This commit was manufactured by cvs2svn to create tag

'egcs_1_0_3_prerelease'.

From-SVN: r19382

COPYING.LIB

@@ -1,481 +1,3 @@
-		  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!
+ 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA
+    Software Foundation, Inc., 59 Temple Place - Suite 330, Boston,
+    MA 02111-1307, USA

ChangeLog

@@ -1,3 +1,53 @@
+Thu Apr 23 01:40:32 1998  Jeffrey A Law  (law@cygnus.com)
+
+	* egcs-1.0.3 release.
+
+Sun Mar 15 23:38:21 1998  Jeffrey A Law  (law@cygnus.com)
+
+	* egcs-1.0.2 release.
+
+Sun Mar 15 23:32:17 1998 Manfred Hollstein  <manfred@s-direktnet.de>
+			 Pieter Nagel <pnagel@epiuse.co.za>
+
+	* Makefile.in (BASE_FLAGS_TO_PASS): Pass gxx_include_dir down
+	to sub-makes.
+
+Sun Feb 22 15:23:27 1998  H.J. Lu  (hjl@gnu.ai.mit.edu)
+
+	* configure.in (alpha*-*-linux*): Treat alpha*-*-linux* as
+	alpha*-*-linux* not alpha*-*-*.
+
+Fri Jan  2 11:34:38 1998  Jeffrey A Law  (law@cygnus.com)
+
+	* egcs-1.0.1 release.
+
+Wed Dec  3 07:55:59 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* egcs-1.0 release.
+
+	* configure (gxx_include_dir): Fix thinko.
+
+Thu Nov 27 01:31:30 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* Makefile.in (INSTALL_TARGET): Do install-gcc first.
+	* configure (gxx_include_dir): Provide a definition for subdirs
+	which do not use autoconf.
+	
+Wed Nov 26 16:08:50 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* From Franz Sirl.
+	* config.guess (powerpc*-*-linux): Handle glibc2 beta release
+	found on RedHat Linux systems.
+
+Fri Nov 21 09:51:01 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* config.guess (alpha stuff): Merge with FSF to avoid incorrect
+	guesses.
+
+Thu Nov 13 11:38:37 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* configure.in (i[3456]86-ncr-sysv4.3*): Tweak.
+
 Mon Oct 27 12:03:53 1997  Jason Merrill  <jason@yorick.cygnus.com>
 
 	* Makefile.in: check-target-libio depends on all-target-libstdc++.

INSTALL/BUILD

@@ -0,0 +1,50 @@
+
+                            Building egcs-1.0.3
+                                      
+   Now that egcs is configured, you are ready to build the compiler and
+   runtime libraries.
+   
+   We highly recommend that egcs be built using gnu-make; other versions
+   make work, then again they might not. To be safe build with gnu-make.
+   
+   Building a native compiler
+   
+   For a native build issue the command "make bootstrap". This will build
+   the entire egcs compiler system, which includes the following steps:
+     * Build host tools necessary to build the compiler such as texinfo,
+       bison, gperf.
+     * Build target tools for use by the compiler such as gas, gld, and
+       binutils.
+     * Perform a 3-stage bootstrap of the compiler.
+     * Perform a comparison test of the stage2 and stage3 compilers.
+     * Build runtime libraries using the stage3 compiler from the
+       previous step.
+       
+   If you are short on disk space you might consider "make
+   bootstrap-lean" instead. This is identical to "make bootstrap" except
+   that object files from the stage1 and stage2 of the 3-stage bootstrap
+   of the compiler are deleted as soon as they are no longer needed.
+   
+   Building a cross compiler
+   
+   We recommend reading the [1]crossgcc FAQ for information about
+   building cross compilers.
+   
+   For a cross build, issue the command "make cross", which performs the
+   following steps:
+     * Build host tools necessary to build the compiler such as texinfo,
+       bison, gperf.
+     * Build target tools for use by the compiler such as gas, gld, and
+       binutils.
+     * Build the compiler (single stage only).
+     * Build runtime libraries using the compiler from the previous step.
+       
+   Note that if an error occurs in any step the make process will exit.
+   
+     _________________________________________________________________
+                                      
+   Last modified on Jan 2, 1998.
+
+References
+
+   1. ftp://ftp.cygnus.com/pub/embedded/crossgcc/FAQ-0.8.1

INSTALL/CONFIGURE

@@ -0,0 +1,90 @@
+
+                           Configuring egcs-1.0.3
+                                      
+   Like most GNU software, egcs must be configured before it can be
+   built. This document attempts to describe the recommended
+   configuration procedure for both native and cross targets.
+   
+   We use srcdir to refer to the toplevel source directory for egcs; we
+   use objdir to refer to the toplevel build/object directory for egcs.
+   
+   First, we highly recommend that egcs be built into a separate
+   directory than the sources. This is how we generally build egcs;
+   building where srcdir == objdir should still work, but doesn't get
+   extensive testing.
+   
+   Second, when configuring a native system, either "cc" must be in your
+   path or you must set CC in your environment before running configure.
+   Otherwise the configuration scripts may fail.
+   
+   To configure egcs:
+   
+     
+     % mkdir objdir
+     % cd objdir
+     % srcdir/configure [target] [options]
+     
+   target specification
+     * egcs has code to correctly determine the correct value for target
+       for nearly all native systems. Therefore, we highly recommend you
+       not provide a configure target when configuring a native compiler.
+     * target must be specified when configuring a cross compiler;
+       examples of valid targets would be i960-rtems, m68k-coff, sh-elf,
+       etc.
+       
+   options specification
+   
+   Use options to override several configure time options for egcs. A
+   partial list of supported options:
+     * --prefix=dirname -- Specify the toplevel installation directory.
+       This is the recommended way to install the tools into a directory
+       other than the default. The toplevel installation directory
+       defaults to /usr/local.
+       These additional options control where certain parts of the
+       distribution are installed. Normally you should not need to use
+       these options.
+          + --with-local-prefix=dirname -- Specify the installation
+            directory for local include files. The default is /usr/local.
+          + --with-gxx-include-dir=dirname -- Specify the installation
+            directory for g++ header files. The default is
+            /usr/local/include/g++.
+     * --enable-shared -- Build shared versions of the C++ runtime
+       libraries if supported --disable-shared is the default.
+     * --enable-haifa -- Enable the new Haifa instruction scheduler in
+       the compiler; the new scheduler can significantly improve code on
+       some targets. --disable-haifa is currently the default on all
+       platforms except the HPPA.
+     * --with-gnu-as -- Specify that the compiler should assume the GNU
+       assembler (aka gas) is available.
+     * --with-gnu-ld -- Specify that the compiler should assume the GNU
+       linker (aka gld) is available.
+     * --with-stabs -- Specify that stabs debugging information should be
+       used instead of whatever format the host normally uses. Normally
+       GCC uses the same debug format as the host system.
+     * --enable-multilib -- Specify that multiple target libraries should
+       be built to support different target variants, calling
+       conventions, etc. This is the default.
+     * --enable-threads -- Specify that the target supports threads. This
+       only effects the Objective-C compiler and runtime library.
+     * --enable-threads=lib -- Specify that lib is the thread support
+       library. This only effects the Objective-C compiler and runtime
+       library.
+     * --with-cpu=cpu -- Specify which cpu variant the compiler should
+       generate code for by default. This is currently only supported on
+       the RS6000/PowerPC ports.
+       
+   Some options which only apply to building cross compilers:
+     * --with-headers=dir -- Specifies a directory which has target
+       include files.
+     * --with-libs=dirs -- Specifies a list of directories which contain
+       the target runtime libraries.
+     * --with-newlib -- Specifies that "newlib" is being used as the
+       target C library. This causes __eprintf to be omitted from
+       libgcc.a on the assumption that it will be provided by newlib.
+       
+   Note that each --enable option has a corresponding --disable option
+   and that each --with option has a corresponding --without option.
+   
+     _________________________________________________________________
+                                      
+   Last modified on Jan 2, 1998.

INSTALL/FAQ

@@ -0,0 +1,633 @@
+
+                      egcs Frequently Asked Questions
+                                      
+    1. [1]How is egcs different from gcc2?
+    2. [2]What is an open development model?
+    3. [3]Releases and Forking
+    4. [4]bits/libc-lock.h: No such file or directory
+    5. [5]`_IO_stdfile_0_lock' was not declared in this scope
+    6. [6]Problems building the Fortran compiler
+    7. [7]Problems building on MIPS platforms
+    8. [8]Problems with exception handling on x86 platforms
+    9. [9]Bootstrap comparison failures on HPs
+   10. [10]Bootstrap loops rebuilding cc1 over and over
+   11. [11]Dynamic linker is unable to find GCC libraries
+   12. [12]libstdc++/libio tests fail badly with --enable-shared
+   13. [13]Unable to run the testsuite
+   14. [14]How to build a cross compiler
+   15. [15]How to install both gcc2 and egcs
+   16. [16]Snapshots, how, when, why
+   17. [17]Problems building Linux kernels
+   18. [18]Virtual memory exhausted
+   19. [19]GCC can not find GAS
+   20. [20]egcs does not work on Red Hat 5.0
+   21. [21]Unable to bootstrap on x86 Solaris2.{5,6}
+   22. [22]EGCS with Windows
+   23. [23]EGCS with OS/2
+   24. [24]cpp: Usage:... Error
+   25. [25]EGCS will not build KDE
+   26. [26]Friend Templates
+   27. [27]Where to find libg++
+   28. [28]Why do I need autoconf & bison
+   29. [29]EGCS does not work on AIX 4.3
+   30. [30]Problems debugging egcs code
+   31. [31]Conflicts when using cvs update 
+       
+     _________________________________________________________________
+                                      
+How is egcs be different from gcc2?
+
+   Six years ago, gcc version 1 had reached a point of stability. For the
+   targets it could support, it worked well. It had limitations inherent
+   in its design that would be difficult to resolve, so a major effort
+   was made and gcc version 2 was the result. When we had gcc2 in a
+   useful state, development efforts on gcc1 stopped and we all
+   concentrated on making gcc2 better than gcc1 could ever be. This is
+   the kind of step forward we want to make with egcs.
+   
+   In brief, the three biggest differences between egcs and gcc2 are:
+     * More rexamination of basic architectual decisions of gcc and an
+       interest in adding new optimizations;
+     * working with the groups who have fractured out from gcc2 (like the
+       Linux folks, the Intel optimizations folks, Fortran folks)
+       including more front-ends; and finally
+     * An open development model ([32]see below) for the development
+       process.
+       
+   These three differences will work together to result in a more useful
+   compiler, a more stable compiler, a central compiler that works for
+   more people, a compiler that generates better code.
+   
+   There are a lot of exciting compiler optimizations that have come out.
+   We want them in gcc. There are a lot of front ends out there for gcc
+   for languages like Fortran or Pascal. We want them easily installable
+   by users. After six years of working on gcc2, we've come to see
+   problems and limitations in the way gcc is architected; it is time to
+   address these again.
+     _________________________________________________________________
+                                      
+What is an open development model?
+
+   With egcs, we are going to try a bazaar style[33][1] approach to its
+   development: We're going to be making snapshots publicly available to
+   anyone who wants to try them; we're going to welcome anyone to join
+   the development mailing list. All of the discussions on the
+   development mailing list are available via the web. We're going to be
+   making releases with a much higher frequency than they have been made
+   in the past: We're shooting for three by the end of 1997.
+   
+   In addition to weekly snapshots of the egcs development sources, we
+   are going to look at making the sources readable from a CVS server by
+   anyone. We want to make it so external maintainers of parts of egcs
+   are able to commit changes to their part of egcs directly into the
+   sources without going through an intermediary.
+   
+   There have been many potential gcc developers who were not able to
+   participate in gcc development in the past. We these people to help in
+   any way they can; we ultimately want gcc to be the best compiler in
+   the world.
+   
+   A compiler is a complicated piece of software, there will still be
+   strong central maintainers who will reject patches, who will demand
+   documentation of implementations, and who will keep the level of
+   quality as high as it is today. Code that could use wider testing may
+   be intergrated--code that is simply ill-conceived won't be.
+   
+   egcs is not the first piece of software to use this open development
+   process; FreeBSD, the Emacs lisp repository, and Linux are a few
+   examples of the bazaar style of development.
+   
+   With egcs, we will be adding new features and optimizations at a rate
+   that has not been done since the creation of gcc2; these additions
+   will inevitably have a temporarily destabilizing effect. With the help
+   of developers working together with this bazaar style development, the
+   resulting stability and quality levels will be better than we've had
+   before.
+   
+     [1] We've been discussing different development models a lot over
+     the past few months. The paper which started all of this introduced
+     two terms: A cathedral development model versus a bazaar
+     development model. The paper is written by Eric S. Raymond, it is
+     called ``[34]The Cathedral and the Bazaar''. The paper is a useful
+     starting point for discussions.
+     
+     _________________________________________________________________
+                                      
+Releases and Forking?
+
+   Some folks have questioned whether or not making releases is
+   consistent with the goals of the egcs project and whether or not
+   making releases is a fork from gcc2.
+
+The egcs project has several goals, including:
+
+  * Experimenting with a new development model, release process and
+  release packaging,
+
+  * Using the new development model to accelerate development of new
+  features, optimizations, etc for future inclusion in gcc,
+
+  * Providing high quality releases to the public.
+
+An egcs release is a copy of the egcs sources that the developers have
+tested and are believed to be suitable for wider scale use and testing.
+
+Making releases of stable, tested sources is both a goal and a means by
+which we hope to achieve other goals of the egcs project.
+
+The existence of a stable tested release allows egcs to be more thoroughly
+used and tested by a wider audience than is capable of testing snapshots.
+The expanded audience provides developers with critical feedback in a
+timely manner, which is beneficial to GCC as a whole and is consistent with
+the stated goals of egcs.
+
+The gcc maintainers are encouraged to migrate tested fixes and new features
+from egcs into gcc at their discretion.  egcs maintainers are willing to
+assist the gcc maintainers as time permits.  egcs periodically merges in
+changes from gcc into the egcs sources.
+
+What will keep egcs from becoming a fork is cooperation between the
+developers of gcc and egcs.
+
+We don't see this situation as significantly different than other projects
+that make releases based on some version of the gcc sources (Cygnus, g77,
+etc).  All the code is still available for inclusion in gcc at the discretion
+of the gcc maintainers.
+
+     _________________________________________________________________
+                                      
+bits/libc-lock.h: No such file or directory
+
+   This entry should be obsolete, egcs should handle these beta versions
+   of glibc2 correctly.
+   
+   egcs includes a tightly integrated libio and libstdc++ implementation
+   which can cause problems on hosts which have libio integrated into
+   their C library (most notably Linux).
+   
+   We believe that we've solved the major technical problems for the most
+   common versions of libc found on Linux systems. However, some versions
+   of Linux use pre-release versions of glibc2, which egcs has trouble
+   detecting and correctly handling.
+   
+   If you're using one of these pre-release versions of glibc2, you may
+   get a message "bits/libc-lock.h: No such file or directory" when
+   building egcs. Unfortunately, to fix this problem you will need to
+   update your C library to glibc2.0.5c.
+     _________________________________________________________________
+                                      
+`_IO_stdfile_0_lock' was not declared in this scope
+
+   If you get this error, it means either egcs incorrectly guessed what
+   version of libc is installed on your linux system, or you incorrectly
+   specified a version of glibc when configuring egcs.
+   
+   If you did not provide a target name when configuring egcs, then
+   you've found a bug which needs to be reported. If you did provide a
+   target name at configure time, then you should reconfigure without
+   specifying a target name.
+     _________________________________________________________________
+                                      
+Problems building the Fortran compiler
+
+   The Fortran front end can not be built with most vendor compilers; it
+   must be built with gcc. As a result, you may get an error if you do
+   not follow the install instructions carefully.
+   
+   In particular, instead of using "make" to build egcs, you should use
+   "make bootstrap" if you are building a native compiler or "make cross"
+   if you are building a cross compiler.
+   
+   It has also been reported that the Fortran compiler can not be built
+   on Red Hat 4.X linux for the Alpha. Fixing this may require upgrading
+   binutils or to Red Hat 5.0; we'll provide more information as it
+   becomes available.
+     _________________________________________________________________
+                                      
+Problems building on MIPS platforms
+
+   egcs requires the use of GAS on all versions of Irix, except Irix 6
+   due to limitations in older Irix assemblers.
+   
+   Either of these messages indicates that you are using the MIPS
+   assembler when instead you should be using GAS.
+
+    as0: Error: ./libgcc2.c, line 1:Badly delimited numeric literal
+          .4byte $LECIE1-$LSCIE1
+    as0: Error: ./libgcc2.c, line 1:malformed statement
+
+     _________________________________________________________________
+                                      
+    as0: Error: /home/law/egcs_release/gcc/libgcc2.c, line 1:undefined symbol i
+n expression
+    .word $LECIE1-$LSCIE1
+
+   For Irix 6, you should use the native assembler as GAS is not
+   supported on Irix 6.
+     _________________________________________________________________
+                                      
+Problems with exception handling on x86 platforms
+
+   If you are using the GNU assembler (aka gas) on an x86 platform and
+   exception handling is not working correctly, then odds are you're
+   using a buggy assembler.
+   
+   We recommend binutils-2.8.1.0.15 or newer.
+   [35]binutils-2.8.1.0.20 source
+   [36]binutils-2.8.1.0.20 x86 binary for libc5
+   [37]binutils-2.8.1.0.20 x86 binary for glibc2 Or, you can try a
+   [38]binutils snapshot; however, be aware that the binutils snapshot is
+   untested and may not work (or even build). Use it at your own risk.
+     _________________________________________________________________
+                                      
+Bootstrap comparison failures on HPs
+
+   If you bootstrap the compiler on hpux10 using the HP assembler instead
+   of gas, every file will fail the comparison test.
+   
+   The HP asembler inserts timestamps into object files it creates,
+   causing every file to be different. The location of the timestamp
+   varies for each object file, so there's no real way to work around
+   this mis-feature.
+   
+   Odds are your compiler is fine, but there's no way to be certain.
+   
+   If you use GAS on HPs, then you will not run into this problem because
+   GAS never inserts timestamps into object files. For this and various
+   other reasons we highly recommend using GAS on HPs.
+     _________________________________________________________________
+                                      
+Bootstrap loops rebuilding cc1 over and over
+
+   When building egcs, the build process loops rebuilding cc1 over and
+   over again. This happens on mips-sgi-irix5.2, and possibly other
+   platforms.
+   
+   This is probably a bug somewhere in the egcs Makefile. Until we find
+   and fix this bug we recommend you use GNU make instead of vendor
+   supplied make programs.
+     _________________________________________________________________
+                                      
+Dynamic linker is unable to find GCC libraries
+
+   This problem manifests itself by programs not finding shared libraries
+   they depend on when the programs are started. Note this problem often
+   manifests itself with failures in the libio/libstdc++ tests after
+   configuring with --enable-shared and building egcs.
+   
+   GCC does not specify a runpath so that the dynamic linker can find
+   dynamic libraries at runtime.
+   
+   The short explaination is that if you always pass a -R option to the
+   linker, then your programs become dependent on directories which may
+   be NFS mounted, and programs may hang unnecessarily when an NFS server
+   goes down.
+   
+   The problem is not programs that do require the directories; those
+   programs are going to hang no matter what you do. The problem is
+   programs that do not require the directories.
+   
+   SunOS effectively always passed a -R option for every -L option; this
+   was a bad idea, and so it was removed for Solaris. We should not
+   recreate it.
+     _________________________________________________________________
+                                      
+Unable to run the testsuite
+
+   If you get a message about unable to find "standard.exp" when trying
+   to run the egcs testsuites, then your dejagnu is too old to run the
+   egcs tests. You will need to get a newer version of dejagnu; we've
+   made a [39]dejagnu snapshot available until a new version of dejagnu
+   can be released.
+     _________________________________________________________________
+                                      
+How to build a cross compiler
+
+   Building cross compilers is a rather complex undertaking because they
+   usually need additional software (cross assembler, cross linker,
+   target libraries, target include files, etc).
+   
+   We recommend reading the [40]crossgcc FAQ for information about
+   building cross compilers.
+   
+   If you have all the pieces available, then `make cross' should build a
+   cross compiler. `make LANGUAGES="c c++" install'will install the cross
+   compiler.
+   
+   Note that if you're trying to build a cross compiler in a tree which
+   includes binutils-2.8 in addition to egcs, then you're going to need
+   to make a couple minor tweaks so that the cross assembler, linker and
+   nm utilities will be found.
+   
+   binutils-2.8 builds those files as gas.new, ld.new and nm.new; egcs
+   gcc looks for them using gas-new, ld-new and nm-new, so you may have
+   to arrange for any symlinks which point to &ltfile>.new to be changed
+   to &ltfile>-new.
+     _________________________________________________________________
+                                      
+Snapshots, how, when, why
+
+   We make snapshots of the egcs sources about once a week; there is no
+   predetermined schedule. These snapshots are intended to give everyone
+   access to work in progress. Any given snapshot may generate incorrect
+   code or even fail to build.
+   
+   If you plan on downloading and using snapshots, we highly recommend
+   you subscribe to the egcs mailing lists. See [41]mailing lists on the
+   main egcs page for instructions on how to subscribe.
+   
+   When using the diff files to update from older snapshots to newer
+   snapshots, make sure to use "-E" and "-p" arguments to patch so that
+   empty files are deleted and full pathnames are provided to patch. If
+   your version of patch does not support "-E", you'll need to get a
+   newer version. Also note that you may need autoconf, autoheader and
+   various other programs if you use diff files to update from one
+   snapshot to the next.
+     _________________________________________________________________
+                                      
+How to install both egcs and gcc2
+
+   It may be desirable to install both egcs and gcc2 on the same system.
+   This can be done by using different prefix paths at configure time and
+   a few symlinks.
+   
+   Basically, configure the two compilers with different --prefix
+   options, then build and install each compiler. Assume you want "gcc"
+   to be the egcs compiler and available in /usr/local/bin; also assume
+   that you want "gcc2" to be the gcc2 compiler and also available in
+   /usr/local/bin.
+   
+   The easiest way to do this is to configure egcs with
+   --prefix=/usr/local/egcs and gcc2 with --prefix=/usr/local/gcc2. Build
+   and install both compilers. Then make a symlink from
+   /usr/local/bin/gcc to /usr/local/egcs/bin/gcc and from
+   /usr/local/bin/gcc2 to /usr/local/gcc2/bin/gcc. Create similar links
+   for the "g++", "c++" and "g77" compiler drivers.
+     _________________________________________________________________
+                                      
+Problems building Linux kernels
+
+   If you installed a recent binutils/gas snapshot on your Linux system,
+   you may not be able to build the kernel because objdump does not
+   understand the "-k" switch. The solution for this problem is to remove
+   /usr/bin/encaps.
+   
+   The reason you must remove /usr/bin/encaps is because it is an
+   obsolete program that was part of older binutils distributions; the
+   Linux kernel's Makefile looks for this program to decide if you have
+   an old or a new binutils. Problems occur if you installed a new
+   binutils but haven't removed encaps, because the Makefile thinks you
+   have the old one. So zap it; trust us, you won't miss it.
+   
+   You may get an internal compiler error compiling process.c in newer
+   versions of the Linux kernel on x86 machines. This is a bug in an asm
+   statement in process.c, not a bug in egcs. XXX How to fix?!?
+   
+   You may get errors with the X driver of the form
+_X11TransSocketUNIXConnect: Can't connect: errno = 111
+
+   It's a kernel bug. The function sys_iopl in arch/i386/kernel/ioport.c
+   does an illegal hack which used to work but is now broken since GCC
+   optimizes more aggressively . The newer 2.1.x kernels already have a
+   fix which should also work in 2.0.32.
+     _________________________________________________________________
+                                      
+Virtual memory exhausted error
+
+   This error means your system ran out of memory; this can happen for
+   large files, particularly when optimizing. If you're getting this
+   error you should consider trying to simplify your files or reducing
+   the optimization level.
+   
+   Note that using -pedantic or -Wreturn-type can cause an explosion in
+   the amount of memory needed for template-heavy C++ code, such as code
+   that uses STL. Also note that -Wall includes -Wreturn-type, so if you
+   use -Wall you will need to specify -Wno-return-type to turn it off.
+     _________________________________________________________________
+                                      
+GCC can not find GAS
+
+   Some configurations like irix4, irix5, hpux* require the use of the
+   GNU assembler intead of the system assembler. To ensure that egcs
+   finds the GNU assembler, you should configure the GNU assembler with
+   the same --prefix option as you used for egcs. Then build & install
+   the GNU assembler. After the GNU assembler has been installed, proceed
+   with building egcs.
+     _________________________________________________________________
+                                      
+egcs does not work on Red Hat 5.0
+
+   This entry is obsolete with the release of egcs-1.0.1 which should
+   handle Red Hat 5.0 correctly.
+   
+   egcs-1.0 does not currently work with Red Hat 5.0 on some platforms;
+   we'll update this entry with more information as it becomes available.
+   
+   You may want to try this [42]proposed patch for Red Hat 5.0. Please
+   let us know if you use this patch and whether or not it works.
+     _________________________________________________________________
+                                      
+Unable to bootstrap on x86 Solaris 2.{5,6}
+
+   This entry is obsolete with the release of egcs-1.0.1 which should
+   handle x86 Solaris systems correctly.
+   
+   This patch should fix the problem
+Index: t-sol2
+===================================================================
+RCS file: /cvs/cvsfiles/egcs/gcc/config/i386/t-sol2,v
+retrieving revision 1.2
+diff -c -3 -p -r1.2 t-sol2
+*** t-sol2      1997/09/04 23:54:04     1.2
+--- t-sol2      1997/12/04 07:19:07
+*************** crtn.o: $(srcdir)/config/i386/sol2-cn.as
+*** 31,36 ****
+  # to produce a shared library, but since we don't know ahead of time when
+  # we will be doing that, we just always use -fPIC when compiling the
+  # routines in crtstuff.c.
+
+! CRTSTUFF_T_CFLAGS = -fPIC
+  TARGET_LIBGCC2_CFLAGS = -fPIC
+--- 31,40 ----
+  # to produce a shared library, but since we don't know ahead of time when
+  # we will be doing that, we just always use -fPIC when compiling the
+  # routines in crtstuff.c.
++ #
++ # We must also enable optimization to avoid having any code appear after
++ # the call & alignment statement, but before we switch back to the
++ # .text section.
+
+! CRTSTUFF_T_CFLAGS = -fPIC -O2
+  TARGET_LIBGCC2_CFLAGS = -fPIC
+
+     _________________________________________________________________
+                                      
+EGCS with Windows
+
+   egcs does not currently support windows, either natively or with the
+   cygwin32 dll. However Mumit Khan has been working on supporting
+   Windows with egcs. You should check out his site if you're interested
+   in Windows support. [43]GNU Win32 related projects
+     _________________________________________________________________
+                                      
+EGCS with OS/2
+
+   egcs does not currently support OS/2. However, Andrew Zabolotny has
+   been working on a generic os/2 port with pgcc. The current code code
+   can be found at [44]http://www.goof.com/pcg/os2.
+     _________________________________________________________________
+                                      
+cpp: Usage:... Error
+
+   If you get an error like this when building egcs (particularly when
+   building __mulsi3), then you likely have a problem with your
+   environment variables.
+
+cpp: Usage: /usr/lib/gcc-lib/i586-unknown-linux-gnulibc1/2.7.2.3/cpp
+[switches] input output
+
+   First look for an explicit '.' in either LIBRARY_PATH or
+   GCC_EXEC_PREFIX from your environment. If you do not find an explicit
+   '.', look for an empty pathname in those variables. Note that ':' at
+   either the start or end of these variables is an implicit '.' and will
+   cause problems.
+     _________________________________________________________________
+                                      
+EGCS will not build KDE
+
+   Previous versions of g++ accepted (as a GNU extension)
+   constructor-arguments for the objects in an array of objects
+   dynamically allocated with new. Here's an example of this construct:
+
+    struct S { S(int); }
+    void f() { new S[3](6); }
+
+   However, this construct is not allowed by the ANSI/ISO Standard, and
+   is no longer accepted by g++.
+   
+   KDE uses such constructs and therefore will not build with egcs; note
+   patches are available to fix KDE.
+     _________________________________________________________________
+                                      
+Friend Templates
+
+In order to make a specialization of a template function a friend of a
+(possibly template) class, you must explicitly state that the friend function
+is a template, by appending angle brackets to its name, and this template
+function must have been declared already. An error in the last public comment
+draft of the ANSI/ISO C++ Standard has led people to believe that was not
+necessary, but it is, and it was fixed in the final version of the Standard.
+  __________________________________________________________________________
+                                       
+Where to find libg++
+
+Many folks have been asking where to find libg++ for egcs. First we should
+point out that few programs actually need libg++; most only need
+libstdc++/libio which are included in the egcs distribution.
+
+If you do need libg++ you can get a libg++ snapshot which works with egcs from
+[45]ftp://ftp.yggdrasil.com/private/hjl/libg++-2.8.1-980119.tar.gz
+  __________________________________________________________________________
+                                       
+Why do I need autoconf/bison
+
+If you're using diffs up dated from one snapshot to the next, or if you're
+using the CVS repository, you may need autoconf, bison, or possibly other tools
+to rebuild egcs.
+
+This is necessary because neither diff nor cvs keep timestamps correct. So it
+is possible for "make" to think a generated file is out of date.
+
+If you do not have autoconf, bison, etc, then you can issue the following
+commands to touch all the generated files.
+
+
+touch `find egcs -name configure -print`
+touch egcs/gcc/c-parse.y
+touch egcs/gcc/objc/objc-parse.y
+touch egcs/gcc/{cstamp-h.in,c-gperf.h,c-parse.c,c-parse.h,cexp.c}
+touch egcs/gcc/cp/{parse.c,parse.h}
+touch egcs/gcc/objc/objc-parse.c
+
+  __________________________________________________________________________
+                                       
+EGCS does not work on AIX 4.3
+
+EGCS does not currently support AIX4.3; however, if you want to try and make it
+work with AIX 4.3 we highly recommend you get the fix for APAR IX74254 (64BIT
+DISASSEMBLED OUPUT FROM COMPILER FAILS TO ASSEMBLE/BIND) which is available
+from IBM Customer Support and IBM's service.boulder.ibm.com website.
+  __________________________________________________________________________
+                                       
+Problems debugging egcs code
+
+On some systems egcs will produce dwarf debug records by default; however the
+current gdb-4.16 release may not be able to read such debug records.
+
+You can either use the argument "-gstabs" instead of "-g" or pick up the
+current beta copy of gdb-4.17 to work around the problem.
+  __________________________________________________________________________
+                                       
+Conflicts when using cvs update
+
+It is not uncommon to get cvs conflict messages for some generated files when
+updating your local sources from the CVS repository. Typically such conflicts
+occur with bison or autoconf generated files.
+
+As long as you haven't been making modifications to the generated files or the
+generator files, it is safe to delete the offending file, then run cvs update
+again to get a new copy.
+  __________________________________________________________________________
+                                       
+[46]Return to the egcs home page
+
+Last modified: March 04, 1998
+
+References
+
+   1. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#gcc-2-diff
+   2. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#open-development
+   3. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#release-fork
+   4. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#libc-lock
+   5. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#morelibc
+   6. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#fortran
+   7. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#mips
+   8. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#x86eh
+   9. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#hpcompare
+  10. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#makebugs
+  11. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#rpath
+  12. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#rpath
+  13. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#dejagnu
+  14. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#cross
+  15. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#multiple
+  16. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#snapshot
+  17. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#linuxkernel
+  18. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#memexhausted
+  19. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#gas
+  20. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#rh5.0
+  21. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#x86solaris
+  22. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#windows
+  23. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#os2
+  24. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#environ
+  25. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#kde
+  26. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#friend
+  27. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#libg++
+  28. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#autoconf/bison++
+  29. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#aix
+  30. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#gdb
+  31. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#conflicts
+  32. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#open-development
+  33. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html#cathedral-vs-bazaar
+  34. http://locke.ccil.org/~esr/writings/cathedral.html
+  35. ftp://tsx-11.mit.edu/pub/linux/packages/GCC/binutils-2.8.1.0.20.tar.gz
+  36. ftp://tsx-11.mit.edu/pub/linux/packages/GCC/binutils-2.8.1.0.20.bin.tar.gz
+  37. ftp://tsx-11.mit.edu/pub/linux/packages/GCC/binutils-2.8.1.0.20.glibc.bin.tar.gz
+  38. ftp://egcs.cygnus.com/pub/egcs/infrastructure/gas-970915.tar.gz
+  39. ftp://egcs.cygnus.com/pub/egcs/infrastructure/dejagnu-971222.tar.gz
+  40. ftp://ftp.cygnus.com/pub/embedded/crossgcc/FAQ-0.8.1
+  41. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/index.html#mailinglists
+  42. http://www.cygnus.com/ml/egcs/1997-Dec/0594.html
+  43. http://www.xraylith.wisc.edu/~khan/software/gnu-win32
+  44. http://www.goof.com/pcg/os2
+  45. ftp://ftp.yggdrasil.com/private/hjl/libg++-2.8.1-980119.tar.gz
+  46. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/index.html

INSTALL/FINALINSTALL

@@ -0,0 +1,26 @@
+
+                          Final install egcs-1.0.3
+                                      
+   Now that egcs has been built and tested, you can install it with `cd
+   objdir; make install' for a native compiler or `cd objdir; make
+   install LANGUAGES="c c++"' for a cross compiler (note installing cross
+   compilers will be easier in the next release!).
+   
+   That step completes the installation of egcs; user level binaries can
+   be found in prefix/bin where prefix is the value you specified with
+   the --prefix to configure (or /usr/local by default).
+   
+   If you don't mind, please send egcs@cygnus.com a short mail message
+   indicating that you successfully built and installed egcs. Include the
+   output from running srcdir/config.guess.
+   
+   If you find a bug in egcs, please report it to
+   [1]egcs-bugs@cygnus.com.
+   
+     _________________________________________________________________
+                                      
+   Last modified on Jan 2, 1998.
+
+References
+
+   1. mailto:egcs-bugs@cygnus.com

