RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-07-25 12:37:44 +00:00
Code Issues Activity Actions

Kernel/FileSystem/FATFS: Support for FAT12/16 DOS BIOS Parameter Blocks

Browse source
This commit is contained in:
Taj Morton 2023-07-08 22:10:10 -07:00 • committed by Andrew Kaster
parent 02edd240ae
commit 67f567348f
4 changed files with 332 additions and 44 deletions
Download patch file Download diff file Expand all files Collapse all files

33
Kernel/FileSystem/FATFS/Inode.cpp
View file

@ -58,12 +58,28 @@ ErrorOr<Vector<BlockBasedFileSystem::BlockIndex>> FATInode::compute_block_list()
while (cluster < no_more_clusters) {
dbgln_if(FAT_DEBUG, "FATFS: Appending cluster {} to inode {}'s cluster chain", cluster, index());
BlockBasedFileSystem::BlockIndex first_block = fs().first_block_of_cluster(cluster);
for (u8 i = 0; i < fs().boot_record()->sectors_per_cluster; i++)
block_list.append(BlockBasedFileSystem::BlockIndex { first_block.value() + i });
auto first_block_and_length = fs().first_block_of_cluster(cluster);
for (u8 i = 0; i < first_block_and_length.number_of_sectors; i++)
block_list.append(BlockBasedFileSystem::BlockIndex { first_block_and_length.start_block.value() + i });
// Clusters 0 and 1 are reserved in the FAT, and their entries in the FAT will
// not point to another valid cluster in the chain (Cluster 0 typically holds
// the "FAT ID" field with some flags, Cluster 1 should be the end of chain
// marker).
// Internally, we use `cluster == 0` to represent the root directory Inode,
// which is a signal to read the root directory region blocks on FAT12/16
// file systems. (`fs().first_block_of_cluster` will return the appropriate
// block/sectors to read given cluster == 0).
// Therefore, we read one set of sectors for these invalud cluster numbers,
// and then terminate the loop becuase the FAT entry at `cluster` for these
// values does not represent the next step in the chain (because there is
// nothing else to read).
if (cluster <= 1) {
break;
}
u32 fat_offset = cluster * sizeof(u32);
u32 fat_sector_index = fs().boot_record()->reserved_sector_count + (fat_offset / fs().m_device_block_size);
u32 fat_sector_index = fs().m_parameter_block->common_bpb()->reserved_sector_count + (fat_offset / fs().m_device_block_size);
u32 entry_offset = fat_offset % fs().m_device_block_size;
TRY(fs().raw_read(fat_sector_index, fat_sector_buffer));
@ -126,6 +142,15 @@ ErrorOr<RefPtr<FATInode>> FATInode::traverse(Function<ErrorOr<bool>(RefPtr<FATIn
} else if (entry->attributes == FATAttributes::LongFileName) {
dbgln_if(FAT_DEBUG, "FATFS: Found LFN entry");
TRY(lfn_entries.try_append(*reinterpret_cast<FATLongFileNameEntry*>(entry)));
} else if ((entry->first_cluster_high << 16 | entry->first_cluster_low) <= 1 && entry->file_size > 0) {
// Because clusters 0 and 1 are reserved, only empty files (size == 0 files)
// should specify these clusters.
// This driver uses a cluster number == 0 to represent the root directory inode
// on FAT12/16 file systems (a signal to look in the root directory region),
// so we ensure that no entries read off the file system have a cluster number
// that would also point to this region.
dbgln_if(FAT_DEBUG, "FATFS: Invalid cluster for entry");
return EINVAL;
} else {
dbgln_if(FAT_DEBUG, "FATFS: Found 8.3 entry");
lfn_entries.reverse();