ARMv8 & ARMv9 아키텍처 extension

 

ARM Cortex-A 시리즈

참고: ARM Cortex-A 시리즈 | 나무위키

• Cortex A57 (Little A53)
  • SoC) 퀄컴 스냅드래곤 808, 810, 삼성 엑시노스 7 Octa 5433, 7420
  • 참고
    • ARM Cortex-A53 | Wiki
• Cortex A72
  • Cortex-A57 대비 동클럭당 23% 향상(일반 연산 16%, 암호화 38%, 메모리 I/O 50%, 실수 연산 26%, 정수 연산 16%)
  • 참고:
    • Fast Processing Designed for Multiple Application Areas | ARM
    • ARM Cortex-A72 | Wiki
• Cortex A73
  • Cortex A72와 성능 유사, 단 면적당 성능비 40% 향상, 20% 전력 효율 개선
  • 최대 2.8Ghz 클럭
  • SoC) 퀄컴 스냅드래곤 835
  • 참고:
    • Sustained Performance for Mobile Processing | ARM
    • ARM Cortex-A73 | Wiki
• Cortex A75 (Little A55)
  • A73 대비 20% 향상
  • ARMV8.2, DynamIQ, L3 캐시 첫 채용
  • 최대 3.0Ghz 클럭
  • 참고:
    • First Generation, High-Performance CPU Based on DynamIQ Technology | ARM
    • ARM Cortex-A75 | Wiki
• Cortex A76
  • Cortex-A75 대비 최대 35% 향상, 40% 전력 효율 개선, 머신 러닝 4배 향상, 메모리 대역폭 90% 확장
  • 최대 3.0/3.3Ghz 클럭 (phone/tablets)
  • 참고:
    • Second-Generation, High-Performance CPU Based on DynamIQ Technology | ARM
    • ARM Cortex-A76 | Wiki
• Cortex A77
  • A76 대비 20% 향상
  • 최대 3.0/3.3Ghz 클럭 (phone/tablets)
  • 참고:
    • Third-Generation, High-Performance CPU Based on DynamIQ Technology | ARM
    • ARM Cortex-A77 | Wiki
• Cortex A78 (Little A55, Custom X1)
  • A77 대비 5% 향상
  • 마지막 ARMv8 아키텍처
  • 최대 3.0/3.3Ghz 클럭 (phone/tablets)
  • 참고:
    • Fourth-Generation, High-Performance CPU Based on DynamIQ Technology | ARM
    • ARM Cortex-A78 | Wiki
• Cortex A710 (Little A510, Custom X2)
  • A78 대비 10% 향상, 소비전력 30% 개선, 머신 러닝 2배 향상
  • 참고:
    • First-Generation Armv9 “big” Cortex CPU Based on Arm DynamIQ Technology | ARM

 

ARMv8 & ARMv9.x 아키텍처 extension

• ARMv8.1
  • ARM Cortex-A32, A35, A53, A57, A72, A73
  • Atomic(LSE) memory access instructions (AArch64)
  • Limited Order regions (AArch64)
  • Increased Virtual Machine Identifier (VMID) size, and Virtualization Host Extensions (AArch64)
  • Privileged Access Never (PAN) (AArch32 and AArch64)
• ARMv8.2 (52bits, share TLB, RAS)
  • ARM Cortex-A55, A75, A76, A77, A78
  • Support for 52-bit addresses (AArch64)
  • The ability for PEs to share Translation Lookaside Buffer (share TLB) entries (AArch32 and AArch64)
  • FP16 data processing instructions (AArch32 and AArch64)
  • Statistical profiling (AArch64)
  • Reliability Availability Serviceabilty (RAS) support becomes mandatory (AArch32 and AArch64)
• ARMv8.3 (Pointer Authentification)
  • ARM Cortex 미채택
  • Pointer authentication (AArch64)
  • Nested virtualization (AArch64)
  • Advanced Single Instruction Multiple Data (SIMD) complex number support (AArch32 and AArch64)
  • Improved JavaScript data type conversion support (AArch32 and AArch64)
  • A change to the memory consistency model (AArch64)
  • ID mechanism support for larger system-visible caches (AArch32 and AArch64)
• ARMv8.4
  • ARM Cortex 미채택, Apple 사에서만 사용
  • Secure virtualization (AArch64)
  • Nested virtualization enhancements (AArch64)
  • Small translation table support (AArch64)
  • Relaxed alignment restrictions (AArch32 and AArch64)
  • Memory Partitioning and Monitoring (MPAM) (AArch32 and AArch64)
  • Additional crypto support (AArch32 and AArch64)
  • Generic counter scaling (AArch32 and AArch64)
  • Instructions to accelerate SHA512 and SHA3 (AArch64 only)
• ARMv8.5 & ARMv9.0
  • ARM Cortex A510, 710
  • Memory Tagging (AArch64)
  • Branch Target Identification (AArch64)
  • Random Number Generator instructions (AArch64)
  • Cache Clean to Point of Deep Persistence (AArch64)
• ARMv8.6 & ARMv9.1
  • General Matrix Multiply (GEMM) instructions (AArch64)
  • Fine grained traps for virtualization (AArch64)
  • High precision Generic Timer
  • Data Gathering Hint (AArch64)
• ARMv8.7 & ARMv9.2
  • Enhanced support for PCIe hot plug (AArch64)
  • Atomic 64-byte load and stores to accelerators (AArch64)
  • Wait For Instruction (WFI) and Wait For Event (WFE) with timeout (AArch64)
  • Branch-Record recording (Armv9.2 only)

 

Feature Name

• 참고: Feature names for A-profile | ARM

 

예) ARM Cortex A57 (

# cat cpuinfo
processor       : 0
BogoMIPS        : 50.00
Features        : fp asimd aes pmull sha1 sha2 crc32
CPU implementer : 0x41
CPU architecture: 8
CPU variant     : 0x1
CPU part        : 0xd07
CPU revision    : 3

 

예) rpi4 – ARM Cortex A72

$ cat cpuinfo
processor       : 0
BogoMIPS        : 108.00
Features        : fp asimd evtstrm crc32 cpuid
CPU implementer : 0x41
CPU architecture: 8
CPU variant     : 0x0
CPU part        : 0xd08
CPU revision    : 3

 

예) Odroid-N2 – ARM Cortex A73

# cat /proc/cpuinfo
processor       : 0
BogoMIPS        : 48.00
Features        : fp asimd evtstrm aes pmull sha1 sha2 crc32 cpuid
CPU implementer : 0x41
CPU architecture: 8
CPU variant     : 0x0
CPU part        : 0xd03
CPU revision    : 4

 

참고

• Understanding the Armv8.x extensions | ARM – 다운로드 pdf
• ARMv9: What is the Big Deal? | gitconnected
• Arm, 모바일 Armv9 CPU 마이크로 아키텍처 : Cortex-X2, Cortex-A710 및 Cortex-A510 발표 | ANNDTECH
  • (1), (2), (3), (4), (5), (6), (7)