From: Patrick Kirsch
> Jenda Krynicky schrieb:
> > Why do you think weakref would help you? Do you have any cyclic data
> > structures that are not getting released? Other than memory leaks
> > like that, weakref is not gonna do anything about the memory
> > footprint. See Devel::Leak to find o
Jenda Krynicky schrieb:
> Why do you think weakref would help you? Do you have any cyclic data
> structures that are not getting released? Other than memory leaks
> like that, weakref is not gonna do anything about the memory
> footprint. See Devel::Leak to find out if you have leaks.
>
As th
zentara schrieb:
> Almost, the create/undef must be in the same scope ( block{} ).
> This works, however, it can't be relied upon.
>
> #!/usr/bin/perl -w
> use strict;
> $| = 1;
> print "$$\n"; #top -p $$
>
> {
> my $string;
> for ( 1 .. 10 ) {
> $string .= ( 'x' x 1000 );
>
From: Patrick Kirsch
> Yes, I do have a problem with that situation.
> There is an (huge) application written in Perl. It includes a couple of
> modules from CPAN and does analyzing stuff. As a result the memory
> footprint is (in my expectation) too high (I mean throwing hardware at
> it does not
Jay Savage schrieb:
> Hi Patrick,
>
> This is hardly a beginner question, and as such is a bit out of scope
> for this list. If you're really interested in Perl internals, you'll
> probably want to check out the alt.lang.perl.misc newsgroup.
>
Yes, you're right. I thought a couple of people did
Mr. Shawn H. Corey schrieb:
> I did not expect this behaviour either. I would think it would give
> none back or most of it; there would be some overhead.
>
> Question : Are you having a problem with memory or is this just to
> satisfy your curiosity?
>
Yes, I do have a problem with that situa
> This is hardly a beginner question, and as such is a bit out of scope
> for this list. If you're really interested in Perl internals, you'll
> probably want to check out the alt.lang.perl.misc newsgroup.
comp.lang.perl.misc might be better. I think Randal S is the only one
to look at the alt
On Mon, 2008-12-22 at 15:02 -0500, Jay Savage wrote:
> If you're really interested in Perl internals, you'll
> probably want to check out the alt.lang.perl.misc newsgroup.
Perl.org maintains a number of mailings list; one of the perl-porters
http://lists.cpan.org/showlist.cgi?name=perl5-porters is
On Mon, Dec 22, 2008 at 12:36 PM, Patrick Kirsch
wrote:
> Mr. Shawn H. Corey schrieb:
>> On Mon, 2008-12-22 at 18:10 +0100, Patrick Kirsch wrote:
>>
>>> Is there a possibility to influence it, to free memory (in the sense
>>> of
>>> give it back to the OS)?
>>>
>>>
>>
>> Does your OS have a functi
On Mon, 2008-12-22 at 18:36 +0100, Patrick Kirsch wrote:
> I'm using: Linux 2.6.27.6 #2 SMP x86_64; Distribution: OpenSUSE 11.0 .
> In my opinion the Linux kernel is/should be able to rearrange freed
> memory between processes (I think e.g. of the slab).
>
> But let me show another example:
> $foo
Mr. Shawn H. Corey schrieb:
> On Mon, 2008-12-22 at 18:10 +0100, Patrick Kirsch wrote:
>
>> Is there a possibility to influence it, to free memory (in the sense
>> of
>> give it back to the OS)?
>>
>>
>
> Does your OS have a function that allows processes to return memory to
> it? Many do
On Mon, 2008-12-22 at 18:10 +0100, Patrick Kirsch wrote:
> Is there a possibility to influence it, to free memory (in the sense
> of
> give it back to the OS)?
>
Does your OS have a function that allows processes to return memory to
it? Many do not. Most of the time, when a process frees memory
12 matches
Mail list logo
