ffi 1.0.11 → 1.0.12.pre

data/ext/ffi_c/libffi/msvcc.sh

@@ -0,0 +1,197 @@
+#!/bin/sh
+
+# ***** BEGIN LICENSE BLOCK *****
+# Version: MPL 1.1/GPL 2.0/LGPL 2.1
+#
+# The contents of this file are subject to the Mozilla Public License Version
+# 1.1 (the "License"); you may not use this file except in compliance with
+# the License. You may obtain a copy of the License at
+# http://www.mozilla.org/MPL/
+#
+# Software distributed under the License is distributed on an "AS IS" basis,
+# WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
+# for the specific language governing rights and limitations under the
+# License.
+#
+# The Original Code is the MSVC wrappificator.
+#
+# The Initial Developer of the Original Code is
+# Timothy Wall <twalljava@dev.java.net>.
+# Portions created by the Initial Developer are Copyright (C) 2009
+# the Initial Developer. All Rights Reserved.
+#
+# Contributor(s):
+#   Daniel Witte <dwitte@mozilla.com>
+#
+# Alternatively, the contents of this file may be used under the terms of
+# either the GNU General Public License Version 2 or later (the "GPL"), or
+# the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
+# in which case the provisions of the GPL or the LGPL are applicable instead
+# of those above. If you wish to allow use of your version of this file only
+# under the terms of either the GPL or the LGPL, and not to allow others to
+# use your version of this file under the terms of the MPL, indicate your
+# decision by deleting the provisions above and replace them with the notice
+# and other provisions required by the GPL or the LGPL. If you do not delete
+# the provisions above, a recipient may use your version of this file under
+# the terms of any one of the MPL, the GPL or the LGPL.
+#
+# ***** END LICENSE BLOCK *****
+
+#
+# GCC-compatible wrapper for cl.exe and ml.exe. Arguments are given in GCC
+# format and translated into something sensible for cl or ml.
+#
+
+args="-nologo -W3"
+md=-MD
+cl="cl"
+ml="ml"
+safeseh="-safeseh"
+output=
+
+while [ $# -gt 0 ]
+do
+  case $1
+  in
+    -fexceptions)
+      # Don't enable exceptions for now.
+      #args="$args -EHac"
+      shift 1
+    ;;
+    -m32)
+      shift 1
+    ;;
+    -m64)
+      cl="cl"   # "$MSVC/x86_amd64/cl"
+      ml="ml64" # "$MSVC/x86_amd64/ml64"
+      safeseh=
+      shift 1
+    ;;
+    -O0)
+      args="$args -Od"
+      shift 1
+    ;;
+    -O*)
+      # If we're optimizing, make sure we explicitly turn on some optimizations
+      # that are implicitly disabled by debug symbols (-Zi).
+      args="$args $1 -OPT:REF -OPT:ICF -INCREMENTAL:NO"
+      shift 1
+    ;;
+    -g)
+      # Enable debug symbol generation.
+      args="$args -Zi -DEBUG"
+      shift 1
+    ;;
+    -DFFI_DEBUG)
+      # Link against debug CRT and enable runtime error checks.
+      args="$args -RTC1"
+      defines="$defines $1"
+      md=-MDd
+      shift 1
+    ;;
+    -c)
+      args="$args -c"
+      args="$(echo $args | sed 's%/Fe%/Fo%g')"
+      single="-c"
+      shift 1
+    ;;
+    -D*=*)
+      name="$(echo $1|sed 's/-D\([^=][^=]*\)=.*/\1/g')"
+      value="$(echo $1|sed 's/-D[^=][^=]*=//g')"
+      args="$args -D${name}='$value'"
+      defines="$defines -D${name}='$value'"
+      shift 1
+    ;;
+    -D*)
+      args="$args $1"
+      defines="$defines $1"
+      shift 1
+    ;;
+    -I)
+      args="$args -I$2"
+      includes="$includes -I$2"
+      shift 2
+    ;;
+    -I*)
+      args="$args $1"
+      includes="$includes $1"
+      shift 1
+    ;;
+    -W|-Wextra)
+      # TODO map extra warnings
+      shift 1
+    ;;
+    -Wall)
+      # -Wall on MSVC is overzealous, and we already build with -W3. Nothing
+      # to do here.
+      shift 1
+    ;;
+    -Werror)
+      args="$args -WX"
+      shift 1
+    ;;
+    -W*)
+      # TODO map specific warnings
+      shift 1
+    ;;
+    -S)
+      args="$args -FAs"
+      shift 1
+    ;;
+    -o)
+      outdir="$(dirname $2)"
+      base="$(basename $2|sed 's/\.[^.]*//g')"
+      if [ -n "$single" ]; then 
+        output="-Fo$2"
+      else
+        output="-Fe$2"
+      fi
+      if [ -n "$assembly" ]; then
+        args="$args $output"
+      else
+        args="$args $output -Fd$outdir/$base -Fp$outdir/$base -Fa$outdir/$base"
+      fi
+      shift 2
+    ;;
+    *.S)
+      src=$1
+      assembly="true"
+      shift 1
+    ;;
+    *.c)
+      args="$args $1"
+      shift 1
+    ;;
+    *)
+      # Assume it's an MSVC argument, and pass it through.
+      args="$args $1"
+      shift 1
+    ;;
+  esac
+done
+
+if [ -n "$assembly" ]; then
+    if [ -z "$outdir" ]; then
+      outdir="."
+    fi
+    ppsrc="$outdir/$(basename $src|sed 's/.S$/.asm/g')"
+    echo "$cl -nologo -EP $includes $defines $src > $ppsrc"
+    "$cl" -nologo -EP $includes $defines $src > $ppsrc || exit $?
+    output="$(echo $output | sed 's%/F[dpa][^ ]*%%g')"
+    args="-nologo $safeseh $single $output $ppsrc"
+
+    echo "$ml $args"
+    eval "\"$ml\" $args"
+    result=$?
+
+    # required to fix ml64 broken output?
+    #mv *.obj $outdir
+else
+    args="$md $args"
+    echo "$cl $args"
+    eval "\"$cl\" $args"
+    result=$?
+fi
+
+exit $result
+

