# Pre-4.2 versions of Zsh do word splitting on ${1+"$@"}, which
# Prefer a ksh shell builtin over an external printf program on Solaris,
  # Pre-4.2 versions of Zsh do word splitting on \${1+\"\$@\"}, which
  $as_echo "$0: the shells that I found on your system."
# Perform arithmetic evaluation on the ARGs, and store the result in the
  # original and so on.  Autoconf is especially sensitive to this).
  *)   echo `echo ksh88 bug on AIX 6.1` > /dev/null
# hostname on some systems (SVR3.2, old GNU/Linux) returns a bogus exit status,
# and all the variables that are supposed to be based on exec_prefix
# There might be people who depend on the old broken behavior: `$host'
  --program-transform-name=PROGRAM   run sed PROGRAM on installed program names
  --build=BUILD     configure for building on BUILD [guessed]
  --host=HOST       cross-compile to build programs to run on HOST [BUILD]
    <limits.h> exists even on freestanding compilers.  */
# AmigaOS /C/install, which installs bootblocks on floppy discs
# We need awk for the "check" target.  The system "awk" is bad on
  # instance it was reported that on HP-UX the gcc test will end up
    # like to wrap large dependency lists on column 80 (with \), and
      # icc doesn't choke on unknown options, it will just issue warnings
# FIXME: we rely on the cache variable name because
  # instance it was reported that on HP-UX the gcc test will end up
    # like to wrap large dependency lists on column 80 (with \), and
      # icc doesn't choke on unknown options, it will just issue warnings
	# Adding the `sed 1q' prevents false positives on HP-UX, which says:
    # all of the supported platforms: 8192 (on NT/2K/XP).
    # due to this test when exec_disable_arg_limit is 1 on Tru64. It is not
      # Add a significant safety factor because C++ compilers can tack on
# Need to set the preceding variable on all platforms that support
# If you have `file' or equivalent on your system and you're not sure
  # func_win32_libid shell function, so use a weaker test based on 'objdump',
  # use the weaker test based on 'objdump'. See mingw*.
  # PIC code is broken on Interix 3.x, that's why |\.a not |_pic\.a here
# These are sane defaults that work on at least a few old systems.
# Some systems (esp. on ia64) link data and code symbols differently,
    darwin*) # darwin 5.x on
      # if running on 10.5 or later, the deployment target defaults
      # to the OS version, if on x86, and 10.4, the deployment
  # <limits.h> exists even on freestanding compilers.
  # Broken: fails on valid input.
  # OK, works on sane cases.  Now check whether nonexistent headers
  # Broken: success on invalid input.
  # <limits.h> exists even on freestanding compilers.
  # Broken: fails on valid input.
  # OK, works on sane cases.  Now check whether nonexistent headers
  # Broken: success on invalid input.
  # instance it was reported that on HP-UX the gcc test will end up
    # like to wrap large dependency lists on column 80 (with \), and
      # icc doesn't choke on unknown options, it will just issue warnings
  # <limits.h> exists even on freestanding compilers.
  # Broken: fails on valid input.
  # OK, works on sane cases.  Now check whether nonexistent headers
  # Broken: success on invalid input.
  # <limits.h> exists even on freestanding compilers.
  # Broken: fails on valid input.
  # OK, works on sane cases.  Now check whether nonexistent headers
  # Broken: success on invalid input.
*** depend on shared libraries, but programs linked with such libtool
*** depend on shared libraries, but programs linked with such libtool
      # PIC is the default on this platform
      # on systems that don't support them.
	# IBM XL C 8.0/Fortran 10.1 on PPC
  # rely on this symbol name, it's probably fine to never include it in
*** to be unable to reliably create shared libraries on AIX.
	ecc*,ia64* | icc*,ia64*)	# Intel C compiler on ia64
	efc*,ia64* | ifort*,ia64*)	# Intel Fortran compiler on ia64
	xl[cC]*)			# IBM XL C 8.0 on PPC (deal with xlf below)
	  # IBM XL Fortran 10.1 on PPC cannot create shared libs itself
*** create shared libraries on Solaris systems.  Therefore, libtool
*** reliably create shared libraries on SCO systems.  Therefore, libtool
	# We only want to do this on AIX 4.2 and lower, the check
	# chokes on -Wl,-G. The following line is correct:
      # Test whether the compiler implicitly links with -lc since on some
    # depend on `.', always an invalid library.  This was fixed in
    # AIX (on Power*) has no versioning support, so currently we can not hardcode correct
      # If using run time linking (on AIX 4.2 or later) use lib<name>.so
        # mingw gcc, but we are running on Cygwin. Gcc prints its search
  *) # from 4.6 on, and DragonFly
  # We used to test for /lib/ld.so.1 and disable shared libraries on
# are currently assumed to apply to all compilers on this platform,
# and will be corrupted by setting them based on a non-working compiler.
        # overridden to take advantage of the native linker features on
        # the platform it is being used on.
          # We only want to do this on AIX 4.2 and lower, the check
	  # chokes on -Wl,-G. The following line is correct:
	    # version 8.0 and above of icpc choke on multiply defined symbols
	    # IBM XL 8.0 on PPC, with GNU ld
	      # Not sure whether something based on
	        # g++ 2.7 appears to require `-G' NOT `-shared' on this
# PORTME: override above test on systems where it is broken
    # -library=stlport4 depends on it.
    # -library=stlport4 depends on it.
      # PIC is the default on this platform
	    # IBM XL 8.0 on PPC
      # Test whether the compiler implicitly links with -lc since on some
    # depend on `.', always an invalid library.  This was fixed in
    # AIX (on Power*) has no versioning support, so currently we can not hardcode correct
      # If using run time linking (on AIX 4.2 or later) use lib<name>.so
        # mingw gcc, but we are running on Cygwin. Gcc prints its search
  *) # from 4.6 on, and DragonFly
  # We used to test for /lib/ld.so.1 and disable shared libraries on
# Turn on all warnings for gcc.  Turn off optimisation for the test build.
# tests run on this system so they can be shared between configure
# It is not useful on other systems.  If it contains results you don't
  # Pre-4.2 versions of Zsh do word splitting on ${1+"$@"}, which
# Prefer a ksh shell builtin over an external printf program on Solaris,
# Perform arithmetic evaluation on the ARGs, and store the result in the
  *)   echo `echo ksh88 bug on AIX 6.1` > /dev/null
# Save the log message, to keep $0 and so on meaningful, and to
on `(hostname || uname -n) 2>/dev/null | sed 1q`
# See if we are running on zsh, and set the options which allow our
# bizarre bug on SunOS 4.1.3.
    # on some systems where configure will not decide to define it.
    # each Makefile.in and add a new line on top of each file to say so.
    # See if we are running on zsh, and set the options which allow our
# Libtool was configured on host `(hostname || uname -n) 2>/dev/null | sed 1q`:
# Tool to manipulate archived DWARF debug symbol files on Mac OS X.
# Tool to change global to local symbols on Mac OS X.
# Tool to manipulate fat objects and archives on Mac OS X.
# ldd/readelf like tool for Mach-O binaries on Mac OS X.
# ldd/readelf like tool for 64 bit Mach-O binaries on Mac OS X 10.4.
  # We use sed instead of cat because bash on DJGPP gets confused if
  # is reportedly fixed, but why not run on old versions too?
# Unfortunately, on DOS this fails, as config.log is still kept open