-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2896/#review6525
-----------------------------------------------------------


I'd like to see Brad's comment on this. I know he pushed hard for this 
functionality a few years ago. 

If I remember correctly, the purpose of the WireBuffer was to allow complicated 
inter-hardware communication without an explosion of state spaces. For 
instance, a probe filter co-located with the L3 cache. When you have these two 
structures implemented a single state machine you have a multiplicative effect 
on the number of states. But the WireBuffer would (theoretically) allow you to 
separate these two tightly-integrated structures into two different files.

Personally, I don't have any protocols that use the WireBuffers. However, I 
have often suggested to others that they use WireBuffers.

- Jason Power


On June 19, 2015, 4:09 a.m., Nilay Vaish wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> http://reviews.gem5.org/r/2896/
> -----------------------------------------------------------
> 
> (Updated June 19, 2015, 4:09 a.m.)
> 
> 
> Review request for Default.
> 
> 
> Repository: gem5
> 
> 
> Description
> -------
> 
> Changeset 10881:4962564c4ce6
> ---------------------------
> ruby: remove wire buffer
> 
> The structure is not being used anywhere.
> 
> 
> Diffs
> -----
> 
>   src/mem/protocol/RubySlicc_Types.sm ebb3d0737aa7 
>   src/mem/ruby/SConscript ebb3d0737aa7 
>   src/mem/ruby/structures/SConscript ebb3d0737aa7 
>   src/mem/ruby/structures/WireBuffer.hh ebb3d0737aa7 
>   src/mem/ruby/structures/WireBuffer.cc ebb3d0737aa7 
>   src/mem/ruby/structures/WireBuffer.py ebb3d0737aa7 
>   src/mem/slicc/symbols/StateMachine.py ebb3d0737aa7 
> 
> Diff: http://reviews.gem5.org/r/2896/diff/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Nilay Vaish
> 
>

_______________________________________________
gem5-dev mailing list
[email protected]
http://m5sim.org/mailman/listinfo/gem5-dev

Reply via email to