diff options
Diffstat (limited to 'Debug')
| -rw-r--r-- | Debug/Nuttx.py | 88 | ||||
| -rwxr-xr-x | Debug/poor-mans-profiler.sh | 260 |
2 files changed, 299 insertions, 49 deletions
diff --git a/Debug/Nuttx.py b/Debug/Nuttx.py index cf86dd668..826e12c97 100644 --- a/Debug/Nuttx.py +++ b/Debug/Nuttx.py @@ -1,7 +1,10 @@ # GDB/Python functions for dealing with NuttX +from __future__ import print_function import gdb, gdb.types +parse_int = lambda x: int(str(x), 0) + class NX_register_set(object): """Copy of the registers for a given context""" @@ -155,7 +158,7 @@ class NX_task(object): result = [] for i in range(pidhash_type.range()[0],pidhash_type.range()[1]): entry = pidhash_value[i] - pid = int(entry['pid']) + pid = parse_int(entry['pid']) if pid is not -1: result.append(pid) return result @@ -184,7 +187,7 @@ class NX_task(object): self.__dict__['stack_used'] = 0 else: stack_limit = self._tcb['adj_stack_size'] - for offset in range(0, int(stack_limit)): + for offset in range(0, parse_int(stack_limit)): if stack_base[offset] != 0xff: break self.__dict__['stack_used'] = stack_limit - offset @@ -244,7 +247,7 @@ class NX_task(object): filearray = filelist['fl_files'] result = dict() for i in range(filearray.type.range()[0],filearray.type.range()[1]): - inode = int(filearray[i]['f_inode']) + inode = parse_int(filearray[i]['f_inode']) if inode != 0: result[i] = inode return result @@ -299,7 +302,7 @@ class NX_show_task (gdb.Command): super(NX_show_task, self).__init__("show task", gdb.COMMAND_USER) def invoke(self, arg, from_tty): - t = NX_task.for_pid(int(arg)) + t = NX_task.for_pid(parse_int(arg)) if t is not None: my_fmt = 'PID:{pid} name:{name} state:{state}\n' my_fmt += ' stack used {stack_used} of {stack_limit}\n' @@ -435,12 +438,12 @@ class NX_tcb(object): first_tcb = tcb_ptr.dereference() tcb_list.append(first_tcb); next_tcb = first_tcb['flink'].dereference() - while not self.is_in(int(next_tcb['pid']),[int(t['pid']) for t in tcb_list]): + while not self.is_in(parse_int(next_tcb['pid']),[parse_int(t['pid']) for t in tcb_list]): tcb_list.append(next_tcb); old_tcb = next_tcb; next_tcb = old_tcb['flink'].dereference() - return [t for t in tcb_list if int(t['pid'])<2000] + return [t for t in tcb_list if parse_int(t['pid'])<2000] def getTCB(self): list_of_listsnames = ['g_pendingtasks','g_readytorun','g_waitingforsemaphore','g_waitingforsignal','g_inactivetasks'] @@ -469,12 +472,12 @@ class NX_check_stack_order(gdb.Command): first_tcb = tcb_ptr.dereference() tcb_list.append(first_tcb); next_tcb = first_tcb['flink'].dereference() - while not self.is_in(int(next_tcb['pid']),[int(t['pid']) for t in tcb_list]): + while not self.is_in(parse_int(next_tcb['pid']),[parse_int(t['pid']) for t in tcb_list]): tcb_list.append(next_tcb); old_tcb = next_tcb; next_tcb = old_tcb['flink'].dereference() - return [t for t in tcb_list if int(t['pid'])<2000] + return [t for t in tcb_list if parse_int(t['pid'])<2000] def getTCB(self): list_of_listsnames = ['g_pendingtasks','g_readytorun','g_waitingforsemaphore','g_waitingforsignal','g_inactivetasks'] @@ -488,7 +491,7 @@ class NX_check_stack_order(gdb.Command): def getSPfromTask(self,tcb): regmap = NX_register_set.v7em_regmap a =tcb['xcp']['regs'] - return int(a[regmap['SP']]) + return parse_int(a[regmap['SP']]) def find_closest(self,list,val): tmp_list = [abs(i-val) for i in list] @@ -525,8 +528,8 @@ class NX_check_stack_order(gdb.Command): for t in tcb: p = []; #print(t.name,t._tcb['stack_alloc_ptr']) - p.append(int(t['stack_alloc_ptr'])) - p.append(int(t['adj_stack_ptr'])) + p.append(parse_int(t['stack_alloc_ptr'])) + p.append(parse_int(t['adj_stack_ptr'])) p.append(self.getSPfromTask(t)) stackadresses[str(t['name'])] = p; address = int("0x30000000",0) @@ -594,12 +597,12 @@ class NX_search_tcb(gdb.Command): first_tcb = tcb_ptr.dereference() tcb_list.append(first_tcb); next_tcb = first_tcb['flink'].dereference() - while not self.is_in(int(next_tcb['pid']),[int(t['pid']) for t in tcb_list]): + while not self.is_in(parse_int(next_tcb['pid']),[parse_int(t['pid']) for t in tcb_list]): tcb_list.append(next_tcb); old_tcb = next_tcb; next_tcb = old_tcb['flink'].dereference() - return [t for t in tcb_list if int(t['pid'])<2000] + return [t for t in tcb_list if parse_int(t['pid'])<2000] def invoke(self,args,sth): list_of_listsnames = ['g_pendingtasks','g_readytorun','g_waitingforsemaphore','g_waitingforsignal','g_inactivetasks'] @@ -612,7 +615,7 @@ class NX_search_tcb(gdb.Command): # filter for tasks that are listed twice tasks_filt = {} for t in tasks: - pid = int(t['pid']); + pid = parse_int(t['pid']); if not pid in tasks_filt.keys(): tasks_filt[pid] = t['name']; print('{num_t} Tasks found:'.format(num_t = len(tasks_filt))) @@ -653,62 +656,49 @@ class NX_my_bt(gdb.Command): tcb_ptr = addr_value.cast(gdb.lookup_type('struct tcb_s').pointer()) return tcb_ptr.dereference() - def print_instruction_at(self,addr,stack_percentage): + def resolve_file_line_func(self,addr,stack_percentage): gdb.write(str(round(stack_percentage,2))+":") str_to_eval = "info line *"+hex(addr) #gdb.execute(str_to_eval) res = gdb.execute(str_to_eval,to_string = True) # get information from results string: words = res.split() - valid = False - if words[0] == 'No': - #no line info... - pass - else: - valid = True + if words[0] != 'No': line = int(words[1]) - idx = words[3].rfind("/"); #find first backslash - if idx>0: - name = words[3][idx+1:]; - path = words[3][:idx]; - else: - name = words[3]; - path = ""; block = gdb.block_for_pc(addr) func = block.function if str(func) == "None": func = block.superblock.function - - if valid: - print("Line: ",line," in ",path,"/",name,"in ",func) - return name,path,line,func - - - + return words[3].strip('"'), line, func def invoke(self,args,sth): - addr_dec = int(args[2:],16) - _tcb = self.get_tcb_from_address(addr_dec) + try: + addr_dec = parse_int(args) # Trying to interpret the input as TCB address + except ValueError: + for task in NX_task.tasks(): # Interpreting as a task name + if task.name == args: + _tcb = task._tcb + break + else: + _tcb = self.get_tcb_from_address(addr_dec) + print("found task with PID: ",_tcb["pid"]) - up_stack = int(_tcb['adj_stack_ptr']) - curr_sp = int(_tcb['xcp']['regs'][0]) #curr stack pointer - other_sp = int(_tcb['xcp']['regs'][8]) # other stack pointer - stacksize = int(_tcb['adj_stack_size']) # other stack pointer + up_stack = parse_int(_tcb['adj_stack_ptr']) + curr_sp = parse_int(_tcb['xcp']['regs'][0]) #curr stack pointer + other_sp = parse_int(_tcb['xcp']['regs'][8]) # other stack pointer + stacksize = parse_int(_tcb['adj_stack_size']) # other stack pointer print("tasks current SP = ",hex(curr_sp),"stack max ptr is at ",hex(up_stack)) - if curr_sp == up_stack: - sp = other_sp - else: - sp = curr_sp; - - while(sp < up_stack): + item = 0 + for sp in range(other_sp if curr_sp == up_stack else curr_sp, up_stack, 4): mem = self.readmem(sp) #print(hex(sp)," : ",hex(mem)) if self.is_in_bounds(mem): # this is a potential instruction ptr stack_percentage = (up_stack-sp)/stacksize - name,path,line,func = self.print_instruction_at(mem,stack_percentage) - sp = sp + 4; # jump up one word + filename,line,func = self.resolve_file_line_func(mem, stack_percentage) + print('#%-2d ' % item, '0x%08x in ' % mem, func, ' at ', filename, ':', line, sep='') + item += 1 NX_my_bt() diff --git a/Debug/poor-mans-profiler.sh b/Debug/poor-mans-profiler.sh new file mode 100755 index 000000000..ab06a1b66 --- /dev/null +++ b/Debug/poor-mans-profiler.sh @@ -0,0 +1,260 @@ +#!