Init commit

programs/parva/add_carry_1.lbpasm

@@ -0,0 +1 @@
+.asm_version 0.1

programs/parva/assembler_1.lbpasm

@@ -0,0 +1,48 @@
+hash_table:
+	.data blablabla
+
+
+parse_line:
+	li 
+	rw t0, 0(a0)
+	srli t1, 18
+	beqz end_of_line
+	sub t1, t0, t1
+
+end_of_line:
+
+.char_encoding lbpc1
+
+get_instruction:
+
+a-z 0-9 () , space enter
+
+li x2, 0 ; cursor position in line
+
+listen:
+iord x0, 1, 0 ; get char from keyboard (device 1, buffer 0)
+beqz listen
+sub x1, x0, 'a'
+.eq diff 'z' - 'a'
+bleu x1, diff, okay
+sub x1, x0, '0'
+.eq diff '9' - '0'
+bleu x1, diff, okay
+
+beq x0, ' ', okay
+beq x0, '\n', okay
+beq x0, '(', okay
+beq x0, ')', okay
+beq x0, ',', okay
+b invalid_input
+
+invalid_input:
+iow 3, 1, 0o1214 ; play buzzing sound (sound card is device 3, play direct value command is 1)
+b listen
+
+okay:
+iow 2, 2, x0, 0 ; print character x0 (gpu is device 2, print char command is 2)
+addi x2, x2, 1 ; move cursor right
+
+
+

programs/parva/binary_to_decimal_1.lbpasm

@@ -0,0 +1,103 @@
+lw x0, value
+li x1, 0
+
+# https://ridiculousfish.com/blog/posts/labor-of-division-episode-i.html
+dec_to_bin_loop:
+	lw x4, ten_powers(x1)
+	beqz x4, end_dec_to_bin_loop
+	slli x5, x1, 1
+	lw x3, ten_divisors(x5)
+	lw x5, (ten_divisors + 1)(x5)
+	mulhu x6, x0, x3
+	sub x7, x0, x6
+	srli x7, x7, 1
+	add x3, x7, x6
+	srl x3, x3, x5 # x3 = digit of result
+	sw x3, result_string(x1)
+	mulu x3, x3, x4
+	sub x0, x0, x3
+	addi x1, x1, 1
+	b dec_to_bin_loop
+end_dec_to_bin_loop:
+
+sw x0, result_string(x1)
+li x0, -1
+li x6, 1 # cursor x
+li x7, 1 # cursor y
+sw x0, 0b01_0100_000000(upper) # gpu clear screen
+
+skip_zeroes:
+	addi x0, x0, 1
+	lw x1, result_string(x0)
+	beqz x1, skip_zeroes
+print_loop:
+	sd x67, 0b01_0000_000000(upper) # gpu move cursor
+	lw x1, result_string(x0)
+	bltz x1, end_print_loop
+	slli x1, x1, 1
+	lw x2, char_pixels(x1)
+	lw x3, (char_pixels + 1)(x1)
+	sd x23, 0b01_0010_000000(upper) # gpu print char
+	addi x6, x6, 4
+	addi x0, x0, 1
+	b print_loop
+end_print_loop:
+
+sd x01, 0b01_0011_000000(upper) # gpu show buffer
+wfi
+
+ten_powers:
+.data 10000000
+.data 1000000
+.data 100000
+.data 10000
+.data 1000
+.data 100
+.data 10
+.data 0
+
+ten_divisors:
+.data 0xad7f2a 23 # 10000000
+.data 0x0c6f7b 19 # 1000000
+.data 0x4f8b59 16 # 100000
+.data 0xa36e2f 13 # 10000
+.data 0x0624de 9 # 1000
+.data 0x47ae15 6 # 100
+.data 0x99999a 3 # 10
+
+char_pixels:
+.data 0b111000_101000_101000_101000 # 0
+.data 0b111000_000000_000000_000000
+
+.data 0b001000_001000_001000_001000 # 1
+.data 0b001000_000000_000000_000000
+
+.data 0b111000_001000_111000_100000 # 2
+.data 0b111000_000000_000000_000000
+
+.data 0b111000_001000_111000_001000 # 3
+.data 0b111000_000000_000000_000000
+
+.data 0b101000_101000_111000_001000 # 4
+.data 0b001000_000000_000000_000000
+
+.data 0b111000_100000_111000_001000 # 5
+.data 0b111000_000000_000000_000000
+
+.data 0b111000_100000_111000_101000 # 6
+.data 0b111000_000000_000000_000000
+
+.data 0b111000_001000_001000_010000 # 7
+.data 0b010000_000000_000000_000000
+
+.data 0b111000_101000_111000_101000 # 8
+.data 0b111000_000000_000000_000000
+
+.data 0b111000_101000_111000_001000 # 9
+.data 0b001000_000000_000000_000000
+result_string:
+.repeat 0 8
+.data -1
+
+value:
+.data 6942069

