## ## This file is part of the libsigrokdecode project. ## ## Copyright (C) 2011-2014 Uwe Hermann ## ## 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, see . ## import sigrokdecode as srd from math import floor, ceil ''' OUTPUT_PYTHON format: Packet: [, , ] This is the list of s and their respective values: - 'STARTBIT': The data is the (integer) value of the start bit (0/1). - 'DATA': This is always a tuple containing two items: - 1st item: the (integer) value of the UART data. Valid values range from 0 to 511 (as the data can be up to 9 bits in size). - 2nd item: the list of individual data bits and their ss/es numbers. - 'PARITYBIT': The data is the (integer) value of the parity bit (0/1). - 'STOPBIT': The data is the (integer) value of the stop bit (0 or 1). - 'INVALID STARTBIT': The data is the (integer) value of the start bit (0/1). - 'INVALID STOPBIT': The data is the (integer) value of the stop bit (0/1). - 'PARITY ERROR': The data is a tuple with two entries. The first one is the expected parity value, the second is the actual parity value. - TODO: Frame error? The field is 0 for RX packets, 1 for TX packets. ''' # Used for differentiating between the two data directions. RX = 0 TX = 1 # Given a parity type to check (odd, even, zero, one), the value of the # parity bit, the value of the data, and the length of the data (5-9 bits, # usually 8 bits) return True if the parity is correct, False otherwise. # 'none' is _not_ allowed as value for 'parity_type'. def parity_ok(parity_type, parity_bit, data, num_data_bits): # Handle easy cases first (parity bit is always 1 or 0). if parity_type == 'zero': return parity_bit == 0 elif parity_type == 'one': return parity_bit == 1 # Count number of 1 (high) bits in the data (and the parity bit itself!). ones = bin(data).count('1') + parity_bit # Check for odd/even parity. if parity_type == 'odd': return (ones % 2) == 1 elif parity_type == 'even': return (ones % 2) == 0 class SamplerateError(Exception): pass class ChannelError(Exception): pass class Decoder(srd.Decoder): api_version = 3 id = 'uart' name = 'UART' longname = 'Universal Asynchronous Receiver/Transmitter' desc = 'Asynchronous, serial bus.' license = 'gplv2+' inputs = ['logic'] outputs = ['uart'] optional_channels = ( # Allow specifying only one of the signals, e.g. if only one data # direction exists (or is relevant). {'id': 'rx', 'name': 'RX', 'desc': 'UART receive line'}, {'id': 'tx', 'name': 'TX', 'desc': 'UART transmit line'}, ) options = ( {'id': 'baudrate', 'desc': 'Baud rate', 'default': 115200}, {'id': 'num_data_bits', 'desc': 'Data bits', 'default': 8, 'values': (5, 6, 7, 8, 9)}, {'id': 'parity_type', 'desc': 'Parity type', 'default': 'none', 'values': ('none', 'odd', 'even', 'zero', 'one')}, {'id': 'parity_check', 'desc': 'Check parity?', 'default': 'yes', 'values': ('yes', 'no')}, {'id': 'num_stop_bits', 'desc': 'Stop bits', 'default': 1.0, 'values': (0.0, 0.5, 1.0, 1.5)}, {'id': 'bit_order', 'desc': 'Bit order', 'default': 'lsb-first', 'values': ('lsb-first', 'msb-first')}, {'id': 'format', 'desc': 'Data format', 'default': 'hex', 'values': ('ascii', 'dec', 'hex', 'oct', 'bin')}, {'id': 'invert_rx', 'desc': 'Invert RX?', 'default': 'no', 'values': ('yes', 'no')}, {'id': 'invert_tx', 'desc': 'Invert TX?', 'default': 'no', 'values': ('yes', 'no')}, ) annotations = ( ('rx-data', 'RX data'), ('tx-data', 