data/ext/ffi_c/libffi/src/alpha/osf.S

@@ -1,5 +1,5 @@
 /* -----------------------------------------------------------------------
-   osf.S - Copyright (c) 1998, 2001, 2007, 2008 Red Hat
+   osf.S - Copyright (c) 1998, 2001, 2007, 2008, 2011 Red Hat
    
    Alpha/OSF Foreign Function Interface 
 
@@ -298,34 +298,52 @@ $load_table:
 #error "osf.S out of sync with ffi.h"
 #endif
 
+#ifdef __ELF__
+# define UA_SI		.4byte
+# define FDE_ENCODING	0x1b	/* pcrel sdata4 */
+# define FDE_ENCODE(X)	.4byte X-.
+# define FDE_ARANGE(X)	.4byte X
+#elif defined __osf__
+# define UA_SI		.align 0; .long
+# define FDE_ENCODING	0x50	/* aligned absolute */
+# define FDE_ENCODE(X)	.align 3; .quad X
+# define FDE_ARANGE(X)	.align 0; .quad X
+#endif
+
 #ifdef __ELF__
 	.section	.eh_frame,EH_FRAME_FLAGS,@progbits
+#elif defined __osf__
+	.data
+	.align 3
+	.globl _GLOBAL__F_ffi_call_osf
+_GLOBAL__F_ffi_call_osf:
+#endif
 __FRAME_BEGIN__:
-	.4byte	$LECIE1-$LSCIE1	# Length of Common Information Entry
+	UA_SI	$LECIE1-$LSCIE1	# Length of Common Information Entry
 $LSCIE1:
-	.4byte	0x0		# CIE Identifier Tag
+	UA_SI	0x0		# CIE Identifier Tag
 	.byte	0x1		# CIE Version
 	.ascii "zR\0"		# CIE Augmentation
 	.byte	0x1		# uleb128 0x1; CIE Code Alignment Factor
 	.byte	0x78		# sleb128 -8; CIE Data Alignment Factor
 	.byte	26		# CIE RA Column
 	.byte	0x1		# uleb128 0x1; Augmentation size
-	.byte	0x1b		# FDE Encoding (pcrel sdata4)
+	.byte	FDE_ENCODING	# FDE Encoding
 	.byte	0xc		# DW_CFA_def_cfa
 	.byte	30		# uleb128 column 30
 	.byte	0		# uleb128 offset 0
 	.align 3
 $LECIE1:
 $LSFDE1:
-	.4byte	$LEFDE1-$LASFDE1		# FDE Length
+	UA_SI	$LEFDE1-$LASFDE1		# FDE Length
 $LASFDE1:
-	.4byte	$LASFDE1-__FRAME_BEGIN__	# FDE CIE offset
-	.4byte	$LFB1-.		# FDE initial location
-	.4byte	$LFE1-$LFB1	# FDE address range
+	UA_SI	$LASFDE1-__FRAME_BEGIN__	# FDE CIE offset
+	FDE_ENCODE($LFB1)			# FDE initial location
+	FDE_ARANGE($LFE1-$LFB1)			# FDE address range
 	.byte	0x0		# uleb128 0x0; Augmentation size
 
