> On Mon, 24 Apr 2000, Jakob Østergaard wrote:
> >
> > Dual PPro stepping 1 here. I was just finishing my ``success report'' as the
> > program locked up :(
>
> Good. Not because I like the result, but because this finally puts the
> thing to rest with not a shadow of a doubt.
Sorry, I'm afraid it doesn't. Manfred's test was assuming a model even
more strict than Intel's "processor ordering". Unfortunately, the code in
question is a little confusing. Hopefully people can agree that my new
code works as advertised and does indeed test the issue in question.
Thing is, if my test code breaks, it almost certainly indicates big
problems with the "processor ordering" model so we're liable to have races
in other things we thought were safe.
> The question is really only how the config option should be presented to
> the user (a "generic" kernel would obviously have the CONFIG_SLOW_UNLOCK
> thing enabled, but should be for example just split the "PPro" config
> entry into "PPro" and "PII", and then for "PPro" have a furter "Slow but
> safe" option?)
I hope that if a bug exists we can narrow it down to a stepping level.
Early PPros are quite rare, SMP systems more so, but I've already seen
results from the last test for 3 of the 5 known stepping levels.
Hopefully, they'll test again.
-- "Love the dolphins," she advised him. "Write by W.A.S.T.E.."
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/