'TX data'), ('rx-start', 'RX start bits'), ('tx-start', 'TX start bits'), ('rx-parity-ok', 'RX parity OK bits'), ('tx-parity-ok', 'TX parity OK bits'), ('rx-parity-err', 'RX parity error bits'), ('tx-parity-err', 'TX parity error bits'), ('rx-stop', 'RX stop bits'), ('tx-stop', 'TX stop bits'), ('rx-warnings', 'RX warnings'), ('tx-warnings', 'TX warnings'), ('rx-data-bits', 'RX data bits'), ('tx-data-bits', 'TX data bits'), ) annotation_rows = ( ('rx-data', 'RX', (0, 2, 4, 6, 8)), ('rx-data-bits', 'RX bits', (12,)), ('rx-warnings', 'RX warnings', (10,)), ('tx-data', 'TX', (1, 3, 5, 7, 9)), ('tx-data-bits', 'TX bits', (13,)), ('tx-warnings', 'TX warnings', (11,)), ) binary = ( ('rx', 'RX dump'), ('tx', 'TX dump'), ('rxtx', 'RX/TX dump'), ) idle_state = ['WAIT FOR START BIT', 'WAIT FOR START BIT'] def putx(self, rxtx, data): s, halfbit = self.startsample[rxtx], self.bit_width / 2.0 self.put(s - floor(halfbit), self.samplenum + ceil(halfbit), self.out_ann, data) def putpx(self, rxtx, data): s, halfbit = self.startsample[rxtx], self.bit_width / 2.0 self.put(s - floor(halfbit), self.samplenum + ceil(halfbit), self.out_python, data) def putg(self, data): s, halfbit = self.samplenum, self.bit_width / 2.0 self.put(s - floor(halfbit), s + ceil(halfbit), self.out_ann, data) def putp(self, data): s, halfbit = self.samplenum, self.bit_width / 2.0 self.put(s - floor(halfbit), s + ceil(halfbit), self.out_python, data) def putbin(self, rxtx, data): s, halfbit = self.startsample[rxtx], self.bit_width / 2.0 self.put(s - floor(halfbit), self.samplenum + ceil(halfbit), self.out_binary, data) def __init__(self): self.samplerate = None self.samplenum = 0 self.frame_start = [-1, -1] self.startbit = [-1, -1] self.cur_data_bit = [0, 0] self.datavalue = [0, 0] self.paritybit = [-1, -1] self.stopbit1 = [-1, -1] self.startsample = [-1, -1] self.state = ['WAIT FOR START BIT', 'WAIT FOR START BIT'] self.databits = [[], []] def start(self): self.out_python = self.register(srd.OUTPUT_PYTHON) self.out_binary = self.register(srd.OUTPUT_BINARY) self.out_ann = self.register(srd.OUTPUT_ANN) self.bw = (self.options['num_data_bits'] + 7) // 8 def metadata(self, key, value): if key == srd.SRD_CONF_SAMPLERATE: self.samplerate = value # The width of one UART bit in number of samples. self.bit_width = float(self.samplerate) / float(self.options['baudrate']) def get_sample_point(self, rxtx, bitnum): """Determine absolute sample number of a bit slot's sample point.""" # bitpos is the samplenumber which is in the middle of the # specified UART bit (0 = start bit, 1..x = data, x+1 = parity bit # (if used) or the first stop bit, and so on). # The samples within bit are 0, 1, ..., (bit_width - 1), therefore # index of the middle sample within bit window is (bit_width - 1) / 2. bitpos = self.frame_start[rxtx] + (self.bit_width - 1) / 2.0 bitpos += bitnum * self.bit_width return bitpos # Return true if we reached the middle of the desired bit, false otherwise. def reached_bit(self, rxtx, bitnum): bitpos = self.get_sample_point(rxtx, bitnum) if self.samplenum >= bitpos: return True return False def wait_for_start_bit(self, rxtx, signal): # The caller already has detected an edge. Strictly speaking this # check on the current signal level is redundant. But it does not # harm either. if signal != 0: return # Save the sample number where the start bit