 	.byte	0x4		# DW_CFA_advance_loc4
-	.4byte	$LCFI1-$LFB1
+	UA_SI	$LCFI1-$LFB1
 	.byte	0x9a		# DW_CFA_offset, column 26
 	.byte	4		# uleb128 4*-8
 	.byte	0x8f		# DW_CFA_offset, column 15
@@ -335,32 +353,35 @@ $LASFDE1:
 	.byte	32		# uleb128 offset 32
 
 	.byte	0x4		# DW_CFA_advance_loc4
-	.4byte	$LCFI2-$LCFI1
+	UA_SI	$LCFI2-$LCFI1
 	.byte	0xda		# DW_CFA_restore, column 26
 	.align 3
 $LEFDE1:
 
 $LSFDE3:
-	.4byte	$LEFDE3-$LASFDE3		# FDE Length
+	UA_SI	$LEFDE3-$LASFDE3		# FDE Length
 $LASFDE3:
-	.4byte	$LASFDE3-__FRAME_BEGIN__	# FDE CIE offset
-	.4byte	$LFB2-.		# FDE initial location
-	.4byte	$LFE2-$LFB2	# FDE address range
+	UA_SI	$LASFDE3-__FRAME_BEGIN__	# FDE CIE offset
+	FDE_ENCODE($LFB2)			# FDE initial location
+	FDE_ARANGE($LFE2-$LFB2)			# FDE address range
 	.byte	0x0		# uleb128 0x0; Augmentation size
 
 	.byte	0x4		# DW_CFA_advance_loc4
-	.4byte	$LCFI5-$LFB2
+	UA_SI	$LCFI5-$LFB2
 	.byte	0xe		# DW_CFA_def_cfa_offset
 	.byte	0x80,0x1	# uleb128 128
 
 	.byte	0x4		# DW_CFA_advance_loc4
-	.4byte	$LCFI6-$LCFI5
+	UA_SI	$LCFI6-$LCFI5
 	.byte	0x9a		# DW_CFA_offset, column 26
 	.byte	16		# uleb128 offset 16*-8
 	.align 3
 $LEFDE3:
+#if defined __osf__
+	.align 0
+	.long	0		# End of Table
+#endif
 
-#ifdef __linux__
+#if defined __ELF__ && defined __linux__
 	.section	.note.GNU-stack,"",@progbits
 #endif
-#endif

data/ext/ffi_c/libffi/src/arm/ffi.c

@@ -1,6 +1,9 @@
 /* -----------------------------------------------------------------------
-   ffi.c - Copyright (c) 1998, 2008  Red Hat, Inc.
-   
+   ffi.c - Copyright (c) 2011 Plausible Labs Cooperative, Inc.
+           Copyright (c) 2011 Anthony Green
+	   Copyright (c) 2011 Free Software Foundation
+           Copyright (c) 1998, 2008, 2011  Red Hat, Inc.
+	   
    ARM Foreign Function Interface 
 
    Permission is hereby granted, free of charge, to any person obtaining
@@ -29,12 +32,20 @@
 
 #include <stdlib.h>
 
-/* ffi_prep_args is called by the assembly routine once stack space
-   has been allocated for the function's arguments */
+/* Forward declares. */
+static int vfp_type_p (ffi_type *);
+static void layout_vfp_args (ffi_cif *);
 
-void ffi_prep_args(char *stack, extended_cif *ecif)
+/* ffi_prep_args is called by the assembly routine once stack space
+   has been allocated for the function's arguments
+   
+   The vfp_space parameter is the load area for VFP regs, the return
+   value is cif->vfp_used (word bitset of VFP regs used for passing
+   arguments). These are only used for the VFP hard-float ABI.
+*/
+int ffi_prep_args(char *stack, extended_cif *ecif, float *vfp_space)
 {
-  register unsigned int i;
+  register unsigned int i, vi = 0;
   register void **p_argv;
   register char *argp;
   register ffi_type **p_arg;
@@ -54,6 +65,21 @@ void ffi_prep_args(char *stack, extended_cif *ecif)
     {
       size_t z;
 
