Memory with processing in memory architecture and operating method thereof

1. A memory with a processing in memory architecture, comprising:

a memory array, comprising a plurality of memory regions;

a mode register, configured to store a plurality of memory mode settings;

a memory interface, coupled to the memory array and the mode register, and externally coupled to a special function processing core; and

an artificial intelligence core, coupled to the memory array and the mode register,

wherein the plurality of memory regions are respectively selectively assigned to the special function processing core or the artificial intelligence core according to the plurality of memory mode settings of the mode register, so that the special function processing core and the artificial intelligence core respectively access different memory regions in the memory array according to the plurality of memory mode settings,

wherein the plurality of memory regions comprise a first memory region and a second memory region, the first memory region is configured for exclusive access by the artificial intelligence core, and the second memory region is configured for exclusive access by the special function processing core,

wherein the plurality of memory regions further comprise a plurality of data buffer regions, and the artificial intelligence core and the memory interface alternately access different data in the plurality of data buffer regions,

wherein when the artificial intelligence core performs a neural network operation, the artificial intelligence core reads input data of one of the plurality of data buffer regions as an input parameter, and reads weight data of the first memory region, wherein the artificial intelligence core outputs feature map data to the first memory region.

2. The memory according to claim 1 , wherein the special function processing core and the artificial intelligence core simultaneously access the different memory regions in the memory array respectively via their respective dedicated memory buses.

3. The memory according to claim 1 , wherein when the artificial intelligence core performs the neural network operation, the artificial intelligence core reads the feature map data of the first memory region as a next input parameter, and reads another weight data of the first memory region, wherein the artificial intelligence core outputs next feature map data to one of the plurality of data buffer regions to overwrite the one of the plurality of data buffer regions.

4. The memory according to claim 1 , wherein the plurality of data buffer regions are respectively and alternately assigned to the special function processing core or the artificial intelligence core, so that a first memory space corresponding to the artificial intelligence core comprises the first memory region and one of the plurality of data buffer regions, and a second memory space corresponding to the special function processing core comprises the second memory region and another one of the plurality of data buffer regions.

5. The memory according to claim 1 , wherein a width of a bus dedicated between the artificial intelligence core and the plurality of memory regions is greater than a width of an external bus between the special function processing core and the memory interface.

6. The memory according to claim 1 , wherein the plurality of memory regions respectively correspond to a plurality of row buffer blocks, and the plurality of memory regions each comprise a plurality of memory banks, wherein a width of a bus dedicated between the artificial intelligence core and the plurality of memory regions is greater than or equal to data amount of an entire row of the plurality of memory banks.

7. The memory according to claim 1 , wherein the memory is a dynamic random access memory chip.

8. An operating method of a memory with a processing in memory architecture, the memory comprising a memory array, a mode register, a memory interface and an artificial intelligence core, wherein the operating method comprises:

respectively selectively assigning a plurality of memory regions in the memory to a special function processing core and the artificial intelligence core according to a plurality of memory mode settings of the mode register; and

respectively accessing different memory regions in the memory array by the special function processing core and the artificial intelligence core according to the plurality of memory mode settings,

wherein the plurality of memory regions comprise a first memory region and a second memory region, the first memory region is configured for exclusive access by the artificial intelligence core, and the second memory region is configured for exclusive access by the special function processing core,

wherein the plurality of memory regions further comprise a plurality of data buffer regions, and the artificial intelligence core and the memory interface alternately access different data in the plurality of data buffer regions,

wherein when the artificial intelligence core performs a neural network operation, the step of respectively accessing the different memory regions in the memory array by the special function processing core and the artificial intelligence core according to the plurality of memory mode settings of the mode register comprises:

reading input data of one of the plurality of data buffer regions as an input parameter by the artificial intelligence core;

reading weight data of the first memory region by the artificial intelligence core; and

outputting feature map data to the first memory region by the artificial intelligence core.

9. The operating method of a memory according to claim 8 , wherein the special function processing core and the artificial intelligence core simultaneously access the different memory regions in the memory array respectively via their respective dedicated memory buses.

10. The operating method of a memory according to claim 8 , wherein when the artificial intelligence core performs the neural network operation, the step of respectively accessing the different memory regions in the memory array by the special function processing core and the artificial intelligence core according to the plurality of memory mode settings of the mode register further comprises:

reading the feature map data of the first memory region as a next input parameter by the artificial intelligence core;

reading another weight data of the first memory region by the artificial intelligence core; and

outputting next feature map data to one of the plurality of data buffer regions by the artificial intelligence core to overwrite the one of the plurality of data buffer regions.

11. The operating method of a memory according to claim 8 , wherein the plurality of data buffer regions are respectively and alternately assigned to the special function processing core or the artificial intelligence core, so that a first memory space corresponding to the artificial intelligence core comprises the first memory region and one of the plurality of data buffer regions, and a second memory space corresponding to the special function processing core comprises the second memory region and another one of the plurality of data buffer regions.

12. The operating method of a memory according to claim 8 , wherein a width of a bus dedicated between the artificial intelligence core and the plurality of memory regions is greater than a width of an external bus between the special function processing core and the memory interface.

13. The operating method of a memory according to claim 8 , wherein the plurality of memory regions respectively correspond to a plurality of row buffer blocks, and the plurality of memory regions each comprise a plurality of memory banks, wherein a width of a bus dedicated between the artificial intelligence core and the plurality of memory regions is greater than or equal to data amount of an entire row of the plurality of memory banks.

14. The operating method of a memory according to claim 8 , wherein the memory is a dynamic random access memory chip.