFreeThreads="4"
appRequestQueueLimit="100" />
<pages buffer="true"
enableSessionState="true"
autoEventWireup="true"
validateRequest="true" />
<clientTarget>
<add alias="ie5" userAgent="Mozilla/4.0 (compatible; MSIE 5.5; Windows NT 4.0)" />
<add alias="ie4" userAgent="Mozilla/4.0 (compatible; MSIE 4.0; Windows NT 4.0)" />
<add alias="uplevel" userAgent="Mozilla/4.0 (compatible; MSIE 4.0; Windows NT 4.0)" />
<add alias="downlevel" userAgent="Unknown" />
</clientTarget>
</system.web>
<system.runtime.remoting>
<application>
<channels>
<channel ref="http client" displayName="http client (delay loaded)" delayLoadAsClientChannel="true" />
<channel ref="tcp client" displayName="tcp client (delay loaded)" delayLoadAsClientChannel="true" />
</channels>
</application>
<channels>
<channel id="http" type="System.Runtime.Remoting.Channels.Http.HttpChannel, System.Runtime.Remoting, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
<channel id="http client" type="System.Runtime.Remoting.Channels.Http.HttpClientChannel, System.Runtime.Remoting, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
<channel id="http server" type="System.Runtime.Remoting.Channels.Http.HttpServerChannel, System.Runtime.Remoting, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
<channel id="tcp" type="System.Runtime.Remoting.Channels.Tcp.TcpChannel, System.Runtime.Remoting, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
<channel id="tcp client" type="System.Runtime.Remoting.Channels.Tcp.TcpClientChannel, System.Runtime.Remoting, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
<channel id="tcp server" type="System.Runtime.Remoting.Channels.Tcp.TcpServerChannel, System.Runtime.Remoting, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
</channels>
<channelSinkProviders>
<clientProviders>
<formatter id="soap" type="System.Runtime.Remoting.Channels.SoapClientFormatterSinkProvider, System.Runtime.Remoting, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
<formatter id="binary" type="System.Runtime.Remoting.Channels.BinaryClientFormatterSinkProvider, System.Runtime.Remoting, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
</clientProviders>
<serverProviders>
<formatter id="soap" type="System.Runtime.Remoting.Channels.SoapServerFormatterSinkProvider, System.Runtime.Remoting, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
<formatter id="binary" type="System.Runtime.Remoting.Channels.BinaryServerFormatterSinkProvider, System.Runtime.Remoting, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
<provider id="wsdl" type="System.Runtime.Remoting.MetadataServices.SdlChannelSinkProvider, System.Runtime.Remoting, Version=1.0.5000.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
</serverProviders>
</channelSinkProviders>
</system.runtime.remoting>
<appSettings>
<!--<add key="yourkey" value="your value" /> -->
<!--<remove key="a key defined higher in the hierarchy" /> -->
<!--<clear/> Removes all defined settings -->
</appSettings>
<system.diagnostics>
<trace autoflush="false" indentsize="4" />
</system.diagnostics>
<system.drawing>
</system.drawing>
<strongNames>
<pubTokenMapping>
<!-- ECMA key -->
<map Token="b77a5c561934e089" PublicKey="002400000480000094000000060200000024000052534131000400000100010079159977d2d03a8e6bea7a2e74e8d1afcc93e8851974952bb480a12c9134474d04062447c37e0e68c080536fcf3c3fbe2ff9c979ce998475e506e8ce82dd5b0f350dc10e93bf2eeecf874b24770c5081dbea7447fddafa277b22de47d6ffea449674a4f9fccf84d15069089380284dbdd35f46cdff12a1bd78e4ef0065d016df" />
<!-- Microsoft (final) key -->
<map Token="b03f5f7f11d50a3a" PublicKey="002400000480000094000000060200000024000052534131000400000100010079159977d2d03a8e6bea7a2e74e8d1afcc93e8851974952bb480a12c9134474d04062447c37e0e68c080536fcf3c3fbe2ff9c979ce998475e506e8ce82dd5b0f350dc10e93bf2eeecf874b24770c5081dbea7447fddafa277b22de47d6ffea449674a4f9fccf84d15069089380284dbdd35f46cdff12a1bd78e4ef0065d016df" />
<!-- Microsoft (Web Service Enhancement) key -->
<map Token="31bf3856ad364e35" PublicKey="002400000480000094000000060200000024000052534131000400000100010079159977d2d03a8e6bea7a2e74e8d1afcc93e8851974952bb480a12c9134474d04062447c37e0e68c080536fcf3c3fbe2ff9c979ce998475e506e8ce82dd5b0f350dc10e93bf2eeecf874b24770c5081dbea7447fddafa277b22de47d6ffea449674a4f9fccf84d15069089380284dbdd35f46cdff12a1bd78e4ef0065d016df" />
<!-- IBM (DB2 Data Provider) key -->
<map Token="7c307b91aa13d208" PublicKey="002400000480000094000000060200000024000052534131000400000100010079159977d2d03a8e6bea7a2e74e8d1afcc93e8851974952bb480a12c9134474d04062447c37e0e68c080536fcf3c3fbe2ff9c979ce998475e506e8ce82dd5b0f350dc10e93bf2eeecf874b24770c5081dbea7447fddafa277b22de47d6ffea449674a4f9fccf84d15069089380284dbdd35f46cdff12a1bd78e4ef0065d016df" />
</pubTokenMapping>
</strongNames>
</configuration>
+3
View File
@@ -0,0 +1,3 @@
Makefile.in
Makefile
all-docs
+19
View File
@@ -0,0 +1,19 @@
WEB_FILES= \
c-sharp class-library contact contributing documentation download \
faq gcc-frontend ideas index passport rationale resources \
roadmap runtime status team testing thanks tools
OTHERS= pending resources-pending todo
EXTRA_DIST = $(WEB_FILES) README $(OTHERS)
all-docs: $(WEB_FILES)
cat rationale roadmap c-sharp tools class-library \
runtime documentation download faq contributing \
resources status > all-docs
webit:
(cd web; make && make push)
push-notes:
scp release-notes/mono* www@www:/web/cvsmodules/mono/archive
+17
View File
@@ -0,0 +1,17 @@
To edit the website appearance:
cd doc/web
edit template.html.in
make
To edit website content:
cd doc (this directory)
edit files carefully
cd web
make
To publish changes:
cd web
make push
+151
View File
@@ -0,0 +1,151 @@
* MCS: The Ximian C# compiler
MCS began as an experiment to learn the features of C# by
writing a large C# program. MCS is currently able to parse C#
programs and create an internal tree representation of the
program. MCS can parse itself.
Work is progressing quickly on various fronts in the C#
compiler. Recently I started using the System.Reflection API
to load system type definitions and avoid self-population of
types in the compiler and dropped my internal Type
representation in favor of using the CLI's System.Type.
** Phases of the compiler
The compiler has a number of phases:
<ul>
* Lexical analyzer: hand-coded lexical analyzer that
provides tokens to the parser.
* The Parser: the parser is implemented using Jay (A
Berkeley Yacc port to Java, that I ported to C#).
The parser does minimal work and syntax checking,
and only constructs a parsed tree.
Each language element gets its own class. The code
convention is to use an uppercase name for the
language element. So a C# class and its associated
information is kept in a "Class" class, a "struct"
in a "Struct" class and so on. Statements derive
from the "Statement" class, and Expressions from the
Expr class.
* Parent class resolution: before the actual code
generation, we need to resolve the parents and
interfaces for interface, classe and struct
definitions.
* Semantic analysis: since C# can not resolve in a
top-down pass what identifiers actually mean, we
have to postpone this decision until the above steps
are finished.
* Code generation: nothing done so far, but I do not
expect this to be hard, as I will just use
System.Reflection.Emit to generate the code.
</ul>
<a name="tasks">
** Current pending tasks
Simple tasks:
<ul>
* Array declarations are currently being ignored,
* PInvoke declarations are not supported.
* Pre-processing is not supported.
* Attribute declarations and passing currently ignored.
* Compiler does not pass around line/col information from tokenizer for error reporting.
* Jay does not work correctly with `error'
productions, making parser errors hard to point. It
would be best to port the Bison-To-Java compiler to
become Bison-to-C# compiler ([email protected]
might have more information)
</ul>
Critical tasks:
<ul>
* Resolve "base" classes and "base" interfaces for
classes, structs and interfaces.
Once this is done, we can actually do the semantic
analysis, because otherwise we do not know who our
parents are.
</ul>
Interesting tasks:
<ul>
* Finishing the JB port from Java to C#. If you are
interested in working on this, please contact Brian
Jepson (bjepson at oreilly d-o-t com).
More on JB at: <a href="http://www.cs.colorado.edu/~dennis/software/jb.html">
http://www.cs.colorado.edu/~dennis/software/jb.html</a>
JB will allow us to move from the Berkeley Yacc
based Jay to a Bison-based compiler (better error
reporting and recovery).
</ul>
** Questions and Answers
Q: Why not write a C# front-end for GCC?
A: I wanted to learn about C#, and this was an exercise in this
task. The resulting compiler is highly object-oriented, which has
lead to a very nice, easy to follow and simple implementation of
the compiler.
I found that the design of this compiler is very similar to
Guavac's implementation.
Targeting the CIL/MSIL byte codes would require to re-architecting
GCC, as GCC is mostly designed to be used for register machines.
The GCC Java engine that generates Java byte codes cheats: it does
not use the GCC backend; it has a special backend just for Java, so
you can not really generate Java bytecodes from the other languages
supported by GCC.
Q: If your C# compiler is written in C#, how do you plan on getting
this working on a non-Microsoft environment.
We will do this through an implementation of the CLI Virtual
Execution System for Unix (our JIT engine).
Q: Do you use Bison?
A: No, currently I am using Jay which is a port of Berkeley Yacc to
Java that I later ported to C#. This means that error recovery is
not as nice as I would like to, and for some reason error
productions are not being caught.
In the future I want to port one of the Bison/Java ports to C# for
the parser.
Q: How do I compile it?
A: Compiling MCS currently requires you to run my port of <a
href="http://primates.ximian.com/~miguel/code/jay.cs.tar.gz">Jay to
C#</a> on a Unix system to generate the parser, and then you need
to use Microsoft's .NET csc.exe compiler to compile the compiler.
You only need to compile the compiler compiler (C code), the
samples are Java samples that I did not port, and you do not need
them.
It might be simple to port Jay.cs to Windows, but I have not tried
this.
You might also want to look at the <a href="faq.html#gcc">GCC</a>
section on the main FAQ
+174
View File
@@ -0,0 +1,174 @@
* The Class Library
The Class Library should be compatible with Microsoft's .NET
implementation.
Please see the <a href="class-status.html">Class Status</a>
page for a status of who is working on which classes.
We will write as much code as possible in C#. We may need to
interface with code written in C to gain access to the
functionality of libraries like libart, Gtk+, and libc.
** Contributing
We welcome contributions to the the Class Library. To get
started, check the status page for information about which
APIs are being worked on, and how to get in touch with
individual maintainers.
If you want to work on a class, first check the <a
href="download.html">Classes Distribution</a> to see if it is
not implemented yet, if not, check the <a
href="class-status.html">Class Status</a> to see if someone is
already working on it, and maybe contact them.
If nobody is working on it, mail <a
href="mailto:[email protected]">[email protected]</a>
with the class you want to implement and CC <a
href="mailto:[email protected]">[email protected]</a>
** Layout
The Class Library resides in the `mcs' module in the directoy
`class'.
Each directory in the directory represents the assembly where
the code belongs to, and inside each directory we divide the
code based on the namespace they implement.
There are two cases when we should consider portability: when
we are dealing with a couple of classes only that differ from
system to system (Consider System.Net and System.IO for Win32
and Unix). In those cases we will just place the files for
example on <t>corlib/System/System.IO/Unix-Console.cs</t> and
<t>corlib/System/System.IO/Win32-Console.cs</t>.
For classes that might differ more (for example, the
implementation of Windows.Forms), we might have different
directories altogether: <t>System.Windows.Forms/Win32</t>,
<t>System.Windows.Forms/Gtk+</t> and
<t>System.Windows.Forms/Cocoa</t>.
** Using existing components from GNOME.
Our current plan is to implement the GUI tools on top of
Gtk+. The only obstacle here is that applications from Windows
might expect to be able to pull the HWND property from the
widgets and use PInvoke to call Windows functions.
** Class Library and Win32 dependencies.
There are a few spots where the Win32 foundation is exposed to
the class library (for example, the HDC and HWND properties in
the GDI+). Casual inspection suggests that these can be
safely mapped to Gdk's GC and GdkWindow pointers without
breaking anything.
The only drawback is that support for PInvoke of Win32 code
won't be available. An alternate solution would be to use
portions of Wine, or even to use Wine as our toolkit.
*** Initial GDI+ and WinForms implementation
The initial implementation will use Gtk+ as the underlying
toolkit. Since GTK+ has already been ported to many windowing
systems other than X (including frame buffer, Win32, and BeOS)
its use should cover most applications for most users.
*** Database access
We will implement ADO.NET functionality by reusing <a
href="http://www.gnome-db.org">GNOME-DB</a>. This is an ideal
choice, since GNOME-DB was implemented precisely to provide an
ADO-like system for GNOME.
*** Component Integration
We will provide a new namespace to use GNOME specific features
as well as a namespace to host Bonobo interfaces and classes
in Mono.
** Licensing
The class library will be licensed under the terms of the GNU
LGPL. Some people have pointed out that the plain LGPL is
troublesome for embedded use of the Mono class libraries. So
we are considering to use the GPL with a special exception
(like the <a
href="http://www.gnu.org/software/classpath/classpath.html">GNU
Classpath</a> project did.
The exception to the GPL would be:
<i>The library is distributed under the terms of the GNU General
Public License with the following exception:
If you link this library against your own program, then you do not
need to release the source code for that program. However, any
changes that you make to the library itself, or to any native
methods upon which the library relies, must be re-distributed in
accordance with the terms of the GPL.</i>
If you are going to contribute, please keep in mind that we
might require you to agree that Ximian might adjust the
license to enable the use of the class libraries on embedded
systems or to develop proprietary applications using Mono.
We suggest that you assign the copyright of your work to the
GNOME Foundation or the Free Software Foundation to simplify
defending the code in case it is used inappropiately.
** Class Library testing
We need to write regression tests that will verify
the correctness of the class library, compiler, and JIT
engine.
** Coding conventions
Please follow the conventions on the ECMA specification (On
the Annex Partition) for your coding your libraries.
Use 8 space tabs for writing your code (hopefully we can keep
this consistent). If you are modifying someone else's code, try
to keep the coding style similar.
For a rationale on 8 space tabs, read Linus Torvald's Coding
Style guidelines in the Linux kernel source for a rationale.
*** Missing implementation bits
If you implement a class and you are missing implementation bits,
please put in the code the word "TODO" and a description of what
is missing to be implemented.
*** Tagging buggy code
If there is a bug in your implementation tag the problem by using
the word "FIXME" in the code, together with a description of the
problem.
Do not use XXX or obscure descriptions, because otherwise people
will not be able to understand what you mean.
*** Tagging Lame specs
Sometimes the specification will be lame (consider Version.ToString (fieldCount)
where there is no way of knowing how many fields are available, making the API
not only stupid, but leading to unreliable code).
In those cases, use the keyword "LAMESPEC".
** FAQ
Frequently asked questions about the class library:
Q: I am writing a new class that overrides one of the system classes,
and I am getting a 1595 warning from the compiler. Should we use a
different namespace?
A: There is a quick solution to the problem, you can pass the command
line argument /nowarn:1595 and this will effectively let you use
your implementation of the code, while overriding the ones from the
system assemblies.
+27
View File
@@ -0,0 +1,27 @@
* Status of the various pieces of the class library
You can browse the status of the class library and see who has
registered to work on what parts of the system. These list
work-in-progress components currently.
<ul>
* Per <a href="class-status/per-assembly.html">Assembly</a>
* Per <a href="class-status/per-completion.html">Completion</a>
* Per <a href="class-status/per-maintainer.html">Maintainer</a>
* Per <a href="class-status/per-unimplemented.html">Unimplemented</a>
</ul>
A better mechanism will soon be in place, I just did this
really quickly, as we already got one duplicated implementation.
You can also download the XML <a
href="class-status/maintainers.xml">maintainers</a> file that
contains the actual maintainers list.
I am sure we can generate better reports, if you are
interested in contributing, let <a
href="mailto:[email protected]">me</a> know.
+16
View File
@@ -0,0 +1,16 @@
* Contacting the Mono team.
You can contact the Mono Team by sending e-mail to `<a
href="mailto:[email protected]">mailto:[email protected]</a>'.
You can contact the general forum of discussion by sending
e-mail to <a href="mailto:[email protected]">[email protected]</a>
You can contact me (Miguel de Icaza) by sending e-mail to <a
href="mailto:[email protected]">[email protected]</a>. My web
page is <a
href="http://primates.ximian.com/~miguel">http://primates.ximian.com/~miguel"</a>
You can also <a
href="http://www.ximian.com/about/contact.php3">reach Ximian.</a>
+53
View File
@@ -0,0 +1,53 @@
* Contributing to the Mono project
There are many ways in which you can help in the Mono project:
<ul>
* <b>Programmers:</b> You can work on a free
implementation of the <a
href="class-library.html">class libraries</a>, the
<a href="runtime.html">runtime engine</a>, <a
href="tools.html">the tools</a>, the <a
href="testing.html">testing framework</a>
* <b>Writers:</b> You can help us with writing <a
href="documentation.html">documentation</a>.
</ul>
Those are just broad things that need to be worked on, but
something that would help tremendously would be to help with
small duties in the project that need to be addressed.
** Small tasks
A few smaller tasks are here, dropped in no particular order:
<ul>
* <b>MCS compilation process:</b> Currently MCS does
not build with a single `make' command. This should
be fixed (this being part of the `Joel Test' that
software has to pass).
* <b>Mono/doc and web site:</b> They need to be
packaged up in the official `distribution'
* <b>Enthusiasts: </b>
</ul>
* Special note
If you have looked at Microsoft's implementation of .NET or
their shared source code, you may not be able to contribute
to Mono. Details will follow when we know more about this.
In general be careful when you are implementing free software
and you have access to proprietary code. We need to make sure
that we are not using someone else's copyrighted code
accidentally.
Please do not use the <b>ildasm</b> program to disassemble
proprietary code when you are planning to reimplement a class
for Mono. If you have done this, we might not be able to use
your code.
Please stick to published documentation for implementing any
classes.
+50
View File
@@ -0,0 +1,50 @@
* Documentation
Although most of the concepts from Microsoft.NET can
be applied to the completed Mono platform, we do need to
have a complete set of free documentation written specifically
for Mono.
The documentation license we have chosen is the GNU Free
Documentation License (FDL), the standard for most documents
in the free software world.
We need documentation on a number of topics:
<ul>
* The development tools (compilers, assembler tools,
language reference, design time features).
* Frequently Asked Question compilations.
* HOWTO documents.
* The Class Libraries
* Tutorials on Mono and the specifics of running it.
* A guide to Mono as compared to the Microsoft.NET
Framework SDK
</ul>
** Class Library documentation
When contributing to the Class Library effort, please use the
inline XML documentation tags to document your classes so we
can automatically generate the documentation from the class
libraries.
If you provide examples, please do not embed them into the
source code, as that will make the source code harder to read
and maintain. Instead, put examples for your code into a
subdirectory of the class libraries. Make your sample
code a full standalone application that people can compile.
Ideally the Mono documentation browser will let you edit, modify
and run the sample programs.
+52
View File
@@ -0,0 +1,52 @@
* Software Availability
The Virtual Execution System is available in package `mono'.
Currently this contains a metadata library and the
disassembler. Please reffer to our <a
href="runtime.html">Runtime</a> description for more details
on this part of the project.
The code for the C# compiler as well as the language error
test suite and the class library are in the `mcs' package, we
will move this later into `mono' itself.
The mcs compiler requires the `jay' parser generator, this is
available in module `jay'.
** Sources
Currently we will be providing CVS snapshots every day, until
we make the first tarball release.
<ul>
<a name="july-15">
<b>July 15th, 2001</b>
<ul>
* <a href="archive/mcs-15-Jul-2001.tar.gz">mcs-15-Jul-2001.tar.gz</a>: CVS snapshot.
* <a href="archive/mono-0.4.tar.gz">mono-0.4.tar.gz</a>: Packaged Source Code.
* <a href="archive/mono-0.4">Release Notes</a>
</ul>
<a name="july-14">
<b>July 14th, 2001</b>
<ul>
* <a href="archive/jay.cs.tar.gz">archive/jay.cs.tar.gz</a>: Sources to Jay.
* <a href="archive/mcs-12-Jul-2001.tar.gz">mcs-12-Jul-2001.tar.gz</a>: CVS snapshot.
* <a href="archive/mono-0.3.tar.gz">mono-0.3.tar.gz</a>: Packaged Source Code.
</ul>
<a name="july-8">
<b>July 8th, 2001</b>
<ul>
* <a href="archive/jay.cs.tar.gz">archive/jay.cs.tar.gz</a>: Sources.
* <a href="archive/mcs-08-Jul-2001.tar.gz">mcs-08-Jul-2001.tar.gz</a>: CVS Snapshot
* <a href="archive/mono-08-Jul-2001.tar.gz">mono-08-Jul-2001.tar.gz</a>: CVS Snapshot
</ul>
</ul>
** CVS
We are trying to figure out where to put our CVS repository.
We are debating between the GNOME CVS or SourceForge. Watch
this spot.
+609
View File
@@ -0,0 +1,609 @@
** Basics
Q: Is Mono the same as Microsoft's .NET initiative?
A: It is not.
.NET is a company-wide initiative at Microsoft that
encompasses many different areas. The .NET development framework,
Passport, Biztalk, new server products, and anything that is
remotely connected to .NET gets the ".NET-stamping" treatment.
Some components of Microsoft's .NET initiative have been announced
and some others are in the works.
Mono is a project to implement several technologies developed by
Microsoft that have now been submitted to the ECMA Standards Body.
Q: What technologies are included in Mono?
A: Mono contains a number of components useful for building new
software:
<ul>
* A Common Language Infrastructure (CLI) virtual
machine that contains a class loader, Just-in-time
compiler, and a garbage collecting runtime.
* A class library that can work with any language
which works on the CLR.
* A compiler for the C# language. In the future we
might work on other compilers that target the Common
Language Runtime.
Windows has compilers that target the
virtual machine from
<a href="http://msdn.microsoft.com/net/thirdparty/default.asp#lang">a
number of languages:</a> Managed C++, Java Script,
Eiffel, Component Pascal, APL, Cobol, Oberon, Perl,
Python, Scheme, Smalltalk, Standard ML, Haskell,
Mercury and Oberon.
</ul>
The CLR and the Common Type System (CTS) enables applications and
libraries to be written in a number of languages. Classes and
methods created in one language can be used from a different
language.
This means for example that if you define a class to do algebraic
manipulation in C#, that class can be reused from any other
language that supports the CLI. You could create a class in C#,
subclass it in C++ and instantiate it in an Eiffel program.
A single object system, threading system, class libraries, and
garbage collection system can be shared across all these languages.
Q: Where can I find the specification for these technologies?
A: You can find the work-in-progress documentation from the T3G ECMA
group here:
<a href="http://www.dotnetexperts.com">http://www.dotnetexperts.com</a>
Q: Will you implement the .NET Framework SDK class libraries?
A: Yes, we will be implementing the APIs of the .NET Framework SDK
class libraries.
Q: What does Mono stand for?
A: Mono is the word for `Monkey' in Spanish. We like monkeys.
Q: When will you ship it?
A: We do not know when the code will be shipped. The more
contributions we get to the project, the sooner it will ship.
A rough estimate is that we might be able to run our C# compiler on
Linux by the end of the year. That means running the Windows
Executable generated by a Microsoft .NET compiler on the Linux
platform.
We expect that doing GUI applications will require more work on the
class libraries. That could take another six months.
Q: How can I contribute?
A: Check the <a href="contributing.html">contributing</a> section.
Q: You guys should innovate instead of copying.
A: In this particular case, we see a clear advantage in the platform
and we are interested in using the features of the CLI on open source systems.
We have decided that we should spend our limited resources towards
implementing an existing specification instead of designing and
implementing our own.
Designing and implementing our own would be possible, but it doesn't make
sense to do that just because the specification comes from a
proprietary vendor.
** Ximian
Q: Why is Ximian working on .NET?
A: We are interested in providing the best tools for programmers to
develop applications for Free Operating Systems.
For more information, read the project <a
href="rationale.html">rationale</a> page.
Q: Will Ximian be able to take on a project of this size?
A: Ximian will not be able to taken on the whole project on its own.
Mono will be a free software/open source community project, that is
the only way we can hope to implement something of this size. You
can <a href="contributing.html">contribute</a> to this effort.
Q: What pieces will Ximian be working on?
A: We will focus on building a development and execution
environment.
The idea is to get Mono to a state of that would allow
third parties to actually be able to use it real-world development.
Q: Why does Ximian even care?
A: We like the features that the CLI and its related technologies
bring to the table. An exciting addition to the developer toolkit.
The goal of Mono is to bring this technology to non-Windows
platforms (although we hope Mono will also run on Windows, for
debugging and comparative purposes).
** Licensing
Q: Will I be able to write proprietary applications that run with
Mono?
A: Yes. We will make sure that the licensing will allow developers to
create proprietary applications.
Q: What license is Mono on?
A: The C# Compiler is released under the terms of the GPL. The class
libraries will be under the LGPL or the GPL with a special
exception. The runtime libraries are under the LGPL.
Here is the exception for the class libraries:
The library is distributed under the terms of the GNU General
Public License with the following exception:
If you link this library against your own program, then you do not
need to release the source code for that program. However, any
changes that you make to the library itself, or to any native
methods upon which the library relies, must be re-distributed in
accordance with the terms of the GPL.
This is similar in spirit to <a
href="http://www.gnu.org/software/classpath/classpath.html">GNU
Classpath.</a>
Q: Will you accept code under the XXX License?
A: If the XXX License is compatible with the license we use in that
specific piece of code, then yes. If you want to use the BSD license, make
sure you use the BSD license without the advertisement clause (The
`Ousterhout License').
** Mono and .NET
Q: If applications use Mono, does that mean that I have to pay a service fee?
A: No. Mono is not related to Microsoft's initiative of
software-as-a-service.
Q: If you implement .NET, will I depend on Microsoft Passport to run my software?
A: No. The .NET Framework is a runtime infrastructure and collection
of class libraries. Passport may be required to access certain web
services written for that framework, but only if the programmer
chooses Passport as the authentication mechanism.
Q: Is .NET just a bunch of marketing slogans?
A: Although the `.NET initiative' is still quite nebulous, The .NET Framework
has been available for some time. Mono is not an implementation of the .NET
initiative, just the development framework.
Q: What is a 100% .NET application?
A: A `100% .NET application' is one that only uses the APIs defined
under the System namespace and does not use PInvoke. These
applications would in theory run unmodified on Windows, Linux,
HP-UX, Solaris, MacOS X and others.
Q: But Microsoft will release a port of the real thing under the
`Shared Source' license, why bother with anything else?
A: The Shared Source implementation will not be usable for commercial
purposes. We are working towards an implementation that will grant
a number of rights to recipients: use for any purpose,
redistribution, modification, and redistribution of modifications.
This is what we call <a
href="http://www.gnu.org/philosophy/free-sw.html">Free Software</a>
** Passport
Q: Is this a free implementation of Passport?
A: No. Passport is part of Microsoft's Hailstorm initiative. Mono
is just a runtime, a compiler and a set of class libraries.
Q: Will the System.Web.Security.PassportIdentity class, mean
that my software will depend on Passport?
A: No. That just means that applications might use that API to
contact a Passport site.
As long as your application does not use Passport, you will not
need Passport.
It might even be possible to implement that class with
a set of dummy functions, or use an alternate Passport implementation.
We do not know at this time whether the Passport protocol is
documented and whether we will be able to talk to
passport.com
Q: What is your opinion?
A: You can read my personal <a href="passport.html">opinion on
passport</a>.
** Mono and Windows
Q: Will Mono allow me to run Microsoft Office on Linux?
A: No, it will not. Microsoft Office is a Windows application. To
run Windows applications on Intel Unix systems refer to <a
href="http://www.winehq.com">the Wine Project</a>
** GNOME
Q: How is this related to GNOME?
A: In a number of ways:
* Mono will use existing
components that have been developed for GNOME when it makes
sense. For example on X systems, we will use Gtk+ and
Libart to implement Winforms and the Drawing2D API.
For database access, we will use LibGDA (not really
depending on GNOME, but related to).
* This project was born out of the need of providing improved
tools for the GNOME community.
* We would like to add support to our CLR implementation to
deal with GObjects (in GNOME 1.x, they are called
GtkObjects), and allow Mono developers to provide GObjects
or use and extend existing GObjects.
Q: Has the GNOME Foundation or the GNOME team adopted Mono?
A: Mono is too new to be adopted by those groups. We hope that the
tools that we will provide will be adopted by free software
programmers including the GNOME Foundation members and the GNOME
project generally.
Q: Should GNOME programmers switch over to Mono?
A: Mono will not be ready even within the next six months, and a
complete implementation is probably one year away.
We encourage GNOME developers to continue using the existing tools,
libraries and components. Improvements made to GNOME will have an
impact on Mono, as they will provide the "backend" for various
classes.
Q: Will Mono include compatibility with Bonobo components?
A: Yes, we will provide a set of classes for implementing and using
Bonobo components from within Mono.
** Mono and the Web
Q: Is Mono a way of running Java applets?
A: No.
** Web Services
Q: Is Mono just a new way of writing Web Services?
A: No.
Q: If this implements the SDK classes, will I be able to write and
execute .NET Web Services with this?
A: Yes, you will.
When the project is finished, you will be able to use the same
technologies that are available through the .NET Framework SDK on
Windows to write Web Services.
Q: What about Soup?
A: Soup is a library for GNOME applications to create SOAP server and
SOAP clients. You can browse the source code for soup using <a
href="http://cvs.gnome.org/bonsai">GNOME's Bonsai</a>
Q: Can I use CORBA?
A: Yes. The CLI contains enough information about a class that
exposing it to other RPC systems (like CORBA) is really simple, and
does not even require support from an object.
We will be implementing CORBA interoperation as an extension to the
Mono classes so that we can integrate with Bonobo, just like
Microsoft provides COM interoperation classes and support
mechanisms.
Q: Can I serialize my objects to other things other than XML?
A: Yes, although the serializing tools have not yet been planned, and
you would probably have to implement them yourself.
** Development Tools
Q: Will it be possible to use the CLI features without using bytecodes
or the JIT?
A: Yes. The CLI engine will be made available as a shared library.
The garbage collection engine, the threading abstraction, the
object system, the dynamic type code system and the JIT will be
available for C developers to integreate with their applications if
they wish to do so.
Q: Will you have new development tools?
A: Hopefully Free Software enthusiasts will contribute tools to
improve the developer environment. These tools could be developed
initially using Microsoft implementation of the CLI and then
executed later with Mono.
** Mono and Java
Q: What about using Java? After all there are many languages that
target the Java VM.
A: You can get very good tools for doing Java development on free
systems right now. <a href="http://www.redhat.com">Red Hat</a> has
contributed a <a href="http://gcc.gnu.org">GCC</a> <a
href="http://gcc.gnu.org/java">frontend for Java</a> that can take
Java sources or Java byte codes and generate native executables; <a
href="http://www.transvirtual.com">Transvirtual</a> has implemented
<a href="http://www.kaffe.org">Kaffe</a> a JIT engine for Java;
Intel also has a Java VM called <a
href="http://www.intel.com/research/mrl/orp">ORP</a>.
The JVM is not designed to be a general purpose virtual machine.
The Common Intermediate Language (CIL), on the other hand, is
designed to be a target for a
wide variety of programming languages, and has a set of rules
designed to be optimal for JITers.
Q: What kind of rules make the Common Intermediate Language useful for
JITers?
A: The main rule is that the stack in the CLI is not a general purpose
stack. You are not allowed to use it for other purposes than
computing values and passing arguments to functions or return
values.
At any given call or return instruction, the types on the stack
have to be the same independently of the flow of execution of your
code.
Q: I heard that the CIL is ideal for JITing and not efficient for
interpreters, is this the case?
A: The CIL is better suited to be JITed than JVM byte codes, but you
can interpret them as trivially as you can interpret JVM byte
codes.
** Extending Mono
Q: Would you allow other classes other than those in the
specification?
A: Yes. The Microsoft class collection is very big, but it is by no
means complete. It would be nice to have a port of `Camel' (the
Mail API used by Evolution inspired by Java Mail) for Mono
applications.
You might also want to look into implementing CORBA for Mono. Not
only because it would be useful, but because it sounds like a fun
thing to do, given the fact that the CLI is such a type rich
system.
For more information on extending Mono, see our <a
href="ideas.html">ideas</a> page.
** Mono and portability
Q: Will Mono only work on Linux?
A: Currently, we are doing our work on Linux-based systems and
Windows. We do not expect many Linux-isms in the code, so it
should be easy to port Mono to other UNIX variants.
Q: What about Mono on non X-based systems?
A: Our main intention at Ximian is to be able to develop GNOME
applications with Mono, but if you are interested in providing a
port of the Winform classes to other platforms (frame buffer or
MacOS X for example), we would gladly integrate them, as long
they are under a Free Software License.
** Reusing existing Code
Q: What projects will you reuse or build upon?
A: We want to get Mono in the hands of programmers soon. We are
interested in reusing existing open source software.
Q: What about Intel's research JIT framework, ORP?
A: At this time, we are investigating whether we can use elements of
ORP for Mono. ORP is a research JIT engine that has a clear
defined API that splits the JIT from the GC system and the actual
byte code implementation. It is a research product.
Q: What about using GNU Lightning?
A: We are also researching <a
href="http://www.gnu.org/software/lightning/lightning.html">GNU
Lightning</a>.
** Ximian and Microsoft
Q: Didn't Miguel de Icaza say that `Unix Sucks'?
A: Yes, he did, as a catch phrase in his opening remark on the Ottawa
Linux Symposium. His talk focused on various ways to improve Unix.
There is a paper describing some ways to improve Unix at:
<a href="http://primates.ximian.com/~miguel/bongo-bong.html">
http://primates.ximian.com/~miguel/bongo-bong.html</a>
Q: Didn't Ximian's Miguel work for Microsoft?
A: Actually, Nat Friedman (Ximian's co-founder) did work as an
intern for Microsoft for one summer but Miguel did not.
Q: Did Nat and Miguel meet at Microsoft?
A: They met online on the Linux IRC network; They met in person for
the first time in 1997.
** Mono and Microsoft
Q: How can you expect Mono to compete with Microsoft, wont this
require an effort too large?
A: You are right. Mono will never become a reality without the help
of other contributors. Ximian is a small company that can not
finish Mono alone. We will be working with members of the
community to deliver the product.
Q: Is Microsoft and Corel involved in the Mono implementation?
A: No, they are not.
Q: Are you writing Mono from the ECMA specs?
A: yes, we are writing them from the ECMA specs and the published
materials in print about .NET
** Acronyms
Q: What is the difference between CLR (Common Language Runtime) and
CLI (Common Language Infrastructure)?
A: CLI is the specification of an execution system. The Microsoft
implementation of this specification is named CLR.
Unless we come up with our own acronym, we could just call ours
also CLR, just because it would do exactly the same thing the
Microsoft implementation does.
<a name="gcc">
** Mono and GCC
Q: Should someone work on a GCC front-end to C#?
A: I would love if someone does, and we would love to help anyone that
takes on that task, but we do not have the time or expertise to
build a C# compiler with the GCC engine. I find it a lot more fun
personally to work on C# on a C# compiler, which has an intrinsic
beauty.
Q: Should someone make a GCC backend that will generate CIL images?
A: I would love to see a backend to GCC that generates CIL images. It
would provide a ton of free compilers that would generate CIL
code. This is something that people would want to look into
anyways for Windows interoperation in the future.
Q: What about making a front-end to GCC that takes CIL images and
generates native code?
A: I would love to see this, specially since GCC supports this same
feature for Java Byte Codes. You could use the metadata library
from Mono to read the byte codes (ie, this would be your
"front-end") and generate the trees that get passed to the
optimizer.
Ideally our implementation of the CLI will be available as a shared
library that could be linked with your application as its runtime
support.
Q: But would this work around the GPL in the GCC compiler and allow
people to work on non-free front-ends?
A: People can already do this by targeting the JVM byte codes (there
are about 130 compilers for various languages that target the JVM).
Q: Why are you writing a JIT engine instead of a front-end to GCC?
A: The JIT engine and runtime engine will be able to execute CIL
executables generated on Windows.
** Mono and GNU
Q: Is Mono part of the GNU project?
A: Mono is a Ximian project, that is also part of the GNU project.
** Mono and Portability
Q: Will Mono work on other variants of Unix?
A: Yes. We do not expect to add any gratuitous incompatibilities.
Q: Will Mono run on Windows?
A: Hopefully yes. Currently some parts of Mono only run on Windows
(the C# compiler is a .NET executable) and other parts have only
been compiled on Linux, but work on Windows with Cygwin.
Q: Will Mono depend on GNOME?
A: It will depend only if you are using a particular assembly (for
example, for doing GUI applications). If you are just interested
in Mono for implementing a `Hello World Enterprise P2P Web
Service', you will not need any GNOME component.
** Performance
Q: How fast will be Mono?
A: We can not predict the future, but a conservative estimate is that
it would be at least `as fast as other JIT engines'.
Now, wishfully thinking I hope that we will ship various JITs with
Mono just like Microsoft has done. A fast JITer when maximum
performance is not needed, but fast load times are important; And
an optimizing JITer that would be slower at generating code but
produce more optimal output.
The CIL has some advantages over the Java byte code: it is really
an intermediate representation and there are a number of
restrictions on how you can emit CIL code that simplify creating
better JIT engines.
For example, on the CIL the stack is not really an abstraction
available for the code generator to use at will: it is just a way
of creating a postfix representation of the parsed tree. At any
given call point or return point, the contents of the stack are
expected to contain the same object types independently of how the
instructions was reached.
** Assorted questions
Q: You say that the CLI allows multiple languages to execute on the
same environment. Isn't this the purpose of CORBA?
A: CORBA can be used for this purpose, but think of CORBA mostly as a
inter-process (local machine or remote machine) communication
system.
Think of the CLI to be targeted to interoperation of languages in
the same process, and CORBA as being a nifty addition for
out-of-process.
Sure, we have used CORBA for in-proc components, but the lack of an
ABI for in-proc components does not translate into a good framework
for in-proc systems
+9
View File
@@ -0,0 +1,9 @@
* The GCC front-end
The GCC front-end will accept input in a binary file with
codes in the Common Intermediate Language (CIL), and generate
native code.
This will allow pre-compilation and full optimization to take
place before a program is executed.
+40
View File
@@ -0,0 +1,40 @@
* Ideas
Here are a few ideas of tools, classes and projects that you
could start. More are forthcoming.
<a name="classes">
** Classes
<ul>
* Implement a JXTA protocol implementation:
<a href="http://www.jxta.org">http://www.jxta.org</a>
* Implement a Mail API, similar to Camel or JavaMail (Camel has
significant architecture features that are required on a real
mailer).
You can check the current C
<a href="http://cvs.gnome.org/bonsai/rview.cgi?dir=evolution%2Fcamel">
Camel implementation</a>.
* Interfacing to Multimedia systems. You might want
to look into the Quicktime API. I know <a
href="mailto:[email protected]">Vladimir</a> has
researched the problem before
</ul>
<a name="projects">
** Projects
<ul>
* Implement an xmlStorageSystem for the CLI:
<a href="http://www.soapware.org/xmlStorageSystem">
http://www.soapware.org/xmlStorageSystem</a>
* Build a CORBA interoperability engine for the CLR.
You do not need to do all of the work, just talking
the protocol will get us a long way (<a
href="http://www.omg.org">The OMG site</a> has the CORBA specs).
</ul>
+45
View File
@@ -0,0 +1,45 @@
** Jul 15, 2001
Another release of Mono is out, check the <a
href="archive/mono-0.4">Mono 0.4 Release Notes</a>. Get it <a
href="download.html#july-15">here</a>.
** Jul 14, 2001
A <a
href="http://mail.ximian.com/archives/public/mono-list/2001-July/000399.html">new
release</a> of the
runtime, compiler and classes has been made. Get it <a href="download.html#july-14">here</a>
** Jul 12, 2001
I keep getting questions about my opinion on Passport, even when
Mono has <b>nothing</b> to do with it. I finally <a
href="passport.html">wrote something.</a>
** Jul 9, 2001
Today <a href="http://www.ximian.com">Ximian</a> announced the
launch of the Mono
project, an effort to create a free software
(sometimes called open source) implementation of the .NET
Development Framework.
If you have questions about the
project, please read our list of <a
href="faq.html">Frequently Asked Questions</a>.
You might also want to <a href="download.html">Download the
source</a> for our work so far.
You might want to <a
href="resources.html#mailing">subscribe</a> to our mono-list
and mono-announce-list
You can contact the team at: <a
href="mailto:[email protected]">[email protected]</a>
** O'Reilly
Brian posted a story on <a
href="http://www.oreillynet.com/dotnet">O'Reilly Network .NET</a>
+231
View File
@@ -0,0 +1,231 @@
* Microsoft Hailstorm and Passport
Microsoft Passport is a centralized database hosted by
Microsoft that enhances the consumer experience with the Web
by providing a single logon system that they can use across a
number of participant web sites.
As you might know by now from our extensive <a
href="faq.html">FAQ</a>, the Mono project has nothing to do
with Microsoft Hailstorm or <a
href="http://www.passport.com">Microsoft Passport.</a>
Still a lot of people have asked us our opinion on them.
** Passport
Passport is important not because of it being breaktrough
technologically speaking, but because the company is in a
position to drive most people toward being suscribers of it.
At the time of this writing passport is required to use the
free mail service <a href="http://www.hotmail.com">Hotmail</a>
to get customized support for the <a
href="http://www.msn.com">MSN portal</a>, <a
href="http://msdn.microsoft.com">Microsoft Developers
Network</a> and according to the original announcement from
Microsoft <a href="http://www.americanexpress.com">American
Express</a> and <a href="http://www.ebay.com">EBay</a> will be
adopting it.
There is already a <a
href="http://www.passport.com/Directory/Default.asp?PPDir=C&lc=1033">Large
list</a> of participating sites.
There are many current users of it and Microsoft will be
driving more users towards Passport as it <a
href="http://news.cnet.com/news/0-1003-200-6343275.html">integrates
it</a> in their upcoming release of Windows.
Microsoft has also <a
href="http://www.passport.com/Business/JoinPassportNetwork.asp?lc=1033">developed
a toolkit</a> to enable current web merchants to integrate
their services with passport.
To the end user, there is a clear benefit: they only have to
log into a single network and not remember multiple passwords
across sites on the internet. Companies that adopt passport
will have a competition advantage over those that dont.
Microsoft lists a list of <a
href="http://www.passport.com/Business/Default.asp?lc=1033">benefits</a>
to companies.
** The problems of Passport
There are a number of concerns that different groups have over
Passport. Sometimes I have some, sometimes I do not. But
overall, consumers and businesses can have better solutions.
<ul>
* <b>Single Point of Failure:</b> As more services and
components depend on remote servers, functionality can
grind to a halt if there is a failure on the
centralized Passport system.
Such a failure was predicted, and we recently <a
href="http://news.cnet.com/news/0-1005-200-6473003.html">witnessed</a>
got a lot of people worried.
The outgage lasted for seven days. Think what this
could do to your business.
* <b>Trust:</b> Not everyone trusts Microsoft to keep
their information confidential. Concerns are not only
at the corporate level policy, but also the fact that
the source code for Microsoft products is not
available, means that trojans or worms could be built
into the products by malicious engineers. This is not
unheard of, as the <a
href="http://slashdot.org/articles/00/04/14/0619206.shtml">Microsoft
Internet Server</a> had a trojan horse built into that
allowed anyone that knew about this to control any
server running IIS.
* <b>Security:</b> With a centralized system like
Passport, imagine the repercussions of a malicious
hacker gaining access to the Passport database.
Personal information and credit card information about
almost everyone using a computer could be stored there.
Hackers have already <a
href="http://slashdot.org/articles/00/10/27/1147248.shtml">broken
into Microsoft</a> in the past. And the company was
unable to figure out for how long their systems had been hacked.
</ul>
Microsoft might or might not realize this. The idea behind
Passport is indeed a good one (I can start to get rid of my
file that keeps track of the 30 logins and passwords or so
that I use across the various services on the net myself).
** Alternatives to Microsoft Passport
An alternative to Microsoft Passport needs to take the above
problems into consideration. Any solution of the form `We
will just have a competing offering' will not work.
The system thus has to be:
<ul>
* <b>Distributed:</b> The entire authentication
system should not create an internet `blackout' in the
case of failure.
* <b>Allow for multiple registrars:</b> Users should
be able to choose a registrar (their banks, local
phone company, service provider, Swiss bank, or any
other entity they trust.
* <b>Mandate good security measures:</b> As a
principle, only Open Source software should be used
for servers in the registrar, and they should conform
to a standard set of tools and software that can be
examined by third parties.
</ul>
An implementation of this protocol could use the DNS or a
DNS-like setup to distribute the information of users with the
possibility of replicating and caching public information
about the user.
For instant messaging (another pieces of the Hailstorm bit),
you want to use a non-centralized system like Sun's <a
href="http://www.jxta.org">JXTA</a>. Some people mailed me to
mention Jabber as a messaging platform.
It could also just use the user e-mail address as the `key' to
choose the registrar (msn.com, hotmail.com -> passport.com;
aol.com -> aol.passport.com; you get the idea).
The <a
href="http://www.soapware.org/xmlStorageSystem">xmlStorage</a>
idea from <a href="http://www.scripting.com">Dave Winer</a>
could be used to store the information.
A toolkit for various popular web servers could be provided,
authenticated and should be open sourced (for those of you who
think that a binary program would give more security and would
prevent people from tampering: you are wrong. You can always
use a proxy system that "behaves" like the binary, and passes
information back and forth from the real program, and snoops
in-transit information).
Good cryptographers need to be involved in this problem to
figure out the details and the possible insecure pieces of a
proposal like this.
** Implementation: In short
To keep it short: <b>DNS, JXTA, xmlStorage.</b>
** Deploying it
The implementation of such a system should be a pretty
straightforward tasks once security cryptographers have
designed such a beast.
The major problems are:
<ul>
* <b>People might just not care:</b> In a poll to US
citizens a couple of decades ago, it was found that
most people did not care about the rights they were
given by the Bill of Rights, which lead to a number of
laws to be passed in the US that eliminated most of
the rights people had.
* <b>The industry will move way too slow:</b>
Microsoft's implementation is out in the open now: it
is being deployed, and soon it will be insinuated to
many, many users. The industry needs to get together
soon if they care about this issue.
By the time the industry reacts, it might be too
late.
</ul>
** Passport and Mono
The .NET class libraries includes a Passport class that
applications might use to authenticate with Passport. Since
we do not have information at this point on the exact protocol
of Passport, it is not even feasible to implement it.
If at some point the information is disclosed, it could be
implemented.
If a competing system to Passport existed, we could probably
hide all the authentication information to use a number of
different passport-like systems.
If a user does not want to use Passport at all, he could
always turn it off (or completely remove the class from the
library). After all, this is free software.
Currently, we are too far from the point where this is a real
issue.
** Disclaimer
This is just a group of personal thoughts of mine that I have
placed here because I get asked this question a lot lately.
The views of this page are not a statement from my employer
(Ximian, Inc).
This is not part of Mono. We are not trying to deal with this
problem.
Nat Friedman (Ximian's co-founder) has his own ideas on how a
competing system to Passport could be designed, but I will let
<a href="http://www.nat.org/gym">him</a> post his own story.
** Other Alternatives
Some people have pointed out <a
href="http://www.xns.org">XNS</a>
Send comments to me: Miguel de Icaza (<a
href="mailto:[email protected]">[email protected]</a>)
+14
View File
@@ -0,0 +1,14 @@
** Microsoft and GNU and Linux.
Q: Does this mean that Microsoft is better than Linux?
A: Many of us are working on <a
href="http://www.gnu.org/philosophy/free-sw.html">free software<a>
and want to have an <a href="http://www.opensource.org">open
source</a> environment that we can change, modify, improve, learn
from, and share with others. Some of us also think that this will
lead on the long run to better software: more efficient, faster,
more robust and more.
We are willing to take good ideas from any source they come from.
-246
View File
@@ -1,246 +0,0 @@
System.Object
System.Exception
System.ValueType
System.Delegate
System.MulticastDelegate
System.Enum
System.Activator
System.ArgIterator
System.__ComObject
System.TypedReference
System.Security.AllowPartiallyTrustedCallersAttribute
System.Runtime.Serialization.Formatter
System.Runtime.Serialization.FormatterConverter
System.Runtime.Serialization.FormatterServices
System.Runtime.Serialization.ObjectIDGenerator
System.Runtime.Serialization.ObjectManager
System.Reflection.ModuleResolveEventHandler
System.Reflection.Pointer
System.Globalization.CompareInfo
System.Globalization.HebrewCalendar
System.Globalization.HijriCalendar
System.Globalization.JapaneseCalendar
System.Globalization.KoreanCalendar
System.Globalization.SortKey
System.Globalization.StringInfo
System.Globalization.TaiwanCalendar
System.Globalization.TextElementEnumerator
System.Globalization.TextInfo
System.Globalization.ThaiBuddhistCalendar
System.IO.IsolatedStorage.IsolatedStorageFile
System.Reflection.Emit.MethodRental
System.Runtime.CompilerServices.AccessedThroughPropertyAttribute
System.Runtime.CompilerServices.CallConvCdecl
System.Runtime.CompilerServices.CallConvStdcall
System.Runtime.CompilerServices.CallConvThiscall
System.Runtime.CompilerServices.CallConvFastcall
System.Runtime.CompilerServices.CustomConstantAttribute
System.Runtime.CompilerServices.DateTimeConstantAttribute
System.Runtime.CompilerServices.DiscardableAttribute
System.Runtime.CompilerServices.DecimalConstantAttribute
System.Runtime.CompilerServices.CompilationRelaxationsAttribute
System.Runtime.CompilerServices.CompilerGlobalScopeAttribute
System.Runtime.CompilerServices.IDispatchConstantAttribute
System.Runtime.CompilerServices.IsVolatile
System.Runtime.CompilerServices.IUnknownConstantAttribute
System.Runtime.CompilerServices.RequiredAttributeAttribute
System.Runtime.InteropServices.ArrayWithOffset
System.Runtime.InteropServices.DispIdAttribute
System.Runtime.InteropServices.ClassInterfaceType
System.Runtime.InteropServices.ClassInterfaceAttribute
System.Runtime.InteropServices.ComVisibleAttribute
System.Runtime.InteropServices.LCIDConversionAttribute
System.Runtime.InteropServices.ComRegisterFunctionAttribute
System.Runtime.InteropServices.ComUnregisterFunctionAttribute
System.Runtime.InteropServices.ProgIdAttribute
System.Runtime.InteropServices.ImportedFromTypeLibAttribute
System.Runtime.InteropServices.IDispatchImplType
System.Runtime.InteropServices.IDispatchImplAttribute
System.Runtime.InteropServices.ComSourceInterfacesAttribute
System.Runtime.InteropServices.ComConversionLossAttribute
System.Runtime.InteropServices.TypeLibTypeFlags
System.Runtime.InteropServices.TypeLibFuncFlags
System.Runtime.InteropServices.TypeLibVarFlags
System.Runtime.InteropServices.TypeLibTypeAttribute
System.Runtime.InteropServices.TypeLibFuncAttribute
System.Runtime.InteropServices.TypeLibVarAttribute
System.Runtime.InteropServices.ComImportAttribute
System.Runtime.InteropServices.PreserveSigAttribute
System.Runtime.InteropServices.ComAliasNameAttribute
System.Runtime.InteropServices.AutomationProxyAttribute
