Semiconductor & Computer Engineering
From WikiChip
Difference between revisions of "nvidia/microarchitectures/carmel"
(→Performance claims) |
(→Overview) |
||
| Line 50: | Line 50: | ||
== Overview == | == Overview == | ||
Carmel is a CPU microarchitecture designed by Nvidia for their SoCs. The design consists of an 8-core cluster made of 4 core duplexes. The entire complex has [[cache coherency]] as well as an I/O coherent memory subsystem which is designed for communication with the various other accelerators on their SoCs such as the vision accelerator, deep learning accelerator, multimedia engine, and the GPU. | Carmel is a CPU microarchitecture designed by Nvidia for their SoCs. The design consists of an 8-core cluster made of 4 core duplexes. The entire complex has [[cache coherency]] as well as an I/O coherent memory subsystem which is designed for communication with the various other accelerators on their SoCs such as the vision accelerator, deep learning accelerator, multimedia engine, and the GPU. | ||
| + | |||
| + | The cores consume 500-1,500 mW per core. | ||
=== Performance claims === | === Performance claims === | ||
Revision as of 16:36, 4 October 2018
| Edit Values | |
| Carmel µarch | |
| General Info | |
| Arch Type | CPU |
| Designer | Nvidia |
| Manufacturer | TSMC |
| Introduction | January 7, 2018 |
| Process | 12 nm |
| Core Configs | 8 |
| Pipeline | |
| Type | Superscalar |
| Instructions | |
| ISA | ARMv8 |
| Cache | |
| L2 Cache | 2 MiB/cluster |
| L3 Cache | 4 MiB/complex |
| Succession | |
Carmel is a the successor to Denver 2, an ARM microarchitecture for Nvidia's Tegra series of SoCs.
Contents
Process Technology
Carmel is integrated into chips fabricated on TSMC 12 nm process (12FFN)
Architecture
Nvidia disclosed very few details regarding Carmel. Carmel is a 10-wide superscalar with each core supporting dual execution mode.
Key changes from Denver 2
- 12 nm (12FFN)
- ARMv8.2
- ARM RAS standard support
- Wider dispatch (10, up from 7)
Memory Hierarchy
- Cache
- Parity & ECC
- L1
- L2
- 2 MiB
- Shared per duplex
- 2 MiB
- L3
- 4 MiB
- Shared by entire cluster
- Exclusive
- 4 MiB
Block Diagram
CPU Complex
Overview
Carmel is a CPU microarchitecture designed by Nvidia for their SoCs. The design consists of an 8-core cluster made of 4 core duplexes. The entire complex has cache coherency as well as an I/O coherent memory subsystem which is designed for communication with the various other accelerators on their SoCs such as the vision accelerator, deep learning accelerator, multimedia engine, and the GPU.
The cores consume 500-1,500 mW per core.
Performance claims
| SPECint 2000 Rate | SPECint 2006 Rate |
|---|---|
| 2700 | 21 |
Die
CPU Complex
- 8 cores
- 4 duplexes
- shared L3
- ~62.25 mm² die size area
.png)
CPU Duplex
- 2 cores
- ~11.4 mm² die size area
.png)
Core
- ~5.75 mm² die size area
Bibliography
- IEEE Hot Chips 30 Symposium (HCS) 2018.
Facts about "Carmel - Microarchitectures - Nvidia"
| codename | Carmel + |
| core count | 8 + |
| designer | Nvidia + |
| first launched | January 7, 2018 + |
| full page name | nvidia/microarchitectures/carmel + |
| instance of | microarchitecture + |
| instruction set architecture | ARMv8 + |
| manufacturer | TSMC + |
| microarchitecture type | CPU + |
| name | Carmel + |
| process | 12 nm (0.012 μm, 1.2e-5 mm) + |