New thresholds in table

[Niels Möller]

Torbjorn Granlund <tg at gmplib.org> writes:

> When there is no value, it is absent because it would not make sense.
> When there is a value, it is from real hardware.

Ok, I see. 

> (In a few cases, we have suppressed tunuep since there are timing
> problems in the kernel Linux.  We then run tests on similar hardware
> using a kernel without such problems.)

On my linux boxes, running tuneup with -o cycles-broken is about as
reliable as it can be. Might still work badly on old versions or on
certain hardware, I guess.

Regards,
/Niels

-- 
Niels Möller. PGP-encrypted email is preferred. Keyid C0B98E26.
Internet email is subject to wholesale government surveillance.