JishinMaster

CGEMM performance lower than SGEMM?

5

Hi, Thanks again for this great library. I used clblas a lot before switching to clBLAST, especially because it behaves better on other than AMD platforms due to the autotuning,...

upgrade to latest lvgl, micropython, ulab

8

I upgraded the micropython submodule to the latest commit. I had to change somes files due to some renaming in micropython 1.12, upgrade ulab and lvgl. For lvgl I also...

Upgrade to GCC 10.2?

1

The compiler version is laggin behind the GCC releases. Could it be possible to upgrade to GCC 9.3.0 or even 10.2, with newlib 3.2? Will the specifics from the Kendryte...

shuffles and other horizontal operations

2

Dear all, I work on an open source library called simd_utils (https://github.com/JishinMaster/simd_utils), which gives some optimised mathematical functions using SIMD/Vector intrinsics. I have a few functions using riscv vector extensions...

Support for Vector Instructions

3

I have not been able to use the vector instructions with the already built image. It worked with the bl808_linux kernel and Xuantie-900-gcc-linux-5.10.4-glibc-x86_64-V2.6.1 cross-compiler with options : "riscv64-unknown-linux-gnu-gcc -mcpu=c906 -mtune=c906v...

Hi, Thanks a lot for the good work. On pine64 Ox64 the vector instructions do not seem to work, but they do work on bl808_linux (linux 5.10.4). Is this a...