ffi 1.11.2

data/ext/ffi_c/libffi/src/mips/o32.S

@@ -0,0 +1,502 @@
+/* -----------------------------------------------------------------------
+   o32.S - Copyright (c) 1996, 1998, 2005  Red Hat, Inc.
+   
+   MIPS Foreign Function Interface 
+
+   Permission is hereby granted, free of charge, to any person obtaining
+   a copy of this software and associated documentation files (the
+   ``Software''), to deal in the Software without restriction, including
+   without limitation the rights to use, copy, modify, merge, publish,
+   distribute, sublicense, and/or sell copies of the Software, and to
+   permit persons to whom the Software is furnished to do so, subject to
+   the following conditions:
+
+   The above copyright notice and this permission notice shall be included
+   in all copies or substantial portions of the Software.
+
+   THE SOFTWARE IS PROVIDED ``AS IS'', WITHOUT WARRANTY OF ANY KIND,
+   EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+   MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+   NONINFRINGEMENT.  IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
+   HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+   WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+   OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+   DEALINGS IN THE SOFTWARE.
+   ----------------------------------------------------------------------- */
+
+#define LIBFFI_ASM	
+#include <fficonfig.h>
+#include <ffi.h>
+
+/* Only build this code if we are compiling for o32 */	
+
+#if defined(FFI_MIPS_O32)
+	
+#define callback a0
+#define bytes	 a2
+#define flags	 a3
+		
+#define SIZEOF_FRAME	(4 * FFI_SIZEOF_ARG + 2 * FFI_SIZEOF_ARG)
+#define A3_OFF		(SIZEOF_FRAME + 3 * FFI_SIZEOF_ARG)
+#define FP_OFF		(SIZEOF_FRAME - 2 * FFI_SIZEOF_ARG)
+#define RA_OFF		(SIZEOF_FRAME - 1 * FFI_SIZEOF_ARG)
+
+	.abicalls
+	.text
+	.align	2
+	.globl	ffi_call_O32
+	.ent	ffi_call_O32
+ffi_call_O32:	
+$LFB0:
+	# Prologue
+	SUBU	$sp, SIZEOF_FRAME	# Frame size
+$LCFI00:
+	REG_S	$fp, FP_OFF($sp)	# Save frame pointer
+$LCFI01:
+	REG_S	ra, RA_OFF($sp)		# Save return address
+$LCFI02:
+	move	$fp, $sp
+
+$LCFI03:
+	move	t9, callback		# callback function pointer
+	REG_S	flags, A3_OFF($fp)	# flags
+
+	# Allocate at least 4 words in the argstack
+	LI	v0, 4 * FFI_SIZEOF_ARG
+	blt	bytes, v0, sixteen
+
+	ADDU	v0, bytes, 7	# make sure it is aligned 
+	and	v0, -8		# to an 8 byte boundry
+
+sixteen:
+	SUBU	$sp, v0		# move the stack pointer to reflect the
+				# arg space
+
+	ADDU	a0, $sp, 4 * FFI_SIZEOF_ARG
+
+	jalr	t9
+	
+	REG_L	t0, A3_OFF($fp)		# load the flags word
+	SRL	t2, t0, 4		# shift our arg info
+	and     t0, ((1<<4)-1)          # mask out the return type
+		
+	ADDU	$sp, 4 * FFI_SIZEOF_ARG		# adjust $sp to new args
+
+#ifndef __mips_soft_float
+	bnez	t0, pass_d			# make it quick for int
+#endif
+	REG_L	a0, 0*FFI_SIZEOF_ARG($sp)	# just go ahead and load the
+	REG_L	a1, 1*FFI_SIZEOF_ARG($sp)	# four regs.
