On Fri, Jan 23, 2015 at 3:12 AM, Dmitry Osipenko <digetx@xxxxxxxxx> wrote:
> Support CPU BE mode by adding endianness conversion for memcpy interactions.
>
> Signed-off-by: Dmitry Osipenko <digetx@xxxxxxxxx>
> ---
> Changelog:
> v2: For consistency cpu_to_le32() changed to le32_to_cpu() because
> i2c_writel() takes BE value in BE CPU mode and value is in LE format.
>
> drivers/i2c/busses/i2c-tegra.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/drivers/i2c/busses/i2c-tegra.c b/drivers/i2c/busses/i2c-tegra.c
> index 28b87e6..b5afb9f 100644
> --- a/drivers/i2c/busses/i2c-tegra.c
> +++ b/drivers/i2c/busses/i2c-tegra.c
> @@ -286,6 +286,7 @@ static int tegra_i2c_empty_rx_fifo(struct tegra_i2c_dev *i2c_dev)
> if (rx_fifo_avail > 0 && buf_remaining > 0) {
> BUG_ON(buf_remaining > 3);
> val = i2c_readl(i2c_dev, I2C_RX_FIFO);
> + val = cpu_to_le32(val);
> memcpy(buf, &val, buf_remaining);
> buf_remaining = 0;
> rx_fifo_avail--;
> @@ -343,7 +344,9 @@ static int tegra_i2c_fill_tx_fifo(struct tegra_i2c_dev *i2c_dev)
> */
> if (tx_fifo_avail > 0 && buf_remaining > 0) {
> BUG_ON(buf_remaining > 3);
> + val = 0;
> memcpy(&val, buf, buf_remaining);
> + val = le32_to_cpu(val);
Ok, with le32_to_cpu() this makes sense now. memcpying part of an int
will keep the lowest byte at the lowest address, so this should work
for little-endian.
Acked-by: Alexandre Courbot <acourbot@xxxxxxxxxx>
--
To unsubscribe from this list: send the line "unsubscribe linux-i2c" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
