Bananymous
/
banan-os
1
1
Fork 1
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Kernel: Rewrite a lot of ext2 code 

This commit consists of multiple big changes

1. blocks for inodes are now allocated on demand
  - reading from non allocated block will just return zeroes
  - writing to non allocated block allocates it

2. code doesn't really use raw pointers anymore
  - all casts to uint32_t or structures are now replaced with
    spans. either as<T> or as_span<T> which both are bounds
	checked

3. code doesn't depend on random macros for accessing indirect blocks
  - i added some recursive functions which take care of this :)
This commit is contained in:
Bananymous 2023-10-28 22:12:33 +03:00
parent d98f84f9d3
commit 76f48f095c
2 changed files with 169 additions and 219 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
kernel/include/kernel/FS/Ext2/Inode.h
View File

@ -44,7 +44,12 @@ namespace Kernel
virtual BAN::ErrorOr<void> chmod_impl(mode_t) override;
private:
uint32_t fs_block_of_data_block_index(uint32_t data_block_index);
// Returns maximum number of data blocks in use
// NOTE: the inode might have more blocks than what this suggests if it has been shrinked
uint32_t max_used_data_block_count() const { return size() / blksize(); }
BAN::Optional<uint32_t> block_from_indirect_block(uint32_t block, uint32_t index, uint32_t depth);
BAN::Optional<uint32_t> fs_block_of_data_block_index(uint32_t data_block_index);
BAN::ErrorOr<void> link_inode_to_directory(Ext2Inode&, BAN::StringView name);
BAN::ErrorOr<bool> is_directory_empty();
@ -53,7 +58,8 @@ namespace Kernel
BAN::ErrorOr<void> cleanup_default_links();
void cleanup_from_fs();
BAN::ErrorOr<uint32_t> allocate_new_block();
BAN::ErrorOr<uint32_t> allocate_new_block_to_indirect_block(uint32_t& block, uint32_t index, uint32_t depth);
BAN::ErrorOr<uint32_t> allocate_new_block(uint32_t data_block_index);
void sync();
uint32_t block_group() const;

378
kernel/kernel/FS/Ext2/Inode.cpp
View File

@ -32,12 +32,9 @@ namespace Kernel
auto block_buffer = fs.get_block_buffer();
fs.read_block(inode_location.block, block_buffer);
auto& inode = *(Ext2::Inode*)(block_buffer.data() + inode_location.offset);
auto& inode = block_buffer.span().slice(inode_location.offset).as<Ext2::Inode>();
Ext2Inode* result_ptr = new Ext2Inode(fs, inode, inode_ino);
if (result_ptr == nullptr)
return BAN::Error::from_errno(ENOMEM);
auto result = BAN::RefPtr<Ext2Inode>::adopt(result_ptr);
auto result = TRY(BAN::RefPtr<Ext2Inode>::create(fs, inode, inode_ino));
TRY(fs.inode_cache().insert(inode_ino, result));
return result;
}
@ -48,57 +45,56 @@ namespace Kernel
cleanup_from_fs();
}
#define VERIFY_AND_READ_BLOCK(expr) do { const uint32_t block_index = expr; ASSERT(block_index); m_fs.read_block(block_index, block_buffer); } while (false)
#define VERIFY_AND_RETURN(expr) ({ const uint32_t result = expr; ASSERT(result); return result; })
uint32_t Ext2Inode::fs_block_of_data_block_index(uint32_t data_block_index)
BAN::Optional<uint32_t> Ext2Inode::block_from_indirect_block(uint32_t block, uint32_t index, uint32_t depth)
{
ASSERT(data_block_index < blocks());
if (block == 0)
return {};
ASSERT(depth >= 1);
auto block_buffer = m_fs.get_block_buffer();