INSTALL/INDEX

@@ -0,0 +1,46 @@
+
+                           Installing egcs-1.0.3
+                                      
+   This document describes the generic installation procedure for egcs as
+   well as detailing some target specific installation instructions for
+   egcs.
+   
+   egcs includes several components that previously were separate
+   distributions with their own installation instructions. This document
+   supercedes all package specific installation instructions. We provide
+   the component specific installation information in the source
+   distribution for historical reference purposes only.
+   
+   We recommend you read the entire generic installation instructions as
+   well as any target specific installation instructions before you
+   proceed to configure, build, test and install egcs.
+   
+   If something goes wrong in the configure, build, test or install
+   procedures, first double check that you followed the generic and
+   target specific installation instructions carefully. Then check the
+   [1]FAQ to see if your problem is covered before you file a bug report.
+   
+   The installation procedure is broken into four steps.
+     * [2]configure
+     * [3]build
+     * [4]test (optional)
+     * [5]install
+       
+   Before starting the build/install procedure please browse the
+   [6]host/target specific installation notes.
+     _________________________________________________________________
+                                      
+   [7]Return to the egcs home page
+     _________________________________________________________________
+                                      
+   Last modified on Jan 2, 1998.
+
+References
+
+   1. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/faq.html
+   2. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/configure.html
+   3. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/build.html
+   4. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/test.html
+   5. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/finalinstall.html
+   6. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/INSTALL/specific.html
+   7. file://localhost/.automount/hurl/root/puke/law/egcs-1.0.1/index.html

INSTALL/README

@@ -0,0 +1,14 @@
+This directory contains installation instrutions for egcs-1.00.
+
+We're providing installation instructions in two forms, html and
+plaintext.
+
+index.html is the toplevel install file for html browsers.
+
+INDEX is the toplevel install file in plaintext form.
+
+The most recent HTML installation instructions for egcs can be obtained from
+the egcs web site:
+
+http://www.cygnus.com/egcs/install
+

INSTALL/SPECIFIC

@@ -0,0 +1,122 @@
+
+           Host/Target specific installation notes for egcs-1.0.1
+                                      
+   alpha*-*-*
+   No specific installation needs/instructions.
+   
+   i?86-*-linux*
+   You will need binutils-2.8.1.0.15 or newer for exception handling to
+   work.
+   
+   i?86-*-sco3.2v5*
+   The SCO assembler is currently required. The GNU assembler is not up
+   to the task of switching between ELF and COFF at runtime.
+   Unlike various prereleases of GCC, that used '-belf' and defaulted to
+   COFF, you must now use the '-melf' and '-mcoff' flags to toggle
+   between the two object file formats. ELF is now the default.
+   Look in gcc/config/i386/sco5.h (search for "messy") for additional
+   OpenServer-specific flags.
+   
+   i?86-pc-solaris*
+   You'll need a patch to fix an egcs bug on this platform. [1]x86
+   solaris patch
+   
+   hppa*-hp-hpux*
+   We highly recommend using gas/binutils-2.8 on all hppa platforms; you
+   may encounter a variety of problems when using the HP assembler.
+   
+   hppa*-hp-hpux9
+   The HP assembler has major problems on this platform. We've tried to
+   work around the worst of the problems. However, those workarounds may
+   be causing linker crashes in some circumstances; the workarounds also
+   probably prevent shared libraries from working. Use the GNU assembler
+   to avoid these problems.
+   The configuration scripts for egcs will also trigger a bug in the
+   hpux9 shell. To avoid this problem set CONFIG_SHELL to /bin/ksh and
+   SHELL to /bin/ksh in your environment.
+   
+   hppa*-hp-hpux10
+   For hpux10.20, we highly recommend you pick up the latest sed patch
+   from HP. HP has two sites which provide patches free of charge.
+   [2]US, Canada, Asia-Pacific, and Latin-America
+   [3]Europe
+   
+   Retrieve patch PHCO_12862.
+   
+   The HP assembler on these systems is much better than the hpux9
+   assembler, but still has some problems. Most notably the assembler
+   inserts timestamps into each object file it creates, causing the
+   3-stage comparison test to fail during a "make bootstrap". You should
+   be able to continue by saying "make all" after getting the failure
+   from "make bootstrap".
+   
+   m68k-*-nextstep*
+   You absolutely must use GNU sed and GNU make on this platform.
+   
+   If you try to build the integrated C++ & C++ runtime libraries on this
+   system you will run into trouble with include files. The way to get
+   around this is to use the following sequence. Note you must have write
+   permission to prefix for this sequence to work.
+   
+   cd objdir
+   make all-texinfo all-bison all-byacc all-binutils all-gas all-ld
+   cd gcc
+   make bootstrap
+   make install-headers-tar
+   cd ..
+   make bootstrap3
+   
+   m68k-sun-sunos4.1.1
+   It is reported that you may need the GNU assembler on this platform.
+   
+   mips*-sgi-irix4
+   mips*-sgi-irix5
+   You must use GAS on these platforms, the native assembler can not
+   handle the code for exception handling support on this platform.
+   
+   These systems don't have ranlib, which various components in egcs
+   need; you should be able to avoid this problem by installing GNU
+   binutils, which includes a functional ranlib for this system.
+   
+   You may get the following warning on irix4 platforms, it can be safely
+   ignored.
+
+    warning: foo.o does not have gp tables for all its sections.
+
+   mips*-sgi-irix6
+   You must not use GAS on irix6 platforms; doing so will only cause
+   problems.
+   
+   These systems don't have ranlib, which various components in egcs
+   need; you should be able to avoid this problem by making a dummy
+   script called ranlib which just exits with zero status and placing it
+   in your path.
+   
+   rs6000-ibm-aix*
+   powerpc-ibm-aix*
+   At least one person as reported problems with older versions of
+   gnu-make on this platform. make-3.76 is reported to work correctly.
+   
+   powerpc-*-linux-gnu*
+   You will need [4]binutils-2.8.1.0.17 for a working egcs. It is
+   strongly recommended to recompile binutils with egcs if you initially
+   built it with gcc-2.7.2.*.
+   
+   sparc-unkonwn-linux-gnulibc1
+   It has been reported that you might need binutils-2.8.1.0.17 for this
+   platform too. [5]binutils-2.8.1.0.17
+   
+   exception handling
+   
+   XXX Linux stuff -k encaps stuff
+     _________________________________________________________________
+                                      
+   Last modified on Jan 2, 1998.
+
+References
+
+   1. http://www.cygnus.com/egcs/faq.html#x86solaris
+   2. http://us-support.external.hp.com/
+   3. http://europe-support.external.hp.com/
+   4. ftp://ftp.yggdrasil.com/private/hjl
+   5. ftp://ftp.yggdrasil.com/private/hjl

INSTALL/TEST

@@ -0,0 +1,34 @@
+
+                             Testing egcs-1.0.3
+                                      
+   Before you install egcs, you might wish to run the egcs testsuite;
+   this step is optional and may require you to download additional
+   software.
+   
+   First, you must have downloaded the egcs testsuites; the full
+   distribution contains testsuites. If you downloaded the "core"
+   compiler plus any front ends, then you do not have the testsuites. You
+   can download the testsuites from the same site where you downloaded
+   the core distribution and language front ends.
+   
+   Second, you must have a new version of dejagnu on your system;
+   dejagnu-1.3 will not work. We have made a [1]dejagnu snapshot
+   available in egcs.cygnus.com:/pub/egcs/infrastructure until a new
+   version of dejagnu can be released.
+   
+   Assuming you've got the testsuites unpacked and have installed an
+   appropriate dejagnu, you can run the testsuite with "cd objdir; make
+   -k check". This may take a long time. Go get some lunch.
+   
+   The testing process will try to test as many components in the egcs
+   distrubution as possible, including the C, C++ and Fortran compiler as
+   well as the C++ runtime libraries.
+   
+   How to interpret test results TBD.
+     _________________________________________________________________
+                                      
+   Last modified on Jan 2, 1998.
+
+References
+
+   1. ftp://egcs.cygnus.com/pub/egcs/infrastructure/dejagnu-971222.tar.gz

INSTALL/build.html

@@ -0,0 +1,67 @@
+<html>
+<head>
+<title>Building egcs-1.0.3 </title>
+</head>
+<body bgcolor="white">
+<h1 align="center">Building egcs-1.0.3</h1>
+
+<p>Now that egcs is configured, you are ready to build the compiler and
+runtime libraries.
+
+<p>We <b>highly</b> recommend that egcs be built using gnu-make; other
+versions make work, then again they might not.  To be safe build with gnu-make.
+
+<p><b>Building a native compiler</b>
+<p>For a native build issue the command "make bootstrap".  This will build
+the entire egcs compiler system, which includes the following steps:
+
+<ul>
+  <li> Build host tools necessary to build the compiler such as texinfo, bison,
+  gperf.<p>
+
+  <li> Build target tools for use by the compiler such as gas, gld, and
+  binutils.<p>
+
+  <li> Perform a 3-stage bootstrap of the compiler.<p>
+
+  <li> Perform a comparison test of the stage2 and stage3 compilers.<p>
+
+  <li> Build runtime libraries using the stage3 compiler from the previous
+  step.<p>
+</ul>
+
+<p>If you are short on disk space you might consider "make bootstrap-lean"
+instead.  This is identical to "make bootstrap" except that object files
+from the stage1 and stage2 of the 3-stage bootstrap of the compiler are
+deleted as soon as they are no longer needed.
+
+<p><b>Building a cross compiler</b>
+
+<p> We recommend reading the
+<a href="ftp://ftp.cygnus.com/pub/embedded/crossgcc/FAQ-0.8.1">
+crossgcc FAQ</a> for information about building cross compilers.
+
+<p>For a cross build, issue the command "make cross", which performs the
+following steps:
+<ul>
+  <li> Build host tools necessary to build the compiler such as texinfo, bison,
+  gperf.<p>
+
+  <li> Build target tools for use by the compiler such as gas, gld, and
+  binutils.<p>
+
+  <li> Build the compiler (single stage only).<p>
+
+  <li> Build runtime libraries using the compiler from the previous
+  step.<p>
+</ul>
+
+<p>Note that if an error occurs in any step the make process will exit.
+
+<p>
+<hr>
+<i>Last modified on Jan 2, 1998.</i>
+
+<!--#include virtual="/glimpsebox.html"-->
+</body>
+</html>

INSTALL/configure.html

@@ -0,0 +1,122 @@
+<html>
+<head>
+<title>Configuring egcs-1.0.3 </title>
+</head>
+<body bgcolor="white">
+<h1 align="center">Configuring egcs-1.0.3</h1>
+
+<p>Like most GNU software, egcs must be configured before it can be built.
+This document attempts to describe the recommended configuration procedure
+for both native and cross targets.
+
+<p>We use <i>srcdir</i> to refer to the toplevel source directory for
+egcs; we use <i>objdir</i> to refer to the toplevel build/object
+directory for egcs.
+
+<p>First, we <b>highly</b> recommend that egcs be built into a separate
+directory than the sources.  This is how we generally build egcs; building
+where <i>srcdir</i> == <i>objdir</i> should still work, but doesn't get
+extensive testing.
+
+<p>Second, when configuring a native system, either "cc" must be in your
+path or you must set CC in your environment before running configure.
+Otherwise the configuration scripts may fail.
+
+<p>To configure egcs:
+
+<blockquote>
+<tt>
+  <br>% mkdir <i>objdir</i>
+  <br>% cd <i>objdir</i>
+  <br>% <i>srcdir</i>/configure <b>[target]</b> <b>[options]</b>
+</tt>
+</blockquote>
+
+
+<p><b>target specification</b>
+<ul>
+  <li> egcs has code to correctly determine the correct value for
+  <b>target</b> for nearly all native systems.  Therefore, we highly
+  recommend you not provide a configure target when configuring a
+  native compiler.
+
+  <li> <b>target</b> must be specified when configuring a cross compiler;
+  examples of valid targets would be i960-rtems, m68k-coff, sh-elf, etc.
+</ul>
+
+
+<p><b> options specification</b>
+
+<p>Use <b>options</b> to override several configure time options for
+egcs.  A partial list of supported <tt>options</tt>:
+
+<ul>
+  <li> <tt>--prefix=</tt><i>dirname</i> -- Specify the toplevel installation
+  directory.  This is the recommended way to install the tools into a directory
+  other than the default.  The toplevel installation directory defaults to
+  /usr/local.
+
+  <br>These additional options control where certain parts of the distribution
+  are installed.  Normally you should not need to use these options.
+  <ul>
+     <li> <tt>--with-local-prefix=</tt><i>dirname</i> -- Specify the installation
+     directory for local include files.  The default is /usr/local.
+
+     <li> <tt>--with-gxx-include-dir=</tt><i>dirname</i> -- Specify the installation
+     directory for g++ header files.  The default is /usr/local/include/g++.
+  </ul>
+
+  <li> <tt>--enable-shared</tt> -- Build shared versions of the C++ runtime
+  libraries if supported <tt>--disable-shared</tt> is the default.
+
+  <li> <tt>--enable-haifa</tt> -- Enable the new Haifa instruction scheduler in the
+  compiler; the new scheduler can significantly improve code on some targets.
+  <tt>--disable-haifa</tt> is currently the default on all platforms except the HPPA.
+
+  <li> <tt>--with-gnu-as</tt> -- Specify that the compiler should assume the GNU
+  assembler (aka gas) is available. 
+
+  <li> <tt>--with-gnu-ld</tt> -- Specify that the compiler should assume the GNU
+  linker (aka gld) is available. 
+
+  <li> <tt>--with-stabs</tt> -- Specify that stabs debugging information should be used
+  instead of whatever format the host normally uses.  Normally GCC uses the
+  same debug format as the host system. 
+
+  <li> <tt>--enable-multilib</tt> -- Specify that multiple target libraries
+  should be built to support different target variants, calling conventions,
+  etc.  This is the default. 
+
+  <li> <tt>--enable-threads</tt> -- Specify that the target supports threads.
+  This only effects the Objective-C compiler and runtime library.
+
+  <li> <tt>--enable-threads=</tt><i>lib</i> -- Specify that <i>lib</i> is the
+  thread support library.  This only effects the Objective-C compiler  and
+  runtime library.
+
+  <li> <tt>--with-cpu=</tt><i>cpu</i> -- Specify which cpu variant the compiler should
+  generate code for by default.  This is currently only supported on the
+  RS6000/PowerPC ports.
+</ul>
+
+<p>Some options which only apply to building cross compilers:
+<ul>
+  <li> <tt>--with-headers=</tt><i>dir</i> -- Specifies a directory which has target
+  include files.
+  <li> <tt>--with-libs=</tt><i>dirs</i> -- Specifies a list of directories which contain
+  the target runtime libraries.
+  <li> <tt>--with-newlib</tt> -- Specifies that "newlib" is being used as the target
+  C library.   This causes __eprintf to be omitted from libgcc.a on the
+  assumption that it will be provided by newlib.
+</ul>
+ 
+<p>Note that each <tt>--enable</tt> option has a corresponding <tt>--disable</tt> option and
+that each <tt>--with</tt> option has a corresponding <tt>--without</tt> option.
+
+
+<p>
+<hr>
+<i>Last modified on Jan 2, 1998.</i>
+
+</body>
+</html>

INSTALL/faq.html

@@ -0,0 +1,583 @@
+<html>
+<head>
+<title>egcs Frequently Asked Questions</title>
+</head>
+<body bgcolor="white">
+<h1 align="center">egcs Frequently Asked Questions</h1>
+
+<ol>
+  <li><a href="#gcc-2-diff">How is egcs different from gcc2?</a>
+  <li><a href="#open-development">What is an open development model?</a>
+  <li><a href="#release-fork">Releases and Forking</a>
+  <li><a href="#libc-lock">bits/libc-lock.h:  No such file or directory</a>
+  <li><a href="#morelibc">`_IO_stdfile_0_lock' was not declared in this scope</a>
+  <li><a href="#fortran">Problems building the Fortran compiler</a>
+  <li><a href="#mips">Problems building on MIPS platforms</a>
+  <li><a href="#x86eh">Problems with exception handling on x86 platforms</a>
+  <li><a href="#hpcompare">Bootstrap comparison failures on HPs</a>
+  <li><a href="#makebugs">Bootstrap loops rebuilding cc1 over and over</a>
+  <li><a href="#rpath">Dynamic linker is unable to find GCC libraries</a>
+  <li><a href="#rpath">libstdc++/libio tests fail badly with --enable-shared</a>
+  <li><a href="#dejagnu">Unable to run the testsuite</a>
+  <li><a href="#cross">How to build a cross compiler</a>
+  <li><a href="#multiple">How to install both gcc2 and egcs</a>
+  <li><a href="#snapshot">Snapshots, how, when, why</a>
+  <li><a href="#linuxkernel">Problems building Linux kernels</a>
+  <li><a href="#memexhausted">Virtual memory exhausted</a>
+  <li><a href="#gas">GCC can not find GAS</a>
+  <li><a href="#rh5.0">egcs does not work on Red Hat 5.0</a>
+  <li><a href="#x86solaris">Unable to bootstrap on x86 Solaris2.{5,6}</a>
+  <li><a href="#windows">EGCS with Windows</a>
+  <li><a href="#os2">EGCS with OS/2</a>
+  <li><a href="#environ">cpp: Usage:... Error</a>
+  <li><a href="#kde">EGCS will not build KDE</a>
+  <li><a href="#friend">Friend Templates</a>
+  <li><a href="#libg++">Where to find libg++</a>
+  <li><a href="#autoconf/bison++">Why do I need autoconf & bison</a>
+  <li><a href="#aix">EGCS does not work on AIX 4.3</a>
+  <li><a href="#gdb">Problems debugging egcs code</a>
+  <li><a href="#conflicts">Conflicts when using cvs update </a>
+</ol>
+
+<hr>
+<h2><a name="gcc-2-diff">How is egcs be different from gcc2?</a></h2>
+
+<p>Six years ago, gcc version 1 had reached a point of stability.  For the
+targets it could support, it worked well.  It had limitations inherent in
+its design that would be difficult to resolve, so a major effort was made
+and gcc version 2 was the result.  When we had gcc2 in a useful state,
+development efforts on gcc1 stopped and we all concentrated on making
+gcc2 better than gcc1 could ever be.  This is the kind of step forward
+we want to make with egcs.
+
+<p>In brief, the three biggest differences between egcs and gcc2 are:  
+
+<ul>
+  <li>More rexamination of basic architectual decisions of
+    gcc and an interest in adding new optimizations;
+
+  <li>working with the groups who have fractured out from gcc2 (like
+   the Linux folks, the Intel optimizations folks, Fortran folks)
+   including more front-ends; and finally
+
+  <li>An open development model (<a
+   href="#open-development">see below</a>) for the development process.
+</ul>
+
+<p>These three differences will work together to result in a more
+useful compiler, a more stable compiler, a central compiler that works
+for more people, a compiler that generates better code.  
+
+
+<p>There are a lot of exciting compiler optimizations that have come
+out.  We want them in gcc.  There are a lot of front ends out there for
+gcc for languages like Fortran or Pascal.  We want them easily
+installable by users.  After six years of working on gcc2, we've come
+to see problems and limitations in the way gcc is architected; it is
+time to address these again.
+
+<hr>
+<h2><a name="open-development">What is an open development model?</a></h2>
+
+<p>With egcs, we are going to try a bazaar style<a
+href="#cathedral-vs-bazaar"><b>[1]</b></a> approach to its
+development:  We're going to be making snapshots publicly available
+to anyone who wants to try them; we're going to welcome anyone to join
+the development mailing list.  All of the discussions on the
+development mailing list are available via the web.  We're going to be
+making releases with a much higher frequency than they have been made
+in the past:  We're shooting for three by the end of 1997.
+
+<p>In addition to weekly snapshots of the egcs development sources, we
+are going to look at making the sources readable from a CVS server by
+anyone.  We want to make it so external maintainers of parts of egcs
+are able to commit changes to their part of egcs directly into the
+sources without going through an intermediary.
+
+<p>There have been many potential gcc developers who were not able to
+participate in gcc development in the past.  We these people to help in
+any way they can; we ultimately want gcc to be the best compiler in the
+world.
+
+<p>A compiler is a complicated piece of software, there will still be
+strong central maintainers who will reject patches, who will demand
+documentation of implementations, and who will keep the level of
+quality as high as it is today.  Code that could use wider testing may
+be intergrated--code that is simply ill-conceived won't be.
+
+<p>egcs is not the first piece of software to use this open development
+process; FreeBSD, the Emacs lisp repository, and Linux are a few
+examples of the bazaar style of development.
+
+<p>With egcs, we will be adding new features and optimizations at a
+rate that has not been done since the creation of gcc2; these additions
+will inevitably have a temporarily destabilizing effect.  With the help
+of developers working together with this bazaar style development, the
+resulting stability and quality levels will be better than we've had
+before.
+
+<blockquote>
+<a name="cathedral-vs-bazaar"><b>[1]</b></a> 
+  We've been discussing different development models a lot over the
+  past few months.  The paper which started all of this introduced two
+  terms:  A <b>cathedral</b> development model versus a <b>bazaar</b>
+  development model.  The paper is written by Eric S. Raymond, it is
+  called ``<a
+  href="http://locke.ccil.org/~esr/writings/cathedral.html">The
+  Cathedral and the Bazaar</a>''.  The paper is a useful starting point
+  for discussions.
+</blockquote>
+
+<hr>
+<h2><a name="release-fork">Releases and Forking?</a></h2>
+<p>Some folks have questioned whether or not making releases is consistent
+with the goals of the egcs project and whether or not making releases is
+a fork from gcc2.
+
+<pre>
+The egcs project has several goals, including:
+
+  * Experimenting with a new development model, release process and
+  release packaging,
+
+  * Using the new development model to accelerate development of new
+  features, optimizations, etc for future inclusion in gcc,
+
+  * Providing high quality releases to the public.
+
+An egcs release is a copy of the egcs sources that the developers have
+tested and are believed to be suitable for wider scale use and testing.
+
+Making releases of stable, tested sources is both a goal and a means by
+which we hope to achieve other goals of the egcs project.
+
+The existence of a stable tested release allows egcs to be more thoroughly
+used and tested by a wider audience than is capable of testing snapshots.
+The expanded audience provides developers with critical feedback in a
+timely manner, which is beneficial to GCC as a whole and is consistent with
+the stated goals of egcs.
+
+The gcc maintainers are encouraged to migrate tested fixes and new features
+from egcs into gcc at their discretion.  egcs maintainers are willing to
+assist the gcc maintainers as time permits.  egcs periodically merges in
+changes from gcc into the egcs sources.
+
+What will keep egcs from becoming a fork is cooperation between the
+developers of gcc and egcs.
+
+We don't see this situation as significantly different than other projects
+that make releases based on some version of the gcc sources (Cygnus, g77,
+etc).  All the code is still available for inclusion in gcc at the discretion
+of the gcc maintainers.
+</pre>
+
+<hr>
+<h2><a name="libc-lock">bits/libc-lock.h: No such file or directory</a></h2>
+<p>This entry should be obsolete, egcs should handle these beta versions of
+glibc2 correctly.
+
+<p>egcs includes a tightly integrated libio and libstdc++ implementation which
+can cause problems on hosts which have libio integrated into their C library
+(most notably Linux).
+
+<p>We believe that we've solved the major technical problems for the most
+common versions of libc found on Linux systems.  However, some versions
+of Linux use pre-release versions of glibc2, which egcs has trouble detecting
+and correctly handling.
+
+<p>If you're using one of these pre-release versions of glibc2, you may get
+a message "bits/libc-lock.h: No such file or directory" when building egcs.
+Unfortunately, to fix this problem you will need to update your C library to
+glibc2.0.5c.
+
+<hr>
+<h2><a name="morelibc">`_IO_stdfile_0_lock' was not declared in this scope</a></h2>
+<p>If you get this error, it means either egcs incorrectly guessed what version
+of libc is installed on your linux system, or you incorrectly specified a
+version of glibc when configuring egcs.
+
+<p>If you did not provide a target name when configuring egcs, then you've
+found a bug which needs to be reported.  If you did provide a target name at
+configure time, then you should reconfigure without specifying a target name.
+
+<hr>
+<h2><a name="fortran">Problems building the Fortran compiler</a></h2>
+<p>The Fortran front end can not be built with most vendor compilers; it must
+be built with gcc.  As a result, you may get an error if you do not follow
+the install instructions carefully.
+
+<p>In particular, instead of using "make" to build egcs, you should use
+"make bootstrap" if you are building a native compiler or "make cross"
+if you are building a cross compiler.
+
+<p>It has also been reported that the Fortran compiler can not be built     
+on Red Hat 4.X linux for the Alpha.  Fixing this may require upgrading
+binutils or to Red Hat 5.0; we'll provide more information as it becomes  
+available.                                                               
+
+<hr>
+<h2><a name="mips">Problems building on MIPS platforms</a></h2>
+<p>egcs requires the use of GAS on all versions of Irix, except Irix 6 due
+to limitations in older Irix assemblers.
+
+<p> Either of these messages indicates that you are using the MIPS assembler
+when instead you should be using GAS.
+
+<pre>
+    as0: Error: ./libgcc2.c, line 1:Badly delimited numeric literal
+          .4byte $LECIE1-$LSCIE1
+    as0: Error: ./libgcc2.c, line 1:malformed statement
+</pre>
+
+<hr>
+<pre>
+    as0: Error: /home/law/egcs_release/gcc/libgcc2.c, line 1:undefined symbol in expression
+    .word $LECIE1-$LSCIE1
+
+</pre>
+
+
+<p> For Irix 6, you should use the native assembler as GAS is not supported
+on Irix 6.
+
+<hr>
+<h2> <a name="x86eh">Problems with exception handling on x86 platforms</a></h2>
+<p>If you are using the GNU assembler (aka gas) on an x86 platform and
+exception handling is not working correctly, then odds are you're using a
+buggy assembler.
+
+<p>We recommend binutils-2.8.1.0.15 or newer.
+<br><a href="ftp://tsx-11.mit.edu/pub/linux/packages/GCC/binutils-2.8.1.0.20.tar.gz"> binutils-2.8.1.0.20 source</a>
+<br><a href="ftp://tsx-11.mit.edu/pub/linux/packages/GCC/binutils-2.8.1.0.20.bin.tar.gz"> binutils-2.8.1.0.20 x86 binary for libc5</a>
+<br><a href="ftp://tsx-11.mit.edu/pub/linux/packages/GCC/binutils-2.8.1.0.20.glibc.bin.tar.gz"> binutils-2.8.1.0.20 x86 binary for glibc2</a>
+Or, you can try a
+<a href="ftp://egcs.cygnus.com/pub/egcs/infrastructure/gas-970915.tar.gz"> binutils snapshot</a>; however, be aware that the binutils snapshot is untested
+and may not work (or even build).  Use it at your own risk.
+
+<hr>
+<h2> <a name="hpcompare">Bootstrap comparison failures on HPs</a></h2>
+<p>If you bootstrap the compiler on hpux10 using the HP assembler instead of
+gas, every file will fail the comparison test.
+
+<p>The HP asembler inserts timestamps into object files it creates, causing
+every file to be different.  The location of the timestamp varies for each
+object file, so there's no real way to work around this mis-feature.
+
+<p>Odds are your compiler is fine, but there's no way to be certain.
+
+<p>If you use GAS on HPs, then you will not run into this problem because
+GAS never inserts timestamps into object files.  For this and various other
+reasons we highly recommend using GAS on HPs.
+
+<hr>
+<h2> <a name="makebugs">Bootstrap loops rebuilding cc1 over and over</a></h2>
+<p>When building egcs, the build process loops rebuilding cc1 over and
+over again.  This happens on mips-sgi-irix5.2, and possibly other platforms.
+
+<p>This is probably a bug somewhere in the egcs Makefile.  Until we find and
+fix this bug we recommend you use GNU make instead of vendor supplied make
+programs.
+
+<hr>
+<h2> <a name="rpath">Dynamic linker is unable to find GCC libraries</a></h2>
+<p>This problem manifests itself by programs not finding shared libraries
+they depend on when the programs are started.  Note this problem often manifests
+itself with failures in the libio/libstdc++ tests after configuring with
+--enable-shared and building egcs.
+
+<p>GCC does not specify a runpath so that the dynamic linker can find dynamic
+libraries at runtime.
+
+<p>The short explaination is that if you always pass a -R option to the
+linker, then your programs become dependent on directories which
+may be NFS mounted, and programs may hang unnecessarily when an
+NFS server goes down.
+
+<p>The problem is not programs that do require the directories; those
+programs are going to hang no matter what you do.  The problem is
+programs that do not require the directories.
+
+<p>SunOS effectively always passed a -R option for every -L option;
+this was a bad idea, and so it was removed for Solaris.  We should
+not recreate it.
+
+<hr>
+<h2> <a name="dejagnu">Unable to run the testsuite</a></h2>
+<p>If you get a message about unable to find "standard.exp" when trying to
+run the egcs testsuites, then your dejagnu is too old to run the egcs tests.
+You will need to get a newer version of dejagnu; we've made a
+<a href="ftp://egcs.cygnus.com/pub/egcs/infrastructure/dejagnu-971222.tar.gz">
+dejagnu snapshot</a> available until a new version of dejagnu can be released.
+
+<hr>
+<h2> <a name="cross">How to build a cross compiler</a></h2>
+<p> Building cross compilers is a rather complex undertaking because they
+usually need additional software (cross assembler, cross linker, target
+libraries, target include files, etc).
+
+<p> We recommend reading the <a href="ftp://ftp.cygnus.com/pub/embedded/crossgcc/FAQ-0.8.1">
+crossgcc FAQ</a> for information about building cross compilers.
+
+<p> If you have all the pieces available, then `make cross' should build a
+cross compiler.  `make LANGUAGES="c c++" install'will install the cross
+compiler.
+
+<p> Note that if you're trying to build a cross compiler in a tree which
+includes binutils-2.8 in addition to egcs, then you're going to need to
+make a couple minor tweaks so that the cross assembler, linker and
+nm utilities will be found.
+
+<p>binutils-2.8 builds those files as gas.new, ld.new and nm.new; egcs gcc
+looks for them using gas-new, ld-new and nm-new, so you may have to arrange
+for any symlinks which point to &ltfile&gt.new to be changed to &ltfile&gt-new.
+
+<hr>
+<h2> <a name="snapshot">Snapshots, how, when, why</a></h2>
+<p> We make snapshots of the egcs sources about once a week; there is no
+predetermined schedule.  These snapshots are intended to give everyone
+access to work in progress.  Any given snapshot may generate incorrect code
+or even fail to build.
+
+<p>If you plan on downloading and using snapshots, we highly recommend you
+subscribe to the egcs mailing lists.  See <a href="index.html#mailinglists">
+mailing lists</a> on the main egcs page for instructions on how to subscribe.
+
+<p>When using the diff files to update from older snapshots to newer snapshots,
+make sure to use "-E" and "-p" arguments to patch so that empty files are
+deleted and full pathnames are provided to patch.  If your version of
+patch does not support "-E", you'll need to get a newer version.  Also note
+that you may need autoconf, autoheader and various other programs if you use
+diff files to update from one snapshot to the next.
+
+<hr>
+<h2> <a name="multiple">How to install both egcs and gcc2</a></h2>
+<p>It may be desirable to install both egcs and gcc2 on the same system.  This
+can be done by using different prefix paths at configure time and a few
+symlinks.
+
+<p>Basically, configure the two compilers with different --prefix options,
+then build and install each compiler.  Assume you want "gcc" to be the egcs
+compiler and available in /usr/local/bin; also assume that you want "gcc2"
+to be the gcc2 compiler and also available in /usr/local/bin.
+
+<p>The easiest way to do this is to configure egcs with --prefix=/usr/local/egcs
+and gcc2 with --prefix=/usr/local/gcc2.  Build and install both compilers.
+Then make a symlink from /usr/local/bin/gcc to /usr/local/egcs/bin/gcc and
+from /usr/local/bin/gcc2 to /usr/local/gcc2/bin/gcc.  Create similar links
+for the "g++", "c++" and "g77" compiler drivers.
+
+<hr>
+<h2> <a name="linuxkernel">Problems building Linux kernels</a></h2>
+<p>If you installed a recent binutils/gas snapshot on your Linux system,
+you may not be able to build the kernel because objdump does not understand
+the "-k" switch.  The solution for this problem is to remove /usr/bin/encaps.
+
+<p>The reason you must remove /usr/bin/encaps is because it is an obsolete
+program that was part of older binutils distributions; the Linux kernel's
+Makefile looks for this program to decide if you have an old or a new
+binutils.  Problems occur if you installed a new binutils but haven't
+removed encaps, because the Makefile thinks you have the old one.  So zap
+it; trust us, you won't miss it.
+
+<p>You may get an internal compiler error compiling process.c in newer
+versions of the Linux kernel on x86 machines.  This is a bug in an asm
+statement in process.c, not a bug in egcs.  XXX How to fix?!?
+
+<p>You may get errors with the X driver of the form 
+<pre>
+_X11TransSocketUNIXConnect: Can't connect: errno = 111
+</pre>
+
+<p>It's a kernel bug. The function sys_iopl in arch/i386/kernel/ioport.c
+does an illegal hack which used to work but is now broken since GCC optimizes
+more aggressively . The newer 2.1.x kernels already have a fix which should
+also work in 2.0.32.
+
+<hr>
+<h2> <a name="memexhausted">Virtual memory exhausted error</a></h2>
+<p> This error means your system ran out of memory; this can happen for large
+files, particularly when optimizing.  If you're getting this error you should
+consider trying to simplify your files or reducing the optimization level.
+
+<p>Note that using -pedantic or -Wreturn-type can cause an explosion in the
+amount of memory needed for template-heavy C++ code, such as code that uses
+STL.  Also note that -Wall includes -Wreturn-type, so if you use -Wall you
+will need to specify -Wno-return-type to turn it off.
+
+<hr>
+<h2> <a name="gas">GCC can not find GAS</a></h2>
+<p>Some configurations like irix4, irix5, hpux* require the use of the GNU
+assembler intead of the system assembler.  To ensure that egcs finds the GNU
+assembler, you should configure the GNU assembler with the same --prefix
+option as you used for egcs.  Then build & install the GNU assembler.  After
+the GNU assembler has been installed, proceed with building egcs.
+
+<hr>
+<h2> <a name="rh5.0">egcs does not work on Red Hat 5.0</a></h2>
+<p> This entry is obsolete with the release of egcs-1.0.1 which should
+handle Red Hat 5.0 correctly.
+
+<p> egcs-1.0 does not currently work with Red Hat 5.0 on some platforms; we'll
+update this entry with more information as it becomes available.
+
+<p> You may want to try this
+<a href="http://www.cygnus.com/ml/egcs/1997-Dec/0594.html"> proposed patch</a>
+for Red Hat 5.0.  Please let us know if you use this patch and whether or
+not it works.
+
+<hr>
+<h2> <a name="x86solaris">Unable to bootstrap on x86 Solaris 2.{5,6}</a></h2>
+<p> This entry is obsolete with the release of egcs-1.0.1 which should
+handle x86 Solaris systems correctly.
+
+<p>This patch should fix the problem
+
+<pre>
+Index: t-sol2
+===================================================================
+RCS file: /cvs/cvsfiles/egcs/gcc/config/i386/t-sol2,v
+retrieving revision 1.2
+diff -c -3 -p -r1.2 t-sol2
+*** t-sol2	1997/09/04 23:54:04	1.2
+--- t-sol2	1997/12/04 07:19:07
+*************** crtn.o: $(srcdir)/config/i386/sol2-cn.as
+*** 31,36 ****
+  # to produce a shared library, but since we don't know ahead of time when
+  # we will be doing that, we just always use -fPIC when compiling the
+  # routines in crtstuff.c.
+  
+! CRTSTUFF_T_CFLAGS = -fPIC
+  TARGET_LIBGCC2_CFLAGS = -fPIC
+--- 31,40 ----
+  # to produce a shared library, but since we don't know ahead of time when
+  # we will be doing that, we just always use -fPIC when compiling the
+  # routines in crtstuff.c.
++ #
++ # We must also enable optimization to avoid having any code appear after
++ # the call & alignment statement, but before we switch back to the
++ # .text section.
+  
+! CRTSTUFF_T_CFLAGS = -fPIC -O2
+  TARGET_LIBGCC2_CFLAGS = -fPIC
+</pre>
+
+<hr>
+<h2> <a name="windows">EGCS with Windows</a></h2>
+<p>egcs does not currently support windows, either natively or with the
+cygwin32 dll.  However Mumit Khan has been working on supporting Windows
+with egcs.  You should check out his site if you're interested in Windows
+support.
+<a href="http://www.xraylith.wisc.edu/~khan/software/gnu-win32">GNU Win32 related projects</a>
+
+<hr>
+<h2> <a name="os2">EGCS with OS/2</a></h2>
+<p>egcs does not currently support OS/2.  However, Andrew Zabolotny has been
+working on a generic os/2 port with pgcc.  The current code code can be found
+at <a href="http://www.goof.com/pcg/os2">http://www.goof.com/pcg/os2</a>.
+
+<hr>
+<h2> <a name="environ">cpp: Usage:... Error</a></h2>
+<p>If you get an error like this when building egcs (particularly when building
+__mulsi3), then you likely have a problem with your environment variables.
+<pre>
+cpp: Usage: /usr/lib/gcc-lib/i586-unknown-linux-gnulibc1/2.7.2.3/cpp
+[switches] input output
+</pre>
+
+<p>First look for an explicit '.' in either LIBRARY_PATH or GCC_EXEC_PREFIX
+from your environment.  If you do not find an explicit '.', look for 
+an empty pathname in those variables.  Note that ':' at either the start
+or end of these variables is an implicit '.' and will cause problems.
+
+<hr>
+<h2> <a name="kde">EGCS will not build KDE</a></h2>
+<p> Previous versions of g++ accepted (as a GNU extension)
+constructor-arguments for the objects in an array of objects
+dynamically allocated with new.  Here's an example of this construct:
+
+<pre>
+    struct S { S(int); }
+    void f() { new S[3](6); }
+</pre>
+
+<p>However, this construct is not allowed by the ANSI/ISO Standard, and
+is no longer accepted by g++.
+
+<p> KDE uses such constructs and therefore will not build with egcs; note
+patches are available to fix KDE.
+
+<hr>
+<h2> <a name="friend">Friend Templates<a></h2>
+<p>In order to make a specialization of a template function a friend of a
+(possibly template) class, you must explicitly state that the friend
+function is a template, by appending angle brackets to its name, and
+this template function must have been declared already.  An error in
+the last public comment draft of the ANSI/ISO C++ Standard has led
+people to believe that was not necessary, but it is, and it was fixed
+in the final version of the Standard.
+
+<hr>
+<h2> <a name="libg++">Where to find libg++<a></h2>
+<p>Many folks have been asking where to find libg++ for egcs.  First we
+should point out that few programs actually need libg++; most only need
+libstdc++/libio which are included in the egcs distribution.
+
+<p>If you do need libg++ you can get a libg++ snapshot  which works with egcs
+from <a href="ftp://ftp.yggdrasil.com/private/hjl/libg++-2.8.1-980119.tar.gz">
+ftp://ftp.yggdrasil.com/private/hjl/libg++-2.8.1-980119.tar.gz</a>
+
+<hr>
+<h2> <a name="autoconf/bison++">Why do I need autoconf/bison<a></h2>
+<p>If you're using diffs up dated from one snapshot to the next, or
+if you're using the CVS repository, you may need autoconf, bison, or
+possibly other tools to rebuild egcs.
+
+<p>This is necessary because neither diff nor cvs keep timestamps
+correct.  So it is possible for "make" to think a generated file is
+out of date.
+
+<p>If you do not have autoconf, bison, etc, then you can issue the
+following commands to touch all the generated files.
+
+<pre>
+touch `find egcs -name configure -print`
+touch egcs/gcc/c-parse.y
+touch egcs/gcc/objc/objc-parse.y
+touch egcs/gcc/{cstamp-h.in,c-gperf.h,c-parse.c,c-parse.h,cexp.c}
+touch egcs/gcc/cp/{parse.c,parse.h}
+touch egcs/gcc/objc/objc-parse.c
+</pre>
+
+<hr>
+<h2> <a name="aix">EGCS does not work on AIX 4.3<a></h2>
+<p>EGCS does not currently support AIX4.3; however, if you want to try
+and make it work with AIX 4.3 we highly recommend you get the
+fix for APAR IX74254 (64BIT DISASSEMBLED OUPUT FROM COMPILER FAILS TO
+ASSEMBLE/BIND) which is available from IBM Customer Support and IBM's
+service.boulder.ibm.com website.
+
+<hr>
+<h2><a name="gdb">Problems debugging egcs code</a></h2>
+<p>On some systems egcs will produce dwarf debug records by default; however
+the current gdb-4.16 release may not be able to read such debug records.
+
+<p>You can either use the argument "-gstabs" instead of "-g" or pick up
+the current beta copy of gdb-4.17 to work around the problem.
+
+<hr>
+<h2><a name="conflicts">Conflicts when using cvs update</a></h2>
+<p>It is not uncommon to get cvs conflict messages for some generated files
+when updating your local sources from the CVS repository.  Typically such
+conflicts occur with bison or autoconf generated files.
+
+<p>As long as you haven't been making modifications to the generated files
+or the generator files, it is safe to delete the offending file, then run
+cvs update again to get a new copy.
+
+<hr>
+<p><a href="index.html">Return to the egcs home page</a>
+<p><i>Last modified:  March 04, 1998</i>
+
+<!--#include virtual="/glimpsebox.html"-->
+</body>
+</html>

