BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260202T201804Z
LOCATION:261
DTSTART;TZID=America/Chicago:20251117T103000
DTEND;TZID=America/Chicago:20251117T105500
UID:submissions.supercomputing.org_SC25_sess214_ws_corr101@linklings.com
SUMMARY:Using Code Coverage to Assess Feature Gaps in MPI Correctness Tool
  Classification Tests
DESCRIPTION:Alexander Hück (Technical University of Darmstadt), Simon Schw
 itanski (RWTH Aachen University), Tim Jammer (Technical University of Darm
 stadt), Joachim Jenke (RWTH Aachen University), and Yussur Mustafa Oraji a
 nd Christian Bischof (Technical University of Darmstadt)\n\nWe examine the
  code generator-based MPI correctness benchmark MPI-BugBench (MBB) by anal
 yzing the code coverage it triggers in three tools: MUST, PARCOACH, and cl
 ang-tidy. Our analysis complements MBB’s design, which prunes potentially 
 exhaustive test sets based on real-world MPI usage. Our assessment identif
 ies two key limitations in the generated tests: incomplete coverage of MPI
  features, such\nas varying-count collectives, and limited structural dive
 rsity of the generated tests, such as lack of loops and lack of array-base
 d MPI handles. We find increasing test volume alone offers limited benefit
  for exercising the tool's analysis code in our assessment.\nTo address th
 ese gaps, we propose a new generation level with the missing features and 
 more varied code structures. To that end, we implemented 34 additional tes
 ts to exercise previously uncovered analysis code, adding as many as 770 l
 ines of code coverage in MUST with a single test for varying-count collect
 ives.\n\nTag: Debugging & Correctness Tools, HPC Software & Runtime System
 s\n\nRecording: Livestreamed, Recorded\n\nRegistration Category: Technical
  Program Reg Pass, Workshop Reg Pass\n\nIn-Person / Remote: In-person\n\nS
 ession Chairs: Ignacio Laguna (Lawrence Livermore National Laboratory (LLN
 L)) and Cindy Rubio-González (University of California, Davis)\n\n
END:VEVENT
END:VCALENDAR