m_fs.read_block(block, block_buffer);
const uint32_t indices_per_block = blksize() / sizeof(uint32_t);
// Direct block
if (data_block_index < 12)
VERIFY_AND_RETURN(m_inode.block[data_block_index]);
uint32_t divisor = 1;
for (uint32_t i = 1; i < depth; i++)
divisor *= indices_per_block;
const uint32_t next_block = block_buffer.span().as_span<uint32_t>()[(index / divisor) % indices_per_block];
if (next_block == 0)
return {};
if (depth == 1)
return next_block;
return block_from_indirect_block(next_block, index, depth - 1);
}
BAN::Optional<uint32_t> Ext2Inode::fs_block_of_data_block_index(uint32_t data_block_index)
{
const uint32_t indices_per_block = blksize() / sizeof(uint32_t);
if (data_block_index < 12)
{
if (m_inode.block[data_block_index] == 0)
return {};
return m_inode.block[data_block_index];
}
data_block_index -= 12;
auto block_buffer = m_fs.get_block_buffer();
// Singly indirect block
if (data_block_index < indices_per_block)
{
VERIFY_AND_READ_BLOCK(m_inode.block[12]);
VERIFY_AND_RETURN(((uint32_t*)block_buffer.data())[data_block_index]);
}
return block_from_indirect_block(m_inode.block[12], data_block_index, 1);
data_block_index -= indices_per_block;
// Doubly indirect blocks
if (data_block_index < indices_per_block * indices_per_block)
{
VERIFY_AND_READ_BLOCK(m_inode.block[13]);
VERIFY_AND_READ_BLOCK(((uint32_t*)block_buffer.data())[data_block_index / indices_per_block]);
VERIFY_AND_RETURN(((uint32_t*)block_buffer.data())[data_block_index % indices_per_block]);
}
return block_from_indirect_block(m_inode.block[13], data_block_index, 2);
data_block_index -= indices_per_block * indices_per_block;
// Triply indirect blocks
if (data_block_index < indices_per_block * indices_per_block * indices_per_block)
{
VERIFY_AND_READ_BLOCK(m_inode.block[14]);
VERIFY_AND_READ_BLOCK(((uint32_t*)block_buffer.data())[data_block_index / (indices_per_block * indices_per_block)]);
VERIFY_AND_READ_BLOCK(((uint32_t*)block_buffer.data())[(data_block_index / indices_per_block) % indices_per_block]);
VERIFY_AND_RETURN(((uint32_t*)block_buffer.data())[data_block_index % indices_per_block]);
}
return block_from_indirect_block(m_inode.block[14], data_block_index, 3);
ASSERT_NOT_REACHED();
}
#undef VERIFY_AND_READ_BLOCK
#undef VERIFY_AND_RETURN
BAN::ErrorOr<BAN::String> Ext2Inode::link_target_impl()
{
ASSERT(mode().iflnk());
@ -119,7 +115,7 @@ namespace Kernel
if (offset >= m_inode.size)
return 0;
uint32_t count = buffer.size();
if (offset + buffer.size() > m_inode.size)
count = m_inode.size - offset;
@ -135,8 +131,11 @@ namespace Kernel
for (uint32_t data_block_index = first_block; data_block_index < last_block; data_block_index++)
{
uint32_t block_index = fs_block_of_data_block_index(data_block_index);
m_fs.read_block(block_index, block_buffer);
auto block_index = fs_block_of_data_block_index(data_block_index);
if (block_index.has_value())
m_fs.read_block(block_index.value(), block_buffer);
else
memset(block_buffer.data(), 0x00, block_buffer.size());
uint32_t copy_offset = (offset + n_read) % block_size;
uint32_t to_copy = BAN::Math::min<uint32_t>(block_size - copy_offset, count - n_read);
@ -171,14 +170,20 @@ namespace Kernel
// Write partial block
if (offset % block_size)