System.Runtime.InteropServices.PrimaryInteropAssemblyAttribute
System.Runtime.InteropServices.CoClassAttribute
System.Runtime.InteropServices.ComEventInterfaceAttribute
System.Runtime.InteropServices.COMException
System.Runtime.InteropServices.ComMemberType
System.Runtime.InteropServices.CurrencyWrapper
System.Runtime.InteropServices.DispatchWrapper
System.Runtime.InteropServices.ErrorWrapper
System.Runtime.InteropServices.ExtensibleClassFactory
System.Runtime.InteropServices.HandleRef
System.Runtime.InteropServices.InvalidComObjectException
System.Runtime.InteropServices.InvalidOleVariantTypeException
System.Runtime.InteropServices.TypeLibImporterFlags
System.Runtime.InteropServices.ExporterEventKind
System.Runtime.InteropServices.ITypeLibExporterNotifySink
System.Runtime.InteropServices.ITypeLibConverter
System.Runtime.InteropServices.MarshalDirectiveException
System.Runtime.InteropServices.ObjectCreationDelegate
System.Runtime.InteropServices.RuntimeEnvironment
System.Runtime.InteropServices.RegistrationServices
System.Runtime.InteropServices.SafeArrayRankMismatchException
System.Runtime.InteropServices.SafeArrayTypeMismatchException
System.Runtime.InteropServices.SEHException
System.Runtime.InteropServices.TypeLibConverter
System.Runtime.InteropServices.BIND_OPTS
System.Runtime.InteropServices.UCOMIBindCtx
System.Runtime.InteropServices.UCOMIConnectionPointContainer
System.Runtime.InteropServices.UCOMIConnectionPoint
System.Runtime.InteropServices.UCOMIEnumMoniker
System.Runtime.InteropServices.CONNECTDATA
System.Runtime.InteropServices.UCOMIEnumConnections
System.Runtime.InteropServices.UCOMIEnumConnectionPoints
System.Runtime.InteropServices.UCOMIEnumString
System.Runtime.InteropServices.UCOMIEnumVARIANT
System.Runtime.InteropServices.FILETIME
System.Runtime.InteropServices.UCOMIMoniker
System.Runtime.InteropServices.UCOMIPersistFile
System.Runtime.InteropServices.UCOMIRunningObjectTable
System.Runtime.InteropServices.STATSTG
System.Runtime.InteropServices.UCOMIStream
System.Runtime.InteropServices.DESCKIND
System.Runtime.InteropServices.BINDPTR
System.Runtime.InteropServices.UCOMITypeComp
System.Runtime.InteropServices.TYPEKIND
System.Runtime.InteropServices.TYPEFLAGS
System.Runtime.InteropServices.IMPLTYPEFLAGS
System.Runtime.InteropServices.TYPEATTR
System.Runtime.InteropServices.FUNCDESC
System.Runtime.InteropServices.IDLFLAG
System.Runtime.InteropServices.IDLDESC
System.Runtime.InteropServices.PARAMFLAG
System.Runtime.InteropServices.PARAMDESC
System.Runtime.InteropServices.TYPEDESC
System.Runtime.InteropServices.ELEMDESC
System.Runtime.InteropServices.VARDESC
System.Runtime.InteropServices.DISPPARAMS
System.Runtime.InteropServices.EXCEPINFO
System.Runtime.InteropServices.FUNCKIND
System.Runtime.InteropServices.INVOKEKIND
System.Runtime.InteropServices.CALLCONV
System.Runtime.InteropServices.FUNCFLAGS
System.Runtime.InteropServices.VARFLAGS
System.Runtime.InteropServices.UCOMITypeInfo
System.Runtime.InteropServices.SYSKIND
System.Runtime.InteropServices.LIBFLAGS
System.Runtime.InteropServices.TYPELIBATTR
System.Runtime.InteropServices.UCOMITypeLib
System.Runtime.InteropServices.UnknownWrapper
System.Runtime.Remoting.IObjectHandle
System.Runtime.Remoting.IRemotingTypeInfo
System.Runtime.Remoting.IChannelInfo
System.Runtime.Remoting.IEnvoyInfo
System.Runtime.Remoting.RemotingConfiguration
System.Runtime.Remoting.TypeEntry
System.Runtime.Remoting.ActivatedClientTypeEntry
System.Runtime.Remoting.ActivatedServiceTypeEntry
System.Runtime.Remoting.WellKnownClientTypeEntry
System.Runtime.Remoting.WellKnownServiceTypeEntry
System.Runtime.Remoting.RemotingException
System.Runtime.Remoting.ServerException
System.Runtime.Remoting.RemotingTimeoutException
System.Runtime.Remoting.RemotingServices
System.Runtime.Remoting.InternalRemotingServices
System.Runtime.Remoting.SoapServices
System.Runtime.Remoting.Activation.UrlAttribute
System.Runtime.Remoting.Messaging.IMessageSink
System.Runtime.Remoting.Messaging.AsyncResult
System.Runtime.Remoting.Messaging.CallContext
System.Runtime.Remoting.Messaging.ILogicalThreadAffinative
System.Runtime.Remoting.Messaging.InternalMessageWrapper
System.Runtime.Remoting.Messaging.IMethodCallMessage
System.Runtime.Remoting.Messaging.MethodCallMessageWrapper
System.Runtime.Remoting.Messaging.HeaderHandler
System.Runtime.Remoting.Messaging.IMessageCtrl
System.Runtime.Remoting.Messaging.IRemotingFormatter
System.Runtime.Remoting.Messaging.ReturnMessage
System.Runtime.Remoting.Messaging.MethodCall
System.Runtime.Remoting.Messaging.ConstructionCall
System.Runtime.Remoting.Messaging.MethodResponse
System.Runtime.Remoting.Messaging.ConstructionResponse
System.Runtime.Remoting.Messaging.MethodReturnMessageWrapper
System.Runtime.Remoting.Messaging.OneWayAttribute
System.Runtime.Remoting.Messaging.MessageSurrogateFilter
System.Runtime.Remoting.Messaging.RemotingSurrogateSelector
System.Runtime.Remoting.Contexts.CrossContextDelegate
System.Runtime.Remoting.Contexts.ContextProperty
System.Runtime.Remoting.Contexts.IContextPropertyActivator
System.Runtime.Remoting.Contexts.IContributeClientContextSink
System.Runtime.Remoting.Contexts.IContributeDynamicSink
System.Runtime.Remoting.Contexts.IContributeEnvoySink
System.Runtime.Remoting.Contexts.IContributeObjectSink
System.Runtime.Remoting.Contexts.IContributeServerContextSink
System.Runtime.Serialization.Formatters.InternalRM
System.Runtime.Serialization.Formatters.InternalST
System.Runtime.Serialization.Formatters.SoapMessage
System.Runtime.Serialization.Formatters.SoapFault
System.Runtime.Serialization.Formatters.ServerFault
System.Runtime.Serialization.Formatters.Binary.BinaryFormatter
System.Security.Cryptography.CryptoConfig
System.Security.Cryptography.KeyedHashAlgorithm
System.Security.Cryptography.HMACSHA1
System.Security.Cryptography.MACTripleDES
System.Security.Cryptography.MaskGenerationMethod
System.Security.Cryptography.PasswordDeriveBytes
System.Security.Cryptography.PKCS1MaskGenerationMethod
System.Security.Cryptography.RC2
System.Security.Cryptography.RC2CryptoServiceProvider
System.Security.Cryptography.RSACryptoServiceProvider
System.Security.Cryptography.RSAOAEPKeyExchangeDeformatter
System.Security.Cryptography.RSAOAEPKeyExchangeFormatter
System.Security.Cryptography.RSAPKCS1KeyExchangeDeformatter
System.Security.Cryptography.RSAPKCS1KeyExchangeFormatter
System.Security.Cryptography.RSAPKCS1SignatureDeformatter
System.Security.Cryptography.RSAPKCS1SignatureFormatter
System.Security.Cryptography.SHA1Managed
System.Security.Cryptography.TripleDES
System.Security.Cryptography.TripleDESCryptoServiceProvider
System.Security.Permissions.EnvironmentPermission
System.Security.Permissions.FileDialogPermission
System.Security.Permissions.IsolatedStorageFilePermission
System.Security.Permissions.PrincipalPermissionAttribute
System.Security.Permissions.SecurityPermissionAttribute
System.Security.Permissions.PublisherIdentityPermissionAttribute
System.Security.Permissions.PublisherIdentityPermission
System.Security.Permissions.ReflectionPermission
System.Security.Permissions.RegistryPermission
System.Security.Permissions.PrincipalPermission
System.Security.Permissions.SiteIdentityPermission
System.Security.Permissions.StrongNameIdentityPermission
System.Security.Permissions.StrongNamePublicKeyBlob
System.Security.Permissions.UIPermission
System.Security.Permissions.UrlIdentityPermission
System.Security.Permissions.ZoneIdentityPermission
System.Security.Policy.ApplicationDirectory
System.Security.Policy.ApplicationDirectoryMembershipCondition
System.Security.Policy.FirstMatchCodeGroup
System.Security.Policy.Hash
System.Security.Policy.HashMembershipCondition
System.Security.Policy.NetCodeGroup
System.Security.Policy.PermissionRequestEvidence
System.Security.Policy.Publisher
System.Security.Policy.PublisherMembershipCondition
System.Security.Policy.Site
System.Security.Policy.SiteMembershipCondition
System.Security.Policy.StrongName
System.Security.Policy.StrongNameMembershipCondition
System.Security.Policy.UnionCodeGroup
System.Security.Policy.Url
System.Security.Policy.UrlMembershipCondition
System.Security.Policy.Zone
System.Security.Policy.ZoneMembershipCondition
System.Security.Principal.WindowsIdentity
System.Security.Principal.WindowsImpersonationContext
System.Security.Principal.WindowsPrincipal
System.Threading.CompressedStack
+173
View File
@@ -0,0 +1,173 @@
* The Mono Project
** Background.
The GNOME project goal was to bring missing technologies to
Unix and make it competitive in the current market place for
desktop applications. We also realized early on that language
independence was important, and that is why GNOME APIs were
coded using a standard that allowed the APIs to be easily
wrapped for other languages. Our APIs are available to most
programming languages on Unix (Perl, Python, Scheme, C++,
Objective-C, Ada).
Later on we decided to use better methods for encapsulating
our APIs, and we started to use CORBA to define interfaces to
components. We complemented it with policy and a set of
standard GNOME interfaces for easily creating reusable,
language independent components, controls and compound
documents. This technology is known as <a
href="http://www.ximian.com/tech/bonobo.php3">Bonobo</a>.
Interfaces to Bonobo exist for C, Perl, Python, and
Java.
CORBA is good when you define coarse interfaces, and most
Bonobo interfaces are coarse. The only problem is that
Bonobo/CORBA interfaces are not good for small interfaces.
For example, an XML parsing Bonobo/CORBA component would be
inefficient compared to a C API.
** Another explanation
I recently explained our motivations to Dave Winer, and he posted
it <a
href="http://scriptingnews.userland.com/stories/storyReader$1275">here</a>
** Microsoft's .NET
The Microsoft .NET initiative is confusing because it is a
company wide effort that ranges from development tools to end
user applications. .NET is a branding formative that
has been applied to:
<ul>
* The .NET development platform, a new platform for
writing software.
* Web services.
* Microsoft Server Applications.
* New tools that use the new development platform.
* Hailstorm, the Passport centralized single-signon
system that is being integrated into Windows XP.
</ul>
Mono is an implementation of the .NET development platform.
** The Common Language Infrastructure platform.
Microsoft has created a new development platform. The
highlights of this new development platform are:
<ul>
* A runtime environment that provides garbage
collection, threading and a virtual machine
specification (The Virtual Execution System, VES)
* A comprehensive class library.
* A new language, C#. Very similar to Java, C#
allows programmers to use all the features available
on the .NET runtime.
* A language specification that compilers can
follow if they want to generate classes and code
that can interoperate with other programming
languages (The Common Language Specification: CLS)
</ul>
The Common Language Infrastructure platform is similar to the
goals we had in GNOME of giving language independence to
programmers. It is more mature, documented, larger in scope,
and has a consistent design.
Any API that is written using a CLS provider language can be
used by any language that is a CLS consumer. Compilers
generate code in a format called Common Intermediate Language
(CIL) which is an intermediate representation of a compiled
program and is easy to compile to native code or compiled
using Just-in-Time (JIT) engines. The restrictions placed by
the runtime on the CIL byte codes ensures that it is possible
to do a good job at optimizing the code in a JIT compiler.
There is not really a lot of innovation in this platform: we
have seen all of these concepts before, and we are all
familiar with how these things work.
What makes the Common Language Infrastructure development
platform interesting is that it is a good mix of technologies
that have been nicely integrated.
The .NET development platform is essentially a new foundation
for program development that gives Microsoft a room to grow
for the coming years.
** ECMA standards.
Microsoft has submitted the
specifications of C#, the runtime, the metadata and the
other various bits of the .NET development platform to the
<a href="http://www.ecma.ch">ECMA</a> for standarization.
You can get a copy of the specifications submitted to ECMA
from: <a href="http://www.dotnetexperts.com/ecma">http://www.dotnetexperts.com/ecma</a>
** Mono: an Open Source Common Language Infrastructure implementation.
Ximian has begun work on Mono, a project that aims to bring
the Common Language Infrastructure platform to free systems.
When the GNU project was launched, they picked the best
operating system that was available out there, and they
began to clone it: Unix.
The .NET development platform is a very rich, powerful, and
well designed platform that would help improve the free
software development platform. Just like the GNU project
began to clone Unix sixteen years ago, we will be cloning the
.NET development platform because it is a great platform to
build on.
** What makes up Mono?
There are various pieces that will make up Mono:
<ul>
* A C# compiler.
* The Virtual Execution System: that will have the
Just-in-Time compiler, garbage collector, loader,
threading engine.
A byte code interpreter will be provided for quickly
porting Mono to new systems and debugging the JIT
purposes, but it is not intended to be the ideal
execution environment.
* An implemenation of the .NET class library.
* Visual development tools.
* A CIL GCC frontend.
</ul>
** Why use GNOME components?
GNOME is an umbrella project that consists of infrastructural
components (GUI toolkit, XML libraries, CORBA implementation,
printing architecture, imaging system), a desktop environment,
and productivity applications.
The GNOME infrastructural components can be used to quickly
implement various pieces of the class libraries without reinventing
the wheel, and since all those components are licensed under
the terms of the GNU LGPL it is a perfect fit.
Libart will be used to implement the Drawing.2D API; Gtk+ and
the GNOME libraries will be used to implement the WinForms
API and of course Glib and libxml will be used in various
places.
+46
View File
@@ -0,0 +1,46 @@
To: [email protected], [email protected]
Subject: July 12 snapshots.
FCC: ~/Mail/outbox.txt
X-Windows: Sometimes you fill a vacuum and it still sucks.
--text follows this line--
Hey!
July 12 snapshots of class libraries, the compiler and the mono
runtime are available.
New on this release:
* Runtime (module: mono)
The beginning of a simple interpreter that Paolo started
workign on (can run really simple .NET programs).
Disassembler copes with more elements of the binary format and
more tokens are decoded. Paolo is working now on moving some
of these to the metadata library.
More tables are dumped.
* Class libraries (module: mcs/class)
Many new more classes are in from Joe, Vladimir, Jeff, Sean
and yours truly.
Sean fixed the build process, and it is now possible to
compile with a single command the assemblies. We will be
revisiting this mechanism in the future to compile per-OS
assemblies (ie, Unix, Windows, MacOS, etc).
* Compiler (module mcs/mcs)
Not much done this week, just a few fixes here and there, and
more work to make it easy to compiler.
* Documentation (module: mono/doc)
All the changes to the web site are there for your browsing
pleasure. We still need to integrate the status system in
there.
Miguel.
+43
View File
@@ -0,0 +1,43 @@
To: [email protected]
Subject: Sunday snapshot available.
Gcc: mail.2001-07
--text follows this line--
Hey guys,
I promise I will not be doing these so often once we have the CVS
server up. In the meantime:
* MCS
Sean got the classes to compile in a single go. You
will need CygWin (www.cygwin.org) to compile though
(GNU make and stuff is required).
System.Xml.XmlReader contribution from Jason
(WOOHHOO!!). It also contains a nice test-suite for
his functions, and in his new code bit, his
implementation is faster than Microsoft's
We now ship `jay' as part of the distribution to allow
you to compile the compiler with the same `make'
command. Small fixes to the parser as well were
introduced.
* Mono 0.4
Paolo's interpreter supports call instructions and has
the test suite program that he posted about.
All documentation ships now in the mono-0.4.tar.gz
Notes:
As usual, MCS is targeted to be compiled on a Windows machine
(you will need Cygwin).
Mono is targeted to be compiled on a Unix machine or a Windows
machine running Cygwin.
Miguel.
+109
View File
@@ -0,0 +1,109 @@
* Resources
There are a number of resources available for those of you who
want to contribute to the Mono project. Here are a few links.
If you want to send suggestions for links, address them to <a
mailto="[email protected]">[email protected]</a>.
** ECMA Documentation.
You can get the documentation for the ECMA specs from a number of sites:
<ul>
* <a href="http://msdn.microsoft.com/net/ecma">At MSDN</a>
* <a href="http://www.dotnetexperts.com">Dot Net Experts</a>
* <a href="http://developer.intel.com/software/idap/ecma">Intel</a>
* <a href="http://lightning.csse.monash.edu.au/.net/CLI">Monash University</a>
</ul>
These contain specifications for the assembler, the metadata,
byte codes supported by the CLI virtual machine, the C#
language and the core class libraries.
For details on the .NET class libraries, you can
visit the Microsoft's Developer Network:
<ul>
* <a href="http://msdn.microsoft.com/library/default.asp?url=/library/en-us/cpref/html/cpref_start.asp">.NET Framework Class Library</a>
</ul>
You can also get this information if you install the Beta2
release of the .NET Framework.
<a name="mailing">
** Mailing Lists
There are a number of mailing lists for Mono
<ul>
* <b><a
href="http://mail.ximian.com/mailman/listinfo/mono-list">mono-list:</a></b>
The general Mono discussion list.
* <b><a
href="http://mail.ximian.com/mailman/listinfo/mono-announce-list">mono-announce-list:</a></b>
Announcements of Mono developments.
* <b><a
href="http://mail.ximian.com/mailman/listinfo/mono-docs-list">mono-docs-list:</a></b>
Discussion on the documentation of Mono.
* <b><a
href="http://discuss.develop.com/dotnet.html">Dotnet mailing
list at Develop.com:</a></b> The guys at Develop Mentor run
this general purpose mailing list.
</ul>
** Discussion Groups.
<ul>
* <a href="http://www.oreillynet.com">O'Reilly
Network</a> has a <a
href="http://www.oreillynet.com/dotnet">section devoted to
.NET</a>
* <a
href="http://msdn.microsoft.com/newsgroups">MSDN</a> also
lists various newsgroups related to .NET</ul>
</ul>
** Other .NET related projects
There are a number of related projects to Mono:
<ul>
* <a
href="http://www.icsharpcode.net/OpenSource/SD/default.asp">Sharp
Develop:</a> an IDE for the C# language written in C#.
* <a
href="http://www.southern-storm.com.au/portable_net.html">Portable.NET:</a>
* <a href="http://nunit.sourceforge.net">NUnit:</a> A
testing framework for .NET classes.
* <a href="http://www.kaffe.org">Kaffe:</a> A popular
Free Software JIT engine for Java.
* <a href="http://www.intel.com/research/mrl/orp">ORP:</a> A research
JIT/VM/GC system from Intel.
</ul>
** GNOME Documentation
Documnetation on GNOME, and the GNOME APIs is available from
the <a href="http://developer.gnome.org">developer</a> site at
GNOME:
<ul>
* <a href="http://developer.gnome.org/doc/API/">GNOME
API documentation</a>
* <a href="http://developer.gnome.org/doc/books">GNOME
Online books</a>
* <A
href="http://developer.gnome.org/arch/">Architecture Overview</a>
</ul>
+30
View File
@@ -0,0 +1,30 @@
** MacOS Documentation
** Assembly Language Manuals online
Intel
MIPS
SPARC
** Microsoft
msdn.microsoft.com/net
Research.microsoft.com
** Related Technologies
<ul>
* The CLI allows people to create Web Services using the SOAP
protocol. SOAP is based on XML, XML schemas an the HTTP
protocol.
<ul>
* XML specification.
* XML Namespaces.
* XML Schemas.
* SOAP Specification.
</ul>
</ul>
** Compiler Information
GCC
GCC Sample front-end tutorial
+12
View File
@@ -0,0 +1,12 @@
* Roadmap
We are working on the following three projects at Ximian:
The C# Compiler (mcs/mcs)
A .NET compatible Class Library (mcs/class)
The JIT/interpreter (mono)
+133
View File
@@ -0,0 +1,133 @@
The MonoNet runtime
The MonoNet runtime will implement the JIT engine (and a byte
code interpreter for quickly porting to new systems), the
class loader, the garbage collector, threading system and
metadata access libraries.
Currently the runtime contains the beginning of an image
loader and metadata access entry points. Since Beta2 has been
now released, it is possible to resume work using the ECMA
specs and testing with Beta2-generated executables.
The runtime core will be implemented in C, in a library
"libMonoVES.so".
** Executing MSIL/CIL images
The code will load an executable and map the references to
external assemblies to our own version of the assemblies on
GNU/Linux.
Our roadmap looks like this, this has been updated as of
<b>Jul 15, 2001</b>:
<ul>
* Milestone 1: Fully read and parse all CIL byte-codes
and metadata tokens (ie, a disassembler).
* Milestone 2: Complete an interpreter for CIL byte
codes. This interpreter can be used temporarly to
run CIL byte code on a system where no JIT is
available.
* Milestone 3: Define an <i>lburg</i> instruction
selector for the JITer for Intel. Although slower
at JITing than a streaming JITer, it generates
better code. The same grammar can later be used for
the stream jitter.
* Milestone 4: Implement JITer.
* Milestone 5: Port of the JITer to non IA32 systems.
</ul>
A setup similar to the Kaffe JIT engine can be used to
layout the code to support non-IA32 architectures. Our work
will be focused on getting a IA32 version running first.
The JIT engine should work on Linux and Win32, although you
will need to install the CygWin32 development tools to get a
Unix-like compilation environment.
** JIT Engine (<b>updated, Jul 14th, 2001</b>)
We will be using a code-generator generator approach for our
JITer. Given the properties of CIL byte codes, we can take
full advantage of a real instruction selector for our code
generator.
There are a couple of books that deal with this technique: "A
Retargetable C Compiler" and "Advanced Compiler Design and
Implementation" are good references. You can also get a
technical description of <a
href="http://research.microsoft.com/copyright/accept.asp?path=http://www.research.microsoft.com/~drh/pubs/iburg.pdf&pub=ACM">lbrug</a>
Previously we had looked at a number of JIT engines and tools,
but they would not take full advantage of the CIL properties:
<ul>
* <a
href="http://www.intel.com/research/mrl/orp/">ORP</a>
* <a
href="http://www.gnu.org/software/lightning/">GNU
Lightning</a>
* href="http://www.eecs.harvard.edu/~nr/toolkit/">NJ Machine
Toolkit</a>.).
* VCODE.
</ul>
** Garbage Collection
We have decided to implement a generational tracing garbage
collector, which is very similar to the one being used by
.NET. For an introduction to the garbage collection system
used by Microsoft's CLR implementation, you can read this book
on <a
href="http://www.amazon.com/exec/obidos/ASIN/0471941484/o/qid=992556433/sr=2-1/ref=aps_sr_b_1_1/103-5866388-0492603">Garbage
Collection.</a>
Although using a conservative garbage collector like Bohem's
would work, all the type information is available at runtime,
so we can actually implement a better collector than a
conservative collector.
<ul>
* Garbage collection list and FAQ:<br>
<a href="http://www.iecc.com/gclist/">http://www.iecc.com/gclist/</a>
* The Microsoft .NET Garbage Collection Implementation:<br>
<a href="http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnmag00/html/GCI.asp">http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnmag00/html/GCI.asp</a>
<a href="http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnmag00/html/GCI.asp">http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnmag00/html/GCI.asp</a>
</ul>
** Useful links
Paolo Molaro found a few interesting links:
<ul>
* On compilation of stack-based languages:<br>
<a href="http://www.complang.tuwien.ac.at/projects/rafts.html">
http://www.complang.tuwien.ac.at/projects/rafts.html</a>
* A paper on fast JIT compilation of a stack-based language:<br>
<a href="http://www.research.microsoft.com/~cwfraser/pldi99codegen.pdf">
http://www.research.microsoft.com/~cwfraser/pldi99codegen.pdf</a>
* Vmgen generates much of the code for efficient virtual machine (VM)
interpreters from simple descriptions of the VM instructions:<br>
<a href="http://www.complang.tuwien.ac.at/anton/vmgen/">
http://www.complang.tuwien.ac.at/anton/vmgen</a>
</ul>
** PInvoke
PInvoke will be supported, and will be used to wrap Unix API
calls, these in turn are required for reusing some of the
GNOME libraries that will reduce the work we have to do to
deliver a complete class library.
+44
View File
@@ -0,0 +1,44 @@
* Project Status
Pieces of Mono that have been implemented:
<ul>
* C# compiler: The C# parser can now generate parse
trees. Next up: semantic analysis, compiler lever optimizations
and code generation.
* Metadata library: Can currently parse and load
information from .NET modules (executables and DLL
files). It has a few bugs outstanding (table
dimension computation is incorrect).
Parsing of Exception tables is missing.
It lacks Token to (Table, Index) mapping, but this
should be trivial to write.
* Disassembler: Can disassemble .NET modules. Still
lacking exception handling as well as useful debugging
tools (hex dumping, token dumping).
* Class Libraries: Only a few classes have been
implemented.
</ul>
Tasks on the critical path:
<ul>
* Method and Field resolution.
* Bytecode interpreter: Not implemented yet.
* Simple JIT: Not implemented yet.
* Garbage collection engine.
</ul>
+2
View File
@@ -0,0 +1,2 @@
* The MonoNet Team
+21
View File
@@ -0,0 +1,21 @@
* Testing
All classes in Mono libraries should have comprehensive unit test
suites to go with them. Unit testing is a software engineering
methodology that makes it easier to build correct code. Every
method in every class should have a set of tests to verify
that they work correctly. Mono also needs a testing framework
to make it easy to write and run lots of tests.
Try <a href="http://nunit.sourceforge.net">NUnit</a>
Why do unit testing? It becomes simple to run automated tests
for the whole library. Unit tests are a safety net - you can
change part of the code and verify that you haven't broken
anything. Ideally, tests are written before the actual library
code itself. And every time a bug is discovered, a test should
be written to demonstrate the bug and its fix. Then, if
you ever reintroduce the bug, you will know immediately. For
more info, read <a
href="http://nunit.sourceforge.net/doc/testinfected/testing.html">
JUnit Test Infected: Programmers Love Writing Tests</a>.
+6
View File
@@ -0,0 +1,6 @@
* Thanks
We would like to thank Tim O'Reilly, Brian Jepson and Nathan
Torkington for their help.
Dave Winer for provided interesting comments and a to read.
+1
View File
@@ -0,0 +1 @@
Discuss with write new JIT
+25
View File
@@ -0,0 +1,25 @@
* Tools
We need a number of tools to make people productive using a
Mono-based solution. Some of these tools can be developed on
Windows before Mono is fully finished.
** Debugger
We will need a debugging API to debug CLI applications and
then a debugger component that can be used in an IDE
environment.
** Integrated Development Environment
There is already a project to create a C# development
environment (SharpDevelop). People could work with the
SharpDevelop hackers to produce a unified development environment.
** Help Browser
We need a good help browser that can be used to browse all the
Mono documentation.
+2
View File
@@ -0,0 +1,2 @@
*.src
*.html
+17
View File
@@ -0,0 +1,17 @@
0,Home,index.html,index.src
1,FAQ,faq.html,faq.src
0,Mono,rationale.html,rationale.src
1,Runtime,runtime.html,runtime.src
1,Classes,class-library.html,class-library.src
2,Class Status,class-status.html,class-status.src
1,C# Compiler,c-sharp.html,c-sharp.src
1,Status,status.html,status.src
0,Contributing,contributing.html,contributing.src
1,Documentation,documentation.html,documentation.src
1,Test Suite,testing.html,testing.src
1,Tools,tools.html,tools.src
0,Download,download.html,download.src
0,Resources,resources.html,resources.src
1,Ideas,ideas.html,ideas.src
1,Passport,passport.html,passport.src
0,Contact,contact.html,contact.src
+24
View File
@@ -0,0 +1,24 @@
#!/usr/bin/perl
$q = 1;
while (<>){
chop;
if (/^\* (.*)$/){
print "<h1>$1</h1>\n";
} elsif (/^\*\* (.*)$/) {
print "<h2>$1</h2>\n";
} elsif (/^\*\*\* (.*)$/) {
print "<h3>$1</h3>\n";
} elsif (/^$/) {
print "<p>\n";
} elsif (/^\t\t\* (.*)$/) {
print "<li>$1\n";
} elsif (/^Q: (.*)$/){
print "<p><a name=\"q$q\"><b>Question $q:</b> $1\n";
$q++;
} elsif (/^A: (.*)$/){
print "$1\n";
} else {
print "$_\n";
}
}
File renamed without changes.

