third_party_littlefs/scripts/readmdir.py
Christopher Haster 5137e4b0ba Last minute tweaks to debug scripts
- Standardized littlefs debug statements to use hex prefixes and
  brackets for printing pairs.

- Removed the entry behavior for readtree and made -t the default.
  This is because 1. the CTZ skip-list parsing was broken, which is not
  surprising, and 2. the entry parsing was more complicated than useful.
  This functionality may be better implemented as a proper filesystem
  read script, complete with directory tree dumping.

- Changed test.py's --gdb argument to take [init, main, assert],
  this matches the names of the stages in C's startup.

- Added printing of tail to all mdir dumps in readtree/readmdir.

- Added a print for if any mdirs are corrupted in readtree.

- Added debug script side-effects to .gitignore.
2020-03-29 21:19:33 -05:00

368 lines
11 KiB
Python
Executable File

#!/usr/bin/env python3
import struct
import binascii
import sys
import itertools as it
TAG_TYPES = {
'splice': (0x700, 0x400),
'create': (0x7ff, 0x401),
'delete': (0x7ff, 0x4ff),
'name': (0x700, 0x000),
'reg': (0x7ff, 0x001),
'dir': (0x7ff, 0x002),
'superblock': (0x7ff, 0x0ff),
'struct': (0x700, 0x200),
'dirstruct': (0x7ff, 0x200),
'ctzstruct': (0x7ff, 0x202),
'inlinestruct': (0x7ff, 0x201),
'userattr': (0x700, 0x300),
'tail': (0x700, 0x600),
'softtail': (0x7ff, 0x600),
'hardtail': (0x7ff, 0x601),
'gstate': (0x700, 0x700),
'movestate': (0x7ff, 0x7ff),
'crc': (0x700, 0x500),
}
class Tag:
def __init__(self, *args):
if len(args) == 1:
self.tag = args[0]
elif len(args) == 3:
if isinstance(args[0], str):
type = TAG_TYPES[args[0]][1]
else:
type = args[0]
if isinstance(args[1], str):
id = int(args[1], 0) if args[1] not in 'x.' else 0x3ff
else:
id = args[1]
if isinstance(args[2], str):
size = int(args[2], str) if args[2] not in 'x.' else 0x3ff
else:
size = args[2]
self.tag = (type << 20) | (id << 10) | size
else:
assert False
@property
def isvalid(self):
return not bool(self.tag & 0x80000000)
@property
def isattr(self):
return not bool(self.tag & 0x40000000)
@property
def iscompactable(self):
return bool(self.tag & 0x20000000)
@property
def isunique(self):
return not bool(self.tag & 0x10000000)
@property
def type(self):
return (self.tag & 0x7ff00000) >> 20
@property
def type1(self):
return (self.tag & 0x70000000) >> 20
@property
def type3(self):
return (self.tag & 0x7ff00000) >> 20
@property
def id(self):
return (self.tag & 0x000ffc00) >> 10
@property
def size(self):
return (self.tag & 0x000003ff) >> 0
@property
def dsize(self):
return 4 + (self.size if self.size != 0x3ff else 0)
@property
def chunk(self):
return self.type & 0xff
@property
def schunk(self):
return struct.unpack('b', struct.pack('B', self.chunk))[0]
def is_(self, type):
return (self.type & TAG_TYPES[type][0]) == TAG_TYPES[type][1]
def mkmask(self):
return Tag(
0x700 if self.isunique else 0x7ff,
0x3ff if self.isattr else 0,
0)
def chid(self, nid):
ntag = Tag(self.type, nid, self.size)
if hasattr(self, 'off'): ntag.off = self.off
if hasattr(self, 'data'): ntag.data = self.data
if hasattr(self, 'crc'): ntag.crc = self.crc
return ntag
def typerepr(self):
if self.is_('crc') and getattr(self, 'crc', 0xffffffff) != 0xffffffff:
return 'crc (bad)'
reverse_types = {v: k for k, v in TAG_TYPES.items()}
for prefix in range(12):
mask = 0x7ff & ~((1 << prefix)-1)
if (mask, self.type & mask) in reverse_types:
type = reverse_types[mask, self.type & mask]
if prefix > 0:
return '%s %#0*x' % (
type, prefix//4, self.type & ((1 << prefix)-1))
else:
return type
else:
return '%02x' % self.type
def idrepr(self):
return repr(self.id) if self.id != 0x3ff else '.'
def sizerepr(self):
return repr(self.size) if self.size != 0x3ff else 'x'
def __repr__(self):
return 'Tag(%r, %d, %d)' % (self.typerepr(), self.id, self.size)
def __lt__(self, other):
return (self.id, self.type) < (other.id, other.type)
def __bool__(self):
return self.isvalid
def __int__(self):
return self.tag
def __index__(self):
return self.tag
class MetadataPair:
def __init__(self, blocks):
if len(blocks) > 1:
self.pair = [MetadataPair([block]) for block in blocks]
self.pair = sorted(self.pair, reverse=True)
self.data = self.pair[0].data
self.rev = self.pair[0].rev
self.tags = self.pair[0].tags
self.ids = self.pair[0].ids
self.log = self.pair[0].log
self.all_ = self.pair[0].all_
return
self.pair = [self]
self.data = blocks[0]
block = self.data
self.rev, = struct.unpack('<I', block[0:4])
crc = binascii.crc32(block[0:4])
# parse tags
corrupt = False
tag = Tag(0xffffffff)
off = 4
self.log = []
self.all_ = []
while len(block) - off >= 4:
ntag, = struct.unpack('>I', block[off:off+4])