programs/parva/binary_to_decimal_2.lbpasm

@@ -0,0 +1,109 @@
+li sp, stack
+
+j main
+
+main:
+	sw x0, gpu_clear_screen(upper)
+
+	li x6, 1
+	li x7, 1
+	sd x67, cursor_pos
+	li a0, 6942069
+	call print_number
+	
+	li x6, 1
+	li x7, 7
+	sd x67, cursor_pos
+	li a0, 12345678
+	call print_number
+
+	wfi
+
+.eq gpu_clear_screen 0b01_0100_000000
+.eq gpu_move_cursor 0b01_0000_000000
+.eq gpu_print_char 0b01_0010_000000
+.eq gpu_show_buffer 0b01_0011_000000
+
+
+# void print_number(*void return: ra, int number: a0)
+.address 128
+print_number:
+	push ra
+	mv x0, a0
+	li x2, 0
+	li x4, 10000000
+
+	.align 2
+	print_number__bin_to_dec_loop:
+		divu x3, x0, x4
+		remu x0, x0, x4
+		sw x3, print_number__result_string(x2)
+		divui x4, x4, 10
+		addi x2, x2, 1
+		bnez x4, print_number__bin_to_dec_loop
+
+	li x0, -1
+	ld x67, cursor_pos
+	print_number__skip_zeroes:
+		addi x0, x0, 1
+		lw x4, print_number__result_string(x0)
+		beqz x4, print_number__skip_zeroes
+	print_number__print_loop:
+		lw x4, print_number__result_string(x0)
+		bltz x4, print_number__end_print_loop
+		sd x67, gpu_move_cursor(upper)
+		slli x4, x4, 1
+		ld x23, print_number__char_pixels(x4)
+		addi x6, x6, 4
+		addi x0, x0, 1
+		sd x23, gpu_print_char(upper)
+		b print_number__print_loop
+	print_number__end_print_loop:
+	sd x01, gpu_show_buffer(upper) # gpu show buffer
+	sw x6, (cursor_pos + 0)
+	pop ra
+	j ra
+
+	print_number__char_pixels:
+	.data 0b111000_101000_101000_101000 # 0
+	.data 0b111000_000000_000000_000000
+
+	.data 0b001000_001000_001000_001000 # 1
+	.data 0b001000_000000_000000_000000
+
+	.data 0b111000_001000_111000_100000 # 2
+	.data 0b111000_000000_000000_000000
+
+	.data 0b111000_001000_111000_001000 # 3
+	.data 0b111000_000000_000000_000000
+
+	.data 0b101000_101000_111000_001000 # 4
+	.data 0b001000_000000_000000_000000
+
+	.data 0b111000_100000_111000_001000 # 5
+	.data 0b111000_000000_000000_000000
+
+	.data 0b111000_100000_111000_101000 # 6
+	.data 0b111000_000000_000000_000000
+
+	.data 0b111000_001000_001000_010000 # 7
+	.data 0b010000_000000_000000_000000
+
+	.data 0b111000_101000_111000_101000 # 8
+	.data 0b111000_000000_000000_000000
+
+	.data 0b111000_101000_111000_001000 # 9
+	.data 0b001000_000000_000000_000000
+	print_number__result_string:
+	.repeat 0 8
+	.data -1
+# end print_number()
+
+.align 2
+cursor_pos:
+.data 1
+.data 1
+
+.align 2
+stack:
+.data 0

programs/parva/calculator_1.lbpasm

@@ -0,0 +1,96 @@
+# target parva_0_1
+
+li sp, stack
+
+# function
+tokenize:
+	push ra
+	addi x2, a0, -1 # x2 = pointer to input string
+
+	li x0, -1
+	push x0 # mark the beginning of the token stack
+
+tokenize__loop:
+	addi x2, x2, 1
+	lw x3, x2
+	beqz x3, tokenize__end
+	li x4, '0'
+	blt x3, x4, tokenize__symbol
+	li x5, '9'
+	bgt x3, x5, tokenize__symbol
+
+	mv x6, sp
+.align 2
+tokenize__number_gather:
+	push x3
+	addi x2, x2, 1
+	addi x3, x3, -'0'
+	lw x3, x2
+	blt x3, x4, tokenize__number_gather_end
+	bgt x3, x5, tokenize__number_gather_end
+	b tokenize__number_gather
+tokenize__number_gather_end:
+	li x0, 0 # x0 = sum
+	li x7, 1 # x7 = power of 10
+.align 2
+tokenize__number_sum:
+	pop x3
+	mulu x3, x3, x7
+	add x0, x0, x3
+	mului x7, x7, 10
+	bne x6, sp, tokenize__number_sum
+
+	li x3, 1 # token type = 1 = number
+	push x3
+	push x0
+	b tokenize__loop
+
+tokenize__symbol:
+	xori x3, x3, 42 # todo, set to appropriate value for hashing symbols
+	lw x3, tokenize__symbol_hashtable(x3)
+	beqz x3, error_invalid_input
+	push x3
+	b tokenize__loop
+
+tokenize__end:
+
+tokenize__shunting_yard:
+	pop x3
+	bnez x3, 3
+	
+
+
+b end
+
+tokenize__output_queue:
+	.repeat 0 40
+
+tokenize__symbol_hashtable:
+	.data 0
+	.data 1
+	.repeat 0 30 # todo
+
+error_invalid_input:
+	li a0, error_message_invalid_input
+	li a1, 1
+	call print
+	b end
+
+# function
+print:
+	# todo
+	ret
+
+end:
+	wfi
+
+input_string:
+.string_encoding terminal
+.string "125+23*8"
+.string "\0"
+
+error_message_invalid_input:
+.string_encoding ascii # todo, use correct encoding
+.string "Invalid input\0"
+
+stack:

programs/parva/cli_1.lbpasm

@@ -0,0 +1,13 @@
+split_args: ; fn([str8; len]) -> ( [[str8; len]; len] )
+	push ra
+	ld x01, 0(r0) ; x0 = pointer to first char, x1 = length
+loop:
+	lw t0, 0(a0)
+	andi t0, a0, 0o77
+	beqz end
+	
+	end:
+
+hash:
+	
+

programs/parva/fibonacci_1.lbpasm

@@ -0,0 +1,12 @@
+li x0, 0
+li x1, 1
+li x2, 20
+
+loop:
+
+add x0, x0, x1
+add x1, x0, x1
+subi x2, x2, 1
+bnez x2, loop
+
+wfi

programs/parva/fibonacci_2.lbpasm

@@ -0,0 +1,24 @@
+li x0, 1
+li x1, 0
+
+li x2, 30 ; nth fibonacci number to calculate
+
+andi x4, x2, 1 ; x4 = n odd?
+srli x2, x2, 1 ; n /= 2
+
+loop:
+
+add x0, x0, x1
+add x1, x0, x1
+subi x2, x2, 1
+bnez x2, loop
+
+beqz x4, no_final_iteration
+
+add x1, x0, x1
+
+no_final_iteration:
+
+mv x0, x1
+li x1, 0
+wfi

programs/parva/fibonacci_3.lbpasm

@@ -0,0 +1,24 @@
+li x0, 1
+li x1, 0
+
+li x2, 30 ; nth fibonacci number to calculate
+
+andi x4, x2, 1 ; x4 = n odd?
+srli x2, x2, 1 ; n /= 2
+
+loop:
+
+add x0, x0, x1
+add x1, x0, x1
+subi x2, x2, 1
+bnez x2, loop
+
+beqz x4, no_final_iteration
+
+add x1, x0, x1
+
+no_final_iteration:
+
+mv x0, x1
+li x1, 0
+wfi

programs/parva/gpu_hello_world_1.lbpasm

@@ -0,0 +1,93 @@
+; CPU version 0.1
+; GPU version 0.1 on IO port 1
+
+nop
+b _start
+
+characters:
+char_H: ; 0
+.data 0b000000_100010_100010_111110 0b100010_100010_100010_000000
+char_e: ; 1
+.data 0b000000_000000_011000_100100 0b111100_100000_011100_000000
+char_l: ; 2
+.data 0b000000_100000_100000_100000 0b100000_100000_010000_000000
+char_o: ; 3
+.data 0b000000_000000_011000_100100 0b100100_100100_011000_000000
+char_w: ; 4
+.data 0b000000_000000_101010_101010 0b101010_101010_010100_000000
+char_r: ; 5
+.data 0b000000_000000_111000_100000 0b100000_100000_100000_000000
+char_d: ; 6
+.data 0b000100_011100_100100_100100 0b100100_100100_011100_000000
+
+_start:
+
+li x6, 1 ; x position
+li x7, 1 ; y position
+
+sw zero, 0b01_0100_000000(upper) ; gpu clear screen
+sd x67, 0b01_0000_000000(upper) ; gpu move cursor
+
+; H
+lw x0, 0(char_H)(zero) ; upper 24 pixels
+lw x1, 1(char_H)(zero) ; lower 24 pixels
+sd x01, 0b01_0010_000000(upper) ; gpu print 6x8 char
+addi x6, x6, 6 ; x position += 6
+
+; e
+lw x0, 0(char_e)(zero)
+lw x1, 1(char_e)(zero)
+sd x01, 0b01_0010_000000(upper)
+addi x6, x6, 6
+
+; l
+lw x0, 0(char_l)(zero)
+lw x1, 1(char_l)(zero)
+sd x01, 0b01_0010_000000(upper)
+addi x6, x6, 6
+
+; l
+lw x0, 0(char_l)(zero)
+lw x1, 1(char_l)(zero)
+sd x01, 0b01_0010_000000(upper)
+addi x6, x6, 6
+
+; o
+lw x0, 0(char_o)(zero)
+lw x1, 1(char_o)(zero)
+sd x01, 0b01_0010_000000(upper)
+addi x6, x6, 6
+
+; w
+lw x0, 0(char_w)(zero)
+lw x1, 1(char_w)(zero)
+sd x01, 0b01_0010_000000(upper)
+addi x6, x6, 6
+
+; o
+lw x0, 0(char_o)(zero)
+lw x1, 1(char_o)(zero)
+sd x01, 0b01_0010_000000(upper)
+addi x6, x6, 6
+
+; r
+lw x0, 0(char_r)(zero)
+lw x1, 1(char_r)(zero)
+sd x01, 0b01_0010_000000(upper)
+addi x6, x6, 6
+
+; l
+lw x0, 0(char_l)(zero)
+lw x1, 1(char_l)(zero)
+sd x01, 0b01_0010_000000(upper)
+addi x6, x6, 6
+
+; d
+lw x0, 0(char_d)(zero)
+lw x1, 1(char_d)(zero)
+sd x01, 0b01_0010_000000(upper)
+addi x6, x6, 6
+
+sd x01, 0b01_0011_000000(upper) ; gpu print buffer to screen
+
+wfi

