That will make them develop specific IO dies only, on cheap as f*** 14 nm process. Which will in return turn into more profit, and saving moneyz on manufacturing.
#20180102338 Circuit Board with Bridge Chiplets
Actually the interconnect is also a chiplet (
55/50) according to the patent and is fabricated using a
higher density process than the rest of the board
The circuit structures of the chiplets 50 and 55 may be constructed using one more design rules for higher density circuit structures while the circuit structures of the remainder of the circuit board...
The interconnect chiplets provide connectivity from one regular chiplet (
20/25/30) to another and to the pin arrays (
45)
The chiplets 50 and 55 may be used for a variety of purposes. For example, the chiplet 50 may be used to provide large numbers of electrical pathways between the semiconductor chip 20 and 25 as well as electrical pathways to and from the semiconductor chips 20 and 25, through the circuit board 15 and out to the I/O's 45 if desired. The chiplet 55 may be used to provide large numbers of electrical pathways between the semiconductor chip 25 and the semiconductor chip 30 as well as electrical pathways to and from the semiconductor chips 25 and 30 through the circuit board 15 and out to the I/O's 45 if desired.
This might not be the definitive layout for Zen 2 but as it looks, seem complicated.
The patent is fairly comprehensive in the sense it covers not only how the chip are connected to one another but also how these pockets will be created on the substrate and how to insert these interconnect onto the substrate and putting the chiplets on top.
#20180096938 Circuit Board with Multiple Density regions
Here another layout with the
50/52/53/55 interconnect chiplets being used differently.
The patent is all about how the chips
20/25/30 can be of different density and using the interconnects
50/52/53/55 to connect to one another.
#20180239708 Acceleration of cache-to-cache data transfers for producer-consumer communication
[0020] Referring to FIG. 1, processing system 100 (e.g., a server) includes multiple processing nodes (e.g., node 0 and node 1). Each processing node includes multiple caching agents (e.g., processors 102 and 104 coupled to main memory 110). For example, caching agent 102, is a processor including core 0, core 1, core 2, . . . core 7 and caching agent 104, is a processor including core 0, core 1, core 2, . . . core 7) and a memory system. Each of the nodes accesses its own memory within corresponding coherence domain 122 faster than memory in non-coherence domain 124 (e.g., main memory 110) or memory in another node. As referred to herein, a coherence domain refers to a subset of memory (e.g., cache memory of node 0) for which a cache coherence mechanism maintains a coherent view of copies of shared data. A non-coherence domain refers to memory not included in the coherence domain (e.g., main memory 110 or cache memory in another node). Each of the caching agents in a node includes a last-level cache shared by the cores of the caching agent. Each core includes a private penultimate-level cache. For example, caching agent 102 includes last-level cache 128, which is a level-three cache shared by core 0, core 1, core 2, . . . core 7, and includes a level-two cache within each of core 0, core 1, core 2, . . . core 7. Last-level cache 128 and each level-two cache of caching agent 102 includes storage elements, e.g., storage implemented in fast static Random Access Memory (RAM) or other suitable storage elements. Cache control logic is distributed across last-level cache 128 and each level-two cache of caching agent 102. The caching agents use inter-processor communication via directory controller 121 to maintain coherency of a memory image in main memory 110 when caches of more than one caching agent contain the same cache line (i.e., a copy of contents of the same location of main memory 110) of coherence domain 122. In at least one embodiment of processing system 100, probe filter 112 includes storage for a cache directory used to implement a directory-based cache coherency policy. Probe filter 112 is implemented in fast static RAM associated with directory controller 121 or by other suitable storage technique.
#20180239702 Locality-aware and sharing-aware cache coherence for collections of processors
Cache coherence across multiple processors on an interconnect network (in one case an interposer 412)
[0019] Processing system 100 includes a distributed, shared memory system. For example, all memory locations of memory system 108 are accessible by each of processors 102, 104, and 106. Memory system 108 includes multiple memory portions, which are distributed across processors 102, 104, and 106. Memory portion 110 is local to processor 102 (e.g., tightly integrated with processor 102) and remote to processors 104 and 106 (e.g., within processing system 100 and accessible to processors 104 and 106, but not local to processors 104 and 106).
Some more applications which are interesting to read regarding AMDs ideas/attempts at memory access optimization in a CPU and/or with GPUs
#20180039587 Network of memory modules with logarithmic access
#20180018105 Memory controller with virtual controller mode
#20180019006 Memory controller with flexible address decoding
#20180143905 Network-aware cache coherence protocol enhancement
#20180239722 Allocation of memory buffers in computing system with multiple memory channels
Search in USPTO website here
http://appft.uspto.gov/netahtml/PTO/srchnum.html
TL;DR;
In simple terms,
#20180102338 describes how to create cavities on the substrate to put connecting chiplets with conducting pads on both surfaces on these cavities and the real chips are then placed on top of them. Due to the high density of the connecting chiplets complex routing between the chiplets can be achieved as well as connections to the pin arrays are also provided.
#20180096938 describes chiplets of different process nodes integrated in one package
#20180239702 describes cache coherence across multiple processors.
#20180239708 talks of a node with multiple processor/dies each with multiple cores and all of them are using a single Memory controller and LLC sync across the dies
#20180239722 talks of a single Memory controller in one Computing devices but not sure if only for APU use case.