+	REG_L	a2, 2*FFI_SIZEOF_ARG($sp)
+	REG_L	a3, 3*FFI_SIZEOF_ARG($sp)
+	b	call_it
+
+#ifndef __mips_soft_float
+pass_d:
+	bne	t0, FFI_ARGS_D, pass_f
+	l.d	$f12, 0*FFI_SIZEOF_ARG($sp)	# load $fp regs from args
+	REG_L	a2,   2*FFI_SIZEOF_ARG($sp)	# passing a double
+	REG_L	a3,   3*FFI_SIZEOF_ARG($sp)
+	b	call_it
+
+pass_f:	
+	bne	t0, FFI_ARGS_F, pass_d_d
+	l.s	$f12, 0*FFI_SIZEOF_ARG($sp)	# load $fp regs from args
+	REG_L	a1,   1*FFI_SIZEOF_ARG($sp)	# passing a float
+	REG_L	a2,   2*FFI_SIZEOF_ARG($sp)
+	REG_L	a3,   3*FFI_SIZEOF_ARG($sp)
+	b	call_it		
+
+pass_d_d:		
+	bne	t0, FFI_ARGS_DD, pass_f_f
+	l.d	$f12, 0*FFI_SIZEOF_ARG($sp)	# load $fp regs from args
+	l.d	$f14, 2*FFI_SIZEOF_ARG($sp)	# passing two doubles
+	b	call_it
+
+pass_f_f:	
+	bne	t0, FFI_ARGS_FF, pass_d_f
+	l.s	$f12, 0*FFI_SIZEOF_ARG($sp)	# load $fp regs from args
+	l.s	$f14, 1*FFI_SIZEOF_ARG($sp)	# passing two floats
+	REG_L	a2,   2*FFI_SIZEOF_ARG($sp)
+	REG_L	a3,   3*FFI_SIZEOF_ARG($sp)
+	b	call_it
+
+pass_d_f:		
+	bne	t0, FFI_ARGS_DF, pass_f_d
+	l.d	$f12, 0*FFI_SIZEOF_ARG($sp)	# load $fp regs from args
+	l.s	$f14, 2*FFI_SIZEOF_ARG($sp)	# passing double and float
+	REG_L	a3,   3*FFI_SIZEOF_ARG($sp)
+	b	call_it
+
+pass_f_d:		
+ # assume that the only other combination must be float then double
+ #	bne	t0, FFI_ARGS_F_D, call_it
+	l.s	$f12, 0*FFI_SIZEOF_ARG($sp)	# load $fp regs from args
+	l.d	$f14, 2*FFI_SIZEOF_ARG($sp)	# passing double and float
+#endif
+
+call_it:	
+	# Load the static chain pointer
+	REG_L	t7, SIZEOF_FRAME + 6*FFI_SIZEOF_ARG($fp)
+
+	# Load the function pointer
+	REG_L	t9, SIZEOF_FRAME + 5*FFI_SIZEOF_ARG($fp)
+
+	# If the return value pointer is NULL, assume no return value.
+	REG_L	t1, SIZEOF_FRAME + 4*FFI_SIZEOF_ARG($fp)
+	beqz	t1, noretval
+
+	bne     t2, FFI_TYPE_INT, retlonglong
+	jalr	t9
+	REG_L	t0, SIZEOF_FRAME + 4*FFI_SIZEOF_ARG($fp)
+	REG_S	v0, 0(t0)
+	b	epilogue
+
+retlonglong:
+	# Really any 64-bit int, signed or not.
