mbaret commented on a change in pull request #6222:
URL: https://github.com/apache/incubator-tvm/pull/6222#discussion_r467068235
##########
File path: cmake/config.cmake
##########
@@ -198,6 +198,16 @@ set(USE_DNNL_CODEGEN OFF)
set(USE_ARM_COMPUTE_LIB OFF)
set(USE_ARM_COMPUTE_LIB_GRAPH_RUNTIME OFF)
+# Whether to build with Arm Ethos-N support
+# Possible values:
+# - OFF: disable Arm Ethos-N support
+# - path/to/arm-ethos-N-stack: use a specific version of the
+# Ethos-N driver stack
+set(USE_ETHOSN OFF)
+# If USE_ETHOSN is enabled, use Ethos-N hardware (ON) or
+# software test infrastructure (OFF)
+set(USE_ETHOSN_HW ON)
Review comment:
Good catch, yes it should be off by default. The naming of USE_ETHOSN is
really because it toggles both the codegen and runtime support. USE_ETHOSN_HW
determines whether or not we mock the inference function in the runtime or use
real hardware.
----------------------------------------------------------------
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]