tag = Tag(int(tag) ^ ntag)
tag.off = off + 4
tag.data = block[off+4:off+tag.dsize]
if tag.is_('crc'):
crc = binascii.crc32(block[off:off+4+4], crc)
else:
crc = binascii.crc32(block[off:off+tag.dsize], crc)
tag.crc = crc
off += tag.dsize
self.all_.append(tag)
if tag.is_('crc'):
# is valid commit?
if crc != 0xffffffff:
corrupt = True
if not corrupt:
self.log = self.all_.copy()
# reset tag parsing
crc = 0
tag = Tag(int(tag) ^ ((tag.type & 1) << 31))
# find active ids
self.ids = list(it.takewhile(
lambda id: Tag('name', id, 0) in self,
it.count()))
# find most recent tags
self.tags = []
for tag in self.log:
if tag.is_('crc') or tag.is_('splice'):
continue
elif tag.id == 0x3ff:
if tag in self and self[tag] is tag:
self.tags.append(tag)
else:
# id could have change, I know this is messy and slow
# but it works
for id in self.ids:
ntag = tag.chid(id)
if ntag in self and self[ntag] is tag:
self.tags.append(ntag)
self.tags = sorted(self.tags)
def __bool__(self):
return bool(self.log)
def __lt__(self, other):
# corrupt blocks don't count
if not self or not other:
return bool(other)
# use sequence arithmetic to avoid overflow
return not ((other.rev - self.rev) & 0x80000000)
def __contains__(self, args):
try:
self[args]
return True
except KeyError:
return False
def __getitem__(self, args):
if isinstance(args, tuple):
gmask, gtag = args
else:
gmask, gtag = args.mkmask(), args
gdiff = 0
for tag in reversed(self.log):
if (gmask.id != 0 and tag.is_('splice') and
tag.id <= gtag.id - gdiff):
if tag.is_('create') and tag.id == gtag.id - gdiff:
# creation point
break
gdiff += tag.schunk
if ((int(gmask) & int(tag)) ==
(int(gmask) & int(gtag.chid(gtag.id - gdiff)))):
if tag.size == 0x3ff:
# deleted
break
return tag
raise KeyError(gmask, gtag)
def _dump_tags(self, tags, f=sys.stdout, truncate=True):
f.write("%-8s %-8s %-13s %4s %4s" % (
'off', 'tag', 'type', 'id', 'len'))
if truncate:
f.write(' data (truncated)')
f.write('\n')
for tag in tags:
f.write("%08x: %08x %-13s %4s %4s" % (
tag.off, tag,
tag.typerepr(), tag.idrepr(), tag.sizerepr()))
if truncate:
f.write(" %-23s %-8s\n" % (
' '.join('%02x' % c for c in tag.data[:8]),
''.join(c if c >= ' ' and c <= '~' else '.'
for c in map(chr, tag.data[:8]))))
else:
f.write("\n")
for i in range(0, len(tag.data), 16):
f.write(" %08x: %-47s %-16s\n" % (
tag.off+i,
' '.join('%02x' % c for c in tag.data[i:i+16]),
''.join(c if c >= ' ' and c <= '~' else '.'
for c in map(chr, tag.data[i:i+16]))))
def dump_tags(self, f=sys.stdout, truncate=True):
self._dump_tags(self.tags, f=f, truncate=truncate)
def dump_log(self, f=sys.stdout, truncate=True):
self._dump_tags(self.log, f=f, truncate=truncate)
def dump_all(self, f=sys.stdout, truncate=True):
self._dump_tags(self.all_, f=f, truncate=truncate)
def main(args):
blocks = []
with open(args.disk, 'rb') as f:
for block in [args.block1, args.block2]:
if block is None:
continue
f.seek(block * args.block_size)
blocks.append(f.read(args.block_size)
.ljust(args.block_size, b'\xff'))
# find most recent pair
mdir = MetadataPair(blocks)
try:
mdir.tail = mdir[Tag('tail', 0, 0)]
if mdir.tail.size != 8 or mdir.tail.data == 8*b'\xff':
mdir.tail = None
except KeyError:
mdir.tail = None
print("mdir {%s} rev %d%s%s%s" % (
', '.join('%#x' % b
for b in [args.block1, args.block2]
if b is not None),
mdir.rev,
' (was %s)' % ', '.join('%d' % m.rev for m in mdir.pair[1:])
if len(mdir.pair) > 1 else '',
' (corrupted!)' if not mdir else '',
' -> {%#x, %#x}' % struct.unpack('<II', mdir.tail.data)
if mdir.tail else ''))
if args.all:
mdir.dump_all(truncate=not args.no_truncate)
elif args.log:
mdir.dump_log(truncate=not args.no_truncate)
else:
mdir.dump_tags(truncate=not args.no_truncate)
return 0 if mdir else 1
if __name__ == "__main__":
import argparse
import sys
parser = argparse.ArgumentParser(
description="Dump useful info about metadata pairs in littlefs.")
parser.add_argument('disk',
help="File representing the block device.")
parser.add_argument('block_size', type=lambda x: int(x, 0),
help="Size of a block in bytes.")
parser.add_argument('block1', type=lambda x: int(x, 0),
help="First block address for finding the metadata pair.")
parser.add_argument('block2', nargs='?', type=lambda x: int(x, 0),
help="Second block address for finding the metadata pair.")
parser.add_argument('-l', '--log', action='store_true',
help="Show tags in log.")
parser.add_argument('-a', '--all', action='store_true',
help="Show all tags in log, included tags in corrupted commits.")
parser.add_argument('-T', '--no-truncate', action='store_true',
help="Don't truncate large amounts of data.")
sys.exit(main(parser.parse_args()))