(Created page with " == Abstract == Making full use of the capabilities of the FPGA as an accelerator is difficult for non hardware experts, especially if partial reconfiguration is to be employ...") |
|||
(One intermediate revision by the same user not shown) | |||
Line 3: | Line 3: | ||
Making full use of the capabilities of the FPGA as an accelerator is difficult for non hardware experts, especially if partial reconfiguration is to be employed. One of the issues that arise is to physically implement modules into bounding boxes of minimum size for improving fragmentation cost and reconfiguration time. In this paper we present a method which automates the modules designing step, fulfilling module resource requirements and architectural FPGA constraints. We present a case study that shows how our automatic module implementation flow can be used to generate run-time reconfigurable bitstreams that are suited for stitching together processing pipelines directly from a Maxeler MaxJ HLS specification. This takes into consideration design alternatives, fragmentation, and routing failure mitigation strategies. | Making full use of the capabilities of the FPGA as an accelerator is difficult for non hardware experts, especially if partial reconfiguration is to be employed. One of the issues that arise is to physically implement modules into bounding boxes of minimum size for improving fragmentation cost and reconfiguration time. In this paper we present a method which automates the modules designing step, fulfilling module resource requirements and architectural FPGA constraints. We present a case study that shows how our automatic module implementation flow can be used to generate run-time reconfigurable bitstreams that are suited for stitching together processing pipelines directly from a Maxeler MaxJ HLS specification. This takes into consideration design alternatives, fragmentation, and routing failure mitigation strategies. | ||
− | |||
− | |||
− | |||
− | |||
− | |||
Line 17: | Line 12: | ||
* [https://www.research.manchester.ac.uk/portal/files/64823020/hls_enabled_partially.pdf https://www.research.manchester.ac.uk/portal/files/64823020/hls_enabled_partially.pdf] | * [https://www.research.manchester.ac.uk/portal/files/64823020/hls_enabled_partially.pdf https://www.research.manchester.ac.uk/portal/files/64823020/hls_enabled_partially.pdf] | ||
+ | |||
+ | * [http://link.springer.com/content/pdf/10.1007/978-3-319-77610-1_20 http://link.springer.com/content/pdf/10.1007/978-3-319-77610-1_20], | ||
+ | : [http://dx.doi.org/10.1007/978-3-319-77610-1_20 http://dx.doi.org/10.1007/978-3-319-77610-1_20] under the license http://www.springer.com/tdm | ||
+ | |||
+ | * [https://link.springer.com/chapter/10.1007/978-3-319-77610-1_20 https://link.springer.com/chapter/10.1007/978-3-319-77610-1_20], | ||
+ | : [https://www.research.manchester.ac.uk/portal/files/64823020/hls_enabled_partially.pdf https://www.research.manchester.ac.uk/portal/files/64823020/hls_enabled_partially.pdf], | ||
+ | : [https://dblp.uni-trier.de/db/conf/arcs/arcs2018.html#GrigoreKK18 https://dblp.uni-trier.de/db/conf/arcs/arcs2018.html#GrigoreKK18], | ||
+ | : [https://www.research.manchester.ac.uk/portal/en/publications/hls-enabled-partially-reconfigurable-module-implementation(83d3aced-2407-4b35-a56a-e199806e1874).html https://www.research.manchester.ac.uk/portal/en/publications/hls-enabled-partially-reconfigurable-module-implementation(83d3aced-2407-4b35-a56a-e199806e1874).html], | ||
+ | : [https://rd.springer.com/chapter/10.1007/978-3-319-77610-1_20 https://rd.springer.com/chapter/10.1007/978-3-319-77610-1_20], | ||
+ | : [https://academic.microsoft.com/#/detail/2786613774 https://academic.microsoft.com/#/detail/2786613774] | ||
+ | |||
+ | * [https://www.research.manchester.ac.uk/portal/en/publications/hls-enabled-partially-reconfigurable-module-implementation(83d3aced-2407-4b35-a56a-e199806e1874).html https://www.research.manchester.ac.uk/portal/en/publications/hls-enabled-partially-reconfigurable-module-implementation(83d3aced-2407-4b35-a56a-e199806e1874).html], | ||
+ | : [https://doi.org/10.1007/978-3-319-77610-1 https://doi.org/10.1007/978-3-319-77610-1], | ||
+ | : [https://pure.manchester.ac.uk/ws/files/64823020/hls_enabled_partially.pdf https://pure.manchester.ac.uk/ws/files/64823020/hls_enabled_partially.pdf] | ||
− | DOIS: 10.1007/978-3-319-77610- | + | DOIS: 10.1007/978-3-319-77610-1_20 10.1007/978-3-319-77610-1 |
Making full use of the capabilities of the FPGA as an accelerator is difficult for non hardware experts, especially if partial reconfiguration is to be employed. One of the issues that arise is to physically implement modules into bounding boxes of minimum size for improving fragmentation cost and reconfiguration time. In this paper we present a method which automates the modules designing step, fulfilling module resource requirements and architectural FPGA constraints. We present a case study that shows how our automatic module implementation flow can be used to generate run-time reconfigurable bitstreams that are suited for stitching together processing pipelines directly from a Maxeler MaxJ HLS specification. This takes into consideration design alternatives, fragmentation, and routing failure mitigation strategies.
The different versions of the original document can be found in:
DOIS: 10.1007/978-3-319-77610-1_20 10.1007/978-3-319-77610-1
Published on 01/01/2018
Volume 2018, 2018
DOI: 10.1007/978-3-319-77610-1
Licence: CC BY-NC-SA license
Are you one of the authors of this document?