INSTALL/finalinstall.html

@@ -0,0 +1,30 @@
+<html>
+<head>
+<title>Final install egcs-1.0.3 </title>
+</head>
+<body bgcolor="white">
+<h1 align="center">Final install egcs-1.0.3</h1>
+
+<p>Now that egcs has been built and tested, you can install it with
+`cd <i>objdir</i>; make install' for a native compiler or
+`cd <i>objdir</i>; make install LANGUAGES="c c++"' for a cross compiler
+(note installing cross compilers will be easier in the next release!).
+
+
+<p>That step completes the installation of egcs; user level binaries can
+be found in <i>prefix</i>/bin where <i>prefix</i> is the value you specified
+with the --prefix to configure (or /usr/local by default).
+
+<p>If you don't mind, please send egcs@cygnus.com a short mail message
+indicating that you successfully built and installed egcs.  Include
+the output from running <i>srcdir</i>/config.guess.
+
+<p>If you find a bug in egcs, please report it to
+<a href="mailto:egcs-bugs@cygnus.com">egcs-bugs@cygnus.com</a>.
+
+<p>
+<hr>
+<i>Last modified on Jan 2, 1998.</i>
+
+</body>
+</html>

INSTALL/index.html

@@ -0,0 +1,47 @@
+<html>
+<head>
+<title>Installing egcs-1.0.3 </title>
+</head>
+<body bgcolor="white">
+<h1 align="center">Installing egcs-1.0.3</h1>
+
+<p>This document describes the generic installation procedure for egcs as
+well as detailing some target specific installation instructions for egcs.
+
+<p>egcs includes several components that previously were separate distributions
+with their own installation instructions.  This document supercedes all 
+package specific installation instructions.  We provide the component specific
+installation information in the source distribution for historical reference
+purposes only.
+
+<p>We recommend you read the entire generic installation instructions as
+well as any target specific installation instructions before you proceed
+to configure, build, test and install egcs.
+
+<p>If something goes wrong in the configure, build, test or install
+procedures, first double check that you followed the generic and target
+specific installation instructions carefully.  Then check the 
+<a href="faq.html">FAQ</a> to see if your problem is covered before you file
+a bug report.
+
+<p>The installation procedure is broken into four steps.
+
+<ul>
+
+  <li> <a href="configure.html">configure</a>
+  <li> <a href="build.html">build</a>
+  <li> <a href="test.html">test</a> (optional)
+  <li> <a href="finalinstall.html">install</a>
+
+</ul>
+
+<p>Before starting the build/install procedure <b>please</b> browse the
+<a href="specific.html">host/target specific installation notes</a>.
+
+<hr>
+<a href="../index.html">Return to the egcs home page</a>
+</body>
+</html>
+<hr>
+<i>Last modified on Jan 2, 1998.</i>
+

INSTALL/specific.html

@@ -0,0 +1,130 @@
+<html>
+<head>
+<title>Host/Target specific installation notes for egcs-1.0.3 </title>
+</head>
+<body bgcolor="white">
+<h1 align="center">Host/Target specific installation notes for egcs-1.0.3</h1>
+
+<p><b>alpha*-*-*</b><br>
+No specific installation needs/instructions.
+
+
+<p><b>i?86-*-linux*</b><br>
+You will need binutils-2.8.1.0.15 or newer for exception handling to work.
+
+<p><b>i?86-*-sco3.2v5*</b><br>
+The SCO assembler is currently required.    The GNU assembler is not up
+to the task of switching between ELF and COFF at runtime.
+
+<br>Unlike various prereleases of GCC, that used '-belf' and defaulted to 
+COFF, you must now use the '-melf' and '-mcoff' flags to toggle between 
+the two object file formats.     ELF is now the default.
+
+<br>Look in gcc/config/i386/sco5.h (search for "messy") for additional
+OpenServer-specific flags.
+
+<br>Systems based on OpenServer before 5.0.4 (<code>uname -X</code> will
+tell you what you're running) require TLS597 from ftp.sco.com/TLS for
+C++ constructors and destructors to work right.
+
+<p><b>i?86-pc-solaris*</b><br>
+You'll need a patch to fix an egcs bug on this platform.
+<a href="http://www-egcs.cygnus.com/faq.html#x86solaris"> x86 solaris patch</a>
+
+<p><b>hppa*-hp-hpux*</b><br>
+We <b>highly</b> recommend using gas/binutils-2.8 on all hppa platforms; you
+may encounter a variety of problems when using the HP assembler.
+
+<p><b>hppa*-hp-hpux9</b><br>
+The HP assembler has major problems on this platform.  We've tried to work
+around the worst of the problems.  However, those workarounds may be causing
+linker crashes in some circumstances; the workarounds also probably prevent
+shared libraries from working.  Use the GNU assembler to avoid these problems.
+
+<br>The configuration scripts for egcs will also trigger a bug in the hpux9
+shell.  To avoid this problem set CONFIG_SHELL to /bin/ksh and SHELL to
+/bin/ksh in your environment.
+
+<p><b>hppa*-hp-hpux10</b><br>
+For hpux10.20, we <b>highly</b> recommend you pick up the latest sed
+patch from HP.  HP has two sites which provide patches free of charge.
+
+<br><a href="http://us-support.external.hp.com">US, Canada, Asia-Pacific, and
+Latin-America</a>
+<br><a href="http://europe-support.external.hp.com">Europe</a>
+
+<p>Retrieve patch PHCO_12862.
+
+<p>The HP assembler on these systems is much better than the hpux9 assembler,
+but still has some problems.  Most notably the assembler inserts timestamps
+into each object file it creates, causing the 3-stage comparison test to fail
+during a "make bootstrap".  You should be able to continue by saying "make all"
+after getting the failure from "make bootstrap".
+
+<p><b>m68k-*-nextstep*</b><br>
+You absolutely must use GNU sed and GNU make on this platform.  
+
+<p>If you try to build the integrated C++ & C++ runtime libraries on this system
+you will run into trouble with include files.  The way to get around this is
+to use the following sequence.  Note you must have write permission to
+<i>prefix</i> for this sequence to work.
+
+<p>cd <i>objdir</i><br>
+make all-texinfo all-bison all-byacc all-binutils all-gas all-ld<br>
+cd gcc<br>
+make bootstrap<br>
+make install-headers-tar<br>
+cd ..<br>
+make bootstrap3<br>
+
+<p><b>m68k-sun-sunos4.1.1</b><br>
+It is reported that you may need the GNU assembler on this platform.
+
+<p><b>mips*-sgi-irix4</b><br>
+<b>mips*-sgi-irix5</b><br>
+You must use GAS on these platforms, the native assembler can not handle the
+code for exception handling support on this platform.
+
+<p>These systems don't have ranlib, which various components in egcs need; you
+should be able to avoid this problem by installing GNU binutils, which includes
+a functional ranlib for this system.
+
+<p>You may get the following warning on irix4 platforms, it can be safely
+ignored.
+<pre>
+    warning: foo.o does not have gp tables for all its sections.
+</pre>
+
+<p><b>mips*-sgi-irix6</b><br>
+You must not use GAS on irix6 platforms; doing so will only cause problems.
+
+<p>These systems don't have ranlib, which various components in egcs need; you
+should be able to avoid this problem by making a dummy script called ranlib
+which just exits with zero status and placing it in your path.
+
+<p><b>rs6000-ibm-aix*</b><br>
+<b>powerpc-ibm-aix*</b><br>
+At least one person as reported problems with older versions of gnu-make on
+this platform.  make-3.76 is reported to work correctly.
+
+<p><b>powerpc-*-linux-gnu*</b><br>
+You will need
+<a href="ftp://ftp.yggdrasil.com/private/hjl">binutils-2.8.1.0.20</a> for
+a working egcs. It is strongly recommended to recompile binutils with egcs
+if you initially built it with gcc-2.7.2.*.
+
+<p><b>sparc-unkonwn-linux-gnulibc1</b><br>
+It has been reported that you might need binutils-2.8.1.0.20 for this
+platform too.
+<a href="ftp://ftp.yggdrasil.com/private/hjl">binutils-2.8.1.0.20</a>
+
+<p>
+exception handling 
+<p>XXX Linux stuff
+-k encaps stuff
+<hr>
+<i>Last modified on February 7, 1998.</i>
+
+<!--#include virtual="/glimpsebox.html"-->
+</body>
+</html>

INSTALL/test.html

@@ -0,0 +1,38 @@
+<html>
+<head>
+<title>Testing egcs-1.0.3 </title>
+</head>
+<body bgcolor="white">
+<h1 align="center">Testing egcs-1.0.3</h1>
+
+<p>Before you install egcs, you might wish to run the egcs testsuite; this
+step is optional and may require you to download additional software.
+
+<p>First, you must have downloaded the egcs testsuites; the full distribution
+contains testsuites.  If you downloaded the "core" compiler plus any front
+ends, then you do not have the testsuites.  You can download the testsuites
+from the same site where you downloaded the core distribution and language
+front ends.
+
+<p>Second, you must have a new version of dejagnu on your system; dejagnu-1.3
+will not work.  We have made a 
+<a href="ftp://egcs.cygnus.com/pub/egcs/infrastructure/dejagnu-971222.tar.gz">
+dejagnu snapshot</a> available in egcs.cygnus.com:/pub/egcs/infrastructure until
+a new version of dejagnu can be released.
+
+<p>Assuming you've got the testsuites unpacked and have installed an appropriate
+dejagnu, you can run the testsuite with "cd <i>objdir</i>; make -k check".
+This may take a long time.  Go get some lunch.
+
+<p>The testing process will try to test as many components in the egcs
+distrubution as possible, including the C, C++ and Fortran compiler as
+well as the C++ runtime libraries.
+
+<p> How to interpret test results TBD.
+
+<hr>
+<i>Last modified on Jan 2, 1998.</i>
+
+<!--#include virtual="/glimpsebox.html"-->
+</body>
+</html>

Makefile.in

@@ -175,10 +175,10 @@ REALLY_SET_LIB_PATH = \
 
 ALL = all.normal
 INSTALL_TARGET = installdirs \
+	install-gcc \
 	$(INSTALL_MODULES) \
 	$(INSTALL_TARGET_MODULES) \
 	$(INSTALL_X11_MODULES) \
-	install-gcc \
 	$(INSTALL_DOSREL)
 
 
@@ -349,7 +349,8 @@ BASE_FLAGS_TO_PASS = \
 	"YACC=$(YACC)" \
 	"exec_prefix=$(exec_prefix)" \
 	"prefix=$(prefix)" \
-	"tooldir=$(tooldir)" 
+	"tooldir=$(tooldir)" \
+	"gxx_include_dir=$(gxx_include_dir)"
 
 # Flags to pass down to most sub-makes, in which we're building with
 # the host environment.

config.guess

@@ -52,6 +52,9 @@ trap 'rm -f dummy.c dummy.o dummy; exit 1' 1 2 15
 
 case "${UNAME_MACHINE}:${UNAME_SYSTEM}:${UNAME_RELEASE}:${UNAME_VERSION}" in
     alpha:OSF1:*:*)
+	if test $UNAME_RELEASE = "V4.0"; then
+		UNAME_RELEASE=`/usr/sbin/sizer -v | awk '{print $3}'`
+	fi
 	# A Vn.n version is a released version.
 	# A Tn.n version is a released field test version.
 	# A Xn.n version is an unreleased experimental baselevel.
@@ -62,9 +65,14 @@ case "${UNAME_MACHINE}:${UNAME_SYSTEM}:${UNAME_RELEASE}:${UNAME_VERSION}" in
 main:
 	.frame \$30,0,\$26,0
 	.prologue 0
-	.long 0x47e03d84
-	cmoveq \$4,0,\$3
-	addl \$3,\$31,\$0
+	.long 0x47e03d80 # implver $0
+	lda \$2,259
+	.long 0x47e20c21 # amask $2,$1
+	srl \$1,8,\$2
+	sll \$2,2,\$2
+	sll \$0,3,\$0
+	addl \$1,\$0,\$0
+	addl \$2,\$0,\$0
 	ret \$31,(\$26),1
 	.end main
 EOF
@@ -72,16 +80,25 @@ EOF
 	if test "$?" = 0 ; then
 		./dummy
 		case "$?" in
-			1)
+			7)
+				UNAME_MACHINE="alpha"
+				;;
+			15)
 				UNAME_MACHINE="alphaev5"
 				;;
-			2)
+			14)
 				UNAME_MACHINE="alphaev56"
 				;;
+			10)
+				UNAME_MACHINE="alphapca56"
+				;;
+			16)
+				UNAME_MACHINE="alphaev6"
+				;;
 		esac
 	fi
 	rm -f dummy.s dummy
-	echo ${UNAME_MACHINE}-dec-osf`echo ${UNAME_RELEASE} | sed -e 's/^[VTX]//'`
+	echo ${UNAME_MACHINE}-dec-osf`echo ${UNAME_RELEASE} | sed -e 's/^[VTX]//' | tr [[A-Z]] [[a-z]]`
 	exit 0 ;;
     21064:Windows_NT:50:3)
 	echo alpha-dec-winnt3.5
@@ -490,7 +507,36 @@ EOF
 	  i?86coff)   echo "${UNAME_MACHINE}-pc-linux-gnucoff"      ; exit 0 ;;
 	  sparclinux) echo "${UNAME_MACHINE}-unknown-linux-gnuaout" ; exit 0 ;;
 	  m68klinux)  echo "${UNAME_MACHINE}-unknown-linux-gnuaout" ; exit 0 ;;
-	  elf32ppc)   echo "powerpc-unknown-linux-gnu"              ; exit 0 ;;
+	  elf32ppc)
+		# Determine Lib Version
+		cat >dummy.c <<EOF
+#include <features.h>
+#if defined(__GLIBC__)
+extern char __libc_version[];
+extern char __libc_release[];
+#endif
+main(argc, argv)
+     int argc;
+     char *argv[];
+{
+#if defined(__GLIBC__)
+  printf("%s %s\n", __libc_version, __libc_release);
+#else
+  printf("unkown\n");
+#endif
+  return 0;
+}
+EOF
+		LIBC=""
+		${CC-cc} dummy.c -o dummy 2>/dev/null
+		if test "$?" = 0 ; then
+			./dummy | grep 1\.99 > /dev/null
+			if test "$?" = 0 ; then
+				LIBC="libc1"
+			fi
+		fi	
+		rm -f dummy.c dummy
+		echo powerpc-unknown-linux-gnu${LIBC} ; exit 0 ;;
 	esac
 
 	if test "${UNAME_MACHINE}" = "alpha" ; then
@@ -500,9 +546,14 @@ EOF
 	main:
 		.frame \$30,0,\$26,0
 		.prologue 0
-		.long 0x47e03d84
-		cmoveq \$4,0,\$3
-		addl \$3,\$31,\$0
+		.long 0x47e03d80 # implver $0
+		lda \$2,259
+		.long 0x47e20c21 # amask $2,$1
+		srl \$1,8,\$2
+		sll \$2,2,\$2
+		sll \$0,3,\$0
+		addl \$1,\$0,\$0
+		addl \$2,\$0,\$0
 		ret \$31,(\$26),1
 		.end main
 EOF
@@ -511,12 +562,21 @@ EOF
 		if test "$?" = 0 ; then
 			./dummy
 			case "$?" in
-				1)
-					UNAME_MACHINE="alphaev5"
-				;;	
-			2)	
-					UNAME_MACHINE="alphaev56"
-				;;	
+			7)
+				UNAME_MACHINE="alpha"
+				;;
+			15)
+				UNAME_MACHINE="alphaev5"
+				;;
+			14)
+				UNAME_MACHINE="alphaev56"
+				;;
+			10)
+				UNAME_MACHINE="alphapca56"
+				;;
+			16)
+				UNAME_MACHINE="alphaev6"
+				;;
 			esac	
 
 			objdump --private-headers dummy | \

configure

@@ -81,7 +81,7 @@ subdirs=
 target_alias=NOTARGET
 target_makefile_frag=
 undefs=NOUNDEFS
-version="$Revision: 1.1.1.1 $"
+version="$Revision: 1.2.2.1 $"
 x11=default
 
 ### we might need to use some other shell than /bin/sh for running subshells
@@ -1171,6 +1171,15 @@ EOF
                 echo site_makefile_frag = ${invsubdir}${site_makefile_frag} >>${Makefile}
             fi 
 
+	    # provide a proper gxx_include_dir to all subdirs.
+	    # Note, if you change the default, make sure to fix both here
+	    # and in the gcc subdirectory.
+	    if test -z "${with_gxx_include_dir}"; then
+		echo gxx_include_dir = '${prefix}/include/g++' >> ${Makefile}
+	    else
+		echo gxx_include_dir = ${with_gxx_include_dir} >> ${Makefile}
+	    fi
+
             # reset prefix, exec_prefix, srcdir, SUBDIRS, NONSUBDIRS,
             # remove any form feeds.
             if [ -z "${subdirs}" ]; then

configure.in

@@ -122,7 +122,7 @@ case "${host}" in
   mips*-*-sysv4*)	  host_makefile_frag=config/mh-sysv4 ;;
   mips*-*-sysv*)	  host_makefile_frag=config/mh-riscos ;;
   i[3456]86-*-dgux*)	  host_makefile_frag=config/mh-dgux386 ;;
-  i[3456]86-ncr-sysv4.3)  host_makefile_frag=config/mh-ncrsvr43 ;;
+  i[3456]86-ncr-sysv4.3*) host_makefile_frag=config/mh-ncrsvr43 ;;
   i[3456]86-ncr-sysv4*)   host_makefile_frag=config/mh-ncr3000 ;;
   i[3456]86-*-sco3.2v5*)  host_makefile_frag=config/mh-sysv ;;
   i[3456]86-*-sco*)       host_makefile_frag=config/mh-sco ;;
@@ -416,6 +416,12 @@ case "${target}" in
   alpha*-*-*vms*)
     noconfigdirs="$noconfigdirs gdb ld target-newlib target-libgloss"
     ;;
+  alpha*-*-linux*)
+    # newlib is not 64 bit ready
+    noconfigdirs="$noconfigdirs target-newlib target-libgloss"
+    # linux has rx in libc
+    skipdirs="$skipdirs target-librx"
+    ;;
   alpha*-*-*)
     # newlib is not 64 bit ready
     noconfigdirs="$noconfigdirs target-newlib target-libgloss"

etc/ChangeLog

@@ -1,3 +1,8 @@
+Fri Nov 21 12:54:58 1997  Manfred Hollstein <manfred@s-direktnet.de>
+
+	* Makefile.in: Add --no-split argument to avoid creating files
+	with names longer than 14 characters.
+
 Tue Oct  7 16:27:34 1997  Manfred Hollstein  <manfred@s-direktnet.de>
 
         * aclocal.m4: Substitute INSTALL.

etc/Makefile.in

@@ -60,7 +60,7 @@ install-info: info
 dvi: $(DVIFILES)
 
 standards.info: $(srcdir)/standards.texi
-	$(MAKEINFO) -I$(srcdir) -o standards.info $(srcdir)/standards.texi
+	$(MAKEINFO) --no-split -I$(srcdir) -o standards.info $(srcdir)/standards.texi
 
 standards.dvi: $(srcdir)/standards.texi
 	TEXINPUTS=$(TEXIDIR):$$TEXINPUTS $(TEXI2DVI) $(srcdir)/standards.texi

