[ofa-general] [PATCH] opensm: Parallelize (Stripe) LFT sets across switches

[Hal Rosenstock]

On Thu, Jul 23, 2009 at 6:14 PM, Jason
Gunthorpe<jgunthorpe at obsidianresearch.com> wrote:
> On Thu, Jul 23, 2009 at 05:43:38PM -0400, Hal Rosenstock wrote:
>
>> The proposed algorithm is less aggressive in terms of sending DR SMPs
>> than the current one which seems to work well enough based on field
>> experience. If the concern with this approach is breaking things over
>> the current approach, I suppose a config variable could be added for
>
> No, it just wasn't clear to me what the substantive difference was
> over the current approach. Reading your email made it seem like it was
> more agressive not less.

Sorry for the confusion. The proposed algorithm is less aggressive as
it limits things based on the max_smps_per_node which defaults to 4
whereas the current algorithm in the tree will send all blocks.

> Anyhow, I had no particularly strong objection

Glad to hear this.

> just an observation that this could be done much better.
>
> Jason

-- Hal