[zircon][core-test]: Fix flaky ChannelInternalTest
There is a race condition in a channel call when the channel is written
to and the thread state is set to blocked. Instead of just asserting
for the thread state, we wait for the channel to become readable, and
then wait for the thread state to be updated.
Essentially the expectation the thread is blocked switches to the thread
will eventually become blocked when issuing a channel_call.
ZX-4422 #done
TEST=channel-test
Change-Id: Iffa110f67a63cc6ba450631ceafa7430aa44f885
diff --git a/zircon/system/utest/core/channel/channel-internal.cc b/zircon/system/utest/core/channel/channel-internal.cc
index 56a9dd2..087fce5 100644
--- a/zircon/system/utest/core/channel/channel-internal.cc
+++ b/zircon/system/utest/core/channel/channel-internal.cc
@@ -65,8 +65,20 @@
&act_bytes, &act_handles));
}
-// TODO(ZX-4422) Test is flaky
-#if 0
+void WaitForThreadState(zx_handle_t thread_handle, zx_thread_state_t state) {
+ // Make sure the original thread is still blocked.
+ // It is safe to read from caller_thread_handle since this is set before
+ // the read happends, and we waited until the remote endpoint became readable.
+ zx_info_thread_t info = {};
+ while (info.state != state) {
+ uint64_t actual = 0;
+ uint64_t actual_2 = 0;
+ ASSERT_OK(zx_object_get_info(thread_handle, ZX_INFO_THREAD, &info, sizeof(info), &actual,
+ &actual_2));
+ }
+ return;
+}
+
// Test current behavior when transferring a channel with pending calls out of the current process.
// TODO(ZX-4233): This test ensures that currently undefined behavior does not change
// unexpectedly. Once the behavior is properly undefined, this test should be updated.
@@ -127,6 +139,8 @@
});
ASSERT_OK(remote.wait_one(ZX_CHANNEL_READABLE, zx::time::infinite(), nullptr));
+ ASSERT_NO_FATAL_FAILURES(
+ WaitForThreadState(caller_thread_handle.load(), ZX_THREAD_STATE_BLOCKED_CHANNEL));
// Read the message from the test thread.
Message request = {};
@@ -150,9 +164,8 @@
ASSERT_EQ(zx::thread::create(proc, "mini-p-channel-test-thrd", 2u, 0u, &thread), ZX_OK);
zx::channel cmd_channel;
- ASSERT_OK(start_mini_process_etc(proc.get(), thread.get(), vmar.get(),
- local.release(), true,
- cmd_channel.reset_and_get_address()));
+ ASSERT_OK(start_mini_process_etc(proc.get(), thread.get(), vmar.get(), local.release(),
+ true, cmd_channel.reset_and_get_address()));
auto cleanup = fbl::MakeAutoCall([&cmd_channel]() {
ASSERT_EQ(mini_process_cmd(cmd_channel.get(), MINIP_CMD_EXIT_NORMAL, nullptr),
@@ -176,9 +189,8 @@
zx_info_thread_t info = {};
uint64_t actual = 0;
uint64_t actual_2 = 0;
- ASSERT_EQ(zx_object_get_info(caller_thread_handle.load(), ZX_INFO_THREAD, &info,
- sizeof(info), &actual, &actual_2),
- ZX_OK);
+ ASSERT_OK(zx_object_get_info(caller_thread_handle.load(), ZX_INFO_THREAD, &info,
+ sizeof(info), &actual, &actual_2));
EXPECT_EQ(info.state, ZX_THREAD_STATE_BLOCKED_CHANNEL);
// Write a response to the original call after we know the endpoint has been
@@ -192,7 +204,6 @@
FAIL("caller_thread encountered an error on channel::call: %s", caller_error.load());
}
}
-#endif
} // namespace
} // namespace channel