{
uint32_t block_index = fs_block_of_data_block_index(offset / block_size);
uint32_t block_offset = offset % block_size;
auto block_index = fs_block_of_data_block_index(offset / block_size);
if (block_index.has_value())
m_fs.read_block(block_index.value(), block_buffer);
else
{
block_index = TRY(allocate_new_block(offset / block_size));;
memset(block_buffer.data(), 0x00, block_buffer.size());
}
uint32_t block_offset = offset % block_size;
uint32_t to_copy = BAN::Math::min<uint32_t>(block_size - block_offset, to_write);
m_fs.read_block(block_index, block_buffer);
memcpy(block_buffer.data() + block_offset, buffer.data(), to_copy);
m_fs.write_block(block_index, block_buffer);
m_fs.write_block(block_index.value(), block_buffer);
written += to_copy;
offset += to_copy;
@ -187,10 +192,12 @@ namespace Kernel
while (to_write >= block_size)
{
uint32_t block_index = fs_block_of_data_block_index(offset / block_size);
auto block_index = fs_block_of_data_block_index(offset / block_size);
if (!block_index.has_value())
block_index = TRY(allocate_new_block(offset / block_size));
memcpy(block_buffer.data(), buffer.data() + written, block_buffer.size());
m_fs.write_block(block_index, block_buffer);
m_fs.write_block(block_index.value(), block_buffer);
written += block_size;
offset += block_size;
@ -199,11 +206,17 @@ namespace Kernel
if (to_write > 0)
{
uint32_t block_index = fs_block_of_data_block_index(offset / block_size);
auto block_index = fs_block_of_data_block_index(offset / block_size);
if (block_index.has_value())
m_fs.read_block(block_index.value(), block_buffer);
else
{
block_index = TRY(allocate_new_block(offset / block_size));
memset(block_buffer.data(), 0x00, block_buffer.size());
}
m_fs.read_block(block_index, block_buffer);
memcpy(block_buffer.data(), buffer.data() + written, to_write);
m_fs.write_block(block_index, block_buffer);
m_fs.write_block(block_index.value(), block_buffer);
}
return buffer.size();
@ -214,34 +227,7 @@ namespace Kernel
if (m_inode.size == new_size)
return {};
const uint32_t block_size = blksize();
const uint32_t current_data_blocks = blocks();
const uint32_t needed_data_blocks = BAN::Math::div_round_up<uint32_t>(new_size, block_size);
if (new_size < m_inode.size)
{
m_inode.size = new_size;
sync();
return {};
}
auto block_buffer = m_fs.get_block_buffer();
if (uint32_t rem = m_inode.size % block_size)
{
uint32_t last_block_index = fs_block_of_data_block_index(current_data_blocks - 1);
m_fs.read_block(last_block_index, block_buffer);
memset(block_buffer.data() + rem, 0, block_size - rem);
m_fs.write_block(last_block_index, block_buffer);
}
memset(block_buffer.data(), 0, block_size);
while (blocks() < needed_data_blocks)
{
uint32_t block_index = TRY(allocate_new_block());
m_fs.write_block(block_index, block_buffer);
}
// TODO: we should remove unused blocks on shrink
m_inode.size = new_size;
sync();
@ -275,9 +261,10 @@ namespace Kernel
const uint32_t ids_per_block = blksize() / sizeof(uint32_t);
for (uint32_t i = 0; i < ids_per_block; i++)
{
const uint32_t idx = ((uint32_t*)block_buffer.data())[i];
if (idx > 0)
cleanup_indirect_block(idx, depth - 1);
const uint32_t next_block = block_buffer.span().as_span<uint32_t>()[i];
if (next_block == 0)