Before

Width:  |  Height:  |  Size: 4.8 KiB

After

Width:  |  Height:  |  Size: 4.8 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 4.5 KiB

After

Width:  |  Height:  |  Size: 4.5 KiB

File renamed without changes.
Binary file not shown.

After

Width:  |  Height:  |  Size: 3.3 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 4.4 KiB

After

Width:  |  Height:  |  Size: 4.4 KiB

File renamed without changes.

Before

Width:  |  Height:  |  Size: 49 B

After

Width:  |  Height:  |  Size: 49 B

Binary file not shown.

Before

Width:  |  Height:  |  Size: 156 B

After

Width:  |  Height:  |  Size: 157 B

+39
View File
@@ -0,0 +1,39 @@
SOURCES= \
../contributing \
../class-library \
../class-status \
../contact \
../c-sharp \
../documentation \
../download \
../faq \
../gcc-frontend \
../index \
../ideas \
../passport \
../rationale \
../resources \
../roadmap \
../runtime \
../status \
../testing \
../tools
all:
-mkdir site
for i in $(SOURCES); do \
perl htmlify $$i > `basename $$i`.src; \
done;
perl process.pl commands template.html.in .
clean:
for i in $(SOURCES); do \
rm -f `basename $$i`.src `basename $$i`.html; \
done;
push:
scp *.html [email protected]:/web/cvsmodules/mono
push2:
scp *.html primates:public_html/xxx
+73
View File
@@ -0,0 +1,73 @@
#!/usr/bin/perl
#
# Author:
# Sean MacIsaac
#
use strict;
my $full_expand = 1;
my @template;
my $n;
if ($#ARGV != 2) {
print "process.pl command_file template_file directory_prefix\n";
exit ();
}
my $menu = "";
open COMMANDS, $ARGV[0] || die "Can not open $ARGV[0]";
while (<COMMANDS>) {
chop;
my @command = split /,/;
if ($command[0] != -1) {
$menu .= "\t\t";
if ($command[0] == 0){
$menu .= "<tr><td valign=\"top\" class=\"navi\"><a class=\"navi\"";
} else {
$menu .= "<tr><td valign=\"top\" class=\"subnavi\">&nbsp;&nbsp;&nbsp<a class=\"subnavi\"";
}
$menu .= "HREF=\"$command[2]\">$command[1]</A></td></tr>\n\n";
}
}
close COMMANDS;
open TEMPLATE, $ARGV[1] || die "Can not open $ARGV[1]";
while (<TEMPLATE>) {
push @template, $_;
}
close TEMPLATE;
open COMMANDS, $ARGV[0] || die "Can not open $ARGV[0]";
while (<COMMANDS>) {
chop;
my @command = split /,/;
$n = $ARGV[2] . "/" . $command[2];
open OUTPUT, ">" . $n || die "Can not create $n";
my $content = "";
open INPUT, $command[3] || die "Can not open $command[3]";
while (<INPUT>) {
$content .= $_;
}
close INPUT;
my $line;
my $temp;
my $tit;
my $title;
$tit = $command[1];
foreach $line (@template) {
$temp = $line;
$title = "$tit / Mono";
$temp =~ s/#TITLE#/$title/;
$temp =~ s/#CONTENT#/$content/;
$temp =~ s/#MENU#/$menu/;
print OUTPUT $temp;
}
close OUTPUT;
}
+78
View File
@@ -0,0 +1,78 @@
<html>
<head>
<title>#TITLE#</title>
<style type="text/css">
<!--
body { font-family: "trebuchet ms", lucida, verdana, helvetica;
background-image: url("images/bgsquares.gif");
background-attachment: fixed; }
body, td, table { font-family: "trebuchet ms", lucida, verdana, helvetica;
font-size: 12px; }
.navi { font-size: 14px; background: #444444; }
.subnavi { font-size: 12px; }
.footnote { font-size: 10px; color: #aaaaaa; }
a.navi { color: #ffffff; text-decoration: none; font-weight: bold; }
a.navi:visited { color: #cccccc; }
a.navi:hover { color: #ee9900; text-decoration: underline; }
a.subnavi { color: #ffffff; text-decoration: none; font-weight: bold; }
a.subnavi:visited { color: #cccccc; }
a.subnavi:hover { color: #ee9900; text-decoration: underline; }
// -->
</style>
</head>
<body bgcolor="#555555" text="#000000">
<table cellpadding="0" cellspacing="0" border="0" width="100%">
<tr>
<td><img src="images/pixel.gif"></td><!-- left border -->
<td colspan="4">
<a href="/"><img src="images/mono.gif" border="0"></a></td>
<td><img src="images/pixel.gif"></td><!-- right border -->
</tr>
<tr>
<td><img src="images/pixel.gif" width="1" height="1"></td>
<td colspan="3" bgcolor="black"><img src="images/pixel.gif" height="2"></td>
<td bgcolor="black"><img src="images/pixel.gif" width="1"></td>
<td><img src="images/pixel.gif"></td>
</tr>
<tr>
<td width="100"><img src="images/pixel.gif"></td>
<td valign="top">
<table cellpadding="2" valign="top" cellspacing="0" border="0">
#MENU#
</table>
</td>
<td bgcolor="black" width="1"><img src="images/pixel.gif" width="1"></td>
<td bgcolor="white">
<table cellpadding="16">
<tr><td>
#CONTENT#
</td></tr>
</table>
</td>
<td bgcolor="black"><img src="images/pixel.gif" width="1"></td>
<td width="100"><img src="images/pixel.gif"></td>
</tr>
<tr>
<td colspan="2">
<img src="images/pixel.gif"></td>
<td colspan="2" bgcolor="black"><img src="images/pixel.gif" height="1"></td>
<td bgcolor="black"><img src="images/pixel.gif" width="1"></td>
<td><img src="images/pixel.gif"></td>
</tr>
<td colspan="2"></td>
<td colspan="2" align="center">
<a class="footnote" href="mailto: [email protected]">[email protected]</a>
</td>
<td colspan="2"></td>
</tr>
</tr>
</table>
</body>
</html>
-2
View File
@@ -1,2 +0,0 @@
Makefile
Makefile.in
-15
View File
@@ -1,15 +0,0 @@
EXTRA_DIST = \
aot-compiler.txt \
assembly-bundle \
embedded-api \
exceptions \
gc-issues \
jit-thoughts \
jit-trampolines \
local-regalloc.txt \
mini-doc.txt \
mini-porting.txt \
opcode-decomp.txt \
object-layout \
unmanaged-calls \
stack-alignment
-525
View File
@@ -1,525 +0,0 @@
Here "abc" stays for "array bounds check", or "array bound checks", or
some combination of the two...
------------------------------------------------------------------------------
USAGE
Simply use the "abcrem" optimization invoking mono.
To see if bound checks are actually removed, use "mono -v" and grep for
"ARRAY-ACCESS" in the output, there should be a message for each check
that has been removed.
To trace the algorithm execution, use "-v -v -v", and be prepared to be
totally submersed by debugging messages...
------------------------------------------------------------------------------
EFFECTIVENESS
The abc removal code can now always remove bound checks from "clean"
array scans in loops, and generally anyway there are clear conditions that
already state that the index is "safe".
To be clearer, and give an idea of what the algorithm can and cannot do
without describing it in detail... keep in mind that only "redunant" checks
cannot be removed. By "redundant", I mean "already explicitly checked" in
the method code.
Unfortunately, analyzing complex expressions is not so easy (see below for
details), so the capabilities of this "abc remover" are limited.
These are simple guidelines:
- Only simple comparisons between variables are understood.
- Only increment-decrement operations are handled.
- "switch" statements are not yet handled.
This means that code like this will be handled well:
for (int i = 0; i < a.Length; i++) {
a [i] = .....
}
The "i" variable could be declared out of the "for", the "for" could be a
"while", and maybe even implemented with "goto", the array could be scanned
in reverse order, and everything would still work.
I could have a temporary variable storing the array length, and check on
it inside the loop, and the abc removal would still occurr, like this:
int i = 0;
int l = a.Length;
while ( i < l ) {
a [i] ......
}
or this:
int l = a.Length;
for (int i = l; i > 0; i--) {
a [i] = .....
}
But just something like this
for (int i = 0; i < (a.Length -1); i++) .....
or like this
for (int i = 0; i < a.Length; i += 2) .....
would not work, the check would stay there, sorry :-(
Just to make you understand how things are tricky... this would work!
int limit = a.Length - 1;
for (int i = 0; i < limit; i++) {
a [i] = .....
}
A detailed explanation of the reason why things are done like this is given
below...
All in all, the compiling phase is generally non so fast, and the abc removed
are so few, that there is hardly a performance gain in using the "abcrem"
flag in the compiling options for short programs like the ones I tried.
Anyway, various bound checks *are* removed, and this is good :-)
------------------------------------------------------------------------------
THE ALGORITHM
Unfortunately, this time google has not been my friend, and I did not
find an algorithm I could reasonably implement in the time frame I had,
so I had to invent one. The two cleasses of algorithms I found were either
simply an analisys of variable ranges (which is not very different from
what I'm doing), or they were based on type systems richer than the .NET
one (particularly, they required integer types to have an explicit range,
which could be related to array ranges). By the way, it seems that the
gcj people do not have an array bound check removal optimization (they
just have a compiler flag that unconditionally removes all the checks,
but this is not what we want).
This array bound check removal (abc removal) algorithm is based on
symbolic execution concepts. I handle the removal like an "unreachable
code elimination" (and in fact the optimization could be extended to remove
also other unreachable sections of code, due to branches that "always go the
same way").
In symbolic execution, variables do not have actual (numeric) values, but
instead symbolic expressions (like "a", or "x+y").
Also, branch conditions are handled like symbolic conditions (like "i<k"),
which state relations between variable values.
The SSA representation inside mini is somewhat close to a symbolic
representation of the execution of the compiled method.
Particularly, the representation of variable values is exactly a symbolic
one. It is enough to find all CEE_STIND_* instructions which store to a
local variable, and their second argument is exactly the variable value.
Actually, "cfg->vars [<variable-index>]->def" should contain exactly
those store instructions, and the "get_variable_value_from_ssa_store"
function extracts the variable value from there.
On the other hand, the conditions under which each basic block is
executed are not made fully explicit.
However, it is not difficult to make them so.
Each BB that has more than one exit BB, in practice must end either with
a conditional branch instruction or with a switch instruction.
In the first case, the BB has exactly two exit BBs, and their execution
conditions are easy to get from the condition of the branch (see the
"get_relation_from_branch_instruction" function, and expecially the end of
"analyze_block" in abcremoval.c.
If there is a switch, the jump condition of every exit BB is the equality
of the switch argument with the particular index associated with its case
(but the current implementation does not handle switch statements yet).
These individual conditions are in practice associated to each arc that
connects BBs in the CFG (with the simple case that unconditional branches
have a "TRUE" condition, because they always happen).
So, for each BB, its *proper* entry condition is the union of all the
conditions associated to arcs that enter the BB. The "union" is like a
logical "or", in the sense that either of the condition could be true,
they are not necessarily all true. This means that if I can enter a BB
in two ways, and in one case I know that "x>0", and in the other that
"x==0", actually in the BB I know that "x>=0", which is a weaker
condition (the union of the two).
Also, the *complete* entry condition for a BB is the "intersection" of all
the entry conditions of its dominators. This is true because each dominator
is the only way to reach the BB, so the entry condition of each dominator
must be true if the control flow reached the BB. This translates to the
logical "and" of all the "proper" conditions of the BBs met walking up in the
dominator tree. So, if one says "x>0", and another "x==1", then I know
that "x==1", which is a stronger condition (the intersection of the two).
Note that, if the two conditions were "x>0" and "x==0", then the block would
be unreachable (the intersection is empty), because some branch is impossible.
Another observation is that, inside each BB, every variable is subject to the
complete entry condition of that very same BB, and not the one in which it is
defined (with the "complete entry condition" being the thing I defined before,
sorry if these terms "proper" and "complete" are strange, I found nothing
better).
This happens because the branch conditions are related to the control flow.
I can define "i=a", and if I am in a BB where "a>0", then "i>0", but not
otherwise.
So, intuitively, if the available conditions say "i>=0", and i is used as an
index in an array access, then the lower bound check can be omitted.
If the condition also says "(i>=0)&&(i<array.length)", the abc removal can
occur.
So, a complete solution to the problem of abc removal would be the following:
for each array access, build a system of equations containing:
[1] all the symbolic variable definitions
[2] the complete entry condition of the BB in which the array access occurs
[3] the two "goal functions" ("index >=0" and "index < array.length")
If the system is valid for *each possible* variable value, then the goal
functions are always true, and the abc can be removed.
All this discussion is useful to give a precise specification to the problem
we are trying to solve.
The trouble is that, in the general case, the resulting system of equations
is like a predicate in first order logic, which is semi-decidable, and its
general solution is anyway too complex to be attempted in a JIT compiler
(which should not contain a full fledged theorem prover).
Therefore, we must cut some corner.
By the way, there is also another big problem, which is caused by "recursive"
symbolic definitions. These definition can (and generally do) happen every
time there is a loop. For instance, in the following piece of code
for ( int i = 0; i < array.length; i++ ) {
Console.WriteLine( "array [i] = " + array [i] );
}
one of the definitions of i is a PHI that can be either 0 or "i + 1".
Now, we know that mathematically "i = i + 1" does not make sense, and in
fact symbolic values are not "equations", they are "symbolic definitions".
The symbolic value of i is a generic "n", where "n" is the number of
iterations of the loop, but this is terrible to handle (and in more complex
examples the symbolic definition of i simply cannot be written, because i is
calculated in an iterative way).
However, the definition "i = i + 1" tells us something about i: it tells us
that i "grows". So (from the PHI definition) we know that i is either 0, or
"grows". This is enough to tell that "i>=0", which is what we want!
It is important to note that recursive definitions can only occurr inside
PHI definitions, because actually a variable cannot be defined *only* in terms
of itself!
At this point, I can explain which corners I want to cut to make the
problem solvable. It will not remove all the abc that could theoretically
be removed, but at least it will work.
The easiest way to cut corners is to only handle expressions which are
"reasonably simple", and ignore the rest.
Keep in mind that ignoring an expression is not harmful in itself.
The algorithm will be simply "less powerful", because it will ignore
conditions that could have caused to the removal of an abc, but will
not remove checks "by mistake" (so the resulting code will be in any case
correct).
In a first attempt, we can consider only conditions that have the simple
possible form, which is "(compare simpleExpression simpleExpression)",
where "simpleExpression" is either "(ldind.* local[*])" or
"(ldlen (ldind.ref local[*]))" (or a constant, of course).
We can also "summarize" variable definitions, keeping only what is relevant
to know: if they are greater or smaller than zero or other variables.
One particular note on PHI functions: they work (obviously) like the logical
"or" of their definitions, and therefore are equivalent to the "logical or"
of the summarization of their definitions.
About recursive definitions (which, believe me, are the worst thing in all
this mess), we handle only "monotonic" ones. That is, we try to understand
if the recursive definition (which, as we said above, must happen because
of a loop) always "grows" or "gets smaller". In all other cases, we decide
we cannot handle it.
So, the algorithm to optimize one method could be something like:
[1] Build the SSA representation.
[2] Analyze each BB. Particularly, do the following:
[2a] summarize its exit conditions
[2b] summarize all the variable definitions occurring in the BB
[2c] check if it contains array accesses (only as optimization, so that
step 3 will be performed only when necessary)
[3] Perform the removal. For each BB that contains array accesses, do:
[3a] build an evaluation area where for each variable we keep track
of its relations with zero and with each other variable.
[3b] populate the evaluation area with the initial data (the variable
definitions and the complete entry condition of the BB)
[3c] propagate the relations between variables (so that from "a<b" and
"b<c" we know that "a<c") in the evaluation area
[3d] for each array access in the BB, use the evaluation area to "match
conditions", to see if the goal functions are satisfied
------------------------------------------------------------------------------
LOW LEVEL DESIGN DECISIONS
One of the problems I had to solve was the representation of the relations
between symbolic values.
The "canonical" solution would be to have a complex, potentially recursive
data structure, somewhat similar to MonoInst, and use it all the time (maybe
MonoInst could be used "as is").
The problem is that, at this point, the software would have to "play" with
these structures, for instance to deduce "b < a-1" from "a > b+1", because we
are interested in what value b has, and not a. And maybe the software should
be able to do this without modifying the original condition, because it was
used elsewhere (or could be reused).
Then, the software should also be able to apply logical relations to
conditions (like "(a>0)||(c<1)"), and manipulate them, too.
In my opinion, an optimizer written in this way risks to be too complex, and
its performance not acceptable in a JIT compiler.
Therefore, I decided to represent values in a "summarized" way.
For instance, "x = a + 1" becomes "x > a" (and, by the way, "x = x + 1"
would look like "x > x", which means that "x grows").
So, each variable value is simply a "relation" of the variable with other
variables or constants. Anything more complex is ignored.
Relations are represented with flags (EQ, LT and GT), so that it is easy
to combine them logically (it is enough to perform the bitwise operations
on the flags). The condition (EQ|LT|GT) means we know nothing special, any
value is possible. The empty condition would mean an unreachable statement
(because no value is acceptable).
There is still the problem of identifying variables, and where to store all
these relations. After some thinking, I decided that a "brute force" approach
is the easier, and probably also the fastest. In practice, I keep in memory
an array of "variable relations", where for each variable I record its
relation with the constants 0 and 1 and an array of its relations with all
other variables. The evaluation area, therefore, looks like a square matrix
as large as the number of variables. I *know* that the matrix is rather sparse
(in the sense that only few of its cells are significant), but I decided that
this data structure was the simplest and more efficient anyway. After all,
each cell takes just one byte, and any other data structure must be some
kind of associative array (like a g_hash_table). Such a data structue is
difficult to use with a MonoMemPool, and in the end is slower than a plain
array, and has the storage overhead of all the pointers... in the end, maybe
the full matrix is not a waste at all.
<note>
After a few benchmarks, I clearly see that the matrix works well if the
variables are not too much, otherwise it gets "too sparse", and much time
is wasted in the propagation phase. Maybe the data structure could be changed
after a more careful evaluation of the problem...
</note>
With these assumptions, all the memory used to represent values, conditions
and the evaluation area can be allocated in advance (from a MonoMemPool), and
used for all the analysis of one method. Moreover, the performance of logical
operations is high (they are simple bitwise operations on bytes).
<note>
For the propagation phase, there was a function I could not easily code
with bitwise operations. As that phase is a real pain for the performance
(the complexity is quadratic, and it must be executed iteratively until
nothing changes), I decided to code it in perl, and generate a precomputed
table then used by the C code (just 64 bytes). I think this is faster than
all those branches... Really, I could think again to phases 3c and 3d (see
above), maybe there is a better way to do them.
</note>
Of course not all expressions can be handled by the summarization, so the
optimizer is not as effective as a full blown theorem prover...
Another thing related to design... the implementation is as less intrusive
as possible.
I "patched" the main files to add an option for abc removal, and handle it
just like any other option, and then wrote all the relevant code in two
separated files (one header for data structures, and a C file for the code).
If the patch turns out to be useful also for other pourposes (like a more
generic "unreachable code elimination", that detects branches that will never
be taken using the analysis of variable values), it will not be difficult
to integrate the branch summarizations into the BasicBlock data structures,
and so on.
One last thing... in some places, I knowingly coded in an inefficient way.
For instance, "remove_abc_from_block" does something only if the flag
"has_array_access_instructions" is true, but I call it anyway, letting it
do nothing (calling it conditionally would be better). There are several
things like this in the code, but I was in a hurry, and preferred to make
the code more readable (it is already difficult to understand as it is).
------------------------------------------------------------------------------
WHEN IS ABC REMOVAL POSSIBLE? WHEN IS IT USEFUL?
Warning! Random ideas ahead!
In general, managed code should be safe, so abc removal is possible only if
there already is some branch in the method that does the check.
So, one check on the index is "mandatory", the abc removal can only avoid a
second one (which is useless *because* of the first one).
If in the code there is not a conditional branch that "avoids" to access
an array out of its bounds, in general the abc removal cannot occur.
It could happen, however, that one method is called from another one, and
the check on the index is performed in the caller.
For instance:
void
caller( int[] a, int n ) {
if ( n <= a.Length ) {
for ( int i = 0; i < n; i++ ) {
callee( a, i );
}
}
}
void callee( int[] a, int i ) {
Console.WriteLine( "a [i] = " + a [i] );
}
In this case, it should be possible to have two versions of the callee, one
without checks, to be called from methods that do the check themselves, and
another to be called otherwise.
This kind of optimization could be profile driven: if one method is executed
several times, wastes CPU time for bound checks, and is mostly called from
another one, it could make sense to analyze the situation in detail.
However, one simple way to perform this optimization would be to inline
the callee (so that the abc removal algorithm can see both methods at once).
The biggest benefits of abc removal can be seen for array accesses that
occur "often", like the ones inside inner loops. This is why I tried to
handle "recursive" variable definitions, because without them you cannot
optimize inside loops, which is also where you need it most.
Another possible optimization (alwais related to loops) is the following:
void method( int[] a, int n ) {
for ( int i = 0; i < n; i++ ) {
Console.WriteLine( "a [i] = " + a [i] );
}
}
In this case, the check on n is missing, so the abc could not be removed.
However, this would mean that i will be checked twice at each loop iteration,
one against n, and the other against a.Length. The code could be transformed
like this:
void method( int[] a, int n ) {
if ( n < a.Length ) {
for ( int i = 0; i < n; i++ ) {
Console.WriteLine( "a [i] = " + a [i] ); // Remove abc
}
} else {
for ( int i = 0; i < n; i++ ) {
Console.WriteLine( "a [i] = " + a [i] ); // Leave abc
}
}
}
This could result in performance improvements (again, probably this
optimization should be profile driven).
------------------------------------------------------------------------------
OPEN ISSUES
There are several issues that should still be addressed...
One is related to aliasing. For now, I decided to operate only on local
variables, and ignore aliasing problems (and alias analisys has not yet
been implemented anyway).
Actually, I identified the local arrays with their length, because I
totally ignore the contents of arrays (and objects in general).
Also, in several places in the code I only handle some cases, and ignore
other more complex ones, which could anyway work (there are comments where
this happens). Anyway, this is not harmful (the code is not as effective
as it could be).
Another possible improvement is the explicit handling of all constants.
For now, code like
void
method () {
int[] a = new int [16];
for ( int i = 0; i < 16; i++ ) {
a [i] = i;
}
}
is not handled, because the two constants "16" are lost when variable
relations are summarized (actually they are not lost, they are stored in the
summarized values, but they are not yet used correctly).
The worst thing, anyway, is that for now I fail completely in distinguishing
between signed and unsigned variables/operations/conditions, and between
different integer sizes (in bytes). I did this on pourpose, just for lack of
time, but this can turn out to be terribly wrong.
The problem is caused by the fact that I handle arithmetical operations and
conditions as "ideal" operations, but actually they can overflow and/or
underflow, giving "surprising" results.
For instance, look at the following two methods:
public static void testSignedOverflow()
{
Console.WriteLine( "Testing signed overflow..." );
int[] ia = new int[70000];
int i = 1;
while ( i <ia.Length )
{
try
{
Console.WriteLine( " i = " + i );
ia[i] = i;
}
catch ( IndexOutOfRangeException e )
{
Console.WriteLine( "Yes, we had an overflow (i = " + i + ")" );
}
i *= 60000;
}
Console.WriteLine( "Testing signed overflow done." );
}
public static void testUnsignedOverflow()
{
Console.WriteLine( "Testing unsigned overflow..." );
int[] ia = new int[70000];
uint i = 1;
while ( i <ia.Length )
{
try
{
Console.WriteLine( " i = " + i );
ia[i] = (int) i;
}
catch ( IndexOutOfRangeException e )
{
Console.WriteLine( "Yes, we had an overflow (i = " + i + ")" );
}
i *= 60000;
}
Console.WriteLine( "Testing unsigned overflow done." );
}
In the "testSignedOverflow" method, the exception will be thrown, while
in the "testUnsignedOverflow" everything will go on smoothly.
The problem is that the test "i <ia.Length" is a signed comparison in the
first case, and will not exit from the loop if the index is negative.
Therefore, in the first method, the abc should not be removed.
In practice, the abc removal code should try to predict if any given
expression could over/under-flow, and act accordingly.
For now, I made so that the only accepted expressions are plain increments
and decrements of variables (which cannot over/under-flow without being
trapped by the existing conditions). Anyway, this issue should be analyzed
better.
By the way, in summarizations there is a field that keeps track of the
relation with "1", which I planned to use exactly to summarize products and
quotients, but it is never used (as I said, only increment and decrement
operations are properly summarized).
-44
View File
@@ -1,44 +0,0 @@
Mono Ahead Of Time Compiler
===========================
The new mono JIT has sophisticated optimization features. It uses SSA and has a
pluggable architecture for further optimizations. This makes it possible and
efficient to use the JIT also for AOT compilation.
* file format: We use the native object format of the platform. That way it is
possible to reuse existing tools like objdump and the dynamic loader. All we
need is a working assembler, i.e. we write out a text file which is then
passed to gas (the gnu assembler) to generate the object file.
* file names: we simply add ".so" to the generated file. For example:
basic.exe -> basic.exe.so
corlib.dll -> corlib.dll.so
* staring the AOT compiler: mini --aot assembly_name
The following things are saved in the object file:
* version infos:
* native code: this is labeled with method_XXXXXXXX: where XXXXXXXX is the
hexadecimal token number of the method.
* additional informations needed by the runtime: For example we need to store
the code length and the exception tables. We also need a way to patch
constants only available at runtime (for example vtable and class
addresses). This is stored i a binary blob labeled method_info_XXXXXXXX:
PROBLEMS:
- all precompiled methods must be domain independent, or we add patch infos to
patch the target doamin.
- the main problem is how to patch runtime related addresses, for example:
- current application domain
- string objects loaded with LDSTR
- address of MonoClass data
- static field offsets
- method addreses
- virtual function and interface slots
-57
View File
@@ -1,57 +0,0 @@
HOWTO bundle assemblies inside the mono runtime.
Paolo Molaro ([email protected])
* Intent
Bundling assemblies inside the mono runtime may be useful for a number
of reasons:
* creating a standalone complete runtime that can be more easily
distributed
* having an application run against a known set of assemblies
that has been tested
Of course, there are drawbacks, too: if there has been fixes
to the assemblies, replacing them means recompiling the
runtime as well and if there are other mono apps, unless they
use the same mono binary, there will be less opportunities for
the operating system to optimize memory usage. So use this
feature only when really needed.
* Creating the Bundle
To bundle a set of assemblies, you need to create a file that
lists the assembly names and the relative files. Empty lines
and lines starting with # are ignored:
== cut cut ==
# Sample bundle template
mscorlib: /path/to/mscorlib/assembly.dll
myapp: /path/to/myapp.exe
== cut cut ==
Next you need to build the mono runtime using a special configure option:
./configure --with-bundle=/path/to/bundle/template
The path to the template should be an absolute path.
The script metadata/make-bundle.pl will take the specifie
assemblies and embed them inside the runtime where the loading
routines can find them before searching for them on disk.
* Open Issues
There are still two issues to solve:
* config files: sometimes they are needed but they are
not yet bundled inside the library ()
* building with the included libgc makes it not
possible to build a mono binary statically linked to
libmono: this needs to be fixed to make bundles
really useful.
-271
View File
@@ -1,271 +0,0 @@
Embedding the Mono runtime, preliminary version
Miguel de Icaza ([email protected]),
Paolo Molaro ([email protected])
This document describes how to embed the Mono runtime in your
application, and how to invoke CIL methods from C, and how to
invoke C code from CIL. Both the JIT and interpreter can be
embedded in very similar ways so most of what is described
here can be used in either case.
* Embedding the runtime.
Embedding the runtime consists of various steps:
* Compiling and linking the Mono runtime
* Initializing the Mono runtime
* Optionally expose C code to the C#/CIL universe.
These are discussed in detail next.
** Compiling and Linking
To embed the runtime, you have to link your code against the
Mono runtime libraries. To do this, you want to pass the
