1 / 7

Jet-algorithm / FIO-tests

Jet-algorithm / FIO-tests. Attila Hidvégi Stockholm University. Jet-algorithm. Last time: The Jet-algorithm was running at 110 MHz. The Summation-algorithm was running at 42 MHz. After merging the two codes together the speed went down to 64/32 MHz. Jet-algorithm.

Download Presentation

Jet-algorithm / FIO-tests

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Jet-algorithm / FIO-tests Attila Hidvégi Stockholm University

  2. Jet-algorithm • Last time: • The Jet-algorithm was running at 110 MHz. • The Summation-algorithm was running at 42 MHz. • After merging the two codes together the speed went down to 64/32 MHz.

  3. Jet-algorithm • Understanding the problem: • Leonardo did not understood that there were two clocks, and did a bad optimization. • Place-and-Route could understand that but at that point it was already too late. • Leonardo’s GUI have several bugs.

  4. Jet-algorithm • Solving the problem: • Global optimization could reveal that the design were able to achieve delays of 6.7/17.8 ns. • Moving to scripts enabled us to get around bugs in the GUI and make settings which were not possible with it. • The commands for Leonardo were carefully studied, and finally the proper requirements were defined.

  5. Jet-algorithm • Standalone tests in a real FPGA: • Last time: • The bigger clustersizes did not seemed to work properly. • At 100 MHz we saw some timing problems. • Explanations: • The clustersizes was a minor software issue. • The timing problem is due to the long path from the BlockRAMs to the adders, in this particular chip.

  6. FIO-tests • A Debug version to the main-FPGA has been created to monitor all the inputs to the JEP. • A delay-scan has been done for several reasons. • Numbers of bits that are wrong are calculated for each deskew value.

  7. FIO-tests

More Related