Diff: NeoCore.s

Revision:

0:9f237b11f0a8

Child:

2:cf5c8199c733

--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/NeoCore.s	Wed Mar 12 18:36:58 2014 +0000
@@ -0,0 +1,218 @@
+;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
+; NeoCore.s
+; 
+; Allen Wild
+; March 2014
+;
+; ARM assembly functions for writing to Adafruit NeoPixels
+; with the mbed NXP LPC1768
+;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
+	AREA neo_core, CODE, READONLY
+	
+	IMPORT neo_fio_reg
+	IMPORT neo_bitmask
+	
+;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
+neo_write_pin
+; Set the GPIO pin to the value passed in R0
+; Registers and bitmasks are stored in variables set by the C++ library
+	LDR		R1, =neo_fio_reg	; load pointers to register values
+	LDR		R2, =neo_bitmask
+	LDR		R1, [R1]			; load actual values from memory
+	LDR		R2, [R2]
+
+	CMP		R0, #0			; VALUE == 0 ?
+	ITE EQ					; (IF-THEN-ELSE) ON NEXT TWO INSTRUCTIONS USING "EQ" FLAG
+	STREQ	R2, [R1,#0x1C]	; if==0, CLEAR BIT
+	STRNE	R2, [R1,#0x18]	; if==1, SET BIT
+	BX		LR
+
+;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
+neo_zero
+; Output a NeoPixel zero, composed of a short
+; HIGH pulse and a long LOW pulse
+	PUSH	{LR}
+	MOV		R0, #1
+	BL		neo_write_pin	; set pin high
+	
+	MOV		R0, #10			; delay for long enough
+	BL		neo_delay
+	
+	MOV		R0, #0			; set pin low
+	BL		neo_write_pin
+	
+	MOV		R0, #20			; delay
+	BL		neo_delay
+	
+	POP		{LR}
+	BX		LR
+
+;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
+neo_one
+; Output a NeoPixel one, composed of a long
+; HIGH pulse and a short LOW pulse
+	PUSH	{LR}
+	MOV		R0, #1
+	BL		neo_write_pin
+	
+	MOV		R0, #86
+	BL		neo_delay
+	
+	MOV		R0, #0
+	BL		neo_write_pin
+	
+	NOP		; really short delay
+	NOP
+	NOP
+	
+	POP		{LR}
+	BX		LR
+	
+;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
+	EXPORT neo_out ; void neo_out(int *data, int n);
+; Main function called from the C++ library
+; R0 contains a pointer to the array of color data to send
+; R1 contains the number of bytes of data to send
+neo_out
+	PUSH	{LR, R4, R5, R6, R7, R8}
+	MOV		R7, R1		; move length to R7
+	MOV		R6, R0		; move address to R6
+	
+neo_byteloop
+	LDRB	R5, [R6]	; load byte to send
+	MOV		R4, #0x80	; load initial bitmask
+	
+neo_bitloop
+	AND		R3, R5, R4	; mask current byte
+	CMP		R3, #0
+	BLEQ	neo_zero	; send current bit
+	BLNE	neo_one
+	
+	LSR		R4, R4, #1	; shift bitmask right one
+	CMP		R4, #0		; if still more bits, loop back
+	BNE		neo_bitloop
+	
+	ADD		R6, R6, #1	; increment address
+	SUB		R7, R7, #1	; decrement count
+	CMP		R7, #0
+	BNE		neo_byteloop	; continue if not done
+
+	MOV		R0, #0
+	BL		neo_write_pin
+	POP		{R8, R7, R6, R5, R4, LR}
+	BX		LR
+
+;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
+ 
+neo_delay
+; delay the specified number of cycles in R0 with a bunch of nops
+	LDR		R2, =neo_delay_end
+	SUB		R2, R2, R0
+	BX		R2
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+	NOP
+neo_delay_end
+	BX	 LR
+
+	END ; end code region
+
+
+