+	bne	t2, FFI_TYPE_UINT64, retfloat
+	jalr	t9
+	REG_L	t0, SIZEOF_FRAME + 4*FFI_SIZEOF_ARG($fp)
+	REG_S	v1, 4(t0)
+	REG_S	v0, 0(t0)
+	b	epilogue
+
+retfloat:
+	bne     t2, FFI_TYPE_FLOAT, retdouble
+	jalr	t9
+	REG_L	t0, SIZEOF_FRAME + 4*FFI_SIZEOF_ARG($fp)
+#ifndef __mips_soft_float
+	s.s	$f0, 0(t0)
+#else
+	REG_S	v0, 0(t0)
+#endif
+	b	epilogue
+
+retdouble:	
+	bne	t2, FFI_TYPE_DOUBLE, noretval
+	jalr	t9
+	REG_L	t0, SIZEOF_FRAME + 4*FFI_SIZEOF_ARG($fp)
+#ifndef __mips_soft_float
+	s.d	$f0, 0(t0)
+#else
+	REG_S	v1, 4(t0)
+	REG_S	v0, 0(t0)
+#endif
+	b	epilogue
+	
+noretval:	
+	jalr	t9
+	
+	# Epilogue
+epilogue:	
+	move	$sp, $fp	
+	REG_L	$fp, FP_OFF($sp)	# Restore frame pointer
+	REG_L	ra, RA_OFF($sp)		# Restore return address
+	ADDU	$sp, SIZEOF_FRAME	# Fix stack pointer
+	j	ra
+
+$LFE0:
+	.end	ffi_call_O32
+
+
+/* ffi_closure_O32. Expects address of the passed-in ffi_closure
+	in t4 ($12). Stores any arguments passed in registers onto the
+	stack, then calls ffi_closure_mips_inner_O32, which
+	then decodes them.
+	
+	Stack layout:
+
+	 3 - a3 save
+	 2 - a2 save
+	 1 - a1 save
+	 0 - a0 save, original sp
+	-1 - ra save
+	-2 - fp save
+	-3 - $16 (s0) save
+	-4 - cprestore
+	-5 - return value high (v1)
+	-6 - return value low (v0)
+	-7 - f14 (le high, be low)
+	-8 - f14 (le low, be high)
+	-9 - f12 (le high, be low)
+       -10 - f12 (le low, be high)
+       -11 - Called function a5 save
+       -12 - Called function a4 save
+       -13 - Called function a3 save
+       -14 - Called function a2 save
+       -15 - Called function a1 save
+       -16 - Called function a0 save, our sp and fp point here
+	 */
+	
+#define SIZEOF_FRAME2	(16 * FFI_SIZEOF_ARG)
+#define A3_OFF2		(SIZEOF_FRAME2 + 3 * FFI_SIZEOF_ARG)
+#define A2_OFF2		(SIZEOF_FRAME2 + 2 * FFI_SIZEOF_ARG)
+#define A1_OFF2		(SIZEOF_FRAME2 + 1 * FFI_SIZEOF_ARG)
+#define A0_OFF2		(SIZEOF_FRAME2 + 0 * FFI_SIZEOF_ARG)
+#define RA_OFF2		(SIZEOF_FRAME2 - 1 * FFI_SIZEOF_ARG)
+#define FP_OFF2		(SIZEOF_FRAME2 - 2 * FFI_SIZEOF_ARG)
+#define S0_OFF2		(SIZEOF_FRAME2 - 3 * FFI_SIZEOF_ARG)
+#define GP_OFF2		(SIZEOF_FRAME2 - 4 * FFI_SIZEOF_ARG)
+#define V1_OFF2		(SIZEOF_FRAME2 - 5 * FFI_SIZEOF_ARG)
+#define V0_OFF2		(SIZEOF_FRAME2 - 6 * FFI_SIZEOF_ARG)
+#define FA_1_1_OFF2	(SIZEOF_FRAME2 - 7 * FFI_SIZEOF_ARG)
+#define FA_1_0_OFF2	(SIZEOF_FRAME2 - 8 * FFI_SIZEOF_ARG)
+#define FA_0_1_OFF2	(SIZEOF_FRAME2 - 9 * FFI_SIZEOF_ARG)
+#define FA_0_0_OFF2	(SIZEOF_FRAME2 - 10 * FFI_SIZEOF_ARG)
+#define CALLED_A5_OFF2  (SIZEOF_FRAME2 - 11 * FFI_SIZEOF_ARG)
+#define CALLED_A4_OFF2  (SIZEOF_FRAME2 - 12 * FFI_SIZEOF_ARG)
+
+	.text
+
+	.align	2
+	.globl	ffi_go_closure_O32
+	.ent	ffi_go_closure_O32
+ffi_go_closure_O32:
+$LFB1:
+	# Prologue
+	.frame	$fp, SIZEOF_FRAME2, ra
+	.set	noreorder
+	.cpload	t9
+	.set	reorder
+	SUBU	$sp, SIZEOF_FRAME2
+	.cprestore GP_OFF2
+$LCFI10:
+
+	REG_S	$16, S0_OFF2($sp)	 # Save s0
+	REG_S	$fp, FP_OFF2($sp)	 # Save frame pointer
+	REG_S	ra, RA_OFF2($sp)	 # Save return address
+$LCFI11:
+
+	move	$fp, $sp
+$LCFI12:
+
+	REG_S	a0, A0_OFF2($fp)
+	REG_S	a1, A1_OFF2($fp)
+	REG_S	a2, A2_OFF2($fp)
+	REG_S	a3, A3_OFF2($fp)
+
+	# Load ABI enum to s0
+	REG_L	$16, 4($15)	# cif 
+	REG_L	$16, 0($16)	# abi is first member.
