Project

General

Profile

Bug #680

DistrMPolyClean does not use MemPool for summands?

Added by John Abbott about 9 years ago. Updated about 1 year ago.

Status:
Resolved
Priority:
Normal
Assignee:
John Abbott
Category:
Tidying
Target version:
CoCoALib-0.99880
Start date:
13 Apr 2015
Due date:
% Done:

70%

Estimated time:
Spent time:
3.30 h

Description

It seems that DistrMPolyClean does not use MemPool for summands even though a suitable MemPool as passed as argument to the ctor.

Related issues

Related to CoCoALib - Slug #329: Why is DMPI slower than DMPClean?New2013-03-12

Related to CoCoALib - Support #791: Clean code for DistrMPolyCleanNew2015-10-29

History

#1 Updated by John Abbott about 9 years ago

Why is the MemPool object which is given to DistrMPoly ignored?

MemPool logging shows that the summand pool is not used.

#2 Updated by John Abbott over 8 years ago

  • Status changed from New to In Progress
  • % Done changed from 0 to 10

I have mimicked the idea of NewSummandPtr from DistrMPolyInlFpPP. It all compiles, and the tests pass (miracle?).

I was a bit surprised to find that DistrMPolyClean is not as clean as I had expected.

Now I have to check whether MemPool is really being used for the summands...

#3 Updated by Anna Maria Bigatti over 8 years ago

John Abbott wrote:

I was a bit surprised to find that DistrMPolyClean is not as clean as I had expected.

oops. I do remember that I noticed that summand are treated differently in the DistrMPolyXXX implementations, and I did think of comparing them and unify them. But one always finds good excuses not to do dangerous work on pointers ;-)

#4 Updated by John Abbott over 8 years ago

  • Status changed from In Progress to Resolved
  • Assignee set to John Abbott
  • % Done changed from 10 to 70

I have now revised the code so that it uses its MemPool for summands; essentially I had to "steal" the idea of NewSummandPtr from DistrMPolyInlFpPP. After some SEGV excitement, it now seems to work.

I wonder if changing to MemPool will make DMPClean slower than DMPI -- see #329.

I have cleaned my changes, but there is still a lot of cruft.

#5 Updated by John Abbott over 3 years ago

  • Target version changed from CoCoALib-1.0 to CoCoALib-0.99850

#6 Updated by John Abbott about 1 year ago

  • Target version changed from CoCoALib-0.99850 to CoCoALib-0.99880

Also available in: Atom PDF