ARM Overview

Report
ARM Processor Overview
Prof. Taeweon Suh
Computer Science Education
Korea University
ARM (www.arm.com)
2
Korea Univ
ARM
Source: 2008 Embedded SW Insight Conference
3
Korea Univ
ARM Partners
Source: 2008 Embedded SW Insight Conference
4
Korea Univ
ARM (as of 2008)
Source: 2008 Embedded SW Insight Conference
5
Korea Univ
ARM Brief
• ARM architecture was first developed in the 1980s by
Acorn
• Spin off from Acron in 1990
• Released ARM6 in early 1992
• …
• As of 2013, ARM architecture is the most widely used
32-bit ISA in terms of quantity produced
• In 2010 alone, 6.1 billion ARM-based processors
shipped, representing
 95% of smartphones
 35% of digital TV and set-top boxes
 10% of mobile computers
Source: Wikipedia
6
Korea Univ
ARM Architecture
• ARM is RISC (Reduced Instruction Set Computer)
 x86 ISA is based on CISC (Complex Instruction Set
Computer) even though x86 internally implements RISClike microcode and pipelining
• Suitable for embedded systems
 Very small die size (low price)
 Low power consumption (longer battery life)
7
Korea Univ
ARM Processor Portfolio
Source: 2008 Embedded SW Insight Conference
8
Korea Univ
Product Code
•
•
•
•
•
•
•
T: Thumb
T2: Thumb-2 Enhancement
D: Debug
M: Multiplier
I: Embedded ICE (In-Circuit Emulation)
E: Enhanced DPS Extension
J: Jazelle
 Direct execution of 8-bit Java bytecode in hardware
• S: Synthesizable core
• Z: Should be TrustZone?
9
Korea Univ
ARM Cortex Series
• ARM Cortex-A family:
• ARM Cortex-R family:
Cortex-A15
...2.5GHz
x1-4
Cortex-A9
Cortex-A8
x1-4
Cortex-A5
 Embedded processors for real-time
signal processing, control applications
1-2
Cortex-R7
1-2
Cortex-R5
Cortex-R4
• ARM Cortex-M family:
 Microcontroller-oriented processors
for MCU, ASSP, and SoC applications
Cortex-M4
SC300
Cortex-M3
Cortex-M1
SC000
Unparalleled Applicability
 Applications processors for featurerich OS and 3rd party applications
x1-4
Cortex-M0
12k gates...
Source: ARM Processor Portfolio 2011
10
Korea Univ
ARMv7-A
 ACP: Accelerator Coherency Port
www.arm.com
 SCU: Snoop Control Unit
11
Korea Univ
ARM Processor Brief
#pipeline stages
Frequency
Architecture
Process
ARM6 (1992)
3
~33MHz
ARMv3
1.2μm
ARM7TDMI
3
~70MHz
ARMv4
0.13nm
ARM920T
5
~400MHz
ARMv4
90nm
ARM1136J
8
~1Ghz
ARMv6
65nm
Cortex-A9
8~11 (OoO)
~2GHz
ARMv7
32nm
Cortex-A15
15~24 (OoO)
~2.5GHz
ARMv7
22nm
 OOO: Out Of Order
12
Korea Univ
ARM Instruction Overview
• ARM is a RISC machine, so the instruction length is fixed
 In ARM mode, instructions are 32-bit wide
 In Thumb mode, instructions are 16-bit wide
• Most ARM instructions can be conditionally
executed
 It means that they have their normal effect only if the N
(Negative), Z (Zero), C (Carry) and V (Overflow) flags in the CPSR
satisfy a condition specified in the instruction
• If the flags do not satisfy this condition, the instruction acts as a NOP (No
Operation)
• In other words, the instruction has no effect and advances to the next
instruction
13
Korea Univ
ARM Instructions
• For the complete instruction set, refer to the
“ARM Architecture Reference Manual”
• We are going to cover essential and important
instructions in this course
 If you completely understand one CPU, it is pretty
straightforward to understand other CPUs
14
Korea Univ
Essential Instructions
• Instruction categories
 Data processing instructions: add, sub, cmp, and, or
 Memory access instructions: ldr, str
 Branch instructions: b, bl
 Miscellaneous instructions:
Real-PC system
CPU
FSB
(Front-Side Bus)
Main
Memory
(DDR)
Address Bus
Simplified
ARM CPU
North
Bridge
Data Bus
DMI
(Direct Media I/F)
Memory
(Instruction,
data)
South
Bridge
15
Korea Univ
A Memory Hierarchy
DDR3
HDD
2nd Gen. Core i7
(2011)
16
Korea Univ
A Memory Hierarchy
lower level
higher level
On-Chip Components
CPU Core
Reg
File
L1I
(Instr )
L2
L1D
(Data)
Speed (cycles):
½’s
1’s
Size (bytes):
100’s
10K’s
Cost:
L3
Main
Memory
(DRAM)
Secondary
Storage
(Disk)
10’s
100’s
10,000’s
M’s
G’s
T’s
highest
lowest
17
Korea Univ
ARM Registers
• ARM has 31 general purpose registers and 6 status
registers (32-bit each)
18
Korea Univ
ARM Registers
• Unbanked registers: R0 ~ R7
 Each of them refers to the same 32-bit
physical register in all processor modes.
