| 1 |
Andy Dustman said: |
| 2 |
> On Sun, 2004-02-29 at 01:30, Sancho2k.net Lists wrote: |
| 3 |
>> We have a range of hardware from Pentium3 to Pentium4, some 2x CPU, most |
| 4 |
>> one CPU though. All common packages such as apache2, mysql, php/mod_php, |
| 5 |
>> and the like would be built there first and then distributed to the |
| 6 |
>> servers as needed (they would emerge the tbz2s from thier nfs-mount |
| 7 |
>> /usr/portage/packages/All directory). Since we are using -mcpu pentium3 |
| 8 |
>> in the make.conf, our binaries should be portable across p3 and p4 |
| 9 |
>> platforms, correct? |
| 10 |
> |
| 11 |
> Correct, but you would be better off using -march=pentium3, which |
| 12 |
> implies -mcpu=pentium3. If you tend to have mostly Pentium 4s, then you |
| 13 |
> can use -march=pentium3 -mcpu=pentium4, which causes it to only generate |
| 14 |
> instructions for pentium3, but optimize for pentium4. The Pentium IIIs |
| 15 |
> might take a slight performance hit in this case. |
| 16 |
|
| 17 |
Also along these lines: we're trying to develop an SOP document detailing |
| 18 |
how to bootstrap one of these systems prior to fleshing it out from the |
| 19 |
master server. For the same given group of P3s and P4s, would there be |
| 20 |
problems booting these systems from the same Pentium3 Live CD? |
| 21 |
|
| 22 |
DS |
Archives