gcc/ChangeLog

@@ -1,3 +1,919 @@
+Thu Apr 23 01:41:02 1998  Jeffrey A Law  (law@cygnus.com)
+
+	* version.c: Bump for 1.0.3 pre-release.
+
+Mon Apr 20 02:17:37 1998  Richard Henderson  <rth@cygnus.com>               
+			  Jim Wilson  <wilson@cygnus.com>
+			  J"orn Rennecke <amylaar@cygnus.co.uk>
+
+	* reload1.c (eliminate_regs): Delete LOAD_EXTENDED_OP code that
+	boiled down to && ! 0.                                         
+
+	* reload.c (find_reloads): Always force (subreg (mem)) to be
+	reloaded if WORD_REGISTER_OPERATIONS.
+
+	* reload.c (find_reloads_toplev): Handle arbitrary non-paradoxical    
+	SUBREGs of CONST_INTs.                                         
+
+	* reload.c (push_reload): If WORD_REGISTER_OPERATIONS, reload the 
+	SUBREG_REG if the word count is unchanged, also in the input reload   
+	case.  Disable non-applicable sanity checks.
+
+	* reload.c (push_reload): In WORD_REGISTER_OPERATIONS code, add test
+	to require the SUBREG mode to be smaller than the SUBREG_REG mode.
+	* reload1.c (eliminate_regs): Likewise.                               
+        
+	* reload.c (push_reload): If WORD_REGISTER_OPERATIONS, reload the
+	SUBREG_REG if the word count is unchanged.           
+	* reload1.c (eliminate_regs) [case SET]: If W_R_O, preserve
+	subregs of identical word size for push_reload.                 
+
+Mon Apr 20 00:58:48 1998  H.J. Lu  (hjl@gnu.org)
+
+	* reg-stack.c (subst_asm_stack_regs): Change to return the last
+	new insn generated by this function.
+	(subst_stack_regs): Likewise.
+	(convert_regs): Record the last newly generated insn and use
+	it for change_stack () instead of INSN.
+
+Sun Apr 19 00:23:23 1998  Richard Henderson  <rth@cygnus.com>
+
+        * alpha.md (extendsidi2): Kill bogus cvtql+cvtlq case.
+
+        * alpha.h (PRINT_OPERAND_PUNCT_VALID_P): Accept '(' for s/sv/svi.
+        * alpha.c (print_operand): Handle it.
+        * alpha.md (fix_truncsfdi2): Use it.  Add earlyclobber pattern
+        for ALPHA_TP_INSN.
+        (fix_truncdfdi2): Likewise.
+
+	* alpha/linux.h (FUNCTION_PROFILER): _mcount expects its pv in $28.
+
+Sat Apr 18 19:06:59 1998  David Edelsohn  <edelsohn@mhpcc.edu>
+
+	* rs6000.md (floatsidf2_loadaddr): rs6000_fpmem_offset will be
+	negative in a stackless frame.
+	* rs6000.c (rs6000_stack_info): Don't include fixed-size link area
+	in stackless frame size.  Support 64-bit stackless frame size.
+	Combine fpmem offset calculations and don't add total_size to
+	offset if not pushing a stack frame.
+
+Sat Apr 18 17:55:57 1998  Jim Wilson  <wilson@cygnus.com>
+
+	* i386.md (fix_truncsfdi2+[123]): Add + to operand 1 constraints.
+
+Sun Mar 15 23:39:10 1998  Jeffrey A Law  (law@cygnus.com)
+
+	* version.c: Update for egcs-1.0.2 release.
+
+Sun Mar 15 23:29:25 1998  Manfred Hollstein  <manfred@s-direktnet.de>
+
+	* config/m68k/mot3300.h (ASM_BYTE_OP): Don't include '\t' in the
+	definition.
+	(ASM_OUTPUT_ASCII): Prefix ASM_BYTE_OP by one single '\t'.
+
+Sun Mar 15 23:29:18 1998  Jeffrey A Law  (law@cygnus.com)
+
+        * m68k.h: Fix mis-applied fix from the mainline branch.
+
+Wed Mar 10 14:26:52 1998  Jeffrey A Law  (law@cygnus.com)
+
+	* rtlanal.c (optimize): Declare.
+
+Sun Mar  8 11:16:23 1998  Jeffrey A Law  (law@cygnus.com)
+
+	* emit-rtl.c (gen_lowpart_common): Handle more case where converting
+	a CONST_INT into SFmode.
+
+Sat Mar  7 01:12:46 1998  Jeffrey A Law  (law@cygnus.com)
+
+	* i386/aix386ng.h (CPP_PREDEFINES): Put back -Di386.
+	* i386/freebsd-elf.h: Likewise.
+	* i386/gas.h: Likewise.
+	* i386/linux-aout.h: Likewise.
+	* i386/linux-oldld.h: Likewise.
+	* i386/linux.h: Likewise.
+	* i386/osfelf.h: Likewise.
+	* i386/osfrose.h: Likewise.
+	* i386/sco.h: Likewise.
+	* i386/sco4.h: Likewise.
+	* i386/sco4dbx.h: Likewise.
+	* i386/sco5.h: Likewise.
+	* i386/scodbx.h: Likewise.
+	* i386/sol2.h: Likewise.
+	* i386/sysv3.h: Likewise.
+
+Wed Mar  4 22:03:38 1998 Franz Sirl <franz.sirl-kernel@lauterbach.com>
+
+	* rs6000/linux.h: don't define DEFAULT_VTABLE_THUNKS to 1 if 
+	USE_GNULIBC_1 is defined
+	* configure.in: add a new case powerpc-*-linux-gnulibc1 which
+	includes the t-linux-gnulibc1 fragment
+
+Tue Mar  3 00:06:45 1998  Jeffrey A Law  (law@cygnus.com)
+
+	* version.c: Bump for 1.0.2 prerelease diffs.
+
+Sun Mar  1 18:14:46 1998  Jeffrey A Law  (law@cygnus.com)
+
+	* ginclude/va-ppc.h (va_arg): Fix typo in long long support.
+
+Tue Feb 24 09:32:24 1998  Jeffrey A Law  (law@cygnus.com)
+
+	* expr.c (emit_move_insn_1): Fix minor buglet in last change
+	caused by importint it into the release branch.
+
+Sun Feb 22 20:46:31 1998  Bernd Schmidt <crux@ohara.Informatik.RWTH-Aachen.DE>
+
+	* expr.c (emit_move_insn_1): When moving complex values in several
+	steps, emit a CLOBBER to show the destination dies.
+
+Sun Feb 22 20:06:34 1998  Jim Wilson  <wilson@cygnus.com>
+
+	* iris5.h (DWARF2_UNWIND_INFO): Define to 0.
+	* iris5gas.h (DWARF2_UNWIND_INFO): Define to 1.
+
+Thu Feb 19 09:41:23 1998  Mumit Khan <khan@xraylith.wisc.edu>
+
+	* i386/x-cygwin32 (USE_COLLECT2): Disable collect2 for now on
+	i386-{cygwin32,mingw32}.
+
+Sun Feb 22 09:45:39 1998  Kaveh R. Ghazi  <ghazi@caip.rutgers.edu>
+
+	* collect2.c (scan_prog_file): Completely cover uses of variable
+	`exports' with macro COLLECT_EXPORT_LIST.
+
+1998-02-20  Jason Merrill  <jason@yorick.cygnus.com>
+
+	* collect2.c (main): Still handle !do_collecting for non-AIX targets.
+
+Thu Feb 19 22:36:53 1998  Andrey Slepuhin  <pooh@msu.net>
+                          David Edelsohn  <edelsohn@mhpcc.edu>
+
+	* collect2.c (XCOFF_SCAN_LIBS): Remove.
+	(export_flag): New variable.
+	(export_file): #ifdef COLLECT_EXPORT_LIST.
+	(import_file, exports, imports, undefined): New variables.
+	(libs, cmdline_lib_dirs, libpath_lib_dirs, libpath, libexts): Same.
+	(dump_list, dump_prefix_list, is_in_list): New functions.
+	(write_export_file): $ifdef COLLECT_EXPORT_LIST.
+	(write_import_file, resolve_lib_name): New functions.
+	(use_import_list, ignore_library): Same.
+	(collect_exit): maybe_unlink import_file and #ifdef.
+	(handler): Same.
+	(main): New variable importf, #ifdef exportf.  Move parsing of
+	-shared before general argument parsing.  Resolve AIX library
+	paths and import libgcc.a symbols.  Treat .so shared libraries the
+	same as objects and .a libraries.  Create alias for object_lst and
+	increment it instead of original pointer.  Scan AIX libraries as
+	objects earlier instead of using scan_libraries.  Perform AIX
+	tlink later to resolve templates instead of forking ld.
+	(GCC_OK_SYMBOL): Ensure symbol not in undef section.
+	(GCC_UNDEF_SYMBOL): New macro.
+	(scan_prog_file): Loop for members of AIX libraries.  Handle
+	export/import of ctors/dtors.
+	(aix_std_libs): New variable.
+	(scan_libraries, XCOFF): Delete.
+
+1998-02-19  Mike Stump  <mrs@wrs.com>
+
+	* Makefile.in: Use $tooldir for sys-include to match toplevel
+	configure.
+
+Thu Feb 19 01:41:47 1998  Robin Kirkham <rjk@mlb.dmt.csiro.au>
+
+	* m68k.h (TARGET_SWITCHES): -mcpu32 now clears MASK_68881.
+	(MACHINE_STATE_m68010_up): Replaced __mc68332__ with __mcpu32__.
+	* m68k/m68k-none.h(CPP_FPU_SPEC): Update relative to TARGET_SWITCHES.
+	(CPP_SPEC, ASM_SPEC, CC1_SPEC): Likewise.
+	(CPP_SPEC): -m68332 defines both __mc68332 and __mcpu32__.
+	* m68k/t-m68kbare (MULTILIB_OPTIONS): Add mcpu32.
+	(MULTILIB_MATCHES): -m68332 now uses mcpu32 libraries, not m68000.
+	(MULTILIB_EXCEPTIONS): Don't build 68881 libraries for m68000,
+	mcpu32 or m5200.
+	* longlong.h: Replace __mc68332__ with __mcpu32__.
+
+Thu Feb 19 01:32:37 1998  Jeffrey A Law  (law@cygnus.com)
+			  Richard Kenner  <kenner@vlsi1.ultra.nyu.edu>
+
+	* emit-rtl.c (gen_lowpart_common): Suppress last change if __complex__.
+
+	* emit-rtl.c (hard-reg-set.h): Include.
+	(get_lowpart_common): Don't make new REG for hard reg in a 
+	class that cannot change size.
+	* Makefile.in (emit-rtl.o): Depend on hard-reg-set.h.
+
+	* combine.c: Revert previous patch.
+
+Tue Feb 17 23:34:29 1998 David Edelsohn  <edelsohn@mhpcc.edu>
+
+	* rs6000.h (MY_ISCOFF): Add numeric value of U803XTOCMAGIC.
+        * x-aix31 (INSTALL): Delete.
+
+Tue Feb 17 22:56:14 1998  Richard Henderson  <rth@cygnus.com>
+
+	* combine.c (simplify_rtx): Obey CLASS_CANNOT_CHANGE_SIZE when
+	simplifying a subreg of a hard reg.
+	(expand_compound_operation): Likewise.
+	(force_to_mode): Likewise.
+
+Sun Feb 15 21:07:48 1998  Jeffrey A Law  (law@cygnus.com)
+
+	* arm/netbsd.h (DWARF2_UNWIND_INFO): Define as zero for now.
+	* i386/netbsd.h, m68k/netbsd.h, ns32k/netbsd.h: Likewise.
+	* sparc/netbsd.h, vax/netbsd.h: Likewise.
+
+Sun Feb 15 20:55:23 1998  Bernd Schmidt <crux@ohara.Informatik.RWTH-Aachen.DE>
+
+	* i386.c (notice_update_cc): Use reg_overlap_mentioned_p.
+
+Sat Feb 14 15:54:28 1998  H.J. Lu  (hjl@gnu.org)
+
+	* config/alpha/elf.h (LIB_SPEC): New. Defined if
+	USE_GNULIBC_1 is not defined.
+
+	* config/linux.h (LIB_SPEC): Add -lc for -shared if
+	USE_GNULIBC_1 is not defined.
+	* config/sparc/linux.h: Ditto.
+
+	* config/sparc/linux64.h (LIB_SPEC): Add -lc for -shared.
+
+	* config/sparc/linux64.h (LIBGCC_SPEC): Removed.
+	(CPP_SUBTARGET_SPEC): Add %{pthread:-D_REENTRANT}.
+	(LIB_SPEC): Updated for glibc 2.
+
+Sat Nov 29 12:45:51 1997  Mumit Khan  <khan@xraylith.wisc.edu>
+
+	* config/i386/cygwin32.h (DWARF2_UNWIND): Exception handling 
+	doesn't work with it yet, so set it to 0.
+	* config/i386/x-cygwin32 (USE_COLLECT2): Delete.
+	* config/i386/xm-cygwin32.h (NO_SYS_SIGLIST): Define.
+
+Fri Feb 13 01:29:29 1998 Franz Sirl <franz.sirl-kernel@lauterbach.com>
+
+	* rs6000/sysv4.h (ENDFILE_SPEC): add missing %(endfile_linux)
+	for -mcall-linux
+
+Fri Feb 13 00:47:21 1998  Jeffrey A Law  (law@cygnus.com)
+
+	* cccp.c (new_include_prefix): Correctly handle -I./.
+
+Sun Dec 28 00:32:16 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* flow.c (find_basic_blocks): Don't create a new basic block
+	for calls in a LIBCALL block.
+
+Fri Dec 12 01:19:48 1997  Jason Merrill  <jason@yorick.cygnus.com>
+
+	* flow.c (flow_analysis): Be consistent with find_basic_blocks in
+	determining when a new basic block starts.
+
+Thu Dec 11 22:02:10 1997  Jason Merrill  <jason@yorick.cygnus.com>
+
+	* flow.c (find_basic_blocks): A CALL_INSN that can throw starts
+	a new basic block.
+	(find_basic_blocks_1): Likewise.
+
+Thu Dec  4 11:51:00 1997  Jason Merrill  <jason@yorick.cygnus.com>
+
+	* except.c (get_dynamic_handler_chain): Only make the call once per
+ 	function.
+
+Wed Dec  3 12:01:56 1997  Jason Merrill  <jason@yorick.cygnus.com>
+
+	* except.c (expand_fixup_region_end): New fn.
+	(expand_fixup_region_start): Likewise.
+	(expand_eh_region_start_tree): Store cleanup into finalization here.
+	* stmt.c (expand_cleanups): Use them to protect fixups.
+
+Mon Nov 24 22:41:55 1997  Jason Merrill  <jason@yorick.cygnus.com>
+
+	* except.c (get_dynamic_handler_chain): Build up a FUNCTION_DECL.
+	* optabs.c (init_optabs): Lose get_dynamic_handler_chain_libfunc.
+	* expr.h: Likewise.
+
+Wed Feb 11 01:39:47 1998  Richard Henderson  <rth@cygnus.com>
+
+	* stor-layout.c (layout_type): Do upper - lower in the native type,
+	so as to properly handle negative indices.
+
+Tue Feb 10 00:26:25 1998  Richard Kenner  <kenner@vlsi1.ultra.nyu.edu>
+
+	* rs6000.c (setup_incoming_varargs): Always set rs6000_sysv_varargs_p.
+
+Mon Feb  9 23:15:14 1998 Franz Sirl <franz.sirl-kernel@lauterbach.com>
+
+	* rs6000/linux.h (CPP_PREDEFINES): Add -D__ELF__.
+
+Mon Feb  9 23:08:56 1998  Richard Henderson  <rth@cygnus.com>
+
+	* i386.md (push): Prohibit symbolic constants if flag_pic.
+	(movsi+1): Likewise for move to non-register.
+
+Mon Feb  9 23:07:04 1998  John Wehle  (john@feith.com)
+
+	* i386.md: Remove redundant integer push patterns.
+	Don't bother checking for TARGET_PUSH_MEMORY when
+	pushing constants or registers.
+
+1998-01-28  Mike Stump  <mrs@wrs.com>
+ 
+	* rtlanal.c (dead_or_set_regno_p): Ignore REG_DEAD notes after
+	reload completes.
+	* genattrtab.c (reload_completed, optimize): Define.
+
+Mon Feb  9 22:16:04 1998  J"orn Rennecke <amylaar@cygnus.co.uk>
+
+	* c-decl.c (grokdeclarator): Get parameter tags from
+	last_function_parm_tags.
+	* dwarfout.c (output_formal_types): Set TREE_ASM_WRITTEN before
+	traversing the parameter types.
+	(output_type): No early exit for FUNCTION_TYPE / METHOD_TYPE context.
+
+Mon Feb  9 22:10:58 1998  Todd Vierling <tv@pobox.com>
+
+	* fixincludes: Tweak fix for struct exception in math.h
+
+Mon Feb  9 01:15:08 1998  Mark Mitchell  <mmitchell@usa.net>
+
+        * integrate.c (get_label_from_map): New function.
+        (expand_inline_function): Use it.  Initialize the label_map to
+        NULL_RTX instead of gen_label_rtx.
+        (copy_rtx_and_substitute): Use get_label_from_map.
+        * integrate.h (get_label_from_map): New function.
+        (set_label_from_map): New macro.
+        * unroll.c (unroll_loop): Use them.
+        (copy_loop_body): Ditto.
+
+        * toplev.c (rest_of_compilation): Don't call save_for_inline_copy
+        if all we're doing is dealing with -Wreturn-type.
+
+Mon Feb  9 01:07:41 1998  Jeffrey A Law  (law@cygnus.com)
+
+        * i386/x-sco5 (CC): Remove trailing whitespace.
+
+Mon Feb  9 01:07:37 1998  J. Kean Johnston  <jkj@sco.com>
+
+	* i386/sco5.h (STARTFILE_SPEC, ENDFILE_SPEC): Correctly handle
+	"-static".
+
+Mon Feb  9 00:02:53 1998  Jim Wilson  <wilson@cygnus.com>
+
+	* function.c (assign_parms): New variable named_arg, with value
+	depending on STRICT_ARGUMENT_NAMING.  Use instead of ! last_named.
+
+Mon Feb  9 00:01:00 1998  Richard Kenner  <kenner@vlsi1.ultra.nyu.edu>
+
+	* stmt.c (pushcase_range): Clean up handling of "infinite" values.
+
+	* loop.c (strength_reduce): When placing increment for auto-inc
+	case, do comparison in loop order.
+
+Fri Jan  2 23:40:09 1998  Jim Wilson  (wilson@cygnus.com)
+			  Jeffrey A Law  (law@cygnus.com)
+
+	* crtstuff.c (__frame_dummy): New function for irix6.
+	(__do_global_ctors): Call __frame_dummy for irix6.
+	* iris6.h (LINK_SPEC): Hide __frame_dummy too.
+	 
+Wed Dec 24 23:03:42 1997  Jim Wilson  (wilson@cygnus.com)
+
+	* mips.c (mips_expand_epilogue): Emit blockage insn before call to
+	save_restore_insns if no FP and GP will be restored.
+
+	* abi64.h (LONG_MAX_SPEC): Check MIPS_ABI_DEFAULT and TARGET_DEFAULT,
+	and define __LONG_MAX__ appropriately.  Add support for -mabi=X,
+	-mlong64, and -mgp{32,64} options.
+	* mips.c (mips_abi): Change type to int.
+	* mips.h (enum mips_abi_type): Delete.
+	(ABI_32, ABI_N32, ABI_64, ABI_EABI): Define as constants.
+	(mips_abi): Change type to int.
+
+Mon Dec 22 09:55:01 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* version.c: Bump for beta release of egcs-1.0.1.
+
+	* haifa-sched.c (create_reg_dead_note): Detect and handle another
+	case where we kill more regs after sched than were killed before
+	sched.
+	* sched.c (create_reg_dead_note): Similarly.
+
+Sun Dec 21 21:47:38 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* pa.c (emit_move_sequence): Handle a function label source
+	operand.
+
+Sat Dec 20 16:14:50 1997  David Edelsohn  <edelsohn@mhpcc.edu>
+
+	* rs6000.h (FUNCTION_ARG_PADDING): Define.
+	* rs6000.c (function_arg_padding): New function.
+
+Sat Dec 20 13:36:06 1997  Paul Eggert  <eggert@twinsun.com>
+                          Richard Kenner  <kenner@vlsi1.ultra.nyu.edu>
+                          H.J. Lu  (hjl@gnu.org)
+
+	* frame.h (__register_frame, __register_frame_table,
+	__deregister_frame): New.
+	* frame.c (__register_frame, __register_frame_table,
+	__deregister_frame): New.
+	* frame.c (__deregister_frame_info): Return void *.
+	* frame.h (__deregister_frame_info): Ditto.
+	* collect2.c (__deregister_frame_info): Ditto.
+
+	* frame.h (__register_frame_info_table): Fix typo in declaration.
+
+	* frame.c (__register_frame_info): Renamed from __register_frame.
+	(__register_frame_info_table, __deregister_frame_info): Similarly.
+	* frame.h (__{,de}register_frame_info): Likewise.
+	(__register_frame_info_table): New declaration.
+	* crtstuff.c (__do_global_dtors{,_aux}): Rename __deregister_frame.
+	(frame_dummy, __do_global_ctors): Likewise for __register_frame.
+	* collect2.c (write_c_file_{stat,glob}): Rename __register_frame
+	to __register_frame_info and similarly for __deregister_frame and
+	__register_frame_table.
+
+	* collect2.c (write_c_file_glob):
+	Allocate initial frame object in static storage and pass its address.
+
+	* crtstuff.c (__do_global_ctors): Fix typo in last change.
+
+	* crtstuff.c (__do_global_ctors): Add missing arg to __register_frame.
+
+	* collect2.c (write_c_file_stat): Fix error in last change;
+	use __SIZE_TYPE__, not size_t.
+
+	Alter C startup code so that it doesn't invoke malloc on Solaris.
+	* frame.h (struct object): Decl moved here from frame.c.
+	* frame.c (struct object): Move decl to frame.h.
+	("frame.h"): Include after <stddef.h>, so that size_t is defined.
+	(__register_frame, __register_frame_table, __deregister_frame):
+	It's now the caller's responsibility to allocate storage for object.
+	* crtstuff.c (frame_dummy), collect2.c (write_c_file_stat):
+	Allocate initial frame object in static storage and pass its address.
+	* crtstuff.c (<stddef.h>, "frame.h"): Include.
+	* Makefile.in ($(T)crtbegin.o, $(T)crtend.o, stamp-crtS):
+	Depend on defaults.h and frame.h.
+
+Fri Dec 19 10:02:57 1997  Richard Kenner  <kenner@vlsi1.ultra.nyu.edu>
+
+	* i386.h (INITIAL_ELIMINATION_OFFSET): Correctly test for PIC
+	register used.
+
+Fri Dec 19 09:50:30 1997  Bernd Schmidt <crux@ohara.Informatik.RWTH-Aachen.DE>
+
+	* combine.c (simplify_rtx, case ABS): Don't get confused by a
+	VOIDmode operand.
+
+	* i386.c (notice_update_cc): Remove bogus pentium GCC code.
+
+Fri Dec 19 09:39:48 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* i386/t-sol2 (CRTSTUFF_T_CFLAGS): Turn on the optimizer.
+
+	* stmt.c (warn_if_unused_value): Don't warn for TRY_CATCH_EXPR.
+
+	* stmt.c (expand_asm_operands): If an ASM has no outputs, then treat
+	it as volatile.
+
+Wed Dec  3 01:13:45 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* version.c: Update for egcs-1.0 release.
+
+Mon Dec  1 15:36:13 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* version.c: Bump for snapshot.
+
+Tue Nov 25 14:08:12 1997  Jim Wilson  <wilson@cygnus.com>
+
+	* mips.md (fix_truncdfsi2, fix_truncsfsi2, fix_truncdfdi2,
+	fix_truncsfdi2): Change *.
+
+Sun Nov 30 19:59:20 1997  Jason Merrill  <jason@yorick.cygnus.com>
+
+	* stmt.c (expand_decl_cleanup): Update thisblock after eh_region_start.
+
+Sat Nov 29 12:44:57 1997  David Edelsohn  <edelsohn@mhpcc.edu>
+
+	* rs6000.c (function_arg_partial_nregs): Undo Nov. 26 patch.
+
+	* rs6000/aix41.h (ASM_CPU_SPEC): Define.
+
+Fri Nov 28 10:00:27 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* sh/elf.h (PREFERRED_DEBUGGING_TYPE): Prefer stabs again.
+
+	* arm.md (movsfcc, movdfcc): Fix "patch" lossage.
+
+	* configure.in: Fix NCR entries.
+
+Thu Nov 27 12:20:19 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* version.c: Bump for snapshot.
+	
+	* flow.c (find_basic_blocks): Handle cfg issues for rethrows and
+	nested exceptions correctly.
+
+	* unroll.c (find_splittable_givs): Don't split givs with a dest_reg
+	that was created by loop.
+
+Thu Nov 27 09:34:58 1997  Jason Merrill  <jason@yorick.cygnus.com>
+
+	* expr.c (preexpand_calls): Don't look past a TRY_CATCH_EXPR.
+
+	* except.c (expand_start_all_catch): One more do_pending_stack_adjust.
+
+Thu Nov 27 09:32:39 1997  Richard Earnshaw  (rearnsha@arm.com)
+
+	* arm.md (movsfcc): If not TARGET_HARD_FLOAT, ensure operand[3] 
+	valid.
+
+Wed Nov 26 13:04:46 1997  Michael Meissner  <meissner@cygnus.com>
+
+	* rs6000.c (SMALL_DATA_REG): Register to use for small data relocs.
+	(print_operand): Use SMALL_DATA_REG for the register involved in
+	small data relocations.
+	(print_operand_address): Ditto.
+
+	* rs6000/linux.h (LINK_SPEC): Pass -dynamic-linker /lib/ld.so.1 if
+	-dynamic linker is not used.
+
+	* rs6000.md (call insns): For local calls, use @local suffix under
+	System V.  Don't use @plt under Solaris.
+
+	* rs6000.c (output_function_profiler): Put label address in r0, and
+	store LR in 4(sp) for System V/eabi.
+
+	* rs6000.h (ASM_OUTPUT_REG_{PUSH,POP}): Keep stack aligned to 16
+	byte boundary, and maintain stack backchain.
+
+	(patches originally from Geoffrey Keating)
+	* rs6000.c (function_arg): Excess floating point arguments don't
+	go into GPR registers after exhausting FP registers under the
+	System V.4 ABI.
+	(function_arg_partial_nregs): Ditto.
+
+	* rs6000.md (call insns): If -fPIC or -mrelocatable, add @plt
+	suffix to calls.
+
+Wed Nov 26 13:04:46 1997  Jason Merrill  <jason@yorick.cygnus.com>
+
+	* toplev.c (main): Complain about -gdwarfn.
+
+Tue Nov 25 22:43:30 1997  Jason Merrill  <jason@yorick.cygnus.com>
+
+	* dwarf2out.c (outout_call_frame_info): Ensure that the info has
+	proper alignment.
+
+	* libgcc2.c (__throw): Initialize HANDLER.
+
+	* dwarfout.c (output_type): If finalizing, write out nested types 
+	of types we've already written.
+
+Tue Nov 25 10:00:42 1997  J"orn Rennecke <amylaar@cygnus.co.uk>
+
+	* sh/elf.h (PREFERRED_DEBUGGING_TYPE): Undefine before including
+	svr4.h.
+
+	* sh/elf.h (PREFERRED_DEBUGGING_TYPE): Don't redefine.
+
+	* sh/elf.h (HANDLE_SYSV_PRAGMA): Undefine.
+
+	* sh/elf.h: (LINK_SPEC): Use shlelf.
+	(USER_LABEL_PREFIX, LOCAL_LABEL_PREFIX, ASM_FILE_START): Redefine.
+
+	* sh.h: dwarf patches from Fred Fish.
+
+	* va-sh.h (__va_arg_sh1): Define.
+	(va_arg): Use it.
+	SH3E doesn't use any integer registers for subsequent arguments
+	once a non-float value was passed in the stack.
+	* sh.h (PASS_IN_REG_P): Fix SH3E case.
+
+Tue Nov 25 10:00:42 1997  Richard Henderson  (rth@cygnus.com)
+
+	* alpha.h (CONST_OK_FOR_LETTER): Fix 'L' handling.
+
+Tue Nov 25 10:00:42 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* configure.in (alphaev56*-*-*): Disable MASK_BYTE_OPS until
+	ev56 support works correctly.
+
+	* crtstuff.c (do_global_dtors_aux): Handle multiple calls better.
+
+Sun Nov 23 13:01:48 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* version.c: Bump for snapshot.
+
+Sat Nov 22 18:58:20 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* pa-hpux10.h (NEW_HP_ASSEMBLER): Define.
+	* pa.h (LEGITIMATE_CONSTANT_P): Reject LABEL_REFs if not using
+	gas and not using the new HP assembler.
+
+Fri Nov 21 15:20:05 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* Makefile.in (program_transform_cross_name): Clean up "-e" confusion.
+	(GCC_INSTALL_NAME, GCC_CROSS_NAME): Likewise.
+
+	* i386.h (TARGET_CMOV): Disable conditional moves for this release.
+
+Fri Nov 21 12:18:51 1997  Jason Merrill  <jason@yorick.cygnus.com>
+
+	* except.h: Add outer_context_label_stack.
+	* except.c: Likewise.
+	(expand_start_all_catch): Push the outer_context for the try block
+ 	onto outer_context_label_stack.
+	(expand_end_all_catch): Use it and pop it.
+
+Fri Nov 21 10:13:11 1997  Robert Lipe (robertl@dgii.com)
+
+	* i386/sco5.h (HAVE_ATEXIT): Revert last change.
+
+Thu Nov 20 16:11:50 1997  Richard Henderson  <rth@cygnus.com>
+
+	* alpha.c (alpha_emit_set_const_1): Handle narrow hosts better.
+
+Thu Nov 20 16:11:50 1997  Klaus Kaempf  <kkaempf@progis.de>
+
+	* alpha/vms.h (ASM_OUTPUT_ADDR_VEC_ELT): Add an L for the local label
+	to correspond with the change to ASM_GENERATE_INTERNAL_LABEL.
+
+Thu Nov 20 14:42:15 1997  Jason Merrill  <jason@yorick.cygnus.com>
+
+	* except.c (output_exception_table): Don't bother with
+ 	__EXCEPTION_END__.
+
+Thu Nov 20 16:11:50 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* pa.md (pre_stwm, post_stwm, pre_ldwm, post_ldwm): Base register
+	is an in/out operand. 
+	(zero extended variants of stwm/stwm patterns): Similarly.
+
+	* mips/x-iris (FIXPROTO_DEFINES): Add -D_SGI_SOURCE.
+
+Thu Nov 20 13:19:32 1997  Jason Merrill  <jason@yorick.cygnus.com>
+
+	* dwarf2out.c (ASM_OUTPUT_DWARF_OFFSET4): Rename from VALUE4.
+	Use assemble_name.
+	(ASM_OUTPUT_DEFINE_LABEL_DIFFERENCE_SYMBOL): Use assemble_name.
+	(output_call_frame_info): Emit a \n after using it.
+
+Thu Nov 20 00:38:46 1997  Dave Love  <d.love@dl.ac.uk>
+
+	* configure.in: Add AC_ARG_ENABLE for Haifa as documentation.
+
+Wed Nov 19 12:03:04 1997  Philippe De Muyter  <phdm@macqel.be>
+
+	* dwarf2out.c (CIE_LENGTH_LABEL, FDE_LENGTH_LABEL): New macros.
+	(ASM_OUTPUT_DWARF_VALUE4): New macro.
+	(ASM_OUTPUT_DEFINE_LABEL_DIFFERENCE_SYMBOL): Define if SET_ASM_OP is
+	defined.
+	(output_call_frame_info): Do not output forward label differences
+	if ASM_OUTPUT_DEFINE_LABEL_DIFFERENCE_SYMBOL is defined.
+	* m68k/mot3300.h (SET_ASM_OP): Define when not using gas.
+
+Tue Nov 18 22:50:24 1997  Jim Wilson  <wilson@cygnus.com>
+
+	* mips/mips.c (save_restore_insns): If gp_offset or fp_offset are
+	large_int, emit two insns instead of one splitable insn.
+	* dwarf2out.c (dwarf2out_frame_debug): When set cfa_store_offset
+	from cfa_temp_value, use cfa_offset.  Abort if cfa_reg is not SP.
+
+Fri Nov  7 15:33:11 1997  Robert Lipe (robertl@dgii.com)
+
+        * i386/sco5.h (HAVE_ATEXIT): Delete definition.
+
+Sun Nov 16 23:52:48 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* cse.c (cse_insn): Don't look at JUMP_LABEL field of a conditionl
+	return.
+	(cse_end_of_basic_block): Similarly.
+	
+Sun Nov 16 23:01:40 1997  J. Kean Johnston  <jkj@sco.com>
+
+	* i386/sco5.h (ASM_OUTPUT_ALIGNED_BSS): Define.
+	(SELECT_RTX_SECTION): Define.
+	(LIBGCC_SPEC, LIB_SPEC): Do the right thing for PIC.
+
+Sun Nov 16 22:47:03 1997  Manfred Hollstein  <manfred@s-direktnet.de>
+
+	* Makefile.in (compare, compare-lean): Define $stage for each
+	shell command.
+	(gnucompare, gnucompare-lean): Likewise.
+
+Sun Nov 16 22:02:16 1997  Richard Henderson  (rth@cygnus.com)
+
+	* alpha/win-nt.h (TRAMPOLINE_TEMPLATE): Fix offsets.
+
+	* alpha.h (ASM_OUTPUT_ADDR_DIFF_ELT): Add an L for the local label
+	to correspond with the change to ASM_GENERATE_INTERNAL_LABEL.
+
+Fri Nov 14 08:01:25 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* version.c: Bump for snapshot.
+
+1997-11-14  Paul Eggert  <eggert@twinsun.com>
+
+        Fix some confusion with IEEE minus zero.
+
+        * real.h (REAL_VALUES_IDENTICAL): New macro.
+
+        * expr.c (is_zeros_p): Don't consider -0.0 to be all zeros.
+        * fold-const.c (operand_equal_p): Don't consider -0.0 to be
+        identical to 0.0.
+        * tree.c (simple_cst_equal): Don't consider -0.0 to have the
+        same tree structure as 0.0.
+
+        * varasm.c (immed_real_const_1): Use new REAL_VALUES_IDENTICAL
+        macro instead of doing it by hand.
+
+Fri Nov 14 07:24:20 1997  Richard Henderson  <rth@cygnus.com>
+
+	* expr.c (expand_builtin_setjmp): Set
+	current_function_has_nonlocal_label.
+	* stupid.c (stupid_life_analysis): If has_nonlocal_label, kill
+	call-saved registers across calls.
+
+	* alpha.md (exception_receiver): Remove.
+	(nonlocal_goto_receiver_osf): New
+	(nonlocal_goto_receiver_vms): Renamed from nonlocal_goto_receiver.
+	(nonlocal_goto_receiver): New, select _osf or _vms.
+
+	* alpha.c (output_prolog [*]): Prefix entry labels with '$' to 
+	keep them from being propogated to the object file.
+	(alpha_write_linkage): Likewise.
+	* alpha.md (call_vms): Likewise.
+	(call_value_vms): Likewise.
+	(unnamed osf call insns): Likewise.
+
+	* alpha.h (ASM_OUTPUT_INTERNAL_LABEL): Don't omit L from local label.
+	(ASM_GENERATE_INTERNAL_LABEL): Likewise.
+
+	* alpha.c (call_operand): Any reg is valid for WinNT.
+	* alpha.md (call_nt, call_value_nt): Don't force address into $27.
+	(anon nt calls): Add 'R' alternative.
+	* alpha/win-nt.h (TRAMPOLINE_TEMPLATE, TRAMPOLINE_SIZE,
+	INITIALIZE_TRAMPOLINE): Handle lack of original $27 and 32-bit ptrs.
+
+Thu Nov 13 12:53:44 1997  Jim Wilson  <wilson@cygnus.com>
+
+	* mips.h (GO_IF_LEGITIMATE_ADDRESS): Delete code swapping xplus0 and
+	xplus1 when xplus0 is not a register.
+
+Thu Nov 13 11:51:32 1997  David Edelsohn  <edelsohn@mhpcc.edu>
+
+	* rs6000.md (lshrdi3_power): Delete '&' from first alternative and
+	swap instruction order.
+
+Thu Nov 13 11:47:55 1997  Michael Meissner  <meissner@cygnus.com>
+
+	* rs6000.c (num_insns_constant): Use REAL_VALUE_FROM_CONST_DOUBLE to
+	pick apart floating point values, instead of using CONST_DOUBLE_LOW
+	and CONST_DOUBLE_HIGH.
+
+	* rs6000.md (define_splits for DF constants): Use the appropriate
+	REAL_VALUE_* interface to pick apart DF floating point constants in
+	a machine independent fashion.
+
+Thu Nov 13 11:41:42 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* flow.c (find_basic_blocks): During marking phase, if we encounter
+	an insn with a REG_LABEL note, make the target block live and
+	create an edge from the insn to the target block.  Do not make
+	edges from all blocks to the target block.
+
+	* m68k/x-next (OTHER_FIXINCLUDES_DIRS): Include /NextDeveloper/Headers.
+
+	* configure.in: Tweak NCR entries.
+	* configure: Rebuilt.
+
+Wed Nov 12 09:37:01 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* except.c: Do not include "assert.h".
+	(save_eh_status): Turn asserts into conditional aborts.
+	(restore_eh_status, scan_region): Likewise.
+	* dwarfout.c: Do not include "assert.h".
+	(bit_offset_attribute): Turn asserts into conditional aborts.
+	(bit_size_attribute, output_inlined_enumeration_type_die): Likewise.
+	(output_inlined_structure_type_die): Likewise.
+	(output_inlined_union_type_die): Likewise
+	(output_tagged_type_instantiation): Likewise.
+	(dwarfout_file_scope_decl): Likewise.
+	* dwarf2out.c: Do not include "assert.h"
+	(expand_builtin_dwarf_reg_size): Turn asserts into conditional aborts.
+	(reg_save, initial_return_save, dwarf2out_frame_debug): Likewise.
+	(add_child_die, modified_type_die, add_bit_offset_attribute): Likewise.
+	(add_bit_size_attribute, scope_die_for): Likewise.
+	(output_pending_types_for_scope): Likewise.
+	(get_inlined_enumeration_type_die): Likewise.
+	(get_inlined_structure_type_die): Likewise.
+	(get_inlined_union_type_die, gen_subprogram_die): Likewise.
+	(gen_tagged_type_instantiation_die): Likewise.
+
+	* flow.c (find_basic_blocks): Refine further to get a more correct
+	cfg, especially in the presense of exception handling, computed
+	gotos, and other non-trivial cases.  Call abort if an inaccuracy
+	is detected in the cfg.
+
+Mon Nov 10 03:02:19 1997  Jason Merrill  <jason@yorick.cygnus.com>
+
+	* stmt.c (expand_decl_cleanup_no_eh): New fn.
+
+	* except.c (expand_leftover_cleanups): do_pending_stack_adjust.
+
+	Sun Oct 19 09:07:38 1997  Richard Kenner  <kenner@vlsi1.ultra.nyu.edu>
+
+	* stmt.c (using_eh_for_cleanups_p): New variable.
+	(using_eh_for_cleanups): New function.
+	(expand_decl_cleanup): Don't call expand_eh_region_start_tree
+	unless using EH for cleanups.
+
+Mon Nov 10 00:05:56 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* alias.c (MAX_ALIAS_LOOP_PASSES): Define.
+	(init_alias_analysis): Break out of loops after MAX_ALIAS_LOOP_PASSES.
+
+Sun Nov  9 02:07:16 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* fixinc.svr4 (__STDC__): Add another case.
+
+Sun Nov  9 02:00:29 1997  J"orn Rennecke <amylaar@cygnus.co.uk>
+
+	* a29k.h (ELIGIBLE_FOR_EPILOGUE_DELAY): Avoid loads from varying
+	addresses in the epilogue delay slot.
+
+Sun Nov  9 01:40:40 1997  Manfred Hollstein  (manfred@s-direktnet.de)
+
+	* m88k/dgux.h (ASM_CPU_SPEC): Reformatted to suppress wrong whitespace
+	in generated `specs' file.
+
+Fri Nov  7 16:23:20 1997  Jim Wilson  (wilson@cygnus.com)
+
+	* flags.h (flag_rerun_loop_opt): Declare.
+	* loop.c (invariant_p, case LABEL_REF): Check flag_rerun_loop_opt.
+	* toplev.c (flag_rerum_loop_opt): Delete static.
+
+Fri Nov  7 10:22:24 1997  Jason Merrill  <jason@yorick.cygnus.com>
+
+	* frame.c (add_fdes, count_fdes): Go back to checking pc_begin for
+ 	linked once FDEs.
+
+Wed Nov  5 01:42:12 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* version.c: Bump for snapshot.
+
+	* alias.c (find_base_value): Only return the known base value for
+	pseudo registers.
+
+Tue Nov  4 16:07:15 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* fixincludes: Fix "hypot" prototype in NeXT math.h.
+
+	* Makefile.in (USE_ALLOCA): Always include alloca.o.
+	(USE_HOST_ALLOCA): Likewise.
+
+	* alias.c (find_base_value): When copying arguments, return the
+	tenative value for a hard register.
+
+Tue Nov  4 14:12:30 1997  Richard Henderson  (rth@cygnus.com)
+			  Jim Wilson  (wilson@cygnus.com)
+
+	* alpha.c (summarize_insn): Handle ASM_OPERANDS.  Don't recurse
+	for SUBREG, just fall through.
+
+	* alpha.c (alpha_handle_trap_shadows): Init sum.defd to zero.
+
+	* alpha.md (attr trap): Make TRAP_YES non-zero for sanity's sake.
+
+	* combine.c (try_combine): When setting elim_i1, check if newi2pat
+	sets i1dest.
+
+        * combine.c (try_combine): When setting elim_i2, check whether newi2pat
+        sets i2dest.  When calling distribute_notes for i3dest_killed, pass
+        elim_i2 and elim_i1.
+
+Mon Nov  3 14:36:50 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* configure.in (sco5): Use cpio to install header files.
+
+Sun Nov  2 23:30:31 1997  Manfred Hollstein  <manfred@s-direktnet.de>
+
+	* aclocal.m4 (conftestdata_from, conftestdata_to): Names shortened to
+	14 char length.
+	* configure: Rebuild.
+
+Sun Nov  2 22:53:16 1997  Richard Earnshaw  (rearnsha@arm.com)
+
+	* arm.c (load_multiple_sequence): Support SUBREG of MEM.
+	(store_multiple_sequence): Likewise.
+
+Sun Nov  2 19:44:00 1997  Robert Lipe (robertl@dgii.com)
+
+	* i386/sco5.h: enable -gstabs once again.
+
+Sun Nov  2 19:27:21 1997  Jeffrey A Law  (law@cygnus.com)
+
+	* arm.c (output_move_double): Allocate 3 entries in otherops array.
+
 Sat Nov  1 21:43:00 1997  Mike Stump  (mrs@wrs.com)
 
 	* except.c (expand_ex_region_start_for_decl): Emit EH_REGION_BEG

gcc/INSTALL

@@ -3,6 +3,9 @@ This file documents the installation of the GNU compiler.  Copyright
 may copy, distribute, and modify it freely as long as you preserve this
 copyright notice and permission notice.
 
