Initial commit.

11 files changed, 1270 insertions, 0 deletions

diff --git a/gmp-6.3.0/mpn/m68k/README b/gmp-6.3.0/mpn/m68k/README
new file mode 100644
index 0000000..5261564
--- /dev/null
+++ b/gmp-6.3.0/mpn/m68k/README
@@ -0,0 +1,138 @@
+Copyright 2001, 2003, 2004 Free Software Foundation, Inc.
+
+This file is part of the GNU MP Library.
+
+The GNU MP Library is free software; you can redistribute it and/or modify
+it under the terms of either:
+
+  * the GNU Lesser General Public License as published by the Free
+    Software Foundation; either version 3 of the License, or (at your
+    option) any later version.
+
+or
+
+  * the GNU General Public License as published by the Free Software
+    Foundation; either version 2 of the License, or (at your option) any
+    later version.
+
+or both in parallel, as here.
+
+The GNU MP 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 General Public License
+for more details.
+
+You should have received copies of the GNU General Public License and the
+GNU Lesser General Public License along with the GNU MP Library.  If not,
+see https://www.gnu.org/licenses/.
+
+
+
+
+
+                      M68K MPN SUBROUTINES
+
+
+This directory contains mpn functions for various m68k family chips.
+
+
+CODE ORGANIZATION
+
+	m68k             m68000, m68010, m68060
+	m68k/mc68020     m68020, m68030, m68040, and CPU32
+
+
+The m5200 "coldfire", which is m68000 less a few instructions, currently has
+no assembler code support.
+
+
+STATUS
+
+The code herein is old and poorly maintained.  If somebody really cared, it
+could be optimized substantially.  For example,
+
+* mpn_add_n and mpn_sub_n could, with more unrolling be improved from 6 to
+  close to 4 c/l (on m68040).
+
+* The multiplication loops could be sped up by using the FPU.
+
+* mpn_lshift by 31 should use the special-case mpn_rshift by 1 code, and
+  vice versa mpn_rshift by 31 use the special lshift by 1, when operand
+  overlap permits.
+
+* On 68000, mpn_mul_1, mpn_addmul_1 and mpn_submul_1 could check for a
+  16-bit multiplier and use two multiplies per limb, not four.
+
+  Similarly various other _1 operations like mpn_mod_1, mpn_divrem_1,
+  mpn_divexact_1, mpn_modexact_1c_odd.
+
+* On 68000, mpn_lshift and mpn_rshift could use a roll and mask instead of
+  lsrl and lsll.  This promises to be a speedup, effectively trading a 6+2*n
+  shift for one or two 4 cycle masks.  Suggested by Jean-Charles Meyrignac.
+
+* config.guess detects 68000, 68010, CPU32 and 68020 by running some code,
+  but relies on system information for 030, 040 and 060.  Can they be
+  identified by running some code?  Currently this only makes a difference
+  to the compiler options selected, since we have no specific asm code for
+  those chips.
+
+One novel idea for 68000 would be to use a 16-bit limb instead of 32-bits.
+This would suit the native 16x16 multiply, but might make it difficult to
+get full value from the native 32x32 add/sub/etc.  This would be an ABI
+option, and would select "__GMP_SHORT_LIMB" in gmp.h.
+
+Naturally an entirely new set of asm subroutines would be needed for a
+16-bit limb.  Also there's various places in the C code assuming limb>=long,
+which would need to be updated, eg. mpz_set_ui.  Some of the nails changes
+may have helped cover some of this.
+
+
+ASM FILES
+
+The .asm files are put through m4 for macro processing, and with the help of
+configure give either MIT or Motorola syntax.  The generic mpn/asm-defs.m4
+is used, together with mpn/m68k/m68k-defs.m4.  See comments in those files.
+
+Not all possible syntax variations are covered.  GCC config/m68k for
+instance has things like $ for immediates on CRDS or reversed cmp order for
+AT&T SGS.  These could probably be handled if anyone really needs it.
+
+
+CALLING CONVENTIONS
+
+The SVR4 standard has an int of 32 bits, and all parameters 32-bit aligned
+on the stack.
+
+PalmOS and perhaps various embedded systems intended for 68000 however use
+an int of 16 bits and parameters only 16-bit aligned on the stack.  This is
+generated by "gcc -mshort" (and is the default for the PalmOS gcc port, we
+believe).
+
+The asm files adapt to these two ABIs by checking sizeof(unsigned), coming
+through config.m4 as SIZEOF_UNSIGNED.  Only mpn_lshift and mpn_rshift are
+affected, all other routines take longs and pointers, which are 32-bits in
+both cases.
+
+Strictly speaking the size of an int doesn't determine the stack padding
+convention.  But if int is 16 bits then we can definitely say the host
+system is not SVR4, and therefore may as well assume we're in 16-bit stack
+alignment.
+
+
+REFERENCES
+
+"Motorola M68000 Family Programmer's Reference Manual", available online,
+
+	http://e-www.motorola.com/brdata/PDFDB/docs/M68000PM.pdf
+
+"System V Application Binary Interface: Motorola 68000 Processor Family
+Supplement", AT&T, 1990, ISBN 0-13-877553-6.  Has details of calling
+conventions and ELF style PIC coding.
+
+
+
+----------------
+Local variables:
+mode: text
+fill-column: 76
+End:
diff --git a/gmp-6.3.0/mpn/m68k/aors_n.asm b/gmp-6.3.0/mpn/m68k/aors_n.asm
new file mode 100644
index 0000000..f7d379e
--- /dev/null
+++ b/gmp-6.3.0/mpn/m68k/aors_n.asm
@@ -0,0 +1,99 @@
+dnl  mc68020 mpn_add_n, mpn_sub_n -- add or subtract limb vectors
+
+dnl  Copyright 1992, 1994, 1996, 1999-2003, 2005 Free Software Foundation, Inc.
+
+dnl  This file is part of the GNU MP Library.
+dnl
+dnl  The GNU MP Library is free software; you can redistribute it and/or modify
+dnl  it under the terms of either:
+dnl
+dnl    * the GNU Lesser General Public License as published by the Free
+dnl      Software Foundation; either version 3 of the License, or (at your
+dnl      option) any later version.