+      /* Allocated in VFP registers. */
+      if (ecif->cif->abi == FFI_VFP
+	  && vi < ecif->cif->vfp_nargs && vfp_type_p (*p_arg))
+	{
+	  float* vfp_slot = vfp_space + ecif->cif->vfp_args[vi++];
+	  if ((*p_arg)->type == FFI_TYPE_FLOAT)
+	    *((float*)vfp_slot) = *((float*)*p_argv);
+	  else if ((*p_arg)->type == FFI_TYPE_DOUBLE)
+	    *((double*)vfp_slot) = *((double*)*p_argv);
+	  else
+	    memcpy(vfp_slot, *p_argv, (*p_arg)->size);
+	  p_argv++;
+	  continue;
+	}
+
       /* Align if necessary */
       if (((*p_arg)->alignment - 1) & (unsigned) argp) {
 	argp = (char *) ALIGN(argp, (*p_arg)->alignment);
@@ -103,13 +129,15 @@ void ffi_prep_args(char *stack, extended_cif *ecif)
 	  p_argv++;
 	  argp += z;
     }
-  
-  return;
+
+  /* Indicate the VFP registers used. */
+  return ecif->cif->vfp_used;
 }
 
 /* Perform machine dependent cif processing */
 ffi_status ffi_prep_cif_machdep(ffi_cif *cif)
 {
+  int type_code;
   /* Round the stack up to a multiple of 8 bytes.  This isn't needed 
      everywhere, but it is on some platforms, and it doesn't harm anything
      when it isn't needed.  */
@@ -130,7 +158,14 @@ ffi_status ffi_prep_cif_machdep(ffi_cif *cif)
       break;
 
     case FFI_TYPE_STRUCT:
-      if (cif->rtype->size <= 4)
+      if (cif->abi == FFI_VFP
+	  && (type_code = vfp_type_p (cif->rtype)) != 0)
+	{
+	  /* A Composite Type passed in VFP registers, either
+	     FFI_TYPE_STRUCT_VFP_FLOAT or FFI_TYPE_STRUCT_VFP_DOUBLE. */
+	  cif->flags = (unsigned) type_code;
+	}
+      else if (cif->rtype->size <= 4)
 	/* A Composite Type not larger than 4 bytes is returned in r0.  */
 	cif->flags = (unsigned)FFI_TYPE_INT;
       else
@@ -145,11 +180,18 @@ ffi_status ffi_prep_cif_machdep(ffi_cif *cif)
       break;
     }
 
+  /* Map out the register placements of VFP register args.
+     The VFP hard-float calling conventions are slightly more sophisticated than
+     the base calling conventions, so we do it here instead of in ffi_prep_args(). */
+  if (cif->abi == FFI_VFP)
+    layout_vfp_args (cif);
+
   return FFI_OK;
 }
 
-extern void ffi_call_SYSV(void (*)(char *, extended_cif *), extended_cif *,
-			  unsigned, unsigned, unsigned *, void (*fn)(void));
+/* Prototypes for assembly functions, in sysv.S */
+extern void ffi_call_SYSV (void (*fn)(void), extended_cif *, unsigned, unsigned, unsigned *);
+extern void ffi_call_VFP (void (*fn)(void), extended_cif *, unsigned, unsigned, unsigned *);
 
 void ffi_call(ffi_cif *cif, void (*fn)(void), void *rvalue, void **avalue)
 {
@@ -157,6 +199,8 @@ void ffi_call(ffi_cif *cif, void (*fn)(void), void *rvalue, void **avalue)
 
   int small_struct = (cif->flags == FFI_TYPE_INT 
 		      && cif->rtype->type == FFI_TYPE_STRUCT);
+  int vfp_struct = (cif->flags == FFI_TYPE_STRUCT_VFP_FLOAT
+		    || cif->flags == FFI_TYPE_STRUCT_VFP_DOUBLE);
 
   ecif.cif = cif;
   ecif.avalue = avalue;
@@ -173,38 +217,51 @@ void ffi_call(ffi_cif *cif, void (*fn)(void), void *rvalue, void **avalue)
     }
   else if (small_struct)
     ecif.rvalue = &temp;
+  else if (vfp_struct)
+    {
+      /* Largest case is double x 4. */
+      ecif.rvalue = alloca(32);
+    }
   else
     ecif.rvalue = rvalue;
 
   switch (cif->abi) 
     {
     case FFI_SYSV:
-      ffi_call_SYSV(ffi_prep_args, &ecif, cif->bytes, cif->flags, ecif.rvalue,
-		    fn);
+      ffi_call_SYSV (fn, &ecif, cif->bytes, cif->flags, ecif.rvalue);
+      break;
 
