There are SoC specific ways to optimize transfers. Make the way free to implement these by creating SoC specific transfer functions. Signed-off-by: Sascha Hauer <s.hauer@xxxxxxxxxxxxxx> --- drivers/spi/imx_spi.c | 45 +++++++++++++++++++++++++++++++-------------- 1 file changed, 31 insertions(+), 14 deletions(-) diff --git a/drivers/spi/imx_spi.c b/drivers/spi/imx_spi.c index 81fa3b0..9a3f05c 100644 --- a/drivers/spi/imx_spi.c +++ b/drivers/spi/imx_spi.c @@ -39,11 +39,18 @@ struct imx_spi { struct clk *clk; unsigned int (*xchg_single)(struct imx_spi *imx, u32 data); + void (*do_transfer)(struct spi_device *spi); void (*chipselect)(struct spi_device *spi, int active); + + const void *tx_buf; + void *rx_buf; + int xfer_len; + int bits_per_word; }; struct spi_imx_devtype_data { unsigned int (*xchg_single)(struct imx_spi *imx, u32 data); + void (*do_transfer)(struct spi_device *spi); void (*chipselect)(struct spi_device *spi, int active); void (*init)(struct imx_spi *imx); }; @@ -332,38 +339,39 @@ static u32 imx_xchg_single(struct spi_device *spi, u32 tx_val) return imx_spi_maybe_reverse_bits(spi, rx_val); } -static void imx_spi_do_transfer(struct spi_device *spi, struct spi_transfer *t) +static void imx_spi_do_transfer(struct spi_device *spi) { + struct imx_spi *imx = container_of(spi->master, struct imx_spi, master); unsigned i; - if (spi->bits_per_word <= 8) { - const u8 *tx_buf = t->tx_buf; - u8 *rx_buf = t->rx_buf; + if (imx->bits_per_word <= 8) { + const u8 *tx_buf = imx->tx_buf; + u8 *rx_buf = imx->rx_buf; u8 rx_val; - for (i = 0; i < t->len; i++) { + for (i = 0; i < imx->xfer_len; i++) { rx_val = imx_xchg_single(spi, tx_buf ? tx_buf[i] : 0); if (rx_buf) rx_buf[i] = rx_val; } - } else if (spi->bits_per_word <= 16) { - const u16 *tx_buf = t->tx_buf; - u16 *rx_buf = t->rx_buf; + } else if (imx->bits_per_word <= 16) { + const u16 *tx_buf = imx->tx_buf; + u16 *rx_buf = imx->rx_buf; u16 rx_val; - for (i = 0; i < t->len >> 1; i++) { + for (i = 0; i < imx->xfer_len >> 1; i++) { rx_val = imx_xchg_single(spi, tx_buf ? tx_buf[i] : 0); if (rx_buf) rx_buf[i] = rx_val; } - } else if (spi->bits_per_word <= 32) { - const u32 *tx_buf = t->tx_buf; - u32 *rx_buf = t->rx_buf; + } else if (imx->bits_per_word <= 32) { + const u32 *tx_buf = imx->tx_buf; + u32 *rx_buf = imx->rx_buf; u32 rx_val; - for (i = 0; i < t->len >> 2; i++) { + for (i = 0; i < imx->xfer_len >> 2; i++) { rx_val = imx_xchg_single(spi, tx_buf ? tx_buf[i] : 0); if (rx_buf) @@ -395,7 +403,12 @@ static int imx_spi_transfer(struct spi_device *spi, struct spi_message *mesg) cs_change = t->cs_change; - imx_spi_do_transfer(spi, t); + imx->tx_buf = t->tx_buf; + imx->rx_buf = t->rx_buf; + imx->xfer_len = t->len; + imx->bits_per_word = spi->bits_per_word; + imx->do_transfer(spi); + mesg->actual_length += t->len; if (cs_change) @@ -411,17 +424,20 @@ static int imx_spi_transfer(struct spi_device *spi, struct spi_message *mesg) static __maybe_unused struct spi_imx_devtype_data spi_imx_devtype_data_0_0 = { .chipselect = cspi_0_0_chipselect, .xchg_single = cspi_0_0_xchg_single, + .do_transfer = imx_spi_do_transfer, .init = cspi_0_0_init, }; static __maybe_unused struct spi_imx_devtype_data spi_imx_devtype_data_0_7 = { .chipselect = cspi_0_7_chipselect, .xchg_single = cspi_0_7_xchg_single, + .do_transfer = imx_spi_do_transfer, .init = cspi_0_7_init, }; static __maybe_unused struct spi_imx_devtype_data spi_imx_devtype_data_2_3 = { .chipselect = cspi_2_3_chipselect, + .do_transfer = imx_spi_do_transfer, .xchg_single = cspi_2_3_xchg_single, }; @@ -486,6 +502,7 @@ static int imx_spi_probe(struct device_d *dev) imx->chipselect = devdata->chipselect; imx->xchg_single = devdata->xchg_single; + imx->do_transfer = devdata->do_transfer; imx->regs = dev_request_mem_region(dev, 0); if (devdata->init) -- 2.5.0 _______________________________________________ barebox mailing list barebox@xxxxxxxxxxxxxxxxxxx http://lists.infradead.org/mailman/listinfo/barebox