fix sw imm 2x problem, and fix jalr rs1 and rd mixup problem

src/riscv_assembler/convert.py

@@ -157,9 +157,11 @@ def write_to_file(output : list, file : str) -> None:
 		raise NotImplementedError()
 
 	def mod(self, output : list) -> list:
+		print('inside mod:', output)
 		if self.__nibble_mode:
 			output = AssemblyConverter.apply_nibble(output)
 		elif self.__hex_mode:
+			print('here we convert to hex', output)
 			output = AssemblyConverter.apply_hex(output)
 		return output
 

src/riscv_assembler/instr_arr.py

@@ -97,9 +97,8 @@ def immediate(imm, n):
 
 								return mod_imm, mod_imm_2'''
 		mod_imm = format(((1 << 12) - 1) & int(imm), '012b')
-		if n == 1:
-			return mod_imm[0] + mod_imm[12-10 : 12-4]
-		return mod_imm[12-4 : 12 - 0] + mod_imm[1]
+		if n == 1: return mod_imm[0:7]
+		else: return mod_imm[7:12]
 
 
 class _SB(Instruction):
@@ -139,7 +138,6 @@ def __str__(self):
 	def compute_instr(self, instr, imm, rd):
 		instr = super().check_instr_valid(instr, U_instr)
 		opcode = 0
-
 		return "".join([
 			_U.immediate(imm),
 			super().reg(rd),
@@ -148,7 +146,10 @@ def compute_instr(self, instr, imm, rd):
 
 	@staticmethod
 	def immediate(imm):
-		return format(int(imm) >> 12, '013b')
+		# return format(int(imm) >> 12, '013b')
+		high_20 = int(imm)
+		mod_imm = format(((1 << 20) - 1) & high_20, '020b')
+		return mod_imm
 
 class _UJ(Instruction):
 	def __repr__(self):
@@ -219,13 +220,19 @@ def organize(self, tokens):
 		line_num, code = tokens[-2], tokens[-1]
 		instr, rs1, imm, rd = tokens[0], None, None, None
 		if instr == "jalr":
-			if len(tokens) == 4+2:
-				rs1, imm, rd = reg_map[tokens[2]], super().JUMP(tokens[3], line_num, code), reg_map[tokens[1]]
-			else:
-				rs1, imm, rd = reg_map[tokens[1]], 0, reg_map["x1"]
+			if tokens[3][0] =='x':
+				rs1, imm, rd = reg_map[tokens[3]], tokens[2], reg_map[tokens[1]] # use for jalr rd, offset(rs1)
+			else: 
+				rs1, imm, rd = reg_map[tokens[2]], tokens[3], reg_map[tokens[1]] # normal jalr rs1, rs2, offset	
 		elif instr == "lw":
 			rs1, imm, rd = reg_map[tokens[3]], tokens[2], reg_map[tokens[1]]
-		elif instr == 'ld':
+		elif instr == 'lh':
+			rs1, imm, rd = reg_map[tokens[3]], tokens[2], reg_map[tokens[1]]
+		elif instr == 'lb':
+			rs1, imm, rd = reg_map[tokens[3]], tokens[2], reg_map[tokens[1]]
+		elif instr == 'lbu':
+			rs1, imm, rd = reg_map[tokens[3]], tokens[2], reg_map[tokens[1]]
+		elif instr == 'lhu':
 			rs1, imm, rd = reg_map[tokens[3]], tokens[2], reg_map[tokens[1]]
 		else:
 			rs1, imm, rd = reg_map[tokens[2]], tokens[3], reg_map[tokens[1]]
@@ -266,7 +273,8 @@ def __str__(self):
 		return "U Parser"
 
 	def organize(self, tokens):
-		instr, imm, rd = tokens[0], tokens[1], reg_map[tokens[2]]
+		print('in U type:', tokens[0], tokens[1], tokens[2])
+		instr, rd, imm = tokens[0], reg_map[tokens[1]], tokens[2]
 		return U(instr, imm, rd)
 
 class _UJ_parse(InstructionParser):
@@ -280,10 +288,13 @@ def __str__(self):
 	def organize(self, tokens):
 		line_num, code = tokens[-2], tokens[-1]
 		instr, imm, rd = tokens[0], None, None
+		print('see token len', len(tokens), 'and jal get:',tokens[0], tokens[1], tokens[2])
 		if len(tokens) == 3:
+			print('here')
 			imm, rd = super().JUMP(tokens[2], line_num, code), reg_map[tokens[1]]
 		else:
-			imm, rd = super().JUMP(tokens[1], line_num, code), reg_map["x1"]
+			print('there????')
+			imm, rd = super().JUMP(tokens[1], line_num, code), reg_map[tokens[2]]
 
 		return UJ(instr, imm, rd)
 