continue;
cleanup_indirect_block(next_block, depth - 1);
}
m_fs.release_block(block);
@ -299,14 +286,14 @@ namespace Kernel
cleanup_indirect_block(m_inode.block[13], 2);
if (m_inode.block[14])
cleanup_indirect_block(m_inode.block[14], 3);
// mark blocks as deleted
memset(m_inode.block, 0x00, sizeof(m_inode.block));
// FIXME: this is only required since fs does not get
// deleting inode from its cache
sync();
m_fs.delete_inode(ino());
}
@ -322,8 +309,8 @@ namespace Kernel
return {};
}
const uint32_t block_size = blksize();
const uint32_t block_index = fs_block_of_data_block_index(offset);
// FIXME: can we actually assume directories have all their blocks allocated
const uint32_t block_index = fs_block_of_data_block_index(offset).value();
auto block_buffer = m_fs.get_block_buffer();
@ -331,16 +318,15 @@ namespace Kernel
// First determine if we have big enough list
{
const uint8_t* block_buffer_end = block_buffer.data() + block_size;
const uint8_t* entry_addr = block_buffer.data();
BAN::ConstByteSpan entry_span = block_buffer.span();
size_t needed_size = sizeof(DirectoryEntryList);
while (entry_addr < block_buffer_end)
while (entry_span.size() >= sizeof(Ext2::LinkedDirectoryEntry))
{
auto& entry = *(Ext2::LinkedDirectoryEntry*)entry_addr;
auto& entry = entry_span.as<const Ext2::LinkedDirectoryEntry>();
if (entry.inode)
needed_size += sizeof(DirectoryEntry) + entry.name_len + 1;
entry_addr += entry.rec_len;
entry_span = entry_span.slice(entry.rec_len);
}
if (needed_size > list_size)
@ -352,11 +338,10 @@ namespace Kernel
DirectoryEntry* ptr = list->array;
list->entry_count = 0;
const uint8_t* block_buffer_end = block_buffer.data() + block_size;
const uint8_t* entry_addr = block_buffer.data();
while (entry_addr < block_buffer_end)
BAN::ConstByteSpan entry_span = block_buffer.span();
while (entry_span.size() >= sizeof(Ext2::LinkedDirectoryEntry))
{
auto& entry = *(Ext2::LinkedDirectoryEntry*)entry_addr;
auto& entry = entry_span.as<const Ext2::LinkedDirectoryEntry>();
if (entry.inode)
{
ptr->dirent.d_ino = entry.inode;
@ -368,7 +353,7 @@ namespace Kernel
ptr = ptr->next();
list->entry_count++;
}
entry_addr += entry.rec_len;
entry_span = entry_span.slice(entry.rec_len);
}
}
@ -460,7 +445,7 @@ namespace Kernel
TRY(inode->link_inode_to_directory(*inode, "."sv));
TRY(inode->link_inode_to_directory(*this, ".."sv));
TRY(link_inode_to_directory(*inode, name));
cleanup.disable();
@ -505,7 +490,7 @@ namespace Kernel
: typed_mode.iflnk() ? Ext2::Enum::SYMLINK
: 0;
auto& new_entry = *(Ext2::LinkedDirectoryEntry*)(block_buffer.data() + entry_offset);
auto& new_entry = block_buffer.span().slice(entry_offset).as<Ext2::LinkedDirectoryEntry>();
new_entry.inode = inode.ino();
new_entry.rec_len = entry_rec_len;
new_entry.name_len = name.size();
@ -523,17 +508,18 @@ namespace Kernel
if (auto rem = needed_entry_len % 4)
needed_entry_len += 4 - rem;
const uint32_t data_block_count = blocks();
// FIXME: can we actually assume directories have all their blocks allocated
const uint32_t data_block_count = max_used_data_block_count();
if (data_block_count == 0)
goto needs_new_block;
// Try to insert inode to last data block
block_index = fs_block_of_data_block_index(data_block_count - 1);