programs/parva/gpu_hello_world_2.lbpasm

@@ -0,0 +1,117 @@
+; CPU version 0.1
+; GPU version 0.1 on IO port 1
+
+nop
+b _start
+
+characters:
+char_H_top: ; 0
+.data 0b000000_100010_100010_111110
+char_H_bottom:
+.data 0b100010_100010_100010_000000
+char_e_top: ; 1
+.data 0b000000_000000_011000_100100
+char_e_bottom:
+.data 0b111100_100000_011100_000000
+char_l_top: ; 2
+.data 0b000000_100000_100000_100000
+char_l_bottom:
+.data 0b100000_100000_010000_000000
+char_o_top: ; 3
+.data 0b000000_000000_011000_100100
+char_o_bottom:
+.data 0b100100_100100_011000_000000
+char_w_top: ; 4
+.data 0b000000_000000_101010_101010
+char_w_bottom:
+.data 0b101010_101010_010100_000000
+char_r_top: ; 5
+.data 0b000000_000000_111000_100000
+char_r_bottom:
+.data 0b100000_100000_100000_000000
+char_d_top: ; 6
+.data 0b000100_011100_100100_100100
+char_d_bottom:
+.data 0b100100_100100_011100_000000
+
+_start:
+
+li x6, 1 ; x position
+li x7, 1 ; y position
+
+sw x0, 0b01_0100_000000(upper) ; gpu clear screen
+sd x67, 0b01_0000_000000(upper) ; gpu move cursor
+
+; H
+lw x0, char_H_top(zero) ; upper 24 pixels
+lw x1, char_H_bottom(zero) ; lower 24 pixels
+sd x01, 0b01_0010_000000(upper) ; gpu print 6x8 char
+addi x6, x6, 6 ; x position += 6
+sd x67, 0b01_0000_000000(upper) ; gpu move cursor
+
+; e
+lw x0, char_e_top(zero)
+lw x1, char_e_bottom(zero)
+sd x01, 0b01_0010_000000(upper)
+addi x6, x6, 6
+sd x67, 0b01_0000_000000(upper) ; gpu move cursor
+
+; l
+lw x0, char_l_top(zero)
+lw x1, char_l_bottom(zero)
+sd x01, 0b01_0010_000000(upper)
+addi x6, x6, 6
+sd x67, 0b01_0000_000000(upper) ; gpu move cursor
+
+; l
+lw x0, char_l_top(zero)
+lw x1, char_l_bottom(zero)
+sd x01, 0b01_0010_000000(upper)
+addi x6, x6, 6
+sd x67, 0b01_0000_000000(upper) ; gpu move cursor
+
+; o
+lw x0, char_o_top(zero)
+lw x1, char_o_bottom(zero)
+sd x01, 0b01_0010_000000(upper)
+addi x6, x6, 6
+sd x67, 0b01_0000_000000(upper) ; gpu move cursor
+
+; w
+lw x0, char_w_top(zero)
+lw x1, char_w_bottom(zero)
+sd x01, 0b01_0010_000000(upper)
+addi x6, x6, 6
+sd x67, 0b01_0000_000000(upper) ; gpu move cursor
+
+; o
+lw x0, char_o_top(zero)
+lw x1, char_o_bottom(zero)
+sd x01, 0b01_0010_000000(upper)
+addi x6, x6, 6
+sd x67, 0b01_0000_000000(upper) ; gpu move cursor
+
+; r
+lw x0, char_r_top(zero)
+lw x1, char_r_bottom(zero)
+sd x01, 0b01_0010_000000(upper)
+addi x6, x6, 6
+sd x67, 0b01_0000_000000(upper) ; gpu move cursor
+
+; l
+lw x0, char_l_top(zero)
+lw x1, char_l_bottom(zero)
+sd x01, 0b01_0010_000000(upper)
+addi x6, x6, 6
+sd x67, 0b01_0000_000000(upper) ; gpu move cursor
+
+; d
+lw x0, char_d_top(zero)
+lw x1, char_d_bottom(zero)
+sd x01, 0b01_0010_000000(upper)
+nop
+nop
+nop ; wait for print to finish
+sd x01, 0b01_0011_000000(upper) ; gpu flush buffer to screen
+
+wfi

