diff mbox series

[v6,07/17] spi: axi-spi-engine: implement offload support

Message ID 20241211-dlech-mainline-spi-engine-offload-2-v6-7-88ee574d5d03@baylibre.com (mailing list archive)
State Changes Requested
Headers show
Series spi: axi-spi-engine: add offload support | expand

Commit Message

David Lechner Dec. 11, 2024, 8:54 p.m. UTC
Implement SPI offload support for the AXI SPI Engine. Currently, the
hardware only supports triggering offload transfers with a hardware
trigger so attempting to use an offload message in the regular SPI
message queue will fail. Also, only allows streaming rx data to an
external sink, so attempts to use a rx_buf in the offload message will
fail.

Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: David Lechner <dlechner@baylibre.com>
---

v6 changes:
* Update for split spi/offload headers.

v5 changes:
* Set offload capability flags based on DT properties.
* Add support for TX DMA since the hardware supports that now.
* Update for changes in other patches in the series.

v4 changes:
* Adapted to changes in other patches in the series.
* Moved trigger enable/disable to same function as offload
  enable/disable.

v3 changes:
* Added clk and dma_chan getter callbacks.
* Fixed some bugs.

v2 changes:

This patch has been reworked to accommodate the changes described in all
of the other patches.
---
 drivers/spi/Kconfig              |   1 +
 drivers/spi/spi-axi-spi-engine.c | 314 ++++++++++++++++++++++++++++++++++++++-
 2 files changed, 308 insertions(+), 7 deletions(-)

Comments

Nuno Sá Dec. 17, 2024, 11:48 a.m. UTC | #1
On Wed, 2024-12-11 at 14:54 -0600, David Lechner wrote:
> Implement SPI offload support for the AXI SPI Engine. Currently, the
> hardware only supports triggering offload transfers with a hardware
> trigger so attempting to use an offload message in the regular SPI
> message queue will fail. Also, only allows streaming rx data to an
> external sink, so attempts to use a rx_buf in the offload message will
> fail.
> 
> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
> Signed-off-by: David Lechner <dlechner@baylibre.com>
> ---

