On 05/20/2016 08:54 PM, Simon Glass wrote:
Hi Vignesh,
On 19 May 2016 at 23:01, Vignesh R vigneshr@ti.com wrote:
Convert davinci_spi driver so that it complies with SPI DM framework.
Signed-off-by: Vignesh R vigneshr@ti.com Reviewed-by: Tom Rini trini@konsulko.com
v5: correct error message.
v4: Check error returned by dev_map_physmem().
v3: No changes
v2: Add comments to struct davinci_spi_slave members. Use dev_map_physmem() added by previous patch.
drivers/spi/davinci_spi.c | 329 +++++++++++++++++++++++++++++++++------------- 1 file changed, 240 insertions(+), 89 deletions(-)
Sorry I missed something below. [snip]
+static int davinci_ofdata_to_platadata(struct udevice *bus) +{
struct davinci_spi_slave *ds = dev_get_priv(bus);const void *blob = gd->fdt_blob;int node = bus->of_offset;ds->regs = dev_map_physmem(bus, sizeof(struct davinci_spi_regs));if (!ds->regs) {printf("%s: could not map device address\n", __func__);return -ENODEV;-EINVAL
We use -ENODEV to say there is no device. Here, we have a device but the configuration is wrong.
Ok, I posted in v6 with above change.