 They are completely general-purpose
registers, with no special uses implied by
the architecture
• Banked registers: R8 ~ R14
 R8 ~ R12 have no dedicated special
purposes
• FIQ mode has dedicated registers for fast
interrupt processing
 R13 and R14 are dedicated for special
purposes for each mode
19
Korea Univ
R13, R14, and R15
• Some registers in ARM are used for special purposes
 R15 == PC (Program Counter)
• x86 uses a terminology called IP (Instruction Pointer)
 R14 == LR (Link Register)
 R13 == SP (Stack Pointer)
20
Korea Univ
ARM9 Register File
• A set of architectural (programmervisible) registers inside CPU is called
register file
Register File
src1 addr
 Register file can be implemented with
flip-flops or SRAM
 ARM9 register file has 16 32-bit registers
• 3 read ports
• 2 write ports
 Register file access is much faster than
main memory or cache because there
are a very limited number of registers
and they reside inside CPU
 So, compilers strive to use the register
file when translating high-level code to
assembly code
src2 addr
src3 addr
dst1 addr
write1 data
dst2 addr
write2 data
4
32 bits
R0
4
src1
data
32
src2
data
32
src3
data
R1
R2
4
R3
4
…
32
4
R14
32
R15
write1
21
32
write2
Korea Univ
CPSR
• Current Program Status Register (CPSR) is accessible in all modes
• Contains all condition flags, interrupt disable bits, the current
processor mode
22
Korea Univ
CPSR in ARM
23
Korea Univ
CPSR bits
24
Korea Univ
CPSR bits
• ARM: 32-bit mode
• Thumb: 16-bit mode
• Jazelle: Special mode for JAVA acceleration
25
Korea Univ
ARM Instruction Format
Arithmetic and Logical
Instructions
Memory Access
Instructions
(Load/Store)
Branch Instructions
Software Interrupt
Instruction
26
Korea Univ
ARM Instruction Fields
32-bit
opcode
operation code
Rn
4-bits
first source register
Rm
4-bits
second source register
Rs
4-bits
third source register
Rd
4-bits
destination register
shift
2-bits
shift type*
shift amount
5-bits
shift by how many bits
* Shift type: Arithmetic, logical (left, right)
27
Korea Univ
Condition Field
28
Korea Univ
Overview of ARM Operation
• ARM arithmetic in assembly form
add R3, R1, R5
# R3 = R1 + R5
 R1 and R5 are source operands, and R3 is destination
 # indicate a comment, so assembler ignores it
• Operands of arithmetic instructions come from special
locations called registers inside CPU or from the immediate
field in instructions
 All CPUs (x86, PowerPC, MIPS, ARM…) have registers inside
• Registers are visible to the programmers
 ARM has a register file consisting of 16 registers
29
Korea Univ
Simplified Version of CPU Internal
add
Registers
R3, R1, R5
# R3 = R1 + R5
ARM CPU
32 bits
R0
R1
R2
R1
Memory
Address Bus
R3
…
+
R3
add R3, R1, R5
R5
Data Bus
R14
R15
30
Korea Univ
ARM Register Convention
Name
R0~R3
•
•
R13
R14
R15
Usage
Arguments passed to a subroutine
Results returned from a subroutine
Stack pointer
Link Register
Program Counter
31
Korea Univ
Backup Slides
32
Korea Univ
ARM Processor Family
Source: Wikipedia
33
Korea Univ
NEON & VFP
www.arm.com
34
Korea Univ
Register Mapping
• NEON Advanced SIMD and
VFP use the same register set
35
Korea Univ
NEON
• Advanced SIMD (Single Instruction Multiple Data)
• It supports 8, 16, 32 and 64-bit integer and singleprecision (32-bit) floating point data
• Up to 16 operations at the same time
 1B x 16 = 16B (= 1 quad word)
http://en.wikipedia.org/wiki/ARM_architecture
36
Korea Univ
VFP (Vector Floating Point)
• FPU (Floating Point Unit) coprocessor extension to ARM
architecture
• Single-precision and double-precision FP computation
 Compliant with IEEE 754-1985
• Intended to support execution of short “vector mode”
instructions, but operated on “each” vector element
sequentially
 Thus, did not offer the performance of true SIMD
 This vector mode was thus removed shortly after its introduction, to
be replaced with the much more powerful NEON Advanced SIMD
http://en.wikipedia.org/wiki/ARM_architecture
37
Korea Univ
ARM Processor Selector
www.arm.com
38
Korea Univ

similar documents