Hi Pengfei,

We would definitely welcome your contribution. Normally patches are sent to
the m5-dev list. While this is pretty reasonable for a few patches to a
handful of files, it's more complex with all the files I imagine you'll be
changing/patching/etc. Please give us a few days to see if we can come up
with a better method to evaluate your code for inclusion in M5. At that
time we'll be very interested to see your work and thank you for taking the
time to contribute it back to the M5 community. 

Ali

On Tue, 13 Apr 2010 21:51:33 +0800, Pengfei Gou <[email protected]>
wrote:
> Hello, guys:
> 
> For studying EDGE ( Explicit Data Graph Execution ) architecture, I spent
> some time to make M5 simulator to support it.
> 
> I have done two things:
> 1). Adding a CPU model, which is based on O3-CPU model of M5, to support
> explicitly data-flow execution. This model includes an inst-block
> predictor,
> an IQ with explicit data-flow execution supporting, an LSQ, an inst-block
> RoB, etc.
> 
> 2). Describing TRIPS ISA ( an instance of EDGE ) using M5 ISA description
> language. I have made some trivial changes in isa_parser.py to support
> describing TRIPS ISA which is rather different with conventional
RISC/CISC
> ISA.
> 
> This model is able to execute binaries generated by TRIPS tool-chain. All
> the CPU2000 benchmarks supported by TRIPS tool-chain have ran
successfully
> in a reasonable speed.
> 
> I'm willing to contribute to the M5 community however, I have no idea how
> to
> share my codes with you guys.
> 
> If my achievement is worth to share, could anybody give me some guides of
> sharing my codes. If my achievement is not so interesting, could anybody
> give me some suggestions of how to improve it.
> 
> Thank you all.
> 
> Best wishes to all of you guys.
_______________________________________________
m5-dev mailing list
[email protected]
http://m5sim.org/mailman/listinfo/m5-dev

Reply via email to