qihaiyan / fat12 Goto Github PK

When I tried copying a file bigger than one cluster into a floppy image with the following conditions, it did not work. The FAT was not updated. Looking at the image using a hex editor, I saw a weird part was updated as if it were the FAT.

1. The number of bytes per sector is 512.
2. The number of sectors per cluster is 4.
3. The number of FATs is 2.
4. The number of reserved sectors is 1.

Reading the source code, I found two functions calculating the offset of an entry in the FAT: get_fat_entry() and set_fat_entry(), on dos.c. For some reason, the number of sectors per cluster was used. I found this odd, since the FAT entry offset is (usually) calculated by finding the FAT offset and then adding entry_number * 1.5.

By removing the count of sectors per cluster, the problem was fixed. If you'd like, I can just make a pull request with this change.