| 1 |
Am 10.08.2012 08:56, schrieb Jesús J. Guerrero Botella: |
| 2 |
> It could be anything. Maybe some orphaned process was running in the |
| 3 |
> background and leaking ram, or something. It's futile to speculate now |
| 4 |
> about that. |
| 5 |
> |
| 6 |
> Also, the -recently added- "pgo" USE flag could have something to do |
| 7 |
> with that. Not sure, since I didn't bother to investigate it's true |
| 8 |
> purpose on firefox. |
| 9 |
> |
| 10 |
|
| 11 |
It does two compilations with a headless firefox benchmark in between. |
| 12 |
Except of doubling the compilation time, there is little difference in |
| 13 |
the compilation itself. |
| 14 |
|
| 15 |
> I really don't think that the kernel has changed in a significant way |
| 16 |
> in this regard since the latests 2.6.x releases. But I certainly |
| 17 |
> didn't read *all* the kernel changelogs. |
| 18 |
> |
| 19 |
|
| 20 |
The latest thing of any significance I can think of is the removal of |
| 21 |
lumpy reclaim in 3.4 which has something to do with reducing memory |
| 22 |
fragmentation in systems under memory stress. LWN has a subscriber-only |
| 23 |
article about the change causing performance regressions. From my |
| 24 |
understanding of the code, I doubt it could cause an improvement in this |
| 25 |
particular situation. |
| 26 |
|
| 27 |
Regards, |
| 28 |
Florian Philipp |
Archives