COSC 423 - Midterm Cards 2.0

What does IVT stand for?

Interrupt Vector Table

What is the motivation of IVT?

Create global functions that can be called from any segment

Where is IVT located in memory?

0×0000:0×0000 to 0×0000:0×03FF (first 1024 bytes of memory)

How many IVT entries are there?

256 entries (0×00 to 0xFF)

How many bytes per IVT entry?

4 bytes per entry

What does each 4-byte IVT entry contain?

Bytes 0-1: Offset (IP - instruction pointer)

Bytes 2-3: Segment (CS - code segment)

Formula to calculate IVT entry address for interrupt N?

Address = interrupt number x 4

is IVT used in modern protected mode systems?

No, only used in 16-bit real mode

What replaced IVT in protected mode?

IDT (Interrupt Descriptor Table)

Is IVT still used anywhere today?

Yes, during boot process in real mode before switching to protected mode

Why is CLI necessary before modifying IVT?

Disables interrupts

Why is SIT necessary after modifying IVT?

Re-enables interrupts

What are the critical steps to register an IVT handler?

1. CLI (disable interrupts)

2. Set ES to 0×0000

3. Write offset to IVT entry

4. Write segment to IVT entry

5. STI (enable interrupts)

What instruction must end an interrupt handler?

IRET

Why can’t you use RET in an interrupt handler

RET only pops IP (and CS for far return)
IRET pops IP, CS, and FLAGS

What does IRET restore that RET doesn’t

FLAGS register

What are the 3 values IRET pops from stack?

IP

CS

FLAGS

What is the major secutiry flaw of real-mode IVT

No protection, any program can override any interrupt

How does protected mode (IDT) fix IVT security problem

IDT uses Ring levels (0-3). Only Ring 0 (kernel) can modify IDT. User programs (Ring 3) can’t access or modify interrupt handlers

What are Ring levels in protected mode?

Ring 0 = Kernel (highest privilege)

Ring 1-2 = Device drivers 

Ring 3 = User applications (lowest privilege)

Which Ring level can modify the IDT

Ring 0 (kernel mode) only

What is First-Fit allocation

Scan from beginning, allocate first hole large enough.

Fast but creates small holes at start of memory

What is Best-Fit allocation

Scan entire list, choose smallest hole that fits.

Minimizes wasted space but creates tiny unusable holes

What is Worst-Fit allocation

Choose largest available hole.

Leftover hole more likely usable but wastes large holes

External Fragmentation

Allocated block larger than needed, waste inside allocation

Ex: Need 1000 bytes, allocate 1024-byte block, 24 bytes wasted inside

Can FAT12 have external fragementation?

No, FAT12 uses chained allocation (not contiguous)

Files can be scattered across non-contiguous clusters

What is the identical concept: Sector (disk) vs ? (memory)

Page

Why does FAT use chained allocation instead of indexed

1. Index section must be contiguous - same fragmentation problem as data

2. Space constraints in FAT12

Why does memory use indexed (page tables) instead of chained

Time issue, memory needs fast random access

What are 2 problems of page allocation?

1. Internal fragmentation 

2. Grouped data (arrays) get scattered

Whats the advantage of segment over page

Only 2 numbers to describe: START address + LIMIT (size)

Segment is contiguous which is better for grouped data structures

Search Complexity: Linear search in page table

O(n)

Search Complexity: Binary search in page table

O(log n) but requires O(n log n) sort first

Search Complexity: Hash table in page table

O(1)

Which search method is typically used for page tables

Hash Table (O(1) lookup)

Can FAT12 have 525 files in root directory? Why?

No, root directory fixed at 224 entries maximum

Can FAT12 have 525 files total? Why?

Yes, by using subdirectories which have unlimited space

Can FAT12 have 525 subdirectories? Why?

Yes, subdirectories unlimited

How does FAT12 handle more than 224 files?

By using subdirectories

Is FAT12 purely chained allocation?

No, chained + indexed

Does FAT use best-fit allocation

No, uses chained allocation

Does FAT12 suffer from external fragmentation

No, doesn’t use contiguous allocation.

