On 11/18/14, 5:02 PM, "Steve Reinhardt via gem5-dev" <[email protected]> wrote: > >I don't follow your question, Nilay. The former suggestion is really a >generalization of Andreas's idea, so the same answer holds. In the latter >case, you'd still need to ifdef some C++ code with USE_KVM (as Ali >originally suggested), but you wouldn't need a KvmVM param in the python, >which Gabe pointed out was a problem with Ali's suggestion. > >As Nate points out, if USE_KVM is exposed as a top-level config variable, >then it should be accessible from Python too, which perhaps a better way >of >fixing Ali's approach. > >But I still think the former approach is preferable, if you can get to the >point where the device doesn't really know if KVM is configured or not, >but >just gives some other part of the system the information it needs to do >the >right thing as appropriate. > >Steve >
Just to be clear my preferred solution is to provide a members/data-structures in a System object that objects can register ranges with and that can be queried by KVM. The skeleton header file was simply a straw man. Thanks, Ali -- IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you. ARM Limited, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, Registered in England & Wales, Company No: 2557590 ARM Holdings plc, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, Registered in England & Wales, Company No: 2548782 _______________________________________________ gem5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/gem5-dev
