Closed mmrahorovic closed 3 months ago
Interleave output of RTL-SWG in CPPsim (according to SIMD) such that following VVAU (which is assumed to be VVAU_hls, VVAU_rtl or VVAU and not a MatMul node) executes the matrix multiplication correctly given it's PE parallelism.
Interleave output of RTL-SWG in CPPsim (according to SIMD) such that following VVAU (which is assumed to be VVAU_hls, VVAU_rtl or VVAU and not a MatMul node) executes the matrix multiplication correctly given it's PE parallelism.