begins. self.frame_start[rxtx] = self.samplenum self.state[rxtx] = 'GET START BIT' def get_start_bit(self, rxtx, signal): # Skip samples until we're in the middle of the start bit. if not self.reached_bit(rxtx, 0): return self.startbit[rxtx] = signal # The startbit must be 0. If not, we report an error and wait # for the next start bit (assuming this one was spurious). if self.startbit[rxtx] != 0: self.putp(['INVALID STARTBIT', rxtx, self.startbit[rxtx]]) self.putg([rxtx + 10, ['Frame error', 'Frame err', 'FE']]) self.state[rxtx] = 'WAIT FOR START BIT' return self.cur_data_bit[rxtx] = 0 self.datavalue[rxtx] = 0 self.startsample[rxtx] = -1 self.state[rxtx] = 'GET DATA BITS' self.putp(['STARTBIT', rxtx, self.startbit[rxtx]]) self.putg([rxtx + 2, ['Start bit', 'Start', 'S']]) def get_data_bits(self, rxtx, signal): # Skip samples until we're in the middle of the desired data bit. if not self.reached_bit(rxtx, 1 + self.cur_data_bit[rxtx]): return # Save the sample number of the middle of the first data bit. if self.startsample[rxtx] == -1: self.startsample[rxtx] = self.samplenum # Get the next data bit in LSB-first or MSB-first fashion. if self.options['bit_order'] == 'lsb-first': self.datavalue[rxtx] >>= 1 self.datavalue[rxtx] |= \ (signal << (self.options['num_data_bits'] - 1)) else: self.datavalue[rxtx] <<= 1 self.datavalue[rxtx] |= (signal << 0) self.putg([rxtx + 12, ['%d' % signal]]) # Store individual data bits and their start/end samplenumbers. s, halfbit = self.samplenum, int(self.bit_width / 2) self.databits[rxtx].append([signal, s - halfbit, s + halfbit]) # Return here, unless we already received all data bits. self.cur_data_bit[rxtx] += 1 if self.cur_data_bit[rxtx] < self.options['num_data_bits']: return # Skip to either reception of the parity bit, or reception of # the STOP bits if parity is not applicable. self.state[rxtx] = 'GET PARITY BIT' if self.options['parity_type'] == 'none': self.state[rxtx] = 'GET STOP BITS' self.putpx(rxtx, ['DATA', rxtx, (self.datavalue[rxtx], self.databits[rxtx])]) b = self.datavalue[rxtx] formatted = self.format_value(b) if formatted is not None: self.putx(rxtx, [rxtx, [formatted]]) bdata = b.to_bytes(self.bw, byteorder='big') self.putbin(rxtx, [rxtx, bdata]) self.putbin(rxtx, [2, bdata]) self.databits[rxtx] = [] def format_value(self, v): # Format value 'v' according to configured options. # Reflects the user selected kind of representation, as well as # the number of data bits in the UART frames. fmt, bits = self.options['format'], self.options['num_data_bits'] # Assume "is printable" for values from 32 to including 126, # below 32 is "control" and thus not printable, above 127 is # "not ASCII" in its strict sense, 127 (DEL) is not printable, # fall back to hex representation for non-printables. if fmt == 'ascii': if v in range(32, 126 + 1): return chr(v) hexfmt = "[{:02X}]" if bits <= 8 else "[{:03X}]" return hexfmt.format(v) # Mere number to text conversion without prefix and padding # for the "decimal" output format. if fmt == 'dec': return "{:d}".format(v) # Padding with leading zeroes for hex/oct/bin formats, but # without a prefix for density -- since the format is user # specified, there is no ambiguity. if fmt == 'hex': digits = (bits + 4 - 1) // 4 fmtchar = "X" elif fmt == 'oct': digits = (bits + 3 - 1) // 3 fmtchar = "o" elif fmt == 'bin': digits = bits fmtchar = "b" else: fmtchar = None