programs/parva/gpu_hello_world_3.lbpasm

@@ -0,0 +1,59 @@
+; Parva CPU version 0.1
+; Parva GPU version 0.1 on IO port 0x1
+
+nop
+b _start
+
+string:
+.data 1 2 3 3 4 8 5 4 6 3 7 0
+
+characters:
+; upper 24 pixels, lower 24 pixels, width
+char_null: ; 0
+.data 0
+characters_plus_1:
+.data 0
+characters_plus_2:
+.data 0
+char_H: ; 1
+.data 0b000000_100010_100010_111110 0b100010_100010_100010_000000 6
+char_e: ; 2
+.data 0b000000_000000_011000_100100 0b111100_100000_011100_000000 5
+char_l: ; 3
+.data 0b000000_100000_100000_100000 0b100000_100000_010000_000000 3
+char_o: ; 4
+.data 0b000000_000000_011000_100100 0b100100_100100_011000_000000 5
+char_w: ; 5
+.data 0b000000_000000_101010_101010 0b101010_101010_010100_000000 6
+char_r: ; 6
+.data 0b000000_000000_011000_100000 0b100000_100000_100000_000000 4
+char_d: ; 7
+.data 0b000000_000100_011100_100100 0b100100_100100_011100_000000 5
+char_space: ; 8
+.data 0 0 4
+
+_start:
+
+sw x0, 0b01_0100_000000(upper) ; gpu clear screen
+li x6, 1 ; x position
+li x7, 1 ; y position
+li x0, string
+loop:
+sd x67, 0b01_0000_000000(upper) ; gpu move cursor
+lw x1, 0(x0) ; x1 = current character
+beqz x1, end
+add x2, x1, x1
+add x1, x2, x1 ; x1 *= 3
+lw x2, characters(x1) ; upper 24 pixels
+lw x3, characters_plus_1(x1) ; lower 24 pixels
+sd x01, 0b01_0011_000000(upper) ; gpu print buffer to screen
+sd x23, 0b01_0010_000000(upper) ; gpu print 6x8 char
+lw x4, characters_plus_2(x1) ; width
+add x6, x6, x4 ; x position += width
+addi x0, x0, 1
+b loop
+
+end:
+
+sd x01, 0b01_0011_000000(upper) ; gpu print buffer to screen
+wfi

programs/parva/gpu_print_1

@@ -0,0 +1,35 @@
+font_data:
+	.data 0x8b18a9 0xf9a9d7 0xeba9a0 0xdeadbe 0xefdead 0xbeefde...
+
+.export _start:
+	li x0, font_data
+	li x1, 16 # font data array length
+	add x3, x0, x1
+
+loop_1:
+	lw x2, 0(x0)
+	gpu.sw x2, 0(x1)
+	addi x0, 1
+	blt x0, x3 loop_1
+
+	li x6, 0
+loop_2:
+	kb.rw x0
+	beq x0, x6, loop_2
+
+	and x1, x0, 0x0f
+	li x4, 2 # x position
+	li x5, 2 # y position
+
+	# first char
+	srl x1, x0, 4
+	gpu.move x4, x5
+	gpu.pchar x1
+
+	# second char
+	addi x4, x4, 6
+	gpu.move x4, x5
+	gpu.pchar x1
+
+	gpu.swap
+	b loop_2

programs/parva/memtest_1.lbpasm

@@ -0,0 +1,33 @@
+li x0, program_end ; x0 = start
+li x1, 0xff ; x1 = end
+
+mv x2, x0
+
+loop_write:
+
+xori x3, x2, 0xaaa
+sw x3, 0(x2)
+addi x2, x2, 1
+ble x2, x1, loop_write
+
+mv x2, x0
+
+loop_read:
+
+xori x3, x2, 0xaaa
+lw x4, 0(x2)
+bne x3, x4, error
+addi x2, x2, 1
+ble x2, x1, loop_read
+
+success:
+
+li x0, 0
+wfi
+
+error:
+
+mv x0, x2
+wfi
+
+program_end:

programs/parva/memtest_2.lbpasm

@@ -0,0 +1,62 @@
+li x0, program_end
+li x1, 512
+
+mv x2, x0
+
+.align 2
+loop_write_word:
+
+xori x3, x2, 0xaaa
+sw x3, 0(x2)
+addi x2, x2, 1
+blt x2, x1, loop_write_word
+
+mv x2, x0
+
+.align 2
+loop_read_word:
+
+xori x3, x2, 0xaaa
+lw x4, 0(x2)
+bne x3, x4, error
+addi x2, x2, 1
+blt x2, x1, loop_read_word
+
+.align 2
+loop_write_double:
+
+xori x6, x2, 0xaaa
+xori x7, x2, 0x555
+sd x67, 0(x2)
+addi x2, x2, 2
+blt x2, x1, loop_write_double
+
+mv x2, x0
+
+.align 2
+loop_read_double:
+
+xori x6, x2, 0xaaa
+xori x7, x2, 0x555
+ld x45, 0(x2)
+bne x4, x6, error
+bne x5, x7, error
+lw x4, 0(x2)
+bne x4, x6, error
+lw x5, 1(x2)
+bne x5, x7, error
+addi x2, x2, 2
+blt x2, x1, loop_read_double
+
+success:
+
+li x0, 0
+wfi
+
+error:
+
+mv x0, x2
+wfi
+
+.align 2
+program_end:

programs/parva/multiply_1.lbpasm

@@ -0,0 +1,16 @@
+li a0, 420
+li a1, 69
+
+multiply:
+	li t0, 0
+loop:
+	andi t1, a1, 1
+	beqz t1, no_add
+	add t0, t0, a0
+no_add:
+	slli a0, a0, 1
+	srli a1, a1, 1
+	bnez a1, loop
+	mv a0, t0
+	j ra
+