flags returned by pkg-config to your compiler:
pkg-config --cflags --libs mono
is used to get the flags for the JIT runtime and
pkg-config --cflags --libs mint
for the interpreted runtime.
Like this:
gcc sample.c `pkg-config --cflags --libs mono`
You can separate the compilation flags from the linking flags, for
instance, you can use the following macros in your makefile:
CFLAGS=`pkg-config --cflags mono`
LDFLAGS=`pkg-config --libs mono`
** Initializing the Mono runtime
To initialize the JIT runtime, call mono_jit_init, like this:
#include <mono/mini/jit.h>
MonoDomain *domain;
domain = mono_jit_init ("domain-name");
For the interpreted runtime use mono_interp_init instead:
#include <mono/interpreter/embed.h>
MonoDomain *domain;
domain = mono_interp_init ("domain-name");
That will return a MonoDomain where your code will be
executed. You can create multiple domains. Each domain is
isolated from the other domains and code in one domain will
not interfere with code in other domains. This is useful if
you want to host different applications in your program.
Then you can load an assembly containing code into the domain:
MonoAssembly *assembly;
assembly = mono_domain_assembly_open (domain, "file.dll");
if (!assembly)
error ();
In the above example, the contents of `file.dll' will be
loaded into the domain. This only loads the code, but it will
not execute anything yet. You can replace `file.dll' with
another transport file, like `file.exe'
To start executing code, you must invoke a method in the
assembly, or if you have provided a static Main method (an
entry point), you can use the convenience function:
retval = mono_jit_exec (domain, assembly, argc - 1, argv + 1);
or when using the interpreter use:
retval = mono_interp_exec (domain, assembly, argc - 1, argv + 1);
If you want to invoke a different method, look at the
`Invoking Methods in the CIL universe' section later on.
** Shutting down the runtime
To shutdown the Mono runtime, you have to clean up all the
domains that were created, use this function:
mono_jit_cleanup (domain);
Or in the case of the interpreted runtime use:
mono_interp_cleanup (domain);
** Applications that use threads.
The Boehm GC system needs to catch your calls to the pthreads
layer, so in each file where you use pthread.h you should
include the <gc/gc.h> file.
* Exposing C code to the CIL universe
The Mono runtime provides two mechanisms to expose C code to
the CIL universe: internal calls and native C code. Internal
calls are tightly integrated with the runtime, and have the
least overhead, as they use the same data types that the
runtime uses.
The other option is to use the Platform Invoke (P/Invoke) to
call C code from the CIL universe, using the standard P/Invoke
mechanisms.
To register an internal call, use this call in the C code:
mono_add_internal_call ("Hello::Sample", sample);
Now, you need to declare this on the C# side:
using System;
using System.Runtime.CompilerServices;
class Hello {
[MethodImplAttribute(MethodImplOptions.InternalCall)]
extern static string Sample ();
}
Since this routine returns a string, here is the C definition:
static MonoString*
Sample ()
{
return mono_string_new (mono_domain_get (), "Hello!");
}
Notice that we have to return a `MonoString', and we use the
`mono_string_new' API call to obtain this from a string.
* Invoking Methods in the CIL universe
Calling a method in the CIL universe from C requires a number of steps:
* Obtaining the MonoMethod handle to the method.
* The method invocation.
** Obtaining a MonoMethod
To get a MonoMethod there are several ways.
You can get a MonoClass (the structure representing a type)
using:
MonoClass *
mono_class_from_name (MonoImage *image, const char* name_space, const char *name);
and then loop in the returned class method array until you get
the one you're looking for. There are examples of such
searches as static functions in several C files in
metadata/*.c: we need to expose one through the API and remove
the duplicates.
The other, simpler, way is to use the functions in
debug-helpers.h: there are examples of their use in monograph,
mint and the jit as well. You basically use a string
description of the method, like:
"System.Object:GetHashCode()"
and create a MonoMethodDesc out of it with:
MonoMethodDesc* mono_method_desc_new (const char *name, gboolean include_namespace);
You can then use:
MonoMethod* mono_method_desc_search_in_class (MonoMethodDesc *desc, MonoClass *klass);
MonoMethod* mono_method_desc_search_in_image (MonoMethodDesc *desc, MonoImage *image);
to search for the method in a class or in an image. You would
tipically do this just once at the start of the program and
store the result for reuse somewhere.
** Invoking a Method
There are two functions to call a managed method:
MonoObject*
mono_runtime_invoke (MonoMethod *method, void *obj, void **params,
MonoObject **exc);
and
MonoObject*
mono_runtime_invoke_array (MonoMethod *method, void *obj, MonoArray *params,
MonoObject **exc);
obj is the 'this' pointer, it should be NULL for static
methods, a MonoObject* for object instances and a pointer to
the value type for value types.
These functions can be used in both the JIT and the interpreted
environments.
The params array contains the arguments to the method with the
same convention: MonoObject* pointers for object instances and
pointers to the value type otherwise. The _invoke_array
variant takes a C# object[] as the params argument (MonoArray
*params): in this case the value types are boxed inside the
respective reference representation.
From unmanaged code you'll usually use the
mono_runtime_invoke() variant.
Note that this function doesn't handle virtual methods for
you, it will exec the exact method you pass: we still need to
expose a function to lookup the derived class implementation
of a virtual method (there are examples of this in the code,
though).
You can pass NULL as the exc argument if you don't want to
catch exceptions, otherwise, *exc will be set to the exception
thrown, if any. if an exception is thrown, you can't use the
MonoObject* result from the function.
If the method returns a value type, it is boxed in an object
reference.
We have plans for providing an additional method that returns
an unmanaged->managed thunk like this:
void* mono_method_get_unmanaged_thunk (MonoMethod *method);
You'll be able to store the returned pointer in a function
pointer with the proper signature and call that directly from
C:
typedef gint32 (*GetHashCode) (MonoObject *obj);
GetHashCode func = mono_method_get_unmanaged_thunk (System_Object_GetHashCode_method);
gint32 hashvalue = func (myobject);
It may not be possible to manage exceptions in that case,
though. I need to think more about it.
** Threading issues
If your application creates threads on its own, and you want them to
be able to call code into the CIL universe with Mono, you have to
register the thread with Mono before issuing the call.
To do so, call the mono_thread_attach() function before you execute
any managed code from the thread
* Samples
See the sample programs in mono/sample/embed for examples of
embedding the Mono runtime in your application.
-110
View File
@@ -1,110 +0,0 @@
Exception Implementation in the Mono Runtime
Dietmar Maurer ([email protected])
(C) 2001 Ximian, Inc.
Exception implementation (jit):
===============================
Stack unwinding:
================
We record the code address (start_address, size) of all methods. That way it is
possible to map an instruction pointer (IP) to the method information needed
for unwinding the stack:
We also save a Last Managed Frame (LMF) structure at each call from managed to
unmanaged code. That way we can recover from exceptions inside unmanaged code.
void handle_exception ((struct sigcontext *ctx, gpointer obj)
{
if (ctx->bp < mono_end_of_stack) {
/* unhandled exception */
abort ();
}
info = mono_jit_info_table_find (mono_jit_info_table, ctx->ip);
if (info) { // we are inside managed code
if (ch = find_catch_handler ())
execute_catch_handler (ch, ctx, obj);
execute_all_finally_handler ();
// restore register, including IP and Frame pointer
ctx = restore_caller_saved_registers_from_ctx (ji, ctx);
// continue unwinding
handle_exception (ctx, obj);
} else {
lmf = get_last_managed_frame ();
// restore register, including IP and Frame pointer
ctx = restore_caller_saved_registers_from_lmf (ji, lmf);
// continue unwinding
handle_exception (ctx, obj);
}
}
Code generation:
================
leave: is simply translated into a branch to the target. If the leave
instruction is inside a finally block (but not inside another handler)
we call the finally handler before we branch to the target.
finally/endfinally, filter/endfilter: is translated into subroutine ending with
a "return" statement. The subroutine does not save EBP, because we need access
to the local variables of the enclosing method. Its is possible that
instructions inside those handlers modify the stack pointer, thus we save the
stack pointer at the start of the handler, and restore it at the end. We have
to use a "call" instruction to execute such finally handlers. This makes it
also possible to execute them inside the stack unwinding code. The exception
object for filters is passed in a local variable (cfg->exvar).
throw: we first save all regs into a sigcontext struct and then call the stack
unwinding code.
catch handler: catch hanlders are always called from the stack unwinding
code. The exception object is passed in a local variable (cfg->exvar).
gcc support for Exceptions
==========================
gcc supports exceptions in files compiled with the -fexception option. gcc
generates DWARF exceptions tables in that case, so it is possible to unwind the
stack. The method to read those exception tables is contained in libgcc.a, and
in newer versions of glibc (glibc 2.2.5 for example), and it is called
__frame_state_for(). Another usable glibc function is backtrace_symbols() which
returns the function name corresponding to a code address.
We dynamically check if those features are available using g_module_symbol(),
and we use them only when available. If not available we use the LMF as
fallback.
Using gcc exception information prevents us from saving the LMF at each native
call, so this is a way to speed up native calls. This is especially valuable
for internal calls, because we can make sure that all internal calls are
compiled with -fexceptions (we compile the whole mono runtime with that
option).
All native function are able to call function without exception tables, and so
we are unable to restore all caller saved registers if an exception is raised
in such function. Well, its possible if the previous function already saves all
registers. So we only omit the the LMF if a function has an exception table
able to restore all caller saved registers.
One problem is that gcc almost never saves all caller saved registers, because
it is just unnecessary in normal situations. But there is a trick forcing gcc
to save all register, we just need to call __builtin_unwind_init() at the
beginning of a function. That way gcc generates code to save all caller saved
register on the stack.
-173
View File
@@ -1,173 +0,0 @@
Rules for opening shared files
==============================
File is already open, with share set to none:
Can not open again
File is already open for reading, with read share:
Can open for reading only, share must include read (can have write too)
File is already open for reading, with write share:
Can open for writing only, share must include read (can have write too)
File is already open for reading, with read + write share:
Can open for read, writing or both, share must include read (can have write too)
File is already open for writing, with read share:
Can open for reading only, share must include write (can have read too)
File is already open for writing, with write share:
Can open for writing only, share must include write (can have read too)
File is already open for writing, with read + write share:
Can open for reading, writing or both, share must include write (can have read too)
File is already open for reading + writing, with read share:
Can open for reading only, share must be read + write
File is already open for reading + writing, with write share:
Can open for for writing only, share must be read + write
File is already open for reading + writing, with read + write share:
Can open for read, writing or both, share must be read + write
Executive Summary
-----------------
Second open must have access within first share, must set second share to at least first access
Documenting code
----------------
#include <stdio.h>
#include <windows.h>
int access[] = {
GENERIC_READ,
GENERIC_WRITE,
GENERIC_READ | GENERIC_WRITE
};
char *access_names[] = {
"G_READ",
"G_WRITE",
"G_READ|G_WRITE"
};
int share[] = {
FILE_SHARE_READ,
FILE_SHARE_WRITE,
FILE_SHARE_READ | FILE_SHARE_WRITE
};
char *share_names[] = {
"SHARE_READ",
"SHARE_WRITE",
"SHARE_READ|SHARE_WRITE"
};
void lockfiles(int access1, int share1, int access2, int share2)
{
HANDLE h1, h2;
BOOL ret;
if (access2 == 0 && share2 == 0) {
printf("\n");
printf("%22.22s\n%22.22s", access_names[access1], share_names[share1]);
}
h1 = CreateFile("lockedfile",
access[access1],
share[share1],
NULL, OPEN_ALWAYS, FILE_ATTRIBUTE_NORMAL, NULL);
if (h1 == INVALID_HANDLE_VALUE) {
printf("Open1 failed: %d\n", GetLastError());
return;
}
h2 = CreateFile("lockedfile",
access[access2],
share[share2],
NULL, OPEN_ALWAYS, FILE_ATTRIBUTE_NORMAL, NULL);
if (h2 == INVALID_HANDLE_VALUE) {
printf(" %4.4s", "");
} else {
printf(" %4.4s", "OK");
CloseHandle(h2);
}
CloseHandle(h1);
}
int main(int argc, char **argv)
{
int i, j, k, l;
printf("\t\t\t\t\t\t\tSecond Open\n");
printf("%22.22s G_RE G_RE G_RE G_WR G_WR G_WR G_RW G_RW G_RW\n", "");
printf("%22.22s S_RE S_WR S_RW S_RE S_WR S_RW S_RE S_WR S_RW", "First open --v ");
for (i = 0; i < 3; i++) {
for (j = 0; j < 3; j++) {
for (k = 0; k < 3; k++) {
for (l = 0; l < 3; l++) {
lockfiles(i, j, k, l);
}
}
}
}
return(0);
}
Code output
-----------
Second Open
G_RE G_RE G_RE G_WR G_WR G_WR G_RW G_RW G_RW
First open --v S_RE S_WR S_RW S_RE S_WR S_RW S_RE S_WR S_RW
G_READ
SHARE_READ OK OK
G_READ
SHARE_WRITE OK OK
G_READ
SHARE_READ|SHARE_WRITE OK OK OK OK OK OK
G_WRITE
SHARE_READ OK OK
G_WRITE
SHARE_WRITE OK OK
G_WRITE
SHARE_READ|SHARE_WRITE OK OK OK OK OK OK
G_READ|G_WRITE
SHARE_READ OK
G_READ|G_WRITE
SHARE_WRITE OK
G_READ|G_WRITE
SHARE_READ|SHARE_WRITE OK OK OK
-31
View File
@@ -1,31 +0,0 @@
* GC issues you need to be careful about when hacking on the mono runtime.
mono currently uses the Boehm garbage collection library. This is a conservative
GC package: this means that the memory is searched for possible memory references
to chunks allocated with the GC. Not all the memory is searched, but only the memory
allocated with the GC itself (unless you use the 'atomic' variant of the allocation
routines), the stack and global variables. Also, if the last reference to an object
is stored in an area of themory that is not searched, the object may be freed resulting
in memory corruption ind segfaults.
In particular, memory allocated with system's malloc() is not searched, so you need to be
careful NOT to store object references in memory allocated with malloc() (unless you are sure
that the object reference will be live at the same time in some other area under the control
of the GC (on the stack or in a global variable, for example). Since g_malloc()
ultimately calls system malloc() the data structures in GLib are not safe to
use to store object references.
Occasionally, you'll need to store some object reference from C code: in this case,
you must make sure that the store location is searched by the GC. You can safely
use thread local storage areas, global variables, stack variables. If you need a more
complicated data structure, you can use a hash table: MonoGHashTable.
This hash table has the same interface as GHashTable from GLib, just stick the "mono_"
prefix in function calls and the "Mono" prefix in the hash table type name.
This hash table ensures that object references stored in it are tracked by the GC, as long
as the pointer to the hash is tracked itself.
Other data structures that are allocated with the GC and are safe to use to store pointers
to GC-allocated memory, are MonoDomain (keeps track of many domain specfic objects)
and MonoVTable (referenced by MonoDomain: keeps track of the static data of a type
since that can hold references to objects).
-84
View File
@@ -1,84 +0,0 @@
Internal Call Topics
* Introduction
The Common Language Infrastructure allows for methods to be
implemented in unmanaged code. Unlike the Platform Invocation
services which provide marshalling and unmarshalling of data
from managed to unmanaged and viceversa the Internal calls do
not perform any kind of marshalling.
* Basic Type mapping
The following lists how the C# types are exposed to the C API.
C# type C type
-----------------------------
char gunichar2
bool MonoBoolean
sbyte signed char
byte guchar
short gint16
ushort guint16
int gint32
uint guint32
long gint64
ulong guint64
IntPtr/UIntPtr gpointer
object MonoObject*
string MonoString*
* Pointers
For ref and out paramaters you'll use the corresponding
pointer type.
So if you have a C# type listed as "ref int", you should use
"int *" in your implementation.
* Arrays
Arrays of any type must be described with a MonoArray* and the
elements must be accessed with the mono_array_* macros.
* Other Structures
Any other type that has a matching C structure representation,
should use a pointer to the struct instead of a generic
MonoObject pointer.
* Instance Methods.
Instance methods that are internal calls will receive as first argument
the instance object, so you must account for it in the C method signature:
[MethodImplAttribute(MethodImplOptions.InternalCall)]
public extern override int GetHashCode ();
becomes:
gint32 ves_icall_System_String_GetHashCode (MonoString *this);
* How to hook internal calls with the runtime
Once you require an internal call in corlib, you need to
create a C implementation for it and register it in a static
table in metadata/icall.c. Add an entry in the table like:
"System.String::GetHashCode", ves_icall_System_String_GetHashCode,
Note that you need to include the full namespace.name of the
class. If there are overloaded methods, you need also to
specify the signature of _all_ of them:
[MethodImplAttribute(MethodImplOptions.InternalCall)]
public extern override void DoSomething ();
[MethodImplAttribute(MethodImplOptions.InternalCall)]
public extern override void DoSomething (bool useful);
should be mapped with:
"Namespace.ClassName::DoSomething()", ves_icall_Namespace_ClassName_DoSomething,
"Namespace.ClassName::DoSomething(bool)", ves_icall_Namespace_ClassName_DoSomething_bool,
-54
View File
@@ -1,54 +0,0 @@
Mono Intermediate Representation Documentation
Operation: OP_LOCALLOC
inst_left:
The value
ins->inst_left = *sp;
ins->cil_code = ip;
This is the equivalent of alloca() in a C implementation: it
allocates a bunch of bytes on the stack dynamically. Local
variables are allocated statically in the stack instead.
It is implemented by moving the stack pointer according to the
platform ABI: you may need to keep a specific frame register
to address local variables when LOCALLOC is used (the PPC code
does this, for example: ifyou're doing the s390 port I suggest
starting from that)
Operation: OP_CALL_HANDLER.
cil_code:
Points to the CIL where the LEAVE or LEAVE_S
instruction was found.
inst_target_bb:
The target basic block for this invocation.
This instruction must call the target basic block.
Operation: OP_CISINST
Implements the CEE_MONO_CISINST opcode that is decoded from
the CIL stream.
This opcode takes as input an object reference and a class,
and leaves on the assigned register:
0. if the object is an instance of the class,
1. if the object is not instance of the class,
2. if the object is a proxy whose type cannot be determined.
Operation: OP_CCASTCLASS
Implements the CEE_MONO_CCASTCLASS opcode that is decoded from
the CIL stream.
This opcode takes as input an object reference and a class,
and leaves on the assigned register:
0. if the object is an instance of the class,
1. if the object is a proxy whose type cannot be determined
throws an InvalidCastException exception otherwise.
-64
View File
@@ -1,64 +0,0 @@
Register Allocation
===================
The current JIT implementation uses a tree matcher to generate code. We use a
simple algorithm to allocate registers in trees, and limit the number of used
temporary register to 4 when evaluating trees. So we can use 2 registers for
global register allocation.
Register Allocation for Trees
=============================
We always evaluate trees from left to right. When there are no more registers
available we need to spill values to memory. Here is the simplified algorithm.
gboolean
tree_allocate_regs (tree, exclude_reg)
{
if (!tree_allocate_regs (tree->left, -1))
return FALSE;
if (!tree_allocate_regs (tree->right, -1)) {
tree->left->spilled == TRUE;
free_used_regs (tree->left);
if (!tree_allocate_regs (tree->right, tree->left->reg))
return FALSE;
}
free_used_regs (tree->left);
free_used_regs (tree->right);
/* try to allocate a register (reg != exclude_reg) */
if ((tree->reg = next_free_reg (exclude_reg)) != -1)
return TRUE;
return FALSE;
}
The emit routing actually spills the registers:
tree_emit (tree)
{
tree_emit (tree->left);
if (tree->left->spilled)
save_reg (tree->left->reg);
tree_emit (tree->right);
if (tree->left->spilled)
restore_reg (tree->left->reg);
emit_code (tree);
}
Global Register Allocation
==========================
TODO.
-125
View File
@@ -1,125 +0,0 @@
Just some thoughts for the JITer:
General issues:
===============
We are designing a JIT compiler, so we have to consider two things:
- the quality of the generated code
- the time needed to generate that code
The current approach is to keep the JITer as simple as possible, and thus as
fast as possible. The generated code quality will suffer from that.
Register Allocation:
====================
With lcc you can assign a fixed register to a tree before register
allocation. For example this is needed by call, which return the value always
in EAX on x86. The current implementation works without such system, due to
special forest generation.
Different Register Sets:
========================
Most processors have more that one register set, at least one for floating
point values, and one for integers. Should we support architectures with more
that two sets? Does someone knows such an architecture?
64bit Integer Values:
=====================
I can imagine two different implementation. On possibility would be to treat
long (64bit) values simply like any other value type. This implies that we
call class methods for ALU operations like add or sub. Sure, this method will
be be a bit inefficient.
The more performant solution is to allocate two 32bit registers for each 64bit
value. We add a new non terminal to the monoburg grammar called "lreg". The
register allocation routines takes care of this non terminal and allocates two
32 bit registers for them.
Forest generation:
==================
Consider the following code:
OPCODE: STACK LOCALS
LDLOC.0 (5) [5,0]
LDC.1 (5,1) [5,0]
STLOC.0 (5) [1,0]
STLOC.1 () [1,5]
A simple forest generation generates:
STLOC.0(LDC.1)
STLOC.1(LDLOC.0)
Which is wrong, since it stores the wrong value (1 instead of 5). Instead we
must generate something like:
STLOC.TMP(LDLOC.0)
STLOC.0(LDC.1)
STLOC.1(LDLOC.TMP)
Where STLOC.TMP saves the value into a new temporary variable.
We also need a similar solution for basic block boundaries when the stack depth
is not zero. We can simply save those values to new temporary values. Consider
the following basic block with one instruction:
LDLOC.1
This should generate a tree like:
STLOC.TMP(LDLOC.1) Please notice that an intelligent register allocator can
still allocate registers for those new variables.
DAG handling:
=============
Monoburg can't handle DAGs, instead we need real trees as input for
the code generator. So we have two problems:
1.) DUP instruction: This one is obvious - we need to store the value
into a temporary variable to solve the problem.
2.) function calls: Chapter 12.8, page 343 of "A retargetable C compiler"
explains that: "because listing a call node will give it a hidden reference
from the code list". I don't understand that (can someone explain that?), but
there is another reason to save return values to temporaries: Consider the
following code:
x = f(y) + g(z); // all functions return integers
We could generate such a tree for this expression: STLOC(ADD(CALL,CALL))
The problem is that both calls returns the value in the same register,
so it is non trivial to generate code for that tree. We must copy one
register into another one, which make register allocation more complex.
The easier solution is store the result of function calls to
temporaries. This leads to the following forest:
STLOC(CALL)
STLOC(CALL)
STLOC(ADD (LDLOC, LDLOC))
This is what lcc is doing, if I understood 12.8, page 342, 343?
Possible Optimisations:
=======================
Miguel said ORP does some optimisation on IL level, for example moving array
bounds checking out of loops:
for (i = 0; i < N; i++) { check_range (a, i); a [i] = X; }
id transformed to:
if (in_range (a, 0, N)) { for (i = 0; i < N; i++) a[i] = X; }
else for (i = 0; i < N; i++) { check_range (a, i); a [i] = X; }
The "else" is only to keep original semantics (exception handling).
We need loop detection logic in order to implement this (dominator tree).
AFAIK CACAO also implements this.
-52
View File
@@ -1,52 +0,0 @@
Author: Dietmar Maurer ([email protected])
(C) 2001 Ximian, Inc.
Howto trigger JIT compilation
=============================
The JIT translates CIL code to native code on a per method basis. For example
if you have this simple program:
public class Test {
public static void Main () {
System.Console.WriteLine ("Hello");
}
}
the JIT first compiles the Main function. Unfortunately Main() contains another
reference to System.Console.WriteLine(), so the JIT also needs the address for
WriteLine() to generate a call instruction.
The simplest solution would be to JIT compile System.Console.WriteLine()
to generate that address. But that would mean that we JIT compile half of our
class library at once, since WriteLine() uses many other classes and function,
and we have to call the JIT for each of them. Even worse there is the
possibility of cyclic references, and we would end up in an endless loop.
Thus we need some kind of trampoline function for JIT compilation. Such a
trampoline first calls the JIT compiler to create native code, and then jumps
directly into that code. Whenever the JIT needs the address of a function (to
emit a call instruction) it uses the address of those trampoline functions.
One drawback of this approach is that it requires an additional indirection. We
always call the trampoline. Inside the trampoline we need to check if the
method is already compiled or not, and when not compiled we start JIT
compilation. After that we call the code. This process is quite time consuming
and shows very bad performance.
The solution is to add some logic to the trampoline function to detect from
where it is called. It is then possible for the JIT to patch the call
instruction in the caller, so that it directly calls the JIT compiled code
next time.
Implementation for x86
======================
emit-x86.c (arch_create_jit_trampoline): return the JIT trampoline function
emit-x86.c (x86_magic_trampoline): contains the code to detect the caller and
patch the call instruction.
emit-x86.c (arch_compile_method): JIT compile a method
-208
View File
@@ -1,208 +0,0 @@
* Proposal for the local register allocator
The local register allocator deals with allocating registers
for temporaries inside a single basic block, while the global
register allocator is concerned with method-wide allocation of
variables.
The global register allocator uses callee-saved register for it's
purpouse so that there is no need to save and restore these registers
at call sites.
There are a number of issues the local allocator needs to deal with:
*) some instructions expect operands in specific registers (for example
the shl instruction on x86, or the call instruction with thiscall
convention, or the equivalent call instructions on other architectures,
such as the need to put output registers in %oX on sparc)
*) some instructions deliver results only in specific registers (for example
the div instruction on x86, or the call instructionson on almost all
the architectures).
*) it needs to know what registers may be clobbered by an instruction
(such as in a method call)
*) it should avoid excessive reloads or stores to improve performance
While which specific instructions have limitations is architecture-dependent,
the problem shold be solved in an arch-independent way to reduce code duplication.
The register allocator will be 'driven' by the arch-dependent code, but it's
implementation should be arch-independent.
To improve the current local register allocator, we need to
keep more state in it than the current setup that only keeps busy/free info.
Possible state information is:
free: the resgister is free to use and it doesn't contain useful info
freeable: the register contains data loaded from a local (there is
also info about _which_ local it contains) as a result from previous
instructions (like, there was a store from the register to the local)
moveable: it contains live data that is needed in a following instruction, but
the contents may be moved to a different register
busy: the register contains live data and it is placed there because
the following instructions need it exactly in that register
allocated: the register is used by the global allocator
The local register allocator will have the following interfaces:
int get_register ();
Searches for a register in the free state. If it doesn't find it,
searches for a freeable register. Sets the status to moveable.
Looking for a 'free' register before a freeable one should allow for
removing a few redundant loads (though I'm still unsure if such
things should be delegated entirely to the peephole pass).
int get_register_force (int reg);
Returns 'reg' if it is free or freeable. If it is moveable, it moves it
to another free or freeable register.
Sets the status of 'reg' to busy.
void set_register_freeable (int reg);
Sets the status of 'reg' to freeable.
void set_register_free (int reg);
Sets the status of 'reg' to free.
void will_clobber (int reg);
Spills the register to the stack. Sets the status to freeable.
After the clobbering has occurred, set the status to free.
void register_unspill (int reg);
Un-spills register reg and sets the status to moveable.
FIXME: how is the 'local' information represented? Maybe a MonoInst* pointer.
Note: the register allocator will insert instructions in the basic block
during it's operation.
* Examples
Given the tree (on x86 the right argument to shl needs to be in ecx):
store (local1, shl (local1, call (some_arg)))
At the start of the basic block, the registers are set to the free state.
The sequence of instructions may be:
instruction register status -> [%eax %ecx %edx]
start free free free
eax = load local1 mov free free
/* call clobbers eax, ecx, edx */
spill eax free free free
call mov free free
/* now eax contains the right operand of the shl */
mov %eax -> %ecx free busy free
un-spill mov busy free
shl %cl, %eax mov free free
The resulting x86 code is:
mov $fffc(%ebp), %eax
mov %eax, $fff0(%ebp)
push some_arg
call func
mov %eax, %ecx
mov $fff0(%ebp), %eax
shl %cl, %eax
Note that since shl could operate directly on memory, we could have:
push some_arg
call func
mov %eax, %ecx
