Files
2026-07-13 13:04:25 +08:00

449 lines
19 KiB
Python

#!/usr/bin/env python3
# SPDX-FileCopyrightText: 2023-2026 Espressif Systems (Shanghai) CO LTD
# SPDX-License-Identifier: Apache-2.0
from nvs_logger import NVS_Logger
from nvs_parser import NVS_Entry
from nvs_parser import NVS_Page
from nvs_parser import NVS_Partition
from nvs_parser import nvs_const
EMPTY_ENTRY = NVS_Entry(-1, bytearray(32), 'Erased')
used_namespaces: dict[int, str | None] = {}
found_namespaces: dict[int, str] = {}
blobs: dict = {}
blob_chunks: list[NVS_Entry] = []
def check_partition_size(nvs_partition: NVS_Partition, nvs_log: NVS_Logger, read_only: bool = False) -> bool:
"""Checks if the partition is large enough and has enough pages"""
if len(nvs_partition.raw_data) // 0x1000 < 3 and not read_only:
nvs_log.warn('NVS Partition size must be at least 0x3000 (4kiB * 3 pages == 12kiB)!')
return False
if len(nvs_partition.raw_data) % 0x1000 != 0:
nvs_log.warn('NVS Partition size must be a multiple of 0x1000 (4kiB)!')
return False
if len(nvs_partition.pages) < 3 and not read_only:
nvs_log.warn('NVS Partition must contain 3 pages (sectors) at least to function properly!')
return False
return True
def check_empty_page_present(nvs_partition: NVS_Partition, nvs_log: NVS_Logger) -> bool:
if not any(page.header['status'] == 'Empty' for page in nvs_partition.pages):
nvs_log.err(
'No free (empty) page found in the NVS partition, at least one free page is required for proper function!'
)
nvs_log.err('NVS partition possibly truncated?')
return False
return True
def check_empty_page_content(nvs_page: NVS_Page, nvs_log: NVS_Logger) -> bool:
result = True
nvs_log.info(nvs_log.cyan(f'Page {nvs_page.header["status"]}'))
if nvs_page.raw_entry_state_bitmap != bytearray({0xFF}) * nvs_const.entry_size:
result = False
nvs_log.err('The page is reported as Empty but its entry state bitmap is not empty!')
if any([not e.is_empty for e in nvs_page.entries]):
result = False
nvs_log.err('The page is reported as Empty but there are data written!')
return result
def check_page_crc(nvs_page: NVS_Page, nvs_log: NVS_Logger) -> bool:
if nvs_page.header['crc']['original'] == nvs_page.header['crc']['computed']:
nvs_log.info(nvs_log.cyan(f'Page no. {nvs_page.header["page_index"]}'), '\tCRC32: OK')
return True
else:
nvs_log.info(
nvs_log.cyan(f'Page no. {nvs_page.header["page_index"]}'),
'Original CRC32:',
nvs_log.red(f'{nvs_page.header["crc"]["original"]:x}'),
'Generated CRC32:',
nvs_log.green(f'{nvs_page.header["crc"]["computed"]:x}'),
)
return False
def identify_entry_duplicates(entry: NVS_Entry, entry_dict: dict[str, list[NVS_Entry]]) -> dict[str, list[NVS_Entry]]:
"""Identifies and logs written entries
Part 1 of duplicate entry check mechanism
"""
if entry.state == 'Written':
if entry.key in entry_dict:
entry_dict[entry.key].append(entry)
else:
entry_dict[entry.key] = [entry]
return entry_dict
def check_page_entries(nvs_page: NVS_Page, nvs_log: NVS_Logger) -> dict[str, list[NVS_Entry]]:
"""Checks entries in the given page (state, CRC32, type, span; gathers blobs and namespaces)."""
seen_written_entires: dict[str, list[NVS_Entry]] = {}
for entry in nvs_page.entries:
# entry: NVS_Entry
entry.page = nvs_page
# Entries stored in 'page.entries' are primitive data types, blob indexes or string/blob data
# Variable length values themselves occupy whole 32 bytes (therefore metadata values are meaningless)
# and are stored in as entries inside string/blob data entry 'entry.children' list
# Duplicate entry check (1) - same key, different index - find duplicates
seen_written_entires = identify_entry_duplicates(entry, seen_written_entires)
# Entry state check - skip variable-length payload bytes (metadata not meaningful there)
if entry.is_empty:
if entry.state == 'Written':
nvs_log.err(f' Entry #{entry.index:03d} is reported as Written but it is empty!')
