Making a Case for an ARM Cortex-A9 CPU Interlay Replacing the NEON SIMD UnitCitation formats

Standard

Making a Case for an ARM Cortex-A9 CPU Interlay Replacing the NEON SIMD Unit. / Garcia Ordaz, Jose Raul; Koch, Dirk.

International Conference on Field-Programmable Logic and Applications. 2017. (2017 27th International Conference on Field Programmable Logic and Applications (FPL)).

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Harvard

Garcia Ordaz, JR & Koch, D 2017, Making a Case for an ARM Cortex-A9 CPU Interlay Replacing the NEON SIMD Unit. in International Conference on Field-Programmable Logic and Applications. 2017 27th International Conference on Field Programmable Logic and Applications (FPL), Field Programmable Logic and Applications, Ghent, Belgium, 4/09/17. https://doi.org/10.23919/FPL.2017.8056806

APA

Garcia Ordaz, J. R., & Koch, D. (2017). Making a Case for an ARM Cortex-A9 CPU Interlay Replacing the NEON SIMD Unit. In International Conference on Field-Programmable Logic and Applications (2017 27th International Conference on Field Programmable Logic and Applications (FPL)). https://doi.org/10.23919/FPL.2017.8056806

Vancouver

Garcia Ordaz JR, Koch D. Making a Case for an ARM Cortex-A9 CPU Interlay Replacing the NEON SIMD Unit. In International Conference on Field-Programmable Logic and Applications. 2017. (2017 27th International Conference on Field Programmable Logic and Applications (FPL)). https://doi.org/10.23919/FPL.2017.8056806

Author

Garcia Ordaz, Jose Raul ; Koch, Dirk. / Making a Case for an ARM Cortex-A9 CPU Interlay Replacing the NEON SIMD Unit. International Conference on Field-Programmable Logic and Applications. 2017. (2017 27th International Conference on Field Programmable Logic and Applications (FPL)).

Bibtex

@inproceedings{86a2e0d11806489a9596c3e84b7da3cb,
title = "Making a Case for an ARM Cortex-A9 CPU Interlay Replacing the NEON SIMD Unit",
abstract = "As an alternative of adding more and more in-structions to CPU cores in order to address a wide rangeof applications, this paper examines to use a mixed grainedCPU interlay fabric to provide reconfigurable instruction setextensions. In detail, we are examining to replace the hardenedNEON SIMD unit of an ARM Cortex-A9 with an identical sizedFPGA fabric. We show that by applying a set of optimizations, weare able to emulate original applications using NEON instructionsat the same hardware cost and at very little performance drop byan interlay. Moreover we are demonstrating examples where spe-cial custom instructions running on a CPU-Interlay-hybrid aresubstantially outperforming the original hardened CPU-NEON-system, hence making a strong case to embed reconfigurabilityas a beneficial feature in future processors.",
author = "{Garcia Ordaz}, {Jose Raul} and Dirk Koch",
year = "2017",
doi = "10.23919/FPL.2017.8056806",
language = "English",
series = "2017 27th International Conference on Field Programmable Logic and Applications (FPL)",
publisher = "IEEE",
booktitle = "International Conference on Field-Programmable Logic and Applications",
note = "Field Programmable Logic and Applications, FPL ; Conference date: 04-09-2017 Through 08-09-2017",
url = "https://www.fpl2017.org/",

}

RIS

TY - GEN

T1 - Making a Case for an ARM Cortex-A9 CPU Interlay Replacing the NEON SIMD Unit

AU - Garcia Ordaz, Jose Raul

AU - Koch, Dirk

N1 - Conference code: 2017

PY - 2017

Y1 - 2017

N2 - As an alternative of adding more and more in-structions to CPU cores in order to address a wide rangeof applications, this paper examines to use a mixed grainedCPU interlay fabric to provide reconfigurable instruction setextensions. In detail, we are examining to replace the hardenedNEON SIMD unit of an ARM Cortex-A9 with an identical sizedFPGA fabric. We show that by applying a set of optimizations, weare able to emulate original applications using NEON instructionsat the same hardware cost and at very little performance drop byan interlay. Moreover we are demonstrating examples where spe-cial custom instructions running on a CPU-Interlay-hybrid aresubstantially outperforming the original hardened CPU-NEON-system, hence making a strong case to embed reconfigurabilityas a beneficial feature in future processors.

AB - As an alternative of adding more and more in-structions to CPU cores in order to address a wide rangeof applications, this paper examines to use a mixed grainedCPU interlay fabric to provide reconfigurable instruction setextensions. In detail, we are examining to replace the hardenedNEON SIMD unit of an ARM Cortex-A9 with an identical sizedFPGA fabric. We show that by applying a set of optimizations, weare able to emulate original applications using NEON instructionsat the same hardware cost and at very little performance drop byan interlay. Moreover we are demonstrating examples where spe-cial custom instructions running on a CPU-Interlay-hybrid aresubstantially outperforming the original hardened CPU-NEON-system, hence making a strong case to embed reconfigurabilityas a beneficial feature in future processors.

U2 - 10.23919/FPL.2017.8056806

DO - 10.23919/FPL.2017.8056806

M3 - Conference contribution

T3 - 2017 27th International Conference on Field Programmable Logic and Applications (FPL)

BT - International Conference on Field-Programmable Logic and Applications

T2 - Field Programmable Logic and Applications

Y2 - 4 September 2017 through 8 September 2017

ER -