You are viewing a plain text version of this content. The canonical link for it is here.

Posted to discuss-archive@tvm.apache.org by Tristan Konolige via Apache TVM Discuss <no...@discuss.tvm.ai> on 2020/10/13 17:36:42 UTC

[Apache TVM Discuss] [Questions] Dose TVM only support mkldnn dense operation?


To use native libraries, you have to pass them to the libs flag like so: `llvm -libs=mkdnn`. Also make sure you have set `USE_MKLDNN=On` in build/config.cmake.

Right now we only support using mkldnn for matrix multiplication.





---
[Visit Topic](https://discuss.tvm.apache.org/t/dose-tvm-only-support-mkldnn-dense-operation/8160/2) to respond.

You are receiving this because you enabled mailing list mode.

To unsubscribe from these emails, [click here](https://discuss.tvm.apache.org/email/unsubscribe/c98b9313c01e0db0ca0c6ae06377170807744554d2712c5047c045b9ed28d33f).

[Apache TVM Discuss] [Questions] Dose TVM only support mkldnn dense operation?

Posted by yufeng via Apache TVM Discuss <no...@discuss.tvm.ai>.


@tkonolige Thanks for replying!
I read the source code of tvm and find BYOC is exactly what's I need. TVM use dnnl (a new mkldnn name) instead of mkldnn.

Howerver, is there a network benchmark to use dnnl compiler? I can only find snippet network in tests/python/relay/test_external_codegen.py. Is there a resnet or inception serials network to use the BYOC-based TVM?





---
[Visit Topic](https://discuss.tvm.apache.org/t/dose-tvm-only-support-mkldnn-dense-operation/8160/3) to respond.

You are receiving this because you enabled mailing list mode.

To unsubscribe from these emails, [click here](https://discuss.tvm.apache.org/email/unsubscribe/f3edd6dd7f35245e365992b0a16fd815719a2c65d0acdef11855b997c5f5ab11).