[RFC][PATCH 1/3] spi: let master a chance to setup device before registering it

[RFC][PATCH 1/3] spi: let master a chance to setup device before registering it

[Paul Fertser]

Call setup() before registering the device because registering leads
to a probe routine of the slave device and that requires an
initialised master.

Signed-off-by: Paul Fertser <fercerpav@gmail.com>
---

TBH, i do not fully understand the framework, so this and the following patches
might be rather wrong. But the issues i tried to fix were real indeed :)

 drivers/spi/spi.c |    5 ++---
 1 files changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c
index b561f9d..82393ea 100644
--- a/drivers/spi/spi.c
+++ b/drivers/spi/spi.c
@@ -80,7 +80,6 @@ struct spi_device *spi_new_device(struct spi_master *master,
 	strcpy(proxy->dev.name, chip->name);
 	proxy->dev.type_data = proxy;
 	dev_add_child(master->dev, &proxy->dev);
-	status = register_device(&proxy->dev);
 
 	/* drivers may modify this initial i/o setup */
 	status = master->setup(proxy);
@@ -90,10 +89,10 @@ struct spi_device *spi_new_device(struct spi_master *master,
 		goto fail;
 	}
 
-	return proxy;
+	register_device(&proxy->dev);
 
+	return proxy;
 fail:
-	unregister_device(&proxy->dev);
 	free(proxy);
 	return NULL;
 }
-- 
1.5.2.2


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[RFC][PATCH 2/3] spi: enfore default bits_per_word value

[Paul Fertser]

Documentation says it should default to 8 when not specified
explicitly by the device data.

Signed-off-by: Paul Fertser <fercerpav@gmail.com>
---
 drivers/spi/spi.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c
index 82393ea..7a8aed4 100644
--- a/drivers/spi/spi.c
+++ b/drivers/spi/spi.c
@@ -75,7 +75,7 @@ struct spi_device *spi_new_device(struct spi_master *master,
 	proxy->chip_select = chip->chip_select;
 	proxy->max_speed_hz = chip->max_speed_hz;
 	proxy->mode = chip->mode;
-	proxy->bits_per_word = chip->bits_per_word;
+	proxy->bits_per_word = chip->bits_per_word ? chip->bits_per_word : 8;
 	proxy->dev.platform_data = chip->platform_data;
 	strcpy(proxy->dev.name, chip->name);
 	proxy->dev.type_data = proxy;
-- 
1.5.2.2


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[RFC][PATCH 3/3] spi: indicate in the docs that per-transfer bpw setting is not supported

[Paul Fertser]

Currently it's not implemented by any driver so add a note to the
documentation.

Signed-off-by: Paul Fertser <fercerpav@gmail.com>
---
 include/spi/spi.h |    3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)

diff --git a/include/spi/spi.h b/include/spi/spi.h
index 948e050..9d01d06 100644
--- a/include/spi/spi.h
+++ b/include/spi/spi.h
@@ -37,7 +37,8 @@ struct spi_board_info {
  *	powers of two bytes (e.g. 20 bit samples use 32 bits).
  *	This may be changed by the device's driver, or left at the
  *	default (0) indicating protocol words are eight bit bytes.
- *	The spi_transfer.bits_per_word can override this for each transfer.
+ *	The spi_transfer.bits_per_word can override this for each transfer
+ *	(FIXME: not currently implemented).
  * @irq: Negative, or the number passed to request_irq() to receive
  *	interrupts from this device.
  * @controller_state: Controller's runtime state
-- 
1.5.2.2


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox