start.s

嵌入式ARM的一些源代码

	.globl	init_8260_core
init_8260_core:

	/* Initialize machine status; enable machine check interrupt	*/
	/*--------------------------------------------------------------*/

	li	r3, MSR_KERNEL		/* Set ME and RI flags */
	rlwimi	r3, r5, 0, 25, 25	/* preserve IP bit set by HRCW */
	rlwimi	r3, r5, 0, 21, 22	/* debugger might set SE & BE bits */
	SYNC				/* Some chip revs need this... */
	mtmsr	r3
	SYNC
	mtspr	SRR1, r3		/* Make SRR1 match MSR */

	/* Initialize the Hardware Implementation-dependent Registers	*/
	/* HID0 also contains cache control				*/
	/*--------------------------------------------------------------*/

	lis	r3, CFG_HID0_INIT@h
	ori	r3, r3, CFG_HID0_INIT@l
	SYNC
	mtspr	HID0, r3

	lis	r3, CFG_HID0_FINAL@h
	ori	r3, r3, CFG_HID0_FINAL@l
	SYNC
	mtspr	HID0, r3

	lis	r3, CFG_HID2@h
	ori	r3, r3, CFG_HID2@l
	mtspr	HID2, r3

	/* clear all BAT's						*/
	/*--------------------------------------------------------------*/

	li	r0, 0
	mtspr	DBAT0U, r0
	mtspr	DBAT0L, r0
	mtspr	DBAT1U, r0
	mtspr	DBAT1L, r0
	mtspr	DBAT2U, r0
	mtspr	DBAT2L, r0
	mtspr	DBAT3U, r0
	mtspr	DBAT3L, r0
	mtspr	IBAT0U, r0
	mtspr	IBAT0L, r0
	mtspr	IBAT1U, r0
	mtspr	IBAT1L, r0
	mtspr	IBAT2U, r0
	mtspr	IBAT2L, r0
	mtspr	IBAT3U, r0
	mtspr	IBAT3L, r0
	SYNC

	/* invalidate all tlb's						*/
	/*								*/
	/* From the 603e User Manual: "The 603e provides the ability to	*/
	/* invalidate a TLB entry. The TLB Invalidate Entry (tlbie)	*/
	/* instruction invalidates the TLB entry indexed by the EA, and	*/
	/* operates on both the instruction and data TLBs simultaneously*/
	/* invalidating four TLB entries (both sets in each TLB). The	*/
	/* index corresponds to bits 15-19 of the EA. To invalidate all	*/
	/* entries within both TLBs, 32 tlbie instructions should be	*/
	/* issued, incrementing this field by one each time."		*/
	/*								*/
	/* "Note that the tlbia instruction is not implemented on the	*/
	/* 603e."							*/
	/*								*/
	/* bits 15-19 correspond to addresses 0x00000000 to 0x0001F000	*/
	/* incrementing by 0x1000 each time. The code below is sort of	*/
	/* based on code in "flush_tlbs" from arch/ppc/kernel/head.S	*/
	/*								*/
	/*--------------------------------------------------------------*/

	li	r3, 32
	mtctr	r3
	li	r3, 0
1:	tlbie	r3
	addi	r3, r3, 0x1000
	bdnz	1b
	SYNC

	/* Clear the timebase registers					*/
	/*--------------------------------------------------------------*/

	li	r0, 0
	mtspr	TBWL, r0
	mtspr	TBWU, r0
	mtspr	TBWL, r0

	/* Done!							*/
	/*--------------------------------------------------------------*/

	blr

/* Cache functions.
 *
 * Note: requires that all cache bits in
 * HID0 are in the low half word.
 */
	.globl	icache_enable
icache_enable:
	mfspr	r3, HID0
	ori	r3, r3, HID0_ICE
	lis	r4, 0
	ori	r4, r4, HID0_ILOCK
	andc	r3, r3, r4
	ori	r4, r3, HID0_ICFI
	isync
	mtspr	HID0, r4	/* sets enable and invalidate, clears lock */
	isync
	mtspr	HID0, r3	/* clears invalidate */
	blr

	.globl	icache_disable
icache_disable:
	mfspr	r3, HID0
	lis	r4, 0
	ori	r4, r4, HID0_ICE|HID0_ILOCK
	andc	r3, r3, r4
	ori	r4, r3, HID0_ICFI
	isync
	mtspr	HID0, r4	/* sets invalidate, clears enable and lock */
	isync
	mtspr	HID0, r3	/* clears invalidate */
	blr

	.globl	icache_status
icache_status:
	mfspr	r3, HID0
	rlwinm	r3, r3, HID0_ICE_BITPOS + 1, 31, 31
	blr

	.globl	dcache_enable
dcache_enable:
	mfspr	r3, HID0
	ori	r3, r3, HID0_DCE
	lis	r4, 0
	ori	r4, r4, HID0_DLOCK
	andc	r3, r3, r4
	ori	r4, r3, HID0_DCI
	sync
	mtspr	HID0, r4	/* sets enable and invalidate, clears lock */
	sync
	mtspr	HID0, r3	/* clears invalidate */
	blr

	.globl	dcache_disable
dcache_disable:
	mfspr	r3, HID0
	lis	r4, 0
	ori	r4, r4, HID0_DCE|HID0_DLOCK
	andc	r3, r3, r4
	ori	r4, r3, HID0_DCI
	sync
	mtspr	HID0, r4	/* sets invalidate, clears enable and lock */
	sync
	mtspr	HID0, r3	/* clears invalidate */
	blr

	.globl	dcache_status
dcache_status:
	mfspr	r3, HID0
	rlwinm	r3, r3, HID0_DCE_BITPOS + 1, 31, 31
	blr

/*
 * Delay for a number of ticks
 */
	.globl	wait_ticks
wait_ticks:
1:	mftbu	r4
	mftb	r5
	mftbu	r6
	cmp	0,r4,r6
	bne	1b		/* Get [synced] base time */
	addc	r9,r5,r3	/* Compute end time */
	addze	r8,r4
2:	mftbu	r4
	cmp	0,r4,r8
	blt	2b
	bgt	3f
	mftb	r5
	cmp	0,r5,r9
	blt	2b
3:	blr

	.globl get_pvr
get_pvr:
	mfspr	r3, PVR
	blr

/*------------------------------------------------------------------------------*/

/*
 * void relocate_code (addr_sp, bd, addr_moni)
 *
 * This "function" does not return, instead it continues in RAM
 * after relocating the monitor code.
 *
 * r3 = dest
 * r4 = src
 * r5 = length in bytes
 * r6 = cachelinesize
 */
	.globl	relocate_code
relocate_code:
	mr	r1,  r3		/* Set new stack pointer		*/
	mr	r9,  r4		/* Save copy of Board Info pointer	*/
	mr	r10, r5		/* Save copy of Destination Address	*/

	mr	r3,  r5				/* Destination Address	*/
	lis	r4, CFG_MONITOR_BASE@h		/* Source      Address	*/
	ori	r4, r4, CFG_MONITOR_BASE@l
	lis	r5, CFG_MONITOR_LEN@h		/* Length in Bytes	*/
	ori	r5, r5, CFG_MONITOR_LEN@l
	li	r6, CFG_CACHELINE_SIZE		/* Cache Line Size	*/

	/*
	 * Fix GOT pointer:
	 *
	 * New GOT-PTR = (old GOT-PTR - CFG_MONITOR_BASE) + Destination Address
	 *
	 * Offset:
	 */
	sub	r15, r10, r4

	/* First our own GOT */
	add	r14, r14, r15
	/* then the one used by the C code */
	add	r30, r30, r15

	/*
	 * Now relocate code
	 */

	cmplw	cr1,r3,r4
	addi	r0,r5,3
	srwi.	r0,r0,2
	beq	cr1,4f		/* In place copy is not necessary	*/
	beq	7f		/* Protect against 0 count		*/
	mtctr	r0
	bge	cr1,2f

	la	r8,-4(r4)
	la	r7,-4(r3)
1:	lwzu	r0,4(r8)
	stwu	r0,4(r7)
	bdnz	1b
	b	4f

2:	slwi	r0,r0,2
	add	r8,r4,r0
	add	r7,r3,r0
3:	lwzu	r0,-4(r8)
	stwu	r0,-4(r7)
	bdnz	3b

/*
 * Now flush the cache: note that we must start from a cache aligned
 * address. Otherwise we might miss one cache line.
 */
4:	cmpwi	r6,0
	add	r5,r3,r5
	beq	7f		/* Always flush prefetch queue in any case */
	subi	r0,r6,1
	andc	r3,r3,r0
	mfspr	r7,HID0		/* don't do dcbst if dcache is disabled */
	rlwinm	r7,r7,HID0_DCE_BITPOS+1,31,31
	cmpwi	r7,0
	beq	9f
	mr	r4,r3
5:	cmplw	r4,r5
	dcbst	0,r4
	add	r4,r4,r6
	blt	5b
	sync			/* Wait for all dcbst to complete on bus */
9:	mfspr	r7,HID0		/* don't do icbi if icache is disabled */
	rlwinm	r7,r7,HID0_ICE_BITPOS+1,31,31
	cmpwi	r7,0
	beq	7f
	mr	r4,r3
6:	cmplw	r4,r5
	icbi	0,r4
	add	r4,r4,r6
	blt	6b
7:	sync			/* Wait for all icbi to complete on bus	*/
	isync

/*
 * We are done. Do not return, instead branch to second part of board
 * initialization, now running from RAM.
 */

	addi	r0, r10, in_ram - _start + EXC_OFF_SYS_RESET
	mtlr	r0
	blr

in_ram:

	/*
	 * Relocation Function, r14 point to got2+0x8000
	 *
         * Adjust got2 pointers, no need to check for 0, this code
         * already puts a few entries in the table.
	 */
	li	r0,__got2_entries@sectoff@l
	la	r3,GOT(_GOT2_TABLE_)
	lwz	r11,GOT(_GOT2_TABLE_)
	mtctr	r0
	sub	r11,r3,r11
	addi	r3,r3,-4
1:	lwzu	r0,4(r3)
	add	r0,r0,r11
	stw	r0,0(r3)
	bdnz	1b
	
	/*
         * Now adjust the fixups and the pointers to the fixups
	 * in case we need to move ourselves again.
	 */	
2:	li	r0,__fixup_entries@sectoff@l
	lwz	r3,GOT(_FIXUP_TABLE_)
	cmpwi	r0,0
	mtctr	r0
	addi	r3,r3,-4
	beq	4f
3:	lwzu	r4,4(r3)
	lwzux	r0,r4,r11
	add	r0,r0,r11
	stw	r10,0(r3)
	stw	r0,0(r4)
	bdnz	3b
4:
clear_bss:
	/*
	 * Now clear BSS segment
	 */
	lwz	r3,GOT(.bss)
	lwz	r4,GOT(_end)

	cmplw	0, r3, r4
	beq	6f

	li	r0, 0
5:
	stw	r0, 0(r3)
	addi	r3, r3, 4
	cmplw	0, r3, r4
	bne	5b
6:

	mr	r3, r9		/* Board Info pointer		*/
	mr	r4, r10		/* Destination Address		*/
	bl	board_init_r

	/* Problems accessing "end" in C, so do it here */
	.globl	get_endaddr
get_endaddr:
	lwz	r3,GOT(_end)
	blr

	/*
	 * Copy exception vector code to low memory
	 *
	 * r3: dest_addr
	 * r7: source address, r8: end address, r9: target address
	 */
	.globl	trap_init
trap_init:
	lwz	r7, GOT(_start)
	lwz	r8, GOT(_end_of_vectors)

	rlwinm	r9, r7, 0, 18, 31	/* _start & 0x3FFF	*/

	cmplw	0, r7, r8
	bgelr				/* return if r7>=r8 - just in case */

	mflr	r4			/* save link register		*/
1:
	lwz	r0, 0(r7)
	stw	r0, 0(r9)
	addi	r7, r7, 4
	addi	r9, r9, 4
	cmplw	0, r7, r8
	bne	1b

	/*
	 * relocate `hdlr' and `int_return' entries
	 */
	li	r7, .L_MachineCheck - _start + EXC_OFF_SYS_RESET
	li	r8, Alignment - _start + EXC_OFF_SYS_RESET
2:
	bl	trap_reloc
	addi	r7, r7, 0x100		/* next exception vector	*/
	cmplw	0, r7, r8
	blt	2b

	li	r7, .L_Alignment - _start + EXC_OFF_SYS_RESET
	bl	trap_reloc

	li	r7, .L_ProgramCheck - _start + EXC_OFF_SYS_RESET
	bl	trap_reloc

	li	r7, .L_FPUnavailable - _start + EXC_OFF_SYS_RESET
	li	r8, _end_of_vectors - _start + EXC_OFF_SYS_RESET
3:
	bl	trap_reloc
	addi	r7, r7, 0x100		/* next exception vector	*/
	cmplw	0, r7, r8
	blt	3b

	mfmsr	r3			/* now that the vectors have	*/
	lis	r7, MSR_IP@h		/* relocated into low memory	*/
	ori	r7, r7, MSR_IP@l	/* MSR[IP] can be turned off	*/
	andc	r3, r3, r7		/* (if it was on)		*/
	SYNC				/* Some chip revs need this... */
	mtmsr	r3
	SYNC

	mtlr	r4			/* restore link register	*/
	blr

	/*
	 * Function: relocate entries for one exception vector
	 */
trap_reloc:
	lwz	r0, 0(r7)		/* hdlr ...			*/
	add	r0, r0, r3		/*  ... += dest_addr		*/
	stw	r0, 0(r7)

	lwz	r0, 4(r7)		/* int_return ...		*/
	add	r0, r0, r3		/*  ... += dest_addr		*/
	stw	r0, 4(r7)

	blr