The diagram of V100 shows that each SM unit has 64 SP core, 32 DP core, and 8 Tensor core. I am wondering whether these SP and DP core are using the same hardware executaion units or not, i.e., are 2 SP cores logically identified as 1 DP core, or in fact V100 has seperate hardware units for SP and DP instructions.
According to https://images.nvidia.com/content/volta-architecture/pdf/volta-architecture-whitepaper.pdf
Similar to Pascal GP100, the GV100 SM incorporates 64 FP32 cores and 32 FP64 cores per SM.
However, the GV100 SM uses a new partitioning method to improve SM utilization and overall
performance. Note that the GP100 SM is partitioned into two processing blocks, each with 32
FP32 Cores, 16 FP64 Cores, an instruction buffer, one warp scheduler, two dispatch units, and a
128 KB Register File. The GV100 SM is partitioned into four processing blocks, each with 16 FP32
Cores, 8 FP64 Cores, 16 INT32 Cores, two of the new mixed-precision Tensor Cores for deep
learning matrix arithmetic, a new L0 instruction cache, one warp scheduler, one dispatch unit,
and a 64 KB Register File. Note that the new L0 instruction cache is now used in each partition to
provide higher efficiency than the instruction buffers used in prior NVIDIA GPUs. (See the Volta
SM in Figure 5).
I take that as seperate units.
Thanks so much for the information. Suppose my program uses all SP cores for every SM, then is it possible to use DP core at the same time? I am wondering If I have to program in mixed-precision in order to fully exploit the performance of V100.