Thread (20 messages) 20 messages, 2 authors, 2019-03-25

Re: [PATCH V2 6/9] spi: bcm2835aux: warn in dmesg that native cs is not really supported

From: Stefan Wahren <hidden>
Date: 2019-03-25 10:28:31
Also in: linux-spi

Hi Martin,

Am 24.03.19 um 18:49 schrieb kernel@martin.sperl.org:
From: Martin Sperl <redacted>

From personal bad experience (even as the author of the original driver)
it shows that native-cs is "somewhat" supported by the spi bus driver
when using a buggy device tree.

So make sure that the driver is warning in dmesg about this fact
that we are running in a not supported mode that may have surprizing
limitations.

Signed-off-by: Martin Sperl <redacted>
Fixes: 1ea29b39f4c812ece2f936065a0a3d6fe44a263e
please use a proper fixes tag.
quoted hunk ↗ jump to hunk
---
 drivers/spi/spi-bcm2835aux.c | 33 +++++++++++++++++++++++++++++++++
 1 file changed, 33 insertions(+)
diff --git a/drivers/spi/spi-bcm2835aux.c b/drivers/spi/spi-bcm2835aux.c
index c8acde017b6a..6af07dec5fdc 100644
--- a/drivers/spi/spi-bcm2835aux.c
+++ b/drivers/spi/spi-bcm2835aux.c
@@ -396,6 +396,38 @@ static void bcm2835aux_spi_handle_err(struct spi_master *master,
 	bcm2835aux_spi_reset_hw(bs);
 }

+static int bcm2835aux_spi_setup(struct spi_device *spi)
+{
+	int ret;
+
+	if (spi->mode & SPI_NO_CS)
+		return 0;
+	/* sanity check for native cs */
+	if (!gpio_is_valid(spi->cs_gpio)) {
+		/* for dt-backwards compatibility: only support native on CS0
+		 * known things not supported with broken native CS:
+		 * * multiple chip-selects: cs0-cs2 are all
+		 *     simultaniously asserted whenever there is a transfer
+		 *     this even includes SPI_NO_CS
+		 * * SPI_CS_HIGH: cs are always asserted low
+		 * * cs_change: cs is deasserted after each spi_transfer
+		 * * cs_delay_usec: cs is always deasserted one SCK cycle
+		 *     after the last transfer
+		 * probably more...
+		 */
+		if (spi->chip_select == 0) {
+			dev_err(&spi->dev,
+				"native CS is not supported but may work for some use-cases for cs = 0 -  please define a valid value cs-gpios in DT for complete feature set\n");
+			return 0;
These messages are rather too long, please keep it in the ~ 80 char
limit (in source code and dmesg).
Or make it two lines ...

Since we return with 0, i prefer to use dev_warn instead of dev_err.

Thanks
+		}
+		dev_err(&spi->dev,
+			"native CS is not supported for cs > 0 -  please define a valid value cs-gpios in DT to enable multiple cs\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help