LGTM,

Reviewed-by: Nuno Sa <nuno.sa@analog.com>

> 
> v6 changes:
> * Update for split spi/offload headers.
> 
> v5 changes:
> * Set offload capability flags based on DT properties.
> * Add support for TX DMA since the hardware supports that now.
> * Update for changes in other patches in the series.
> 
> v4 changes:
> * Adapted to changes in other patches in the series.
> * Moved trigger enable/disable to same function as offload
>   enable/disable.
> 
> v3 changes:
> * Added clk and dma_chan getter callbacks.
> * Fixed some bugs.
> 
> v2 changes:
> 
> This patch has been reworked to accommodate the changes described in all
> of the other patches.
> ---
>  drivers/spi/Kconfig              |   1 +
>  drivers/spi/spi-axi-spi-engine.c | 314
> ++++++++++++++++++++++++++++++++++++++-
>  2 files changed, 308 insertions(+), 7 deletions(-)
> 
> diff --git a/drivers/spi/Kconfig b/drivers/spi/Kconfig
> index
> 2cfc14be869790f5226130428bb7cb40aadfb031..f496ab127ef011d092f66063e05772725ab8
> 9771 100644
> --- a/drivers/spi/Kconfig
> +++ b/drivers/spi/Kconfig
> @@ -179,6 +179,7 @@ config SPI_AU1550
>  config SPI_AXI_SPI_ENGINE
>  	tristate "Analog Devices AXI SPI Engine controller"
>  	depends on HAS_IOMEM
> +	select SPI_OFFLOAD
>  	help
>  	  This enables support for the Analog Devices AXI SPI Engine SPI
> controller.
>  	  It is part of the SPI Engine framework that is used in some Analog
> Devices
> diff --git a/drivers/spi/spi-axi-spi-engine.c b/drivers/spi/spi-axi-spi-
> engine.c
> index
> 7c252126b33ea83fe6a6e80c6cb87499243069f5..dd6077d3ff7b8d29b0ca2e803a5930c4cedf
> 2e93 100644
> --- a/drivers/spi/spi-axi-spi-engine.c
> +++ b/drivers/spi/spi-axi-spi-engine.c
> @@ -2,11 +2,14 @@
>  /*
>   * SPI-Engine SPI controller driver
>   * Copyright 2015 Analog Devices Inc.
> + * Copyright 2024 BayLibre, SAS
>   *  Author: Lars-Peter Clausen <lars@metafoo.de>
>   */
>  
> +#include <linux/bitops.h>
>  #include <linux/clk.h>
>  #include <linux/completion.h>
> +#include <linux/dmaengine.h>
>  #include <linux/fpga/adi-axi-common.h>
>  #include <linux/interrupt.h>
>  #include <linux/io.h>
> @@ -14,9 +17,11 @@
>  #include <linux/module.h>
>  #include <linux/overflow.h>
>  #include <linux/platform_device.h>
> +#include <linux/spi/offload/provider.h>
>  #include <linux/spi/spi.h>
>  #include <trace/events/spi.h>
>  
> +#define SPI_ENGINE_REG_OFFLOAD_MEM_ADDR_WIDTH	0x10
>  #define SPI_ENGINE_REG_RESET			0x40
>  
>  #define SPI_ENGINE_REG_INT_ENABLE		0x80
> @@ -24,6 +29,7 @@
>  #define SPI_ENGINE_REG_INT_SOURCE		0x88
>  
>  #define SPI_ENGINE_REG_SYNC_ID			0xc0
> +#define SPI_ENGINE_REG_OFFLOAD_SYNC_ID		0xc4
>  
>  #define SPI_ENGINE_REG_CMD_FIFO_ROOM		0xd0
>  #define SPI_ENGINE_REG_SDO_FIFO_ROOM		0xd4
> @@ -34,10 +40,24 @@
>  #define SPI_ENGINE_REG_SDI_DATA_FIFO		0xe8
>  #define SPI_ENGINE_REG_SDI_DATA_FIFO_PEEK	0xec
>  
> +#define SPI_ENGINE_MAX_NUM_OFFLOADS		32
> +
> +#define SPI_ENGINE_REG_OFFLOAD_CTRL(x)		(0x100 +
> SPI_ENGINE_MAX_NUM_OFFLOADS * (x))
> +#define SPI_ENGINE_REG_OFFLOAD_STATUS(x)	(0x104 +
> SPI_ENGINE_MAX_NUM_OFFLOADS * (x))
> +#define SPI_ENGINE_REG_OFFLOAD_RESET(x)		(0x108 +
> SPI_ENGINE_MAX_NUM_OFFLOADS * (x))
> +#define SPI_ENGINE_REG_OFFLOAD_CMD_FIFO(x)	(0x110 +
> SPI_ENGINE_MAX_NUM_OFFLOADS * (x))
> +#define SPI_ENGINE_REG_OFFLOAD_SDO_FIFO(x)	(0x114 +
> SPI_ENGINE_MAX_NUM_OFFLOADS * (x))
> +
> +#define SPI_ENGINE_SPI_OFFLOAD_MEM_WIDTH_SDO	GENMASK(15, 8)
