65 lines
		
	
	
	
		
			2 KiB
			
		
	
	
	
		
			ArmAsm
		
	
	
	
	
	
		
		
			
		
	
	
			65 lines
		
	
	
	
		
			2 KiB
			
		
	
	
	
		
			ArmAsm
		
	
	
	
	
	
| 
								 | 
							
								;;  Copyright 2010  Free Software Foundation, Inc.
							 | 
						||
| 
								 | 
							
								;;  Contributed by Bernd Schmidt <bernds@codesourcery.com>.
							 | 
						||
| 
								 | 
							
								;;
							 | 
						||
| 
								 | 
							
								;; This program is free software; you can redistribute it and/or modify
							 | 
						||
| 
								 | 
							
								;; it under the terms of 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.
							 | 
						||
| 
								 | 
							
								;;
							 | 
						||
| 
								 | 
							
								;; This program 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 a copy of the GNU General Public License
							 | 
						||
| 
								 | 
							
								;; along with this program; if not, write to the Free Software
							 | 
						||
| 
								 | 
							
								;; Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
							 | 
						||
| 
								 | 
							
								
							 | 
						||
| 
								 | 
							
								#include <linux/linkage.h>
							 | 
						||
| 
								 | 
							
								
							 | 
						||
| 
								 | 
							
									;; ABI considerations for the divide functions
							 | 
						||
| 
								 | 
							
									;; The following registers are call-used:
							 | 
						||
| 
								 | 
							
									;; __c6xabi_divi A0,A1,A2,A4,A6,B0,B1,B2,B4,B5
							 | 
						||
| 
								 | 
							
									;; __c6xabi_divu A0,A1,A2,A4,A6,B0,B1,B2,B4
							 | 
						||
| 
								 | 
							
									;; __c6xabi_remi A1,A2,A4,A5,A6,B0,B1,B2,B4
							 | 
						||
| 
								 | 
							
									;; __c6xabi_remu A1,A4,A5,A7,B0,B1,B2,B4
							 | 
						||
| 
								 | 
							
									;;
							 | 
						||
| 
								 | 
							
									;; In our implementation, divu and remu are leaf functions,
							 | 
						||
| 
								 | 
							
									;; while both divi and remi call into divu.
							 | 
						||
| 
								 | 
							
									;; A0 is not clobbered by any of the functions.
							 | 
						||
| 
								 | 
							
									;; divu does not clobber B2 either, which is taken advantage of
							 | 
						||
| 
								 | 
							
									;; in remi.
							 | 
						||
| 
								 | 
							
									;; divi uses B5 to hold the original return address during
							 | 
						||
| 
								 | 
							
									;; the call to divu.
							 | 
						||
| 
								 | 
							
									;; remi uses B2 and A5 to hold the input values during the
							 | 
						||
| 
								 | 
							
									;; call to divu.  It stores B3 in on the stack.
							 | 
						||
| 
								 | 
							
								
							 | 
						||
| 
								 | 
							
									.text
							 | 
						||
| 
								 | 
							
								
							 | 
						||
| 
								 | 
							
								ENTRY(__c6xabi_remi)
							 | 
						||
| 
								 | 
							
									stw	.d2t2	B3, *B15--[2]
							 | 
						||
| 
								 | 
							
								||	cmpgt	.l1	0, A4, A1
							 | 
						||
| 
								 | 
							
								||	cmpgt	.l2	0, B4, B2
							 | 
						||
| 
								 | 
							
								||	mv	.s1	A4, A5
							 | 
						||
| 
								 | 
							
								||	call	.s2	__c6xabi_divu
							 | 
						||
| 
								 | 
							
								
							 | 
						||
| 
								 | 
							
								   [A1]	neg	.l1	A4, A4
							 | 
						||
| 
								 | 
							
								|| [B2]	neg	.l2	B4, B4
							 | 
						||
| 
								 | 
							
								||	xor	.s2x	B2, A1, B0
							 | 
						||
| 
								 | 
							
								||	mv	.d2	B4, B2
							 | 
						||
| 
								 | 
							
								
							 | 
						||
| 
								 | 
							
								   [B0]	addkpc	.s2	_divu_ret_1, B3, 1
							 | 
						||
| 
								 | 
							
								  [!B0] addkpc	.s2	_divu_ret_2, B3, 1
							 | 
						||
| 
								 | 
							
									nop	2
							 | 
						||
| 
								 | 
							
								_divu_ret_1:
							 | 
						||
| 
								 | 
							
									neg	.l1	A4, A4
							 | 
						||
| 
								 | 
							
								_divu_ret_2:
							 | 
						||
| 
								 | 
							
									ldw	.d2t2	*++B15[2], B3
							 | 
						||
| 
								 | 
							
								
							 | 
						||
| 
								 | 
							
									mpy32	.m1x	A4, B2, A6
							 | 
						||
| 
								 | 
							
									nop		3
							 | 
						||
| 
								 | 
							
									ret	.s2	B3
							 | 
						||
| 
								 | 
							
									sub	.l1	A5, A6, A4
							 | 
						||
| 
								 | 
							
									nop	4
							 | 
						||
| 
								 | 
							
								ENDPROC(__c6xabi_remi)
							 |