octave-bug-tracker
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Octave-bug-tracker] [bug #52980] clear in the caller workspace doesn't


From: anonymous
Subject: [Octave-bug-tracker] [bug #52980] clear in the caller workspace doesn't clear all references to a variable.
Date: Fri, 26 Jan 2018 14:00:12 -0500 (EST)
User-agent: Mozilla/5.0 (Windows NT 5.1; rv:51.0) Gecko/20100101 Firefox/51.0

URL:
  <http://savannah.gnu.org/bugs/?52980>

                 Summary: clear in the caller workspace doesn't clear all
references to a variable.
                 Project: GNU Octave
            Submitted by: None
            Submitted on: Fri 26 Jan 2018 07:00:11 PM UTC
                Category: Performance
                Severity: 3 - Normal
                Priority: 5 - Normal
              Item Group: Performance
                  Status: None
             Assigned to: None
         Originator Name: Rahnema
        Originator Email: 
             Open/Closed: Open
         Discussion Lock: Any
                 Release: 4.2.1
        Operating System: GNU/Linux

    _______________________________________________________

Details:

Hi all!
Comparing the following example:

B = rand(5000);
B(1) = 7;

with this one:

function f(A)
    evalin('caller', 'clear V')
    A(1) = 5;
end
V = rand(5000);
f(V);

I can use _time_ to get the maximum memory consumption of a process:

/usr/bin/time -f "%M" octave -q --eval "V=rand(5000);function
f(A);evalin('caller', 'clear V');A(1)=5;end;f(V);"

Interestingly the second example consumes two times more than the first
example.
When 'V' is cleared from the caller workspace we should have only one instance
of the 5000*5000 array in the function scope but when the local 'A' is
modified a copy is created! It seems that 'clear' can't clear all references
to the array object in the caller workspace. 

Best regards,
Rahnema.




    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?52980>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/




reply via email to

[Prev in Thread] Current Thread [Next in Thread]