mirror of
https://github.com/clearlinux/graphene.git
synced 2026-09-04 12:51:41 +00:00
Michał Kowalczyk
e587869e13
Supporting these options complicates the design of Graphene and loading
logic significantly, providing little useful functionality:
- loader.exec:
- the main user of it were our tests
- worked only for the first process spawned inside Graphene, as it
was a unidirectional manifest->binary mapping, so the child
process didn't know about the corresponding manifest.
- sgx.sigfile:
- probably all existing usages of it were completely redundant
- was resolved relatively to CWD instead of the executable location,
which made it mostly useless
From now on, the correct location of the files is:
- either place the manifest and sigfile next to the binary, with a
matching name, or
- create a symlink to the binary in the folder where manifests are
stored and launch it through this symlink
This example demonstrates how to run TensorFlow Lite v1.9. In particular, the
example runs label_image program on Graphene. It reads an input image
image.bmp from the current directory and uses TensorFlow Lite and the
Inception v3 model to label the image.
To install build dependencies on Ubuntu there is a convenience target invoked
with make install-dependencies-ubuntu. This also serves as a starting point to
figure out which packages to install on newer releases of Ubuntu.
To build TensorFlow Lite and Graphene artifacts:
- without SGX do
make - with SGX do
make SGX=1
To run the image labeling example:
- without Graphene do
make run-native - with Graphene do
make run-graphene - with Graphene-SGX do
make SGX=1 run-graphene