README: Update build instructions

Signed-off-by: Suraj Deshmukh <suraj@kinvolk.io>

README.md

@@ -9,14 +9,22 @@ Also, another important consideration is that since a node start till the policy
 is applied, the node should not be exposed. This is not covered by this
 controller.
 
-Building
-========
+## Building
 
 Depending on your architecture you would use one of the commands:
 
+### x86
+
+```
+ARCH=amd64 docker build -t quay.io/kinvolk/calico-hostendpoint-controller:v0.0.4-amd64 \
+    --build-arg ARCH . && docker push quay.io/kinvolk/calico-hostendpoint-controller:v0.0.4-amd64
+```
+
+### arm
+
 ```
-ARCH=amd64 docker build -t kinvolk/calico-hostendpoint-controller:v0.0.2-amd64 --build-arg ARCH . && docker push kinvolk/calico-hostendpoint-controller:v0.0.2-amd64
-ARCH=arm64 docker build -t kinvolk/calico-hostendpoint-controller:v0.0.2-arm64 --build-arg ARCH . && docker push kinvolk/calico-hostendpoint-controller:v0.0.2-arm64
+ARCH=arm64 docker build -t quay.io/kinvolk/calico-hostendpoint-controller:v0.0.4-arm64 \
+    --build-arg ARCH . && docker push quay.io/kinvolk/calico-hostendpoint-controller:v0.0.4-arm64
 ```
 
 There are multiple ways for building an image for another architecture which is
@@ -25,7 +33,7 @@ The most convenient one is to install the `qemu-user-static` package on your
 system to set up binary translation so that the Linux kernel can run binaries
 for other architectures.
 
->NOTE: You may have to run `sudo systemctl restart systemd-binfmt.service` after installing.
+> **NOTE**: You may have to run `sudo systemctl restart systemd-binfmt.service` after installing.
 
 The downside of this approach is that a base image for the non-native
 architecture must be explicitly specified in the `FROM` directive of the
@@ -36,6 +44,8 @@ need to use `arm64v8/debian:9` (prefixed because there is no `debian:9-arm64`).
 When done building, you can change the Dockerfile back to the way it was
 before. Unfortunately the `FROM` directive does not handle build arguments.
 
+### Combined image tag
+
 Now make sure you have `"experimental": "enabled"` in your
 `~/.docker/config.json` (surrounded by `{` and `}` if the file is otherwise
 empty).
@@ -44,8 +54,17 @@ When all images are built on the respective architectures and pushed they can
 be combined through a manifest to build a multiarch image:
 
 ```
-docker manifest create kinvolk/calico-hostendpoint-controller:v0.0.2 --amend kinvolk/calico-hostendpoint-controller:v0.0.2-amd64 --amend kinvolk/calico-hostendpoint-controller:v0.0.2-arm64
-docker manifest annotate kinvolk/calico-hostendpoint-controller:v0.0.2 kinvolk/calico-hostendpoint-controller:v0.0.2-amd64 --arch=amd64 --os=linux
-docker manifest annotate kinvolk/calico-hostendpoint-controller:v0.0.2 kinvolk/calico-hostendpoint-controller:v0.0.2-arm64 --arch=arm64 --os=linux
-docker manifest push kinvolk/calico-hostendpoint-controller:v0.0.2
+docker manifest create quay.io/kinvolk/calico-hostendpoint-controller:v0.0.4 \
+    --amend quay.io/kinvolk/calico-hostendpoint-controller:v0.0.4-amd64 \
+    --amend quay.io/kinvolk/calico-hostendpoint-controller:v0.0.4-arm64
+
+docker manifest annotate quay.io/kinvolk/calico-hostendpoint-controller:v0.0.4 \
+    quay.io/kinvolk/calico-hostendpoint-controller:v0.0.4-amd64 --arch=amd64 --os=linux
+
+docker manifest annotate quay.io/kinvolk/calico-hostendpoint-controller:v0.0.4 \
+    quay.io/kinvolk/calico-hostendpoint-controller:v0.0.4-arm64 --arch=arm64 --os=linux
+
+docker manifest push quay.io/kinvolk/calico-hostendpoint-controller:v0.0.4
 ```
+
+> **NOTE**: Above commands can be run from any machine.