| 1 |
Hi, |
| 2 |
|
| 3 |
On Wed, 19 Oct 2005 12:57:02 -0200 |
| 4 |
Bruno Lustosa <bruno.lists@×××××.com> wrote: |
| 5 |
|
| 6 |
> Just had a look at 'top' here, and was astonished by its output: |
| 7 |
> |
| 8 |
> PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND |
| 9 |
> 19328 isabel.s 15 0 511m 204m 16m S 0.0 27.2 0:02.47 firefox-bin |
| 10 |
> [x4] |
| 11 |
> 22668 lustosa 15 0 129m 97m 17m S 0.0 13.0 11:50.05 firefox-bin |
| 12 |
> [x2] |
| 13 |
> |
| 14 |
> [...] |
| 15 |
> I can't understand how firefox evolved from small and fast phoenix to this |
| 16 |
> memory hungry beast that has a virtual space of half a gigabyte. |
| 17 |
|
| 18 |
Hm, on the hand: yes, it _is_ memory hungry. On the other hand: You've |
| 19 |
already said: virtual address space. And as I interpret the above |
| 20 |
numbers, a lot of things is most probably shared among those processes. |
| 21 |
You can find out e.g. by comparing their /proc/<PID>/maps. |
| 22 |
|
| 23 |
> I mean... even IE doesn't seem to eat all this memory. In fact, I have seen |
| 24 |
> very few processes eating that much memory. |
| 25 |
|
| 26 |
You mean that tiny frontend to the OS' functionality? It certainly |
| 27 |
doesn't need much memory, no. But remember, other Browsers have to |
| 28 |
implement a lot of this stuff themselves and not rely on the |
| 29 |
functionality that specific OS and its API offers. |
| 30 |
|
| 31 |
-hwh |
| 32 |
-- |
| 33 |
gentoo-user@g.o mailing list |
Archives