@@ -366,8 +377,7 @@ def instruction_map():
 	"remu"
 ]
 I_instr = [
-	"addi", "lb", "lw",
-	"ld", "lbu", "lhu",
+	"addi", "lb", "lw", "lh", "lbu", "lhu",
 	"lwu", "fence", "fence.i", 
 	"slli", "slti", "sltiu", 
 	"xori", "slri", "srai",

src/riscv_assembler/parse.py

@@ -58,9 +58,10 @@ def handle_inline_comments(x : str) -> str:
 
 	@staticmethod
 	def handle_specific_instr(x : list) -> list:
-		# for sw, lw, lb, lh, sb, sh
-		if len(x[0]) == 2 and (x[0] in S_instr or x[0] in I_instr):
+		# for sw, lw, lb, lh, sb, sh weird here. It should be rewrited clearer
+		if x[0]=='sw' or x[0]=='sh' or x[0]=='sb' or x[0] == 'lw' or x[0] == 'lh' or x[0] == 'lb' or x[0] == 'lhu' or x[0] == 'lbu' or (x[0]=='jalr' and len(x)==3): # op rd, offset(rs1)
 			y = x[-1].split('('); y[1] = y[1].replace(')','')
+			print('trigger special instr~~~', x[0])
 			return x[0:-1] + y
 		elif 'requires jump' == 5:
 			...
@@ -91,6 +92,7 @@ def interpret_arr(code : list) -> list:
 		for line_num, line in enumerate(code):
 			tokens = _Parser.tokenize(line, line_num, code)
 			int_code += [_Parser.interpret(tokens) for _ in range(1) if len(tokens) != 0]
+			print(int_code)
 
 		return int_code
 
@@ -124,6 +126,7 @@ def determine_type(tk : str) -> function:
 		parsers = [Rp, Ip, Sp, SBp, Up ,UJp, Psp]
 		for i in range(len(instr_sets)):
 			if tk in instr_sets[i]:
+				print('detected instr:', tk)
 				return parsers[i]
 		raise Exception("Bad Instruction Provided: " + tk + "!")
 

tests/assembly/test3.s

@@ -1 +1,7 @@
-sw s0, 0(sp)
+sw s0, 0(sp)
+sw x5, 100(x6)
+sw x5, -300(x6)
+sw x5, -1(x6)
+sw x5, 2047(x6)
+jalr x5, x6, 4
+addi x5, x6, 4

tests/assembly/test8.s

@@ -59,4 +59,4 @@ loop_end:
     lw s4, 16(sp)
     lw s5, 20(sp)
     addi, sp, sp, 24
-    #ret
+    #ret

tests/test_class.py

@@ -34,7 +34,6 @@ def SUITE():
 		results[0] += [cnv(path)] #**
 		# 2.
 		cnv.output_mode = 'p'
-		print('Printing Output')
 		results[1] += [cnv(path)]
 		# 3. 
 		cnv.output_mode = 'f'
@@ -53,7 +52,6 @@ def SUITE():
 		results[4] += [cnv(str_code)] #**
 		# 6.
 		cnv.output_mode = 'p'
-		print('Printing Output')
 		results[5] += [cnv(str_code)]
 		# 7. 
 		cnv.output_mode = 'f'
@@ -82,15 +80,14 @@ def SUITE():
 		results[13] += ['arr{}.bin'.format(i)] #**
 		results[14] += [cnv.output_mode]
 