+dnl
+dnl  or
+dnl
+dnl    * the GNU General Public License as published by the Free Software
+dnl      Foundation; either version 2 of the License, or (at your option) any
+dnl      later version.
+dnl
+dnl  or both in parallel, as here.
+dnl
+dnl  The GNU MP Library is distributed in the hope that it will be useful, but
+dnl  WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+dnl  or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+dnl  for more details.
+dnl
+dnl  You should have received copies of the GNU General Public License and the
+dnl  GNU Lesser General Public License along with the GNU MP Library.  If not,
+dnl  see https://www.gnu.org/licenses/.
+
+include(`../config.m4')
+
+C         cycles/limb
+C 68040:      6
+
+ifdef(`OPERATION_add_n',`
+  define(M4_inst,       addxl)
+  define(M4_function_n, mpn_add_n)
+',`ifdef(`OPERATION_sub_n',`
+  define(M4_inst,       subxl)
+  define(M4_function_n, mpn_sub_n)
+',
+`m4_error(`Need OPERATION_add_n or OPERATION_sub_n
+')')')
+
+MULFUNC_PROLOGUE(mpn_add_n mpn_sub_n)
+
+
+C INPUT PARAMETERS
+C res_ptr	(sp + 4)
+C s1_ptr	(sp + 8)
+C s2_ptr	(sp + 12)
+C size		(sp + 16)
+
+
+PROLOGUE(M4_function_n)
+
+C Save used registers on the stack.
+	movel	d2, M(-,sp)
+	movel	a2, M(-,sp)
+
+C Copy the arguments to registers.  Better use movem?
+	movel	M(sp,12), a2
+	movel	M(sp,16), a0
+	movel	M(sp,20), a1
+	movel	M(sp,24), d2
+
+	eorw	#1, d2
+	lsrl	#1, d2
+	bcc	L(L1)
+	subql	#1, d2	C clears cy as side effect
+
+L(Loop):
+	movel	M(a0,+), d0
+	movel	M(a1,+), d1
+	M4_inst	d1, d0
+	movel	d0, M(a2,+)
+L(L1):	movel	M(a0,+), d0
+	movel	M(a1,+), d1
+	M4_inst	d1, d0
+	movel	d0, M(a2,+)
+
+	dbf	d2, L(Loop)		C loop until 16 lsb of %4 == -1
+	subxl	d0, d0			C d0 <= -cy; save cy as 0 or -1 in d0
+	subl	#0x10000, d2
+	bcs	L(L2)
+	addl	d0, d0			C restore cy
+	bra	L(Loop)
+
+L(L2):
+	negl	d0
+
+C Restore used registers from stack frame.
+	movel	M(sp,+), a2
+	movel	M(sp,+), d2
+
+	rts
+
+EPILOGUE(M4_function_n)
diff --git a/gmp-6.3.0/mpn/m68k/gmp-mparam.h b/gmp-6.3.0/mpn/m68k/gmp-mparam.h
new file mode 100644
index 0000000..9ac7b41
--- /dev/null
+++ b/gmp-6.3.0/mpn/m68k/gmp-mparam.h
@@ -0,0 +1,76 @@
+/* gmp-mparam.h -- Compiler/machine parameter header file.
+
+Copyright 2000-2004 Free Software Foundation, Inc.
+
+This file is part of the GNU MP Library.
+
+The GNU MP Library is free software; you can redistribute it and/or modify
+it under the terms of either:
+
+  * the GNU Lesser General Public License as published by the Free
+    Software Foundation; either version 3 of the License, or (at your
+    option) any later version.
+
+or
+
+  * the GNU General Public License as published by the Free Software
+    Foundation; either version 2 of the License, or (at your option) any
+    later version.
+
+or both in parallel, as here.
+
+The GNU MP 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 General Public License
+for more details.
+
+You should have received copies of the GNU General Public License and the
+GNU Lesser General Public License along with the GNU MP Library.  If not,
+see https://www.gnu.org/licenses/.  */
+
+
+#define GMP_LIMB_BITS 32
+#define GMP_LIMB_BYTES 4
+
+
+/* 25MHz 68040 */
+
+/* Generated by tuneup.c, 2004-02-05, gcc 3.2 */
+
+#define MUL_TOOM22_THRESHOLD             14
+#define MUL_TOOM33_THRESHOLD             90
+
+#define SQR_BASECASE_THRESHOLD            5
+#define SQR_TOOM2_THRESHOLD              28
+#define SQR_TOOM3_THRESHOLD              98
+
+#define DIV_SB_PREINV_THRESHOLD       MP_SIZE_T_MAX  /* never */
+#define DIV_DC_THRESHOLD                 55
+#define POWM_THRESHOLD                   65
+
+#define HGCD_THRESHOLD                  116
+#define GCD_ACCEL_THRESHOLD               3
+#define GCD_DC_THRESHOLD                590
+#define JACOBI_BASE_METHOD                2
+
+#define DIVREM_1_NORM_THRESHOLD       MP_SIZE_T_MAX  /* never */
+#define DIVREM_1_UNNORM_THRESHOLD     MP_SIZE_T_MAX  /* never */
+#define MOD_1_NORM_THRESHOLD          MP_SIZE_T_MAX  /* never */
+#define MOD_1_UNNORM_THRESHOLD        MP_SIZE_T_MAX  /* never */
+#define USE_PREINV_DIVREM_1               0
+#define USE_PREINV_MOD_1                  0
+#define DIVREM_2_THRESHOLD            MP_SIZE_T_MAX  /* never */
+#define DIVEXACT_1_THRESHOLD          MP_SIZE_T_MAX  /* never */
+#define MODEXACT_1_ODD_THRESHOLD      MP_SIZE_T_MAX  /* never */
+
+#define GET_STR_DC_THRESHOLD             18
+#define GET_STR_PRECOMPUTE_THRESHOLD     43
+#define SET_STR_THRESHOLD               937
+
+#define MUL_FFT_TABLE  { 336, 672, 1408, 3584, 10240, 24576, 0 }
+#define MUL_FFT_MODF_THRESHOLD          296
+#define MUL_FFT_THRESHOLD              1728
+
+#define SQR_FFT_TABLE  { 336, 736, 1408, 3584, 10240, 24576, 0 }
+#define SQR_FFT_MODF_THRESHOLD          296
+#define SQR_FFT_THRESHOLD              2304
diff --git a/gmp-6.3.0/mpn/m68k/lshift.asm b/gmp-6.3.0/mpn/m68k/lshift.asm
new file mode 100644
index 0000000..f202abf
--- /dev/null
+++ b/gmp-6.3.0/mpn/m68k/lshift.asm
@@ -0,0 +1,175 @@
+dnl  mc68020 mpn_lshift -- mpn left shift.