block_index = fs_block_of_data_block_index(data_block_count - 1).value();
m_fs.read_block(block_index, block_buffer);
while (entry_offset < block_size)
{
auto& entry = *(Ext2::LinkedDirectoryEntry*)(block_buffer.data() + entry_offset);
auto& entry = block_buffer.span().slice(entry_offset).as<Ext2::LinkedDirectoryEntry>();
uint32_t entry_min_rec_len = sizeof(Ext2::LinkedDirectoryEntry) + entry.name_len;
if (auto rem = entry_min_rec_len % 4)
@ -559,9 +545,10 @@ namespace Kernel
}
needs_new_block:
block_index = TRY(allocate_new_block());
block_index = TRY(allocate_new_block(data_block_count));
m_inode.size += blksize();
m_fs.read_block(block_index, block_buffer);
memset(block_buffer.data(), 0x00, block_buffer.size());
write_inode(0, block_size);
m_fs.write_block(block_index, block_buffer);
@ -571,19 +558,15 @@ needs_new_block:
BAN::ErrorOr<bool> Ext2Inode::is_directory_empty()
{
ASSERT(mode().ifdir());
if (m_inode.flags & Ext2::Enum::INDEX_FL)
{
dwarnln("deletion of indexed directory is not supported");
return BAN::Error::from_errno(ENOTSUP);
}
auto block_buffer = m_fs.get_block_buffer();
// Confirm that this doesn't contain anything else than '.' or '..'
for (uint32_t i = 0; i < blocks(); i++)
for (uint32_t i = 0; i < max_used_data_block_count(); i++)
{
const uint32_t block = fs_block_of_data_block_index(i);
m_fs.read_block(block, block_buffer);
// FIXME: can we actually assume directories have all their blocks allocated
const uint32_t block_index = fs_block_of_data_block_index(i).value();
m_fs.read_block(block_index, block_buffer);
blksize_t offset = 0;
while (offset < blksize())
@ -607,13 +590,19 @@ needs_new_block:
BAN::ErrorOr<void> Ext2Inode::cleanup_default_links()
{
ASSERT(mode().ifdir());
if (m_inode.flags & Ext2::Enum::INDEX_FL)
{
dwarnln("deletion of indexed directory is not supported");
return BAN::Error::from_errno(ENOTSUP);
}
auto block_buffer = m_fs.get_block_buffer();
for (uint32_t i = 0; i < blocks(); i++)
for (uint32_t i = 0; i < max_used_data_block_count(); i++)
{
const uint32_t block = fs_block_of_data_block_index(i);
m_fs.read_block(block, block_buffer);
// FIXME: can we actually assume directories have all their blocks allocated
const uint32_t block_index = fs_block_of_data_block_index(i).value();
m_fs.read_block(block_index, block_buffer);
bool modified = false;
@ -648,7 +637,7 @@ needs_new_block:
}
if (modified)
m_fs.write_block(block, block_buffer);
m_fs.write_block(block_index, block_buffer);
}
return {};
@ -665,10 +654,11 @@ needs_new_block:
auto block_buffer = m_fs.get_block_buffer();
for (uint32_t i = 0; i < blocks(); i++)
for (uint32_t i = 0; i < max_used_data_block_count(); i++)
{
const uint32_t block = fs_block_of_data_block_index(i);
m_fs.read_block(block, block_buffer);
// FIXME: can we actually assume directories have all their blocks allocated
const uint32_t block_index = fs_block_of_data_block_index(i).value();
m_fs.read_block(block_index, block_buffer);
blksize_t offset = 0;
while (offset < blksize())
@ -702,7 +692,7 @@ needs_new_block:
// FIXME: This should expand the last inode if exists
entry.inode = 0;
m_fs.write_block(block, block_buffer);
m_fs.write_block(block_index, block_buffer);
}
offset += entry.rec_len;