continue
elif entry.state == 'Erased':
nvs_log.warn(
f' Entry #{entry.index:03d} is reported as Erased but it is empty! '
'(Only entries reported as Empty should be empty)'
)
if entry.state == 'Written':
# Entry CRC32 check
if entry.metadata['crc']['original'] != entry.metadata['crc']['computed']:
nvs_log.info(
nvs_log.red(f' Entry #{entry.index:03d} {entry.key} has wrong CRC32!'),
'Written:',
nvs_log.red(f'{entry.metadata["crc"]["original"]:x}'),
'Generated:',
nvs_log.green(f'{entry.metadata["crc"]["computed"]:x}'),
)
# Entry children CRC32 check
if entry.metadata['span'] > 1 and (
entry.metadata['crc']['data_original'] != entry.metadata['crc']['data_computed']
):
nvs_log.info(
nvs_log.red(f' Entry #{entry.index:03d} {entry.key} data (string, blob) has wrong CRC32!'),
'Written:',
nvs_log.red(f'{entry.metadata["crc"]["data_original"]:x}'),
'Generated:',
nvs_log.green(f'{entry.metadata["crc"]["data_computed"]:x}'),
)
# Entry type check
if entry.metadata['type'] not in [nvs_const.item_type[key] for key in nvs_const.item_type]:
nvs_log.warn(
f' Type of entry #{entry.index:03d} {entry.key} is unrecognized! Type: {entry.metadata["type"]}'
)
# Span check
if entry.index + entry.metadata['span'] - 1 >= int(nvs_const.page_size / nvs_const.entry_size) - 2:
nvs_log.err(f' Variable length entry #{entry.index:03d} {entry.key} is out of bounds!')
# Spanned entry state checks
elif entry.metadata['span'] > 1:
parent_state = entry.state
for kid in entry.children:
if parent_state != kid.state:
nvs_log.warn(
f' Inconsistent data state! Entry #{entry.index:03d} {entry.key} '
f'state: {parent_state}, Data entry #{kid.index:03d} {entry.key} '
f'state: {kid.state}'
)
# Gather blobs & namespaces
if entry.metadata['type'] == 'blob_index':
blobs[f'{entry.metadata["namespace"]:03d}{entry.key}'] = [entry] + [EMPTY_ENTRY] * entry.data[
'chunk_count'
]
elif entry.metadata['type'] == 'blob_data':
blob_chunks.append(entry)
if entry.metadata['namespace'] == 0:
found_namespaces[entry.data['value']] = entry.key
else:
used_namespaces[entry.metadata['namespace']] = None
return seen_written_entires
def filter_namespaces_fake_duplicates(duplicate_entries_dict: dict[str, list[NVS_Entry]]) -> dict[str, list[NVS_Entry]]:
"""Takes a dictionary of entries (as written) and returns a new dictionary with "fake" duplicates,
where entries which have the same key but under different namespaces are filtered out
Use `filter_entry_duplicates()` to properly filter out all duplicates
"""
new_duplicate_entries_dict: dict[str, list[NVS_Entry]] = {}
for key, duplicate_entries in duplicate_entries_dict.items():
seen_entries: list[NVS_Entry] = []
entry_same_namespace_collisions_list: set[NVS_Entry] = set()
# Search through the "duplicates" and see if there are real duplicates
# E.g. the key can be the same if the namespace is different
for entry in duplicate_entries:
if entry.metadata['type'] in nvs_const.item_type.values():
entry_same_namespace_collisions = set()
for other_entry in seen_entries:
if entry.metadata['namespace'] == other_entry.metadata['namespace']:
entry_same_namespace_collisions.add(entry)
entry_same_namespace_collisions.add(other_entry)
if len(entry_same_namespace_collisions) != 0:
entry_same_namespace_collisions_list.update(entry_same_namespace_collisions)
seen_entries.append(entry)
# Catch real duplicates
new_duplicate_entries: list[NVS_Entry] = []
if len(seen_entries) > 1:
for entry in seen_entries:
if entry in entry_same_namespace_collisions_list:
new_duplicate_entries.append(entry)
if len(new_duplicate_entries) > 0:
new_duplicate_entries_dict[key] = new_duplicate_entries
return new_duplicate_entries_dict
def filter_blob_related_duplicates(duplicate_entries_dict: dict[str, list[NVS_Entry]]) -> dict[str, list[NVS_Entry]]:
"""Takes a dictionary of entries (as written) and returns a new dictionary with "fake" duplicates,
where entries related to blob index and blob data under the same namespace are filtered out