> +#define SPI_ENGINE_SPI_OFFLOAD_MEM_WIDTH_CMD	GENMASK(7, 0)
> +
>  #define SPI_ENGINE_INT_CMD_ALMOST_EMPTY		BIT(0)
>  #define SPI_ENGINE_INT_SDO_ALMOST_EMPTY		BIT(1)
>  #define SPI_ENGINE_INT_SDI_ALMOST_FULL		BIT(2)
>  #define SPI_ENGINE_INT_SYNC			BIT(3)
> +#define SPI_ENGINE_INT_OFFLOAD_SYNC		BIT(4)
> +
> +#define SPI_ENGINE_OFFLOAD_CTRL_ENABLE		BIT(0)
>  
>  #define SPI_ENGINE_CONFIG_CPHA			BIT(0)
>  #define SPI_ENGINE_CONFIG_CPOL			BIT(1)
> @@ -79,6 +99,10 @@
>  #define SPI_ENGINE_CMD_CS_INV(flags) \
>  	SPI_ENGINE_CMD(SPI_ENGINE_INST_CS_INV, 0, (flags))
>  
> +/* default sizes - can be changed when SPI Engine firmware is compiled */
> +#define SPI_ENGINE_OFFLOAD_CMD_FIFO_SIZE	16
> +#define SPI_ENGINE_OFFLOAD_SDO_FIFO_SIZE	16
> +
>  struct spi_engine_program {
>  	unsigned int length;
>  	uint16_t instructions[] __counted_by(length);
> @@ -106,6 +130,17 @@ struct spi_engine_message_state {
>  	uint8_t *rx_buf;
>  };
>  
> +enum {
> +	SPI_ENGINE_OFFLOAD_FLAG_ASSIGNED,
> +	SPI_ENGINE_OFFLOAD_FLAG_PREPARED,
> +};
> +
> +struct spi_engine_offload {
> +	struct spi_engine *spi_engine;
> +	unsigned long flags;
> +	unsigned int offload_num;
> +};
> +
>  struct spi_engine {
>  	struct clk *clk;
>  	struct clk *ref_clk;
> @@ -118,6 +153,11 @@ struct spi_engine {
>  	unsigned int int_enable;
>  	/* shadows hardware CS inversion flag state */
>  	u8 cs_inv;
> +
> +	unsigned int offload_ctrl_mem_size;
> +	unsigned int offload_sdo_mem_size;
> +	struct spi_offload *offload;
> +	u32 offload_caps;
>  };
>  
>  static void spi_engine_program_add_cmd(struct spi_engine_program *p,
> @@ -163,9 +203,9 @@ static void spi_engine_gen_xfer(struct spi_engine_program
> *p, bool dry,
>  		unsigned int n = min(len, 256U);
>  		unsigned int flags = 0;
>  
> -		if (xfer->tx_buf)
> +		if (xfer->tx_buf || (xfer->offload_flags &
> SPI_OFFLOAD_XFER_TX_STREAM))
>  			flags |= SPI_ENGINE_TRANSFER_WRITE;
> -		if (xfer->rx_buf)
> +		if (xfer->rx_buf || (xfer->offload_flags &
> SPI_OFFLOAD_XFER_RX_STREAM))
>  			flags |= SPI_ENGINE_TRANSFER_READ;
>  
>  		spi_engine_program_add_cmd(p, dry,
> @@ -217,16 +257,24 @@ static void spi_engine_gen_cs(struct spi_engine_program
> *p, bool dry,
>   *
>   * NB: This is separate from spi_engine_compile_message() because the latter
>   * is called twice and would otherwise result in double-evaluation.
> + *
> + * Returns 0 on success, -EINVAL on failure.
>   */
> -static void spi_engine_precompile_message(struct spi_message *msg)
> +static int spi_engine_precompile_message(struct spi_message *msg)
>  {
>  	unsigned int clk_div, max_hz = msg->spi->controller->max_speed_hz;
>  	struct spi_transfer *xfer;
>  
>  	list_for_each_entry(xfer, &msg->transfers, transfer_list) {
> +		/* If we have an offload transfer, we can't rx to buffer */
> +		if (msg->offload && xfer->rx_buf)
> +			return -EINVAL;
> +
>  		clk_div = DIV_ROUND_UP(max_hz, xfer->speed_hz);
>  		xfer->effective_speed_hz = max_hz / min(clk_div, 256U);
>  	}
> +
> +	return 0;
>  }
>  
>  static void spi_engine_compile_message(struct spi_message *msg, bool dry,
> @@ -521,11 +569,105 @@ static irqreturn_t spi_engine_irq(int irq, void *devid)
>  	return IRQ_HANDLED;
>  }
>  
> +static int spi_engine_offload_prepare(struct spi_message *msg)
> +{
> +	struct spi_controller *host = msg->spi->controller;
> +	struct spi_engine *spi_engine = spi_controller_get_devdata(host);
> +	struct spi_engine_program *p = msg->opt_state;
