[v2,8/8] spi: loopback-test: added support for HW-loopback mode

Commit Message

Martin Sperl Dec. 21, 2015, 11:52 a.m. UTC

From: Martin Sperl <[e-mail]>

If the module parameter "loopback" for testing RX to match TX
is not set and the spi_master supports SPI_LOOP then
SPI_LOOP as well as RX-data testing will get enabled.

When the "loopback" module parameter is set
then the SPI_LOOP support in spi_master will not get enabled,
which means that MOSI needs to get connected to MISO by some
other means (possibly via a simple jumper connection
or a SN74AHCT125 connecting MOSI/MISO)

Suggested-by: Geert Uytterhoeven <[e-mail]>
Signed-off-by: Martin Sperl <[e-mail]>
---
 drivers/spi/spi-loopback-test.c |   27 ++++++++++++++++++++++++++-
 1 file changed, 26 insertions(+), 1 deletion(-)

Patch

diff --git a/drivers/spi/spi-loopback-test.c b/drivers/spi/spi-loopback-test.c
index b42d1a8..4e4f368 100644
--- a/drivers/spi/spi-loopback-test.c
+++ b/drivers/spi/spi-loopback-test.c
@@ -288,8 +288,34 @@  static struct spi_test spi_tests[] = {
 static int spi_loopback_test_probe(struct spi_device *spi)
 {
 	int ret;
+	char *msg;
 
-	dev_info(&spi->dev, "Executing spi-loopback-tests\n");
+	/*
+	 * Enable HW-loopback automatically if the master supports it
+	 * and we have the have_rx_buf unset (set means that we have
+	 * got an external loopback enabled, so no need to use SPI_LOOP)
+	 */
+	if (!check_rx_buf && (spi->master->mode_bits & SPI_LOOP)) {
+		spi->mode |= SPI_LOOP;
+		ret = spi_setup(spi);
+		if (ret) {
+			dev_err(&spi->dev,
+				"spi_setup failed to enable loopback: %i\n",
+				ret);
+			return ret;
+		}
+
+		/* force checking rx_buf for valid (loopback) data */
+		check_rx_buf = 1;
+
+		msg = "with SPI_LOOP support";
+	} else {
+		msg = (check_rx_buf) ?
+		      "with external loopback" :
+		      "without any loopback - rx_buf content is not checked";
+	}
+
+	dev_info(&spi->dev, "Executing spi-loopback-tests %s\n", msg);
 
 	ret = spi_test_run_tests(spi, spi_tests);