shl %cl, $fffc(%ebp)
The above example with loading the operand in a register is just to complicate
the example and show that the algorithm should be able to handle it.
Let's take another example with the this-call call convention (the first argument
is passed in %ecx).
In this case, will_clobber() will be called only on %eax and %edx, while %ecx
will be allocated with get_register_force ().
Note: when a register is allocated with get_register_force(), it should be set
to a different state as soon as possible.
store (local1, shl (local1, this-call (local1)))
instruction register status -> [%eax %ecx %edx]
start free free free
eax = load local1 mov free free
/* force load in %ecx */
ecx = load local1 mov busy free
spill eax free busy free
call mov free free
/* now eax contains the right operand of the shl */
mov %eax -> %ecx free busy free
un-spill mov busy free
shl %cl, %eax mov free free
What happens when a register that we need to allocate with get_register_force ()
contains an operand for the next instruction?
instruction register status -> [%eax %ecx %edx]
eax = load local0 mov free free
ecx = load local1 mov mov free
get_register_force (ecx) here.
We have two options:
mov %ecx, %edx
or:
spill %ecx
The first option is way better (and allows the peephole pass to
just load the value in %edx directly, instead of loading first to %ecx).
This doesn't work, though, if the instruction clobbers the %edx register
(like in a this-call). So, we first need to clobber the registers
(so the state of %ecx changes to freebale and there is no issue
with get_register_force ()).
What if an instruction both clobbers a register and requires it as
an operand? Lets' take the x86 idiv instruction as an example: it
requires the dividend in edx:eax and returns the result in eax,
with the modulus in edx.
store (local1, div (local1, local2))
instruction register status -> [%eax %ecx %edx]
eax = load local0 mov free free
will_clobber eax, edx free mov free
force mov %ecx, %eax busy free free
set %edx busy free busy
idiv mov free free
Note: edx is set to free after idiv, because the modulus is not needed
(if it was a rem, eax would have been freed).
If we load the divisor before will_clobber(), we'll have to spill
eax and reload it later. If we load it just after the idiv, there is no issue.
In any case, the algorithm should give the correct results and allow the operation.
Working recursively on the isntructions there shouldn't be huge issues
with this algorithm (though, of course, it's not optimal and it may
introduce excessive spills or register moves). The advantage over the current
local reg allocator is that:
1) the number of spills/moves would be smaller anyway
2) a separate peephole pass could be able to eliminate reg moves
3) we'll be able to remove the 'forced' spills we currently do with
the return value of method calls
* Issues
How to best integrate such a reg allocator with the burg stuff.
Think about a call os sparc with two arguments: they got into %o0 and %o1
and each of them sets the register as busy. But what if the values to put there
are themselves the result of a call? %o0 is no problem, but for all the
next argument n the above algorithm would spill all the 0...n-1 registers...
* Papers
More complex solutions to the local register allocator problem:
http://dimacs.rutgers.edu/TechnicalReports/abstracts/1997/97-33.html
Combining register allocation and instruction scheduling:
http://citeseer.nj.nec.com/motwani95combining.html
More on LRA euristics:
http://citeseer.nj.nec.com/liberatore97hardness.html
Linear-time optimal code scheduling for delayedload architectures
http://www.cs.wisc.edu/~fischer/cs701.f01/inst.sched.ps.gz
Precise Register Allocation for Irregular Architectures
http://citeseer.nj.nec.com/kong98precise.html
Allocate registers first to subtrees that need more of them.
http://www.upb.de/cs/ag-kastens/compii/folien/comment401-409.2.pdf
-771
View File
@@ -1,771 +0,0 @@
A new JIT compiler for the Mono Project
Miguel de Icaza (miguel@{ximian.com,gnome.org}),
Paolo Molaro (lupus@{ximian.com,debian.org})
* Abstract
Mini is a new compilation engine for the Mono runtime. The
new engine is designed to bring new code generation
optimizations, portability and pre-compilation.
In this document we describe the design decisions and the
architecture of the new compilation engine.
* Introduction
Mono is a Open Source implementation of the .NET Framework: it
is made up of a runtime engine that implements the ECMA Common
Language Infrastructure (CLI), a set of compilers that target
the CLI and a large collection of class libraries.
This article discusses the new code generation facilities that
have been added to the Mono runtime.
First we discuss the overall architecture of the Mono runtime,
and how code generation fits into it; Then we discuss the
development and basic architecture of our first JIT compiler
for the ECMA CIL framework. The next section covers the
objectives for the work on the new JIT compiler, then we
discuss the new features available in the new JIT compiler,
and finally a technical description of the new code generation
engine.
* Architecture of the Mono Runtime
The Mono runtime is an implementation of the ECMA Common
Language Infrastructure (CLI), whose aim is to be a common
platform for executing code in multiple languages.
Languages that target the CLI generate images that contain
code in high-level intermediate representation called the
"Common Intermediate Language". This intermediate language is
rich enough to allow for programs and pre-compiled libraries
to be reflected. The execution environment allows for an
object oriented execution environment with single inheritance
and multiple interface implementations.
This runtime provides a number of services for programs that
are targeted to it: Just-in-Time compilation of CIL code into
native code, garbage collection, thread management, I/O
routines, single, double and decimal floating point,
asynchronous method invocation, application domains, and a
framework for building arbitrary RPC systems (remoting) and
integration with system libraries through the Platform Invoke
functionality.
The focus of this document is on the services provided by the
Mono runtime to transform CIL bytecodes into code that is
native to the underlying architecture.
The code generation interface is a set of macros that allow a
C programmer to generate code on the fly, this is done
through a set of macros found in the mono/jit/arch/ directory.
These macros are used by the JIT compiler to generate native
code.
The platform invocation code is interesting, as it generates
CIL code on the fly to marshal parameters, and then this
code is in turned processed by the JIT engine.
* Previous Experiences
Mono has built a JIT engine, which has been used to bootstrap
Mono since January, 2002. This JIT engine has reasonable
performance, and uses an tree pattern matching instruction
selector based on the BURS technology. This JIT compiler was
designed by Dietmar Maurer, Paolo Molaro and Miguel de Icaza.
The existing JIT compiler has three phases:
* Re-creation of the semantic tree from CIL
byte-codes.
* Instruction selection, with a cost-driven
engine.
* Code generation and register allocation.
It is also hooked into the rest of the runtime to provide
services like marshaling, just-in-time compilation and
invocation of "internal calls".
This engine constructed a collection of trees, which we
referred to as the "forest of trees", this forest is created by
"hydrating" the CIL instruction stream.
The first step was to identify the basic blocks on the method,
and computing the control flow graph (cfg) for it. Once this
information was computed, a stack analysis on each basic block
was performed to create a forest of trees for each one of
them.
So for example, the following statement:
int a, b;
...
b = a + 1;
Which would be represented in CIL as:
ldloc.0
ldc.i4.1
add
stloc.1
After the stack analysis would create the following tree:
(STIND_I4 ADDR_L[EBX|2] (
ADD (LDIND_I4 ADDR_L[ESI|1])
CONST_I4[1]))
This tree contains information from the stack analysis: for
instance, notice that the operations explicitly encode the
data types they are operating on, there is no longer an
ambiguity on the types, because this information has been
inferred.
At this point the JIT would pass the constructed forest of
trees to the architecture-dependent JIT compiler.
The architecture dependent code then performed register
allocation (optionally using linear scan allocation for
variables, based on life analysis).
Once variables had been assigned, a tree pattern matching with
dynamic programming is used (the tree pattern matcher is
custom build for each architecture, using a code
generator: monoburg). The instruction selector used cost
functions to select the best instruction patterns.
The instruction selector is able to produce instructions that
take advantage of the x86 instruction indexing instructions
for example.
One problem though is that the code emitter and the register
allocator did not have any visibility outside the current
tree, which meant that some redundant instructions were
generated. A peephole optimizer with this architecture was
hard to write, given the tree-based representation that is
used.
This JIT was functional, but it did not provide a good
architecture to base future optimizations on. Also the
line between architecture neutral and architecture
specific code and optimizations was hard to draw.
The JIT engine supported two code generation modes to support
the two optimization modes for applications that host multiple
application domains: generate code that will be shared across
application domains, or generate code that will not be shared
across application domains.
* Objectives of the new JIT engine.
We wanted to support a number of features that were missing:
* Ahead-of-time compilation.
The idea is to allow developers to pre-compile their code
to native code to reduce startup time, and the working
set that is used at runtime in the just-in-time compiler.
Although in Mono this has not been a visible problem, we
wanted to pro-actively address this problem.
When an assembly (a Mono/.NET executable) is installed in
the system, it would then be possible to pre-compile the
code, and have the JIT compiler tune the generated code
to the particular CPU on which the software is
installed.
This is done in the Microsoft.NET world with a tool
called ngen.exe
* Have a good platform for doing code optimizations.
The design called for a good architecture that would
enable various levels of optimizations: some
optimizations are better performed on high-level
intermediate representations, some on medium-level and
some at low-level representations.
Also it should be possible to conditionally turn these on
or off. Some optimizations are too expensive to be used
in just-in-time compilation scenarios, but these
expensive optimizations can be turned on for
ahead-of-time compilations or when using profile-guided
optimizations on a subset of the executed methods.
* Reduce the effort required to port the Mono code
generator to new architectures.
For Mono to gain wide adoption in the Unix world, it is
necessary that the JIT engine works in most of today's
commercial hardware platforms.
* Features of the new JIT engine.
The new JIT engine was architected by Dietmar Maurer and Paolo
Molaro, based on the new objectives.
Mono provides a number of services to applications running
with the new JIT compiler:
* Just-in-Time compilation of CLI code into native code.
* Ahead-of-Time compilation of CLI code, to reduce
startup time of applications.
A number of software development features are also available:
* Execution time profiling (--profile)
Generates a report of the times consumed by routines,
as well as the invocation times, as well as the
callers.
* Memory usage profiling (--profile)
Generates a report of the memory usage by a program
that is ran under the Mono JIT.
* Code coverage (--coverage)
* Execution tracing.
People who are interested in developing and improving the Mini
JIT compiler will also find a few useful routines:
* Compilation times
This is used to time the execution time for the JIT
when compiling a routine.
* Control Flow Graph and Dominator Tree drawing.
These are visual aids for the JIT developer: they
render representations of the Control Flow graph, and
for the more advanced optimizations, they draw the
dominator tree graph.
This requires Dot (from the graphwiz package) and Ghostview.
* Code generator regression tests.
The engine contains support for running regression
tests on the virtual machine, which is very helpful to
developers interested in improving the engine.
* Optimization benchmark framework.
The JIT engine will generate graphs that compare
various benchmarks embedded in an assembly, and run the
various tests with different optimization flags.
This requires Perl, GD::Graph.
* Flexibility
This is probably the most important component of the new code
generation engine. The internals are relatively easy to
replace and update, even large passes can be replaced and
implemented differently.
* New code generator
Compiling a method begins with the `mini_method_to_ir' routine
that converts the CIL representation into a medium
intermediate representation.
The mini_method_to_ir routine performs a number of operations:
* Flow analysis and control flow graph computation.
Unlike the previous version, stack analysis and control
flow graphs are computed in a single pass in the
mini_method_to_ir function, this is done for performance
reasons: although the complexity increases, the benefit
for a JIT compiler is that there is more time available
for performing other optimizations.
* Basic block computation.
mini_method_to_ir populates the MonoCompile structure
with an array of basic blocks each of which contains
forest of trees made up of MonoInst structures.
* Inlining
Inlining is no longer restricted to methods containing
one single basic block, instead it is possible to inline
arbitrary complex methods.
The heuristics to choose what to inline are likely going
to be tuned in the future.
* Method to opcode conversion.
Some method call invocations like `call Math.Sin' are
transformed into an opcode: this transforms the call
into a semantically rich node, which is later inline
into an FPU instruction.
Various Array methods invocations are turned into
opcodes as well (The Get, Set and Address methods)
* Tail recursion elimination
Basic blocks ****
The MonoInst structure holds the actual decoded instruction,
with the semantic information from the stack analysis.
MonoInst is interesting because initially it is part of a tree
structure, here is a sample of the same tree with the new JIT
engine:
(stind.i4 regoffset[0xffffffd4(%ebp)]
(add (ldind.i4 regoffset[0xffffffd8(%ebp)])
iconst[1]))
This is a medium-level intermediate representation (MIR).
Some complex opcodes are decomposed at this stage into a
collection of simpler opcodes. Not every complex opcode is
decomposed at this stage, as we need to preserve the semantic
information during various optimization phases.
For example a NEWARR opcode carries the length and the type of
the array that could be used later to avoid type checking or
array bounds check.
There are a number of operations supported on this
representation:
* Branch optimizations.
* Variable liveness.
* Loop optimizations: the dominator trees are
computed, loops are detected, and their nesting
level computed.
* Conversion of the method into static single assignment
form (SSA form).
* Dead code elimination.
* Constant propagation.
* Copy propagation.
* Constant folding.
Once the above optimizations are optionally performed, a
decomposition phase is used to turn some complex opcodes into
internal method calls. In the initial version of the JIT
engine, various operations on longs are emulated instead of
being inlined. Also the newarr invocation is turned into a
call to the runtime.
At this point, after computing variable liveness, it is
possible to use the linear scan algorithm for allocating
variables to registers. The linear scan pass uses the
information that was previously gathered by the loop nesting
and loop structure computation to favor variables in inner
loops. This process updates the basic block `nesting' field
which is later used during liveness analysis.
Stack space is then reserved for the local variables and any
temporary variables generated during the various
optimizations.
** Instruction selection
At this point, the BURS instruction selector is invoked to
transform the tree-based representation into a list of
instructions. This is done using a tree pattern matcher that
is generated for the architecture using the `monoburg' tool.
Monoburg takes as input a file that describes tree patterns,
which are matched against the trees that were produced by the
engine in the previous stages.
The pattern matching might have more than one match for a
particular tree. In this case, the match selected is the one
whose cost is the smallest. A cost can be attached to each
rule, and if no cost is provided, the implicit cost is one.
Smaller costs are selected over higher costs.
The cost function can be used to select particular blocks of
code for a given architecture, or by using a prohibitive high
number to avoid having the rule match.
The various rules that our JIT engine uses transform a tree of
MonoInsts into a list of monoinsts:
+-----------------------------------------------------------+
| Tree List |
| of ===> Instruction selection ===> of |
| MonoInst MonoInst. |
+-----------------------------------------------------------+
During this process various "types" of MonoInst kinds
disappear and turned into lower-level representations. The
JIT compiler just happens to reuse the same structure (this is
done to reduce memory usage and improve memory locality).
The instruction selection rules are split in a number of
files, each one with a particular purpose:
inssel.brg
Contains the generic instruction selection
patterns.
inssel-x86.brg
Contains x86 specific rules.
inssel-ppc.brg
Contains PowerPC specific rules.
inssel-long32.brg
burg file for 64bit instructions on 32bit architectures.
inssel-long.brg
burg file for 64bit architectures.
inssel-float.brg
burg file for floating point instructions
For a given build, a set of those files would be included.
For example, for the build of Mono on the x86, the following
set is used:
inssel.brg inssel-x86.brg inssel-long32.brg inssel-float.brg
** Native method generation
The native method generation has a number of steps:
* Architecture specific register allocation.
The information about loop nesting that was
previously gathered is used here to hint the
register allocator.
* Generating the method prolog/epilog.
* Optionally generate code to introduce tracing facilities.
* Hooking into the debugger.
* Performing any pending fixups.
* Code generation.
*** Code Generation
The actual code generation is contained in the architecture
specific portion of the compiler. The input to the code
generator is each one of the basic blocks with its list of
instructions that were produced in the instruction selection
phase.
During the instruction selection phase, virtual registers are
assigned. Just before the peephole optimization is performed,
physical registers are assigned.
A simple peephole and algebraic optimizer is ran at this
stage.
The peephole optimizer removes some redundant operations at
this point. This is possible because the code generation at
this point has visibility into the basic block that spans the
original trees.
The algebraic optimizer performs some simple algebraic
optimizations that replace expensive operations with cheaper
operations if possible.
The rest of the code generation is fairly simple: a switch
statement is used to generate code for each of the MonoInsts,
in the mono/mini/mini-ARCH.c files, the method is called
"mono_arch_output_basic_block".
We always try to allocate code in sequence, instead of just using
malloc. This way we increase spatial locality which gives a massive
speedup on most architectures.
*** Ahead of Time compilation
Ahead-of-Time compilation is a new feature of our new
compilation engine. The compilation engine is shared by the
Just-in-Time (JIT) compiler and the Ahead-of-Time compiler
(AOT).
The difference is on the set of optimizations that are turned
on for each mode: Just-in-Time compilation should be as fast
as possible, while Ahead-of-Time compilation can take as long
as required, because this is not done at a time critical
time.
With AOT compilation, we can afford to turn all of the
computationally expensive optimizations on.
After the code generation phase is done, the code and any
required fixup information is saved into a file that is
readable by "as" (the native assembler available on all
systems). This assembly file is then passed to the native
assembler, which generates a loadable module.
At execution time, when an assembly is loaded from the disk,
the runtime engine will probe for the existence of a
pre-compiled image. If the pre-compiled image exists, then it
is loaded, and the method invocations are resolved to the code
contained in the loaded module.
The code generated under the AOT scenario is slightly
different than the JIT scenario. It generates code that is
application-domain relative and that can be shared among
multiple thread.
This is the same code generation that is used when the runtime
is instructed to maximize code sharing on a multi-application
domain scenario.
* SSA-based optimizations
SSA form simplifies many optimization because each variable
has exactly one definition site. This means that each
variable is only initialized once.
For example, code like this:
a = 1
..
a = 2
call (a)
Is internally turned into:
a1 = 1
..
a2 = 2
call (a2)
In the presence of branches, like:
if (x)
a = 1
else
a = 2
call (a)
The code is turned into:
if (x)
a1 = 1;
else
a2 = 2;
a3 = phi (a1, a2)
call (a3)
All uses of a variable are "dominated" by its definition
This representation is useful as it simplifies the
implementation of a number of optimizations like conditional
constant propagation, array bounds check removal and dead code
elimination.
* Register allocation.
Global register allocation is performed on the medium
intermediate representation just before instruction selection
is performed on the method. Local register allocation is
later performed at the basic-block level on the
Global register allocation uses the following input:
1) set of register-sized variables that can be allocated to a
register (this is an architecture specific setting, for x86
these registers are the callee saved register ESI, EDI and
EBX).
2) liveness information for the variables
3) (optionally) loop info to favor variables that are used in
inner loops.
During instruction selection phase, symbolic registers are
assigned to temporary values in expressions.
Local register allocation assigns hard registers to the
symbolic registers, and it is performed just before the code
is actually emitted and is performed at the basic block level.
A CPU description file describes the input registers, output
registers, fixed registers and clobbered registers by each
operation.
* BURG Code Generator Generator
monoburg was written by Dietmar Maurer. It is based on the
papers from Christopher W. Fraser, Robert R. Henry and Todd
A. Proebsting: "BURG - Fast Optimal Instruction Selection and
Tree Parsing" and "Engineering a Simple, Efficient Code
Generator Generator".
The original BURG implementation is unable to work on DAGs, instead only
trees are allowed. Our monoburg implementations is able to generate tree
matcher which works on DAGs, and we use this feature in the new
JIT. This simplifies the code because we can directly pass DAGs and
don't need to convert them to trees.
* Adding IL opcodes: an excercise (from a post by Paolo Molaro)
mini.c is the file that read the IL code stream and decides
how any single IL instruction is implemented
(mono_method_to_ir () func), so you always have to add an
entry to the big switch inside the function: there are plenty
of examples in that file.
An IL opcode can be implemented in a number of ways, depending
on what it does and how it needs to do it.
Some opcodes are implemented using a helper function: one of
the simpler examples is the CEE_STELEM_REF implementation.
In this case the opcode implementation is written in a C
function. You will need to register the function with the jit
before you can use it (mono_register_jit_call) and you need to
emit the call to the helper using the mono_emit_jit_icall()
function.
This is the simpler way to add a new opcode and it doesn't
require any arch-specific change (though it's limited to what
you can do in C code and the performance may be limited by the
function call).
Other opcodes can be implemented with one or more of the already
implemented low-level instructions.
An example is the OP_STRLEN opcode which implements
String.Length using a simple load from memory. In this case
you need to add a rule to the appropriate burg file,
describing what are the arguments of the opcode and what is,
if any, it's 'return' value.
The OP_STRLEN case is:
reg: OP_STRLEN (reg) {
MONO_EMIT_LOAD_MEMBASE_OP (s, tree, OP_LOADI4_MEMBASE, state->reg1,
state->left->reg1, G_STRUCT_OFFSET (MonoString, length));
}
The above means: the OP_STRLEN takes a register as an argument
and returns its value in a register. And the implementation
of this is included in the braces.
The opcode returns a value in an integer register
(state->reg1) by performing a int32 load of the length field
of the MonoString represented by the input register
(state->left->reg1): before the burg rules are applied, the
internal representation is based on trees, so you get the
left/right pointers (state->left and state->right
respectively, the result is stored in state->reg1).
This instruction implementation doesn't require arch-specific
changes (it is using the MONO_EMIT_LOAD_MEMBASE_OP which is
available on all platforms), and usually the produced code is
fast.
Next we have opcodes that must be implemented with new low-level
architecture specific instructions (either because of performance
considerations or because the functionality can't get implemented in
other ways).
You also need a burg rule in this case, too. For example,
consider the OP_CHECK_THIS opcode (used to raise an exception
if the this pointer is null). The burg rule simply reads:
stmt: OP_CHECK_THIS (reg) {
mono_bblock_add_inst (s->cbb, tree);
}
Note that this opcode does not return a value (hence the
"stmt") and it takes a register as input.
mono_bblock_add_inst (s->cbb, tree) just adds the instruction
(the tree variable) to the current basic block (s->cbb). In
mini this is the place where the internal representation
switches from the tree format to the low-level format (the
list of simple instructions).
In this case the actual opcode implementation is delegated to
the arch-specific code. A low-level opcode needs an entry in
the machine description (the *.md files in mini/). This entry
describes what kind of registers are used if any by the
instruction, as well as other details such as constraints or
other hints to the low-level engine which are architecture
specific.
cpu-pentium.md, for example has the following entry:
checkthis: src1:b len:3
This means the instruction uses an integer register as a base
pointer (basically a load or store is done on it) and it takes
3 bytes of native code to implement it.
Now you just need to provide the low-level implementation for
the opcode in one of the mini-$arch.c files, in the
mono_arch_output_basic_block() function. There is a big switch
here too. The x86 implementation is:
case OP_CHECK_THIS:
/* ensure ins->sreg1 is not NULL */
x86_alu_membase_imm (code, X86_CMP, ins->sreg1, 0, 0);
break;
If the $arch-codegen.h header file doesn't have the code to
emit the low-level native code, you'll need to write that as
well.
Complex opcodes with register constraints may require other
changes to the local register allocator, but usually they are
not needed.
* Future
Profile-based optimization is something that we are very
interested in supporting. There are two possible usage
scenarios:
* Based on the profile information gathered during
the execution of a program, hot methods can be compiled
with the highest level of optimizations, while bootstrap
code and cold methods can be compiled with the least set
of optimizations and placed in a discardable list.
* Code reordering: this profile-based optimization would
only make sense for pre-compiled code. The profile
information is used to re-order the assembly code on disk
so that the code is placed on the disk in a way that
increments locality.
This is the same principle under which SGI's cord program
works.
The nature of the CIL allows the above optimizations to be
easy to implement and deploy. Since we live and define our
universe for these things, there are no interactions with
system tools required, nor upgrades on the underlying
infrastructure required.
Instruction scheduling is important for certain kinds of
processors, and some of the framework exists today in our
register allocator and the instruction selector to cope with
this, but has not been finished. The instruction selection
would happen at the same time as local register allocation. <
-424
View File
@@ -1,424 +0,0 @@
Mono JIT porting guide.
Paolo Molaro ([email protected])
* Introduction
This documents describes the process of porting the mono JIT
to a new CPU architecture. The new mono JIT has been designed
to make porting easier though at the same time enable the port
to take full advantage from the new architecture features and
instructions. Knowledge of the mini architecture (described in
the mini-doc.txt file) is a requirement for understanding this
guide, as well as an earlier document about porting the mono
interpreter (available on the web site).
There are six main areas that a port needs to implement to
have a fully-functional JIT for a given architecture:
1) instruction selection
2) native code emission
3) call conventions and register allocation
4) method trampolines
5) exception handling
6) minor helper methods
To take advantage of some not-so-common processor features
(for example conditional execution of instructions as may be
found on ARM or ia64), it may be needed to develop an
high-level optimization, but doing so is not a requirement for
getting the JIT to work.
We'll see in more details each of the steps required, note,
though, that a new port may just as well start from a
cut&paste of an existing port to a similar architecture (for
example from x86 to amd64, or from powerpc to sparc).
The architecture specific code is split from the rest of the
JIT, for example the x86 specific code and data is all
included in the following files in the distribution:
mini-x86.h mini-x86.c
inssel-x86.brg
cpu-pentium.md
tramp-x86.c
exceptions-x86.c
I suggest a similar split for other architectures as well.
Note that this document is still incomplete: some sections are
only sketched and some are missing, but the important info to
get a port going is already described.
* Architecture-specific instructions and instruction selection.
The JIT already provides a set of instructions that can be
easily mapped to a great variety of different processor
instructions. Sometimes it may be necessary or advisable to
add a new instruction that represent more closely an
instruction in the architecture. Note that a mini instruction
can be used to represent also a short sequence of CPU
low-level instructions, but note that each instruction
represents the minimum amount of code the instruction
scheduler will handle (i.e., the scheduler won't schedule the
instructions that compose the low-level sequence as individual