+
+	li	$13, 1		# FFI_O32
+	bne	$16, $13, 1f	# Skip fp save if FFI_O32_SOFT_FLOAT
+	
+	# Store all possible float/double registers.
+	s.d	$f12, FA_0_0_OFF2($fp)
+	s.d	$f14, FA_1_0_OFF2($fp)
+1:
+	# prepare arguments for ffi_closure_mips_inner_O32
+	REG_L	a0, 4($15)	 # cif 
+	REG_L	a1, 8($15)	 # fun
+	move	a2, $15		 # user_data = go closure
+	addu	a3, $fp, V0_OFF2 # rvalue
+
+	addu	t9, $fp, A0_OFF2 # ar
+	REG_S   t9, CALLED_A4_OFF2($fp)
+
+	addu	t9, $fp, FA_0_0_OFF2 #fpr
+	REG_S   t9, CALLED_A5_OFF2($fp)
+
+	b $do_closure
+
+$LFE1:
+	.end ffi_go_closure_O32
+
+	.align	2
+	.globl	ffi_closure_O32
+	.ent	ffi_closure_O32
+ffi_closure_O32:
+$LFB2:
+	# Prologue
+	.frame	$fp, SIZEOF_FRAME2, ra
+	.set	noreorder
+	.cpload	t9
+	.set	reorder
+	SUBU	$sp, SIZEOF_FRAME2
+	.cprestore GP_OFF2
+$LCFI20:
+	REG_S	$16, S0_OFF2($sp)	 # Save s0
+	REG_S	$fp, FP_OFF2($sp)	 # Save frame pointer
+	REG_S	ra, RA_OFF2($sp)	 # Save return address
+$LCFI21:
+	move	$fp, $sp
+
+$LCFI22:
+	# Store all possible argument registers. If there are more than
+	# four arguments, then they are stored above where we put a3.
+	REG_S	a0, A0_OFF2($fp)
+	REG_S	a1, A1_OFF2($fp)
+	REG_S	a2, A2_OFF2($fp)
+	REG_S	a3, A3_OFF2($fp)
+
+	# Load ABI enum to s0
+	REG_L	$16, 20($12)	# cif pointer follows tramp.
+	REG_L	$16, 0($16)	# abi is first member.
+
+	li	$13, 1		# FFI_O32
+	bne	$16, $13, 1f	# Skip fp save if FFI_O32_SOFT_FLOAT
+	
+#ifndef __mips_soft_float
+	# Store all possible float/double registers.
+	s.d	$f12, FA_0_0_OFF2($fp)
+	s.d	$f14, FA_1_0_OFF2($fp)
+#endif
+1:	
+	# prepare arguments for ffi_closure_mips_inner_O32
+	REG_L	a0, 20($12)	 # cif pointer follows tramp.
