(1) In FS mode, it's the Linux kernel running on the simulated machine
that's doing the thread scheduling, so that's where you have to
control CPU assignment.  For example, you can use sched_setaffinity()
.  This has been covered before on the mailing list, I think... maybe
someone should put it on the wiki (if it's not there already).

(2) I don't know why you're getting a segfault; there's no way to tell
without at least a stack backtrace.  Note that the other messages you
showed are warnings, not errors, so they're probably not related.

Steve

On Mon, Mar 15, 2010 at 2:37 PM, Jie Meng <[email protected]> wrote:
> Hi all,
>
> I'm running M5 in Full-System mode with 8 CPUS. I've applied the Linux 
> patches for BigTsunami support up to 64 processors.
> The simulation works well using PARSEC multithreaded benchmark "blackscholes" 
> with small inputs and 64 threads.
>
> In the next step, I'm trying to run with multiprogram workloads on more 
> processors and having the following questions:
>
> (1)     In M5 FS mode, is there a way to assign different multithreaded 
> applications to different groups of CPUs?
>
>        For example, if I have 8 CPUs, and two multithreaded benchmarks ( 
> "blackscholes" and "canneal") each with 64 threads.
>        I'd like to assign "blackscholes"  to cpu0 to cpu3; and assign 
> "canneal"  to cpu4 to cpu7; to let them run in parallel.
>
>        Is there a way I could do that? If there is, how could I configure it 
> in the simulation script?
>        I've learned how to config multiprogram workloads in SE mode. But in 
> FS mode, should I also consider changing the .rcS script?
>
>
> (2)     Also, I'm trying to increase the number of CPUs to 64. (All the 
> configures are the same expect I'm using "-n 64" in the command line.)
>        Firstly, I created a checkpoint at the beginning of the ROI (region of 
> interest--parallel phase), to first run M5 in simple mode.
>        Then, I run M5 in detailed mode beginning at the checkpoint.
>
>        Everything works fine in the first step. But there's an "Segmentation 
> fault" as follows when restoring the checkpoint.
>
>        *****************************************************
>        Restoring from checkpoint
>        warn: Not unserializing 'system.cpu54.dcache': no section found in 
> checkpoint.
>        For more information see: http://www.m5sim.org/warn/dc3e87f8
>        warn: Not unserializing 'system.cpu54.icache': no section found in 
> checkpoint.
>        For more information see: http://www.m5sim.org/warn/dc3e87f8
>        Segmentation fault
>        *****************************************************
>
>        I've been using exactly the same configuration and steps with 8 CPUs 
> and it works well. I was wondering why it doesn't work with 64 cores.
>
>        Is there anybody met the same problem before? Or has any idea what may 
> cause this problem?
>
>        Any comment and suggestion would be appreciated,
>
>
> Thanks in advance,
> Jie
>
> _______________________________________________
> m5-users mailing list
> [email protected]
> http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
>
_______________________________________________
m5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/m5-users

Reply via email to