| 1 |
On 24 Jan 2009 at 10:48, Grant wrote: |
| 2 |
|
| 3 |
> >> Nope, you guys are absolutely right. It falls back to peMRS whether |
| 4 |
> >> or not I enable PAGEEXEC since I don't have the nx flag. |
| 5 |
> > |
| 6 |
> > ok, so coming back to your original problem, are you saying that you |
| 7 |
> > had an observable slowdown due to SEGMEXEC? if so, i'd like to see some |
| 8 |
> > numbers and think about it, if you have some time. |
| 9 |
> |
| 10 |
> I thought there was a slowdown going from no grsecurity to grsecurity |
| 11 |
> "Gentoo (server)", but now I'm not so sure. I'll keep an eye on it |
| 12 |
> for sure. Would you consider an external http monitor's response |
| 13 |
> times over 24 hours a good indicator of whether or not I should |
| 14 |
> investigate further and do real benchmarking? |
| 15 |
|
| 16 |
well, i thought you'd tell me what made you think that things became |
| 17 |
slower than before ;). however if there's a slowdown, i doubt timing |
| 18 |
over a network would be the best way to find out, i'd prefer direct |
| 19 |
measurements on the affected host. so whenever you feel like it's |
| 20 |
still worth doing it... ;) |
Archives