Project

General

Profile

Bug #7689

lattice_simulator.update() leaks memory when the underlying lattice is from a madX file

Added by Eric Stern over 4 years ago. Updated over 4 years ago.

Status:
Closed
Priority:
Normal
Assignee:
-
Start date:
01/28/2015
Due date:
% Done:

0%

Estimated time:
Duration:

Description

repeatedly call stepper.get_lattice_simulator().update() with lattice generated by reading fodo.lat => no memory leak
repeatedly call stepper.get_lattice_simulator().update() with lattice generated by reading fodo.madx => memory leak

History

#1 Updated by Eric Stern over 4 years ago

The problem occurs with quadrupoles and the yoshida propagator. There is no memory leak when the propagator_type is set to basic.

#2 Updated by Eric Stern over 4 years ago

  • Status changed from New to Resolved

The problem was resolved by CHEF commit:
commit:3d51ff75f5c1150fbca048c5a1ffe7ea4571f319

    BUG FIX: memory leak. Removed BmlPtr my_beamline_ptr_ from private data ...

    Removed BmlPtr my_beamline_ptr_ from YoshidaPropagator's
    private data. Combined with lack of virtual destructors
    for base propagators, it resulted in a memory leak.
    (cherry picked from commit 91c7b1cff9fa678f0521fd5557d6ba54f0c7a7c3)
    (cherry picked from commit b906dfcd7a68418d67646df2007986798ba5ec6c)

#3 Updated by Eric Stern over 4 years ago

  • Status changed from Resolved to Closed


Also available in: Atom PDF