Skip to content

GitLab

  • Projects
  • Groups
  • Snippets
  • Help
    • Loading...
  • Help
    • Help
    • Support
    • Community forum
    • Submit feedback
    • Contribute to GitLab
    • Switch to GitLab Next
  • Sign in / Register
eigen
eigen
  • Project overview
    • Project overview
    • Details
    • Activity
    • Releases
  • Repository
    • Repository
    • Files
    • Commits
    • Branches
    • Tags
    • Contributors
    • Graph
    • Compare
    • Locked Files
  • Issues 596
    • Issues 596
    • List
    • Boards
    • Labels
    • Service Desk
    • Milestones
    • Iterations
  • Custom Issue Tracker
    • Custom Issue Tracker
  • Merge Requests 21
    • Merge Requests 21
  • Requirements
    • Requirements
    • List
  • CI / CD
    • CI / CD
    • Pipelines
    • Jobs
    • Schedules
    • Test Cases
  • Security & Compliance
    • Security & Compliance
    • Dependency List
    • License Compliance
  • Operations
    • Operations
    • Incidents
    • Environments
  • Packages & Registries
    • Packages & Registries
    • Package Registry
    • Container Registry
  • Analytics
    • Analytics
    • CI / CD
    • Code Review
    • Insights
    • Issue
    • Repository
    • Value Stream
  • Snippets
    • Snippets
  • Members
    • Members
  • Collapse sidebar
  • Activity
  • Graph
  • Create a new issue
  • Jobs
  • Commits
  • Issue Boards
  • libeigen
  • eigeneigen
  • Issues
  • #1528

Closed
Open
Opened Dec 04, 2019 by Eigen Bugzilla@eigenbzMaintainer

Floating point underflow exception in tolerance computation

Submitted by Anders Sundman

Assigned to Nobody

Link to original bugzilla bug (#1528)
Version: 3.4 (development)
Operating system: Linux

Description

In src/Cholesky/LDLT.h there is the line:

RealScalar tolerance = RealScalar(1) / NumTraits<RealScalar>::highest();

This causes a floating point underflow exception in the CPU. It is normally not handled by applications and is silently ignored. But we run with:

feenableexcept(FE_DIVBYZERO | FE_OVERFLOW | FE_INVALID | FE_UNDERFLOW);

Using min() seems to work fine for us:

RealScalar tolerance = std::numeric_limits<RealScalar>::min();

Edited Dec 05, 2019 by Eigen Bugzilla
Assignee
Assign to
None
Milestone
None
Assign milestone
Time tracking
None
Due date
None
Reference: libeigen/eigen#1528