Get bpf level when check bpf support

Instead of return boolean, bpf support check now returns a integer
represent the current bpf level on device. This level is used to decide
if the device support some advanced bpf feature such as map_in_map and
bpf cgroup socket filter. Delete the binder call for bpf status check
since no one is using it.

Bug: 111441138
Test: libnetdbpf_test, netd_integration_test
Change-Id: Ib70c07647ffe491d493b4582b4b4b0eba7caf3a9

1 files changed, 4 insertions, 1 deletions

diff --git a/server/FirewallControllerTest.cpp b/server/FirewallControllerTest.cpp
index 71cc532f..bd933b47 100644
--- a/server/FirewallControllerTest.cpp
+++ b/server/FirewallControllerTest.cpp
@@ -39,7 +39,10 @@ class FirewallControllerTest : public IptablesBaseTest {
 protected:
     FirewallControllerTest() {
         FirewallController::execIptablesRestore = fakeExecIptablesRestore;
-        mFw.mUseBpfOwnerMatch = false;
+        // This unit test currently doesn't cover the eBPF owner match case so
+        // we have to manually turn eBPF support off.
+        // TODO: find a way to unit test the eBPF code path.
+        mFw.mUseBpfOwnerMatch = android::bpf::BpfLevel::NONE;
     }
     FirewallController mFw;