Re: [Texmacs-dev] Scrolling speed

[David MENTRE]

Hello Joris,

Joris van der Hoeven <address@hidden> writes:

>> Joris van der Hoeven <address@hidden> writes:

>> Is such an article in TeXmacs format is available somewhere?
>
> You may take one of the longer articles from the online examples.
> Help -> Online help -> Browse web -> etc.

Ok, thanks.

>> > then scrolling is completely fluid. However, if I work for a while,
>> > start a browser, launch a few compilations, etc. and then restart
>> > TeXmacs, the scrolling becomes much slower.
>>
>> Yes, but after working for a while, is the scrolling getting better (I
>> would say yes)?
>
> No.

Quite strange. I share your perplexity.

>> Caches are exploiting spatial and temporal locality. Caches take into
>> account physical memory accesses and do not consider how this physical
>> memory is distributed into the virtual address space of the
>> program. So malloc has little to do for cache locality.
>
> Do you have a good reference on how modern memory works more precisely?

No, not at hand. I've learned those things about 10 years ago.

A well known reference is Computer Architecture: A Quantitative Approach
by John L. Hennessy, David A. Patterson, David Goldberg. I've never read
it myself. It is one those very big american book. Do not buy it, take
it at a library. Only first chapters should interest you.

Otherwise, Tannenbaum book on computer architecture should be fairly
correct. 

> OK, but the strange thing here is that the performance depends a lot
> on whether some other processes are or have been running or not.

As first explanation, I would have said that the other applications would
have trashed your cache. But if you work exclusively in TeXmacs and the
performance does not return to the initial state, then something else is
hapenning. This is quite strange. 

Or you have a threshold effect: at the biginning TeXmacs working set
fits in the cache but it grows above cache limits and you observe cache
trashing.


> This is weird; maybe it has something to do with X?

Maybe, but I can't help on X. I know nothing on the X system.


Yours,
d.
-- 
 David Mentré <address@hidden>