Great, thanks for finding this... I was about to ask if you mind posting a
patch to reviews.gem5.org, but I see you already did.  Thanks again!

Steve

On Thu, Jun 30, 2011 at 1:31 AM, Maximilien Breughe <
[email protected]> wrote:

> Hi,
>
> My colleague and I solved the problem by looking at the cache unit. When a
> request is finished, the cache wakes up the CPU (if it was sleeping). We did
> the same for the Multiply Divide Unit: after the mul/div gets execute the
> MDU wakes up the CPU.
>
> By adding the following line to the function MultDivUnit::exeMulDiv(), the
> CPU starts executing instructions again:
> cpu->wakeCPU();
>
> Greets,
>
> Maximilien
>
>
>
> On 06/28/2011 11:02 AM, Maximilien Breughe wrote:
>
>> Hi all,
>>
>> I was playing around with the latency parameters of the MultDiv Unit of
>> the inorder CPU: multLatency, div8Latency, ...
>> Everything works fine for latencies up to 11. However, when you want
>> higher latencies the simulator crashes ("simulate() limit reached"). For
>> some reason the CPU starts being idle after 11 cycles waiting for the MDU to
>> finish and doesn't wake up anymore.
>>
>> Does anybody have a clue why multiply/divide latencies above 11 could
>> cause a problem?
>>
>> Kind regards and thanks in advance,
>>
>> Maximilien
>> ______________________________**_________________
>> gem5-users mailing list
>> [email protected]
>> http://m5sim.org/cgi-bin/**mailman/listinfo/gem5-users<http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users>
>>
>
> ______________________________**_________________
> gem5-users mailing list
> [email protected]
> http://m5sim.org/cgi-bin/**mailman/listinfo/gem5-users<http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users>
>
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Reply via email to