9fans archive / 1997 / 04 / 67 /    prev next

From: miller@ham... miller@ham...
Subject: porting linux programs and drivers to plan9
Date: Mon, 21 Apr 1997 13:50:26 BST

presotto@pla... writes:

> [a fascinating account of how the Pentium Pro's out-of-order
> instruction execution breaks the Plan 9 sleep/wakeup code on
> a multi-CPU system]

It appears that the slightly different version of sleep/wakeup
given in the Volume 2 paper `Process Sleep and Wakeup on a
Shared-memory Multiprocessor' should be immune to the effects
of weak memory coherency, because the shared variables are
referenced only inside a lock/unlock pair.  Is this right?

Perhaps the moral is that it's better to be conservative with
locks than to trust hardware designers to do what we expect.

-- Richard Miller