+Note most of this information is out of date and superceded by the EGCS
+install procedures.  It is provided for historical reference only.
+
 Installing GNU CC
 *****************
 

gcc/LITERATURE

@@ -17,6 +17,10 @@ Collected papers/sites on standards, compilers, optimization, etc.
 	http://www.goof.com/pcg/docs.html
 
 
+- AMD site with optimization guide for x86
+
+	http://www.amd.com/K6/k6docs/pdf/21828a.pdf
+
 - Links related to many compiler topics
 
 	http://www.nullstone.com/htmls/connections.htm
@@ -39,3 +43,16 @@ Collected papers/sites on standards, compilers, optimization, etc.
         http://www.sgi.com/MIPS/arch/ISA5/MIPSVspec.pdf
 
 
+- IBM Journal of Research and Development
+
+  	http://www.almaden.ibm.com/journal/
+
+
+- System V PowerPC ABI 
+
+	http://www.esofta.com/softspecs.html
+
+- C9X draft
+
+	http://www.dkuug.dk/JTC1/SC22/WG14/www/docs/n794.htm
+

gcc/Makefile.in

@@ -156,7 +156,7 @@ ENQUIRE_LDFLAGS = $(LDFLAGS)
 
 # Sed command to transform gcc to installed name.  Overwritten by configure.
 program_transform_name = @program_transform_name@
-program_transform_cross_name = -e s,^,$(target_alias)-,
+program_transform_cross_name = s,^,$(target_alias)-,
 
 # Tools to use when building a cross-compiler.
 # These are used because `configure' appends `cross-make'
@@ -394,10 +394,10 @@ HOST_MALLOC=$(MALLOC)
 HOST_OBSTACK=$(OBSTACK)
 
 # Actual name to use when installing a native compiler.
-GCC_INSTALL_NAME = `t='$(program_transform_name)'; echo gcc | sed $$t`
+GCC_INSTALL_NAME = `t='$(program_transform_name)'; echo gcc | sed -e $$t`
 
 # Actual name to use when installing a cross-compiler.
-GCC_CROSS_NAME = `t='$(program_transform_cross_name)'; echo gcc | sed $$t`
+GCC_CROSS_NAME = `t='$(program_transform_cross_name)'; echo gcc | sed -e $$t`
 
 # Choose the real default target.
 ALL=all.internal
@@ -468,8 +468,8 @@ ALL_CFLAGS = $(INTERNAL_CFLAGS) $(X_CFLAGS) $(T_CFLAGS) $(CFLAGS) $(XCFLAGS) \
 ALL_CPPFLAGS = $(CPPFLAGS) $(X_CPPFLAGS) $(T_CPPFLAGS)
 
 # Even if ALLOCA is set, don't use it if compiling with GCC.
-USE_ALLOCA= ` case "${CC}" in "${OLDCC}") echo "${ALLOCA}" ;; esac `
-USE_HOST_ALLOCA= ` case "${HOST_CC}"@"${HOST_ALLOCA}" in "${OLDCC}"@?*) echo ${HOST_PREFIX}${HOST_ALLOCA} ;; esac `
+USE_ALLOCA= ${ALLOCA}
+USE_HOST_ALLOCA= ` case "${HOST_ALLOCA}" in ?*) echo ${HOST_PREFIX}${HOST_ALLOCA} ;; esac `
 USE_HOST_MALLOC= ` case "${HOST_MALLOC}" in ?*) echo ${HOST_PREFIX}${HOST_MALLOC} ;; esac `
 USE_HOST_OBSTACK= ` case "${HOST_OBSTACK}" in ?*) echo ${HOST_PREFIX}${HOST_OBSTACK} ;; esac `
 
@@ -1304,7 +1304,8 @@ dwarf2out.o : dwarf2out.c $(CONFIG_H) $(TREE_H) $(RTL_H) dwarf2.h flags.h \
 xcoffout.o : xcoffout.c $(CONFIG_H) $(TREE_H) $(RTL_H) xcoffout.h flags.h
 emit-rtl.o : emit-rtl.c $(CONFIG_H) $(RTL_H) $(TREE_H) flags.h \
    function.h regs.h insn-config.h insn-codes.h real.h expr.h bytecode.h \
-   bc-opcode.h bc-typecd.h bc-typecd.def bc-optab.h bc-emit.h bc-opname.h
+   bc-opcode.h bc-typecd.h bc-typecd.def bc-optab.h bc-emit.h bc-opname.h \
+   hard-reg-set.h
 real.o : real.c $(CONFIG_H) $(TREE_H)
 getpwd.o : getpwd.c $(CONFIG_H)
 
@@ -1738,7 +1739,7 @@ cccp.o: cccp.c $(CONFIG_H) pcp.h version.c config.status
 	  -DGPLUSPLUS_INCLUDE_DIR=\"$(gxx_include_dir)\" \
 	  -DOLD_GPLUSPLUS_INCLUDE_DIR=\"$(old_gxx_include_dir)\" \
 	  -DLOCAL_INCLUDE_DIR=\"$(includedir)\" \
-	  -DCROSS_INCLUDE_DIR=\"$(libsubdir)/sys-include\" \
+	  -DCROSS_INCLUDE_DIR=\"$(tooldir)/sys-include\" \
 	  -DTOOL_INCLUDE_DIR=\"$(tooldir)/include\" \
 	  -c `echo $(srcdir)/cccp.c | sed 's,^\./,,'`
 
@@ -1755,7 +1756,7 @@ cpplib.o: cpplib.c $(CONFIG_H) cpplib.h cpphash.h config.status
 	  -DGPLUSPLUS_INCLUDE_DIR=\"$(gxx_include_dir)\" \
 	  -DOLD_GPLUSPLUS_INCLUDE_DIR=\"$(old_gxx_include_dir)\" \
 	  -DLOCAL_INCLUDE_DIR=\"$(includedir)\" \
-	  -DCROSS_INCLUDE_DIR=\"$(libsubdir)/sys-include\" \
+	  -DCROSS_INCLUDE_DIR=\"$(tooldir)/sys-include\" \
 	  -DTOOL_INCLUDE_DIR=\"$(tooldir)/include\" \
 	  -c `echo $(srcdir)/cpplib.c | sed 's,^\./,,'`
 
@@ -1790,7 +1791,7 @@ stamp-proto: protoize.c getopt.h $(CONFIG_H)
 	$(CC) -c $(ALL_CFLAGS) $(ALL_CPPFLAGS) $(INCLUDES) \
           -DGCC_INCLUDE_DIR=\"$(libsubdir)/include\" \
 	  -DGPLUSPLUS_INCLUDE_DIR=\"$(gxx_include_dir)\" \
-          -DCROSS_INCLUDE_DIR=\"$(libsubdir)/sys-include\" \
+          -DCROSS_INCLUDE_DIR=\"$(tooldir)/sys-include\" \
 	  -DTOOL_INCLUDE_DIR=\"$(tooldir)/include\" \
 	  -DLOCAL_INCLUDE_DIR=\"$(includedir)\" \
 	  -DSTD_PROTO_DIR=\"$(libsubdir)\" \
@@ -1799,7 +1800,7 @@ stamp-proto: protoize.c getopt.h $(CONFIG_H)
 	$(CC) -c $(ALL_CFLAGS) $(ALL_CPPFLAGS) $(INCLUDES) \
           -DGCC_INCLUDE_DIR=\"$(libsubdir)/include\" \
 	  -DGPLUSPLUS_INCLUDE_DIR=\"$(gxx_include_dir)\" \
-          -DCROSS_INCLUDE_DIR=\"$(libsubdir)/sys-include\" \
+          -DCROSS_INCLUDE_DIR=\"$(tooldir)/sys-include\" \
 	  -DTOOL_INCLUDE_DIR=\"$(tooldir)/include\" \
 	  -DLOCAL_INCLUDE_DIR=\"$(includedir)\" \
 	  -DSTD_PROTO_DIR=\"$(libsubdir)\" \
@@ -2624,6 +2625,7 @@ compare compare3 compare4 compare-lean compare3-lean compare4-lean: force
 	  fi; \
 	done
 	-rm -f tmp-foo*
+	case "$@" in compare | compare-lean ) stage=2 ;; * ) stage=`echo $@ | sed -e 's,^compare\([0-9][0-9]*\).*,\1,'` ;; esac; \
 	if [ -f .bad_compare ]; then \
 	  echo "Bootstrap comparison failure!"; \
 	  cat .bad_compare; \
@@ -2652,6 +2654,7 @@ gnucompare gnucompare3 gnucompare4 gnucompare-lean gnucompare3-lean gnucompare4-
 	    done; \
 	  fi; \
 	done
+	case "$@" in gnucompare | gnucompare-lean ) stage=2 ;; * ) stage=`echo $@ | sed -e 's,^gnucompare\([0-9][0-9]*\).*,\1,'` ;; esac; \
 	if [ -f .bad_compare ]; then \
 	  echo "Bootstrap comparison failure!"; \
 	  cat .bad_compare; \

gcc/NEWS

@@ -1,4 +1,4 @@
-Noteworthy changes in GCC version 2.8.0:
+Noteworthy changes in GCC for EGCS.
 
 The -specs=file switch allows you to override default specs used in invoking
 programs like cc1, as, etc.

gcc/README

@@ -1,4 +1,4 @@
-This directory contains the version 2.7.2 release of the GNU C
+This directory contains the egcs version 1.0.3 release of the GNU C
 compiler.  It includes all of the support for compiling C++ and
 Objective C, including a run-time library for Objective C.
 

gcc/README.RS6000

@@ -1,3 +1,80 @@
+			    AIX 4.3 assembler
+
+The AIX 4.3.0.0 assembler generates incorrect object files if the ".bs"
+pseudo-op references symbols in certain sections.  If GCC is invoked with
+the -g debugging option (including during bootstrapping), incorrect object
+files will be produced and the AIX linker will fail with a severe error.
+A fix for APAR IX74254 (64BIT DISASSEMBLED OUPUT FROM COMPILER FAILS TO
+ASSEMBLE/BIND) is available from IBM Customer Support and IBM's
+service.boulder.ibm.com website.
+
+
+			      AIX 4.1 binder
+
+Due to changes in the way that GCC invokes the binder (linker) for AIX 4.1,
+the link step now may produce warnings of duplicate symbols which were not
+reported before.  The assembly files generated by GCC for AIX always have
+included multiple symbol definitions for certain global variable and
+function declarations in the original program.  The warnings should not
+prevent the linker from producing a correct library or runnable executable.
+
+
+			     AIX NLS problems
+
+AIX on the RS/6000 provides support (NLS) for environments outside of
+the United States.  Compilers and assemblers use NLS to support
+locale-specific representations of various objects including
+floating-point numbers ("." vs "," for separating decimal fractions).
+There have been problems reported where the library linked with GCC does
+not produce the same floating-point formats that the assembler accepts.
+If you have this problem, set the LANG environment variable to "C" or
+"En_US".
+
+				     
+			 AIX 3.2.5 XLC-1.3 problems
+
+XLC version 1.3.0.0 distributed with AIX 3.2.5 will miscompile jump.c when
+building the stage1 compiler during the bootstrap process.  This will cause
+GCC to crash and the bootstrap to fail later while compiling libgcc2.c.  XLC
+version 1.3.0.1 or later fixes this problem.  XLC-1.3.0.19 also cannot
+bootstrap GCC so please avoid that release as well.  You can obtain
+XLC-1.3.0.24 by requesting PTF 432238 from IBM, or just ask for the latest
+release of XLC-1.3.
+
+There also have been reports of problems bootstrapping GCC with some older
+releases of xlc-1.2.1, including xlc-1.2.1.8.  Newer releases of xlc-1.2.1
+do not exhibit this problem: xlc-1.2.1.28 is known to bootstrap properly.
+
+
+		       AIX 3.2 common-mode support
+
+AIX common-mode providing transparent support of both the POWER and PowerPC
+architectures is usable in AIX 3.2.3 and above but an export file and
+support for hidden export via libc.a will not exist until AIX 4.1.  libgcc.a
+also must be compiled in common-mode.  Note that executables generated for
+the POWER (RIOS1 and RSC) architecture will run directly on systems using
+the MPC601 chip.  Common-mode only improves the performance of a single
+executable run on both POWER and PowerPC architecture platforms by not using
+POWER- or PowerPC-specific instructions and eliminating the need to trap to
+emulation (for POWER instructions run on PowerPC).
+
+To link a common-mode application prior to AIX 4.1 and run it on a system at
+AIX level 3.2.3 or above, use the text between the "<>" as an export file
+(e.g. milli.exp)
+
+<><><><><><><><><><><>
+#!
+__mulh          0x3100
+__mull          0x3180
+__divss         0x3200
+__divus         0x3280
+__quoss         0x3300
+__quous         0x3380
+<><><><><><><><><><><>
+
+and then link with -Wl,-bI:milli.exp.
+
+
 		     AIX 3.1 and 3.2 assembler problems
 
 Specifying the -g flag to GCC on the RS/6000 requires upgrading the
@@ -44,68 +121,3 @@ compiler.  This is because the GNU C compiler wants to build a variant of its
 library, libgcc.a with the -mcpu=common switch to support building programs
 that can run on either the Power or PowerPC machines.
 
-
-			      AIX NLS problems
-
-AIX on the RS/6000 provides support (NLS) for environments outside of
-the United States.  Compilers and assemblers use NLS to support
-locale-specific representations of various objects including
-floating-point numbers ("." vs "," for separating decimal fractions).
-There have been problems reported where the library linked with GCC does
-not produce the same floating-point formats that the assembler accepts.
-If you have this problem, set the LANG environment variable to "C" or
-"En_US".
-
-
-			 AIX 3.2.5 XLC-1.3 problems
-
-XLC version 1.3.0.0 distributed with AIX 3.2.5 will miscompile jump.c when
-building the stage1 compiler during the bootstrap process.  This will cause
-GCC to crash and the bootstrap to fail later while compiling libgcc2.c.  XLC
-version 1.3.0.1 or later fixes this problem.  XLC-1.3.0.19 also cannot
-bootstrap GCC so please avoid that release as well.  You can obtain
-XLC-1.3.0.24 by requesting PTF 432238 from IBM, or just ask for the latest
-release of XLC-1.3.
-
-There also have been reports of problems bootstrapping GCC with some older
-releases of xlc-1.2.1, including xlc-1.2.1.8.  Newer releases of xlc-1.2.1
-do not exhibit this problem: xlc-1.2.1.28 is known to bootstrap properly.
-
-
-			AIX 3.2 common-mode support
-
-AIX common-mode providing transparent support of both the POWER and PowerPC
-architectures is usable in AIX 3.2.3 and above but an export file and
-support for hidden export via libc.a will not exist until AIX 4.1.  libgcc.a
-also must be compiled in common-mode.  Note that executables generated for
-the POWER (RIOS1 and RSC) architecture will run directly on systems using
-the MPC601 chip.  Common-mode only improves the performance of a single
-executable run on both POWER and PowerPC architecture platforms by not using
-POWER- or PowerPC-specific instructions and eliminating the need to trap to
-emulation (for POWER instructions run on PowerPC).
-
-To link a common-mode application prior to AIX 4.1 and run it on a system at
-AIX level 3.2.3 or above, use the text between the "<>" as an export file
-(e.g. milli.exp)
-
-<><><><><><><><><><><>
-#!
-__mulh          0x3100
-__mull          0x3180
-__divss         0x3200
-__divus         0x3280
-__quoss         0x3300
-__quous         0x3380
-<><><><><><><><><><><>
-
-and then link with -Wl,-bI:milli.exp.
-
-
-			       AIX 4.1 binder
-
-Due to changes in the way that GCC invokes the binder (linker) for AIX 4.1,
-the link step now may produce warnings of duplicate symbols which were not
-reported before.  The assembly files generated by GCC for AIX always have
-included multiple symbol definitions for certain global variable and
-function declarations in the original program.  The warnings should not
-prevent the linker from producing a correct library or runnable executable.

gcc/README.g77

@@ -14,6 +14,10 @@ which is the top-level directory containing the gcc back end, the
 gcc C front end, and other non-Fortran files, and gcc/f/, which
 contains all of the Fortran files.
 
+* Note, if this is an egcs release, all the installation information
+  which follows is not needed.  It is provided for historical reference
+  only.
+
 * To build GNU Fortran, you must have a source distribution of gcc
   version 2.7.2.2.  Do not attempt to use any other version
   of gcc, because this version of g77 is designed to work only with
@@ -163,4 +167,4 @@ the compiler is in the second pass, otherwise it is in the first.
 (This information might help you reduce a test case and/or work around
 a bug in g77 until a fix is available.)
 
-Any questions or comments on these topics, email fortran@gnu.ai.mit.edu.
+Any questions or comments on these topics, email egcs@cygnus.com

gcc/aclocal.m4

@@ -30,20 +30,20 @@ dnl See if symbolic links work and if not, try to substitute either hard links o
 AC_DEFUN(GCC_PROG_LN_S,
 [AC_MSG_CHECKING(whether ln -s works)
 AC_CACHE_VAL(gcc_cv_prog_LN_S,
-[rm -f conftestdata_to
-echo >conftestdata_from
-if ln -s conftestdata_from conftestdata_to 2>/dev/null
+[rm -f conftestdata_t
+echo >conftestdata_f
+if ln -s conftestdata_f conftestdata_t 2>/dev/null
 then
   gcc_cv_prog_LN_S="ln -s"
 else
-  if ln conftestdata_from conftestdata_to 2>/dev/null
+  if ln conftestdata_f conftestdata_t 2>/dev/null
   then
     gcc_cv_prog_LN_S=ln
   else
     gcc_cv_prog_LN_S=cp
   fi
 fi
-rm -f conftestdata_from conftestdata_to
+rm -f conftestdata_f conftestdata_t
 ])dnl
 LN_S="$gcc_cv_prog_LN_S"
 if test "$gcc_cv_prog_LN_S" = "ln -s"; then
@@ -62,20 +62,20 @@ dnl See if hard links work and if not, try to substitute either symbolic links o
 AC_DEFUN(GCC_PROG_LN,
 [AC_MSG_CHECKING(whether ln works)
 AC_CACHE_VAL(gcc_cv_prog_LN,
-[rm -f conftestdata_to
-echo >conftestdata_from
-if ln conftestdata_from conftestdata_to 2>/dev/null
+[rm -f conftestdata_t
+echo >conftestdata_f
+if ln conftestdata_f conftestdata_t 2>/dev/null
 then
   gcc_cv_prog_LN="ln"
 else
-  if ln -s conftestdata_from conftestdata_to 2>/dev/null
+  if ln -s conftestdata_f conftestdata_t 2>/dev/null
   then
     gcc_cv_prog_LN="ln -s"
   else
     gcc_cv_prog_LN=cp
   fi
 fi
-rm -f conftestdata_from conftestdata_to
+rm -f conftestdata_f conftestdata_t
 ])dnl
 LN="$gcc_cv_prog_LN"
 if test "$gcc_cv_prog_LN" = "ln"; then

gcc/alias.c

@@ -36,6 +36,12 @@ static int memrefs_conflict_p		PROTO((int, rtx, int, rtx,
 
 #define SIZE_FOR_MODE(X) (GET_MODE_SIZE (GET_MODE (X)))
 
+/* Cap the number of passes we make over the insns propagating alias
+   information through set chains.
+
+   10 is a completely arbitrary choice.  */
+#define MAX_ALIAS_LOOP_PASSES 10
+   
 /* reg_base_value[N] gives an address to which register N is related.
    If all sets after the first add or subtract to the current value
    or otherwise modify it so it does not point to a different top level
@@ -95,16 +101,23 @@ find_base_value (src)
       return src;
 
     case REG:
-      /* If this REG is related to a known base value, return it.  */
-      if (reg_base_value[REGNO (src)])
-	return reg_base_value[REGNO (src)];
-
       /* At the start of a function argument registers have known base
 	 values which may be lost later.  Returning an ADDRESS
 	 expression here allows optimization based on argument values
 	 even when the argument registers are used for other purposes.  */
       if (REGNO (src) < FIRST_PSEUDO_REGISTER && copying_arguments)
 	return new_reg_base_value[REGNO (src)];
+
+      /* If a pseudo has a known base value, return it.  Do not do this
+	 for hard regs since it can result in a circular dependency
+	 chain for registers which have values at function entry.
+
+	 The test above is not sufficient because the scheduler may move
+	 a copy out of an arg reg past the NOTE_INSN_FUNCTION_BEGIN.  */
+      if (REGNO (src) >= FIRST_PSEUDO_REGISTER
+	  && reg_base_value[REGNO (src)])
+	return reg_base_value[REGNO (src)];
+
       return src;
 
     case MEM:
@@ -955,7 +968,7 @@ void
 init_alias_analysis ()
 {
   int maxreg = max_reg_num ();
-  int changed;
+  int changed, pass;
   register int i;
   register rtx insn;
   rtx note;
@@ -995,10 +1008,21 @@ init_alias_analysis ()
 
      We could propagate more information in the first pass by making use
      of REG_N_SETS to determine immediately that the alias information
-     for a pseudo is "constant".  */
+     for a pseudo is "constant".
+
+     A program with an uninitialized variable can cause an infinite loop
+     here.  Instead of doing a full dataflow analysis to detect such problems
+     we just cap the number of iterations for the loop.
+
+     The state of the arrays for the set chain in question does not matter
+     since the program has undefined behavior.  */
   changed = 1;
-  while (changed)
+  pass = 0;
+  while (changed && pass < MAX_ALIAS_LOOP_PASSES)
     {
+      /* Keep track of the pass number so we can break out of the loop.  */
+      pass++;
+
       /* Assume nothing will change this iteration of the loop.  */
       changed = 0;
 
@@ -1123,10 +1147,15 @@ init_alias_analysis ()
 
      In theory this loop can take as long as O(registers^2), but unless
      there are very long dependency chains it will run in close to linear
-     time.  */
+     time.
+
+     This loop may not be needed any longer now that the main loop does
+     a better job at propagating alias information.  */
+  pass = 0;
   do
     {
       changed = 0;
+      pass++;
       for (i = 0; i < reg_base_value_size; i++)
 	{
 	  rtx base = reg_base_value[i];
@@ -1141,7 +1170,7 @@ init_alias_analysis ()
 	    }
 	}
     }
-  while (changed);
+  while (changed && pass < MAX_ALIAS_LOOP_PASSES);
 
   new_reg_base_value = 0;
   reg_seen = 0;

gcc/auto-config.h

@@ -1,43 +0,0 @@
-/* auto-config.h.  Generated automatically by configure.  */
-/* config.in.  Generated automatically from configure.in by autoheader.  */
-/* Whether malloc must be declared even if <stdlib.h> is included.  */
-/* #undef NEED_DECLARATION_MALLOC */
-
-/* Whether realloc must be declared even if <stdlib.h> is included.  */
-/* #undef NEED_DECLARATION_REALLOC */
-
-/* Whether calloc must be declared even if <stdlib.h> is included.  */
-/* #undef NEED_DECLARATION_CALLOC */
-
-/* Whether free must be declared even if <stdlib.h> is included.  */
-/* #undef NEED_DECLARATION_FREE */
-
-/* Define if `sys_siglist' is declared by <signal.h>.  */
-#define SYS_SIGLIST_DECLARED 1
-
-/* Define if you have the strerror function.  */
-#define HAVE_STRERROR 1
-
-/* Define if you have the <inttypes.h> header file.  */
-/* #undef HAVE_INTTYPES_H */
-
-/* Define if you have the <limits.h> header file.  */
-#define HAVE_LIMITS_H 1
-
-/* Define if you have the <stddef.h> header file.  */
-#define HAVE_STDDEF_H 1
-
-/* Define if you have the <stdlib.h> header file.  */
-#define HAVE_STDLIB_H 1
-
-/* Define if you have the <string.h> header file.  */
-#define HAVE_STRING_H 1
-
-/* Define if you have the <strings.h> header file.  */
-#define HAVE_STRINGS_H 1
-
-/* Define if you have the <time.h> header file.  */
-#define HAVE_TIME_H 1
-
-/* Define if you have the <unistd.h> header file.  */
-#define HAVE_UNISTD_H 1

gcc/c-decl.c

@@ -4827,7 +4827,7 @@ grokdeclarator (declarator, declspecs, decl_context, initialized)
 	  {
 	    register tree link;
 
-	    for (link = current_function_parm_tags;
+	    for (link = last_function_parm_tags;
 		 link;
 		 link = TREE_CHAIN (link))
 	      TYPE_CONTEXT (TREE_VALUE (link)) = type;

gcc/cccp.c

@@ -9812,7 +9812,7 @@ new_include_prefix (prev_file_name, prefix, name)
     len = simplify_filename (dir->fname);
 
     /* Convert directory name to a prefix.  */
-    if (dir->fname[len - 1] != '/') {
+    if (len && dir->fname[len - 1] != '/') {
       if (len == 1 && dir->fname[len - 1] == '.')
 	len = 0;
       else

gcc/combine.c

@@ -2141,10 +2141,14 @@ try_combine (i3, i2, i1)
     rtx i3links, i2links, i1links = 0;
     rtx midnotes = 0;
     register int regno;
-    /* Compute which registers we expect to eliminate.  */
-    rtx elim_i2 = (newi2pat || i2dest_in_i2src || i2dest_in_i1src
+    /* Compute which registers we expect to eliminate.  newi2pat may be setting
+       either i3dest or i2dest, so we must check it.  */
+    rtx elim_i2 = ((newi2pat && reg_set_p (i2dest, newi2pat))
+		   || i2dest_in_i2src || i2dest_in_i1src
 		   ? 0 : i2dest);
-    rtx elim_i1 = i1 == 0 || i1dest_in_i1src ? 0 : i1dest;
+    rtx elim_i1 = (i1 == 0 || i1dest_in_i1src
+		   || (newi2pat && reg_set_p (i1dest, newi2pat))
+		   ? 0 : i1dest);
 
     /* Get the old REG_NOTES and LOG_LINKS from all our insns and
        clear them.  */
@@ -2305,7 +2309,7 @@ try_combine (i3, i2, i1)
 	distribute_notes (gen_rtx (EXPR_LIST, REG_DEAD, i3dest_killed,
 				   NULL_RTX),
 			  NULL_RTX, i3, newi2pat ? i2 : NULL_RTX,
-			  NULL_RTX, NULL_RTX);
+			  elim_i2, elim_i1);
       }
 
     /* For I2 and I1, we have to be careful.  If NEWI2PAT exists and sets
@@ -4032,6 +4036,11 @@ simplify_rtx (x, op0_mode, last, in_dest)
       if (GET_CODE (XEXP (x, 0)) == NEG)
 	SUBST (XEXP (x, 0), XEXP (XEXP (x, 0), 0));
 
+      /* If the mode of the operand is VOIDmode (i.e. if it is ASM_OPERANDS),
+	 do nothing.  */
+      if (GET_MODE (XEXP (x, 0)) == VOIDmode)
+	break;
+
       /* If operand is something known to be positive, ignore the ABS.  */
       if (GET_CODE (XEXP (x, 0)) == FFS || GET_CODE (XEXP (x, 0)) == ABS
 	  || ((GET_MODE_BITSIZE (GET_MODE (XEXP (x, 0)))

gcc/config/a29k/a29k.h

@@ -987,13 +987,18 @@ extern char *a29k_function_name;
 
    On the 29k, we must be able to place it in a delay slot, it must
    not use sp if the frame pointer cannot be eliminated, and it cannot
-   use local regs if we need to push the register stack.  */
+   use local regs if we need to push the register stack.
+   If this is a SET with a memory as source, it might load from
+   a stack slot, unless the address is constant.  */
 
 #define ELIGIBLE_FOR_EPILOGUE_DELAY(INSN,N)				\
   (get_attr_in_delay_slot (INSN) == IN_DELAY_SLOT_YES			\
    && ! (frame_pointer_needed						\
 	 && reg_mentioned_p (stack_pointer_rtx, PATTERN (INSN)))	\
-   && ! (needs_regstack_p () && uses_local_reg_p (PATTERN (INSN))))
+   && ! (needs_regstack_p () && uses_local_reg_p (PATTERN (INSN)))	\
+   && (GET_CODE (PATTERN (INSN)) != SET					\
+       || GET_CODE (SET_SRC (PATTERN (INSN))) != MEM			\
+       || ! rtx_varies_p (XEXP (SET_SRC (PATTERN (INSN)), 0))))
 
 /* Output assembler code for a block containing the constant parts
    of a trampoline, leaving space for the variable parts.

gcc/config/alpha/alpha.c

@@ -525,7 +525,8 @@ call_operand (op, mode)
     return 0;
 
   return (GET_CODE (op) == SYMBOL_REF
-	  || (GET_CODE (op) == REG && (TARGET_OPEN_VMS || REGNO (op) == 27)));
+	  || (GET_CODE (op) == REG
+	      && (TARGET_OPEN_VMS || TARGET_WINDOWS_NT || REGNO (op) == 27)));
 }
 
 /* Return 1 if OP is a valid Alpha comparison operator.  Here we know which
@@ -886,12 +887,10 @@ alpha_emit_set_const_1 (target, mode, c, n)
 
   /* If this is a sign-extended 32-bit constant, we can do this in at most
      three insns, so do it if we have enough insns left.  We always have
-     a sign-extended 32-bit constant when compiling on a narrow machine. 
-     Note that we cannot handle the constant 0x80000000.  */
+     a sign-extended 32-bit constant when compiling on a narrow machine.   */
 
-  if ((HOST_BITS_PER_WIDE_INT != 64
-       || c >> 31 == -1 || c >> 31 == 0)
-      && c != 0x80000000U)
+  if (HOST_BITS_PER_WIDE_INT != 64
+      || c >> 31 == -1 || c >> 31 == 0)
     {
       HOST_WIDE_INT low = (c & 0xffff) - 2 * (c & 0x8000);
       HOST_WIDE_INT tmp1 = c - low;
@@ -910,7 +909,18 @@ alpha_emit_set_const_1 (target, mode, c, n)
 	}
 
       if (c == low || (low == 0 && extra == 0))
-	return copy_to_suggested_reg (GEN_INT (c), target, mode);
+	{
+	  /* We used to use copy_to_suggested_reg (GEN_INT (c), target, mode)
+	     but that meant that we can't handle INT_MIN on 32-bit machines
+	     (like NT/Alpha), because we recurse indefinitely through 
+	     emit_move_insn to gen_movdi.  So instead, since we know exactly
+	     what we want, create it explicitly.  */
+
+	  if (target == NULL)
+	    target = gen_reg_rtx (mode);
+	  emit_insn (gen_rtx (SET, VOIDmode, target, GEN_INT (c)));
+	  return target;
+	}
       else if (n >= 2 + (extra != 0))
 	{
 	  temp = copy_to_suggested_reg (GEN_INT (low), subtarget, mode);
@@ -988,9 +998,11 @@ alpha_emit_set_const_1 (target, mode, c, n)
       /* Now try high-order zero bits.  Here we try the shifted-in bits as
 	 all zero and all ones.  Be careful to avoid shifting outside the
 	 mode and to avoid shifting outside the host wide int size.  */
+      /* On narrow hosts, don't shift a 1 into the high bit, since we'll
+	 confuse the recursive call and set all of the high 32 bits.  */
 
       if ((bits = (MIN (HOST_BITS_PER_WIDE_INT, GET_MODE_SIZE (mode) * 8)
-		   - floor_log2 (c) - 1)) > 0)
+		   - floor_log2 (c) - 1 - (HOST_BITS_PER_WIDE_INT < 64))) > 0)
 	for (; bits > 0; bits--)
 	  if ((temp = alpha_emit_set_const (subtarget, mode,
 					    c << bits, i)) != 0
@@ -1405,6 +1417,26 @@ print_operand (file, x, code)
 	fputs ("su", file);
       break;
 
+    case '(':
+      /* Generates trap-mode suffix for instructions that accept the
+	 v, sv, and svi suffix.  The only instruction that needs this
+	 is cvttq.  */
+      switch (alpha_fptm)
+	{
+	case ALPHA_FPTM_N:
+	  break;
+	case ALPHA_FPTM_U:
+	  fputs ("v", file);
+	  break;
+	case ALPHA_FPTM_SU:
+	  fputs ("sv", file);
+	  break;
+	case ALPHA_FPTM_SUI:
+	  fputs ("svi", file);
+	  break;
+	}
+      break;
+
     case ')':
       /* Generates trap-mode suffix for instructions that accept the u, su,
 	 and sui suffix.  This is the bulk of the IEEE floating point
@@ -2020,7 +2052,7 @@ output_prolog (file, size)
   /* Offset during register save.  */
   int reg_offset;
   /* Label for the procedure entry.  */
-  char *entry_label = (char *) alloca (strlen (alpha_function_name) + 5);
+  char *entry_label = (char *) alloca (strlen (alpha_function_name) + 6);
   int i;
 
   sa_size = alpha_sa_size ();
@@ -2033,7 +2065,7 @@ output_prolog (file, size)
   fprintf (file, "\t.ent ");
   assemble_name (file, alpha_function_name);
   fprintf (file, "\n");
-  sprintf (entry_label, "%s..en", alpha_function_name);
+  sprintf (entry_label, "$%s..en", alpha_function_name);
   ASM_OUTPUT_LABEL (file, entry_label);
   inside_function = TRUE;
 
@@ -2087,6 +2119,7 @@ output_prolog (file, size)
 
       fprintf (file, "\tlda $22,4096($30)\n");
 
+      fputc ('$', file);
       assemble_name (file, alpha_function_name);
       fprintf (file, "..sc:\n");
 
@@ -2094,7 +2127,7 @@ output_prolog (file, size)
       fprintf (file, "\tsubq $23,1,$23\n");
       fprintf (file, "\tlda $22,-8192($22)\n");
 
-      fprintf (file, "\tbne $23,");
+      fprintf (file, "\tbne $23,$");
       assemble_name (file, alpha_function_name);
       fprintf (file, "..sc\n");
 