+
+dnl  Copyright 1996, 1999-2003 Free Software Foundation, Inc.
+
+dnl  This file is part of the GNU MP Library.
+dnl
+dnl  The GNU MP Library is free software; you can redistribute it and/or modify
+dnl  it under the terms of either:
+dnl
+dnl    * the GNU Lesser General Public License as published by the Free
+dnl      Software Foundation; either version 3 of the License, or (at your
+dnl      option) any later version.
+dnl
+dnl  or
+dnl
+dnl    * the GNU General Public License as published by the Free Software
+dnl      Foundation; either version 2 of the License, or (at your option) any
+dnl      later version.
+dnl
+dnl  or both in parallel, as here.
+dnl
+dnl  The GNU MP Library is distributed in the hope that it will be useful, but
+dnl  WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+dnl  or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+dnl  for more details.
+dnl
+dnl  You should have received copies of the GNU General Public License and the
+dnl  GNU Lesser General Public License along with the GNU MP Library.  If not,
+dnl  see https://www.gnu.org/licenses/.
+
+include(`../config.m4')
+
+
+C           cycles/limb
+C        shift==1  shift>1
+C 68040:    5         12
+
+
+C mp_limb_t mpn_lshift (mp_ptr res_ptr, mp_srcptr s_ptr, mp_size_t s_size,
+C                       unsigned cnt);
+C
+C The "cnt" parameter is either 16 bits or 32 bits depending on
+C SIZEOF_UNSIGNED (see ABI notes in mpn/m68k/README).  The value is of
+C course only 1 to 31.  When loaded as 16 bits there's garbage in the upper
+C half, hence the use of cmpw.  The shift instructions take the their count
+C modulo 64, so the upper part doesn't matter to them either.
+C
+
+C INPUT PARAMETERS
+C res_ptr	(sp + 4)
+C s_ptr		(sp + 8)
+C s_size	(sp + 12)
+C cnt		(sp + 16)
+
+define(res_ptr, `a1')
+define(s_ptr,   `a0')
+define(s_size,  `d6')
+define(cnt,     `d4')
+
+ifdef(`SIZEOF_UNSIGNED',,
+`m4_error(`SIZEOF_UNSIGNED not defined, should be in config.m4
+')')
+
+PROLOGUE(mpn_lshift)
+C Save used registers on the stack.
+	moveml	d2-d6/a2, M(-,sp)
+
+C Copy the arguments to registers.
+	movel	M(sp,28), res_ptr
+	movel	M(sp,32), s_ptr
+	movel	M(sp,36), s_size
+ifelse(SIZEOF_UNSIGNED,2,
+`	movew	M(sp,40), cnt',
+`	movel	M(sp,40), cnt')
+
+	moveql	#1, d5
+	cmpw	d5, cnt
+	bne	L(Lnormal)
+	cmpl	s_ptr, res_ptr
+	bls	L(Lspecial)		C jump if s_ptr >= res_ptr
+
+ifelse(scale_available_p,1,`
+	lea	M(s_ptr,s_size,l,4), a2
+',`
+	movel	s_size, d0
+	asll	#2, d0
+	lea	M(s_ptr,d0,l), a2
+')
+	cmpl	res_ptr, a2
+	bls	L(Lspecial)		C jump if res_ptr >= s_ptr + s_size
+
+L(Lnormal):
+	moveql	#32, d5
+	subl	cnt, d5
+
+ifelse(scale_available_p,1,`
+	lea	M(s_ptr,s_size,l,4), s_ptr
+	lea	M(res_ptr,s_size,l,4), res_ptr
+',`
+	movel	s_size, d0
+	asll	#2, d0
+	addl	d0, s_ptr
+	addl	d0, res_ptr
+')
+	movel	M(-,s_ptr), d2
+	movel	d2, d0
+	lsrl	d5, d0		C compute carry limb
+
+	lsll	cnt, d2
+	movel	d2, d1
+	subql	#1, s_size
+	beq	L(Lend)
+	lsrl	#1, s_size
+	bcs	L(L1)
+	subql	#1, s_size
+
+L(Loop):
+	movel	M(-,s_ptr), d2
+	movel	d2, d3
+	lsrl	d5, d3
+	orl	d3, d1
+	movel	d1, M(-,res_ptr)
+	lsll	cnt, d2
+L(L1):
+	movel	M(-,s_ptr), d1
+	movel	d1, d3
+	lsrl	d5, d3
+	orl	d3, d2
+	movel	d2, M(-,res_ptr)
+	lsll	cnt, d1
+
+	dbf	s_size, L(Loop)
+	subl	#0x10000, s_size
+	bcc	L(Loop)
+
+L(Lend):
+	movel	d1, M(-,res_ptr)	C store least significant limb
+
+C Restore used registers from stack frame.
+	moveml	M(sp,+), d2-d6/a2
+	rts
+
+C We loop from least significant end of the arrays, which is only
+C permissable if the source and destination don't overlap, since the
+C function is documented to work for overlapping source and destination.
+
+L(Lspecial):
+	clrl	d0			C initialize carry
+	eorw	#1, s_size
+	lsrl	#1, s_size
+	bcc	L(LL1)
+	subql	#1, s_size
+
+L(LLoop):
+	movel	M(s_ptr,+), d2
+	addxl	d2, d2
+	movel	d2, M(res_ptr,+)
+L(LL1):
+	movel	M(s_ptr,+), d2
+	addxl	d2, d2
+	movel	d2, M(res_ptr,+)
+
+	dbf	s_size, L(LLoop)
+	addxl	d0, d0		C save cy in lsb
+	subl	#0x10000, s_size
+	bcs	L(LLend)
+	lsrl	#1, d0		C restore cy
+	bra	L(LLoop)
+
+L(LLend):
+C Restore used registers from stack frame.
