ncv/discmp.py

   1 #!/usr/bin/python
   2 # Copyright 2008, Google Inc.
   3 # All rights reserved.
   4 #
   5 # Redistribution and use in source and binary forms, with or without
   6 # modification, are permitted provided that the following conditions are
   7 # met:
   8 #
   9 #     * Redistributions of source code must retain the above copyright
  10 # notice, this list of conditions and the following disclaimer.
  11 #     * Redistributions in binary form must reproduce the above
  12 # copyright notice, this list of conditions and the following disclaimer
  13 # in the documentation and/or other materials provided with the
  14 # distribution.
  15 #     * Neither the name of Google Inc. nor the names of its
  16 # contributors may be used to endorse or promote products derived from
  17 # this software without specific prior written permission.
  18 #
  19 # THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
  20 # "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
  21 # LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
  22 # A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
  23 # OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
  24 # SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
  25 # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
  26 # DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
  27 # THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
  28 # (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
  29 # OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  30
  31 #
  32 """discmp.py - compare disassembler output
  33
  34 This script compares output from the Native Client decoder to output
  35 from GNU objdump to make sure they agree on instruction lengths. It
  36 returns a zero status on success and non-zero on failure.
  37
  38 Usage:
  39   discmp.py PLATFORM_BASE=<platform_base> <fileordir> <fileordir>
  40 for example:
  41   discmp.py PLATFORM_BASE=/home/who/proj/nacl/googleclient/native_client/scons-out/dbg-linux ncv/testdata/hanoi
  42
  43 """
  44
  45 import re
  46 import sys
  47 import os
  48
  49 def Usage():
  50   print("""
  51 USAGE:
  52   %s PLATFORM_BASE=<platform_base> <fileordir> <fileordir>
  53
  54 For example:
  55   %s PLATFORM_BASE=/home/who/proj/nacl/googleclient/native_client/scons-out/dbg-linux ncv/testdata/hanoi
  56
  57 Commonly this installer will be invoked from hammer as:
  58   ./hammer decoder_test
  59 In this case hammer will take care of the command line args.
  60 """ % (sys.argv[0], sys.argv[0]))
  61   sys.exit(-1)
  62
  63 def FindBinaries(PlatformBase):
  64   ncdis = os.path.join(PlatformBase, "staging", "ncdis")
  65   # Someday if nacl-objdump gets more permissive we can use it
  66   # objdump = os.path.join(nacl.SDK_BASE, "bin", "nacl-objdump")
  67   objdump = "/usr/bin/objdump"
  68   if not os.path.isfile(ncdis):
  69     print "Error: could not find", ncdis
  70     sys.exit(-1)
  71   if not os.path.isfile(objdump):
  72     print "Error: could not find", objdump
  73     sys.exit(-1)
  74   return ncdis, objdump
  75
  76 class DisFile(object):
  77   """Processes variants of disassembler output from various sources,
  78    providing a list of instruction lengths, based on addresses."""
  79
  80   def __init__(self, stream):
  81     self.fd = stream
  82     self.lastline = None
  83     self.lastaddr = None
  84     self.thisaddr = None
  85     self.thisline = None
  86     # note we ignore lines that have only continuation bytes;
  87     # no opcode
  88     self.dislinefmt = re.compile("\s*([0-9a-f]+):\t(.*\t+.*)")
  89     self.data16fmt = re.compile("\s*([0-9a-f]+):\s+66\s+data16")
  90     self.waitfmt = re.compile("\s*([0-9a-f]+):\s+9b\s+[f]?wait")
  91
  92   def NextDisInst(self):
  93     while (True):
  94       line = self.fd.readline()
  95       if line == "": return 0, None
  96       match = self.data16fmt.match(line)
  97       if (match):
  98         addr = match.group(1)
  99         line = self.fd.readline()
 100         match = self.dislinefmt.match(line)
 101         return (int(addr, 16),
 102                 " " + addr + ":\t60 " + match.group(2) + "\n")
 103       match = self.waitfmt.match(line)
 104       if (match):
 105         addr = match.group(1)
 106         line = self.fd.readline()
 107         match = self.dislinefmt.match(line)
 108         return (int(addr, 16),
 109                 " " + addr + ":\t9b " + match.group(2) + "\n")
 110       match = self.dislinefmt.match(line)
 111       if (match):
 112         return int(match.group(1), 16), line
 113
 114   def NextInst(self):
 115     if self.lastaddr is None:
 116       self.lastaddr, self.lastline = self.NextDisInst()
 117     else:
 118       self.lastaddr = self.thisaddr
 119       self.lastline = self.thisline
 120     self.thisaddr, self.thisline = self.NextDisInst()
 121     if self.thisline is None:
 122       # don't know how long the last line was, so just return 1
 123       return (1, self.lastline)
 124     if self.lastline is None:
 125       return 0, None
 126     else:
 127       return (self.thisaddr - self.lastaddr, self.lastline)
 128
 129 def IsElfBinary(fname):
 130   fd = open(fname)
 131   iselfbinary = fd.readline().startswith("\x7fELF")
 132   fd.close()
 133   return iselfbinary
 134
 135 def DoOneFile(fname, ncdis, objdump):
 136   if not IsElfBinary(fname):
 137     print("Error:", fname, "is not an ELF binary\nContinuing...")
 138     return
 139   df1 = DisFile(os.popen(objdump + " -dz " + fname))
 140   df2 = DisFile(os.popen(ncdis + " " + fname))
 141   instcount = 0
 142   while (1):
 143     instcount += 1
 144     len1, line1 = df1.NextInst()
 145     len2, line2 = df2.NextInst()
 146     if line1 is None: break
 147     if line2 is None: break
 148     if (len1 != len2):
 149       sys.stdout.write("ERROR: inst length mistmatch %d != %d\n" %
 150                        (len1, len2))
 151       sys.stdout.write(line1)
 152       sys.stdout.write(line2)
 153       sys.exit(-1)
 154   if line1 or line2:
 155     sys.stdout.write("ERROR: disasm output is different lengths\n")
 156     sys.exit(-1)
 157   sys.stdout.write("%s: %d instructions; 0 errors!\n"
 158                    % (fname, instcount))
 159
 160 def DoOneFileOrDir(name, ncdis, objdump):
 161   if os.path.isfile(name): DoOneFile(name, ncdis, objdump)
 162   elif os.path.isdir(name):
 163     for dirpath, dirlist, filelist in os.walk(name):
 164       for fname in filelist:
 165         DoOneFile(os.path.join(dirpath, fname), ncdis, objdump)
 166   else: print "invalid argument", name
 167
 168 def ParseArgv():
 169   files = []
 170   args = {}
 171   print "***ARGV:"
 172   for arg in sys.argv[1:]:
 173     print "***   ", arg
 174     if arg.find("=") > 0:
 175       name, value = arg.split("=")
 176       print("%s=%s" % (name, value))
 177       args[name] = value
 178     else:
 179       files.append(arg)
 180   # require MODE and PLATFORM_BASE to be set
 181   if "PLATFORM_BASE" not in args: Usage()
 182   return args, files
 183
 184 def main():
 185   args, files = ParseArgv()
 186   ncdis, objdump = FindBinaries(args["PLATFORM_BASE"])
 187   for name in files:
 188     DoOneFileOrDir(name, ncdis, objdump)
 189   sys.exit(0)
 190
 191 if '__main__' == __name__:
 192   main()