Marc, I think if you have LQEntries and SQEntries as 1, then the O3
processor should behave almost like in order processor.
--
Nilay
On Fri, 29 Apr 2011, Marc Orr wrote:
Thanks for your responses!
Yes I did use the most recent version of the code.
I guess I also forgot to mention, changing LQEntries=1 and SQEntries=1 (they
default to 32) seems to make the problem go away.
Anyways, I guess I will have to wait for a newer version of gem5 unless I am
able to figure out the error myself.
Thanks for your help.
Marc
Message: 1
Date: Thu, 28 Apr 2011 18:10:05 -0400
From: Gabriel Michael Black <[email protected]>
To: [email protected]
Subject: Re: [m5-users] issue with O3 CPU and Ruby
Message-ID: <[email protected]>
Content-Type: text/plain; charset=ISO-8859-1; DelSp="Yes";
format="flowed"
Are you using the most recent version of the code? Doing so may fix
some issues with O3. Also, changing the dynamic_cast to a static_cast
probably just suppresses the symptom, not the problem. The
dynamic_cast probably recognizes the cast as illegal and returns a
NULL pointer, and then when you access it you necessarily segfault. If
you use a static_cast, it will merrily convert the value regardless,
and you may not access a bad address, but you'll still do something
you don't want.
Gabe
------------------------------
Message: 2
Date: Thu, 28 Apr 2011 17:22:58 -0500
From: Ali Saidi <[email protected]>
To: M5 users mailing list <[email protected]>
Subject: Re: [m5-users] issue with O3 CPU and Ruby
Message-ID: <[email protected]>
Content-Type: text/plain; charset=UTF-8; format=flowed
I'm fairly certain that Ruby doesn't work with the o3 cpu although one
of the Ruby experts might be able to give you a better status update.
Ali
<http://m5sim.org/cgi-bin/mailman/listinfo/m5-users>
_______________________________________________
m5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