+	moveml	M(sp,+), d2-d6/a2
+	rts
+
+EPILOGUE(mpn_lshift)
diff --git a/gmp-6.3.0/mpn/m68k/m68k-defs.m4 b/gmp-6.3.0/mpn/m68k/m68k-defs.m4
new file mode 100644
index 0000000..15289f6
--- /dev/null
+++ b/gmp-6.3.0/mpn/m68k/m68k-defs.m4
@@ -0,0 +1,230 @@
+divert(-1)
+
+dnl  m4 macros for 68k assembler.
+
+dnl  Copyright 2001-2003 Free Software Foundation, Inc.
+
+dnl  This file is part of the GNU MP Library.
+dnl
+dnl  The GNU MP Library is free software; you can redistribute it and/or modify
+dnl  it under the terms of either:
+dnl
+dnl    * the GNU Lesser General Public License as published by the Free
+dnl      Software Foundation; either version 3 of the License, or (at your
+dnl      option) any later version.
+dnl
+dnl  or
+dnl
+dnl    * the GNU General Public License as published by the Free Software
+dnl      Foundation; either version 2 of the License, or (at your option) any
+dnl      later version.
+dnl
+dnl  or both in parallel, as here.
+dnl
+dnl  The GNU MP Library is distributed in the hope that it will be useful, but
+dnl  WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+dnl  or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+dnl  for more details.
+dnl
+dnl  You should have received copies of the GNU General Public License and the
+dnl  GNU Lesser General Public License along with the GNU MP Library.  If not,
+dnl  see https://www.gnu.org/licenses/.
+
+
+dnl  The default m4 `#' commenting interferes with the assembler syntax for
+dnl  immediates.  `|' would be correct, but it interferes with "||" in
+dnl  eval().  Would like to disable commenting, but that's not possible (see
+dnl  mpn/asm-defs.m4), so use `;' which should be harmless.
+
+changecom(;)
+
+
+dnl  Called: PROLOGUE_cpu(GSYM_PREFIX`'foo)
+dnl
+dnl  Same as the standard PROLOGUE, but align to 2 bytes not 4.
+
+define(`PROLOGUE_cpu',
+m4_assert_numargs(1)
+`	TEXT
+	ALIGN(2)
+	GLOBL	`$1' GLOBL_ATTR
+	TYPE(`$1',`function')
+`$1'LABEL_SUFFIX')
+
+
+dnl  Usage: d0, etc
+dnl
+dnl  Expand to d0 or %d0 according to the assembler's requirements.
+dnl
+dnl  Actually d0 expands to `d0' or %`d0', the quotes protecting against
+dnl  further expansion.  Definitions are made even if d0 is to be just `d0',
+dnl  so that any m4 quoting problems will show up everywhere, not just on a
+dnl  %d0 system.
+dnl
+dnl  Care must be taken with quoting when using these in a definition.  For
+dnl  instance the quotes in the following are essential or two %'s will be
+dnl  produced when `counter' is used.
+dnl
+dnl         define(counter, `d7')
+dnl
+
+dnl  Called: m68k_reg(r)
+define(m68k_reg,
+m4_assert_numargs(1)
+m4_assert_defined(`WANT_REGISTER_PERCENT')
+`ifelse(WANT_REGISTER_PERCENT,yes,%)`$1'')
+
+dnl  Usage: m68k_defreg(r)
+define(m68k_defreg,
+m4_assert_numargs(1)
+`deflit($1,`m68k_reg(`$1')')')
+
+m68k_defreg(d0)
+m68k_defreg(d1)
+m68k_defreg(d2)
+m68k_defreg(d3)
+m68k_defreg(d4)
+m68k_defreg(d5)
+m68k_defreg(d6)
+m68k_defreg(d7)
+
+m68k_defreg(a0)
+m68k_defreg(a1)
+m68k_defreg(a2)
+m68k_defreg(a3)
+m68k_defreg(a4)
+m68k_defreg(a5)
+m68k_defreg(a6)
+m68k_defreg(a7)
+
+m68k_defreg(sp)
+m68k_defreg(pc)
+
+
+dnl  Usage: M(base)
+dnl         M(base,displacement)
+dnl         M(base,index,size)
+dnl         M(base,index,size,scale)
+dnl         M(base,+)
+dnl         M(-,base)
+dnl
+dnl  `base' is an address register, `index' is a data register, `size' is w
+dnl  or l, and scale is 1, 2, 4 or 8.
+dnl
+dnl  M(-,base) has it's arguments that way around to emphasise it's a
+dnl  pre-decrement, as opposed to M(base,+) a post-increment.
+dnl
+dnl  Enhancement: Add the memory indirect modes, if/when they're needed.
+
+define(M,
+m4_assert_numargs_range(1,4)
+m4_assert_defined(`WANT_ADDRESSING')
+`ifelse(WANT_ADDRESSING,mit,
+`ifelse($#,1, ``$1'@')dnl
+ifelse($#,2,
+`ifelse($2,+, ``$1'@+',
+`ifelse($1,-, ``$2'@-',
+              ``$1'@($2)')')')dnl
+ifelse($#,3,  ``$1'@(`$2':`$3')')dnl
+ifelse($#,4,  ``$1'@(`$2':`$3':$4)')',
+
+dnl  WANT_ADDRESSING `motorola'
+`ifelse($#,1, `(`$1')')dnl
+ifelse($#,2,
+`ifelse($2,+, `(`$1')+',
+`ifelse($1,-, `-(`$2')',
+              `$2(`$1')')')')dnl
+ifelse($#,3,  `(`$1',`$2'.$3)')dnl
+ifelse($#,4,  `(`$1',`$2'.$3*$4)')')')
+
+
+dnl  Usage: addl etc
+dnl
+dnl  m68k instructions with special handling for the suffix, with for
+dnl  instance addl expanding to addl or add.l as necessary.