+    case FFI_VFP:
+      ffi_call_VFP (fn, &ecif, cif->bytes, cif->flags, ecif.rvalue);
       break;
+
     default:
       FFI_ASSERT(0);
       break;
     }
   if (small_struct)
     memcpy (rvalue, &temp, cif->rtype->size);
+  else if (vfp_struct)
+    memcpy (rvalue, ecif.rvalue, cif->rtype->size);
 }
 
 /** private members **/
 
 static void ffi_prep_incoming_args_SYSV (char *stack, void **ret,
-					 void** args, ffi_cif* cif);
+					 void** args, ffi_cif* cif, float *vfp_stack);
 
 void ffi_closure_SYSV (ffi_closure *);
 
+void ffi_closure_VFP (ffi_closure *);
+
 /* This function is jumped to by the trampoline */
 
 unsigned int
-ffi_closure_SYSV_inner (closure, respp, args)
+ffi_closure_SYSV_inner (closure, respp, args, vfp_args)
      ffi_closure *closure;
      void **respp;
      void *args;
+     void *vfp_args;
 {
   // our various things...
   ffi_cif       *cif;
@@ -219,7 +276,7 @@ ffi_closure_SYSV_inner (closure, respp, args)
    * a structure, it will re-set RESP to point to the
    * structure return address.  */
 
-  ffi_prep_incoming_args_SYSV(args, respp, arg_area, cif);
+  ffi_prep_incoming_args_SYSV(args, respp, arg_area, cif, vfp_args);
 
   (closure->fun) (cif, *respp, arg_area, closure->user_data);
 
@@ -229,10 +286,12 @@ ffi_closure_SYSV_inner (closure, respp, args)
 /*@-exportheader@*/
 static void 
 ffi_prep_incoming_args_SYSV(char *stack, void **rvalue,
-			    void **avalue, ffi_cif *cif)
+			    void **avalue, ffi_cif *cif,
+			    /* Used only under VFP hard-float ABI. */
+			    float *vfp_stack)
 /*@=exportheader@*/
 {
-  register unsigned int i;
+  register unsigned int i, vi = 0;
   register void **p_argv;
   register char *argp;
   register ffi_type **p_arg;
@@ -249,8 +308,16 @@ ffi_prep_incoming_args_SYSV(char *stack, void **rvalue,
   for (i = cif->nargs, p_arg = cif->arg_types; (i != 0); i--, p_arg++)
     {
       size_t z;
-
-      size_t alignment = (*p_arg)->alignment;
+      size_t alignment;
+  
+      if (cif->abi == FFI_VFP
+	  && vi < cif->vfp_nargs && vfp_type_p (*p_arg))
+	{
+	  *p_argv++ = (void*)(vfp_stack + cif->vfp_args[vi++]);
+	  continue;
+	}
+
+      alignment = (*p_arg)->alignment;
       if (alignment < 4)
 	alignment = 4;
       /* Align if necessary */
@@ -273,18 +340,237 @@ ffi_prep_incoming_args_SYSV(char *stack, void **rvalue,
 
