#!/usr/bin/python3 # # Find holes in structures, so that we can pack them and improve our memory density. # # In order to make this work, you need to # (1) Be operating in a workspace where you have a __NON-DEBUG__ build of LibreOffice, but __WITH SYMBOLS__. # (A debug build has different sizes for some things in the standard library.) # (2) First run the unusedfields loplugin to generate a log file # (3) Install the pahole stuff into your gdb, I used this one: # https://github.com/PhilArmstrong/pahole-gdb # (4) Edit the loop near the top of the script to only produce results for one of our modules. # Note that this will make GDB soak up about 8G of RAM, which is why I don't do more than one module at a time # (5) Run the script # ./compilerplugins/clang/pahole-all-classes.py > ./compilerplugins/clang/pahole.results # import _thread import io import os import subprocess import time import re # search for all the class names in the file produced by the unusedfields loplugin #a = subprocess.Popen("grep 'definition:' workdir/loplugin.unusedfields.log | sort -u", stdout=subprocess.PIPE, shell=True) a = subprocess.Popen("cat n1", stdout=subprocess.PIPE, shell=True) classSet = set() classSourceLocDict = dict() with a.stdout as txt: for line in txt: tokens = line.decode('utf8').strip().split("\t") className = tokens[2].strip() srcLoc = tokens[5].strip() # ignore things like unions if "anonymous" in className: continue # ignore duplicates if className in classSet: continue classSet.add(className) classSourceLocDict[className] = srcLoc a.terminate() # Some of the pahole commands are going to fail, and I cannot read the error stream and the input stream # together because python has no way of (easily) doing a non-blocking read. # So I have to write the commands out using a background thread, and then read the entire resulting # stream out below. def write_pahole_commands(classes): for className in classes: stdin.write("echo " + className + " " + classSourceLocDict[className] + "\n") stdin.write("pahole " + className + "\n") stdin.flush() stdin.write("echo all-done\n") stdin.flush() stdin.close() # only way to make it flush the last echo command # Use generator because lines often end up merged together in gdb's output, and we need # to split them up, and that creates a mess in the parsing logic. def read_generator(gdbOutput): while True: line = gdbOutput.readline().decode('utf8').strip() for split in line.split("(gdb)"): split = split.strip() if len(split) == 0: continue if "all-done" in split: return yield split classList = sorted(classSet) # Process 200 classes at a time, otherwise gdb's memory usage blows up and kills the machine # while len(classList) > 0: currClassList = classList[1:200]; classList = classList[200:] gdbProc = subprocess.Popen("gdb", stdin=subprocess.PIPE, stdout=subprocess.PIPE, stderr=subprocess.STDOUT, shell=True) stdin = io.TextIOWrapper(gdbProc.stdin, 'utf-8') # make gdb load all the debugging info stdin.write("set confirm off\n") for filename in sorted(os.listdir('instdir/program')): if filename.endswith(".so"): stdin.write("add-symbol-file instdir/program/" + filename + "\n") stdin.flush() _thread.start_new_thread( write_pahole_commands, (currClassList,) ) firstLineRegex = re.compile("/\*\s+(\d+)\s+\*/ struct") fieldLineRegex = re.compile("/\*\s+(\d+)\s+(\d+)\s+\*/ ") holeLineRegex = re.compile("/\* XXX (\d+) bit hole, try to pack \*/") # sometimes pahole can't determine the size of a sub-struct, and then it returns bad data bogusLineRegex = re.compile("/\*\s+\d+\s+0\s+\*/") structLines = list() foundHole = False cumulativeHoleBits = 0 structSize = 0 foundBogusLine = False # pahole doesn't report space at the end of the structure, so work it out myself sizeOfFields = 0 for line in read_generator(gdbProc.stdout): structLines.append(line) firstLineMatch = firstLineRegex.match(line) if firstLineMatch: structSize = int(firstLineMatch.group(1)) holeLineMatch = holeLineRegex.match(line) if holeLineMatch: foundHole = True cumulativeHoleBits += int(holeLineMatch.group(1)) fieldLineMatch = fieldLineRegex.match(line) if fieldLineMatch: fieldSize = int(fieldLineMatch.group(2)) sizeOfFields = int(fieldLineMatch.group(1)) + fieldSize if bogusLineRegex.match(line): foundBogusLine = True if line == "}": # Ignore very large structs, packing those is not going to help much, and # re-organising them can make them much less readable. if foundHole and len(structLines) < 12 and structSize < 100 and not foundBogusLine: # Verify that we have enough hole-space that removing it will result in a structure # that still satisfies alignment requirements, otherwise the compiler will just put empty # space at the end of the struct. # TODO improve detection of the required alignment for a structure potentialSpace = (cumulativeHoleBits / 8) + (sizeOfFields - structSize) if potentialSpace >= 8: for line in structLines: print(line) if (sizeOfFields - structSize) > 0: print("hole at end of struct: " + str(sizeOfFields - structSize)) # reset state structLines.clear() foundHole = False cumulativeHoleBits = 0 structSize = 0 foundBogusLine = False actualStructSize = 0 gdbProc.terminate()