+dnl
+dnl  See also t-m68k-defs.pl which verifies all mnemonics used in the asm
+dnl  files have entries here.
+
+dnl  Called: m68k_insn(mnemonic,suffix)
+define(m68k_insn,
+m4_assert_numargs(2)
+m4_assert_defined(`WANT_DOT_SIZE')
+`ifelse(WANT_DOT_SIZE,yes, ``$1'.``$2''',
+                           ``$1$2'')')
+
+dnl  Usage: m68k_definsn(mnemonic,suffix)
+define(m68k_definsn,
+m4_assert_numargs(2)
+`deflit($1`'$2,`m68k_insn(`$1',`$2')')')
+
+m68k_definsn(add,  l)
+m68k_definsn(addx, l)
+m68k_definsn(addq, l)
+m68k_definsn(asl,  l)
+m68k_definsn(cmp,  l)
+m68k_definsn(cmp,  w)
+m68k_definsn(clr,  l)
+m68k_definsn(divu, l)
+m68k_definsn(eor,  w)
+m68k_definsn(lsl,  l)
+m68k_definsn(lsr,  l)
+m68k_definsn(move, l)
+m68k_definsn(move, w)
+m68k_definsn(movem,l)
+m68k_definsn(moveq,l)
+m68k_definsn(mulu, l)
+m68k_definsn(neg,  l)
+m68k_definsn(or,   l)
+m68k_definsn(roxl, l)
+m68k_definsn(roxr, l)
+m68k_definsn(sub,  l)
+m68k_definsn(subx, l)
+m68k_definsn(subq, l)
+
+
+dnl  Usage: bra etc
+dnl
+dnl  Expand to `bra', `jra' or `jbra' according to what the assembler will
+dnl  accept.  The latter two give variable-sized branches in gas.
+dnl
+dnl  See also t-m68k-defs.pl which verifies all the bXX branches used in the
+dnl  asm files have entries here.
+
+dnl  Called: m68k_branch(cond)
+define(m68k_branch,
+m4_assert_numargs(1)
+m4_assert_defined(`WANT_BRANCHES')
+`ifelse(WANT_BRANCHES,jra, `j$1',
+`ifelse(WANT_BRANCHES,jbra,`jb$1',
+                           ``b$1'')')')
+
+dnl  Called: m68k_defbranch(cond)
+define(m68k_defbranch,
+m4_assert_numargs(1)
+`deflit(b$1,`m68k_branch(`$1')')')
+
+m68k_defbranch(ra)
+m68k_defbranch(cc)
+m68k_defbranch(cs)
+m68k_defbranch(ls)
+m68k_defbranch(eq)
+m68k_defbranch(ne)
+
+
+dnl  Usage: scale_available_p
+dnl
+dnl  Expand to 1 if a scale factor can be used in addressing modes, or 0 if
+dnl  not.  M(a0,d0,l,4), meaning a0+d0*4, is not available in 68000 or
+dnl  68010, but is in CPU32 and in 68020 and up.
+
+define(scale_available_p,
+`m4_ifdef_anyof_p(
+`HAVE_HOST_CPU_m68360'
+`HAVE_HOST_CPU_m68020'
+`HAVE_HOST_CPU_m68030'
+`HAVE_HOST_CPU_m68040'
+`HAVE_HOST_CPU_m68060')')
+
+
+divert
diff --git a/gmp-6.3.0/mpn/m68k/mc68020/aorsmul_1.asm b/gmp-6.3.0/mpn/m68k/mc68020/aorsmul_1.asm
new file mode 100644
index 0000000..4ee30ad
--- /dev/null
+++ b/gmp-6.3.0/mpn/m68k/mc68020/aorsmul_1.asm
@@ -0,0 +1,101 @@
+dnl  mc68020 mpn_addmul_1, mpn_submul_1 -- add or subtract mpn multiple.
+
+dnl  Copyright 1992, 1994, 1996, 1999-2002, 2011 Free Software Foundation, Inc.
+
+dnl  This file is part of the GNU MP Library.
+dnl
+dnl  The GNU MP Library is free software; you can redistribute it and/or modify
+dnl  it under the terms of either:
+dnl
+dnl    * the GNU Lesser General Public License as published by the Free
+dnl      Software Foundation; either version 3 of the License, or (at your
+dnl      option) any later version.
+dnl
+dnl  or
+dnl
+dnl    * the GNU General Public License as published by the Free Software
+dnl      Foundation; either version 2 of the License, or (at your option) any
+dnl      later version.
+dnl
+dnl  or both in parallel, as here.
+dnl
+dnl  The GNU MP Library is distributed in the hope that it will be useful, but
+dnl  WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+dnl  or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+dnl  for more details.
+dnl
+dnl  You should have received copies of the GNU General Public License and the
+dnl  GNU Lesser General Public License along with the GNU MP Library.  If not,
+dnl  see https://www.gnu.org/licenses/.
+
+include(`../config.m4')
+
+C         cycles/limb
+C 68040:     25
+
+ifdef(`OPERATION_addmul_1',`
+  define(M4_inst,       addl)
+  define(M4_function_1, mpn_addmul_1)
+',`ifdef(`OPERATION_submul_1',`
+  define(M4_inst,       subl)
+  define(M4_function_1, mpn_submul_1)
+',
+`m4_error(`Need OPERATION_addmul_1 or OPERATION_submul_1
+')')')
+
+
+C INPUT PARAMETERS
+C res_ptr	(sp + 4)
+C s1_ptr	(sp + 8)
+C s1_size	(sp + 12)
+C s2_limb	(sp + 16)
+
+define(res_ptr, `a0')
+define(s1_ptr,  `a1')
+define(s1_size, `d2')
+define(s2_limb, `d4')
+
+MULFUNC_PROLOGUE(mpn_addmul_1 mpn_submul_1)
+
+PROLOGUE(M4_function_1)
+
+C Save used registers on the stack.
+	moveml	d2-d5, M(-,sp)
+
+C Copy the arguments to registers.  Better use movem?
