Semiconductor & Computer Engineering
From WikiChip
Difference between revisions of "movidius/microarchitectures/shave v2.0"
(→Architecture) |
|||
| Line 11: | Line 11: | ||
== Architecture == | == Architecture == | ||
| − | * Hybrid [[RISC]]-[[DSP]]-[[GPU]] [[VLIW]] | + | * Hybrid [[RISC]]-[[DSP]]-[[GPU]] [[VLIW]] architecture |
| + | * [[Predicated execution]] | ||
| + | * [[Branch delay slots]] | ||
* Tailored to streaming workloads | * Tailored to streaming workloads | ||
* 128-bit vector arithmetic | * 128-bit vector arithmetic | ||
** 8/16/32-bit [[integer]] | ** 8/16/32-bit [[integer]] | ||
** 16/32-bit [[floating point]] | ** 16/32-bit [[floating point]] | ||
| + | * Full support for sparse data structures (matrix/array, random access) | ||
| + | |||
| + | === Instruction Set === | ||
| + | SHAVE supports a mixture of many different types of instructions belonging to a number of different classes of architectures. | ||
| + | |||
| + | * RISC style | ||
| + | ** Instruction predication | ||
| + | ** Large set of integer operations | ||
| + | ** C-compiler support | ||
| + | * VLIW style | ||
| + | ** Parallel functional units controlled by VLIW instructions | ||
| + | ** 8/16/32-bit x 1-4 SIMD int | ||
| + | * DSP style | ||
| + | ** Zero overhead looping | ||
| + | ** Modulo addressing | ||
| + | ** Transparent DMA modes | ||
| + | ** FFT, Viterbi, etc.. | ||
| + | ** Parallel comparisons | ||
| + | * GPU style | ||
| + | ** Streaming operations | ||
| + | ** 16/32-bit FP operations | ||
| + | ** Texture management unit | ||
Revision as of 15:20, 10 March 2018
| Edit Values | |
| SHAVE µarch | |
| General Info | |
| Arch Type | DSP |
| Designer | Movidius |
| Manufacturer | TSMC |
| Introduction | 2011 |
| Pipeline | |
| Type | VLIW |
Streaming Hybrid Architecture Vector Engine (SHAVE) is a hybrid microarchitecture designed by Movidius for their vision processors. SHAVE is incorporated into Movidius Myriad family of vision processors.
Architecture
- Hybrid RISC-DSP-GPU VLIW architecture
- Predicated execution
- Branch delay slots
- Tailored to streaming workloads
- 128-bit vector arithmetic
- 8/16/32-bit integer
- 16/32-bit floating point
- Full support for sparse data structures (matrix/array, random access)
Instruction Set
SHAVE supports a mixture of many different types of instructions belonging to a number of different classes of architectures.
- RISC style
- Instruction predication
- Large set of integer operations
- C-compiler support
- VLIW style
- Parallel functional units controlled by VLIW instructions
- 8/16/32-bit x 1-4 SIMD int
- DSP style
- Zero overhead looping
- Modulo addressing
- Transparent DMA modes
- FFT, Viterbi, etc..
- Parallel comparisons
- GPU style
- Streaming operations
- 16/32-bit FP operations
- Texture management unit
Facts about "SHAVE v2.0 - Microarchitectures - Intel Movidius"
| codename | SHAVE + |
| designer | Movidius + |
| first launched | 2011 + |
| full page name | movidius/microarchitectures/shave v2.0 + |
| instance of | microarchitecture + |
| manufacturer | TSMC + |
| name | SHAVE + |