+	REG_L	a1, 24($12)	 # fun
+	REG_L	a2, 28($12)	 # user_data
+	addu	a3, $fp, V0_OFF2 # rvalue
+
+	addu	t9, $fp, A0_OFF2 # ar
+	REG_S   t9, CALLED_A4_OFF2($fp)
+
+	addu	t9, $fp, FA_0_0_OFF2 #fpr
+	REG_S   t9, CALLED_A5_OFF2($fp)
+
+$do_closure:
+	la	t9, ffi_closure_mips_inner_O32
+	# Call ffi_closure_mips_inner_O32 to do the work.
+	jalr	t9
+
+	# Load the return value into the appropriate register.
+	move	$8, $2
+	li	$9, FFI_TYPE_VOID
+	beq	$8, $9, closure_done
+
+	li	$13, 1		# FFI_O32
+	bne	$16, $13, 1f	# Skip fp restore if FFI_O32_SOFT_FLOAT
+
+#ifndef __mips_soft_float
+	li	$9, FFI_TYPE_FLOAT
+	l.s	$f0, V0_OFF2($fp)
+	beq	$8, $9, closure_done
+
+	li	$9, FFI_TYPE_DOUBLE
+	l.d	$f0, V0_OFF2($fp)
+	beq	$8, $9, closure_done
+#endif
+1:	
+	REG_L	$3, V1_OFF2($fp)
+	REG_L	$2, V0_OFF2($fp)
+
+closure_done:
+	# Epilogue
+	move	$sp, $fp
+	REG_L	$16, S0_OFF2($sp)	 # Restore s0
+	REG_L	$fp, FP_OFF2($sp)	 # Restore frame pointer
+	REG_L	ra,  RA_OFF2($sp)	 # Restore return address
+	ADDU	$sp, SIZEOF_FRAME2
+	j	ra
+$LFE2:
+	.end	ffi_closure_O32
+
+/* DWARF-2 unwind info. */
+
+	.section	.eh_frame,"a",@progbits
+$Lframe0:
+	.4byte	$LECIE0-$LSCIE0	 # Length of Common Information Entry
+$LSCIE0:
+	.4byte	0x0	 # CIE Identifier Tag
+	.byte	0x1	 # CIE Version
+	.ascii "zR\0"	 # CIE Augmentation
+	.uleb128 0x1	 # CIE Code Alignment Factor
+	.sleb128 4	 # CIE Data Alignment Factor
+	.byte	0x1f	 # CIE RA Column
+	.uleb128 0x1	 # Augmentation size
+	.byte	0x00	 # FDE Encoding (absptr)
+	.byte	0xc	 # DW_CFA_def_cfa
+	.uleb128 0x1d
+	.uleb128 0x0
+	.align	2
+$LECIE0:
+
+$LSFDE0:
+	.4byte	$LEFDE0-$LASFDE0	 # FDE Length
+$LASFDE0:
+	.4byte	$LASFDE0-$Lframe0	 # FDE CIE offset
+	.4byte	$LFB0	 # FDE initial location
+	.4byte	$LFE0-$LFB0	 # FDE address range
+	.uleb128 0x0	 # Augmentation size
+	.byte	0x4	 # DW_CFA_advance_loc4
+	.4byte	$LCFI00-$LFB0
+	.byte	0xe	 # DW_CFA_def_cfa_offset
+	.uleb128 0x18
+	.byte	0x4	 # DW_CFA_advance_loc4
+	.4byte	$LCFI01-$LCFI00
+	.byte	0x11	 # DW_CFA_offset_extended_sf
+	.uleb128 0x1e	 # $fp
+	.sleb128 -2	 # SIZEOF_FRAME2 - 2*FFI_SIZEOF_ARG($sp)
