I've been testing out the new 3.5 version of POV-Ray
on my laptop that only has 192M RAM (Win98SE).

On a longer render, the memory usage profile has demonstrated some
of the classic symptoms of memory leak typical of C++ programs that
have dangling pointer problems when using dynamically allocated
memory in function calls.

As the render goes on:
1] reduced unused physical memory
2] increased page faults
3] increases swap file in use size
etc.

I have a scene-file to show anyone interested!

Thanks

-Ed

--
Edward Leibnitz

lei### [at] interportnet

Post a reply to this message

"Edward Leibnitz" <lei### [at] rcncom> wrote in news:3d40b6e5@news.povray.org

> I've been testing out the new 3.5 version of POV-Ray
> on my laptop that only has 192M RAM (Win98SE).

> As the render goes on:
> 1] reduced unused physical memory
> 2] increased page faults
> 3] increases swap file in use size

this is a "future" od Win98... maybe ther is a bug in addition, but Win98 
realy likes to lost memory - there are some utilitis to fix this - see my 
question on alt.windows98

-- 
#macro g(U,V)(.4*abs(sin(9*sqrt(pow(x-U,2)+pow(y-V,2))))*pow(1-min(1,(sqrt(
pow(x-U,2)+pow(y-V,2))*.3)),2)+.9)#end#macro p(c)#if(c>1)#local l=mod(c,100
);g(2*div(l,10)-8,2*mod(l,10)-8)*p(div(c,100))#else 1#end#end light_source{
y 2}sphere{z*20 9pigment{function{p(26252423)*p(36455644)*p(66656463)}}}//M

Post a reply to this message

Edward Leibnitz <lei### [at] rcncom> wrote:
> On a longer render, the memory usage profile has demonstrated some
> of the classic symptoms of memory leak typical of C++ programs that
> have dangling pointer problems when using dynamically allocated
> memory in function calls.

  Wow! You were able to deduce the exact cause for the memory leak just
by using a memory usage profile? Amazing.

-- 
#macro N(D)#if(D>99)cylinder{M()#local D=div(D,104);M().5,2pigment{rgb M()}}
N(D)#end#end#macro M()<mod(D,13)-6mod(div(D,13)8)-3,10>#end blob{
N(11117333955)N(4254934330)N(3900569407)N(7382340)N(3358)N(970)}//  - Warp -

Post a reply to this message

"Edward Leibnitz" <lei### [at] rcncom> wrote:
> I've been testing out the new 3.5 version of POV-Ray
> on my laptop that only has 192M RAM (Win98SE).
>
> On a longer render, the memory usage profile has demonstrated some
> of the classic symptoms of memory leak typical of C++ programs that
> have dangling pointer problems when using dynamically allocated
> memory in function calls.
>
> As the render goes on:
> 1] reduced unused physical memory
> 2] increased page faults
> 3] increases swap file in use size
> etc.

IMO, this shows more a big memory usage, which is normal for a ray-tracer,
more than memory leak. BTW, PoV-Ray 3.5 is written in C, not in C++ (that's
for 4.0, AFAIK).
Real leak testing should be done with appropriate tools, like BoundsChecker
or Purify, or one of the numerous open source products, like mpatrol.
And I assume (hope) that PoV-Ray coders has done already such tests...

-- #=--=#=--=#=--=#=--=#=--=#=--=#=--=#=--=# --
Philippe Lhoste (Paris -- France)
Professional programmer and amateur artist
http://jove.prohosting.com/~philho/

Post a reply to this message

>   Wow! You were able to deduce the exact cause for the memory leak
> just by using a memory usage profile? Amazing.

Well, you know what they say, Warp: there's always someone better. ;)

Post a reply to this message

In article <3d40b6e5@news.povray.org>,
 "Edward Leibnitz" <lei### [at] rcncom> wrote:

> As the render goes on:
> 1] reduced unused physical memory
> 2] increased page faults
> 3] increases swap file in use size
> etc.
> 
> I have a scene-file to show anyone interested!

Those are just symptoms of a program that is using lots of memory. POV 
could easily qualify for this. Some things allocate additional memory 
during rendering, so this isn't surprising at all.

-- 
Christopher James Huff <chr### [at] maccom>
POV-Ray TAG e-mail: chr### [at] tagpovrayorg
TAG web site: http://tag.povray.org/

Post a reply to this message

In article <3d412a9c@news.povray.org>,
 "Philippe Lhoste" <Phi### [at] GMXnet> wrote:

> IMO, this shows more a big memory usage, which is normal for a ray-tracer,
> more than memory leak. BTW, PoV-Ray 3.5 is written in C, not in C++ (that's
> for 4.0, AFAIK).

Actually, it uses some C++ features, so it requires a C++ compiler.

-- 
Christopher James Huff <chr### [at] maccom>
POV-Ray TAG e-mail: chr### [at] tagpovrayorg
TAG web site: http://tag.povray.org/

Post a reply to this message