First of all thank you all for your precious support! :-)
The two machines on which I discovered the problem are now in
production and I cannot for the moment make tests, but I hope to have
some other hardware to try in the next week.
I'll let you know ASAP.
Thanks,
Luca
Yes, the information in that thread is the same as what Luca said. It's
a load average problem, not a CPU utilisation problem (those threads are
sleeping!) If Luca could test the msleep_interruptible() version of the
patch and it works (like I said, performance should not change, but the
load average will drop to by 2), then I will ACK the patch for mainline
acceptance.
Thanks,
Nish