Conference paper Open Access

About the functional test of the GPGPU scheduler

B. Du; Josie E. Rodriguez Condia; M. Sonza Reorda; L. Sterpone

DataCite XML Export

<?xml version='1.0' encoding='utf-8'?>
<resource xmlns:xsi="" xmlns="" xsi:schemaLocation="">
  <identifier identifierType="URL"></identifier>
      <creatorName>B. Du</creatorName>
      <affiliation>Politecnico di Torino</affiliation>
      <creatorName>Josie E. Rodriguez Condia</creatorName>
      <affiliation>Politecnico di Torino</affiliation>
      <creatorName>M. Sonza Reorda</creatorName>
      <affiliation>Politecnico di Torino</affiliation>
      <creatorName>L. Sterpone</creatorName>
      <affiliation>Politecnico di Torino</affiliation>
    <title>About the functional test of the GPGPU scheduler</title>
    <date dateType="Issued">2019-07-04</date>
  <resourceType resourceTypeGeneral="ConferencePaper"/>
    <alternateIdentifier alternateIdentifierType="url"></alternateIdentifier>
    <relatedIdentifier relatedIdentifierType="DOI" relationType="IsIdenticalTo">10.1109/IOLTS.2018.8474174</relatedIdentifier>
    <relatedIdentifier relatedIdentifierType="DOI" relationType="IsIdenticalTo">10.1109/IOLTS.2018.8474174</relatedIdentifier>
    <rights rightsURI="">Creative Commons Attribution 4.0 International</rights>
    <rights rightsURI="info:eu-repo/semantics/openAccess">Open Access</rights>
    <description descriptionType="Abstract">&lt;p&gt;General Purpose Graphical Processing Units (GPGPUs) are increasingly used in safety critical applicationssuch as the automotive ones. Hence, techniques are required to test them during the operational phase with respect to possible permanent faults arising when the device is already deployed in the field.&amp;nbsp;Functional tests adopting&amp;nbsp;Software-based Self-test(, or&amp;nbsp;SBST)are an effective solution since they providebenefits in terms of intrusiveness, flexibility and test duration. While the development of the functional test code addressing the several computational cores composing a GPGPU can be done resorting to known&amp;nbsp;&amp;nbsp;&amp;nbsp;methods developed for CPUs, for other modules which are typical of a GPGPU we still miss effective solutions.&amp;nbsp;This paper focuses&amp;nbsp;Oonone of the most relevant module consists on the scheduler core&amp;nbsp;which is in charge of managing different scalar computational cores and the different executed threads.&amp;nbsp;At first, we propose a method for evaluating the fault coverage that can be achieved using an application program. Then, we providesome guidelines for improving the achieved fault coverage. Experimental results are provided on an open-source VHDL model of a GPGPU.&lt;/p&gt;</description>
Views 33
Downloads 79
Data volume 62.0 MB
Unique views 31
Unique downloads 77


Cite as