Use `filter_entry_duplicates()` to properly filter out all duplicates
"""
new_duplicate_entries_dict: dict[str, list[NVS_Entry]] = {}
for key, duplicate_entries in duplicate_entries_dict.items():
seen_blob_index: list[NVS_Entry] = []
seen_blob_data: list[NVS_Entry] = []
seen_another_type_data: list[NVS_Entry] = []
blob_index_chunk_index_collisions_list: set[NVS_Entry] = set()
blob_data_chunk_index_collisions_list: set[NVS_Entry] = set()
# Search through the "duplicates" and see if there are real duplicates
# E.g. the key can be the same for blob_index and blob_data
# (and even for more blob_data entries if they have a different chunk_index)
for entry in duplicate_entries:
if entry.metadata['type'] == 'blob_index':
blob_index_chunk_index_collisions = set()
for other_entry in seen_blob_index:
if entry.metadata['namespace'] == other_entry.metadata['namespace']:
blob_index_chunk_index_collisions.add(entry)
blob_index_chunk_index_collisions.add(other_entry)
if len(blob_index_chunk_index_collisions) != 0:
blob_index_chunk_index_collisions_list.update(blob_index_chunk_index_collisions)
seen_blob_index.append(entry)
elif entry.metadata['type'] == 'blob_data':
blob_data_chunk_index_collisions = set()
for other_entry in seen_blob_data:
if (
entry.metadata['namespace'] == other_entry.metadata['namespace']
and entry.metadata['chunk_index'] == other_entry.metadata['chunk_index']
):
blob_data_chunk_index_collisions.add(entry)
blob_data_chunk_index_collisions.add(other_entry)
if len(blob_data_chunk_index_collisions) != 0:
blob_data_chunk_index_collisions_list.update(blob_data_chunk_index_collisions)
seen_blob_data.append(entry)
else:
seen_another_type_data.append(entry)
# Catch real duplicates
new_duplicate_entries: list[NVS_Entry] = []
if len(seen_blob_index) > 1:
for entry in seen_blob_index:
if entry in blob_index_chunk_index_collisions_list:
new_duplicate_entries.append(entry)
if len(seen_blob_data) > 1:
for entry in seen_blob_data:
if entry in blob_data_chunk_index_collisions_list:
new_duplicate_entries.append(entry)
for entry in seen_another_type_data: # If there are any duplicates of other types
new_duplicate_entries.append(entry)
if len(new_duplicate_entries) > 0:
new_duplicate_entries_dict[key] = new_duplicate_entries
return new_duplicate_entries_dict
def filter_entry_duplicates(entries: dict[str, list[NVS_Entry]]) -> dict[str, list[NVS_Entry]]:
"""Filter fake duplicates; keep real duplicates only (part 2 of duplicate check).
Allowed: same key in different namespaces; blob_index and blob_data sharing a key.
"""
# Only keep seen written entries which have been observerd multiple times (duplicates)
duplicate_entries_list = {key: v for key, v in entries.items() if len(v) > 1}
# Filter out "fake" duplicates 1 (duplicate keys under different namespaces are allowed)
duplicate_entries_list_1 = filter_namespaces_fake_duplicates(duplicate_entries_list)
# Filter out "fake" duplicates 2 (blob index/data may share a key in one namespace)
duplicate_entries_list_2 = filter_blob_related_duplicates(duplicate_entries_list_1)
return duplicate_entries_list_2
def print_entry_duplicates(duplicate_entries_list: dict[str, list[NVS_Entry]], nvs_log: NVS_Logger) -> None:
if len(duplicate_entries_list) > 0:
nvs_log.err('Found duplicate entries:')
nvs_log.err('Entry\tKey\t\t\tType\t\tNamespace idx\tPage\tPage status')
for _, duplicate_entries in duplicate_entries_list.items():
# duplicate_entries: List[NVS_Entry]
for entry in duplicate_entries:
# entry: NVS_Entry
if entry.metadata['namespace'] == 0:
entry_type = f'namespace ({entry.data["value"]})'
else:
entry_type = entry.metadata['type']
if entry.page is not None:
page_num = entry.page.header['page_index']
page_status = entry.page.header['status']
else:
page_num = 'Unknown'
page_status = 'Unknown'
entry_key_tab_cnt = len(entry.key) // 8
entry_key_tab = '\t' * (3 - entry_key_tab_cnt)
namespace_tab_cnt = len(entry_type) // 8
namepace_tab = '\t' * (2 - namespace_tab_cnt)
