board/common: workaround missing --retry=NUM flag to podman

Disable the default "podman pull" retry value.  We use execd to retry
"podman create" on failure.  Wihtout this change, a single container
can block start of other containers by 3 * 20 seconds.  Now we only
block max 20 seconds before we try starting the next container.

Modern versions of podman (>= 5.0) have this --retry option, but it
does not have CNI, so this is a temporary workaround.

Signed-off-by: Joachim Wiberg <troglobit@gmail.com>

board/common/rootfs/usr/sbin/container

@@ -187,6 +187,10 @@ create()
 	logging="--log-driver k8s-file --log-opt path=/run/containers/$name.fifo"
     fi
 
+    # Pull quietly and don't retry on failure, we use execd for this,
+    # or user retry manually when run interactively, we may have other
+    # containers waiting to start that have an image locally already.
+    # Use --retry=0 with Podman 5.0 or later.
     args="$args --read-only --replace --quiet --cgroup-parent=containers $caps"
     args="$args --restart=$restart --systemd=false --tz=local $privileged"
     args="$args $vol $mount $hostname $entrypoint $env $port $logging"

patches/podman/4.5.0/0001-disable-pull-retry.patch

@@ -0,0 +1,21 @@
+This patch disables the default "podman pull" retry value, which otherwise
+blocks execd from trying other jobs in its queue.
+
+As of podman v5.0.0 a --retry=NUM has been added to the podman create, run,
+and pull commands.  However, CNI is no longer supported, and a lot of other
+breaking changes have been made, e.g., output of podman inspect.  So there's
+a lot of work upgrading.
+
+   -- Joachim
+
+--- a/vendor/github.com/containers/common/libimage/copier.go	2023-04-14 15:28:20.000000000 +0200
++++ b/vendor/github.com/containers/common/libimage/copier.go	2024-11-16 13:05:42.207641898 +0100
+@@ -24,7 +24,7 @@
+ )
+ 
+ const (
+-	defaultMaxRetries = 3
++	defaultMaxRetries = 0
+ 	defaultRetryDelay = time.Second
+ )
+