 /* How to make a trampoline.  */
 
+#if FFI_EXEC_TRAMPOLINE_TABLE
+
+#include <mach/mach.h>
+#include <pthread.h>
+#include <stdio.h>
+#include <stdlib.h>
+
+extern void *ffi_closure_trampoline_table_page;
+
+typedef struct ffi_trampoline_table ffi_trampoline_table;
+typedef struct ffi_trampoline_table_entry ffi_trampoline_table_entry;
+
+struct ffi_trampoline_table {
+  /* contigious writable and executable pages */
+  vm_address_t config_page;
+  vm_address_t trampoline_page;
+
+  /* free list tracking */
+  uint16_t free_count;
+  ffi_trampoline_table_entry *free_list;
+  ffi_trampoline_table_entry *free_list_pool;
+
+  ffi_trampoline_table *prev;
+  ffi_trampoline_table *next;
+};
+
+struct ffi_trampoline_table_entry {
+  void *(*trampoline)();
+  ffi_trampoline_table_entry *next;
+};
+
+/* Override the standard architecture trampoline size */
+// XXX TODO - Fix
+#undef FFI_TRAMPOLINE_SIZE
+#define FFI_TRAMPOLINE_SIZE 12
+
+/* The trampoline configuration is placed at 4080 bytes prior to the trampoline's entry point */
+#define FFI_TRAMPOLINE_CODELOC_CONFIG(codeloc) ((void **) (((uint8_t *) codeloc) - 4080));
+
+/* The first 16 bytes of the config page are unused, as they are unaddressable from the trampoline page. */
+#define FFI_TRAMPOLINE_CONFIG_PAGE_OFFSET 16
+
+/* Total number of trampolines that fit in one trampoline table */
+#define FFI_TRAMPOLINE_COUNT ((PAGE_SIZE - FFI_TRAMPOLINE_CONFIG_PAGE_OFFSET) / FFI_TRAMPOLINE_SIZE)
+
+static pthread_mutex_t ffi_trampoline_lock = PTHREAD_MUTEX_INITIALIZER;
+static ffi_trampoline_table *ffi_trampoline_tables = NULL;
+
+static ffi_trampoline_table *
+ffi_trampoline_table_alloc ()
+{
+  ffi_trampoline_table *table = NULL;
+
+  /* Loop until we can allocate two contigious pages */
+  while (table == NULL) {
+    vm_address_t config_page = 0x0;
+    kern_return_t kt;
+
+    /* Try to allocate two pages */
+    kt = vm_allocate (mach_task_self (), &config_page, PAGE_SIZE*2, VM_FLAGS_ANYWHERE);
+    if (kt != KERN_SUCCESS) {
+      fprintf(stderr, "vm_allocate() failure: %d at %s:%d\n", kt, __FILE__, __LINE__);
+      break;
+    }
+
+    /* Now drop the second half of the allocation to make room for the trampoline table */
+    vm_address_t trampoline_page = config_page+PAGE_SIZE;
+    kt = vm_deallocate (mach_task_self (), trampoline_page, PAGE_SIZE);
+    if (kt != KERN_SUCCESS) {
+      fprintf(stderr, "vm_deallocate() failure: %d at %s:%d\n", kt, __FILE__, __LINE__);
+      break;
+    }
+
+    /* Remap the trampoline table to directly follow the config page */
+    vm_prot_t cur_prot;
+    vm_prot_t max_prot;
+
+    kt = vm_remap (mach_task_self (), &trampoline_page, PAGE_SIZE, 0x0, FALSE, mach_task_self (), (vm_address_t) &ffi_closure_trampoline_table_page, FALSE, &cur_prot, &max_prot, VM_INHERIT_SHARE);
+
+    /* If we lost access to the destination trampoline page, drop our config allocation mapping and retry */
+    if (kt != KERN_SUCCESS) {
+      /* Log unexpected failures */
+      if (kt != KERN_NO_SPACE) {
+        fprintf(stderr, "vm_remap() failure: %d at %s:%d\n", kt, __FILE__, __LINE__);
+      }
+
+      vm_deallocate (mach_task_self (), config_page, PAGE_SIZE);
+      continue;
+    }
+
+    /* We have valid trampoline and config pages */
+    table = calloc (1, sizeof(ffi_trampoline_table));
+    table->free_count = FFI_TRAMPOLINE_COUNT;
+    table->config_page = config_page;
+    table->trampoline_page = trampoline_page;
+
+    /* Create and initialize the free list */
+    table->free_list_pool = calloc(FFI_TRAMPOLINE_COUNT, sizeof(ffi_trampoline_table_entry));
+
+    uint16_t i;
+    for (i = 0; i < table->free_count; i++) {
+      ffi_trampoline_table_entry *entry = &table->free_list_pool[i];
+      entry->trampoline = (void *) (table->trampoline_page + (i * FFI_TRAMPOLINE_SIZE));
+
+      if (i < table->free_count - 1)
+        entry->next = &table->free_list_pool[i+1];
+    }
+
