tests: don't swallow stderr when running babeltrace CLI

I have found it very difficult to investigate failing tests if we don't
show stderr, so I suggest we just let it get printed.  Some tests are
expected to print to stderr in their normal course of operation (because
they test babeltrace error cases), so it means there will be some
additional output in a normal test run.  But I think it's a fair
trade-off.

Change-Id: Ia9bc5a706dd756ce0c9d1be4fec7510b12dface7
Signed-off-by: Simon Marchi <simon.marchi@efficios.com>
Reviewed-on: https://review.lttng.org/c/babeltrace/+/1609
CI-Build: Philippe Proulx <eeppeliteloop@gmail.com>
Tested-by: jenkins <jenkins@lttng.org>
Reviewed-by: Philippe Proulx <eeppeliteloop@gmail.com>

diff --git a/tests/utils/utils.sh b/tests/utils/utils.sh
index 5b9f0e96d17d9ab31ccdc647da4cb3ddceb6b70d..65949bd0467828eda3eb262a1b0b925bbf5b07c0 100644 (file)
--- a/tests/utils/utils.sh
+++ b/tests/utils/utils.sh
@@ -124,7 +124,7 @@ bt_diff_cli() {
        # Run the CLI to get a detailed file. Strip any \r present due to
        # Windows (\n -> \r\n). "diff --string-trailing-cr" is not used since it
        # is not present on Solaris.
-       echo "$args" | xargs "$BT_TESTS_BT2_BIN" 2>/dev/null | tr -d "\r" > "$temp_output_file"
+       echo "$args" | xargs "$BT_TESTS_BT2_BIN" | tr -d "\r" > "$temp_output_file"
 
        # Compare output with expected output
        if ! diff -u "$temp_output_file" "$expected_file" 2>/dev/null >"$temp_diff"; then