}
@ -711,114 +701,72 @@ needs_new_block:
return {};
}
#define READ_OR_ALLOCATE_BASE_BLOCK(index_) \
do { \
if (m_inode.block[index_] != 0) \
m_fs.read_block(m_inode.block[index_], block_buffer); \
else \
{ \
m_inode.block[index_] = TRY(m_fs.reserve_free_block(block_group())); \
memset(block_buffer.data(), 0x00, block_buffer.size()); \
} \
} while (false)
#define READ_OR_ALLOCATE_INDIRECT_BLOCK(result_, buffer_index_, parent_block_) \
uint32_t result_ = ((uint32_t*)block_buffer.data())[buffer_index_]; \
if (result_ != 0) \
m_fs.read_block(result_, block_buffer); \
else \
{ \
const uint32_t new_block_ = TRY(m_fs.reserve_free_block(block_group())); \
\
((uint32_t*)block_buffer.data())[buffer_index_] = new_block_; \
m_fs.write_block(parent_block_, block_buffer); \
\
result_ = new_block_; \
memset(block_buffer.data(), 0x00, block_buffer.size()); \
} \
do {} while (false)
#define WRITE_BLOCK_AND_RETURN(buffer_index_, parent_block_) \
do { \
const uint32_t block_ = TRY(m_fs.reserve_free_block(block_group())); \
\
ASSERT(((uint32_t*)block_buffer.data())[buffer_index_] == 0); \
((uint32_t*)block_buffer.data())[buffer_index_] = block_; \
m_fs.write_block(parent_block_, block_buffer); \
\
m_inode.blocks += blocks_per_fs_block; \
update_and_sync(); \
\
return block_; \
} while (false)
BAN::ErrorOr<uint32_t> Ext2Inode::allocate_new_block()
BAN::ErrorOr<uint32_t> Ext2Inode::allocate_new_block_to_indirect_block(uint32_t& block, uint32_t index, uint32_t depth)
{
const uint32_t blocks_per_fs_block = blksize() / 512;
const uint32_t inode_blocks_per_fs_block = blksize() / 512;
const uint32_t indices_per_fs_block = blksize() / sizeof(uint32_t);
uint32_t block_array_index = blocks();
if (depth == 0)
ASSERT(block == 0);
auto update_and_sync =
[&]
{
if (mode().ifdir())
m_inode.size += blksize();
sync();
};
// direct block
if (block_array_index < 12)
if (block == 0)
{
const uint32_t block = TRY(m_fs.reserve_free_block(block_group()));
block = TRY(m_fs.reserve_free_block(block_group()));
m_inode.blocks += inode_blocks_per_fs_block;
ASSERT(m_inode.block[block_array_index] == 0);
m_inode.block[block_array_index] = block;
m_inode.blocks += blocks_per_fs_block;
update_and_sync();
return block;
auto block_buffer = m_fs.get_block_buffer();
memset(block_buffer.data(), 0x00, block_buffer.size());
m_fs.write_block(block, block_buffer);
}
block_array_index -= 12;
if (depth == 0)
return block;
auto block_buffer = m_fs.get_block_buffer();
m_fs.read_block(block, block_buffer);
// singly indirect block
if (block_array_index < indices_per_fs_block)
uint32_t divisor = 1;
for (uint32_t i = 1; i < depth; i++)
divisor *= indices_per_fs_block;
uint32_t& new_block = block_buffer.span().as_span<uint32_t>()[(index / divisor) % indices_per_fs_block];
uint32_t allocated_block = TRY(allocate_new_block_to_indirect_block(new_block, index, depth - 1));
m_fs.write_block(block, block_buffer);
return allocated_block;
}
BAN::ErrorOr<uint32_t> Ext2Inode::allocate_new_block(uint32_t data_block_index)
{
const uint32_t inode_blocks_per_fs_block = blksize() / 512;
const uint32_t indices_per_fs_block = blksize() / sizeof(uint32_t);
BAN::ScopeGuard syncer([&] { sync(); });
if (data_block_index < 12)
{
READ_OR_ALLOCATE_BASE_BLOCK(12);
WRITE_BLOCK_AND_RETURN(block_array_index, m_inode.block[12]);
ASSERT(m_inode.block[data_block_index] == 0);
m_inode.block[data_block_index] = TRY(m_fs.reserve_free_block(block_group()));
m_inode.blocks += inode_blocks_per_fs_block;
return m_inode.block[data_block_index];
}
data_block_index -= 12;
block_array_index -= indices_per_fs_block;
if (data_block_index < indices_per_fs_block)
return TRY(allocate_new_block_to_indirect_block(m_inode.block[12], data_block_index, 1));
data_block_index -= indices_per_fs_block;
// doubly indirect block
if (block_array_index < indices_per_fs_block * indices_per_fs_block)
{
READ_OR_ALLOCATE_BASE_BLOCK(13);
READ_OR_ALLOCATE_INDIRECT_BLOCK(direct_block, block_array_index / indices_per_fs_block, m_inode.block[13]);
WRITE_BLOCK_AND_RETURN(block_array_index % indices_per_fs_block, direct_block);
}
if (data_block_index < indices_per_fs_block * indices_per_fs_block)
return TRY(allocate_new_block_to_indirect_block(m_inode.block[13], data_block_index, 2));
data_block_index -= indices_per_fs_block;
block_array_index -= indices_per_fs_block * indices_per_fs_block;
// triply indirect block
if (block_array_index < indices_per_fs_block * indices_per_fs_block * indices_per_fs_block)
{
READ_OR_ALLOCATE_BASE_BLOCK(14);
READ_OR_ALLOCATE_INDIRECT_BLOCK(indirect_block, block_array_index / (indices_per_fs_block * indices_per_fs_block), m_inode.block[14]);
READ_OR_ALLOCATE_INDIRECT_BLOCK(direct_block, (block_array_index / indices_per_fs_block) % indices_per_fs_block, indirect_block);
WRITE_BLOCK_AND_RETURN(block_array_index % indices_per_fs_block, direct_block);
}
if (data_block_index < indices_per_fs_block * indices_per_fs_block * indices_per_fs_block)
return TRY(allocate_new_block_to_indirect_block(m_inode.block[14], data_block_index, 3));
ASSERT_NOT_REACHED();
}
#undef READ_OR_ALLOCATE_BASE_BLOCK
#undef READ_OR_ALLOCATE_INDIRECT_BLOCK
#undef WRITE_BLOCK_AND_RETURN
void Ext2Inode::sync()
{
auto inode_location = m_fs.locate_inode(ino());
@ -836,26 +784,22 @@ needs_new_block:
{
ASSERT(mode().ifdir());
const uint32_t block_size = blksize();
const uint32_t data_block_count = blocks();
auto block_buffer = m_fs.get_block_buffer();
for (uint32_t i = 0; i < data_block_count; i++)
for (uint32_t i = 0; i < max_used_data_block_count(); i++)
{
const uint32_t block_index = fs_block_of_data_block_index(i);
// FIXME: can we actually assume directories have all their blocks allocated
const uint32_t block_index = fs_block_of_data_block_index(i).value();
m_fs.read_block(block_index, block_buffer);
const uint8_t* block_buffer_end = block_buffer.data() + block_size;
const uint8_t* entry_addr = block_buffer.data();
while (entry_addr < block_buffer_end)
BAN::ConstByteSpan entry_span = block_buffer.span();
while (entry_span.size() >= sizeof(Ext2::LinkedDirectoryEntry))
{
const auto& entry = *(const Ext2::LinkedDirectoryEntry*)entry_addr;
auto& entry = entry_span.as<const Ext2::LinkedDirectoryEntry>();
BAN::StringView entry_name(entry.name, entry.name_len);
if (entry.inode && entry_name == file_name)
return BAN::RefPtr<Inode>(TRY(Ext2Inode::create(m_fs, entry.inode)));
entry_addr += entry.rec_len;
entry_span = entry_span.slice(entry.rec_len);
}
}