programs/parva/pythagorean_triples_1.lbpasm

@@ -0,0 +1,105 @@
+; calculate all primitive pythogorean triples (a^2 + b^2 = c^2) where c <= 32
+
+nop
+b _start
+
+results:
+.repeat 0 15
+
+table:
+.repeat 0 33
+
+result_index:
+.data 0
+
+_start:
+
+li x7, 32 ; limit
+li x0, 1 ; prime
+li x2, 1 ; prime factor bit
+slli x2, x2, 23
+
+loop_prime:
+
+addi x0, x0, 1
+bgt x0, x7, squares
+lw x4, table(x0)
+srli x4, x4, 12
+bnez x4, loop_prime
+mv x3, x0 ; multiple
+
+loop_index:
+
+lw x4, table(x3) ; x4 = current prime factor bits
+or x4, x4, x2 ; x4 |= add current prime factor bit
+sw x4, table(x3)
+add x3, x3, x0
+ble x3, x7, loop_index
+srli x2, x2, 1 ; update prime bit
+b loop_prime
+
+squares:
+
+li x0, 0 ; n
+li x1, 0 ; n^2
+
+loop_squares:
+
+lw x4, table(x0) ; 70
+or x4, x4, x1
+sw x4, table(x0)
+add x1, x1, x0
+add x1, x1, x0
+addi x1, x1, 1
+addi x0, x0, 1
+ble x0, x7, loop_squares
+
+li x0, 1 ; a = 1
+
+iterate_a:
+
+addi x0, x0, 1 ; 79
+bgt x0, x7, end
+lw x2, table(x0) ; a^2 (with prime bits)
+mv x6, x0 ; b = a
+
+iterate_b:
+
+addi x6, x6, 1 ; 83
+bgt x6, x7, iterate_a
+lw x3, table(x6) ; b^2 (with prime bits)
+and x4, x2, x3 ; check if a and b are coprime
+srli x4, x4, 12
+bnez x4, iterate_b ; 87
+add x3, x2, x3
+andi x3, x3, 0o77 ; a^2 + b^2
+mv x5, x6 ; x5 = c = b
+
+; x0 = a, x1 = sp, x2 = a^2, x3 = a^2+b^2, x4 = (unused)
+; x5 = c, x6 = b, x7 = limit
+
+iterate_c:
+
+addi x5, x5, 1 ; c++ ; 92
+bgt x5, x7, iterate_b
+lw x4, table(x5) ; x4 = c^2|cp
+andi x4, x4, 0o77 ; x4 = c^2
+bgt x4, x3, iterate_b
+bne x5, x3, iterate_c
+
+; found triple
+
+; x0 = a, x6 = b, x5 = c
+
+lw x4, result_index
+sw x0, 0(x4)
+sw x6, 1(x4)
+sw x5, 2(x4)
+addi x4, x4, 3
+sw x4, result_index
+
+b iterate_a
+
+end:
+
+wfi

