possible race condition in post-wait-start-complete
Marc Snir (snir@watson.ibm.com)
Sat, 17 Aug 1996 22:37:53 -0400
Rolf, thanks for pointing out that my definitions
for 2-party synchronization are incomplete. I changed
the text to make clear that calls must be interleaved:
a window is available to a "start" call if it is posted, but the same post
cannot satisfy two starts at the same process. I.e., the
post must come after
the wait that matched any previous complete by the same caller. The
implementation must enforce the interleaving, either using a hand-shake
protocol, or using generation counters. I hope that the text to be put out by
Dave next week makes this clear. Please look at it and let me know if it
still has problems.