@@ -2169,7 +2202,7 @@ output_prolog (file, size)
   link_section ();
   fprintf (file, "\t.align 3\n");
   ASM_OUTPUT_LABEL (file, alpha_function_name);
-  fprintf (file, "\t.pdesc ");
+  fprintf (file, "\t.pdesc $");
   assemble_name (file, alpha_function_name);
   fprintf (file, "..en,%s\n", is_stack_procedure ? "stack" : "reg");
   alpha_need_linkage (alpha_function_name, 1);
@@ -2383,6 +2416,7 @@ output_prolog (file, size)
 	fprintf (file, "\tldgp $29,0($27)\n");
 
       /* Put a label after the GP load so we can enter the function at it.  */
+      fputc ('$', file);
       assemble_name (file, alpha_function_name);
       fprintf (file, "..ng:\n");
     }
@@ -2430,6 +2464,7 @@ output_prolog (file, size)
 
       fprintf (file, "\tlda $4,4096($30)\n");
 
+      fputc ('$', file);
       assemble_name (file, alpha_function_name);
       fprintf (file, "..sc:\n");
 
@@ -2437,7 +2472,7 @@ output_prolog (file, size)
       fprintf (file, "\tsubq $5,1,$5\n");
       fprintf (file, "\tlda $4,-8192($4)\n");
 
-      fprintf (file, "\tbne $5,");
+      fprintf (file, "\tbne $5,$");
       assemble_name (file, alpha_function_name);
       fprintf (file, "..sc\n");
 
@@ -2784,11 +2819,20 @@ summarize_insn (x, sum, set)
       summarize_insn (XEXP (x, 0), sum, 0);
       break;
 
+    case ASM_OPERANDS:
+      for (i = ASM_OPERANDS_INPUT_LENGTH (x) - 1; i >= 0; i--)
+	summarize_insn (ASM_OPERANDS_INPUT (x, i), sum, 0);
+      break;
+
     case PARALLEL:
       for (i = XVECLEN (x, 0) - 1; i >= 0; i--)
 	summarize_insn (XVECEXP (x, 0, i), sum, 0);
       break;
 
+    case SUBREG:
+      x = SUBREG_REG (x);
+      /* FALLTHRU */
+
     case REG:
       {
 	int regno = REGNO (x);
@@ -2824,10 +2868,6 @@ summarize_insn (x, sum, set)
       summarize_insn (XEXP (x, 0), sum, 0);
       break;
 
-    case SUBREG:
-      summarize_insn (SUBREG_REG (x), sum, set);
-      break;
-
     case CONST_INT:   case CONST_DOUBLE:
     case SYMBOL_REF:  case LABEL_REF:     case CONST:
       break;
@@ -2958,7 +2998,7 @@ alpha_handle_trap_shadows (insns)
 		  sum.used.i = 0;
 		  sum.used.fp = 0;
 		  sum.used.mem = 0;
-		  sum.defd = shadow.used;
+		  sum.defd = sum.used;
 
 		  switch (GET_CODE (i))
 		    {
@@ -3247,11 +3287,11 @@ alpha_write_linkage (stream)
 	  || ! TREE_SYMBOL_REFERENCED (get_identifier (lptr->name)))
 	continue;
 
-      fprintf (stream, "%s..lk:\n", lptr->name);
+      fprintf (stream, "$%s..lk:\n", lptr->name);
       if (lptr->kind == KIND_LOCAL)   
 	{
 	  /*  Local and used, build linkage pair.  */
-	  fprintf (stream, "\t.quad %s..en\n", lptr->name);
+	  fprintf (stream, "\t.quad $%s..en\n", lptr->name);
 	  fprintf (stream, "\t.quad %s\n", lptr->name);
 	}
       else

gcc/config/alpha/alpha.h

@@ -646,9 +646,7 @@ enum reg_class { NO_REGS, GENERAL_REGS, FLOAT_REGS, ALL_REGS,
    : (C) == 'J' ? (VALUE) == 0					\
    : (C) == 'K' ? (unsigned HOST_WIDE_INT) ((VALUE) + 0x8000) < 0x10000	\
    : (C) == 'L' ? (((VALUE) & 0xffff) == 0			\
-		   && (((VALUE)) >> 31 == -1 || (VALUE) >> 31 == 0) \
-		   && ((HOST_BITS_PER_WIDE_INT == 64		\
-			|| (unsigned) (VALUE) != 0x80000000U)))	\
+                  && (((VALUE)) >> 31 == -1 || (VALUE) >> 31 == 0)) \
    : (C) == 'M' ? zap_mask (VALUE)				\
    : (C) == 'N' ? (unsigned HOST_WIDE_INT) (~ (VALUE)) < 0x100	\
    : (C) == 'O' ? (unsigned HOST_WIDE_INT) (- (VALUE)) < 0x100	\
@@ -1829,10 +1827,7 @@ literal_section ()						\
    PREFIX is the class of label and NUM is the number within the class.  */
 
 #define ASM_OUTPUT_INTERNAL_LABEL(FILE,PREFIX,NUM)	\
-  if ((PREFIX)[0] == 'L')				\
-    fprintf (FILE, "$%s%d:\n", & (PREFIX)[1], NUM + 32); \
-  else							\
-    fprintf (FILE, "%s%d:\n", PREFIX, NUM);
+  fprintf (FILE, "$%s%d:\n", PREFIX, NUM)
 
 /* This is how to output a label for a jump table.  Arguments are the same as
    for ASM_OUTPUT_INTERNAL_LABEL, except the insn for the jump table is
@@ -1847,10 +1842,7 @@ literal_section ()						\
    This is suitable for output with `assemble_name'.  */
 
 #define ASM_GENERATE_INTERNAL_LABEL(LABEL,PREFIX,NUM)	\
-  if ((PREFIX)[0] == 'L')				\
-    sprintf (LABEL, "*$%s%d", & (PREFIX)[1], NUM + 32);	\
-  else							\
-    sprintf (LABEL, "*%s%d", PREFIX, NUM)
+  sprintf (LABEL, "*$%s%d", PREFIX, NUM)
 
 /* Check a floating-point value for validity for a particular machine mode.  */
 
@@ -1987,8 +1979,8 @@ literal_section ()						\
 /* This is how to output an element of a case-vector that is relative.  */
 
 #define ASM_OUTPUT_ADDR_DIFF_ELT(FILE, VALUE, REL) \
-  fprintf (FILE, "\t.%s $%d\n", TARGET_WINDOWS_NT ? "long" : "gprel32", \
-	   (VALUE) + 32)
+  fprintf (FILE, "\t.%s $L%d\n", TARGET_WINDOWS_NT ? "long" : "gprel32", \
+	   (VALUE))
 
 /* This is how to output an assembler line
    that says to advance the location counter
@@ -2060,6 +2052,10 @@ literal_section ()						\
    '	Generates trap-mode suffix for instructions that accept the
         su suffix only (cmpt et al).
 
+   (	Generates trap-mode suffix for instructions that accept the
+	v, sv, and svi suffix.  The only instruction that needs this
+	is cvttq.
+
    )    Generates trap-mode suffix for instructions that accept the
 	u, su, and sui suffix.  This is the bulk of the IEEE floating
 	point instructions (addt et al).
@@ -2075,8 +2071,8 @@ literal_section ()						\
    */
 
 #define PRINT_OPERAND_PUNCT_VALID_P(CODE)				\
-  ((CODE) == '&' || (CODE) == '\'' || (CODE) == ')' || (CODE) == '+'	\
-   || (CODE) == ',' || (CODE) == '-')
+  ((CODE) == '&' || (CODE) == '\'' || (CODE) == '(' || (CODE) == ')'	\
+   || (CODE) == '+' || (CODE) == ',' || (CODE) == '-')
 
 /* Print a memory address as an operand to reference that memory location.  */
 

gcc/config/alpha/alpha.md

@@ -39,7 +39,7 @@
 ;; The TRAP_TYPE attribute marks instructions that may generate traps
 ;; (which are imprecise and may need a trapb if software complention
 ;; is desired).
-(define_attr "trap" "yes,no" (const_string "no"))
+(define_attr "trap" "no,yes" (const_string "no"))
 
 ;; For the EV4 we include four function units: ABOX, which computes
 ;; the address, BBOX, used for branches, EBOX, used for integer
@@ -215,20 +215,14 @@
 ;; First define the arithmetic insns.  Note that the 32-bit forms also
 ;; sign-extend.
 
-;; Note that we can do sign extensions in both FP and integer registers.
-;; However, the result must be in the same type of register as the input.
-;; The register preferencing code can't handle this case very well, so, for
-;; now, don't let the FP case show up here for preferencing.  Also,
-;; sign-extends in FP registers take two instructions.
 (define_insn "extendsidi2"
-  [(set (match_operand:DI 0 "register_operand" "=r,r,*f")
-	(sign_extend:DI (match_operand:SI 1 "nonimmediate_operand" "r,m,*f")))]
+  [(set (match_operand:DI 0 "register_operand" "=r,r")
+	(sign_extend:DI (match_operand:SI 1 "nonimmediate_operand" "r,m")))]
   ""
   "@
    addl %1,$31,%0
-   ldl %0,%1
-   cvtql %1,%0\;cvtlq %0,%0"
-  [(set_attr "type" "iadd,ld,fadd")])
+   ldl %0,%1"
+  [(set_attr "type" "iadd,ld")])
 
 ;; Do addsi3 the way expand_binop would do if we didn't have one.  This
 ;; generates better code.  We have the anonymous addsi3 pattern below in
@@ -1507,20 +1501,39 @@
   [(set_attr "type" "fadd")
    (set_attr "trap" "yes")])
 
+(define_insn ""
+  [(set (match_operand:DI 0 "register_operand" "=&f")
+	(fix:DI (match_operand:DF 1 "reg_or_fp0_operand" "fG")))]
+  "TARGET_FP && alpha_tp == ALPHA_TP_INSN"
+  "cvt%-q%(c %R1,%0"
+  [(set_attr "type" "fadd")
+   (set_attr "trap" "yes")])
+
 (define_insn "fix_truncdfdi2"
   [(set (match_operand:DI 0 "register_operand" "=f")
 	(fix:DI (match_operand:DF 1 "reg_or_fp0_operand" "fG")))]
   "TARGET_FP"
-  "cvt%-qc %R1,%0"
-  [(set_attr "type" "fadd")])
+  "cvt%-q%(c %R1,%0"
+  [(set_attr "type" "fadd")
+   (set_attr "trap" "yes")])
+
+(define_insn ""
+  [(set (match_operand:DI 0 "register_operand" "=&f")
+	(fix:DI (float_extend:DF
+		 (match_operand:SF 1 "reg_or_fp0_operand" "fG"))))]
+  "TARGET_FP && alpha_tp == ALPHA_TP_INSN"
+  "cvt%-q%(c %R1,%0"
+  [(set_attr "type" "fadd")
+   (set_attr "trap" "yes")])
 
 (define_insn "fix_truncsfdi2"
   [(set (match_operand:DI 0 "register_operand" "=f")
 	(fix:DI (float_extend:DF
 		 (match_operand:SF 1 "reg_or_fp0_operand" "fG"))))]
   "TARGET_FP"
-  "cvt%-qc %R1,%0"
-  [(set_attr "type" "fadd")])
+  "cvt%-q%(c %R1,%0"
+  [(set_attr "type" "fadd")
+   (set_attr "trap" "yes")])
 
 (define_insn ""
   [(set (match_operand:SF 0 "register_operand" "=&f")
@@ -3102,22 +3115,17 @@
 }")
 
 (define_expand "call_nt"
-  [(parallel [(call (mem:DI (match_operand:DI 0 "" ""))
+  [(parallel [(call (mem:DI (match_operand 0 "" ""))
 		    (match_operand 1 "" ""))
 	      (clobber (reg:DI 26))])]
   ""
   "
 { if (GET_CODE (operands[0]) != MEM)
     abort ();
-  operands[0] = XEXP (operands[0], 0);
 
-  if (GET_CODE (operands[1]) != SYMBOL_REF
-      && ! (GET_CODE (operands[1]) == REG && REGNO (operands[1]) == 27))
-    {
-      rtx tem = gen_rtx (REG, DImode, 27);
-      emit_move_insn (tem, operands[1]);
-      operands[1] = tem;
-    }
+  operands[0] = XEXP (operands[0], 0);
+  if (GET_CODE (operands[0]) != SYMBOL_REF && GET_CODE (operands[0]) != REG)
+    operands[0] = force_reg (DImode, operands[0]);
 }")
 
 ;;
@@ -3148,12 +3156,13 @@
     {
       extern char *savealloc ();
       char *symbol = XSTR (operands[0], 0);
-      char *linksym = savealloc (strlen (symbol) + 5);
+      char *linksym = savealloc (strlen (symbol) + 6);
       rtx linkage;
 
       alpha_need_linkage (symbol, 0);
 
-      strcpy (linksym, symbol);
+      linksym[0] = '$';
+      strcpy (linksym+1, symbol);
       strcat (linksym, \"..lk\");
       linkage = gen_rtx (SYMBOL_REF, Pmode, linksym);
 
@@ -3215,7 +3224,7 @@
 
 (define_expand "call_value_nt"
   [(parallel [(set (match_operand 0 "" "")
-		   (call (mem:DI (match_operand:DI 1 "" ""))
+		   (call (mem:DI (match_operand 1 "" ""))
 			 (match_operand 2 "" "")))
 	      (clobber (reg:DI 26))])]
   ""
@@ -3224,13 +3233,8 @@
     abort ();
 
   operands[1] = XEXP (operands[1], 0);
-  if (GET_CODE (operands[1]) != SYMBOL_REF
-      && ! (GET_CODE (operands[1]) == REG && REGNO (operands[1]) == 27))
-    {
-      rtx tem = gen_rtx (REG, DImode, 27);
-      emit_move_insn (tem, operands[1]);
-      operands[1] = tem;
-    }
+  if (GET_CODE (operands[0]) != SYMBOL_REF && GET_CODE (operands[0]) != REG)
+    operands[1] = force_reg (DImode, operands[1]);
 }")
 
 (define_expand "call_value_vms"
@@ -3257,11 +3261,12 @@
     {
       extern char *savealloc ();
       char *symbol = XSTR (operands[1], 0);
-      char *linksym = savealloc (strlen (symbol) + 5);
+      char *linksym = savealloc (strlen (symbol) + 6);
       rtx linkage;
 
       alpha_need_linkage (symbol, 0);
-      strcpy (linksym, symbol);
+      linksym[0] = '$';
+      strcpy (linksym+1, symbol);
       strcat (linksym, \"..lk\");
       linkage = gen_rtx (SYMBOL_REF, Pmode, linksym);
 
@@ -3287,18 +3292,19 @@
   "! TARGET_WINDOWS_NT && ! TARGET_OPEN_VMS"
   "@
    jsr $26,($27),0\;ldgp $29,0($26)
-   bsr $26,%0..ng
+   bsr $26,$%0..ng
    jsr $26,%0\;ldgp $29,0($26)"
   [(set_attr "type" "jsr")])
       
 (define_insn ""
-  [(call (mem:DI (match_operand:DI 0 "call_operand" "r,i"))
+  [(call (mem:DI (match_operand:DI 0 "call_operand" "r,R,i"))
 	 (match_operand 1 "" ""))
    (clobber (reg:DI 26))]
   "TARGET_WINDOWS_NT"
   "@
    jsr $26,(%0)
-   bsr $26,%0"
+   bsr $26,%0
+   jsr $26,%0"
   [(set_attr "type" "jsr")])
       
 (define_insn ""
@@ -3323,19 +3329,20 @@
   "! TARGET_WINDOWS_NT && ! TARGET_OPEN_VMS"
   "@
    jsr $26,($27),0\;ldgp $29,0($26)
-   bsr $26,%1..ng
+   bsr $26,$%1..ng
    jsr $26,%1\;ldgp $29,0($26)"
   [(set_attr "type" "jsr")])
 
 (define_insn ""
-  [(set (match_operand 0 "register_operand" "=rf,rf")
-	(call (mem:DI (match_operand:DI 1 "call_operand" "r,i"))
+  [(set (match_operand 0 "register_operand" "=rf,rf,rf")
+	(call (mem:DI (match_operand:DI 1 "call_operand" "r,R,i"))
 	      (match_operand 2 "" "")))
    (clobber (reg:DI 26))]
   "TARGET_WINDOWS_NT"
   "@
    jsr $26,(%1)
-   bsr $26,%1"
+   bsr $26,%1
+   jsr $26,%1"
   [(set_attr "type" "jsr")])
 
 (define_insn ""
@@ -4523,12 +4530,15 @@
     }
 }")
 
-(define_insn "exception_receiver"
+;; Ideally we should be able to define nonlocal_goto and arrange
+;; for the pc to be in a known place.  Or perhaps branch back via
+;; br instead of jmp.
+(define_insn "nonlocal_goto_receiver_osf"
   [(unspec_volatile [(const_int 0)] 2)]
   "! TARGET_OPEN_VMS && ! TARGET_WINDOWS_NT"
-  ".long 0xc3a00000\;ldgp $29,0($29)")
+  "br $29,$LGOTO%=\\n$LGOTO%=:\;ldgp $29,0($29)")
 
-(define_expand "nonlocal_goto_receiver"
+(define_expand "nonlocal_goto_receiver_vms"
   [(unspec_volatile [(const_int 0)] 1)
    (set (reg:DI 27) (mem:DI (reg:DI 29)))
    (unspec_volatile [(const_int 0)] 1)
@@ -4536,6 +4546,18 @@
   "TARGET_OPEN_VMS"
   "")
 
+(define_expand "nonlocal_goto_receiver"
+  [(unspec_volatile [(const_int 0)] 2)]
+  ""
+  "
+{
+  if (TARGET_OPEN_VMS)
+    emit_insn(gen_nonlocal_goto_receiver_vms ());
+  else if (!TARGET_WINDOWS_NT)
+    emit_insn(gen_nonlocal_goto_receiver_osf ());
+  DONE;
+}")
+
 (define_insn "arg_home"
   [(unspec [(const_int 0)] 0)
    (use (reg:DI 1))

gcc/config/alpha/elf.h

@@ -62,6 +62,14 @@ Currently only Linux uses this. */
 #define DEFAULT_VTABLE_THUNKS 1
 #endif
 
+#ifndef USE_GNULIBC_1
+#undef LIB_SPEC
+#define LIB_SPEC \
+  "%{shared: -lc} \
+   %{!shared: %{mieee-fp:-lieee} %{pthread:-lpthread} \
+	%{profile:-lc_p} %{!profile: -lc}}"
+#endif
+
 /* Output at beginning of assembler file.  */
 
 #undef ASM_FILE_START

gcc/config/alpha/linux.h

@@ -57,7 +57,7 @@ the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.  */
 
 #undef FUNCTION_PROFILER
 #define FUNCTION_PROFILER(FILE, LABELNO)			\
-	fputs ("\tjsr $28,_mcount\n", (FILE))
+	fputs ("\tlda $28,_mcount\n\tjsr $28,($28),_mcount\n", (FILE))
 
 /* Generate calls to memcpy, etc., not bcopy, etc. */
 #define TARGET_MEM_FUNCTIONS

gcc/config/alpha/vms-tramp.asm

@@ -1,22 +0,0 @@
-;# New Alpha OpenVMS trampoline
-;#
-	.set noreorder
-	.set volatile
-	.set noat
-	.file 1 "tramp.s"
-.text
-	.align 3
-	.globl __tramp
-	.ent __tramp
-__tramp..en:
-
-.link
-	.align 3
-__tramp:
-	.pdesc __tramp..en,null
-.text
-	ldq $1,24($27)
-	ldq $27,16($27)
-	ldq $28,8($27)
-	jmp $31,($28),0
-	.end __tramp

gcc/config/alpha/vms.h

@@ -238,7 +238,7 @@ link_section ()							\
 
 #undef ASM_OUTPUT_ADDR_VEC_ELT
 #define ASM_OUTPUT_ADDR_VEC_ELT(FILE, VALUE) \
-  fprintf (FILE, "\t.quad $%d\n", (VALUE) + 32)
+  fprintf (FILE, "\t.quad $L%d\n", (VALUE))
 
 #undef READONLY_DATA_SECTION
 #define READONLY_DATA_SECTION readonly_section

gcc/config/alpha/win-nt.h

@@ -70,3 +70,59 @@ Boston, MA 02111-1307, USA.  */
  %{!mwindows:-subsystem console -e _mainCRTStartup} \
  %{mcrtmt:LIBCMT.LIB%s KERNEL32.LIB%s} %{!mcrtmt:LIBC.LIB%s KERNEL32.LIB%s} \
  %{v}"
+
+
+/* Output assembler code for a block containing the constant parts
+   of a trampoline, leaving space for the variable parts.
+
+   The trampoline should set the static chain pointer to value placed
+   into the trampoline and should branch to the specified routine.  */
+
+#undef TRAMPOLINE_TEMPLATE
+#define TRAMPOLINE_TEMPLATE(FILE)			\
+{							\
+  fprintf (FILE, "\tbr $27,$LTRAMPP\n");		\
+  fprintf (FILE, "$LTRAMPP:\n\tldl $1,16($27)\n");	\
+  fprintf (FILE, "\tldl $27,12($27)\n");		\
+  fprintf (FILE, "\tjmp $31,($27),0\n");		\
+  fprintf (FILE, "\t.long 0,0\n");			\
+}
+
+/* Length in units of the trampoline for entering a nested function.  */
+
+#undef TRAMPOLINE_SIZE
+#define TRAMPOLINE_SIZE    24
+
+/* Emit RTL insns to initialize the variable parts of a trampoline.
+   FNADDR is an RTX for the address of the function's pure code.
+   CXT is an RTX for the static chain value for the function. 
+
+   This differs from the standard version in that:
+
+   We are not passed the current address in any register, and so have to 
+   load it ourselves.
+
+   We do not initialize the "hint" field because it only has an 8k
+   range and so the target is in range of something on the stack. 
+   Omitting the hint saves a bogus branch-prediction cache line load.
+
+   Always have an executable stack -- no need for a system call.
+ */
+
+#undef INITIALIZE_TRAMPOLINE
+#define INITIALIZE_TRAMPOLINE(TRAMP, FNADDR, CXT)			\
+{									\
+  rtx _addr, _val;							\
+									\
+  _addr = memory_address (Pmode, plus_constant ((TRAMP), 16));		\
+  _val = force_reg(Pmode, (FNADDR));					\
+  emit_move_insn (gen_rtx (MEM, SImode, _addr),				\
+		  gen_rtx (SUBREG, SImode, _val, 0));			\
+  _addr = memory_address (Pmode, plus_constant ((TRAMP), 20));		\
+  _val = force_reg(Pmode, (CXT));					\
+  emit_move_insn (gen_rtx (MEM, SImode, _addr),				\
+		  gen_rtx (SUBREG, SImode, _val, 0));			\
+									\
+  emit_insn (gen_rtx (UNSPEC_VOLATILE, VOIDmode,			\
+                      gen_rtvec (1, const0_rtx), 0));			\
+}

gcc/config/arm/arm.c

@@ -2352,6 +2352,10 @@ load_multiple_sequence (operands, nops, regs, base, load_offset)
       rtx reg;
       rtx offset;
 
+      /* Convert a subreg of a mem into the mem itself.  */
+      if (GET_CODE (operands[nops + i]) == SUBREG)
+	operands[nops + i] = alter_subreg(operands[nops + i]);
+
       if (GET_CODE (operands[nops + i]) != MEM)
 	abort ();
 
@@ -3913,7 +3917,7 @@ output_move_double (operands)
 {
   enum rtx_code code0 = GET_CODE (operands[0]);
   enum rtx_code code1 = GET_CODE (operands[1]);
-  rtx otherops[2];
+  rtx otherops[3];
 
   if (code0 == REG)
     {

gcc/config/arm/arm.md

@@ -3859,8 +3859,14 @@
   "
 {
   enum rtx_code code = GET_CODE (operands[1]);
-  rtx ccreg = gen_compare_reg (code, arm_compare_op0, arm_compare_op1,
-			       arm_compare_fp);
+  rtx ccreg;
+
+  /* When compiling for SOFT_FLOAT, ensure both arms are in registers.  */
+  if (! TARGET_HARD_FLOAT)
+    operands[3] = force_reg (SFmode, operands[3]);
+
+  ccreg = gen_compare_reg (code, arm_compare_op0, arm_compare_op1,
+			   arm_compare_fp);
 
   operands[1] = gen_rtx (code, VOIDmode, ccreg, const0_rtx);
 }")

gcc/config/arm/netbsd.h

@@ -130,3 +130,8 @@ the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.  */
    requirements.  */
 #undef STRUCTURE_SIZE_BOUNDARY
 #define STRUCTURE_SIZE_BOUNDARY 8
+
+/* Until they use ELF or something that handles dwarf2 unwinds
+   and initialization stuff better.  */
+#define DWARF2_UNWIND_INFO 0
+

gcc/config/float-i128.h

@@ -1,96 +0,0 @@
-/* float.h for target with IEEE 32, 64 and 128 bit floating point formats */
-#ifndef _FLOAT_H_
-#define _FLOAT_H_
-/* Produced by enquire version 4.3, CWI, Amsterdam */
-
-   /* Radix of exponent representation */
-#undef FLT_RADIX
-#define FLT_RADIX 2
-   /* Number of base-FLT_RADIX digits in the significand of a float */
-#undef FLT_MANT_DIG
-#define FLT_MANT_DIG 24
-   /* Number of decimal digits of precision in a float */
-#undef FLT_DIG
-#define FLT_DIG 6
-   /* Addition rounds to 0: zero, 1: nearest, 2: +inf, 3: -inf, -1: unknown */
-#undef FLT_ROUNDS
-#define FLT_ROUNDS 1
-   /* Difference between 1.0 and the minimum float greater than 1.0 */
-#undef FLT_EPSILON
-#define FLT_EPSILON 1.19209290e-07F
-   /* Minimum int x such that FLT_RADIX**(x-1) is a normalised float */
-#undef FLT_MIN_EXP
-#define FLT_MIN_EXP (-125)
-   /* Minimum normalised float */
-#undef FLT_MIN
-#define FLT_MIN 1.17549435e-38F
-   /* Minimum int x such that 10**x is a normalised float */
-#undef FLT_MIN_10_EXP
-#define FLT_MIN_10_EXP (-37)
-   /* Maximum int x such that FLT_RADIX**(x-1) is a representable float */
-#undef FLT_MAX_EXP
-#define FLT_MAX_EXP 128
-   /* Maximum float */
-#undef FLT_MAX
-#define FLT_MAX 3.40282347e+38F
-   /* Maximum int x such that 10**x is a representable float */
-#undef FLT_MAX_10_EXP
-#define FLT_MAX_10_EXP 38
-
-   /* Number of base-FLT_RADIX digits in the significand of a double */
-#undef DBL_MANT_DIG
-#define DBL_MANT_DIG 53
-   /* Number of decimal digits of precision in a double */
-#undef DBL_DIG
-#define DBL_DIG 15
-   /* Difference between 1.0 and the minimum double greater than 1.0 */
-#undef DBL_EPSILON
-#define DBL_EPSILON 2.2204460492503131e-16
-   /* Minimum int x such that FLT_RADIX**(x-1) is a normalised double */
-#undef DBL_MIN_EXP
-#define DBL_MIN_EXP (-1021)
-   /* Minimum normalised double */
-#undef DBL_MIN
-#define DBL_MIN 2.2250738585072014e-308
-   /* Minimum int x such that 10**x is a normalised double */
-#undef DBL_MIN_10_EXP
-#define DBL_MIN_10_EXP (-307)
-   /* Maximum int x such that FLT_RADIX**(x-1) is a representable double */
-#undef DBL_MAX_EXP
-#define DBL_MAX_EXP 1024
-   /* Maximum double */
-#undef DBL_MAX
-#define DBL_MAX 1.7976931348623157e+308
-   /* Maximum int x such that 10**x is a representable double */
-#undef DBL_MAX_10_EXP
-#define DBL_MAX_10_EXP 308
-
-   /* Number of base-FLT_RADIX digits in the significand of a long double */
-#undef LDBL_MANT_DIG
-#define LDBL_MANT_DIG 113
-   /* Number of decimal digits of precision in a long double */
-#undef LDBL_DIG
-#define LDBL_DIG 33
-   /* Difference between 1.0 and the minimum long double greater than 1.0 */
-#undef LDBL_EPSILON
-#define LDBL_EPSILON 1.925929944387235853055977942584927319E-34L
-   /* Minimum int x such that FLT_RADIX**(x-1) is a normalised long double */
-#undef LDBL_MIN_EXP
-#define LDBL_MIN_EXP (-16381)
-   /* Minimum normalised long double */
-#undef LDBL_MIN
-#define LDBL_MIN 3.362103143112093506262677817321752603E-4932L
-   /* Minimum int x such that 10**x is a normalised long double */
-#undef LDBL_MIN_10_EXP
-#define LDBL_MIN_10_EXP (-4931)
-   /* Maximum int x such that FLT_RADIX**(x-1) is a representable long double */
-#undef LDBL_MAX_EXP
-#define LDBL_MAX_EXP 16384
-   /* Maximum long double */
-#undef LDBL_MAX
-#define LDBL_MAX 1.189731495357231765085759326628007016E+4932L
-   /* Maximum int x such that 10**x is a representable long double */
-#undef LDBL_MAX_10_EXP
-#define LDBL_MAX_10_EXP 4932
-
-#endif /*  _FLOAT_H_ */

gcc/config/i386/aix386ng.h

@@ -44,7 +44,7 @@ Boston, MA 02111-1307, USA.  */
 
 /* Specify predefined symbols in preprocessor.  */
 
-#define CPP_PREDEFINES "-Dps2 -Dunix -Asystem(aix)"
+#define CPP_PREDEFINES "-Di386 -Dps2 -Dunix -Asystem(aix)"
 
 #define CPP_SPEC "%(cpp_cpu) %[cpp_cpu] \
   "%{posix:-D_POSIX_SOURCE}%{!posix:-DAIX} -D_I386 -D_AIX -D_MBCS"

gcc/config/i386/cygwin32.h

@@ -199,3 +199,6 @@ do {								\
 	     TREE_CODE (DECL) == FUNCTION_DECL			\
 	     ? "discard" : "same_size");			\
 } while (0)
+
+/* DWARF2 Unwinding doesn't work with exception handling yet. */
+#define DWARF2_UNWIND_INFO 0

gcc/config/i386/freebsd-elf.h

@@ -140,7 +140,7 @@ Boston, MA 02111-1307, USA.  */
 #define WCHAR_TYPE_SIZE BITS_PER_WORD
     
 #undef CPP_PREDEFINES
-#define CPP_PREDEFINES "-Dunix -D__ELF__ -D__FreeBSD__=2 -Asystem(FreeBSD)"
+#define CPP_PREDEFINES "-Di386 -Dunix -D__ELF__ -D__FreeBSD__=2 -Asystem(FreeBSD)"
 
 #undef CPP_SPEC
 #define CPP_SPEC "%(cpp_cpu) %[cpp_cpu] %{fPIC:-D__PIC__ -D__pic__} %{fpic:-D__PIC__ -D__pic__} %{posix:-D_POSIX_SOURCE}"

gcc/config/i386/gas.h

@@ -56,7 +56,7 @@ Boston, MA 02111-1307, USA.  */
 
 /* Specify predefined symbols in preprocessor.  */
 
-#define CPP_PREDEFINES "-Dunix"
+#define CPP_PREDEFINES "-Di386 -Dunix"
 #define CPP_SPEC "%(cpp_cpu) %[cpp_cpu] %{posix:-D_POSIX_SOURCE}"
 
 /* Allow #sccs in preprocessor.  */

gcc/config/i386/i386.c

@@ -3385,16 +3385,7 @@ notice_update_cc (exp)
       /* Jumps do not alter the cc's.  */
       if (SET_DEST (exp) == pc_rtx)
 	return;
-#ifdef IS_STACK_MODE
-      /* Moving into a memory of stack_mode may have been moved
-         in between the use and set of cc0 by loop_spl(). So
-         old value of cc.status must be retained */
-      if(GET_CODE(SET_DEST(exp))==MEM 
-         && IS_STACK_MODE(GET_MODE(SET_DEST(exp))))
-        {
-          return;
-        }
-#endif
+
       /* Moving register or memory into a register:
 	 it doesn't alter the cc's, but it might invalidate
 	 the RTX's which we remember the cc's came from.
@@ -3417,11 +3408,11 @@ notice_update_cc (exp)
 	  && (REG_P (SET_SRC (exp))
 	      || GET_RTX_CLASS (GET_CODE (SET_SRC (exp))) == '<'))
 	{
-	  if (cc_status.value1 && GET_CODE (cc_status.value1) == MEM
-	      || reg_mentioned_p (SET_DEST (exp), cc_status.value1))
+	  if (cc_status.value1
+	      && reg_overlap_mentioned_p (SET_DEST (exp), cc_status.value1))
 	    cc_status.value1 = 0;
-	  if (cc_status.value2 && GET_CODE (cc_status.value2) == MEM
-	      || reg_mentioned_p (SET_DEST (exp), cc_status.value2))
+	  if (cc_status.value2
+	      && reg_overlap_mentioned_p (SET_DEST (exp), cc_status.value2))
 	    cc_status.value2 = 0;
 	  return;
 	}

gcc/config/i386/i386.h

@@ -165,7 +165,9 @@ extern int target_flags;
 #define TARGET_USE_Q_REG (ix86_cpu == PROCESSOR_PENTIUM \
 			  || ix86_cpu == PROCESSOR_PENTIUMPRO)
 #define TARGET_USE_ANY_REG (ix86_cpu == PROCESSOR_I486)
-#define TARGET_CMOVE (ix86_arch == PROCESSOR_PENTIUMPRO)
+/* This is temporary for the release only.  Folks are working on a fix,
+   but the fix will likely not be suitable for the first egcs release.  */
+#define TARGET_CMOVE (0)                                         
 #define TARGET_DEEP_BRANCH_PREDICTION (ix86_cpu == PROCESSOR_PENTIUMPRO)
 #define TARGET_STACK_PROBE (target_flags & MASK_STACK_PROBE)
 
@@ -1522,8 +1524,9 @@ do {						\
 									\
       for (regno = 0; regno < FIRST_PSEUDO_REGISTER; regno++)		\
 	if ((regs_ever_live[regno] && ! call_used_regs[regno])		\
-	    || (current_function_uses_pic_offset_table			\
-		&& regno == PIC_OFFSET_TABLE_REGNUM))			\
+	    || ((current_function_uses_pic_offset_table			\
+		 || current_function_uses_const_pool)			\
+		&& flag_pic && regno == PIC_OFFSET_TABLE_REGNUM))	\
 	  offset += 4;							\
 									\
       (OFFSET) = offset + get_frame_size ();				\

gcc/config/i386/i386.md

@@ -1,5 +1,5 @@
 ; GCC machine description for Intel X86.
-;; Copyright (C) 1988, 1994, 1995, 1996, 1997 Free Software Foundation, Inc.
+;; Copyright (C) 1988, 94-97, 1998 Free Software Foundation, Inc.
 ;; Mostly by William Schelter.
 
