You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
After going through HW reference manual , it's not clear how the Cluster core other than Core 0 would fetch the instruction. In Cluster control subsystem there is boot address configuration for core 0 only there is no mention how the other cores(1-7) will fetch instruction ?
The text was updated successfully, but these errors were encountered:
@mohit3112 , it is not clear to me what is the point of your question. Pulp is made of a MIMD cluster of RISCV cores. Basically each core fetches instruction in the same manner. The instructions reside in the L2 memory and in the pulp cluster you have different cache systems you can explore (private caches, shared cache, hierarchical cache). Please feel free to go more in depth in the conversation, in case I did not answer to your question.
@MAX9596 the cluster domain is under the rtl folder in the pulp project. It instantiates the pulp cluster, which is treated as an IP and fetched from another github project through IPApprox tool
After going through HW reference manual , it's not clear how the Cluster core other than Core 0 would fetch the instruction. In Cluster control subsystem there is boot address configuration for core 0 only there is no mention how the other cores(1-7) will fetch instruction ?
The text was updated successfully, but these errors were encountered: