Date: | Fri, 28 Aug 2009 11:50:37 -0500 |
---|---|
From: | Wael Kdouh <waelkdouh@xxxxxxxxx> |
Subject: | [Gems-users] Using "Opal0.sim-step X" command does not advance the simulation state |
Hi, I have posted this question before, but I haven't got any response yet. I am posting it again with the hope that this time someone will answer my question. Thank you in advance. Here is the question:
I managed to successfully run my benchmarks with Ruby enabled within Simics while disabling Opal. But when I enable opal I am getting confused regarding the synchronization between Simics cycles and Opal cycles. In other words, even though I run opal for billion of instructions, I never get the host machine to execute any command which as I mentioned before would have executed otherwise with Opal disabled.
Furthermore when I check the number of cycles advanced after for example advancing opal for 4_000_0000_0000 instructions using the following command "opal0.sim-step 4_000_0000_0000", I notice that Simics only advances a single instruction which I believe is behind the reason for the host machine not executing any instructions. Your help is highly appreciated. |
[← Prev in Thread] | Current Thread | [Next in Thread→] |
---|---|---|
|
Previous by Date: | [Gems-users] Solaris installation - ruby, niranjan soundararajan |
---|---|
Next by Date: | [Gems-users] question about GEMS and PARSEC, hamid |
Previous by Thread: | [Gems-users] Using "Opal0.sim-step 4_000_0000_0000" command does not advance the simulation state, Wael Kdouh |
Next by Thread: | [Gems-users] Wattch SIGSEGV on Opal init, Wael Kdouh |
Indexes: | [Date] [Thread] |