instructions, but just the whole sequence, as an indivisible
block).
New instructions are created by adding a line in the
mini-ops.h file, assigning an opcode and a name. To specify
the input and output for the instruction, there are two
different places, depending on the context in which the
instruction gets used.
If the instruction is used in the tree representation, the
input and output types are defined by the BURG rules in the
*.brg files (the usual non-terminals are 'reg' to represent a
normal register, 'lreg' to represent a register or two that
hold a 64 bit value, freg for a floating point register).
If an instruction is used as a low-level CPU instruction, the
info is specified in a machine description file. The
description file is processed by the genmdesc program to
provide a data structure that can be easily used from C code
to query the needed info about the instruction.
As an example, let's consider the add instruction for both x86
and ppc:
x86 version:
add: dest:i src1:i src2:i len:2 clob:1
ppc version:
add: dest:i src1:i src2:i len:4
Note that the instruction takes two input integer registers on
both CPU, but on x86 the first source register is clobbered
(clob:1) and the length in bytes of the instruction differs.
Note that integer adds and floating point adds use different
opcodes, unlike the IL language (64 bit add is done with two
instructions on 32 bit architectures, using a add that sets
the carry and an add with carry).
A specific CPU port may assign any meaning to the clob field
for an instruction since the value will be processed in an
arch-specific file anyway.
See the top of the existing cpu-pentium.md file for more info
on other fields: the info may or may not be applicable to a
different CPU, in this latter case the info can be ignored.
The code in mini.c together with the BURG rules in inssel.brg,
inssel-float.brg and inssel-long32.brg provides general
purpose mappings from the tree representation to a set of
instructions that should be easily implemented in any
architecture. To allow for additional arch-specific
functionality, an arch-specific BURG file can be used: in this
file arch-specific instructions can be selected that provide
better performance than the general instructions or that
provide functionality that is needed by the JIT but that
cannot be expressed in a general enough way.
As an example, x86 has the special instruction "push" to make
it easier to implement the default call convention (passing
arguments on the stack): almost all the other architectures
don't have such an instruction (and don't need it anyway), so
we added a special rule in the inssel-x86.brg file for it.
So, one of the first things needed in a port is to write a
cpu-$(arch).md machine description file and fill it with the
needed info. As a start, only a few instructions can be
specified, like the ones required to do simple integer
operations. The default rules of the instruction selector will
emit the common instructions and so we're ready to go for the
next step in porting the JIT.
*) Native code emission
Since the first step in porting mono to a new CPU is to port
the interpreter, there should be already a file that allows
the emission of binary native code in a buffer for the
architecture. This file should be placed in the
mono/arch/$(arch)/
directory.
The bulk of the code emission happens in the mini-$(arch).c
file, in a function called mono_arch_output_basic_block
(). This function takes a basic block, walks the list of
instructions in the block and emits the binary code for each.
Optionally a peephole optimization pass is done on the basic
block, but this can be left for later, when the port actually
works.
This function is very simple, there is just a big switch on
the instruction opcode and in the corresponding case the
functions or macros to emit the binary native code are
used. Note that in this function the lengths of the
instructions are used to determine if the buffer for the code
needs enlarging.
To complete the code emission for a method, a few other
functions need implementing as well:
mono_arch_emit_prolog ()
mono_arch_emit_epilog ()
mono_arch_patch_code ()
mono_arch_emit_prolog () will emit the code to setup the stack
frame for a method, optionally call the callbacks used in
profiling and tracing, and move the arguments to their home
location (in a caller-save register if the variable was
allocated to one, or in a stack location if the argument was
passed in a volatile register and wasn't allocated a
non-volatile one). caller-save registers used by the function
are saved in the prolog as well.
mono_arch_emit_epilog () will emit the code needed to return
from the function, optionally calling the profiling or tracing
callbacks. At this point the basic blocks or the code that was
moved out of the normal flow for the function can be emitted
as well (this is usually done to provide better info for the
static branch predictor). In the epilog, caller-save
registers are restored if they were used.
Note that, to help exception handling and stack unwinding,
when there is a transition from managed to unmanaged code,
some special processing needs to be done (basically, saving
all the registers and setting up the links in the Last Managed
Frame structure).
When the epilog has been emitted, the upper level code
arranges for the buffer of memory that contains the native
code to be copied in an area of executable memory and at this
point, instructions that use relative addressing need to be
patched to have the right offsets: this work is done by
mono_arch_patch_code ().
* Call conventions and register allocation
To account for the differences in the call conventions, a few functions need to
be implemented.
mono_arch_allocate_vars () assigns to both arguments and local
variables the offset relative to the frame register where they
are stored, dead variables are simply discarded. The total
amount of stack needed is calculated.
mono_arch_call_opcode () is the function that more closely
deals with the call convention on a given system. For each
argument to a function call, an instruction is created that
actually puts the argument where needed, be it the stack or a
specific register. This function can also re-arrange th order
of evaluation when multiple arguments are involved if needed
(like, on x86 arguments are pushed on the stack in reverse
order). The function needs to carefully take into accounts
platform specific issues, like how structures are returned as
well as the differences in size and/or alignment of managed
and corresponding unmanaged structures.
The other chunk of code that needs to deal with the call
convention and other specifics of a CPU, is the local register
allocator, implemented in a function named
mono_arch_local_regalloc (). The local allocator deals with a
basic block at a time and basically just allocates registers
for temporary values during expression evaluation, spilling
and unspilling as necessary.
The local allocator needs to take into account clobbering
information, both during simple instructions and during
function calls and it needs to deal with other
architecture-specific weirdnesses, like instructions that take
inputs only in specific registers or output only is some.
Some effort will be put later in moving most of the local
register allocator to a common file so that the code can be
shared more for similar, risc-like CPUs. The register
allocator does a first pass on the instructions in a block,
collecting liveness information and in a backward pass on the
same list performs the actual register allocation, inserting
the instructions needed to spill values, if necessary.
When this part of code is implemented, some testing can be
done with the generated code for the new architecture. Most
helpful is the use of the --regression command line switch to
run the regression tests (basic.cs, for example).
Note that the JIT will try to initialize the runtime, but it
may not be able yet to compile and execute complex code:
commenting most of the code in the mini_init() function in
mini.c is needed to let the JIT just compile the regression
tests. Also, using multiple -v switches on the command line
makes the JIT dump an increasing amount of information during
compilation.
* Method trampolines
To get better startup performance, the JIT actually compiles a
method only when needed. To achieve this, when a call to a
method is compiled, we actually emit a call to a magic
trampoline. The magic trampoline is a function written in
assembly that invokes the compiler to compile the given method
and jumps to the newly compiled code, ensuring the arguments
it received are passed correctly to the actual method.
Before jumping to the new code, though, the magic trampoline
takes care of patching the call site so that next time the
call will go directly to the method instead of the
trampoline. How does this all work?
mono_arch_create_jit_trampoline () creates a small function
that just preserves the arguments passed to it and adds an
additional argument (the method to compile) before calling the
generic trampoline. This small function is called the specific
trampoline, because it is method-specific (the method to
compile is hard-code in the instruction stream).
The generic trampoline saves all the arguments that could get
clobbered and calls a C function that will do two things:
*) actually call the JIT to compile the method
*) identify the calling code so that it can be patched to call directly
the actual method
If the 'this' argument to a method is a boxed valuetype that
is passed to a method that expects just a pointer to the data,
an additional unboxing trampoline will need to be inserted as
well.
* Exception handling
Exception handling is likely the most difficult part of the
port, as it needs to deal with unwinding (both managed and
unmanaged code) and calling catch and filter blocks. It also
needs to deal with signals, because mono takes advantage of
the MMU in the CPU and of the operation system to handle
dereferences of the NULL pointer. Some of the function needed
to implement the mechanisms are:
mono_arch_get_throw_exception () returns a function that takes
an exception object and invokes an arch-specific function that
will enter the exception processing. To do so, all the
relevant registers need to be saved and passed on.
mono_arch_handle_exception () this function takes the
exception thrown and a context that describes the state of the
CPU at the time the exception was thrown. The function needs
to implement the exception handling mechanism, so it makes a
search for an handler for the exception and if none is found,
it follows the unhandled exception path (that can print a
trace and exit or just abort the current thread). The
difficulty here is to unwind the stack correctly, by restoring
the register state at each call site in the call chain,
calling finally, filters and handler blocks while doing so.
As part of exception handling a couple of internal calls need
to be implemented as well.
ves_icall_get_frame_info () returns info about a specific
frame.
mono_jit_walk_stack () walks the stack and calls a callback with info for
each frame found.
ves_icall_get_trace () return an array of StackFrame objects.
** Code generation for filter/finally handlers
Filter and finally handlers are called from 2 different locations:
1.) from within the method containing the exception clauses
2.) from the stack unwinding code
To make this possible we implement them like subroutines,
ending with a "return" statement. The subroutine does not save
the base pointer, because we need access to the local
variables of the enclosing method. Its is possible that
instructions inside those handlers modify the stack pointer,
thus we save the stack pointer at the start of the handler,
and restore it at the end. We have to use a "call" instruction
to execute such finally handlers.
The MIR code for filter and finally handlers looks like:
OP_START_HANDLER
...
OP_END_FINALLY | OP_ENDFILTER(reg)
OP_START_HANDLER: should save the stack pointer somewhere
OP_END_FINALLY: restores the stack pointers and returns.
OP_ENDFILTER (reg): restores the stack pointers and returns the value in "reg".
** Calling finally/filter handlers
There is a special opcode to call those handler, its called
OP_CALL_HANDLER. It simple emits a call instruction.
Its a bit more complex to call handler from outside (in the
stack unwinding code), because we have to restore the whole
context of the method first. After that we simply emit a call
instruction to invoke the handler. Its usually possible to use
the same code to call filter and finally handlers (see
arch_get_call_filter).
** Calling catch handlers
Catch handlers are always called from the stack unwinding
code. Unlike finally clauses or filters, catch handler never
return. Instead we simply restore the whole context, and
restart execution at the catch handler.
** Passing Exception objects to catch handlers and filters.
We use a local variable to store exception objects. The stack
unwinding code must store the exception object into this
variable before calling catch handler or filter.
* Minor helper methods
A few minor helper methods are referenced from the arch-independent code.
Some of them are:
*) mono_arch_cpu_optimizations ()
This function returns a mask of optimizations that
should be enabled for the current CPU and a mask of
optimizations that should be excluded, instead.
*) mono_arch_regname ()
Returns the name for a numeric register.
*) mono_arch_get_allocatable_int_vars ()
Returns a list of variables that can be allocated to
the integer registers in the current architecture.
*) mono_arch_get_global_int_regs ()
Returns a list of caller-save registers that can be
used to allocate variables in the current method.
*) mono_arch_instrument_mem_needs ()
*) mono_arch_instrument_prolog ()
*) mono_arch_instrument_epilog ()
Functions needed to implement the profiling interface.
* Writing regression tests
Regression tests for the JIT should be written for any bug
found in the JIT in one of the *.cs files in the mini
directory. Eventually all the operations of the JIT should be
tested (including the ones that get selected only when some
specific optimization is enabled).
* Platform specific optimizations
An example of a platform-specific optimization is the peephole
optimization: we look at a small window of code at a time and
we replace one or more instructions with others that perform
better for the given architecture or CPU.
-98
View File
@@ -1,98 +0,0 @@
=pod
=head1 Internal design document for the mono_handle_d
This document is designed to hold the design of the mono_handle_d and
not as an api reference.
=head2 Primary goal and purpose
The mono_handle_d is a process which takes care of the (de)allocation
of scratch shared memory and handles (of files, threads, mutexes,
sockets etc. see L<WapiHandleType>) and refcounts of the
filehandles. It is designed to be run by a user and to be fast, thus
minimal error checking on input is done and will most likely crash if
given a faulty package. No effort has been, or should be, made to have
the daemon talking to machine of different endianness/size of int.
=head2 How to start the daemon
To start the daemon you either run the mono_handle_d executable or try
to attach to the shared memory segment via L<_wapi_shm_attach> which
will start a daemon if one does not exist.
=head1 Internal details
The daemon works by opening a socket and listening to clients. These
clients send packages over the socket complying to L<struct
WapiHandleRequest>.
=head2 Possible requests
=over
=item WapiHandleRequest_New
Find a handle in the shared memory segment that is free and allocate
it to the specified type. To destroy use
L</WapiHandleRequest_Close>. A L<WapiHandleResponse> with
.type=WapiHandleResponseType_New will be sent back with .u.new.handle
set to the handle that was allocated. .u.new.type is the type that was
requested.
=item WapiHandleRequestType_Open
Increase the ref count of an already created handle. A
L<WapiHandleResponse> with .type=WapiHandleResponseType_Open will be sent
back with .u.new.handle set to the handle, .u.new.type is set to the
type of handle this is.
=item WapiHandleRequestType_Close
Decrease the ref count of an already created handle. A
L<WapiHandleResponse> with .type=WapiHandleResponseType_Close will be
sent back with .u.close.destroy set to TRUE if ref count for this
client reached 0.
=item WapiHandleRequestType_Scratch
Allocate a shared memory area of size .u.scratch.length in bytes. A
L<WapiHandleResponse> with .type=WapiHandleResponseType_Scratch will be
sent back with .u.scratch.idx set to the index into the shared
memory's scratch area where to memory begins. (works just like
malloc(3))
=item WapiHandleRequestType_Scratch
Deallocate a shared memory area, this must have been allocated before
deallocating. A L<WapiHandleResponse> with
.type=WapiHandleResponseType_ScratchFree will be sent back (works just
like free(3))
=back
=head1 Why a daemon
From an email:
Dennis: I just have one question about the daemon... Why does it
exist? Isn't it better performancewise to just protect the shared area
with a mutex when allocation a new handle/shared mem segment or
changing refcnt? It will however be a less resilient to clients that
crash (the deamon cleans up ref'd handles if socket closes)
Dick: It's precisely because with a mutex the shared memory segment
can be left in a locked state. Also, it's not so easy to clean up
shared memory without it (you can't just mark it deleted when creating
it, because you can't attach any more readers to the same segment
after that). I did some minimal performance testing, and I don't
think the daemon is particularly slow.
=head1 Authors
Documentaion: Dennis Haney
Implementation: Dick Porter
=cut
-63
View File
@@ -1,63 +0,0 @@
Author: Dietmar Maurer ([email protected])
(C) 2001 Ximian, Inc.
Object and VTable layout
========================
The first pointer inside an Object points to a MonoVtable structure. Objects
also contains a MonoThreadsSync structure which is used by the mono Thread
implementation.
typedef struct {
MonoVTable *vtable;
MonoThreadsSync synchronisation;
/* object specific data goes here */
} MonoObject;
The MonoVtable contains the vtable, interface offsets and a pointer to static
class data. Each object/vtable belongs to exactly one AppDomain.
typedef struct {
MonoClass *klass;
MonoDomain *domain;
gpointer *interface_offsets;
/* a pointer to static data */
gpointer data;
/* the variable sized vtable is included at the end */
gpointer vtable [0];
} MonoVTable;
The MonoClass contains domain independent Class infos.
typedef struct {
/* various class infos */
MonoClass *parent;
const char *name;
const char *name_space;
...
} MonoClass;
Calling virtual functions:
==========================
Each MonoMethod (if virtual) has an associated slot, which is an index into the
VTable. So we can use the following code to compute the address of a virtual
function:
method_addr = object->vtable->vtable [method->slot];
Calling interface methods:
==========================
Each interface class is associated with an unique ID. The following code
computes the address of an interface function:
method_addr = *(object->vtable->interface_offsets [interface_id] + method->slot*4);
-113
View File
@@ -1,113 +0,0 @@
* How to handle complex IL opcodes in an arch-independent way
Many IL opcodes are very simple: add, ldind etc.
Such opcodes can be implemented with a single cpu instruction
in most architectures (on some, a group of IL instructions
can be converted to a single cpu op).
There are many IL opcodes, though, that are more complex, but
can be expressed as a series of trees or a single tree of
simple operations. Such simple operations are architecture-independent.
It makes sense to decompose such complex IL instructions in their
simpler equivalent so that we gain in several ways:
*) porting effort is easier, because only the simple instructions
need to be implemented in arch-specific code
*) we could apply BURG rules to the trees and do pattern matching
on them to optimize the expressions according to the host cpu
The issue is: where do we do such conversion from coarse opcodes to
simple expressions?
* Doing the conversion in method_to_ir ()
Some of these conversions can certainly be done in method_to_ir (),
but it's not always easy to decide which are better done there and
which in a different pass.
For example, let's take ldlen: in the mono implementation, ldlen
can be simply implemented with a load from a fixed position in the
array object:
len = [reg + maxlen_offset]
However, ldlen carries also semantics information: the result is the
length of the array, and since in the CLR arrays are of fixed size,
this information can be useful to later do bounds check removal.
If we convert this opcode in method_to_ir () we lost some useful
information for further optimizations.
In some other ways, decomposing an opcode in method_to_ir() may
allow for better optimizations later on (need to come up with an
example here ...).
* Doing the conversion in inssel.brg
Some conversion may be done inside the burg rules: this has the
disadvantage that the instruction selector is not run again on
the resulting expression tree and we could miss some optimization
(this is what effectively happens with the coarse opcodes in the old
jit). This may also interfere with an efficient local register allocator.
It may be possible to add an extension in monoburg that allows a rule
such as:
recheck: LDLEN (reg) {
create an expression tree representing LDLEN
and return it
}
When the monoburg label process gets back a recheck, it will run
the labeling again on the resulting expression tree.
If this is possible at all (and in an efficient way) is a
question for dietmar:-)
It should be noted, though, that this may not always work, since
some complex IL opcodes may require a series of expression trees
and handling such cases in monoburg could become quite hairy.
For example, think of opcode that need to do multiple actions on the
same object: this basically means a DUP...
On the other end, if a complex opcode needs a DUP, monoburg doesn't
actually need to create trees if it emits the instructions in
the correct sequence and maintains the right values in the registers
(usually the values that need a DUP are not changed...). How
this integrates with the current register allocator is not clear, since
that assigns registers based on the rule, but the instructions emitted
by the rules may be different (this already happens with the current JIT
where a MULT is replaced with lea etc...).
* Doing it in a separate pass.
Doing the conversion in a separate pass over the instructions
is another alternative. This can be done right after method_to_ir ()
or after the SSA pass (since the IR after the SSA pass should look
almost like the IR we get back from method_to_ir ()).
This has the following advantages:
*) monoburg will handle only the simple opcodes (makes porting easier)
*) the instruction selection will be run on all the additional trees
*) it's easier to support coarse opcodes that produce multiple expression
trees (and apply the monoburg selector on all of them)
*) the SSA optimizer will see the original opcodes and will be able to use
the semantic info associated with them
The disadvantage is that this is a separate pass on the code and
it takes time (how much has not been measured yet, though).
With this approach, we may also be able to have C implementations
of some of the opcodes: this pass would insert a function call to
the C implementation (for example in the cases when first porting
to a new arch and implemenating some stuff may be too hard in asm).
* Extended basic blocks
IL code needs a lot of checks, bounds checks, overflow checks,
type checks and so on. This potentially increases by a lot
the number of basic blocks in a control flow graph. However,
all such blocks end up with a throw opcode that gives control to the
exception handling mechanism.
After method_to_ir () a MonoBasicBlock can be considered a sort
of extended basic block where the additional exits don't point
to basic blocks in the same procedure (at least when the method
doesn't have exception tables).
We need to make sure the passes following method_to_ir () can cope
with such kinds of extended basic blocks (especially the passes
that we need to apply to all the methods: as a start, we could
skip SSA optimizations for methods with exception clauses...)
-16
View File
@@ -1,16 +0,0 @@
<h1>Mono 1.0 Release Notes</h1>
<h2>What does Mono Include</h2>
<h2>Missing functionality</h2>
<p>COM support.
<p>EnterpriseServices are non-existant.
<p>Windows.Forms is only available as a preview, it is not
completed nor stable.
<h3>Assembly: System.Drawing</h3>
<p>System.Drawing.Printing is not supported.
-26
View File
@@ -1,26 +0,0 @@
Runtime support for Remoting
============================
The runtime supports a special objects called "TransparentProxy". You can
create objects of this type by calling GetTransparentProxy() on a "RealProxy"
object.
LDFLD/STFLD for transparent proxies
===================================
Access to fields must be redirected to the remote object. System.Object has
some special methods for that:
void FieldGetter (string typeName, string fieldName, ref object val);
void FieldSetter (string typeName, string fieldName, object val);
This methods are never called on actual object. The are only used to pack
LDFLD/STFLD operations into method call messages, which are then passed to the
RealProxy::Invoke() method.
There are two helper methods which can be used by the JIT and the interpreter
to convert LDFLD/STFLD operations into messages and then call
RealProxy::Invoke(): mono_store_remote_field() and mono_load_remote_field().
-33
View File
@@ -1,33 +0,0 @@
Size and alignment requirements of stack values
===============================================
P ... System.IntPtr
I1 ... System.Int8
I2 ... System.Int16
I4 ... System.Int32
I8 ... System.Int64
F ... System.Single
D ... System.Double
LD ... native long double
-----------------------------------------------------------
ARCH | P | I1 | I2 | I4 | I8 | F | D | LD |
-----------------------------------------------------------
X86 | 4/4 | 4/4 | 4/4 | 4/4 | 8/4 | 4/4 | 8/4 |12/4 |
-----------------------------------------------------------
X86/W32 | 4/4 | 4/4 | 4/4 | 4/4 | 8/4 | 4/4 | 8/4 |12/4 |
-----------------------------------------------------------
ARM | 4/4 | 4/4 | 4/4 | 4/4 | 8/4 | 4/4 | 8/4 | 8/4 |
-----------------------------------------------------------
M68K | 4/4 | 4/4 | 4/4 | 4/4 | 8/4 | 4/4 | 8/4 |12/4 |
-----------------------------------------------------------
ALPHA | 8/8 | 8/8 | 8/8 | 8/8 | 8/8 | 8/8 | 8/8 | 8/8 |
-----------------------------------------------------------
SPARC | 4/4 | 4/4 | 4/4 | 4/4 | 8/8 | 4/4 | 8/8 |16/8 |
-----------------------------------------------------------
SPARC64 | 8/8 | 8/8 | 8/8 | 8/8 | 8/8 | 8/8 | 8/8 |16/16|
-----------------------------------------------------------
MIPS | 4/4 | 4/4 | 4/4 | 4/4 | ?/? | 4/4 | 8/8 | 8/8 |
-----------------------------------------------------------
| | | | | | | | |
-----------------------------------------------------------
-156
View File
@@ -1,156 +0,0 @@
Threading in Mono
=================
0. Terminology
--------------
"Main thread" - The initial OS-native thread that the
application started with.
"Helper thread" - A native thread created internally
by the runtime, such as the finalizer thread, or an
asynchronous delegate invocation thread. These
threads can run managed code.
"Primary CLR thread" - The native thread that called
the Main() method when executing an assembly.
"Secondary CLR thread" - A native thread created by a
program that instantiates a System.Threading.Thread object
and calls its Start() method.
1. Thread exit behaviour in the standalone mono runtime
-------------------------------------------------------
The correct behaviour of the runtime should be:
a) If Main() returns, the runtime should wait for all
foreground secondary CLR threads to finish. The
wording in the class documentation states: "Once all
foreground threads belonging to a process have
terminated, the common language runtime ends the
process by invoking Abort on any background threads
that are still alive." Testing seems to indicate that
the background thread can't cancel the Abort by
catching the ThreadAbortException and calling
ResetAbort here. Indeed, not even the finally block
seems to be executed.
b) if any of the primary CLR thread, a secondary CLR
thread or a helper thread calls
System.Environment.Exit(), the application should
terminate immediately without waiting for foreground
primary or secondary CLR threads to finish.
c) if the primary CLR thread throws an uncaught
exception, the application should terminate
immediately without waiting for secondary CLR threads
to finish. This might be implemented internally by
pretending that all the running secondary CLR threads
are background threads.
d) if a secondary CLR thread throws an uncaught
exception that thread should terminate and all other
threads should continue to execute.
e) if a helper thread throws an uncaught exception and
that thread happens to be the GC finalizer thread,
testing seems to indicate that the exception stack
trace is displayed as normal, and the exception is
then ignored (as though there is a try {} catch{}
around all finalizers that just prints the stack
trace.) Calling Abort() on the GC finalizer thread
also does not cause it to exit: it behaves as though
the ThreadAbortException is caught and ResetAbort is
called. Asynchronous delegate helper threads should
behave as secondary CLR threads, but uncaught
exceptions should be rethrown on the thread that calls
EndInvoke().
The difficulties happen with cases b) and c):
The current implementation of
System.Environment.Exit() calls exit(2) directly,
which is rather unfriendly: it prevents any runtime
cleanup, statistics gathering, etc. and is pretty
obnoxious to embedded code.
The current exception handling code calls ExitThread()
(emulated with pthread_exit() in the io-layer) if an
exception is not caught.
When called from the main thread, both POSIX
pthread_exit() and w32 ExitThread() block if there are
other threads still running (in the w32 case, if there
are other foreground threads still running; threads
can set as background.) If the main thread is also
the primary CLR thread, then the application will
block until all other threads (including helper
threads) terminate. Some helper threads will not
terminate until specifically told to by the runtime:
for example, the GC finalizer thread needs to run
until all of the primary and secondary CLR threads
have finished.
Also, if the main thread is also the primary CLR
thread, the runtime loses the opportunity to do any
cleaning up. Adding a special case to call exit(2)
instead of ExitThread() in the primary CLR thread
suffers from the same problems as
System.Environment.Exit() calling exit(2).
The simple solution is to run the primary CLR thread
in a new native thread, leaving the main thread free
for housekeeping duties. There still needs to be some
special handling for the case where the primary CLR
thread fails to catch an exception: the secondary CLR
threads then need to be terminated.
When the primary and secondary CLR threads have all
terminated, the helper threads can be killed off and
the runtime can clean itself up and exit.
2. Thread initialisation
------------------------
Threads have to undergo some initialisation before
managed code can be executed. A
System.Threading.Thread object must be created, and
the thread details need to be stored so that the
threads can be managed later. The JIT needs to record
