konqy preloading kills performance

[Thiago Macieira]

Alexander Neundorf wrote:
>Well, as soon as it is swapped in again (because I want to visit a
> webpage) it has an impact again.
>
>Koos wrote:
>> Wouldn't it make more sense to vote for fixing those memory leaks (and
>> weren't those fixed for 3.3 already)?
>
>I don't think these are memory leaks. It's just the memory used by the
> process which isn't given back to the OS until the process exits I
> think.

I don't think so.

Konqueror should properly free the memory that it is no longer using when 
tabs and windows close. That should in turn cause some of that memory to 
be returned to the OS -- not all of it because it can only be returned in 
pagefuls. Some malloc/free methods will, however, keep the entire pool 
and not return a bit.

in any event, the swap-in process should not swap back in the whole memory 
footprint of a sudden. Only what is needed will be swapped back in. And 
if the memory hasn't been returned to the OS, it'll simple be swapped in 
and used, and the process should not grow more.

That is, however, an ideal world...

As far as I know, Konqueror should track its own memory usage and not keep 
itself preloaded if it has grown too big. If it's not happening for you, 
then either the values are too high or instead the reporting procedures 
are underreporting the usage.

-- 
  Thiago Macieira  -  Registered Linux user #65028
   thiago (AT) macieira (DOT) info
    ICQ UIN: 1967141   PGP/GPG: 0x6EF45358; fingerprint:
    E067 918B B660 DBD1 105C  966C 33F5 F005 6EF4 5358

Hwonne seo worulde wæs geong and time_t wæs 0, God gæf þæt Unix þære 
worulde. And æfter se God logode in, lá! þæt Unix cwæð "Hello, World" and 
ǽghwilc wæs glæd.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <https://mail.kde.org/mailman/private/kfm-devel/attachments/20041213/e7a31f27/attachment.sig>