+	movel	M(sp,20), res_ptr
+	movel	M(sp,24), s1_ptr
+	movel	M(sp,28), s1_size
+	movel	M(sp,32), s2_limb
+
+	eorw	#1, s1_size
+	clrl	d1
+	clrl	d5
+	lsrl	#1, s1_size
+	bcc	L(L1)
+	subql	#1, s1_size
+	subl	d0, d0		C (d0,cy) <= (0,0)
+
+L(Loop):
+	movel	M(s1_ptr,+), d3
+	mulul	s2_limb, d1:d3
+	addxl	d0, d3
+	addxl	d5, d1
+	M4_inst	d3, M(res_ptr,+)
+L(L1):	movel	M(s1_ptr,+), d3
+	mulul	s2_limb, d0:d3
+	addxl	d1, d3
+	addxl	d5, d0
+	M4_inst	d3, M(res_ptr,+)
+
+	dbf	s1_size, L(Loop)
+	addxl	d5, d0
+	subl	#0x10000, s1_size
+	bcc	L(Loop)
+
+C Restore used registers from stack frame.
+	moveml	M(sp,+), d2-d5
+
+	rts
+
+EPILOGUE(M4_function_1)
diff --git a/gmp-6.3.0/mpn/m68k/mc68020/mul_1.asm b/gmp-6.3.0/mpn/m68k/mc68020/mul_1.asm
new file mode 100644
index 0000000..f5fbb30
--- /dev/null
+++ b/gmp-6.3.0/mpn/m68k/mc68020/mul_1.asm
@@ -0,0 +1,96 @@
+dnl  mc68020 mpn_mul_1 -- mpn by limb multiply
+
+dnl  Copyright 1992, 1994, 1996, 1999-2002 Free Software Foundation, Inc.
+
+dnl  This file is part of the GNU MP Library.
+dnl
+dnl  The GNU MP Library is free software; you can redistribute it and/or modify
+dnl  it under the terms of either:
+dnl
+dnl    * the GNU Lesser General Public License as published by the Free
+dnl      Software Foundation; either version 3 of the License, or (at your
+dnl      option) any later version.
+dnl
+dnl  or
+dnl
+dnl    * the GNU General Public License as published by the Free Software
+dnl      Foundation; either version 2 of the License, or (at your option) any
+dnl      later version.
+dnl
+dnl  or both in parallel, as here.
+dnl
+dnl  The GNU MP Library is distributed in the hope that it will be useful, but
+dnl  WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+dnl  or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+dnl  for more details.
+dnl
+dnl  You should have received copies of the GNU General Public License and the
+dnl  GNU Lesser General Public License along with the GNU MP Library.  If not,
+dnl  see https://www.gnu.org/licenses/.
+
+include(`../config.m4')
+
+C         cycles/limb
+C 68040:     24
+
+C INPUT PARAMETERS
+C res_ptr	(sp + 4)
+C s1_ptr	(sp + 8)
+C s1_size	(sp + 12)
+C s2_limb	(sp + 16)
+
+
+define(res_ptr, `a0')
+define(s1_ptr,  `a1')
+define(s1_size, `d2')
+define(s2_limb, `d4')
+
+
+PROLOGUE(mpn_mul_1)
+
+C Save used registers on the stack.
+	moveml	d2-d4, M(-,sp)
+
+C	movel	d2, M(-,sp)
+C	movel	d3, M(-,sp)
+C	movel	d4, M(-,sp)
+
+C Copy the arguments to registers.  Better use movem?
+	movel	M(sp,16), res_ptr
+	movel	M(sp,20), s1_ptr
+	movel	M(sp,24), s1_size
+	movel	M(sp,28), s2_limb
+
+	eorw	#1, s1_size
+	clrl	d1
+	lsrl	#1, s1_size
+	bcc	L(L1)
+	subql	#1, s1_size
+	subl	d0, d0		C (d0,cy) <= (0,0)
+
+L(Loop):
+	movel	M(s1_ptr,+), d3
+	mulul	s2_limb, d1:d3
+	addxl	d0, d3
+	movel	d3, M(res_ptr,+)
+L(L1):	movel	M(s1_ptr,+), d3
+	mulul	s2_limb, d0:d3
+	addxl	d1, d3
+	movel	d3, M(res_ptr,+)
+
+	dbf	s1_size, L(Loop)
+	clrl	d3
+	addxl	d3, d0
+	subl	#0x10000, s1_size
+	bcc	L(Loop)
+
+C Restore used registers from stack frame.
+	moveml	M(sp,+), d2-d4
+
+C	movel	M(sp,+),d4
+C	movel	M(sp,+),d3
+C	movel	M(sp,+),d2
+
+	rts
+
+EPILOGUE(mpn_mul_1)
diff --git a/gmp-6.3.0/mpn/m68k/mc68020/udiv.asm b/gmp-6.3.0/mpn/m68k/mc68020/udiv.asm
new file mode 100644
index 0000000..aadeab9
--- /dev/null
+++ b/gmp-6.3.0/mpn/m68k/mc68020/udiv.asm
@@ -0,0 +1,45 @@
+dnl  mc68020 mpn_udiv_qrnnd -- 2x1 limb division
+
+dnl  Copyright 1999-2001 Free Software Foundation, Inc.
+
+dnl  This file is part of the GNU MP Library.
+dnl
+dnl  The GNU MP Library is free software; you can redistribute it and/or modify
+dnl  it under the terms of either:
+dnl
+dnl    * the GNU Lesser General Public License as published by the Free
+dnl      Software Foundation; either version 3 of the License, or (at your
+dnl      option) any later version.
+dnl
+dnl  or
+dnl
+dnl    * the GNU General Public License as published by the Free Software
+dnl      Foundation; either version 2 of the License, or (at your option) any
+dnl      later version.
+dnl
+dnl  or both in parallel, as here.
+dnl
+dnl  The GNU MP Library is distributed in the hope that it will be useful, but
+dnl  WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+dnl  or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+dnl  for more details.
+dnl
+dnl  You should have received copies of the GNU General Public License and the
+dnl  GNU Lesser General Public License along with the GNU MP Library.  If not,
+dnl  see https://www.gnu.org/licenses/.
