runtime: log stderr in TestLockRankGenerated on failure

For #62334.

Change-Id: I6d93dacc914590ef9c5a034dd4633466ec0bde8d
Reviewed-on: https://go-review.googlesource.com/c/go/+/524056
TryBot-Result: Gopher Robot <[email protected]>
Reviewed-by: Ian Lance Taylor <[email protected]>
LUCI-TryBot-Result: Go LUCI <[email protected]>
Auto-Submit: Bryan Mills <[email protected]>
Run-TryBot: Bryan Mills <[email protected]>

Author: Bryan C. Mills

src/runtime/lockrank_test.go

@@ -15,9 +15,13 @@ import (
 // Test that the generated code for the lock rank graph is up-to-date.
 func TestLockRankGenerated(t *testing.T) {
 	testenv.MustHaveGoRun(t)
-	want, err := testenv.CleanCmdEnv(exec.Command(testenv.GoToolPath(t), "run", "mklockrank.go")).CombinedOutput()
+	cmd := testenv.CleanCmdEnv(testenv.Command(t, testenv.GoToolPath(t), "run", "mklockrank.go"))
+	want, err := cmd.Output()
 	if err != nil {
-		t.Fatal(err)
+		if ee, ok := err.(*exec.ExitError); ok && len(ee.Stderr) > 0 {
+			t.Fatalf("%v: %v\n%s", cmd, err, ee.Stderr)
+		}
+		t.Fatalf("%v: %v", cmd, err)
 	}
 	got, err := os.ReadFile("lockrank.go")
 	if err != nil {