> +	struct spi_engine_offload *priv = msg->offload->priv;
> +	struct spi_transfer *xfer;
> +	void __iomem *cmd_addr;
> +	void __iomem *sdo_addr;
> +	size_t tx_word_count = 0;
> +	unsigned int i;
> +
> +	if (p->length > spi_engine->offload_ctrl_mem_size)
> +		return -EINVAL;
> +
> +	/* count total number of tx words in message */
> +	list_for_each_entry(xfer, &msg->transfers, transfer_list) {
> +		/* no support for reading to rx_buf */
> +		if (xfer->rx_buf)
> +			return -EINVAL;
> +
> +		if (!xfer->tx_buf)
> +			continue;
> +
> +		if (xfer->bits_per_word <= 8)
> +			tx_word_count += xfer->len;
> +		else if (xfer->bits_per_word <= 16)
> +			tx_word_count += xfer->len / 2;
> +		else
> +			tx_word_count += xfer->len / 4;
> +	}
> +
> +	if (tx_word_count && !(spi_engine->offload_caps &
> SPI_OFFLOAD_CAP_TX_STATIC_DATA))
> +		return -EINVAL;
> +
> +	if (tx_word_count > spi_engine->offload_sdo_mem_size)
> +		return -EINVAL;
> +
> +	/*
> +	 * This protects against calling spi_optimize_message() with an
> offload
> +	 * that has already been prepared with a different message.
> +	 */
> +	if (test_and_set_bit_lock(SPI_ENGINE_OFFLOAD_FLAG_PREPARED, &priv-
> >flags))
> +		return -EBUSY;
> +
> +	cmd_addr = spi_engine->base +
> +		   SPI_ENGINE_REG_OFFLOAD_CMD_FIFO(priv->offload_num);
> +	sdo_addr = spi_engine->base +
> +		   SPI_ENGINE_REG_OFFLOAD_SDO_FIFO(priv->offload_num);
> +
> +	list_for_each_entry(xfer, &msg->transfers, transfer_list) {
> +		if (!xfer->tx_buf)
> +			continue;
> +
> +		if (xfer->bits_per_word <= 8) {
> +			const u8 *buf = xfer->tx_buf;
> +
> +			for (i = 0; i < xfer->len; i++)
> +				writel_relaxed(buf[i], sdo_addr);
> +		} else if (xfer->bits_per_word <= 16) {
> +			const u16 *buf = xfer->tx_buf;
> +
> +			for (i = 0; i < xfer->len / 2; i++)
> +				writel_relaxed(buf[i], sdo_addr);
> +		} else {
> +			const u32 *buf = xfer->tx_buf;
> +
> +			for (i = 0; i < xfer->len / 4; i++)
> +				writel_relaxed(buf[i], sdo_addr);
> +		}
> +	}
> +
> +	for (i = 0; i < p->length; i++)
> +		writel_relaxed(p->instructions[i], cmd_addr);
> +
> +	return 0;
> +}
> +
> +static void spi_engine_offload_unprepare(struct spi_offload *offload)
> +{
> +	struct spi_engine_offload *priv = offload->priv;
> +	struct spi_engine *spi_engine = priv->spi_engine;
> +
> +	writel_relaxed(1, spi_engine->base +
> +			  SPI_ENGINE_REG_OFFLOAD_RESET(priv->offload_num));
> +	writel_relaxed(0, spi_engine->base +
> +			  SPI_ENGINE_REG_OFFLOAD_RESET(priv->offload_num));
> +
> +	clear_bit_unlock(SPI_ENGINE_OFFLOAD_FLAG_PREPARED, &priv->flags);
> +}
> +
>  static int spi_engine_optimize_message(struct spi_message *msg)
>  {
>  	struct spi_engine_program p_dry, *p;
> +	int ret;
>  
> -	spi_engine_precompile_message(msg);
> +	ret = spi_engine_precompile_message(msg);
> +	if (ret)
> +		return ret;
>  
>  	p_dry.length = 0;
>  	spi_engine_compile_message(msg, true, &p_dry);
> @@ -537,20 +679,61 @@ static int spi_engine_optimize_message(struct
> spi_message *msg)
>  	spi_engine_compile_message(msg, false, p);
>  
>  	spi_engine_program_add_cmd(p, false, SPI_ENGINE_CMD_SYNC(
> -
> 						AXI_SPI_ENGINE_CUR_MSG_SYNC_ID));
> +		msg->offload ? 0 : AXI_SPI_ENGINE_CUR_MSG_SYNC_ID));
>  
>  	msg->opt_state = p;
>  
> +	if (msg->offload) {
> +		ret = spi_engine_offload_prepare(msg);
> +		if (ret) {
> +			msg->opt_state = NULL;
> +			kfree(p);
> +			return ret;
> +		}
> +	}
> +
>  	return 0;
>  }
