spi: omap2-mcspi: Use dma_request_chan() for requesting DMA channel

With the new dma_request_chan() the client driver does not need to look for
the DMA resource and it does not need to pass filter_fn anymore.
By switching to the new API the driver can now support deferred probing
against DMA.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
This commit is contained in:
Peter Ujfalusi 2016-04-29 16:11:56 +03:00 committed by Mark Brown
parent f55532a0c0
commit b085c6129a

View File

@ -23,7 +23,6 @@
#include <linux/delay.h>
#include <linux/dma-mapping.h>
#include <linux/dmaengine.h>
#include <linux/omap-dma.h>
#include <linux/pinctrl/consumer.h>
#include <linux/platform_device.h>
#include <linux/err.h>
@ -103,9 +102,6 @@ struct omap2_mcspi_dma {
struct dma_chan *dma_tx;
struct dma_chan *dma_rx;
int dma_tx_sync_dev;
int dma_rx_sync_dev;
struct completion dma_tx_completion;
struct completion dma_rx_completion;
@ -964,8 +960,7 @@ static int omap2_mcspi_request_dma(struct spi_device *spi)
struct spi_master *master = spi->master;
struct omap2_mcspi *mcspi;
struct omap2_mcspi_dma *mcspi_dma;
dma_cap_mask_t mask;
unsigned sig;
int ret = 0;
mcspi = spi_master_get_devdata(master);
mcspi_dma = mcspi->dma_channels + spi->chip_select;
@ -973,34 +968,25 @@ static int omap2_mcspi_request_dma(struct spi_device *spi)
init_completion(&mcspi_dma->dma_rx_completion);
init_completion(&mcspi_dma->dma_tx_completion);
dma_cap_zero(mask);
dma_cap_set(DMA_SLAVE, mask);
sig = mcspi_dma->dma_rx_sync_dev;
mcspi_dma->dma_rx =
dma_request_slave_channel_compat(mask, omap_dma_filter_fn,
&sig, &master->dev,
mcspi_dma->dma_rx_ch_name);
if (!mcspi_dma->dma_rx)
goto no_dma;
sig = mcspi_dma->dma_tx_sync_dev;
mcspi_dma->dma_tx =
dma_request_slave_channel_compat(mask, omap_dma_filter_fn,
&sig, &master->dev,
mcspi_dma->dma_tx_ch_name);
if (!mcspi_dma->dma_tx) {
dma_release_channel(mcspi_dma->dma_rx);
mcspi_dma->dma_rx = dma_request_chan(&master->dev,
mcspi_dma->dma_rx_ch_name);
if (IS_ERR(mcspi_dma->dma_rx)) {
ret = PTR_ERR(mcspi_dma->dma_rx);
mcspi_dma->dma_rx = NULL;
goto no_dma;
}
return 0;
mcspi_dma->dma_tx = dma_request_chan(&master->dev,
mcspi_dma->dma_tx_ch_name);
if (IS_ERR(mcspi_dma->dma_tx)) {
ret = PTR_ERR(mcspi_dma->dma_tx);
mcspi_dma->dma_tx = NULL;
dma_release_channel(mcspi_dma->dma_rx);
mcspi_dma->dma_rx = NULL;
}
no_dma:
dev_warn(&spi->dev, "not using DMA for McSPI\n");
return -EAGAIN;
return ret;
}
static int omap2_mcspi_setup(struct spi_device *spi)
@ -1039,8 +1025,9 @@ static int omap2_mcspi_setup(struct spi_device *spi)
if (!mcspi_dma->dma_rx || !mcspi_dma->dma_tx) {
ret = omap2_mcspi_request_dma(spi);
if (ret < 0 && ret != -EAGAIN)
return ret;
if (ret)
dev_warn(&spi->dev, "not using DMA for McSPI (%d)\n",
ret);
}
ret = pm_runtime_get_sync(mcspi->dev);
@ -1434,42 +1421,8 @@ static int omap2_mcspi_probe(struct platform_device *pdev)
}
for (i = 0; i < master->num_chipselect; i++) {
char *dma_rx_ch_name = mcspi->dma_channels[i].dma_rx_ch_name;
char *dma_tx_ch_name = mcspi->dma_channels[i].dma_tx_ch_name;
struct resource *dma_res;
sprintf(dma_rx_ch_name, "rx%d", i);
if (!pdev->dev.of_node) {
dma_res =
platform_get_resource_byname(pdev,
IORESOURCE_DMA,
dma_rx_ch_name);
if (!dma_res) {
dev_dbg(&pdev->dev,
"cannot get DMA RX channel\n");
status = -ENODEV;
break;
}
mcspi->dma_channels[i].dma_rx_sync_dev =
dma_res->start;
}
sprintf(dma_tx_ch_name, "tx%d", i);
if (!pdev->dev.of_node) {
dma_res =
platform_get_resource_byname(pdev,
IORESOURCE_DMA,
dma_tx_ch_name);
if (!dma_res) {
dev_dbg(&pdev->dev,
"cannot get DMA TX channel\n");
status = -ENODEV;
break;
}
mcspi->dma_channels[i].dma_tx_sync_dev =
dma_res->start;
}
sprintf(mcspi->dma_channels[i].dma_rx_ch_name, "rx%d", i);
sprintf(mcspi->dma_channels[i].dma_tx_ch_name, "tx%d", i);
}
if (status < 0)