+
+include(`../config.m4')
+
+
+C mp_limb_t mpn_udiv_qrnnd (mp_limb_t *rp,
+C                           mp_limb_t nh, mp_limb_t nl, mp_limb_t d);
+C
+
+PROLOGUE(mpn_udiv_qrnnd)
+	movel	M(sp,4), a0	C rp
+	movel	M(sp,8), d1	C nh
+	movel	M(sp,12), d0	C nl
+	divul	M(sp,16), d1:d0
+	movel	d1, M(a0)	C r
+	rts
+EPILOGUE(mpn_udiv_qrnnd)
diff --git a/gmp-6.3.0/mpn/m68k/mc68020/umul.asm b/gmp-6.3.0/mpn/m68k/mc68020/umul.asm
new file mode 100644
index 0000000..f19314e
--- /dev/null
+++ b/gmp-6.3.0/mpn/m68k/mc68020/umul.asm
@@ -0,0 +1,44 @@
+dnl  mc68020 mpn_umul_ppmm -- limb by limb multiplication
+
+dnl  Copyright 1999-2001 Free Software Foundation, Inc.
+
+dnl  This file is part of the GNU MP Library.
+dnl
+dnl  The GNU MP Library is free software; you can redistribute it and/or modify
+dnl  it under the terms of either:
+dnl
+dnl    * the GNU Lesser General Public License as published by the Free
+dnl      Software Foundation; either version 3 of the License, or (at your
+dnl      option) any later version.
+dnl
+dnl  or
+dnl
+dnl    * the GNU General Public License as published by the Free Software
+dnl      Foundation; either version 2 of the License, or (at your option) any
+dnl      later version.
+dnl
+dnl  or both in parallel, as here.
+dnl
+dnl  The GNU MP Library is distributed in the hope that it will be useful, but
+dnl  WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+dnl  or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+dnl  for more details.
+dnl
+dnl  You should have received copies of the GNU General Public License and the
+dnl  GNU Lesser General Public License along with the GNU MP Library.  If not,
+dnl  see https://www.gnu.org/licenses/.
+
+include(`../config.m4')
+
+
+C mp_limb_t mpn_umul_ppmm (mp_limb_t *lp, mp_limb_t x, mp_limb_t y);
+C
+
+PROLOGUE(mpn_umul_ppmm)
+	movel	M(sp,4), a0	C lp
+	movel	M(sp,8), d1	C x
+	movel	M(sp,12), d0	C y
+	mulul	d0, d0:d1
+	movel	d1, M(a0)	C low
+	rts
+EPILOGUE(mpn_umul_ppmm)
diff --git a/gmp-6.3.0/mpn/m68k/rshift.asm b/gmp-6.3.0/mpn/m68k/rshift.asm
new file mode 100644
index 0000000..21b5f89
--- /dev/null
+++ b/gmp-6.3.0/mpn/m68k/rshift.asm
@@ -0,0 +1,175 @@
+dnl  mc68020 mpn_rshift -- mpn right shift.
+
+dnl  Copyright 1996, 1999-2003 Free Software Foundation, Inc.
+
+dnl  This file is part of the GNU MP Library.
+dnl
+dnl  The GNU MP Library is free software; you can redistribute it and/or modify
+dnl  it under the terms of either:
+dnl
+dnl    * the GNU Lesser General Public License as published by the Free
+dnl      Software Foundation; either version 3 of the License, or (at your
+dnl      option) any later version.
+dnl
+dnl  or
+dnl
+dnl    * the GNU General Public License as published by the Free Software
+dnl      Foundation; either version 2 of the License, or (at your option) any
+dnl      later version.
+dnl
+dnl  or both in parallel, as here.
+dnl
+dnl  The GNU MP Library is distributed in the hope that it will be useful, but
+dnl  WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+dnl  or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+dnl  for more details.
+dnl
+dnl  You should have received copies of the GNU General Public License and the
+dnl  GNU Lesser General Public License along with the GNU MP Library.  If not,
+dnl  see https://www.gnu.org/licenses/.
+
+include(`../config.m4')
+
+
+C           cycles/limb
+C        shift==1  shift>1
+C 68040:    9         12
+
+
+C mp_limb_t mpn_rshift (mp_ptr res_ptr, mp_srcptr s_ptr, mp_size_t s_size,
+C                       unsigned cnt);
+C
+C The "cnt" parameter is either 16 bits or 32 bits depending on
+C SIZEOF_UNSIGNED (see ABI notes in mpn/m68k/README).  The value is of
+C course only 1 to 31.  When loaded as 16 bits there's garbage in the upper
+C half, hence the use of cmpw.  The shift instructions take the their count
+C modulo 64, so the upper part doesn't matter to them either.
+C
+
+C INPUT PARAMETERS
+C res_ptr	(sp + 4)
+C s_ptr		(sp + 8)
+C s_size	(sp + 12)
+C cnt		(sp + 16)
+
+define(res_ptr, `a1')
+define(s_ptr,   `a0')
+define(s_size,  `d6')
+define(cnt,     `d4')
+
+ifdef(`SIZEOF_UNSIGNED',,
+`m4_error(`SIZEOF_UNSIGNED not defined, should be in config.m4
+')')
+
+PROLOGUE(mpn_rshift)
+C Save used registers on the stack.
+	moveml	d2-d6/a2, M(-,sp)
+
+C Copy the arguments to registers.
