chiwwang

Results 24 comments of chiwwang

PoC: Add Qualcomm mobile SoC native backend for GGML

In general I suggest to ask questions on Qualcomm forum. Nonetheless, it's very possible that some fields are wrong in OpConfig structure so it's rejected by QNN CPU. Instead of...

PoC: Add Qualcomm mobile SoC native backend for GGML

> > In general I suggest to ask questions on Qualcomm forum. Nonetheless, it's very possible that some fields are wrong in OpConfig structure so it's rejected by QNN CPU....

[Draft] Qualcomm AI Engine Direct - Support kv_cached llama2 model

Rebased as https://github.com/pytorch/executorch/pull/3656/

[Draft] Qualcomm AI Engine Direct - Support kv_cached llama2 model

Please see https://github.com/pytorch/executorch/pull/4142 instead.

[QNN] Support non-Android aarch64 Linux targets for QNN backend?

May I know if the below work for you? https://github.com/pytorch/executorch/issues/1527#issuecomment-1876428906

Convert scalar to tensor before quantizer annoate

@shewu-quic @haowhsu-quic

ERROR: Overriding output data pointer allocated by memory plan is not allowed.

This looks like a failure in fastrpc. ``` [ERROR] [Qnn ExecuTorch]: SSR Detected - You must free and recreate affected QNN API handles associated with deviceId 0 coreId 0 pdId...

ERROR: Overriding output data pointer allocated by memory plan is not allowed.

``` [ERROR] [Qnn ExecuTorch]: QnnDsp DspTransport call failed, error 0x00000007 ``` mmmmm not sure why the fastrpc is unhappy. This is usually caused by environment settings. @hans00 , Could you...

[Draft] Qualcomm AI Engine Direct - [WIP] llama2...

Please see https://github.com/pytorch/executorch/pull/4142 instead.

prefill model

@shewu-quic @haowhsu-quic @chunit-quic