Re: [Gems-users] Question of protocol testing

[Polina Dudnik <pdudnik@xxxxxxxxx>]

Jerry,

There's no special tricks in debugging the protocols. Just turn on all the debugging and think about what you've changed. Look at the transitions and double-check that the transitions do what you expect. Trace the address for which you get the error backwards through the protocol debug trace.

Polina

On Wed, Oct 21, 2009 at 3:18 AM, Jerry Lin <cosjerry@xxxxxxxxx> wrote:

Hi,
I've modified MOSI_SMP_bcast protocol for my research.

I used the tester to examine the modified protocol.

But when the processor tried to forward data to other processors,

I got error message as following.

====================================================

1070  3  -1            Seq             Done             >        [0x400, line 0x400] 569 cycles NULL LD Yes 

1070  3   0     L1Cache             Data       IS_D>S     [0x400, line 0x400]

failed assertion 'thread != -1' at fn void Sequencer::readCallback(const Address&, DataBlock&) in system/Sequencer.C:556

failed assertion 'thread != -1' at fn void Sequencer::readCallback(const Address&, DataBlock&) in system/Sequencer.C:556

====================================================

After cycle1070, it should forwards the data to another processor.

I used "e_recordForwardingInfo" and "o_cacheToForward" to do the forwarding.

What is the meaning of error message? Did I do something wrong?

Thanks for helping.

_______________________________________________
Gems-users mailing list
Gems-users@xxxxxxxxxxx
https://lists.cs.wisc.edu/mailman/listinfo/gems-users
Use Google to search the GEMS Users mailing list by adding "site:https://lists.cs.wisc.edu/archive/gems-users/" to your search.