+	movel	M(sp,28), res_ptr
+	movel	M(sp,32), s_ptr
+	movel	M(sp,36), s_size
+ifelse(SIZEOF_UNSIGNED,2,
+`	movew	M(sp,40), cnt',
+`	movel	M(sp,40), cnt')
+
+	moveql	#1, d5
+	cmpw	d5, cnt
+	bne	L(Lnormal)
+	cmpl	res_ptr, s_ptr
+	bls	L(Lspecial)		C jump if res_ptr >= s_ptr
+
+ifelse(scale_available_p,1,`
+	lea	M(res_ptr,s_size,l,4), a2
+',`
+	movel	s_size, d0
+	asll	#2, d0
+	lea	M(res_ptr,d0,l), a2
+')
+	cmpl	s_ptr, a2
+	bls	L(Lspecial)		C jump if s_ptr >= res_ptr + s_size
+
+L(Lnormal):
+	moveql	#32, d5
+	subl	cnt, d5
+	movel	M(s_ptr,+), d2
+	movel	d2, d0
+	lsll	d5, d0		C compute carry limb
+
+	lsrl	cnt, d2
+	movel	d2, d1
+	subql	#1, s_size
+	beq	L(Lend)
+	lsrl	#1, s_size
+	bcs	L(L1)
+	subql	#1, s_size
+
+L(Loop):
+	movel	M(s_ptr,+), d2
+	movel	d2, d3
+	lsll	d5, d3
+	orl	d3, d1
+	movel	d1, M(res_ptr,+)
+	lsrl	cnt, d2
+L(L1):
+	movel	M(s_ptr,+), d1
+	movel	d1, d3
+	lsll	d5, d3
+	orl	d3, d2
+	movel	d2, M(res_ptr,+)
+	lsrl	cnt, d1
+
+	dbf	s_size, L(Loop)
+	subl	#0x10000, s_size
+	bcc	L(Loop)
+
+L(Lend):
+	movel	d1, M(res_ptr)	C store most significant limb
+
+C Restore used registers from stack frame.
+	moveml	M(sp,+), d2-d6/a2
+	rts
+
+C We loop from most significant end of the arrays, which is only permissable
+C if the source and destination don't overlap, since the function is
+C documented to work for overlapping source and destination.
+
+L(Lspecial):
+ifelse(scale_available_p,1,`
+	lea	M(s_ptr,s_size,l,4), s_ptr
+	lea	M(res_ptr,s_size,l,4), res_ptr
+',`
+	movel	s_size, d0
+	asll	#2, d0
+	addl	d0, s_ptr
+	addl	d0, res_ptr
+')
+
+	clrl	d0			C initialize carry
+	eorw	#1, s_size
+	lsrl	#1, s_size
+	bcc	L(LL1)
+	subql	#1, s_size
+
+L(LLoop):
+	movel	M(-,s_ptr), d2
+	roxrl	#1, d2
+	movel	d2, M(-,res_ptr)
+L(LL1):
+	movel	M(-,s_ptr), d2
+	roxrl	#1, d2
+	movel	d2, M(-,res_ptr)
+
+	dbf	s_size, L(LLoop)
+	roxrl	#1, d0		C save cy in msb
+	subl	#0x10000, s_size
+	bcs	L(LLend)
+	addl	d0, d0		C restore cy
+	bra	L(LLoop)
+
+L(LLend):
+C Restore used registers from stack frame.
+	moveml	M(sp,+), d2-d6/a2
+	rts
+
+EPILOGUE(mpn_rshift)
diff --git a/gmp-6.3.0/mpn/m68k/t-m68k-defs.pl b/gmp-6.3.0/mpn/m68k/t-m68k-defs.pl
new file mode 100644
index 0000000..91c21fa
--- /dev/null
+++ b/gmp-6.3.0/mpn/m68k/t-m68k-defs.pl
@@ -0,0 +1,91 @@
+#! /usr/bin/perl -w
+
+# Copyright 2001, 2003 Free Software Foundation, Inc.
+#
+#  This file is part of the GNU MP Library.
+#
+#  The GNU MP Library is free software; you can redistribute it and/or modify
+#  it under the terms of either:
+#
+#    * the GNU Lesser General Public License as published by the Free
+#      Software Foundation; either version 3 of the License, or (at your
+#      option) any later version.
+#
+#  or
+#
+#    * the GNU General Public License as published by the Free Software
+#      Foundation; either version 2 of the License, or (at your option) any
+#      later version.
+#
+#  or both in parallel, as here.
+#
+#  The GNU MP 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 General Public License
+#  for more details.
+#
+#  You should have received copies of the GNU General Public License and the
+#  GNU Lesser General Public License along with the GNU MP Library.  If not,
+#  see https://www.gnu.org/licenses/.
+
+
+# Usage:  perl t-m68k-defs.pl [-t]
+#
+# Run this in the mpn/m68k source directory to check that m68k-defs.m4 has
+# m68k_defbranch()s or m68k_definsn()s for each instruction used in *.asm
+# and */*.asm.  Print nothing if everything is ok.  The -t option prints
+# some diagnostic traces.
+
+use strict;
+use Getopt::Std;
+
+my %opt;
+getopts('t', \%opt);
+
+my %branch;
+my %insn;
+
+open(FD, "<m68k-defs.m4")
+    or die "Cannot open m68k-defs.m4: $!\nIs this the mpn/m68k source directory?\n";
+my ($srcdir, $top_srcdir);
+while (<FD>) {
+    if (/^m68k_defbranch\(\s*(.*)\)/) { $branch{"b".$1} = 1; }
+    if (/^m68k_definsn\(\s*(.*),\s*(.*)\)/) { $insn{$1.$2} = 1; }
+}
+close(FD);
+
+print "branches: ", join(" ",keys(%branch)), "\n" if $opt{'t'};
+print "insns: ", join(" ",keys(%insn)), "\n" if $opt{'t'};
+
+
+foreach my $file (glob("*.asm"), glob("*/*.asm")) {
+    print "file $file\n" if $opt{'t'};
+
+    open(FD, "<$file") or die "Cannot open $file: $!";
+    while (<FD>) {
+	if (/^[ \t]*C/) { next; };
+	if (/^\t([a-z0-9]+)/) {
+	    my $opcode = $1;
+	    print "opcode $1\n" if $opt{'t'};
+
+	    # instructions with an l, w or b suffix should have a definsn
+	    # (unless they're already a defbranch)
+	    if ($opcode =~ /[lwb]$/
+		&& ! defined $insn{$opcode}
+		&& ! defined $branch{$opcode})
+	    {
+		print "$file: $.: missing m68k_definsn: $opcode\n";
+	    }
+
+	    # instructions bXX should have a defbranch (unless they're
+	    # already a definsn)
+	    if ($opcode =~ /^b/
+		&& ! defined $insn{$opcode}
+		&& ! defined $branch{$opcode})
+	    {
+		print "$file: $.: missing m68k_defbranch: $opcode\n";
+	    }
+	}
+    }
+    close(FD);
+}