Subject: Re: alternate rough SA patch (works on SMP)
To: Christian Limpach <firstname.lastname@example.org>
From: Stephan Uphoff <email@example.com>
Date: 06/28/2003 22:04:07
[cc line trimmed]
Christian Limpach wrote:
>> - as a target for mi_switch the lwp has to be in memory (l_flag & L_INMEM)
> I noticed these two since I sent the patch.
> Don't we also need the L_INMEM tests on UP?
> The cpu-queue wins over the
> global queue if both have lwp's at the same priority.
This can theoretically cause starvation of LWPs on the global queue.
> Processes with SA enabled have a very strong affinity
You seem to have reduced the SMP case to a UP case with
respect to one SA process.
This means that your patch inherits the UP problems of the current
SA code. (Generally triggered by low memory conditions)