manupa-arm commented on a change in pull request #6950:
URL: https://github.com/apache/tvm/pull/6950#discussion_r529751671
##########
File path: python/tvm/micro/build.py
##########
@@ -162,7 +158,12 @@ def build_static_runtime(
libs.append(compiler.library(lib_build_dir, lib_srcs, lib_opts))
- libs.append(compiler.library(mod_build_dir, [mod_src_path],
generated_lib_opts))
+ libs.append(
Review comment:
In order to satisfy that requirement, we could still make the
export_library create the artifacts in a user given workspace directory. Thus,
we would not have this split flow of exporting the IRModules in tvm vs utvm and
thats what I meant by A2.
On other hand of generating projects, I think we should discuss it in a RFC.
My personal feeling it should be a layer above the TVM stack (similiar to tvmc)
to collect the artifacts produced out of tvm to generate such projects rather
than making tvm a project generator. However, I dont have a strong opinion on
either.
Thoughts ? : @tqchen
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