 ;; This file is part of GNU CC.
@@ -834,24 +834,23 @@
 
 (define_insn ""
   [(set (match_operand:SI 0 "push_operand" "=<")
-	(match_operand:SI 1 "general_operand" "g"))]
+	(match_operand:SI 1 "nonmemory_operand" "rn"))]
+  "flag_pic"
+  "* return AS1 (push%L0,%1);")
+
+(define_insn ""
+  [(set (match_operand:SI 0 "push_operand" "=<")
+	(match_operand:SI 1 "nonmemory_operand" "ri"))]
+  "!flag_pic"
+  "* return AS1 (push%L0,%1);")
+
+;; On a 386, it is faster to push MEM directly.
+
+(define_insn ""
+  [(set (match_operand:SI 0 "push_operand" "=<")
+	(match_operand:SI 1 "memory_operand" "m"))]
   "TARGET_PUSH_MEMORY"
-  "push%L0 %1")
-
-;; If not a 386, it is faster to move MEM to a REG and then push, rather than
-;; push MEM directly.
-
-(define_insn ""
-  [(set (match_operand:SI 0 "push_operand" "=<")
-	(match_operand:SI 1 "nonmemory_operand" "ri"))]
-  "!TARGET_PUSH_MEMORY && TARGET_MOVE"
-  "push%L0 %1")
-
-(define_insn ""
-  [(set (match_operand:SI 0 "push_operand" "=<")
-	(match_operand:SI 1 "nonmemory_operand" "ri"))]
-  "!TARGET_PUSH_MEMORY && !TARGET_MOVE"
-  "push%L0 %1")
+  "* return AS1 (push%L0,%1);")
 
 ;; General case of fullword move.
 
@@ -883,8 +882,10 @@
 
 (define_insn ""
   [(set (match_operand:SI 0 "general_operand" "=g,r")
-	(match_operand:SI 1 "general_operand" "ri,m"))]
-  "(!TARGET_MOVE || GET_CODE (operands[0]) != MEM) || (GET_CODE (operands[1]) != MEM)"
+	(match_operand:SI 1 "general_operand" "rn,im"))]
+  "((!TARGET_MOVE || GET_CODE (operands[0]) != MEM)
+    || (GET_CODE (operands[1]) != MEM))
+   && flag_pic"
   "*
 {
   rtx link;
@@ -903,29 +904,50 @@
     /* Fastest way to change a 0 to a 1.  */
     return AS1 (inc%L0,%0);
 
-  if (flag_pic && SYMBOLIC_CONST (operands[1]))
+  if (SYMBOLIC_CONST (operands[1]))
     return AS2 (lea%L0,%a1,%0);
 
   return AS2 (mov%L0,%1,%0);
 }")
 
+(define_insn ""
+  [(set (match_operand:SI 0 "general_operand" "=g,r")
+	(match_operand:SI 1 "general_operand" "ri,m"))]
+  "((!TARGET_MOVE || GET_CODE (operands[0]) != MEM)
+    || (GET_CODE (operands[1]) != MEM))
+   && !flag_pic"
+  "*
+{
+  rtx link;
+  if (operands[1] == const0_rtx && REG_P (operands[0]))
+    return AS2 (xor%L0,%0,%0);
+
+  if (operands[1] == const1_rtx
+      && (link = find_reg_note (insn, REG_WAS_0, 0))
+      /* Make sure the insn that stored the 0 is still present.  */
+      && ! INSN_DELETED_P (XEXP (link, 0))
+      && GET_CODE (XEXP (link, 0)) != NOTE
+      /* Make sure cross jumping didn't happen here.  */
+      && no_labels_between_p (XEXP (link, 0), insn)
+      /* Make sure the reg hasn't been clobbered.  */
+      && ! reg_set_between_p (operands[0], XEXP (link, 0), insn))
+    /* Fastest way to change a 0 to a 1.  */
+    return AS1 (inc%L0,%0);
+
+  return AS2 (mov%L0,%1,%0);
+}")
+
 (define_insn ""
   [(set (match_operand:HI 0 "push_operand" "=<")
-	(match_operand:HI 1 "general_operand" "g"))]
+	(match_operand:HI 1 "nonmemory_operand" "ri"))]
+  ""
+  "* return AS1 (push%W0,%1);")
+
+(define_insn ""
+  [(set (match_operand:HI 0 "push_operand" "=<")
+	(match_operand:HI 1 "memory_operand" "m"))]
   "TARGET_PUSH_MEMORY"
-  "push%W0 %1")
-
-(define_insn ""
-  [(set (match_operand:HI 0 "push_operand" "=<")
-	(match_operand:HI 1 "nonmemory_operand" "ri"))]
-  "!TARGET_PUSH_MEMORY && TARGET_MOVE"
-  "push%W0 %1")
-
-(define_insn ""
-  [(set (match_operand:HI 0 "push_operand" "=<")
-	(match_operand:HI 1 "nonmemory_operand" "ri"))]
-  "!TARGET_PUSH_MEMORY && !TARGET_MOVE"
-  "push%W0 %1")
+  "* return AS1 (push%W0,%1);")
 
 ;; On i486, an incl and movl are both faster than incw and movw.
 
@@ -1038,22 +1060,12 @@
   [(set (match_operand:QI 0 "push_operand" "=<")
 	(match_operand:QI 1 "const_int_operand" "n"))]
   ""
-  "* return AS1 (push%W0,%1);")
+  "* return AS1(push%W0,%1);")
 
 (define_insn ""
   [(set (match_operand:QI 0 "push_operand" "=<")
 	(match_operand:QI 1 "register_operand" "q"))]
-  "!TARGET_MOVE"
-  "*
-{
-  operands[1] = gen_rtx (REG, HImode, REGNO (operands[1]));
-  return AS1 (push%W0,%1);
-}")
-
-(define_insn ""
-  [(set (match_operand:QI 0 "push_operand" "=<")
-	(match_operand:QI 1 "register_operand" "q"))]
-  "TARGET_MOVE"
+  ""
   "*
 {
   operands[1] = gen_rtx (REG, HImode, REGNO (operands[1]));
@@ -2515,7 +2527,7 @@
 
 (define_insn ""
   [(set (match_operand:DI 0 "nonimmediate_operand" "=rm")
-	(fix:DI (fix:XF (match_operand:XF 1 "register_operand" "f"))))
+	(fix:DI (fix:XF (match_operand:XF 1 "register_operand" "+f"))))
    (clobber (match_dup 1))
    (clobber (match_operand:SI 2 "memory_operand" "m"))
    (clobber (match_operand:DI 3 "memory_operand" "m"))
@@ -2525,7 +2537,7 @@
 
 (define_insn ""
   [(set (match_operand:DI 0 "nonimmediate_operand" "=rm")
-	(fix:DI (fix:DF (match_operand:DF 1 "register_operand" "f"))))
+	(fix:DI (fix:DF (match_operand:DF 1 "register_operand" "+f"))))
    (clobber (match_dup 1))
    (clobber (match_operand:SI 2 "memory_operand" "m"))
    (clobber (match_operand:DI 3 "memory_operand" "m"))
@@ -2535,7 +2547,7 @@
 
 (define_insn ""
   [(set (match_operand:DI 0 "nonimmediate_operand" "=rm")
-	(fix:DI (fix:SF (match_operand:SF 1 "register_operand" "f"))))
+	(fix:DI (fix:SF (match_operand:SF 1 "register_operand" "+f"))))
    (clobber (match_dup 1))
    (clobber (match_operand:SI 2 "memory_operand" "m"))
    (clobber (match_operand:DI 3 "memory_operand" "m"))

gcc/config/i386/linux-aout.h

@@ -28,7 +28,7 @@ Boston, MA 02111-1307, USA.  */
 /* Specify predefined symbols in preprocessor.  */
 
 #undef CPP_PREDEFINES
-#define CPP_PREDEFINES "-Dunix -Dlinux -Asystem(posix)"
+#define CPP_PREDEFINES "-Di386 -Dunix -Dlinux -Asystem(posix)"
 
 #undef CPP_SPEC
 #define CPP_SPEC "%(cpp_cpu) %[cpp_cpu] %{fPIC:-D__PIC__ -D__pic__} %{fpic:-D__PIC__ -D__pic__} %{posix:-D_POSIX_SOURCE}"

gcc/config/i386/linux-oldld.h

@@ -28,7 +28,7 @@ Boston, MA 02111-1307, USA.  */
 /* Specify predefined symbols in preprocessor.  */
 
 #undef CPP_PREDEFINES
-#define CPP_PREDEFINES "-Dunix -Dlinux -Asystem(posix)"
+#define CPP_PREDEFINES "-Di386 -Dunix -Dlinux -Asystem(posix)"
 
 #undef CPP_SPEC
 #define CPP_SPEC "%(cpp_cpu) %[cpp_cpu] %{fPIC:-D__PIC__ -D__pic__} %{fpic:-D__PIC__ -D__pic__} %{posix:-D_POSIX_SOURCE}"

gcc/config/i386/linux.h

@@ -148,7 +148,7 @@ Boston, MA 02111-1307, USA.  */
 #define WCHAR_TYPE_SIZE BITS_PER_WORD
     
 #undef CPP_PREDEFINES
-#define CPP_PREDEFINES "-D__ELF__ -Dunix -Dlinux -Asystem(posix)"
+#define CPP_PREDEFINES "-Di386 -D__ELF__ -Dunix -Dlinux -Asystem(posix)"
 
 #undef CPP_SPEC
 #ifdef USE_GNULIBC_1

gcc/config/i386/netbsd.h

@@ -71,3 +71,8 @@
       fprintf (FILE, "\tcall mcount\n");				\
     }									\
 }
+
+/* Until they use ELF or something that handles dwarf2 unwinds
+   and initialization stuff better.  */
+#define DWARF2_UNWIND_INFO 0
+

gcc/config/i386/osfelf.h

@@ -22,7 +22,7 @@ Boston, MA 02111-1307, USA.  */
 #include "config/i386/osfrose.h"
 
 #undef	CPP_PREDEFINES
-#define CPP_PREDEFINES "-DOSF -DOSF1 -Dunix -Asystem(xpg4)"
+#define CPP_PREDEFINES "-Di386 -DOSF -DOSF1 -Dunix -Asystem(xpg4)"
 
 #undef  CPP_SPEC
 #define CPP_SPEC "%(cpp_cpu) %[cpp_cpu] \

gcc/config/i386/osfrose.h

@@ -90,7 +90,7 @@ Boston, MA 02111-1307, USA.  */
 
 /* Change default predefines.  */
 #undef	CPP_PREDEFINES
-#define CPP_PREDEFINES "-DOSF -DOSF1 -Dunix -Asystem(xpg4)"
+#define CPP_PREDEFINES "-Di386 -DOSF -DOSF1 -Dunix -Asystem(xpg4)"
 
 #undef  CPP_SPEC
 #define CPP_SPEC "%(cpp_cpu) %[cpp_cpu] \

gcc/config/i386/sco.h

@@ -55,7 +55,7 @@ Boston, MA 02111-1307, USA.  */
 /* Specify predefined symbols in preprocessor.  */
 
 #undef CPP_PREDEFINES
-#define CPP_PREDEFINES "-Dunix -DM_UNIX -DM_I386 -DM_COFF -DM_WORDSWAP -Asystem(svr3)"
+#define CPP_PREDEFINES "-Di386 -Dunix -DM_UNIX -DM_I386 -DM_COFF -DM_WORDSWAP -Asystem(svr3)"
 
 #undef CPP_SPEC
 #define CPP_SPEC "%(cpp_cpu) %[cpp_cpu] %{scointl:-DM_INTERNAT}"

gcc/config/i386/sco4.h

@@ -63,7 +63,7 @@ Boston, MA 02111-1307, USA.  */
 
 #undef CPP_PREDEFINES
 #define CPP_PREDEFINES \
- "-Asystem(svr3)"
+ "-Di386 -Asystem(svr3)"
 
 #undef CPP_SPEC
 #define CPP_SPEC "%(cpp_cpu) %[cpp_cpu] \

gcc/config/i386/sco4dbx.h

@@ -61,7 +61,7 @@ Boston, MA 02111-1307, USA.  */
    Specify predefined symbols in preprocessor.  */
 
 #undef CPP_PREDEFINES
-#define CPP_PREDEFINES "-Dunix -Asystem(svr3)"
+#define CPP_PREDEFINES "-Di386 -Dunix -Asystem(svr3)"
 
 #undef CPP_SPEC
 #define CPP_SPEC "%(cpp_cpu) %[cpp_cpu] \

gcc/config/i386/sco5.h

@@ -1,5 +1,5 @@
 /* Definitions for Intel 386 running SCO Unix System V 3.2 Version 5.
-   Copyright (C) 1992, 1995, 1996 Free Software Foundation, Inc.
+   Copyright (C) 1992, 1995, 1996, 1997, 1998 Free Software Foundation, Inc.
    Contributed by Kean Johnston (hug@netcom.com)
 
 This file is part of GNU CC.
@@ -284,6 +284,14 @@ do {									\
    }									\
 } while (0)
 
+/* A C statement (sans semicolon) to output to the stdio stream
+   FILE the assembler definition of uninitialized global DECL named
+   NAME whose size is SIZE bytes and alignment is ALIGN bytes.
+   Try to use asm_output_aligned_bss to implement this macro.  */
+
+#define ASM_OUTPUT_ALIGNED_BSS(FILE, DECL, NAME, SIZE, ALIGN) 		\
+  asm_output_aligned_bss (FILE, DECL, NAME, SIZE, ALIGN)
+
 #undef ESCAPES
 #define ESCAPES \
 "\1\1\1\1\1\1\1\1btn\1fr\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\
@@ -367,6 +375,19 @@ do {									\
         fprintf ((FILE), "\n");						\
 } while (0) 
 
+/* Must use data section for relocatable constants when pic.  */
+#undef SELECT_RTX_SECTION
+#define SELECT_RTX_SECTION(MODE,RTX)					\
+{									\
+  if (TARGET_ELF) {							\
+    if (flag_pic && symbolic_operand (RTX))				\
+      data_section ();							\
+    else								\
+      const_section ();							\
+  } else								\
+    readonly_data_section();						\
+}
+
 #undef ASM_OUTPUT_CASE_LABEL
 #define ASM_OUTPUT_CASE_LABEL(FILE,PREFIX,NUM,JUMPTABLE)		\
 do {									\
@@ -570,6 +591,7 @@ do {									\
 
 #define DWARF_DEBUGGING_INFO 1
 #define SDB_DEBUGGING_INFO   1
+#define DBX_DEBUGGING_INFO   1
 #define PREFERRED_DEBUGGING_TYPE					\
   ((TARGET_ELF) ? DWARF_DEBUG: SDB_DEBUG)
 
@@ -807,18 +829,17 @@ dtors_section ()							\
        %{!Xc:%{Xk:values-Xk.o%s} \
         %{!Xk:%{Xt:values-Xt.o%s} \
          %{!Xt:values-Xa.o%s}}}}}} \
-  %{mcoff:crtbeginS.o%s} \
-  %{!mcoff:%{!static:crtbegin.o%s}%{static:crtbeginS.o%s}}"
+  %{mcoff:crtbeginS.o%s} %{!mcoff:crtbegin.o%s}"
 
 #undef ENDFILE_SPEC
 #define ENDFILE_SPEC \
- "%{!mcoff:%{!static:crtend.o%s}%{static:crtendS.o%s}} \
+ "%{!mcoff:crtend.o%s} \
   %{mcoff:crtendS.o%s} \
   %{pg:gcrtn.o%s}%{!pg:crtn.o%s}"
 
 #undef CPP_PREDEFINES
 #define CPP_PREDEFINES \
- "-Asystem(svr3)"
+ "-Di386 -Asystem(svr3)"
 
 /* You are in a maze of GCC specs ... all alike */
 
@@ -882,7 +903,11 @@ dtors_section ()							\
 
 #undef LIB_SPEC
 #define LIB_SPEC \
- "%{!shared:%{!symbolic:-lcrypt -lgen -lc}}"
+ "%{shared:pic/libgcc.a%s}%{!shared:%{!symbolic:-lcrypt -lgen -lc}}"
+
+#undef LIBGCC_SPEC
+#define LIBGCC_SPEC \
+ "%{!shared:-lgcc}"
 
 #define MASK_COFF     		010000000000	/* Mask for elf generation */
 #define TARGET_COFF             (target_flags & MASK_COFF)
@@ -919,7 +944,7 @@ compiler at the end of the day. Onward we go ...
 
 # if defined (_SCO_ELF)
 #  define OBJECT_FORMAT_ELF
-#  define HAVE_ATEXIT
+#  define HAVE_ATEXIT 1
 #  define INIT_SECTION_ASM_OP INIT_SECTION_ASM_OP_ELF
 #  define FINI_SECTION_ASM_OP FINI_SECTION_ASM_OP_ELF
 #  define DTORS_SECTION_ASM_OP DTORS_SECTION_ASM_OP_ELF

gcc/config/i386/scodbx.h

@@ -50,7 +50,7 @@ Boston, MA 02111-1307, USA.  */
 /* Specify predefined symbols in preprocessor.  */
 
 #undef CPP_PREDEFINES
-#define CPP_PREDEFINES "-Dunix -DM_UNIX -DM_I386 -DM_COFF -DM_WORDSWAP -Asystem(svr3)"
+#define CPP_PREDEFINES "-Di386 -Dunix -DM_UNIX -DM_I386 -DM_COFF -DM_WORDSWAP -Asystem(svr3)"
 
 #undef CPP_SPEC
 #define CPP_SPEC " -Acpu(i386) -Amachine(i386) %{scointl:-DM_INTERNAT}"

gcc/config/i386/sol2.h

@@ -36,7 +36,7 @@ Boston, MA 02111-1307, USA.  */
 /* Add "sun" to the list of symbols defined for SVR4.  */
 #undef CPP_PREDEFINES
 #define CPP_PREDEFINES \
-  "-Dunix -D__svr4__ -D__SVR4 -Dsun -Asystem(svr4)"
+  "-Di386 -Dunix -D__svr4__ -D__SVR4 -Dsun -Asystem(svr4)"
 
 #undef CPP_SPEC
 #define CPP_SPEC "%(cpp_cpu) %[cpp_cpu] \

gcc/config/i386/sysv3.h

@@ -42,7 +42,7 @@ Boston, MA 02111-1307, USA.  */
 
 /* Specify predefined symbols in preprocessor.  */
 
-#define CPP_PREDEFINES "-Dunix -Asystem(svr3)"
+#define CPP_PREDEFINES "-Di386 -Dunix -Asystem(svr3)"
 
 #define CPP_SPEC "%(cpp_cpu) %[cpp_cpu] %{posix:-D_POSIX_SOURCE}"
 

gcc/config/i386/t-sol2

@@ -31,6 +31,10 @@ crtn.o: $(srcdir)/config/i386/sol2-cn.asm $(GCC_PASSES)
 # to produce a shared library, but since we don't know ahead of time when
 # we will be doing that, we just always use -fPIC when compiling the
 # routines in crtstuff.c.
+#
+# We must also enable optimization to avoid having any code appear after
+# the call & alignment statement, but before we switch back to the
+# .text section.
 
-CRTSTUFF_T_CFLAGS = -fPIC
+CRTSTUFF_T_CFLAGS = -fPIC -O2
 TARGET_LIBGCC2_CFLAGS = -fPIC

gcc/config/i386/x-cygwin32

@@ -1,4 +1,3 @@
 # Don't run fixproto
 STMP_FIXPROTO =
-# Don't need collect2
 USE_COLLECT2 =

gcc/config/i386/x-sco5

@@ -1,6 +1,6 @@
 RANLIB = :
 RANLIB_TEST = false
-CC = cc 
+CC = cc
 OLDCC = cc
 CCLIBFLAGS =
 CLIB = -lPW

gcc/config/i386/xm-cygwin32.h

@@ -22,6 +22,7 @@ Boston, MA 02111-1307, USA. */
 #define NO_STAB_H
 #define HAVE_RUSAGE
 #define HAVE_FILE_H
+#define NO_SYS_SIGLIST 1
 #define EXECUTABLE_SUFFIX ".exe"
 
 /* Even though we support "/", allow "\" since everybody tests both.  */

gcc/config/linux.h

@@ -104,7 +104,8 @@ Boston, MA 02111-1307, USA.  */
      %{!profile:%{!ggdb:-lc} %{ggdb:-lg}}}"
 #else
 #define LIB_SPEC \
-  "%{!shared: %{mieee-fp:-lieee} %{pthread:-lpthread} \
+  "%{shared: -lc} \
+   %{!shared: %{mieee-fp:-lieee} %{pthread:-lpthread} \
 	%{profile:-lc_p} %{!profile: -lc}}"
 #endif
 #else

gcc/config/m68k/m68k-none.h

@@ -72,12 +72,12 @@ Boston, MA 02111-1307, USA.  */
 #if TARGET_CPU_DEFAULT == M68K_CPU_m68302
 #define CPP_CPU_DEFAULT_SPEC "%{!ansi:-Dmc68302 } -D__mc68302 -D__mc68302__"
 #define ASM_CPU_DEFAULT_SPEC "-mc68302"
-#define CC1_CPU_DEFAULT_SPEC "-m68000"
+#define CC1_CPU_DEFAULT_SPEC "-m68302"
 #else
 #if TARGET_CPU_DEFAULT == M68K_CPU_m68332
-#define CPP_CPU_DEFAULT_SPEC "%{!ansi:-Dmc68332 } -D__mc68332 -D__mc68332__"
+#define CPP_CPU_DEFAULT_SPEC "%{!ansi:-Dmc68332 -Dmcpu32 } -D__mc68332 -D__mc68332__ -D__mcpu32 -D__mcpu32__"
 #define ASM_CPU_DEFAULT_SPEC "-mc68332"
-#define CC1_CPU_DEFAULT_SPEC "-m68020 -mnobitfield %{!m68881:-msoft-float}"
+#define CC1_CPU_DEFAULT_SPEC "-m68332"
 #else
 Unrecognized value in TARGET_CPU_DEFAULT.
 #endif
@@ -92,20 +92,21 @@ Unrecognized value in TARGET_CPU_DEFAULT.
 #undef CPP_PREDEFINES
 #define CPP_PREDEFINES "-Dmc68000"
 
-/* Define one of __HAVE_68881__, __HAVE_FPA__, or nothing (soft float), appropriately.  */
+/* Define one of __HAVE_68881__, __HAVE_FPA__, __HAVE_SKY__, or nothing 
+   (soft float), appropriately.  */
 #undef CPP_FPU_SPEC
 #if TARGET_DEFAULT & MASK_68881
-/* ??? Why isn't m68302 treated like m68000 here?  */
-#define CPP_FPU_SPEC \
-"%{!mc68000:%{!m68000:%{!m68332:%{!m5200:%{!msoft-float:%{mfpa:-D__HAVE_FPA__ }%{!mfpa:-D__HAVE_68881__ }}}}}}"
+#define CPP_FPU_SPEC "\
+%{!mc68000:%{!m68000:%{!m68302:%{!mcpu32:%{!m68332:%{!m5200:%{!msoft-float:%{!mno-68881:%{!mfpa:%{!msky:-D__HAVE_68881__ }}}}}}}}}} \
+%{m68881:-D__HAVE_68881__ }%{mfpa:-D__HAVE_FPA__ }%{msky:-D__HAVE_SKY__ }"
 #else
 /* This can't currently happen, but we code it anyway to show how it's done.  */
 #if TARGET_DEFAULT & MASK_FPA
 #define CPP_FPU_SPEC \
 "%{!msoft-float:%{m68881:-D__HAVE_68881__ }%{!m68881:-D__HAVE_FPA__ }}"
 #else
-#define CPP_FPU_SPEC \
-"%{m68881:-D__HAVE_68881__ }%{mfpa:-D__HAVE_FPA__ }"
+#define CPP_FPU_SPEC "\
+%{m68881:-D__HAVE_68881__ }%{mfpa:-D__HAVE_FPA__ }%{msky:-D__HAVE_SKY__ }"
 #endif
 #endif
 
@@ -120,8 +121,10 @@ Unrecognized value in TARGET_CPU_DEFAULT.
    -m68040: define mc68040
    -m68060: define mc68060
    -m68020-40: define mc68020 mc68030 mc68040
-   -m68302: define mc68302
-   -m68332: define mc68332
+   -m68020-60: define mc68020 mc68030 mc68040 mc68060
+   -m68302: define mc68302 
+   -m68332: define mc68332 mcpu32
+   -mcpu32: define mcpu32
    -m5200: define mcf5200
    default: define as above appropriately
 
@@ -130,10 +133,9 @@ Unrecognized value in TARGET_CPU_DEFAULT.
 
 #undef CPP_SPEC
 #define CPP_SPEC "\
-%(cpp_fpu) \
-%{!ansi:%{m68010:-Dmc68010 }%{m68020:-Dmc68020 }%{mc68020:-Dmc68020 }%{m68030:-Dmc68030 }%{m68040:-Dmc68040 }%{m68020-40:-Dmc68020 -Dmc68030 -Dmc68040 }%{m68060:-Dmc68060 }%{m68302:-Dmc68302 }%{m68332:-Dmc68332 }%{m5200:-Dmcf5200 }} \
-%{m68010:-D__mc68010__ -D__mc68010 }%{m68020:-D__mc68020__ -D__mc68020 }%{mc68020:-D__mc68020__ -D__mc68020 }%{m68030:-D__mc68030__ -D__mc68030 }%{m68040:-D__mc68040__ -D__mc68040 }%{m68020-40:-D__mc68020__ -D__mc68030__ -D__mc68040__ -D__mc68020 -D__mc68030 -D__mc68040 }%{mc68060:-D__mc68060__ -D__mc68060 }%{m68302:-D__mc68302__ -D__mc68302 }%{m68332:-D__mc68332__ -D__mc68332 }%{m5200:-D__mcf5200__ -D__mcf5200 } \
-%{!mc68000:%{!m68000:%{!m68010:%{!mc68020:%{!m68020:%{!m68030:%{!m68040:%{!m68020-40:%{!m68060:%{!m68302:%{!m68332:%{!m5200:%(cpp_cpu_default)}}}}}}}}}}}} \
+%(cpp_fpu)%{!ansi:%{m68302:-Dmc68302 }%{m68010:-Dmc68010 }%{m68020:-Dmc68020 }%{mc68020:-Dmc68020 }%{m68030:-Dmc68030 }%{m68040:-Dmc68040 }%{m68020-40:-Dmc68020 -Dmc68030 -Dmc68040 }%{m68020-60:-Dmc68020 -Dmc68030 -Dmc68040 -Dmc68060 }%{m68060:-Dmc68060 }%{mcpu32:-Dmcpu32 } %{m68332:-Dmc68332 -Dmcpu32 }%{m5200:-Dmcf5200 }} \
+%{m68302:-D__mc68302__ -D__mc68302 }%{m68010:-D__mc68010__ -D__mc68010 }%{m68020:-D__mc68020__ -D__mc68020 }%{mc68020:-D__mc68020__ -D__mc68020 }%{m68030:-D__mc68030__ -D__mc68030 }%{m68040:-D__mc68040__ -D__mc68040 }%{m68020-40:-D__mc68020__ -D__mc68030__ -D__mc68040__ -D__mc68020 -D__mc68030 -D__mc68040 }%{m68020-60:-D__mc68020__ -D__mc68030__ -D__mc68040__ -D__mc68020 -D__mc68030 -D__mc68040 -D__mc68060__ -D__mc68060 }%{m68060:-D__mc68060__ -D__mc68060 }%{mcpu32:-D__mcpu32__ -D__mcpu32 }%{m68332:-D__mc68332__ -D__mc68332 -D__mcpu32__ -D__mcpu32 }%{m5200:-D__mcf5200__ -D__mcf5200 } \
+%{!mc68000:%{!m68000:%{!m68302:%{!m68010:%{!mc68020:%{!m68020:%{!m68030:%{!m68040:%{!m68020-40:%{!m68020-60:%{!m68060:%{!mcpu32: %{!m68332:%{!m5200:%(cpp_cpu_default)}}}}}}}}}}}}}} \
 %(cpp_subtarget) \
 "
 
@@ -141,18 +143,15 @@ Unrecognized value in TARGET_CPU_DEFAULT.
 
 #undef ASM_SPEC
 #define ASM_SPEC "\
-%{m68851}%{mno-68851}%{m68881}%{mno-68881}%{msoft-float:-mno-68881 } \
-%{m68000}%{mc68000}%{m68010}%{m68020}%{mc68020}%{m68030}%{m68040}%{m68020-40:-mc68040}%{m68060}%{m68302}%{m68332}%{m5200} \
-%{!mc68000:%{!m68000:%{!m68010:%{!mc68020:%{!m68020:%{!m68030:%{!m68040:%{!m68020-40:%{!m68060:%{!m68302:%{!m68332:%{!m5200:%(asm_cpu_default)}}}}}}}}}}}} \
+%{m68851}%{mno-68851}%{m68881}%{mno-68881}%{msoft-float:-mno-68881 }%{m68000}%{m68302}%{mc68000}%{m68010}%{m68020}%{mc68020}%{m68030}%{m68040}%{m68020-40:-mc68040 }%{m68020-60:-mc68040 }%{m68060}%{mcpu32}%{m68332}%{m5200}%{!mc68000:%{!m68000:%{!m68302:%{!m68010:%{!mc68020:%{!m68020:%{!m68030:%{!m68040:%{!m68020-40:%{!m68020-60:%{!m68060:%{!mcpu32:%{!m68332:%{!m5200:%(asm_cpu_default) }}}}}}}}}}}}}} \
 "
 
+/* cc1/cc1plus always receives all the -m flags. If the specs strings above 
+   are consistent with the TARGET_OPTIONS flags in m68k.h, there should be no
+   need for any further cc1/cc1plus specs.  */
+
 #undef CC1_SPEC
-#define CC1_SPEC "\
-%{m68000:%{!m68881:-msoft-float }}%{m68302:-m68000 }%{m68332:-m68020 -mnobitfield %{!m68881:-msoft-float }} \
-"
-/* ??? Is this needed?
-%{!m68000:%{!mc68000:%{!m68010:%{!mc68020:%{!m68020:%{!m68030:%{!m68040:%{!m68020-40:%{!m68302:%{!m68332:%(cc1_cpu_default)}}}}}}}}}}
-*/
+#define CC1_SPEC ""
 
 /* This macro defines names of additional specifications to put in the specs
    that can be used in various specifications like CC1_SPEC.  Its definition

gcc/config/m68k/m68k.h

@@ -165,17 +165,18 @@ extern int target_flags;
     { "68060", - (MASK_5200|MASK_68040)},				\
     { "68060", (MASK_68020|MASK_68881|MASK_BITFIELD			\
 		|MASK_68040_ONLY|MASK_68060)},				\
-    { "5200", - (MASK_68060|MASK_68040|MASK_68020|MASK_BITFIELD|MASK_68881)}, \
+    { "5200", - (MASK_68060|MASK_68040|MASK_68020|MASK_BITFIELD		\
+		|MASK_68881)},						\
     { "5200", (MASK_5200)},						\
     { "68851", 0},							\
     { "no-68851", 0},							\
     { "68302", - (MASK_5200|MASK_68060|MASK_68040|MASK_68040_ONLY	\
 		  |MASK_68020|MASK_BITFIELD|MASK_68881)},		\
     { "68332", - (MASK_5200|MASK_68060|MASK_68040|MASK_68040_ONLY	\
-		  |MASK_BITFIELD)},					\
+		  |MASK_BITFIELD|MASK_68881)},				\
     { "68332", MASK_68020},						\
     { "cpu32", - (MASK_5200|MASK_68060|MASK_68040|MASK_68040_ONLY	\
-		  |MASK_BITFIELD)},					\
+		  |MASK_BITFIELD|MASK_68881)},				\
     { "cpu32", MASK_68020},						\
     { "align-int", MASK_ALIGN_INT },					\
     { "no-align-int", -MASK_ALIGN_INT },				\
@@ -1040,7 +1041,7 @@ while(0)
 	|| defined(__mc68020__) || defined(mc68020) \
 	|| defined(__mc68030__) || defined(mc68030) \
 	|| defined(__mc68040__) || defined(mc68040) \
-	|| defined(__mc68332__) || defined(mc68332)
+	|| defined(__mcpu32__) || defined(mcpu32)
 #define MACHINE_STATE_m68010_up
 #endif
 

gcc/config/m68k/mot3300.h

@@ -184,6 +184,9 @@ Boston, MA 02111-1307, USA.  */
 #undef ASCII_DATA_ASM_OP
 #define	ASCII_DATA_ASM_OP "byte"
 
+#undef SET_ASM_OP
+#define SET_ASM_OP "set"
+
 #endif /* USE_GAS */
 
 #ifdef USE_GLD
@@ -353,7 +356,7 @@ dtors_section ()							\
 #undef ASM_BYTE
 #define ASM_BYTE	".byte"
 #undef ASM_BYTE_OP
-#define ASM_BYTE_OP	"\t.byte"
+#define ASM_BYTE_OP	".byte"
 #else
 #undef ASM_LONG
 #define ASM_LONG	"long"
@@ -364,7 +367,7 @@ dtors_section ()							\
 #undef ASM_BYTE
 #define ASM_BYTE	"byte"
 #undef ASM_BYTE_OP
-#define ASM_BYTE_OP	"\tbyte"
+#define ASM_BYTE_OP	"byte"
 #endif /* USE_GAS */
 
 /* The sysV68 as doesn't know about double's and float's.  */
@@ -473,7 +476,7 @@ do { long l;					\
 #undef ASM_OUTPUT_ASCII
 #define ASM_OUTPUT_ASCII(FILE,PTR,LEN) \
   do { register int sp = 0, lp = 0;				\
-    fprintf ((FILE), "%s\t", ASM_BYTE_OP);			\
+    fprintf ((FILE), "\t%s\t", ASM_BYTE_OP);			\
   loop:								\
     if ((PTR)[sp] > ' ' && ! ((PTR)[sp] & 0x80) && (PTR)[sp] != '\\')	\
       { lp += 3;						\

gcc/config/m68k/netbsd.h

@@ -56,3 +56,8 @@
 /* Don't default to pcc-struct-return, because gcc is the only compiler, and
    we want to retain compatibility with older gcc versions.  */
 #define DEFAULT_PCC_STRUCT_RETURN 0
+
+/* Until they use ELF or something that handles dwarf2 unwinds
+   and initialization stuff better.  */
+#define DWARF2_UNWIND_INFO 0
+

gcc/config/m68k/t-m68kbare

@@ -15,10 +15,10 @@ xfgnulib.c: $(srcdir)/config/m68k/fpgnulib.c
 	echo '#define EXTFLOAT' > xfgnulib.c
 	cat $(srcdir)/config/m68k/fpgnulib.c >> xfgnulib.c
 
-MULTILIB_OPTIONS = m68000/m68020/m5200 m68881/msoft-float
+MULTILIB_OPTIONS = m68000/m68020/m5200/mcpu32 m68881/msoft-float
 MULTILIB_DIRNAMES =
-MULTILIB_MATCHES = m68000=mc68000 m68000=m68302 m68000=m68332 m68020=mc68020 m68020=m68040
-MULTILIB_EXCEPTIONS = *m5200/*m68881 *m5200/*msoft-float
+MULTILIB_MATCHES = m68000=mc68000 m68000=m68302 mcpu32=m68332 m68020=mc68020 m68020=m68040 m68020=m68060
+MULTILIB_EXCEPTIONS = m68000/msoft-float m5200/m68881 m5200/msoft-float mcpu32/m68881 mcpu32/msoft-float
 
 LIBGCC = stmp-multilib
 INSTALL_LIBGCC = install-multilib

gcc/config/m68k/x-next

@@ -3,7 +3,7 @@ CC=cc -traditional-cpp
 OLDCC=CC -traditional-cpp
 
 # Specify other dirs of system header files to be fixed.
-OTHER_FIXINCLUDES_DIRS= /LocalDeveloper/Headers
+OTHER_FIXINCLUDES_DIRS= /LocalDeveloper/Headers /NextDeveloper/Headers
 
 # <limits.h> is sometimes in /usr/include/ansi/limits.h.
 LIMITS_H_TEST = [ -f $(SYSTEM_HEADER_DIR)/limits.h -o -f $(SYSTEM_HEADER_DIR)/ansi/limits.h ]

gcc/config/m88k/dgux.h

@@ -30,7 +30,7 @@ Boston, MA 02111-1307, USA.  */
   (TARGET_SVR4 ? DWARF_DEBUG : SDB_DEBUG)
 
 #ifndef VERSION_INFO2
-#define VERSION_INFO2   "$Revision: 1.24 $"
+#define VERSION_INFO2   "$Revision: 1.1.1.1 $"
 #endif
 #ifndef NO_BUGS
 #define AS_BUG_IMMEDIATE_LABEL
@@ -137,13 +137,13 @@ Boston, MA 02111-1307, USA.  */
 #define ASM_CPU_SPEC "\
 		  %{v:-V}\
 		  %{g:\
-		  %{mno-legend:-Wc,off}\
-		  %{!mno-legend:-Wc,-fix-bb,-s\"%i\"\
-		  %{traditional:,-lc}\
-		  %{!traditional:,-lansi-c}\
-		  %{mstandard:,-keep-std}\
-		  %{mexternal-legend:,-external}\
-		  %{mocs-frame-position:,-ocs}}}"
+%{mno-legend:-Wc,off}\
+%{!mno-legend:-Wc,-fix-bb,-s\"%i\"\
+%{traditional:,-lc}\
+%{!traditional:,-lansi-c}\
+%{mstandard:,-keep-std}\
+%{mexternal-legend:,-external}\
+%{mocs-frame-position:,-ocs}}}"
 
 #define CPP_CPU_SPEC "\
                   %{!m88000:%{!m88100:%{m88110:-D__m88110__}}} \

gcc/config/mips/abi64.h

@@ -213,8 +213,24 @@ extern struct rtx_def *mips_function_value ();
   (mips_abi == ABI_EABI && (NAMED)					\
    && FUNCTION_ARG_PASS_BY_REFERENCE (CUM, MODE, TYPE, NAMED))
 
+/* Define LONG_MAX correctly for all users.  We need to handle 32 bit EABI,
+   64 bit EABI, N32, and N64 as possible defaults.  The checks performed here
+   are the same as the checks in override_options in mips.c that determines
+   whether MASK_LONG64 will be set.
+
+   This does not handle inappropriate options or ununusal option
+   combinations.  */
+
 #undef LONG_MAX_SPEC
-#define LONG_MAX_SPEC "%{!mno-long64:-D__LONG_MAX__=9223372036854775807LL}"
+#if ((MIPS_ABI_DEFAULT == ABI_64) || ((MIPS_ABI_DEFAULT == ABI_EABI) && ((TARGET_DEFAULT | TARGET_CPU_DEFAULT) & MASK_64BIT)))
+#define LONG_MAX_SPEC \
+  "%{!mabi=n32:%{!mno-long64:%{!mgp32:-D__LONG_MAX__=9223372036854775807L}}}"
+#else
+#define LONG_MAX_SPEC \
+  "%{mabi=64:-D__LONG_MAX__=9223372036854775807L} \
+   %{mlong64:-D__LONG_MAX__=9223372036854775807L} \
+   %{mgp64:-D__LONG_MAX__=9223372036854775807L}"
+#endif
 
 /* ??? Unimplemented stuff follows.  */
 

gcc/config/mips/iris5.h

@@ -122,6 +122,9 @@ Boston, MA 02111-1307, USA.  */
 #undef SDB_DEBUGGING_INFO
 #undef MIPS_DEBUGGING_INFO
 
+/* Likewise, the assembler doesn't handle DWARF2 directives.  */
+#define DWARF2_UNWIND_INFO 0
+
 #undef MACHINE_TYPE
 #define MACHINE_TYPE "SGI running IRIX 5.x"
 

gcc/config/mips/iris5gas.h

@@ -7,6 +7,10 @@
 #define SDB_DEBUGGING_INFO
 #define MIPS_DEBUGGING_INFO
 
+/* GNU as does handle DWARF2 directives.  */
+#undef DWARF2_UNWIND_INFO
+#define DWARF2_UNWIND_INFO 1
+
 /* Irix 5 does not have some strange restrictions that Irix 3 had.  */
 #undef SET_FILE_NUMBER
 #define SET_FILE_NUMBER() ++num_source_filenames

gcc/config/mips/iris6.h

@@ -1,5 +1,5 @@
 /* Definitions of target machine for GNU compiler.  Iris version 6.
-   Copyright (C) 1994, 1995, 1996 Free Software Foundation, Inc.
+   Copyright (C) 1994, 1995, 1996, 1998 Free Software Foundation, Inc.
 
 This file is part of GNU CC.
 
@@ -542,5 +542,5 @@ do {									 \
 %{!static: \
   %{!shared: %{!non_shared: %{!call_shared: -call_shared -no_unresolved}}}} \
 %{rpath} -init __do_global_ctors -fini __do_global_dtors \
-%{shared:-hidden_symbol __do_global_ctors,__do_global_dtors,__EH_FRAME_BEGIN__} \
+%{shared:-hidden_symbol __do_global_ctors,__do_global_dtors,__EH_FRAME_BEGIN__,__frame_dummy} \
 -_SYSTYPE_SVR4 %{mabi=32: -32}%{mabi=n32: -n32}%{mabi=64: -64} %{!mabi*: -n32}"

gcc/config/mips/mips.c

@@ -193,9 +193,9 @@ enum processor_type mips_cpu;
 int mips_isa;
 
 #ifdef MIPS_ABI_DEFAULT
-/* which ABI to use.  This is defined to a constant in mips.h if the target
+/* Which ABI to use.  This is defined to a constant in mips.h if the target
    doesn't support multiple ABIs.  */
-enum mips_abi_type mips_abi;
+int mips_abi;
 #endif
 
 /* Strings to hold which cpu and instruction set architecture to use.  */
@@ -5008,9 +5008,32 @@ save_restore_insns (store_p, large_reg, large_offset, file)
 	  base_offset  = gp_offset;
 	  if (file == (FILE *)0)
 	    {
-	      insn = emit_move_insn (base_reg_rtx, GEN_INT (gp_offset));
-	      if (store_p)
-		RTX_FRAME_RELATED_P (insn) = 1;
+	      rtx gp_offset_rtx = GEN_INT (gp_offset);
+
+	      /* Instruction splitting doesn't preserve the RTX_FRAME_RELATED_P
+		 bit, so make sure that we don't emit anything that can be
+		 split.  */
+	      /* ??? There is no DImode ori immediate pattern, so we can only
+		 do this for 32 bit code.  */
+	      if (large_int (gp_offset_rtx)
+		  && GET_MODE (base_reg_rtx) == SImode)
+		{
+		  insn = emit_move_insn (base_reg_rtx,
+					 GEN_INT (gp_offset & 0xffff0000));
+		  if (store_p)
+		    RTX_FRAME_RELATED_P (insn) = 1;
+		  insn = emit_insn (gen_iorsi3 (base_reg_rtx, base_reg_rtx,
+						GEN_INT (gp_offset & 0x0000ffff)));
+		  if (store_p)
+		    RTX_FRAME_RELATED_P (insn) = 1;
+		}
+	      else
+		{
+		  insn = emit_move_insn (base_reg_rtx, gp_offset_rtx);
+		  if (store_p)
+		    RTX_FRAME_RELATED_P (insn) = 1;
+		}
+
 	      if (TARGET_LONG64)
 		insn = emit_insn (gen_adddi3 (base_reg_rtx, base_reg_rtx, stack_pointer_rtx));
 	      else
@@ -5128,7 +5151,32 @@ save_restore_insns (store_p, large_reg, large_offset, file)
 	  base_offset  = fp_offset;
 	  if (file == (FILE *)0)
 	    {
-	      insn = emit_move_insn (base_reg_rtx, GEN_INT (fp_offset));
+	      rtx fp_offset_rtx = GEN_INT (fp_offset);
+
+	      /* Instruction splitting doesn't preserve the RTX_FRAME_RELATED_P
+		 bit, so make sure that we don't emit anything that can be
+		 split.  */
+	      /* ??? There is no DImode ori immediate pattern, so we can only
+		 do this for 32 bit code.  */
+	      if (large_int (fp_offset_rtx)
+		  && GET_MODE (base_reg_rtx) == SImode)
+		{
+		  insn = emit_move_insn (base_reg_rtx,
+					 GEN_INT (fp_offset & 0xffff0000));
+		  if (store_p)
+		    RTX_FRAME_RELATED_P (insn) = 1;
+		  insn = emit_insn (gen_iorsi3 (base_reg_rtx, base_reg_rtx,
+						GEN_INT (fp_offset & 0x0000ffff)));
+		  if (store_p)
+		    RTX_FRAME_RELATED_P (insn) = 1;
+		}
+	      else
+		{
+		  insn = emit_move_insn (base_reg_rtx, fp_offset_rtx);
+		  if (store_p)
+		    RTX_FRAME_RELATED_P (insn) = 1;
+		}
+
 	      if (store_p)
 		RTX_FRAME_RELATED_P (insn) = 1;
 	      if (TARGET_LONG64)
@@ -5579,6 +5627,13 @@ mips_expand_epilogue ()
 	  else
 	    emit_insn (gen_movsi (stack_pointer_rtx, frame_pointer_rtx));
 	}
+      /* The GP/PIC register is implicitly used by all SYMBOL_REFs, so if we
+	 are going to restore it, then we must emit a blockage insn to
+	 prevent the scheduler from moving the restore out of the epilogue.  */
+      else if (TARGET_ABICALLS && mips_abi != ABI_32
+	       && (current_frame_info.mask
+		   & (1L << (PIC_OFFSET_TABLE_REGNUM - GP_REG_FIRST))))
+	emit_insn (gen_blockage ());
 
       save_restore_insns (FALSE, tmp_rtx, tsize, (FILE *)0);
 

gcc/config/mips/mips.h

@@ -77,21 +77,20 @@ enum processor_type {
 /* Recast the cpu class to be the cpu attribute.  */
 #define mips_cpu_attr ((enum attr_cpu)mips_cpu)
 
-/* Which ABI to use.  This is only used by the Irix 6 port currently.  */
+/* Which ABI to use.  These are constants because abi64.h must check their
+   value at preprocessing time.  */
 
-enum mips_abi_type {
-  ABI_32,
-  ABI_N32,
-  ABI_64,
-  ABI_EABI
-};
+#define ABI_32  0
+#define ABI_N32 1
+#define ABI_64  2
+#define ABI_EABI 3
 
 #ifndef MIPS_ABI_DEFAULT
 /* We define this away so that there is no extra runtime cost if the target
    doesn't support multiple ABIs.  */
 #define mips_abi ABI_32
 #else
-extern enum mips_abi_type mips_abi;
+extern int mips_abi;
 #endif
 
 /* Whether to emit abicalls code sequences or not.  */
@@ -2519,14 +2518,6 @@ typedef struct mips_args {
       register enum rtx_code code0 = GET_CODE (xplus0);			\
       register enum rtx_code code1 = GET_CODE (xplus1);			\
 									\
-      if (code0 != REG && code1 == REG)					\
-	{								\
-	  xplus0 = XEXP (xinsn, 1);					\
-	  xplus1 = XEXP (xinsn, 0);					\
-	  code0 = GET_CODE (xplus0);					\
-	  code1 = GET_CODE (xplus1);					\
-	}								\
-									\
       if (code0 == REG && REG_OK_FOR_BASE_P (xplus0))			\
 	{								\
 	  if (code1 == CONST_INT					\

gcc/config/mips/mips.md

@@ -2731,11 +2731,17 @@ move\\t%0,%z4\\n\\
 ;; operand zero, because then the address in the move instruction will be
 ;; clobbered.  We mark the scratch register as early clobbered to prevent this.
 
+;; We need the ?X in alternative 1 so that it will be choosen only if the
+;; destination is a floating point register.  Otherwise, alternative 1 can
+;; have lower cost than alternative 0 (because there is one less loser), and
+;; can be choosen when it won't work (because integral reloads into FP
+;; registers are not supported).
+
 (define_insn "fix_truncdfsi2"
   [(set (match_operand:SI 0 "general_operand" "=d,*f,R,o")
 	(fix:SI (match_operand:DF 1 "register_operand" "f,*f,f,f")))
    (clobber (match_scratch:SI 2 "=d,*d,&d,&d"))
-   (clobber (match_scratch:DF 3 "=f,*X,f,f"))]
+   (clobber (match_scratch:DF 3 "=f,?*X,f,f"))]
   "TARGET_HARD_FLOAT && TARGET_DOUBLE_FLOAT"
   "*
 {
@@ -2760,7 +2766,7 @@ move\\t%0,%z4\\n\\
   [(set (match_operand:SI 0 "general_operand" "=d,*f,R,o")
 	(fix:SI (match_operand:SF 1 "register_operand" "f,*f,f,f")))
    (clobber (match_scratch:SI 2 "=d,*d,&d,&d"))
-   (clobber (match_scratch:SF 3 "=f,*X,f,f"))]
+   (clobber (match_scratch:SF 3 "=f,?*X,f,f"))]
   "TARGET_HARD_FLOAT"
   "*
 {
@@ -2793,7 +2799,7 @@ move\\t%0,%z4\\n\\
 (define_insn "fix_truncdfdi2"
   [(set (match_operand:DI 0 "general_operand" "=d,*f,R,o")
 	(fix:DI (match_operand:DF 1 "register_operand" "f,*f,f,f")))
-   (clobber (match_scratch:DF 2 "=f,*X,f,f"))]
+   (clobber (match_scratch:DF 2 "=f,?*X,f,f"))]
   "TARGET_HARD_FLOAT && TARGET_64BIT && TARGET_DOUBLE_FLOAT"
   "*
 {
@@ -2820,7 +2826,7 @@ move\\t%0,%z4\\n\\
 (define_insn "fix_truncsfdi2"
   [(set (match_operand:DI 0 "general_operand" "=d,*f,R,o")
 	(fix:DI (match_operand:SF 1 "register_operand" "f,*f,f,f")))
-   (clobber (match_scratch:DF 2 "=f,*X,f,f"))]
+   (clobber (match_scratch:DF 2 "=f,?*X,f,f"))]
   "TARGET_HARD_FLOAT && TARGET_64BIT && TARGET_DOUBLE_FLOAT"
   "*
 {

gcc/config/mips/r3900.h

@@ -1,71 +0,0 @@
-/* Definitions of MIPS sub target machine for GNU compiler. 
-   Toshiba r3900.  You should include mips.h after this.
-
-   Copyright (C) 1989, 90-6, 1997 Free Software Foundation, Inc.
-   Contributed by Gavin Koch (gavin@cygnus.com).
-
-This file is part of GNU CC.
-
-GNU CC is free software; you can redistribute it and/or modify
-it under the terms of the GNU General Public License as published by
-the Free Software Foundation; either version 2, or (at your option)
-any later version.
-
-GNU CC 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 General Public License for more details.
-
-You should have received a copy of the GNU General Public License
-along with GNU CC; see the file COPYING.  If not, write to
-the Free Software Foundation, 59 Temple Place - Suite 330,
-Boston, MA 02111-1307, USA.  */
-
-#define SUBTARGET_CPP_SPEC "\
-%{!mabi=32: %{!mabi=n32: %{!mabi=64: -D__mips_eabi}}} \
-%{!msingle-float:-D__mips_soft_float} \
-%{mhard-float:%e-mhard-float not supported.} \
-%{msingle-float:%{msoft-float: \
-  %e-msingle-float and -msoft-float can not both be specified.}}"
-
-/* The following is needed because -mips3 and -mips4 set gp64 which in
-   combination with abi=eabi, causes long64 to be set. */
-#define SUBTARGET_CPP_SIZE_SPEC "\
-%{mips3:-D__SIZE_TYPE__=long\\ unsigned\\ int -D__PTRDIFF_TYPE__=long\\ int} \
-%{mips4:-D__SIZE_TYPE__=long\\ unsigned\\ int -D__PTRDIFF_TYPE__=long\\ int} \
-%{!mips3:%{!mips4:%{!m4650:\
-  -D__SIZE_TYPE__=unsigned\\ int -D__PTRDIFF_TYPE__=int}}} "
-
-/* by default (if not mips-something-else) produce code for the r3900 */
-#define SUBTARGET_CC1_SPEC "\
-%{mhard-float:%e-mhard-float not supported.} \
-%{msingle-float:%{msoft-float: \
-  %e-msingle-float and -msoft-float can not both be specified.}}"
-
-#define TARGET_DEFAULT (MASK_SOFT_FLOAT | MASK_MIPS3900)
-#define MIPS_CPU_STRING_DEFAULT "R3900"
-#define MIPS_ISA_DEFAULT 1
-
-#define MULTILIB_DEFAULTS { "EB", "mips1", "msoft-float" }
-
-/* We use the MIPS EABI by default.  */
-#define MIPS_ABI_DEFAULT ABI_EABI
-
-
-/* Debugging */
-
-#define DWARF2_DEBUGGING_INFO
-#define PREFERRED_DEBUGGING_TYPE DWARF2_DEBUG
-
-/* For the 'preferred' cases ("gN" and "ggdbN") we need to tell the 
-   gnu assembler "dwarf-2" */
-   
-#define SUBTARGET_ASM_DEBUGGING_SPEC "\
-%{!mmips-as: \
-  %{g:-gdwarf-2} %{g0:-gdwarf-2} %{g1:-gdwarf-2} %{g2:-gdwarf-2} %{g3:-gdwarf-2} \
-  %{ggdb:-gdwarf-2} %{ggdb0:-gdwarf-2} %{ggdb1:-gdwarf-2} %{ggdb2:-gdwarf-2} %{ggdb3:-gdwarf-2} \
-  %{gdwarf-2*:-gdwarf-2}} \
-%{gstabs:-g} %{gstabs0:-g0} %{gstabs1:-g1} %{gstabs2:-g2} %{gstabs3:-g3} \
-%{gstabs+:-g} %{gstabs+0:-g0} %{gstabs+1:-g1} %{gstabs+2:-g2} %{gstabs+3:-g3} \
-%{gcoff:-g} %{gcoff0:-g0} %{gcoff1:-g1} %{gcoff2:-g2} %{gcoff3:-g3}"
-

gcc/config/mips/x-iris

@@ -25,4 +25,4 @@ CLIB		= -lmld -lmalloc
 ALLOCA		= alloca.o
 
 # Find all of the declarations from the header files
-FIXPROTO_DEFINES= -D__EXTENSIONS__ -D_LANGUAGE_C_PLUS_PLUS
+FIXPROTO_DEFINES= -D__EXTENSIONS__ -D_SGI_SOURCE -D_LANGUAGE_C_PLUS_PLUS

gcc/config/ns32k/netbsd.h

@@ -105,3 +105,8 @@ Boston, MA 02111-1307, USA.
 
 #undef PCC_STATIC_STRUCT_RETURN
 #define DEFAULT_PCC_STRUCT_RETURN 0
+
+/* Until they use ELF or something that handles dwarf2 unwinds
+   and initialization stuff better.  */
+#define DWARF2_UNWIND_INFO 0
+

gcc/config/pa/pa-hpux10.h

@@ -69,3 +69,6 @@ do {  \
 #define MD_STARTFILE_PREFIX "/usr/ccs/lib/"
 #endif
 
+/* hpux10 has the new HP assembler.  It's still lousy, but it's a whole lot
+   better than the assembler shipped with older versions of hpux.  */
+#define NEW_HP_ASSEMBLER

gcc/config/pa/pa.c

@@ -1244,8 +1244,12 @@ emit_move_sequence (operands, mode, scratch_reg)
 	}
     }
 
-  /* Simplify the source if we need to.  */
+  /* Simplify the source if we need to.
+     Note we do have to handle function labels here, even though we do
+     not consider them legitimate constants.  Loop optimizations can
+     call the emit_move_xxx with one as a source.  */
   if ((GET_CODE (operand1) != HIGH && immediate_operand (operand1, mode))
+      || function_label_operand (operand1, mode)
       || (GET_CODE (operand1) == HIGH
 	  && symbolic_operand (XEXP (operand1, 0), mode)))
     {
@@ -1263,9 +1267,10 @@ emit_move_sequence (operands, mode, scratch_reg)
 
 	     So we force the plabel into memory, load operand0 from
 	     the memory location, then add in the constant part.  */
-	  if (GET_CODE (operand1) == CONST
-	      && GET_CODE (XEXP (operand1, 0)) == PLUS
-	      && function_label_operand (XEXP (XEXP (operand1, 0), 0), Pmode))
+	  if ((GET_CODE (operand1) == CONST
+	       && GET_CODE (XEXP (operand1, 0)) == PLUS
+	       && function_label_operand (XEXP (XEXP (operand1, 0), 0), Pmode))
+	      || function_label_operand (operand1, mode))
 	    {
 	      rtx temp, const_part;
 
@@ -1275,13 +1280,25 @@ emit_move_sequence (operands, mode, scratch_reg)
 	      else if (flag_pic)
 		scratch_reg = gen_reg_rtx (Pmode);
 
-	      /* Save away the constant part of the expression.  */
-	      const_part = XEXP (XEXP (operand1, 0), 1);
-	      if (GET_CODE (const_part) != CONST_INT)
-		abort ();
+	      if (GET_CODE (operand1) == CONST)
+		{
+		  /* Save away the constant part of the expression.  */
+		  const_part = XEXP (XEXP (operand1, 0), 1);
+		  if (GET_CODE (const_part) != CONST_INT)
+		    abort ();
 
-	      /* Force the function label into memory.  */
-	      temp = force_const_mem (mode, XEXP (XEXP (operand1, 0), 0));
+		  /* Force the function label into memory.  */
+		  temp = force_const_mem (mode, XEXP (XEXP (operand1, 0), 0));
+		}
+	      else
+		{
+		  /* No constant part.  */
+		  const_part = NULL_RTX;
+
+		  /* Force the function label into memory.  */
+		  temp = force_const_mem (mode, operand1);
+		}
+		
 
 	      /* Get the address of the memory location.  PIC-ify it if
 		 necessary.  */
@@ -1300,7 +1317,8 @@ emit_move_sequence (operands, mode, scratch_reg)
 	      emit_move_sequence (operands, mode, scratch_reg);
 
 	      /* And add back in the constant part.  */
-	      expand_inc (operand0, const_part);
+	      if (const_part != NULL_RTX)
+		expand_inc (operand0, const_part);
 
 	      return 1;
 	    }

gcc/config/pa/pa.h

@@ -1396,12 +1396,21 @@ extern struct rtx_def *hppa_builtin_saveregs ();
    && (reload_in_progress || reload_completed || ! symbolic_expression_p (X)))
 
 /* Include all constant integers and constant doubles, but not
-   floating-point, except for floating-point zero.  */
+   floating-point, except for floating-point zero.
 
+   Reject LABEL_REFs if we're not using gas or the new HP assembler.  */
+#ifdef NEW_HP_ASSEMBLER
 #define LEGITIMATE_CONSTANT_P(X)  		\
   ((GET_MODE_CLASS (GET_MODE (X)) != MODE_FLOAT	\
     || (X) == CONST0_RTX (GET_MODE (X)))	\
    && !function_label_operand (X, VOIDmode))
+#else
+#define LEGITIMATE_CONSTANT_P(X)  		\
+  ((GET_MODE_CLASS (GET_MODE (X)) != MODE_FLOAT	\
+    || (X) == CONST0_RTX (GET_MODE (X)))	\
+   && (GET_CODE (X) != LABEL_REF || TARGET_GAS)\
+   && !function_label_operand (X, VOIDmode))
+#endif
 
 /* Subroutine for EXTRA_CONSTRAINT.
 

gcc/config/pa/pa.md

@@ -1,6 +1,5 @@
 ;;- Machine description for HP PA-RISC architecture for GNU C compiler
-;;   Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997
-;;   Free Software Foundation, Inc.
+;;   Copyright (C) 1992, 93, 94, 95, 96, 1997 Free Software Foundation, Inc.
 ;;   Contributed by the Center for Software Science at the University
 ;;   of Utah.
 
@@ -1432,7 +1431,7 @@
 
 (define_insn "pre_ldwm"
   [(set (match_operand:SI 0 "register_operand" "=r")
-	(mem:SI (plus:SI (match_operand:SI 1 "register_operand" "=r")
+	(mem:SI (plus:SI (match_operand:SI 1 "register_operand" "+r")
 			 (match_operand:SI 2 "pre_cint_operand" ""))))
    (set (match_dup 1)
 	(plus:SI (match_dup 1) (match_dup 2)))]
@@ -1447,7 +1446,7 @@
    (set_attr "length" "4")])
 
 (define_insn "pre_stwm"
-  [(set (mem:SI (plus:SI (match_operand:SI 0 "register_operand" "=r")
+  [(set (mem:SI (plus:SI (match_operand:SI 0 "register_operand" "+r")
 			 (match_operand:SI 1 "pre_cint_operand" "")))
 	(match_operand:SI 2 "reg_or_0_operand" "rM"))
    (set (match_dup 0)
@@ -1464,7 +1463,7 @@
 
 (define_insn "post_ldwm"
   [(set (match_operand:SI 0 "register_operand" "=r")
-	(mem:SI (match_operand:SI 1 "register_operand" "=r")))
+	(mem:SI (match_operand:SI 1 "register_operand" "+r")))
    (set (match_dup 1)
 	(plus:SI (match_dup 1)
 		 (match_operand:SI 2 "post_cint_operand" "")))]
@@ -1479,7 +1478,7 @@
    (set_attr "length" "4")])
 
 (define_insn "post_stwm"
-  [(set (mem:SI (match_operand:SI 0 "register_operand" "=r"))
+  [(set (mem:SI (match_operand:SI 0 "register_operand" "+r"))
 	(match_operand:SI 1 "reg_or_0_operand" "rM"))
    (set (match_dup 0)
 	(plus:SI (match_dup 0)
@@ -1808,7 +1807,7 @@
 
 (define_insn ""
   [(set (match_operand:HI 0 "register_operand" "=r")
-	(mem:HI (plus:SI (match_operand:SI 1 "register_operand" "=r")
+	(mem:HI (plus:SI (match_operand:SI 1 "register_operand" "+r")
 			 (match_operand:SI 2 "int5_operand" "L"))))
    (set (match_dup 1)
 	(plus:SI (match_dup 1) (match_dup 2)))]
@@ -1822,7 +1821,7 @@
   [(set (match_operand:SI 0 "register_operand" "=r")
 	(zero_extend:SI (mem:HI
 			  (plus:SI
-			    (match_operand:SI 1 "register_operand" "=r")
+			    (match_operand:SI 1 "register_operand" "+r")
 			    (match_operand:SI 2 "int5_operand" "L")))))
    (set (match_dup 1)
 	(plus:SI (match_dup 1) (match_dup 2)))]
@@ -1832,7 +1831,7 @@
    (set_attr "length" "4")])
 
 (define_insn ""
-  [(set (mem:HI (plus:SI (match_operand:SI 0 "register_operand" "=r")
+  [(set (mem:HI (plus:SI (match_operand:SI 0 "register_operand" "+r")
 			 (match_operand:SI 1 "int5_operand" "L")))
 	(match_operand:HI 2 "reg_or_0_operand" "rM"))
    (set (match_dup 0)
@@ -2012,7 +2011,7 @@
 
 (define_insn ""
   [(set (match_operand:QI 0 "register_operand" "=r")
-	(mem:QI (plus:SI (match_operand:SI 1 "register_operand" "=r")
+	(mem:QI (plus:SI (match_operand:SI 1 "register_operand" "+r")
 			 (match_operand:SI 2 "int5_operand" "L"))))
    (set (match_dup 1) (plus:SI (match_dup 1) (match_dup 2)))]
   ""
@@ -2024,7 +2023,7 @@
 (define_insn ""
   [(set (match_operand:SI 0 "register_operand" "=r")
 	(zero_extend:SI (mem:QI (plus:SI
-				  (match_operand:SI 1 "register_operand" "=r")
+				  (match_operand:SI 1 "register_operand" "+r")
 				  (match_operand:SI 2 "int5_operand" "L")))))
    (set (match_dup 1) (plus:SI (match_dup 1) (match_dup 2)))]
   ""
@@ -2035,7 +2034,7 @@
 (define_insn ""
   [(set (match_operand:HI 0 "register_operand" "=r")
 	(zero_extend:HI (mem:QI (plus:SI
-				  (match_operand:SI 1 "register_operand" "=r")
+				  (match_operand:SI 1 "register_operand" "+r")
 				  (match_operand:SI 2 "int5_operand" "L")))))
    (set (match_dup 1) (plus:SI (match_dup 1) (match_dup 2)))]
   ""
@@ -2044,7 +2043,7 @@
    (set_attr "length" "4")])
 
 (define_insn ""
-  [(set (mem:QI (plus:SI (match_operand:SI 0 "register_operand" "=r")
+  [(set (mem:QI (plus:SI (match_operand:SI 0 "register_operand" "+r")
 			 (match_operand:SI 1 "int5_operand" "L")))
 	(match_operand:QI 2 "reg_or_0_operand" "rM"))
    (set (match_dup 0)

gcc/config/rs6000/aix41.h

@@ -36,6 +36,37 @@ Boston, MA 02111-1307, USA.  */
 #undef ASM_SPEC
 #define ASM_SPEC "-u %(asm_cpu)"
 
+/* Common ASM definitions used by ASM_SPEC amonst the various targets
+   for handling -mcpu=xxx switches.  */
+
+#undef ASM_CPU_SPEC
+#define ASM_CPU_SPEC \
+"%{!mcpu*: \
+  %{mpower: %{!mpower2: -mpwr}} \
+  %{mpower2: -mpwr2} \
+  %{mpowerpc*: -mppc} \
+  %{mno-power: %{!mpowerpc*: -mcom}} \
+  %{!mno-power: %{!mpower2: %(asm_default)}}} \
+%{mcpu=common: -mcom} \
+%{mcpu=power: -mpwr} \
+%{mcpu=power2: -mpwr2} \
+%{mcpu=powerpc: -mppc} \
+%{mcpu=rios: -mpwr} \
+%{mcpu=rios1: -mpwr} \
+%{mcpu=rios2: -mpwr2} \
+%{mcpu=rsc: -mpwr} \
+%{mcpu=rsc1: -mpwr} \
+%{mcpu=403: -mppc} \
+%{mcpu=505: -mppc} \
+%{mcpu=601: -m601} \
+%{mcpu=602: -mppc} \
+%{mcpu=603: -m603} \
+%{mcpu=603e: -m603} \
+%{mcpu=604: -m604} \
+%{mcpu=620: -mppc} \
+%{mcpu=821: -mppc} \
+%{mcpu=860: -mppc}"
+
 #undef CPP_PREDEFINES
 #define CPP_PREDEFINES "-D_IBMR2 -D_POWER -D_AIX -D_AIX32 -D_AIX41 \
 -Asystem(unix) -Asystem(aix)"

gcc/config/rs6000/linux.h

@@ -1,6 +1,6 @@
 /* Definitions of target machine for GNU compiler,
    for IBM RS/6000 running AIX version 3.1.
-   Copyright (C) 1996, 1997 Free Software Foundation, Inc.
+   Copyright (C) 1996, 1997, 1998 Free Software Foundation, Inc.
    Contributed by Michael Meissner (meissner@cygnus.com).
 
 This file is part of GNU CC.
@@ -23,12 +23,14 @@ the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.  */
 
 #undef CPP_PREDEFINES
 #define CPP_PREDEFINES \
- "-DPPC -Dunix -Dlinux -Dpowerpc -Asystem(unix) -Asystem(linux) -Acpu(powerpc) -Amachine(powerpc)"
+ "-DPPC -D__ELF__ -Dunix -Dlinux -Dpowerpc -Asystem(unix) -Asystem(linux) -Acpu(powerpc) -Amachine(powerpc)"
 
 #undef LINK_SPEC
 #define LINK_SPEC "-m elf32ppc %{shared:-shared} \
   %{!shared: \
-    %{!static: %{rdynamic:-export-dynamic}} \
+    %{!static: \
+      %{rdynamic:-export-dynamic} \
+      %{!dynamic-linker:-dynamic-linker /lib/ld.so.1}} \
     %{static:-static}}"
 
 #undef	LIB_DEFAULT_SPEC
@@ -62,4 +64,6 @@ the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.  */
 #define	MULTILIB_DEFAULTS { "mbig", "mcall-linux" }
 
 #undef DEFAULT_VTABLE_THUNKS
+#ifndef USE_GNULIBC_1
 #define DEFAULT_VTABLE_THUNKS 1
+#endif