-
 	return results
 
 ans = ['0xfe810113','0x00812023','0x00912223','0x01212423','0x01312623','0x01412823','0x01512a23','0x00100f13','0x01e64863','0x01e6ca63','0x01e74863','0x0140006f','0x00500593','0xfcdff06f','0x00600593','0xfc5ff06f','0x00000413','0x00000293','0x00000313','0x00000393','0x00400e13','0x03c68a33','0x03c70ab3','0x0040006f','0x02c28263','0x00052483','0x0005a903','0x01450533','0x015585b3','0x032489b3','0x01340433','0x00128293','0xfe1ff06f','0x00040513','0x00012403','0x00412483','0x00812903','0x00c12983','0x01012a03','0x01412a83','0x01810113']
 ANSWERS = {
 	0: [['0x000000b3'], None, 'file0.txt', 'file0.bin', ['0x000000b3'], None, 'str0.txt', 'str0.bin', ['0x000000b3'], 'a', None, 'p', 'arr0.txt', 'arr0.bin', 'f'],
 	1: [['0x02040293'], None, 'file1.txt', 'file1.bin',['0x02040293'], None, 'str1.txt', 'str1.bin',['0x02040293'], 'a', None, 'p', 'arr1.txt', 'arr1.bin', 'f'],
 	2: [['0x00a00413', '0x00a00493', '0x00848263', '0xfe040493'], None, 'file2.txt', 'file2.bin', ['0x00a00413', '0x00a00493', '0x00848263', '0xfe040493'], None, 'str2.txt', 'str2.bin', ['0x00a00413', '0x00a00493', '0x00848263', '0xfe040493'], 'a', None, 'p', 'arr2.txt', 'arr2.bin', 'f'],
-	3: [['0x00812023'], None, 'file3.txt', 'file3.bin', ['0x00812023'], None, 'str3.txt', 'str3.bin', ['0x00812023'], 'a', None, 'p', 'arr3.txt', 'arr3.bin', 'f'],
+	3: [['0x00812023', '0x06532223', '0xec532a23','0xfe532fa3','0x7e532fa3','0x004302e7','0x00430293'], None, 'file3.txt', 'file3.bin', ['0x00812023', '0x06532223', '0xec532a23','0xfe532fa3','0x7e532fa3','0x004302e7','0x00430293'], None, 'str3.txt', 'str3.bin', ['0x00812023', '0x06532223', '0xec532a23','0xfe532fa3', '0x7e532fa3','0x004302e7','0x00430293'], 'a', None, 'p', 'arr3.txt', 'arr3.bin', 'f'],
 	4: [['0x000000b3', '0x02040293', '0x02040293','0x00812023'], None, 'file4.txt', 'file4.bin',['0x000000b3', '0x02040293', '0x02040293','0x00812023'], None, 'str4.txt', 'str4.bin',['0x000000b3', '0x02040293', '0x02040293','0x00812023'], 'a', None, 'p', 'arr4.txt', 'arr4.bin', 'f'],
 	5: [['0x00a00413','0x00a00493','0xfff00493','0x00048463','0xfe000ce3','0xfe040493'], None, 'file5.txt', 'file5.bin', ['0x00a00413','0x00a00493','0xfff00493','0x00048463','0xfe000ce3','0xfe040493'], None, 'str5.txt', 'str5.bin', ['0x00a00413','0x00a00493','0xfff00493','0x00048463','0xfe000ce3','0xfe040493'], 'a', None, 'p', 'arr5.txt', 'arr5.bin', 'f'],
 	6: [['0x00a00093','0xfec00113','0x00000663','0x00123023','0x00023083','0xfff00093','0x00123023','0x00023083','0xfe02c6e3'], None, 'file6.txt', 'file6.bin', ['0x00a00093','0xfec00113','0x00000663','0x00123023','0x00023083','0xfff00093','0x00123023','0x00023083','0xfe02c6e3'], None, 'str6.txt', 'str6.bin', ['0x00a00093','0xfec00113','0x00000663','0x00123023','0x00023083','0xfff00093','0x00123023','0x00023083','0xfe02c6e3'], 'a', None, 'p', 'arr6.txt', 'arr6.bin', 'f'],
@@ -101,12 +98,10 @@ def SUITE():
 
 def error_label(q, test):
 	return "Question {q} Failed for Test {test}".format(q = q, test = test)
-
 test_data = []
 for q in range(num_questions):
 	for t in range(num_test_files):
 		test_data += [(q, t)]
-
 @pytest.mark.parametrize("q, test", test_data)
 def test_compute(q: int, test: int):
 	assert RESULTS[q][test] == ANSWERS[test][q], error_label(q, test)