namespace_str = f'{entry.metadata["namespace"]}'
nvs_log.err(
f'#{entry.index:03d}\t{entry.key}{entry_key_tab}{entry_type}{namepace_tab}'
f'{namespace_str}\t\t{page_num}\t{page_status}'
)
def assemble_blobs(nvs_log: NVS_Logger) -> None:
"""Assembles blob data from blob chunks"""
for chunk in blob_chunks:
# chunk: NVS_Entry
parent = blobs.get(f'{chunk.metadata["namespace"]:03d}{chunk.key}', [EMPTY_ENTRY])[0]
# Blob chunk without blob index check
if parent is EMPTY_ENTRY:
nvs_log.err(
f'Blob {chunk.key} chunk has no blob index! '
f'Namespace index: {chunk.metadata["namespace"]:03d} '
f'[{found_namespaces.get(chunk.metadata["namespace"], "undefined")}], '
f'Chunk Index: {chunk.metadata["chunk_index"]:03d}'
)
else:
blob_key = f'{chunk.metadata["namespace"]:03d}{chunk.key}'
chunk_index = chunk.metadata['chunk_index'] - parent.data['chunk_start']
blobs[blob_key][chunk_index + 1] = chunk
def check_blob_data(nvs_log: NVS_Logger) -> None:
"""Checks blob data for missing chunks or data"""
for blob_key in blobs:
blob_index = blobs[blob_key][0]
blob_chunks = blobs[blob_key][1:]
blob_size = blob_index.data['size']
for i, chunk in enumerate(blob_chunks):
# chunk: NVS_Entry
# Blob missing chunk check
if chunk is EMPTY_ENTRY:
nvs_log.err(
f'Blob {blob_index.key} is missing a chunk! '
f'Namespace index: {blob_index.metadata["namespace"]:03d} '
f'[{found_namespaces.get(blob_index.metadata["namespace"], "undefined")}], '
f'Chunk Index: {i:03d}'
)
else:
blob_size -= len(chunk.children) * nvs_const.entry_size
# Blob missing data check
if blob_size > 0:
nvs_log.err(
f'Blob {blob_index.key} is missing {blob_size} B of data! '
f'Namespace index: {blob_index.metadata["namespace"]:03d}'
)
def check_blobs(nvs_log: NVS_Logger) -> None:
# Assemble blobs
assemble_blobs(nvs_log)
# Blob data check
check_blob_data(nvs_log)
def check_namespaces(nvs_log: NVS_Logger) -> None:
"""Checks namespaces (entries using undefined namespace indexes, unused namespaces)"""
# Undefined namespace index check
for used_ns in used_namespaces:
key = found_namespaces.pop(used_ns, None)
if key is None:
nvs_log.err(f'Undefined namespace index! Namespace index: {used_ns:03d} [undefined]')
# Unused namespace index check
for unused_ns in found_namespaces:
nvs_log.warn(f'Found unused namespace. Namespace index: {unused_ns:03d} [{found_namespaces[unused_ns]}]')
def reset_global_variables() -> None:
"""Global variables need to be cleared out before calling `integrity_check()` multiple times from a script
(e.g. when running tests) to avoid incorrect output
"""
global used_namespaces, found_namespaces, blobs, blob_chunks
used_namespaces = {}
found_namespaces = {}
blobs = {}
blob_chunks = []
def integrity_check(nvs_partition: NVS_Partition, nvs_log: NVS_Logger) -> None:
"""Function for multi-stage integrity check of a NVS partition"""
# Partition size check
check_partition_size(nvs_partition, nvs_log)
# Free/empty page check
check_empty_page_present(nvs_partition, nvs_log)
seen_written_entires_all: dict[str, list[NVS_Entry]] = {}
# Loop through all pages in the partition
for page in nvs_partition.pages:
# page: NVS_Page
# Print a page header
if page.header['status'] == 'Empty':
# Check if a page is truly empty
check_empty_page_content(page, nvs_log)
else:
# Check a page header CRC32
check_page_crc(page, nvs_log)
# Check all entries in a page
seen_written_entires = check_page_entries(page, nvs_log)
# Collect all seen written entries
for key in seen_written_entires:
if key in seen_written_entires_all:
seen_written_entires_all[key].extend(seen_written_entires[key])
else:
seen_written_entires_all[key] = seen_written_entires[key]
# Duplicate entry check (2) - same key, different index
duplicates = filter_entry_duplicates(seen_written_entires_all)
# Print duplicate entries
print_entry_duplicates(duplicates, nvs_log)
nvs_log.info() # Empty line
# Blob checks
check_blobs(nvs_log)
# Namespace checks
check_namespaces(nvs_log)
reset_global_variables()