/bin/bash +# +# Poor man's sampling profiler for NuttX. +# +# Usage: Install flamegraph.pl in your PATH, configure your .gdbinit, run the script with proper arguments and go +# have a coffee. When you're back, you'll see the flamegraph. Note that frequent calls to GDB significantly +# interfere with normal operation of the target, which means that you can't profile real-time tasks with it. +# +# Requirements: ARM GDB with Python support +# + +set -e +root=$(dirname $0)/.. + +function die() +{ + echo "$@" + exit 1 +} + +function usage() +{ + echo "Invalid usage. Supported options:" + cat $0 | sed -n 's/^\s*--\([^)\*]*\).*/\1/p' # Don't try this at home. + exit 1 +} + +which flamegraph.pl > /dev/null || die "Install flamegraph.pl first" + +# +# Parsing the arguments. Read this section for usage info. +# +nsamples=0 +sleeptime=0.1 # Doctors recommend 7-8 hours a day +taskname= +elf=$root/Build/px4fmu-v2_default.build/firmware.elf +append=0 +fgfontsize=10 +fgwidth=1900 + +for i in "$@" +do + case $i in + --nsamples=*) + nsamples="${i#*=}" + ;; + --sleeptime=*) + sleeptime="${i#*=}" + ;; + --taskname=*) + taskname="${i#*=}" + ;; + --elf=*) + elf="${i#*=}" + ;; + --append) + append=1 + ;; + --fgfontsize=*) + fgfontsize="${i#*=}" + ;; + --fgwidth=*) + fgwidth="${i#*=}" + ;; + *) + usage + ;; + esac + shift +done + +# +# Temporary files +# +stacksfile=/tmp/pmpn-stacks.log +foldfile=/tmp/pmpn-folded.txt +graphfile=/tmp/pmpn-flamegraph.svg +gdberrfile=/tmp/pmpn-gdberr.log + +# +# Sampling if requested. Note that if $append is true, the stack file will not be rewritten. +# +cd $root + +if [[ $nsamples > 0 ]] +then + [[ $append = 0 ]] && (rm -f $stacksfile; echo "Old stacks removed") + + echo "Sampling the task '$taskname'..." + + for x in $(seq 1 $nsamples) + do + if [[ "$taskname" = "" ]] + then + arm-none-eabi-gdb $elf --batch -ex "set print asm-demangle on" -ex bt \ + 2> $gdberrfile \ + | sed -n 's/\(#.*\)/\1/p' \ + >> $stacksfile + else + arm-none-eabi-gdb $elf --batch -ex "set print asm-demangle on" \ + -ex "source $root/Debug/Nuttx.py" \ + -ex "show mybt $taskname" \ + 2> $gdberrfile \ + | sed -n 's/0\.0:\(#.*\)/\1/p' \ + >> $stacksfile + fi + echo -e '\n\n' >> $stacksfile + echo -ne "\r$x/$nsamples" + sleep $sleeptime + done + + echo + echo "Stacks saved to $stacksfile" +else + echo "Sampling skipped - set 'nsamples' to re-sample." +fi + +# +# Folding the stacks. +# +[ -f $stacksfile ] || die "Where are the stack samples?" + +cat << 'EOF' > /tmp/pmpn-folder.py +# +# This stack folder correctly handles C++ types. +# +from __future__ import print_function, division +import fileinput, collections, os, sys + +def enforce(x, msg='Invalid input'): + if not x: + raise Exception(msg) + +def split_first_part_with_parens(line): + LBRACES = {'(':'()', '<':'<>', '[':'[]', '{':'{}'} + RBRACES = {')':'()', '>':'<>', ']':'[]', '}':'{}'} + QUOTES = set(['"', "'"]) + quotes = collections.defaultdict(bool) + braces = collections.defaultdict(int) + out = '' + for ch in line: + out += ch + # escape character cancels further processing + if ch == '\\': + continue + # special cases + if out.endswith('operator>') or out.endswith('operator>>') or out.endswith('operator->'): # gotta love c++ + braces['<>'] += 1 + if out.endswith('operator<') or out.endswith('operator<<'): + braces['<>'] -= 1 + # switching quotes + if ch in QUOTES: + quotes[ch] = not quotes[ch] + # counting parens only when outside quotes + if sum(quotes.values()) == 0: + if ch in LBRACES.keys(): + braces[LBRACES[ch]] += 1 + if ch in RBRACES.keys(): + braces[RBRACES[ch]] -= 1 + # sanity check + for v in braces.values(): + enforce(v >= 0, 'Unaligned braces: ' + str(dict(braces))) + # termination condition + if ch == ' ' and sum(braces.values()) == 0: + break + out = out.strip() + return out, line[len(out):] + +def parse(line): + def take_path(line, output): + line = line.strip() + if line.startswith('at '): + line = line[3:].strip() + if line: + output['file_full_path'] = line.rsplit(':', 1)[0].strip() + output['file_base_name'] = os.path.basename(output['file_full_path']) + output['line'] = int(line.rsplit(':', 1)[1]) + return output + + def take_args(line, output): + line = line.lstrip() + if line[0] == '(': + output['args'], line = split_first_part_with_parens(line) + return take_path(line.lstrip(), output) + + def take_function(line, output): + output['function'], line = split_first_part_with_parens(line.lstrip()) + return take_args(line.lstrip(), output) + + def take_mem_loc(line, output): + line = line.lstrip() + if line.startswith('0x'): + end = line.find(' ') + num = line[:end] + output['memloc'] = int(num, 16) + line = line[end:].lstrip() + end = line.find(' ') + enforce(line[:end] == 'in') + line = line[end:].lstrip() + return take_function(line, output) + + def take_frame_num(line, output): + line = line.lstrip() + enforce(line[0] == '#') + end = line.find(' ') + num = line[1:end] + output['frame_num'] = int(num) + return take_mem_loc(line[end:], output) + + return take_frame_num(line, {}) + +stacks = collections.defaultdict(int) +current = '' + +stack_tops = collections.defaultdict(int) +num_stack_frames = 0 + +for idx,line in enumerate(fileinput.input()): + try: + line = line.strip() + if line: + inf = parse(line) + fun = inf['function'] + current = (fun + ';' + current) if current else fun + + if inf['frame_num'] == 0: + num_stack_frames += 1 + stack_tops[fun] += 1 + elif current: + stacks[current] += 1 + current = '' + except Exception, ex: + print('ERROR (line %d):' % (idx + 1), ex, file=sys.stderr) + +for s, f in sorted(stacks.items(), key=lambda (s, f): s): + print(s, f) + +print('Total stack frames:', num_stack_frames, file=sys.stderr) +print('Top consumers (distribution of the stack tops):', file=sys.stderr) +for name,num in sorted(stack_tops.items(), key=lambda (name, num): num, reverse=True)[:10]: + print('% 5.1f%% ' % (100 * num / num_stack_frames), name, file=sys.stderr) +EOF + +cat $stacksfile | python /tmp/pmpn-folder.py > $foldfile + +echo "Folded stacks saved to $foldfile" + +# +# Graphing. +# +cat $foldfile | flamegraph.pl --fontsize=$fgfontsize --width=$fgwidth > $graphfile +echo "FlameGraph saved to $graphfile" + +# On KDE, xdg-open prefers Gwenview by default, which doesn't handle interactive SVGs, so we need a browser. +# The current implementation is hackish and stupid. Somebody, please do something about it. +opener=xdg-open +which firefox > /dev/null && opener=firefox +which google-chrome > /dev/null && opener=google-chrome + +$opener $graphfile |