the last managed frame stack pointer in a TLS slot,
and the current Thread object is also recorded.
New threads created by managed calls to
System.Threading.Thread methods will have all needed
initialisation performed. Threads created by the
runtime with calls to mono_thread_create() will too.
Existing threads can be passed to the runtime; these
must call mono_thread_attach() before any CLR code can
be executed on that thread.
3. Constraints on embedding the Mono runtime
--------------------------------------------
The discussion above concerning application behaviour
in the event of threads terminating, whether by
returning from the start function, throwing uncaught
exceptions or by calling System.Environment.Exit(),
only really applies to the standalone Mono runtime.
An embedding application should specify what behaviour
is required when, for example,
System.Environment.Exit() is called. The application
is also responsible for its own thread management, and
it should be prepared for any of the primary CLR
thread or secondary CLR threads to terminate at any
time. The application should also take into account
that the runtime will create helper threads as needed,
as this may cause pthread_exit() or ExitThread() to
block indefinitely, as noted above.
-135
View File
@@ -1,135 +0,0 @@
Author: Dietmar Maurer ([email protected])
(C) 2001 Ximian, Inc.
More about PInvoke and Internal calls
=====================================
1.) What is PInvoke
PInvoke stands for Platform Invoke. It is possible to call functions contained
in native shared libraries, for example you can declare:
[DllImport("cygwin1.dll", EntryPoint="puts"]
public static extern int puts (string name);
If you then call "puts(...)" it invokes the native "puts" functions in
"cygwin1.dll". It is also possible to specify several marshalling attributes
for the arguments, for example you can specify that they puts() function expect
ts the string in Ansi encoding by setting the CharSet attribute field:
[DllImport("cygwin1.dll", EntryPoint="puts", CharSet=CharSet.Ansi)]
public static extern int puts (string name);
2.) What are internal calls
Some class library functions are implemented in C, because it is either not
possible to implement them in C# or because of performance gains. Internal
functions are contained in the mono executable itself. Here is an example form
our array implementation:
[MethodImplAttribute(MethodImplOptions.InternalCall)]
public extern int GetRank ();
If you call this GetRank() function it invokes
ves_icall_System_Array_GetRank() inside the mono runtime.
If you write your own runtime environment you can add internal calls with
mono_add_internal_call().
2.) Runtime considerations
Invoking native (unmanaged) code has several implications:
- We need to handle exceptions inside unmanaged code. The JIT simply saves some
informations at each transition from managed to unmanaged code (in a linked
list), called Last Managed Frame (LMF). If an exception occurs the runtime
first looks if the exception was inside managed code. If not there must be a
LMF entry which contains all necessary information to unwind the stack.
Creation of those LMF structure clearly involves some overhead, so calling
into unmanaged code is not as cheap as it looks like at first glance. Maybe
we can introduce a special attribute to avoid the creation of LMF on internal
call methods that cant raise exceptions.
- PInvoke has the possibility to convert argument types. For example Strings
are marshalled as Char*. So each String argument is translated into a
char*. The encoding is specified in the CharSet of the DllImport attribute.
3.) When/how does the runtime call unmanaged PInvoke code
- LDFTN, CALLI, Delegate::Invoke, Delegate::BeginInvoke: We must generate
wrapper code when we load the function with LDFTN, so that all arguments are
marshalled in the right format. We also need to save/restore the LMF.
- MethodBase::Invoke (runtime invoke): We need to marshal all arguments in
they right format and save/restore the LMF
- CALL: We need to marshal all arguments in they right format and save/restore
the LMF
The easiest way to implement this is to always create a wrapper function for
PInvoke calls, which takes care of argument marshalling and LMF save/restore.
4.) When/how does the runtime call unmanaged internal calls
We don't need to convert any arguments, so we need only take care of the LMF
structure.
- LDFTN, CALLI, Delegate::Invoke, Delegate::BeginInvoke: We must generate
wrapper code when we load the function with LDFTN which saves/restores the
LMF.
- MethodBase::Invoke (runtime invoke): We need to save/restore the LMF.
- CALL: We need to save/restore the LMF.
- CALLVIRT (through the vtable): We must generate wrapper code to save/restore
the LMF.
Please notice that we can call internal function with CALLVIRT, i.e. we can
call those function through a VTable. But we cant know in advance if a vtable
slot contains an internal call or managed code. So again it is best to generate
a wrapper functions for internal calls in order to save/restore the LMF.
Unfortunately we need to push all arguments 2 times, because we have to save
the LMF, and the LMF is currently allocated on the stack. So the stack looks
like:
--------------------
| method arguments |
--------------------
| LMF |
--------------------
| copied arguments |
--------------------
AFAIK this is the way ORP works. Another way is to allocate the LMF not on the
stack, but then we have additional overhead to allocate/free LMF structures
(and another call to arch_get_lmf_addr).
Maybe it is possible to avoid this addiotional copy for internal calls by
including the LMF in the C function signature. Lets say we hav a puts()
function which is a internal call:
ves_icall_puts (MonoString *string);
If we simply modify that to include the LMF we can avoid to copy all arguments:
ves_icall_puts (MonoLMF lmf, MonoString *string);
But this depends somehow on the calling conventions, and I don't know if that
works on all plattforms?
5.) What is stored in the LMF
- all caller saved registers (since we can trust unmanaged code)
- the instruction pointer of the last managed instruction
- a MonoMethod pointer for the unmanaged function
- the address of the thread local lfm_addr pointer (to avoid another call to
arch_get_lmf_addr when restoring LMF)
The LMF is allocated on the stack, so we also know the stack position for
stack unwinding.
-3
View File
@@ -1,3 +0,0 @@
Makefile.in
Makefile
.deps
-24
View File
@@ -1,24 +0,0 @@
2004-06-24 David Waite <[email protected]>
* mono-jni.c: change to C90-style comments from C99/C++-style
Fri Jun 18 13:46:05 CEST 2004 Paolo Molaro <[email protected]>
* mono-jni.c: updates for the API cleanups.
2004-05-05 Zoltan Varga <[email protected]>
* mono-jni.c: Implement some missing JNI functions.
2004-04-13 Zoltan Varga <[email protected]>
* mono-jni.c: Use <stdarg.h> instead of <varargs.h>.
2004-04-12 Zoltan Varga <[email protected]>
* mono-jni.c: Fix compilation under gcc 2.95 + fix warnings.
2004-04-07 Zoltan Varga <[email protected]>
* mono-jni.c Makefile.am: New files.
-15
View File
@@ -1,15 +0,0 @@
INCLUDES = \
-I$(top_srcdir) \
$(IKVM_JNI_CFLAGS) \
$(GMODULE_CFLAGS) \
$(GLIB_CFLAGS)
if JIT_SUPPORTED
lib_LTLIBRARIES = libmono-ikvm-jni.la
endif
libmono_ikvm_jni_la_SOURCES = mono-jni.c
libmono_ikvm_jni_la_LIBADD = $(top_builddir)/mono/mini/libmono.la
-1263
View File
File diff suppressed because it is too large. Load diff
-2
View File
@@ -1,2 +0,0 @@
Makefile
Makefile.in
-5
View File
@@ -1,5 +0,0 @@
man_MANS = mcs.1 mono.1 monostyle.1 mono-config.5 sqlsharp.1 oldmono.1 ilasm.1 \
cert2spc.1 cilc.1 genxs.1 wsdl.1 disco.1 soapsuds.1 makecert.1 \
chktrust.1 setreg.1 sn.1 secutil.1 signcode.1 certmgr.1 monop.1 xsd.1 gacutil.1
EXTRA_DIST = $(man_MANS)
-42
View File
@@ -1,42 +0,0 @@
.\"
.\" cert2spc manual page.
.\" Copyright 2002, 2003 Motus Technologies
.\" Author:
.\" Sebastien Pouliot ([email protected])
.\"
.TH Mono "cert2spc"
.SH NAME
cert2spc \- Transform multiple X.509 certificates to a Software
Publisher Certificate
.SH SYNOPSIS
.PP
.B cert2spc certificate|crl [certificate|crl ...] outputfile.spc
.SH DESCRIPTION
Combine a list of X.509 certificates or X.509 CRL (Certificate
Revocation List) into a Software Publisher Certificate (PKCS#7
file). The SPC file is required for signing a PE file (like an
assembly) using signcode.
.SH PARAMETERS
.TP
.I "certificate"
One (or more) X.509 certificate to add (space separated).
.TP
.I "crl"
One (or more) X.509 CRL to add (space separated).
.TP
.I "output.spc"
Software Publisher Certificate to output. If the specified file
exists it will be overwritten.
.SH AUTHOR
Written by Sebastien Pouliot
.SH COPYRIGHT
Copyright (C) 2002, 2003 Motus Technologies.
Released under BSD license.
.SH MAILING LISTS
Visit http://mail.ximian.com/mailman/mono-list for details.
.SH WEB SITE
Visit: http://www.go-mono.com for details
.SH SEE ALSO
.BR makecert(1), signcode(1)
-75
View File
@@ -1,75 +0,0 @@
.\"
.\" certmgr manual page.
.\" Copyright 2004 Novell
.\" Author:
.\" Sebastien Pouliot <[email protected]>
.\"
.TH Mono "certmgr"
.SH NAME
certmgr \- Mono Certificate Manager (CLI version)
.SH SYNOPSIS
.PP
.B certmgr [action] [object type] [options] store [filename]
.SH DESCRIPTION
This tool allow to add, remove or extract certificates, certificate revocation
lists (CRL) or certificate trust lists (CTL) to/from a certificate store.
Certificate stores are used to build and validate certificate chains for
Authenticode(r) code signing validation and SSL server certificates.
.SH ACTIONS
.TP
.I "-add"
Add a certificate, CRL or CTL to specified store.
.TP
.I "-del"
Remove a certificate, CRL or CTL from specified store.
.TP
.I "-put"
Copy a certificate, CRL or CTL from a store to a file.
.SH OBJECT TYPES
.TP
.I "-c", "-cert", "-certificate"
Add, Delete or Put certificates. That is the specified file must/will contains
X.509 certificates in DER binary encoding.
.TP
.I "-crl"
Add, Delete or Put certificate revocation lists (CRL). That is the specified
file must/will contains X.509 CRL in DER binary encoding.
.TP
.I "-ctl"
Add, Delete or Put certificate trust lists (CRL). UNSUPPORTED.
.SH OPTIONS
.TP
.I "-v", "-verbose"
More details displayed on the console.
.TP
.I "-help", "-h", "-?", "/?"
Display help about this tool.
.SH FILES
.B WARNING: This details the current behavior of Mono and could change between releases.
The only safe way to interact with certificate stores is to use the certmgr
tool. The current releases of Mono keeps all the user certificate stores in
separates directories under
.I ~/.mono/certs/
.TP
For example the trusted root certificates for a user would be kept under
.I ~/.mono/certs/Trust/
.TP
Certificates files are kept in DER (binary) format (extension .cer).
.TP
The filenames either starts with
.I tbp
(thumbprint) or
.I ski
(subject key identifier).
.TP
The rest of the filename is the base64-encoded value (tbp or ski).
.SH AUTHOR
Written by Sebastien Pouliot
.SH COPYRIGHT
Copyright (C) 2004 Novell.
.SH MAILING LISTS
Visit http://mail.ximian.com/mailman/mono-list for details.
.SH WEB SITE
Visit: http://www.go-mono.com for details
.SH SEE ALSO
.BR makecert(1), setreg(1)
-46
View File
@@ -1,46 +0,0 @@
.\"
.\" chktrust manual page.
.\" Copyright 2003 Motus Technologies
.\" Copyright 2004 Novell
.\" Author:
.\" Sebastien Pouliot ([email protected])
.\"
.TH Mono "chktrust"
.SH NAME
chktrust \- Check the trust of a PE executable.
.SH SYNOPSIS
.PP
.B chktrust [options] filename
.SH DESCRIPTION
Verify if an PE executable (CLR assembly, Win32 EXE or DLL) has a valid
Authenticode(r) signature that can be traced back to a trusted certificate
authority (CA). This means that
.TP
(a) the signature is valid (i.e. file integrity) and,
.TP
(b) the code-signing certificate can be chained back to one of the
certificates in the Trust store.
.SH OPTIONS
.TP
.I "-q", "-quiet"
Quiet mode. Limited display on console.
.TP
.I "-v", "-verbose"
Verbose mode. More details displayed on the console.
.TP
.I "-help", "-h", "-?", "/?"
Display help about this tool.
.SH KNOWN RESTRICTIONS
chktrust cannot validate Authenticode signatures on CAB files.
.SH AUTHOR
Written by Sebastien Pouliot
.SH COPYRIGHT
Copyright (C) 2003 Motus Technologies.
Copyright (C) 2004 Novell.
Released under BSD license.
.SH MAILING LISTS
Visit http://mail.ximian.com/mailman/mono-list for details.
.SH WEB SITE
Visit: http://www.go-mono.com for details
.SH SEE ALSO
.BR signcode(1), makecert(1), certmgr(1), setreg(1)
-21
View File
@@ -1,21 +0,0 @@
.TH "cilc" 1
.SH NAME
cilc \- Mono CIL-to-C binding generator
.SH SYNOPSIS
.B cilc
[options]
.I assembly
.I target
.SH DESCRIPTION
cilc is the Mono CIL-to-C binding generator. It takes a CIL assembly as input
and generates a directory of C sources which, when compiled, provide a C
interface to the classes contained within that assembly by means of a shared
object library.
.PP
Generated sources make use of the Mono embedding API. Thus a complete Mono
development environment must be available on the development system and a
complete Mono runtime environment must be available on target systems.
.SH AUTHOR
Written by Alp Toker <[email protected]>
.SH SEE ALSO
mono(1)
-68
View File
@@ -1,68 +0,0 @@
.\"
.\" disco manual page.
.\" (C) 2003 Novell, Inc.
.\" Author:
.\" Lluis Sanchez Gual ([email protected])
.\"
.TH disco 1
.SH NAME
disco \- Mono's Web Service Discovery Tool
.SH SYNOPSIS
.PP
.B disco
[options] url
.SH DESCRIPTION
.I disco
is a tool for discovering web services and for retireving the documents that
describe those services.
.PP
.I url
is the location of a DISCO document, which includes a list of WSDL documents,
XML schemas and references to other DISCO documents.
.PP
.I disco
downloads the DISCO document and all referenced documents (unless
the
.I "-nosave"
option is specified), and saves them to disk.
.SH OPTIONS
The following options are available:
.TP
.I "-nologo"
Supress the startup logo.
.TP
.I "-nosave"
Do not save the discovered documents to disk. The default is to save the documents.
.TP
.I "-o:directory", "-out:directory"
The directory where to save the discovered documents.
By default, documents are saved in the current directory.
.TP
.I "-u:username", "-user:username"
The user name to use when connecting to the server.
.TP
.I "-p:password", "-password:password"
The password to use when connecting to the server.
.TP
.I "-d:domain", "-domain:domain"
The domain to use when connecting to the server.
.TP
.I "-proxy:url"
The url of the proxy server to use for http requests.
.TP
.I "-proxyusername:username"
The user name to use when connecting to the proxy.
.TP
.I "-proxypassword:password"
The password to use when connecting to the proxy.
.TP
.I "-proxydomain:domain"
The domain to use when connecting to the proxy.
.SH AUTHORS
Lluis Sanchez Gual ([email protected])
.PP
.SH LICENSE
disco is released under the terms of the GNU GPL.
.PP
.SH SEE ALSO
wsdl(1), soapsuds(1), mono(1), mcs(1)
-87
View File
@@ -1,87 +0,0 @@
..\"
.\" gacutil manual page.
.\" (C) 2004 Novell, Inc.
.\" Author:
.\" Miguel de Icaza ([email protected])
.\"
.de Sp \" Vertical space (when we can't use .PP)
.if t .sp .5v
.if n .sp
..
.TH gacutil "Mono 1.0"
.SH NAME
gacutil \- Global Asssembly Cache management utility.
.SH SYNOPSIS
.PP
.B gacutil [-user] [command] [options]
.SH DESCRIPTION
\fIgacutil\fP is a tool used by developers to install assemblies into
the system Global Assembly Cache (GAC) to become part of the
assemblies that are available for all applications at runtime.
.PP
Notice that they are not directly available to the compiler, the
convention is that assemblies must also be placed in a separate
directory for being accessed to the compiler. This is done with the
-package directive to gacutil.
.PP
The tool allows for installationl, removal, and listing of the
contents of the assembly cache.
.PP
The GAC is relative to the Mono installation prefix:
mono_prefix/lib/mono.
.SH FLAGS
The "-user" flag enables you to install/remove the assembly into the
user's assembly cache, as opposed to the global one.
.PP
.SH COMMANDS
.I "-i path_to_assembly" [-f] [-root ROOTDIR] [-package NAME]
.Sp
The above installs the assembly into the GAC, use the -f flag to force
an installation an overwrite anything in there.
.Sp
The -package option can be used to also create a directory in in
prefix/lib/mono with the name NAME, and a symlink is created from
NAME/assembly_name to the assembly on the GAC. This is used so
developers can reference a set of libraries at once.
.Sp
The -root option is used by developers integrating this with automake
tools or packaging tools that require a prefix directory to be
specified. The root represents the "libdir" component of a prefix
(typically prefix/lib).
.fi
.PP
.I "-l"
.Sp
Lists the contents of the Global Assembly Cache
.PP
.I "-u assembly_name [-root]"
.Sp
The above removes the given assembly name from the Global Assembly
Cache. As opposed to the install option that takes a filename, this
takes as an argument the assembly name, which looks like this:
.nf
MyLibrary.Something, version=1.0.0.0, publicKeyToken=xxxx,culture=neutral
.fi
.Sp
Notice that you can have spaces in the command line, there is no need
to quote them.
.Sp
The removal performs a greedy removal, if you only specifiy one
component like, "MyLibrary.Something", it will remove all versions of
the library.
.SH FILES
On Unix assemblies are loaded from the installation lib directory. If you set
`prefix' to /usr, the assemblies will be located in /usr/lib. On
Windows, the assemblies are loaded from the directory where mono and
mint live.
.PP
/etc/mono/config, ~/.mono/config
.PP
Mono runtime configuration file. See the mono-config(5) manual page
for more information.
.SH WEB SITE
Visit: http://www.go-mono.com for details
.SH SEE ALSO
.BR mcs(1), mono(1)
-294
View File
@@ -1,294 +0,0 @@
.\"
.\" genxs manual page.
.\" (C) Lluis Sanchez Gual ([email protected])
.\"
.TH genxs 1
.SH NAME
genxs \- Mono's Xml Serializer Generator
.SH SYNOPSIS
.PP
.B genxs
configurationFile [destinationFolder]
.SH DESCRIPTION
.I genxs
is a tool for generating custom XML serialization writers and readers for
classes.
.PP
.I configurationFile
is configuration file which specifies several
information, such as the class for which to generate the reader and writer, the
name and namespace of the classes to generate, and a collection of hooks to
apply. By using hooks it is possible to customize the
behavior of the serializer without needing to modify the generated file, so you
can safely regenerate it if the source class is modified.
.PP
.I destinationFolder
specifies the folder where the files will be generated.
.PP
.B NOTE:
This tool only runs in the Mono runtime, since it uses some internal
classes not available in other runtimes.
.SH CONFIGURATION FILE FORMAT
The configuration file is an xml document based on the following grammar
("?" means optional, "*" 0 or more):
.PP
.nf
<configuration>
<serializer class="name" assembly="name"> *
<reader>name</reader> ?
<writer>name</writer> ?
<namespace>name</namespace> ?
<outFileName>name</outFileName> ?
<readerHooks> ?
<hook ...> *
</readerHooks>
<writerHooks> ?
<hook ...> *
</writerHooks>
</serializer>
</configuration>
.fi
.PP
A configuration file can have multiple "serializer" elements, each of which
specifies the class for which to generate a serializer together with several
generation options. The source class is specified in the following attributes:
.PP
.IP " *" 5
.I class
: name of the class (including namespace).
.IP " *" 5
.I assembly
: assembly name. It can include the complete path.
.PP
Generation options are specified in child elements:
.PP
.IP " *" 5
.I reader
: name of the reader class.
.IP " *" 5
.I writer
: name of the writer class.
.IP " *" 5
.I namespace
: namespace of the reader and writer classes.
.IP " *" 5
.I outFileName
: name of the generated file.
.IP " *" 5
.I readerHooks
: a list of hooks to apply to the reader.
.IP " *" 5
.I writerHooks
: a list of hooks to apply to the writer.
.SH SPECIFYING HOOKS
Using hooks you can customize the behavior of readers and writers.
A hook specification follows this grammar:
.PP
.nf
<hook type="name">
<select> ?
<typeName>name</typeName> ?
<typeAttribute>name</typeAttribute> *
<typeMember>name</typeMember> ?
</select>
<replace>source code</replace> ?
<insertBefore>source code</insertBefore> ?
<insertAfter>source code</insertAfter> ?
</hook>
.fi
.PP
The "type" attribute specifies the context in which the hook is applied. It can
be one of the following:
.PP
.IP " *" 5
.I attributes
: hook is applied where attributes are serialized/deserialized.
.IP " *" 5
.I elements
: hook is applied where elements are serialized/deserialized.
.IP " *" 5
.I unknownAttribute
: hook is applied where unknown attributes are processed.
.IP " *" 5
.I unknownElement
: hook is applied where unknown elements are processed.
.IP " *" 5
.I member
: hook is applied where a member is serialized/deserialized.
.IP " *" 5
.I type
: hook is applied for the whole type.
.PP
The "select" element specifies the classes and members to which the hook has
to be added. It can contain the following elements:
.PP
.IP " *" 5
.I typeName
: the class with that name will be selected (must include namespace)
.IP " *" 5
.I typeAttribute
: all classes which have that attribute applied will be selected (specify the
full attribute class name, including namespace). Several attribute names can be
specified.
.IP " *" 5
.I typeMember
: name of the class member for which the hook must be added.
.PP
The hook source code can be specified using any of the following elements:
.PP
.IP " *" 5
.I replace
: the provided source code will replace all serialization/deserialization
operations in the hook context.
.IP " *" 5
.I insertBefore
: the source code will be added before the hook context.
.IP " *" 5
.I insertAfter
: the source code will be added after the hook context.
.PP
When writing the code for a hook you can use some special variables that are
defined during the code generation process. The variables are the following:
.PP
.IP " *" 5
.I $TYPE:
name of the class being generated, without namespace.
.IP " *" 5
.I $FULLTYPE:
full name of the class being generated, including namespace.
.IP " *" 5
.I $OBJECT:
the object being serialized or deserialized. When using a replace
reader hook of type "type", the hook code must assign the deserialized object
to this variable.
.IP " *" 5
-I $ELEMENT:
name of the element of the object being serialized/deserialized.
.IP " *" 5
.I $NAMESPACE:
namespace of the element of the object being serialized/deserialized.
.IP " *" 5
.I $MEMBER:
name of the member being serialized/deserialized. Only valid in the "member"
context.
.SH HOOK EXAMPLES
The following example adds a call to a Validate method after the deserialization
of any object:
.PP
.nf
<hook type="type">
<insertAfter>
System.Xml.Schema.XmlSchema.Validate$TYPE ($OBJECT);
</insertAfter>
</hook>
.fi
.PP
This example specifies the code to be used to deserialize the XmlSchema class:
.PP
.nf
<hook type="type">
<select>
<typeName>System.Xml.Schema.XmlSchema</typeName>
</select>
<replace>
$OBJECT = System.Xml.Schema.XmlSchema.Read (Reader, null);
</replace>
</hook>
.fi
.PP
That one specifies the code to be used to read XmlSchema instances:
.PP
.nf
<hook type="type">
<select>
<typeName>System.Xml.Schema.XmlSchema</typeName>
</select>
<replace>$OBJECT.Write (Writer);</replace>
</hook>
.fi
.PP
With this two hooks the serializer will print some information when serializing
the class "MyClass":
.PP
.nf
<hook type="type">
<select>
<typeName>MyNamespace.MyClass</typeName>
</select>
<insertBefore>Console.WriteLine ("Serializing MyClass");</replace>
<insertAfter>Console.WriteLine ("MyClass serialized");</insertAfter>
</hook>
<hook type="member">
<select>
<typeName>MyNamespace.MyClass</typeName>
</select>
<insertAfter>
Console.WriteLine ("Serialized member $MEMBER");
</insertAfter>
</hook>
.fi
.PP
This hook writes an additional element for all types that have the custom
attribute "MyAttribute":
.PP
.nf
<hook type="elements">
<select>
<typeAttribute>MyNamespace.MyAttribute</typeAttribute>
</select>
<insertAfter>
Writer.WriteStartElement ("privateData");
Writer.WriteString ($OBJECT.PrivateData);
Writer.WriteEndElement ();
</insertAfter>
</hook>
.fi
.SH CONFIGURATION FILE EXAMPLE
This is the configuration file used to generate the serializer for ServiceDescription:
.PP
.nf
<configuration>
<serializer class="System.Web.Services.Description.ServiceDescription" assembly="System.Web.Services">
<reader>ServiceDescriptionReaderBase</reader>
<writer>ServiceDescriptionWriterBase</writer>
<namespace>System.Web.Services.Description</namespace>
<outFileName>ServiceDescriptionSerializerBase.cs</outFileName>
<readerHooks>
<hook type="unknownElement">
<select>
<typeAttribute>System.Web.Services.Configuration.XmlFormatExtensionPointAttribute</typeAttribute>
</select>
<replace>ServiceDescription.ReadExtension (Reader, $OBJECT);</replace>
</hook>
<hook type="type">
<select>
<typeName>System.Xml.Schema.XmlSchema</typeName>
</select>
<replace>$OBJECT = System.Xml.Schema.XmlSchema.Read (Reader, null);</replace>
</hook>
</readerHooks>
<writerHooks>
<hook type="elements">
<select>
<typeAttribute>System.Web.Services.Configuration.XmlFormatExtensionPointAttribute</typeAttribute>
</select>
<insertBefore>ServiceDescription.WriteExtensions (Writer, $OBJECT);</insertBefore>
</hook>
<hook type="type">
<select>
<typeName>System.Xml.Schema.XmlSchema</typeName>
</select>
<replace>$OBJECT.Write (Writer);</replace>
</hook>
</writerHooks>
</serializer>
</configuration>
.fi
.SH AUTHORS
Lluis Sanchez Gual ([email protected])
.PP
.SH LICENSE
GenXS is released under the terms of the GNU GPL.
.PP
.SH SEE ALSO
mono(1), mcs(1)
-44
View File
@@ -1,44 +0,0 @@
.TH "ilasm" 1
.SH NAME
ilasm \- Mono IL assembler
.SH SYNOPSIS
.B ilasm
[option] [source-files]
.SH DESCRIPTION
ilasm is the Mono ILAsm assembler. You can pass one or more options to
drive the compiler, and a set of source files.
The Mono ILAsm assembler accepts options starting with / or - for mosted
options.
The Mono ILAsm assembler generates images (.exe and .dll files) that contain
CIL byte code that can be executed by any system that implemented a Common
Language Infrastructure virtual machine.
.SH OPTIONS
.TP
.B \//output:<file_name>
Specify the output file name.
.TP
.B \//exe
Generate an exe
.TP
.B \//dll
Generate a library
.TP
.B \//scan_only
Just scan the IL code and display tokens.
.TP
.B \//show_tokens
Show tokens as they are parsed.
.TP
.B \//show_method_def
Display method information when a method is defined.
.TP
.B \//show_method_ref
Display method information when a method is referenced
.TP
.B \--about
Display information about ilasm
.TP
.B \--version
Display ilasm version
-129
View File
@@ -1,129 +0,0 @@
.\"
.\" makecert manual page.
.\" Copyright 2003 Motus Technologies
.\" Copyright 2004 Novell
.\" Author:
.\" Sebastien Pouliot ([email protected])
.\"
.TH Mono "MakeCert"
.SH NAME
MakeCert \- Create X.509 certificates for test purposes
.SH SYNOPSIS
.PP
.B makecert [options] certificate
.SH DESCRIPTION
Create an X.509 certificate using the provided informations. This
is useful for testing Authenticode signatures, SSL and S/MIME
technologies.
.SH PARAMETERS
.TP
.I "-# num"
Specify the certificate serial number.
.TP
.I "-n dn"
Specify the subject Distinguished Name (DN).
.TP
.I "-in dn"
Specify the issuer Distinguished Name (DN).
.TP
.I "-r"
Create a self-signed, also called root, certificate.
.TP
.I "-iv pvkfile"
Specify the private key file (.PVK) for the issuer. The private key in the
specified file will be used to sign the new certificate.
.TP
.I "-ic certfile"
Extract the issuer's name from the specified certificate file - i.e. the
subject name of the specified certificate becomes the issuer name of the
new certificate.
.TP
.I "-in name"
Use the issuer's name from the specified parameter.
.TP
.I "-ik container"
Specify the key container name to be used for the issuer.
.TP
.I "-iky [signature | exchange | #]"
Specify the key number to be used in the provider (when used with -ik).
.TP
.I "-ip provider"
Specify the cryptographic provider to be used for the issuer.
.TP
.I "-ir [localmachine | currentuser]"
Specify the provider will search the user or the machine keys containers for
the issuer.
.TP
.I "-iy number"
Specify the provider type to be used for the issuer.
.TP
.I "-sv pkvfile"
Specify the private key file (.PVK) for the subject. The public part of the
key will be inserted into the created certificate. If non-existant the
specified file will be created with a new key pair (default to 1024 bits RSA
key pair).
.TP
.I "-sk container"
Specify the key container name to be used for the subject.
.TP
.I "-sky [signature | exchange | #]"
Specify the key number to be used in the provider (when used with -sk).
.TP
.I "-sp provider"
Specify the cryptographic provider to be used for the subject.
.TP
.I "-sr [localmachine | currentuser]"
Specify the provider will search the user or the machine keys containers for
the subject.
.TP
.I "-sy number"
Specify the provider type to be used for the issuer.
.TP
.I "-a hash"
Select hash algorithm. Only MD5 and SHA1 algorithms are supported.
.TP
.I "-b date"
The date since when the certificate is valid (notBefore).
.TP
.I "-e date"
The date until when the certificate is valid (notAfter).
.TP
.I "-m number"
Specify the certificate validity period in months. This is added to the
notBefore validity date which can be set with -b or will default to the
current date/time.
.TP
.I "-cy [authority|end]"
Basic constraints. Select Authority or End-Entity certificate. Only Authority
certificates can be used to sign other certificates (-ic). End-Entity can
be used by clients (e.g. Authenticode, S/MIME) or servers (e.g. SSL).
.TP
.I "-h number"
Add a path length restriction to the certificate chain. This is only
applicable for certificates that have BasicConstraint set to Authority (-cy
authority). This is used to limit the chain of certificates than can be
issued under this authority.
.TP
.I "-eku oid[,oid]"
Add some extended key usage OID to the certificate.
.TP
.I "-?"
Help (display this help message)
.TP
.I "-!"
Extended help (for advanced options)
.SH KNOWN RESTRICTIONS
Compared to the Windows version some options aren't supported (-$, -d, -l,
-nscp, -is, -sc, -ss). Also PVK files with passwords aren't supported.
.SH AUTHOR
Written by Sebastien Pouliot
.SH COPYRIGHT
Copyright (C) 2003 Motus Technologies.
Copyright (C) 2004 Novell.
Released under BSD license.
.SH MAILING LISTS
Visit http://mail.ximian.com/mailman/mono-list for details.
.SH WEB SITE
Visit: http://www.go-mono.com for details
.SH SEE ALSO
.BR signcode(1)
-24
View File
@@ -1,24 +0,0 @@
<?xml version="1.0"?>
<manpages>
<manpage name="Mono" page="mono.1" />
<manpage name="Mint" page="mono.1" />
<manpage name="Mcs" page="mcs.1" />
<manpage name="Ilasm" page="ilasm.1" />
<manpage name="Monodis" page="mono.1" />
<manpage name="Cilc" page="cilc.1" />
<manpage name="Sql#" page="sqlsharp.1" />
<manpage name="SecUtil" page="secutil.1" />
<manpage name="MonoStyle" page="monostyle.1" />
<manpage name="genxs" page="genxs.1" />
<manpage name="cert2spc" page="cert2spc.1" />
<manpage name="makecert" page="makecert.1" />
<manpage name="chktrust" page="chktrust.1" />
<manpage name="signcode" page="signcode.1" />
<manpage name="sn" page="sn.1" />
<manpage name="wsdl" page="wsdl.1" />
<manpage name="disco" page="disco.1" />
<manpage name="soapsuds" page="soapsuds.1" />
<manpage name="setreg" page="setreg.1" />
<manpage name="certmgr" page="certmgr.1" />
<manpage name="xsd" page="xsd.1" />
</manpages>
Loaded 100 of 11569 files, more files were not shown because too many files have changed in this diff. Show more