| 1 |
"John R. Dunning" writes: |
| 2 |
> |
| 3 |
> We are working with cfs. That doesn't mean they're doing all our work for us |
| 4 |
> :-} |
| 5 |
> |
| 6 |
> Honestly, a big part of it is just plain old market sensitivity. Cfs is |
| 7 |
> paying attention to where their bread and butter is. So far, that's not |
| 8 |
> gentoo. Perhaps if sicortex is wildly successful we'll be able to change that |
| 9 |
> equation :-} |
| 10 |
> |
| 11 |
> ... |
| 12 |
> |
| 13 |
> So our approach has boiled down to |
| 14 |
> |
| 15 |
> 1. Stick close to vanilla |
| 16 |
> 2. Make mips work |
| 17 |
> 3. Do whatever we need to do to make lustre layer on top of that |
| 18 |
> |
| 19 |
> Based on what you've said, I wouldn't fool around with SLES, I'd just figure |
| 20 |
> out what close-to-vanilla kernel you want to start from (picking one you think |
| 21 |
> you can live with for a while) and do some part of what I described above. |
| 22 |
> You might have a somewhat easier time of it if you started with 2.6.18, as I |
| 23 |
> believe there's a cfs-supplied patchset for that one. |
| 24 |
|
| 25 |
Thank you for your extensive feedback. :) |
| 26 |
|
| 27 |
I think you've done a pretty good job of showing the complications involved in putting together |
| 28 |
your own lustre kernel. It sounds gnarly. What I take away from this is the need for a vanilla |
| 29 |
kernel patchset from CFS, preferably for 2.6.18 or higher. If there is already at least the |
| 30 |
basis for a vanilla 2.6.18 patchset in the current beta, that could be the starting point for an |
| 31 |
ebuild that would be of use for a while. |
| 32 |
|
| 33 |
Its been awhile since I last tried running Lustre. Perhaps it is time I tried building a 2.6.18 |
| 34 |
lustre-ized kernel. It depends on how good the patchset provided by CFS is. I don't have the |
| 35 |
bandwidth to go through the extensive process that you did to get a working kernel. |
| 36 |
|
| 37 |
-bryan |
| 38 |
|
| 39 |
-- |
| 40 |
gentoo-cluster@g.o mailing list |
Archives