Xulin Zhou comments

Results 5 comments of


                                            Xulin Zhou

Errors occur when modify axpy-masked.mlir in some ways

After testing based on the latest repository, the two errors mentioned above still exist and present messages in the following way: 1. Changing `%mask64 = arith.constant dense : vector` to...

Will setting different `lmul` result in obvious performance difference?

After retesting, it is verified that: 1. When setting various lmul, scenarios with higher workload between load&store (e.g. Conv2d) have smaller performance gaps than situations with simple AX plus Y....

Will setting different `lmul` result in obvious performance difference?

The retest's results seem promising. The procedures now run faster than before and the gap between the total cycles of different lmuls has been narrowed (except for the case of...

llvm.sdiv results in ld.lld error: undefined symbol __divdi3

Further observation shows that the generated `sdiv` operation in this demo is of the i64 type, because the array index in MLIR is represented by `index` and will automatically be...

[buddy] The last step to achieve deep learning model e2e inference in vector repo

Further observation shows that additional support for these three functions(`malloc`, `memcpy` and `memrefCopy`) may be required to execute this model. Here, `memrefCopy` is originally implemented in the MLIR shared object...