Files can be scattered

Was FAT created by Microsoft

No, purchased

Why are there 2 copies of FAT

Reliability

How many bits per FAT12 entry

12 bits

What’s the first valid FAT entry number

Entry 2 (entries 0 and 1 are reserved)

Can FAT entry values be negative

No, all FAT values are unsigned

What does FAT entry value 0×000 mean

Free cluster

What does FAT entry value 0xFFF mean

End of file marker

What’s the most influential OS overall

Unix

What OS is most influential for education

Minix

Is iOS Linux based

No (based on BSD/Darwin)

Is Android Linux based

Yes

3 main OS objectives

1. Convenience

2. Efficiency

3. Ability to Evolve

Standard 1.44 MB floppy: Total sectors?

2880 sectors

Standard 1.44 MB floppy: Boot sector count?

1 sector

Standard 1.44 MB floppy: FAT sectors total

18 sectors (9 per FAT x 2 copies)

Standard 1.44 MB floppy: Root directory sectors

14 sectors

Standard 1.44 MB floppy: First data sector

33 (1 boot + 18 FAT + 14 root)

Standard 1.44 MB floppy: First valid cluster number

2 (clusters 0 and 1 reserved)

Boot sector bytes 11-12 contain what

Bytes per sector

Boot sector byte 13 contains what

Sectors per cluster

Boot sector bytes 14-15 contain what

Reserved sectors

Boot sector byte 16 contains what

Number of FAT copies

Boot sector bytes 17-18 contain what

Root directory entries

Boot sector bytes 19-20 contain what

Total sectors

Boot sector byte 21 contains what

Media descriptor

Boot sector bytes 22-23 contain what

Sectors per FAT

Formula to calculate byte offset for FAT entry N

byte offset = entry x 1.5

Formula to find which FAT sector contains entry N

FAT sector = byte offset / 512 (round up)

Formula to find position within FAT sector

position = byte offset mod 512

Formula to calculate row number in hex dump

row = position / 16

Formula to calculate column number in hex dump

col = position mod 16

Formula to convert cluster number to logical sector

logical sector = cluster + 31

Formula to calculate how many sectors a file occupies

sectors = file size / 512 (round up)

Formula to claculate first root directory logical sector

first root = 1 + (2 × sectors per FAT)

For standard floppy: 1 + (2 × 9) = 19

Formula to calculate first data sector (cluster 2)

First data = 1 + (2 x sectors per FAT) + root sectors

For standard floppy: 1 + 18 + 14 = 33

Formula to calculate root directory sectors needed

root sectors = (entries x 32) / 512

For 224 entries: (224 × 32) / 512 = 14

Formula to calculate IVT entry address for interrupt N

address = interrupt number x $

Result format: 0×0000:0×00XX

IVT total size formula

IVT size = 256 entries x 4 bytes = 1024 bytes = 1 KB

Formula to convert manufacturer GB to Windows GB

Windows GB = (Advertised x 1000³) / 1024³

Formula to convert Windows GB to manufacturer GB

Advertised GB = (Windows x 1024³) / 1000³

Formula for external fragmentation calculation

external fragmentation = sum of all remaining holes

Ex: [20,150,30,400] = 20+150+30+400 = 600 KB

What does VFAT stand for

Virtual FAT (extension for long filenames)

Maximum filename length in VFAT

255 characters

How to identify VFAT entry in hexdump

Byte 11 = 0×0F

In what order does VFAT store long filename parts

Reverse order (last part first, then middle, then start)

How many characters per VFAT pseudo-entry

13 Unicode characters per entry

Can DOS read VFAT long filenames

No, only sees 8.3 shortname 

If filename is “My Document.docx” (16 chars), how many VFAT entries needed

[16 / 13] = 2 VFAT entries

What comes after VFAT pseudo-entries

Actual 8.3 directory entry with file data (cluster, size, date, time)

Why does VFAT maintain 8.3 shortname

Backward compatibility with DOS/older systems

What’s stored in VFAT pseudo-entry checksum

Checksum of 8.3 shortname to verify entries belong together

How many bits per FAT16 entry

16 bits (2 bytes)

How many bits per FAT32 entry

32 bits (4 bytes), but only 28 bits actually used

Maximum clusters in FAT12

4096 clusters (2^12)

Maximum clusters in FAT16

65536 clusters (2^16)