Re: [Gems-users] Regarding checkpoints

[Chandra Raghavan <chandrashekar@xxxxxxxxx>]

Hi:
I tried a similar configuration except with ARM processors instead of
x86 and Ruby loaded without any problems. I have a few queries at this
point:
1. I need to run a producer-consumer processor configuration and record
memory access times for each processor separately i.e. how many
instructions cause cache misses and what is the memory access time for
each of those instructions. Can Ruby record statistics at this
granularity? How do I set it up to do so?
2. My configuration defines processors, caches and a physical memory.
My understanding is that when I load Ruby it automatically plugs in to
the configuration and generates timing information. My configuration
file itself is independent of ruby. Is this right? 

Thank you very much for taking the time to reply! I really appreciate the help.
Chandrashekar(Chandra)

On 8/17/05, Mike Marty <mikem@xxxxxxxxxxx> wrote:

Chandra,

 

Currently GEMS only supports target machines with SPARC processors.  We believe that porting Ruby to the x86 target would be easy.  However porting Opal to x86 would likely require a rewrite. 

 

--Mike

 

 

-----Original Message-----
From: gems-users-bounces@xxxxxxxxxxx [mailto: gems-users-bounces@xxxxxxxxxxx] On Behalf Of Chandra Raghavan
Sent: Tuesday, August 16, 2005 9:38 PM
To: gems-users@xxxxxxxxxxx
Subject: [Gems-users] Regarding checkpoints

 

Hi:
I'm trying to get GEMS running with a particular processor configuration I require. I have SIMICS conf file that has the following objects:
1. 3 x86 processors
2. A common cache
3. memory space
4. port-space

I started SIMICS, did a read-configuration and "load-module ruby" which resulted in the error message:
error installing ruby timing model.
Simics license checked in!

What other components do I have to add to the configuration so that Ruby can work? Or am I making someother mistake?
Thank you very much!
Chandrashekar(Chandra)
P.S: I'm a GEMS/SIMICS newbie. So please bear with me if this question is trivial.

_______________________________________________
Gems-users mailing list
Gems-users@xxxxxxxxxxx
https://lists.cs.wisc.edu/mailman/listinfo/ gems-users