Date: | Mon, 28 Jul 2008 16:05:02 +0200 |
---|---|
From: | "Rubén Titos" <rtitos@xxxxxxxxxxx> |
Subject: | [Gems-users] Bug in SimicsProcessor.c |
Apologies for the irrelevant subject of my previous email! :) I thought I'd send it again, to make it easier for others to search the list ------------------ Dear Dan, I just run into a simple bug that remains silent most of the time, but appears when running transactional apps if you previously compile without the -DNO_VECTOR_BOUNDS_CHECKS flag. The error is obvious (in SimicsProcessor.C): When the outcome of the memory transaction is 2, the active request vector is cleared and right after such vector is accessed to set the status (see code fragment below). Cheers, Ruben ----------------------- $GEMS/ruby/simics/SimicsProcessor.C:364 if (XACT_MEMORY){ ... } else if (outcome == 2) { clearActiveRequestVector(); m_active_requests[idx].status = Retry; } else { ... } } -- Rubén Titos Parallel Computing and Architecture Group Computer Engineering Dept. University of Murcia http://skywalker.inf.um.es/~rtitos/ -- Rubén Titos Parallel Computing and Architecture Group Computer Engineering Dept. University of Murcia http://skywalker.inf.um.es/~rtitos/ |
[← Prev in Thread] | Current Thread | [Next in Thread→] |
---|---|---|
|
Previous by Date: | Re: [Gems-users] How to abort a transaction at commit time, Rubén Titos |
---|---|
Next by Date: | Re: [Gems-users] retry hack in ATMTP/Rock, Daniel Nussbaum - Sun Microsystems - Burlington United States |
Previous by Thread: | Re: [Gems-users] Availability of Opensolaris / LogTM changes?, Haris Volos |
Next by Thread: | [Gems-users] control-c not halting simulation in Simics 3, Philip Garcia |
Indexes: | [Date] [Thread] |