programs/parva/pythagorean_triples_2.lbpasm

@@ -0,0 +1,273 @@
+; calculate all primitive pythogorean triples (a^2 + b^2 = c^2) where c <= 32
+
+.eq limit 32
+
+li sp, stack
+
+li x0, 1
+sw x0, cursor_x
+sw x0, cursor_y
+
+sw x0, gpu_clear_screen(upper)
+
+# main stack frame:
+# sp-(limit + 2): table
+# sp-2: a
+# sp-1: a^2
+main: 
+	li x7, limit # limit
+	mv x6, sp # x6 = table base
+	add sp, sp, x7
+	addi sp, sp, 2 # reserve rest of stack frame
+
+	# --- zero-fill table ---
+
+	mv x5, x6
+	li x2, 0
+
+	.align 2
+	main__clear_table:
+	sw x2, 0(x5)
+	addi x5, x5, 1
+	ble x5, sp, main__clear_table
+
+	# --- calculate primes factors ---
+
+	li x0, 1 # x0 = prime
+	li x2, 1 # x2 = prime factor bit
+	
+	main__primes:
+
+	addi x0, x0, 1
+	bgt x0, x7, main__primes_end
+	add x4, x6, x0
+	lw x4, 0(x4)
+	bnez x4, main__primes
+	
+	mv x3, x0 # x3 = multiple
+
+	main__multiples:
+
+	add x5, x6, x3
+	lw x4, 0(x5) # x4 = current prime factor bits
+	or x4, x4, x2 # x4 |= add current prime factor bit
+	sw x4, 0(x5)
+	add x3, x3, x0
+	ble x3, x7, main__multiples
+
+	slli x2, x2, 1 # update prime bit
+	b main__primes
+
+	main__primes_end:
+
+	# --- find pythagorean triples ---
+
+	li x0, 1 # a = 1
+
+	main__iterate_a:
+	
+	lw x0, -2(sp)
+	addi x0, x0, 1 # x0 = a
+	sw x0, -2(sp)
+	bge x0, x7, main__end
+	mulu x2, x0, x0 # x2 = a^2
+	sw x2, -1(sp)
+	add x3, x6, x0
+	lw x3, 0(x3) # x3 = a prime factors
+
+	mv x4, x0 # x4 = b
+
+	main__iterate_b:
+
+	addi x4, x4, 1
+	bgt x4, x7, main__iterate_a
+	add x5, x6, x4
+	lw x5, 0(x5) # x5 = b prime factors
+	and x5, x5, x3
+	bnez x5, main__iterate_b # check if a and b are coprime
+	mulu x5, x4, x4
+	lw x0, -1(sp) # x0 = a^2
+	add x5, x5, x0 # x5 = a^2 + b^2
+
+	mv x0, x4 # x0 = c
+
+	addi x2, x0, 1
+	mulu x2, x2, x2
+	bgt x2, x5, main__iterate_a
+
+	main__iterate_c:
+
+	addi x0, x0, 1 ; c += 1
+	mulu x2, x0, x0 # x2 = c^2
+	bgt x2, x5, main__iterate_b
+	bne x2, x5, main__iterate_c
+
+	# found triple
+	
+	push x6
+	push x7
+	push x0
+	push x4
+	lw a0, -6(sp)
+	call print_number
+
+	lw x6, cursor_x
+	addi x6, x6, 4
+	sw x6, cursor_x
+	
+	pop a0
+	call print_number
+
+	lw x6, cursor_x
+	addi x6, x6, 4
+	sw x6, cursor_x
+	
+	pop a0
+	call print_number
+	
+	# newline
+	li x6, 1
+	sw x6, cursor_x
+	lw x7, cursor_y
+	addi x7, x7, 6
+	sw x7, cursor_y
+
+	pop x7
+	pop x6
+
+	b main__iterate_a
+
+	main__end:
+
+	wfi
+# end main()
+
+.eq gpu_clear_screen 0b01_0100_000000
+.eq gpu_move_cursor 0b01_0000_000000
+.eq gpu_print_char 0b01_0010_000000
+.eq gpu_show_buffer 0b01_0011_000000
+
+cursor_x:
+.data 1
+cursor_y:
+.data 1
+
+# void print_number(*void return: ra, int number: a0)
+print_number:
+	push ra
+	mv x0, a0
+	li x2, 0
+
+	.align 2
+	# https://ridiculousfish.com/blog/posts/labor-of-division-episode-i.html
+	print_number__bin_to_dec_loop:
+		lw x4, print_number__ten_powers(x2)
+
+		beqz x4, print_number__end_bin_to_dec_loop
+		slli x5, x2, 1
+
+		lw x3, print_number__ten_divisors(x5)
+
+		mulhu x6, x0, x3
+		sub x7, x0, x6
+
+		srli x7, x7, 1
+		lw x5, (print_number__ten_divisors + 1)(x5)
+
+		add x3, x7, x6
+		srl x3, x3, x5 # x3 = digit of result
+
+		sw x3, print_number__result_string(x2)
+		mulu x3, x3, x4
+		
+		sub x0, x0, x3
+		addi x2, x2, 1
+		
+		b print_number__bin_to_dec_loop
+	print_number__end_bin_to_dec_loop:
+
+	sw x0, print_number__result_string(x2)
+	li x0, -1
+	lw x6, cursor_x
+	lw x7, cursor_y
+	print_number__skip_zeroes:
+		addi x0, x0, 1
+		lw x4, print_number__result_string(x0)
+		beqz x4, print_number__skip_zeroes
+	print_number__print_loop:
+		sd x67, gpu_move_cursor(upper)
+		nop
+		nop
+		nop # give time for the cursor to move
+		lw x4, print_number__result_string(x0)
+		bltz x4, print_number__end_print_loop
+		slli x4, x4, 1
+		lw x2, print_number__char_pixels(x4)
+		lw x3, (print_number__char_pixels + 1)(x4)
+		addi x6, x6, 4
+		addi x0, x0, 1
+		sd x23, gpu_print_char(upper)
+		b print_number__print_loop
+	print_number__end_print_loop:
+	sd x01, gpu_show_buffer(upper) # gpu show buffer
+	sw x6, cursor_x
+	pop ra
+	j ra
+
+	print_number__ten_powers:
+	.data 10000000
+	.data 1000000
+	.data 100000
+	.data 10000
+	.data 1000
+	.data 100
+	.data 10
+	.data 0
+
+	print_number__ten_divisors:
+	.data 0xad7f2a 23 # 10000000
+	.data 0x0c6f7b 19 # 1000000
+	.data 0x4f8b59 16 # 100000
+	.data 0xa36e2f 13 # 10000
+	.data 0x0624de 9 # 1000
+	.data 0x47ae15 6 # 100
+	.data 0x99999a 3 # 10
+
+	print_number__char_pixels:
+	.data 0b111000_101000_101000_101000 # 0
+	.data 0b111000_000000_000000_000000
+
+	.data 0b001000_001000_001000_001000 # 1
+	.data 0b001000_000000_000000_000000
+
+	.data 0b111000_001000_111000_100000 # 2
+	.data 0b111000_000000_000000_000000
+
+	.data 0b111000_001000_111000_001000 # 3
+	.data 0b111000_000000_000000_000000
+
+	.data 0b101000_101000_111000_001000 # 4
+	.data 0b001000_000000_000000_000000
+
+	.data 0b111000_100000_111000_001000 # 5
+	.data 0b111000_000000_000000_000000
+
+	.data 0b111000_100000_111000_101000 # 6
+	.data 0b111000_000000_000000_000000
+
+	.data 0b111000_001000_001000_010000 # 7
+	.data 0b010000_000000_000000_000000
+
+	.data 0b111000_101000_111000_101000 # 8
+	.data 0b111000_000000_000000_000000
+
+	.data 0b111000_101000_111000_001000 # 9
+	.data 0b001000_000000_000000_000000
+	print_number__result_string:
+	.repeat 0 8
+	.data -1
+# end print_number()
+
+.align 2
+stack:
+.data 0

