Unreviewed 2.95.4 patches since April 1st

[David Korn]

    Hi Bernd and list,

  I've got the following patches outstanding since the start of the month.

http://gcc.gnu.org/ml/gcc-patches/2001-04/msg00015.html

  Uses fixincs to solve a vxworks header problem which leads to va_list 
being undefined.

http://gcc.gnu.org/ml/gcc-patches/2001-04/msg00022.html

  Adds a new target-specific option to the rs6000 backend, which maintains
backward compatilbility with the 2.7.2 implementation of SysV4/EABI.  This
is vital for compatibility with the VxWorks OS libs, which were compiled
under that version.

  There is one possible flaw in this patch; I've noticed since submitting
it that I didn't supply documentation for the option.  However I could
write a few lines at a moment's notice; I'd just like it to be looked over
by an rs6k backend maintainer as I believe a visual inspection should 
easily confirm it won't break existing targets.

  I've got a bunch of other VxWorks-Ppc patches queued up behind these
two, all of which fix testsuite failures.  I can supply further information,
analysis, detailed testsuite results, as required.

  There's also one non-VxWorks-related patch:

http://gcc.gnu.org/ml/gcc-patches/2001-03/msg02033.html

  Fixes the longstanding bug that causes cross compilers to output files
with .exe extensions if that is appropriate for the host, rather than the
target.  This is a backport of DJ Delorie's patch which he posted in
http://gcc.gnu.org/ml/gcc-patches/2001-04/msg00930.html
and which was approved by rth in
http://gcc.gnu.org/ml/gcc-patches/2001-04/msg00957.html

  There is however a minor typo in the ChangeLog entry in that post;
I've appended a corrected version at the end of this post.  The only 
difference is that I forgot to mention tm.texi as being updated as well as
gcc.texi, so I've reworded that entry into two separate entries in the
corrected version at the end of this post.

       DaveK

Here's that corrected changelog entry:

================================================================
2001-04-01 DJ Delorie <dj@redhat.com>, backported from HEAD by Dave Korn
<davek-ml@ntlworld.com>

* config/alpha/vms.h: Change OBJECT_SUFFIX and EXECUTABLE_SUFFIX
to TARGET_OBJECT_SUFFIX and TARGET_EXECUTABLE_SUFFIX.
* config/i386/cygwin.h: Likewise.
* config/i386/mingw32.h: Likewise.
* config/vax/vms.h: Likewise.

* config/alpha/xm-vms.h: Change OBJECT_SUFFIX and EXECUTABLE_SUFFIX
to HOST_OBJECT_SUFFIX and HOST_EXECUTABLE_SUFFIX.
* config/i386/xm-cygwin.h: Likewise.
* config/i386/xm-djgpp.h: Likewise.
* config/i386/xm-dos.h: Likewise.
* config/i386/xm-mingw32.h: Likewise.
* config/i386/xm-os2.h: Likewise.
* config/rs6000/xm-cygwin.h: Likewise.
* config/vax/xm-vms.h: Likewise.
* config/winnt/xm-winnt.h: Likewise.

* cccp.c (main): Use TARGET_OBJECT_SUFFIX instead of OBJECT_SUFFIX.
* cppinit.c (initialize_dependency_output): Likewise.
* collect2.c (find_a_file): Look for files matching the extension
HOST_EXECUTABLE_SUFFIX instead of EXECUTABLE_SUFFIX.
* gcc.c (DEFAULT_SWITCH_CURTAILS_COMPILATION): Depend on 
TARGET_EXECUTABLE_SUFFIX.
(find_a_file): Use HOST_EXECUTABLE_SUFFIX.
(convert_filename): Use TARGET_ suffixes throughout.
(process_command): Likewise.
(do_spec_1): Likewise.
* java/lang.c (init_parse): Likewise.
* gcc.texi: Document new macros HOST_(EXECUTABLE,OBJECT)_SUFFIX.  Remove
documentation for deleted macros OBJECT_SUFFIX and EXECUTABLE_SUFFIX.
* tm.texi: Document new macros TARGET_(EXECUTABLE,OBJECT)_SUFFIX.



**********************************************************************
This email and any files transmitted with it are confidential and
intended solely for the use of the individual or entity to whom they
are addressed. If you have received this email in error please notify
the system manager.

This footnote also confirms that this email message has been swept by
MIMEsweeper for the presence of computer viruses.

www.mimesweeper.com
**********************************************************************