Move the `s8-vclamp` and `u8-vclamp` tests to the `SHARDED_TESTS` since they are very slow on `riscv64-rvv`.

google / XNNPACK

High-efficiency floating-point neural network inference operators for mobile, server, and Web

Other

1.89k stars 376 forks source link

Closed copybara-service[bot] closed 2 weeks ago

copybara-service[bot] commented 2 weeks ago

Move the s8-vclamp and u8-vclamp tests to the SHARDED_TESTS since they are very slow on riscv64-rvv.