Closed Bug 609943 Opened 14 years ago Closed 6 years ago

Selftests to stress memory barriers (fences)

Categories

(Tamarin Graveyard :: Virtual Machine, defect)

Product:
Tamarin Graveyard
Component:
Virtual Machine
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

(Not tracked)

Status:
RESOLVED WONTFIX
Milestone:
Future

People

(Reporter: siwilkin, Assigned: kpalacz)

Details

What does a VM selftest for memory barriers look like?

The one that I have tried in bug 555760 is to protect a shared counter with a Dekker-style lock that has been made sequentially consistent with memory barriers. The idea is that if the barriers are ok, then two threads can compete to update the counter n times each, so that the final counter value is 2n. If the final value is not 2n, then the barriers have failed to ensure sequential consistency.
This seems way too complicated though. We have to be confident in the algorithm before considering the barrier implementations, and I'm not convinced as yet. Is there something simpler?
Assignee: nobody → kpalacz
Flags: flashplayer-qrb+
Flags: flashplayer-injection-
Flags: flashplayer-bug-
Target Milestone: --- → Future
Tamarin is a dead project now. Mass WONTFIX.
Status: NEW → RESOLVED
Closed: 6 years ago
Resolution: --- → WONTFIX
Tamarin isn't maintained anymore. WONTFIX remaining bugs.
You need to log in before you can comment on or make changes to this bug.