+    table->free_list = table->free_list_pool;
+  }
+
+  return table;
+}
+
+void *
+ffi_closure_alloc (size_t size, void **code)
+{
+  /* Create the closure */
+  ffi_closure *closure = malloc(size);
+  if (closure == NULL)
+    return NULL;
+
+  pthread_mutex_lock(&ffi_trampoline_lock);
+
+  /* Check for an active trampoline table with available entries. */
+  ffi_trampoline_table *table = ffi_trampoline_tables;
+  if (table == NULL || table->free_list == NULL) {
+    table = ffi_trampoline_table_alloc ();
+    if (table == NULL) {
+      free(closure);
+      return NULL;
+    }
+
+    /* Insert the new table at the top of the list */
+    table->next = ffi_trampoline_tables;
+    if (table->next != NULL)
+        table->next->prev = table;
+
+    ffi_trampoline_tables = table;
+  }
+
+  /* Claim the free entry */
+  ffi_trampoline_table_entry *entry = ffi_trampoline_tables->free_list;
+  ffi_trampoline_tables->free_list = entry->next;
+  ffi_trampoline_tables->free_count--;
+  entry->next = NULL;
+
+  pthread_mutex_unlock(&ffi_trampoline_lock);
+
+  /* Initialize the return values */
+  *code = entry->trampoline;
+  closure->trampoline_table = table;
+  closure->trampoline_table_entry = entry;
+
+  return closure;
+}
+
+void
+ffi_closure_free (void *ptr)
+{
+  ffi_closure *closure = ptr;
+
+  pthread_mutex_lock(&ffi_trampoline_lock);
+
+  /* Fetch the table and entry references */
+  ffi_trampoline_table *table = closure->trampoline_table;
+  ffi_trampoline_table_entry *entry = closure->trampoline_table_entry;
+
+  /* Return the entry to the free list */
+  entry->next = table->free_list;
+  table->free_list = entry;
+  table->free_count++;
+
+  /* If all trampolines within this table are free, and at least one other table exists, deallocate
+   * the table */
+  if (table->free_count == FFI_TRAMPOLINE_COUNT && ffi_trampoline_tables != table) {
+    /* Remove from the list */
+    if (table->prev != NULL)
+      table->prev->next = table->next;
+
+    if (table->next != NULL)
+      table->next->prev = table->prev;
+
+    /* Deallocate pages */
+    kern_return_t kt;
+    kt = vm_deallocate (mach_task_self (), table->config_page, PAGE_SIZE);
+    if (kt != KERN_SUCCESS)
+      fprintf(stderr, "vm_deallocate() failure: %d at %s:%d\n", kt, __FILE__, __LINE__);
+
+    kt = vm_deallocate (mach_task_self (), table->trampoline_page, PAGE_SIZE);
+    if (kt != KERN_SUCCESS)
+      fprintf(stderr, "vm_deallocate() failure: %d at %s:%d\n", kt, __FILE__, __LINE__);
+
+    /* Deallocate free list */
+    free (table->free_list_pool);
+    free (table);
+  } else if (ffi_trampoline_tables != table) {
+    /* Otherwise, bump this table to the top of the list */
+    table->prev = NULL;
+    table->next = ffi_trampoline_tables;
+    if (ffi_trampoline_tables != NULL)
+      ffi_trampoline_tables->prev = table;
+
+    ffi_trampoline_tables = table;
+  }
+
+  pthread_mutex_unlock (&ffi_trampoline_lock);
+
+  /* Free the closure */
+  free (closure);
+}
+
+#else
+
 #define FFI_INIT_TRAMPOLINE(TRAMP,FUN,CTX)				\
 ({ unsigned char *__tramp = (unsigned char*)(TRAMP);			\
    unsigned int  __fun = (unsigned int)(FUN);				\
    unsigned int  __ctx = (unsigned int)(CTX);				\
+   unsigned char *insns = (unsigned char *)(CTX);                       \
    *(unsigned int*) &__tramp[0] = 0xe92d000f; /* stmfd sp!, {r0-r3} */	\
    *(unsigned int*) &__tramp[4] = 0xe59f0000; /* ldr r0, [pc] */	\
    *(unsigned int*) &__tramp[8] = 0xe59ff000; /* ldr pc, [pc] */	\
    *(unsigned int*) &__tramp[12] = __ctx;				\
    *(unsigned int*) &__tramp[16] = __fun;				\
-   __clear_cache((&__tramp[0]), (&__tramp[19]));			\
+   __clear_cache((&__tramp[0]), (&__tramp[19])); /* Clear data mapping.  */ \
+   __clear_cache(insns, insns + 3 * sizeof (unsigned int));             \
+                                                 /* Clear instruction   \
+                                                    mapping.  */        \
  })
 