programs/parva/tetris_1.lbpasm

@@ -0,0 +1,197 @@
+.arch parva_0_1
+
+# 1(sp): score
+# 2(sp): swap_block
+
+# x4: block lines 0/1
+# x5: block lines 2/3
+# x6: field lines 0/1
+# x7: field lines 2/3
+
+.include consts
+.include sys
+
+.string_encoding sys-6
+
+
+.data.ro
+
+srs_kick_table:
+# XY, 0 = 0, 1 = +1, 2 = +2, 7 = -1, 6 = -2
+# I block
+0o60_10_67_12
+0o70_20_72_27
+# TODO
+# J, L, T, S, Z blocks
+0o70_71_06_76
+0o10_17_02_12
+# TODO
+
+block_table:
+# I 0
+0b_0_0000_000000_0_0_1111_000000_0
+0b_0_1111_000000_0_0_0000_000000_0
+# I 1
+0b_0_0010_000000_0_0_0010_000000_0
+0b_0_0010_000000_0_0_0010_000000_0
+# TODO
+
+hi_score_filename: .cstring "tet_hiscr.dat"
+hi_score_file_error_message: .cstring .line_break 10 "could not open or create hi score file"
+score_string: .cstring "score"
+game_over_string: .cstring "game over"
+.scores_button_string: .cstring "scores"
+.start_button_string: .cstring "start"
+.next_string: .cstring "next"
+
+.data.rw
+
+.align const.cache_line_size
+swap_block: 0
+score: 0
+next_air_drop_time: 0
+next_floor_drop_time: 0
+stack: .repeat 0, 20
+
+hi_scores:
+.repeat 0, 20
+
+.eq score_string_x 30
+.eq score_string_y 10
+
+
+.text 
+
+_start:
+
+push ra
+
+load_hi_scores:
+
+li a0, hi_score_filename
+li a1, hi_scores
+li x0, 20
+call sys.read_file
+li x0, -1
+beq a0, x0, create_hi_score_file
+
+create_hi_score_file:
+
+li a0, hi_score_filename
+call sys.open_file
+
+bltz a0, error_hi_score_file
+
+wait:
+
+wait_loop:
+
+stub.kb.rdchar x0
+
+beqz, x0, wait
+li x1, const.char.left_arrow
+beq x0, x1, move_left
+li x1, const.char.right_arrow
+beq x0, x1, move_right
+li x1, const.char.up_arrow
+
+beq x0, x1, rotate
+li x1, const.char.down_arrow
+beq x0, x1, slow_drop
+li x1, const.char.space
+beq x0, x1, hard_drop
+li x1, const.char.backspace
+
+beq x0, x1, swap
+b wait_loop
+
+rotate:
+
+# x2 = current pos / block / rotation, [0 * 8, 4 bits for position x, 0 * 7, 3 bits for block type, 2 bits for rotation]
+
+addi x3, x2, 1
+andi x3, x3, 0b00011
+andi x2, x2, 0b11100
+or x2, x2, x3
+ll x01 block_table(x2)
+
+.align 2
+move_left:
+
+slli x0, x4, 1
+and x0, x0, x6
+bneqz x0, fail
+slli x0, x5, 1
+and x0, x0, x7
+bneqz x0, fail
+
+slli x4, x4, 1
+slli x5, x5, 1
+b move_successful
+
+.align 2
+move_right:
+
+srli x0, x4, 1
+and x0, x0, x6
+bneqz x0, fail
+srli x0, x5, 1
+and x0, x0, x7
+bneqz x0, fail
+
+srli x4, x4, 1
+srli x5, x5, 1
+b move_successful
+
+move_failed:
+
+# TODO
+
+move_successful:
+
+# TODO
+
+# decimal addition
+# input: x0 = score change
+.align 2
+add_score:
+
+lw x1, score
+add x0, x0, x1
+li x1, 0o1166
+and x2, x0, 0o77
+li x3, 10
+blt x2, x3, 2
+
+add x0, x0, x1
+slli x1, x1, 6
+srli x2, x0, 6
+and x2, x0, 0o77
+blt x2, x3, end_score_change
+add x0, x0, x1
+
+slli x1, x1, 6
+srli x2, x0, 6
+and x2, x0, 0o77
+blt x2, x3, end_score_change
+add x0, x0, x1
+slli x1, x1, 6
+
+srli x2 x0, 6
+and x2, x0, 0o77
+blt x2, x3, end_score_change
+add x0, x0, x1
+
+.align 2
+end_score_change:
+
+sw x0, score
+
+
+
+
+quit:
+
+
+
+pop ra

programs/v8/test

@@ -0,0 +1,30 @@
+LDB #$F0
+JNC bankNoCarry
+NOT B
+bankNoCarry:
+LDA $00
+AND B,A
+OR D,A
+STA $00
+RET
+
+; Memory copy: A=start, B=offset, length=C, destroys D
+copyLoop:
+LDD $00,A
+ADC B,A
+STD $00,A
+DEC C
+JNZ copyLoop
+RET
+
+; Memory copy to screen: A=start, destroys C, D, screen address register
+LDD #$80
+LDC D
+STD $03
+screenCopyLoop:
+
+LDD $00,A
+STD $02
+DEC C
+JNZ screenCopyLoop
+RET