if fmtchar is not None: fmt = "{{:0{:d}{:s}}}".format(digits, fmtchar) return fmt.format(v) return None def get_parity_bit(self, rxtx, signal): # Skip samples until we're in the middle of the parity bit. if not self.reached_bit(rxtx, 1 + self.options['num_data_bits']): return self.paritybit[rxtx] = signal self.state[rxtx] = 'GET STOP BITS' if parity_ok(self.options['parity_type'], self.paritybit[rxtx], self.datavalue[rxtx], self.options['num_data_bits']): self.putp(['PARITYBIT', rxtx, self.paritybit[rxtx]]) self.putg([rxtx + 4, ['Parity bit', 'Parity', 'P']]) else: # TODO: Return expected/actual parity values. self.putp(['PARITY ERROR', rxtx, (0, 1)]) # FIXME: Dummy tuple... self.putg([rxtx + 6, ['Parity error', 'Parity err', 'PE']]) # TODO: Currently only supports 1 stop bit. def get_stop_bits(self, rxtx, signal): # Skip samples until we're in the middle of the stop bit(s). skip_parity = 0 if self.options['parity_type'] == 'none' else 1 b = 1 + self.options['num_data_bits'] + skip_parity if not self.reached_bit(rxtx, b): return self.stopbit1[rxtx] = signal # Stop bits must be 1. If not, we report an error. if self.stopbit1[rxtx] != 1: self.putp(['INVALID STOPBIT', rxtx, self.stopbit1[rxtx]]) self.putg([rxtx + 10, ['Frame error', 'Frame err', 'FE']]) # TODO: Abort? Ignore the frame? Other? self.state[rxtx] = 'WAIT FOR START BIT' self.putp(['STOPBIT', rxtx, self.stopbit1[rxtx]]) self.putg([rxtx + 4, ['Stop bit', 'Stop', 'T']]) def get_wait_cond(self, rxtx, inv): """ Determine Decoder.wait() condition for specified UART line. Returns condititions that are suitable for Decoder.wait(). Those conditions either match the falling edge of the START bit, or the sample point of the next bit time. """ state = self.state[rxtx] if state == 'WAIT FOR START BIT': return {rxtx: 'r' if inv else 'f'} if state == 'GET START BIT': bitnum = 0 elif state == 'GET DATA BITS': bitnum = 1 + self.cur_data_bit[rxtx] elif state == 'GET PARITY BIT': bitnum = 1 + self.options['num_data_bits'] elif state == 'GET STOP BITS': bitnum = 1 + self.options['num_data_bits'] bitnum += 0 if self.options['parity_type'] == 'none' else 1 want_num = self.get_sample_point(rxtx, bitnum) # want_num = int(want_num + 0.5) want_num = ceil(want_num) cond = {'skip': want_num - self.samplenum} return cond def decode(self): if not self.samplerate: raise SamplerateError('Cannot decode without samplerate.') has_pin = [self.has_channel(ch) for ch in (RX, TX)] if has_pin == [False, False]: raise ChannelError('Either TX or RX (or both) pins required.') opt = self.options inv = [opt['invert_rx'] == 'yes', opt['invert_tx'] == 'yes'] while True: conds = [] if has_pin[RX]: conds.append(self.get_wait_cond(RX, inv[RX])) if has_pin[TX]: conds.append(self.get_wait_cond(TX, inv[TX])) (rx, tx) = self.wait(conds) if inv[RX]: rx = not rx if inv[TX]: tx = not tx # State machine. for rxtx in (RX, TX): # Don't try to handle RX (or TX) if not supplied. if not has_pin[rxtx]: continue signal = rx if (rxtx == RX) else tx if self.state[rxtx] == 'WAIT FOR START BIT': self.wait_for_start_bit(rxtx, signal) elif self.state[rxtx] == 'GET START BIT': self.get_start_bit(rxtx, signal) elif self.state[rxtx] == 'GET DATA BITS': self.get_data_bits(rxtx, signal) elif self.state[rxtx] == 'GET PARITY BIT': self.get_parity_bit(rxtx, signal) elif self.state[rxtx] == 'GET STOP BITS': self.get_stop_bits(rxtx, signal)