On 07/04/2012 02:23 AM, John Stultz wrote:

On 07/03/2012 11:21 PM, John Stultz wrote:

quoted

Ok, made a few tweaks to address issues caught by Prarit's and my
testing. This has run for a number of hours now w/ my leap-a-day.c
test on a few machines.

I'd really appreciate any extra testing, review, or acks at this point.
I'm targeting mid-late Thursday (to give folks in the US a chance to
review & test) as a point when I'll submit this upstream if no other
issues are found.

And again, here's my leap-a-day.c test case which can be used to trigger a
leapsecond every day, or every ~13 seconds via the -s option.

I tested this patchset for ~22 hours without any issues using my own leap second
test code, John's previous leaptest.c and this new leap-a-day test with and
without the -s option.

In all x86 systems' cases (both AMD and Intel, large cpu count[1] and small cpu
count) I also ran Firefox in the background and monitored it using top and
occasionally breaking in to check the status of the futexes.  I haven't seen
anything wrong with any of the systems.

John, my apologies for not getting to exactly 24 hours.  I've usurped these
systems long enough and other people need them :/.  I'm going to continue to run
two systems (one AMD and one Intel) for a longer stretch but I don't anticipate
any issues at this point.

So ...

Acked-by: Prarit Bhargava <redacted>

For the entire patchset.

P.