+	.byte	0x11	 # DW_CFA_offset_extended_sf
+	.uleb128 0x1f	 # $ra
+	.sleb128 -1	 # SIZEOF_FRAME2 - 1*FFI_SIZEOF_ARG($sp)
+	.byte	0x4	 # DW_CFA_advance_loc4
+	.4byte	$LCFI02-$LCFI01
+	.byte	0xc	 # DW_CFA_def_cfa
+	.uleb128 0x1e
+	.uleb128 0x18
+	.align	2
+$LEFDE0:
+
+$LSFDE1:
+	.4byte	$LEFDE1-$LASFDE1	 # FDE Length
+$LASFDE1:
+	.4byte	$LASFDE1-$Lframe0	 # FDE CIE offset
+	.4byte	$LFB1	 # FDE initial location
+	.4byte	$LFE1-$LFB1	 # FDE address range
+	.uleb128 0x0	 # Augmentation size
+	.byte	0x4	 # DW_CFA_advance_loc4
+	.4byte	$LCFI10-$LFB1
+	.byte	0xe	 # DW_CFA_def_cfa_offset
+	.uleb128 SIZEOF_FRAME2
+	.byte	0x4	 # DW_CFA_advance_loc4
+	.4byte	$LCFI11-$LCFI10
+	.byte	0x11	 # DW_CFA_offset_extended_sf
+	.uleb128 0x10	 # $16
+	.sleb128 -3	 # SIZEOF_FRAME2 - 3*FFI_SIZEOF_ARG($sp)
+	.byte	0x11	 # DW_CFA_offset_extended_sf
+	.uleb128 0x1e	 # $fp
+	.sleb128 -2	 # SIZEOF_FRAME2 - 2*FFI_SIZEOF_ARG($sp)
+	.byte	0x11	 # DW_CFA_offset_extended_sf
+	.uleb128 0x1f	 # $ra
+	.sleb128 -1	 # SIZEOF_FRAME2 - 1*FFI_SIZEOF_ARG($sp)
+	.byte	0x4	 # DW_CFA_advance_loc4
+	.4byte	$LCFI12-$LCFI11
+	.byte	0xc	 # DW_CFA_def_cfa
+	.uleb128 0x1e
+	.uleb128 SIZEOF_FRAME2
+	.align	2
+$LEFDE1:
+
+$LSFDE2:
+	.4byte	$LEFDE2-$LASFDE2	 # FDE Length
+$LASFDE2:
+	.4byte	$LASFDE2-$Lframe0	 # FDE CIE offset
+	.4byte	$LFB2	 # FDE initial location
+	.4byte	$LFE2-$LFB2	 # FDE address range
+	.uleb128 0x0	 # Augmentation size
+	.byte	0x4	 # DW_CFA_advance_loc4
+	.4byte	$LCFI20-$LFB2
+	.byte	0xe	 # DW_CFA_def_cfa_offset
+	.uleb128 SIZEOF_FRAME2
+	.byte	0x4	 # DW_CFA_advance_loc4
+	.4byte	$LCFI21-$LCFI20
+	.byte	0x11	 # DW_CFA_offset_extended_sf
+	.uleb128 0x10	 # $16
+	.sleb128 -3	 # SIZEOF_FRAME2 - 3*FFI_SIZEOF_ARG($sp)
+	.byte	0x11	 # DW_CFA_offset_extended_sf
+	.uleb128 0x1e	 # $fp
+	.sleb128 -2	 # SIZEOF_FRAME2 - 2*FFI_SIZEOF_ARG($sp)
+	.byte	0x11	 # DW_CFA_offset_extended_sf
+	.uleb128 0x1f	 # $ra
+	.sleb128 -1	 # SIZEOF_FRAME2 - 1*FFI_SIZEOF_ARG($sp)
+	.byte	0x4	 # DW_CFA_advance_loc4
+	.4byte	$LCFI22-$LCFI21
+	.byte	0xc	 # DW_CFA_def_cfa
+	.uleb128 0x1e
+	.uleb128 SIZEOF_FRAME2
+	.align	2
+$LEFDE2:
+
+#endif