+#endif
 
 /* the cif must already be prep'ed */
 
@@ -295,15 +581,148 @@ ffi_prep_closure_loc (ffi_closure* closure,
 		      void *user_data,
 		      void *codeloc)
 {
-  FFI_ASSERT (cif->abi == FFI_SYSV);
+  void (*closure_func)(ffi_closure*) = NULL;
 
+  if (cif->abi == FFI_SYSV)
+    closure_func = &ffi_closure_SYSV;
+  else if (cif->abi == FFI_VFP)
+    closure_func = &ffi_closure_VFP;
+  else
+    return FFI_BAD_ABI;
+    
+#if FFI_EXEC_TRAMPOLINE_TABLE
+  void **config = FFI_TRAMPOLINE_CODELOC_CONFIG(codeloc);
+  config[0] = closure;
+  config[1] = closure_func;
+#else
   FFI_INIT_TRAMPOLINE (&closure->tramp[0], \
-		       &ffi_closure_SYSV,  \
+		       closure_func,  \
 		       codeloc);
-    
+#endif
+
   closure->cif  = cif;
   closure->user_data = user_data;
   closure->fun  = fun;
 
   return FFI_OK;
 }
+
+/* Below are routines for VFP hard-float support. */
+
+static int rec_vfp_type_p (ffi_type *t, int *elt, int *elnum)
+{
+  switch (t->type)
+    {
+    case FFI_TYPE_FLOAT:
+    case FFI_TYPE_DOUBLE:
+      *elt = (int) t->type;
+      *elnum = 1;
+      return 1;
+
+    case FFI_TYPE_STRUCT_VFP_FLOAT:
+      *elt = FFI_TYPE_FLOAT;
+      *elnum = t->size / sizeof (float);
+      return 1;
+
+    case FFI_TYPE_STRUCT_VFP_DOUBLE:
+      *elt = FFI_TYPE_DOUBLE;
+      *elnum = t->size / sizeof (double);
+      return 1;
+
+    case FFI_TYPE_STRUCT:;
+      {
+	int base_elt = 0, total_elnum = 0;
+	ffi_type **el = t->elements;
+	while (*el)
+	  {
+	    int el_elt = 0, el_elnum = 0;
+	    if (! rec_vfp_type_p (*el, &el_elt, &el_elnum)
+		|| (base_elt && base_elt != el_elt)
+		|| total_elnum + el_elnum > 4)
+	      return 0;
+	    base_elt = el_elt;
+	    total_elnum += el_elnum;
+	    el++;
+	  }
+	*elnum = total_elnum;
+	*elt = base_elt;
+	return 1;
+      }
+    default: ;
+    }
+  return 0;
+}
+
+static int vfp_type_p (ffi_type *t)
+{
+  int elt, elnum;
+  if (rec_vfp_type_p (t, &elt, &elnum))
+    {
+      if (t->type == FFI_TYPE_STRUCT)
+	{
+	  if (elnum == 1)
+	    t->type = elt;
+	  else
+	    t->type = (elt == FFI_TYPE_FLOAT
+		       ? FFI_TYPE_STRUCT_VFP_FLOAT
+		       : FFI_TYPE_STRUCT_VFP_DOUBLE);
+	}
+      return (int) t->type;
+    }
+  return 0;
+}
+
+static void place_vfp_arg (ffi_cif *cif, ffi_type *t)
+{
+  int reg = cif->vfp_reg_free;
+  int nregs = t->size / sizeof (float);
+  int align = ((t->type == FFI_TYPE_STRUCT_VFP_FLOAT
+		|| t->type == FFI_TYPE_FLOAT) ? 1 : 2);
+  /* Align register number. */
+  if ((reg & 1) && align == 2)
+    reg++;
+  while (reg + nregs <= 16)
+    {
+      int s, new_used = 0;
+      for (s = reg; s < reg + nregs; s++)
+	{
+	  new_used |= (1 << s);
+	  if (cif->vfp_used & (1 << s))
+	    {
+	      reg += align;
+	      goto next_reg;
+	    }
+	}
+      /* Found regs to allocate. */
+      cif->vfp_used |= new_used;
+      cif->vfp_args[cif->vfp_nargs++] = reg;
+
+      /* Update vfp_reg_free. */
+      if (cif->vfp_used & (1 << cif->vfp_reg_free))
+	{
+	  reg += nregs;
+	  while (cif->vfp_used & (1 << reg))
+	    reg += 1;
+	  cif->vfp_reg_free = reg;
+	}
+      return;
+    next_reg: ;
+    }
+}
+
+static void layout_vfp_args (ffi_cif *cif)
+{
+  int i;
+  /* Init VFP fields */
+  cif->vfp_used = 0;
+  cif->vfp_nargs = 0;
+  cif->vfp_reg_free = 0;
+  memset (cif->vfp_args, -1, 16); /* Init to -1. */
+
+  for (i = 0; i < cif->nargs; i++)
+    {
+      ffi_type *t = cif->arg_types[i];
+      if (vfp_type_p (t))
+	place_vfp_arg (cif, t);
+    }
+}