>  
>  static int spi_engine_unoptimize_message(struct spi_message *msg)
>  {
> +	if (msg->offload)
> +		spi_engine_offload_unprepare(msg->offload);
> +
>  	kfree(msg->opt_state);
>  
>  	return 0;
>  }
>  
> +static struct spi_offload
> +*spi_engine_get_offload(struct spi_device *spi,
> +			const struct spi_offload_config *config)
> +{
> +	struct spi_controller *host = spi->controller;
> +	struct spi_engine *spi_engine = spi_controller_get_devdata(host);
> +	struct spi_engine_offload *priv;
> +
> +	if (!spi_engine->offload)
> +		return ERR_PTR(-ENODEV);
> +
> +	if (config->capability_flags & ~spi_engine->offload_caps)
> +		return ERR_PTR(-EINVAL);
> +
> +	priv = spi_engine->offload->priv;
> +
> +	if (test_and_set_bit_lock(SPI_ENGINE_OFFLOAD_FLAG_ASSIGNED, &priv-
> >flags))
> +		return ERR_PTR(-EBUSY);
> +
> +	return spi_engine->offload;
> +}
> +
> +static void spi_engine_put_offload(struct spi_offload *offload)
> +{
> +	struct spi_engine_offload *priv = offload->priv;
> +
> +	clear_bit_unlock(SPI_ENGINE_OFFLOAD_FLAG_ASSIGNED, &priv->flags);
> +}
> +
>  static int spi_engine_setup(struct spi_device *device)
>  {
>  	struct spi_controller *host = device->controller;
> @@ -583,6 +766,12 @@ static int spi_engine_transfer_one_message(struct
> spi_controller *host,
>  	unsigned int int_enable = 0;
>  	unsigned long flags;
>  
> +	if (msg->offload) {
> +		dev_err(&host->dev, "Single transfer offload not
> supported\n");
> +		msg->status = -EOPNOTSUPP;
> +		goto out;
> +	}
> +
>  	/* reinitialize message state for this transfer */
>  	memset(st, 0, sizeof(*st));
>  	st->cmd_buf = p->instructions;
> @@ -632,11 +821,68 @@ static int spi_engine_transfer_one_message(struct
> spi_controller *host,
>  			trace_spi_transfer_stop(msg, xfer);
>  	}
>  
> +out:
>  	spi_finalize_current_message(host);
>  
>  	return msg->status;
>  }
>  
> +static int spi_engine_trigger_enable(struct spi_offload *offload)
> +{
> +	struct spi_engine_offload *priv = offload->priv;
> +	struct spi_engine *spi_engine = priv->spi_engine;
> +	unsigned int reg;
> +
> +	reg = readl_relaxed(spi_engine->base +
> +			    SPI_ENGINE_REG_OFFLOAD_CTRL(priv->offload_num));
> +	reg |= SPI_ENGINE_OFFLOAD_CTRL_ENABLE;
> +	writel_relaxed(reg, spi_engine->base +
> +			    SPI_ENGINE_REG_OFFLOAD_CTRL(priv->offload_num));
> +	return 0;
> +}
> +
> +static void spi_engine_trigger_disable(struct spi_offload *offload)
> +{
> +	struct spi_engine_offload *priv = offload->priv;
> +	struct spi_engine *spi_engine = priv->spi_engine;
> +	unsigned int reg;
> +
> +	reg = readl_relaxed(spi_engine->base +
> +			    SPI_ENGINE_REG_OFFLOAD_CTRL(priv->offload_num));
> +	reg &= ~SPI_ENGINE_OFFLOAD_CTRL_ENABLE;
> +	writel_relaxed(reg, spi_engine->base +
> +			    SPI_ENGINE_REG_OFFLOAD_CTRL(priv->offload_num));
> +}
> +
> +static struct dma_chan
> +*spi_engine_tx_stream_request_dma_chan(struct spi_offload *offload)
> +{
> +	struct spi_engine_offload *priv = offload->priv;
> +	char name[16];
> +
> +	snprintf(name, sizeof(name), "offload%u-tx", priv->offload_num);
> +
> +	return dma_request_chan(offload->provider_dev, name);
> +}
> +
> +static struct dma_chan
> +*spi_engine_rx_stream_request_dma_chan(struct spi_offload *offload)
> +{
> +	struct spi_engine_offload *priv = offload->priv;
> +	char name[16];
> +
> +	snprintf(name, sizeof(name), "offload%u-rx", priv->offload_num);
> +
> +	return dma_request_chan(offload->provider_dev, name);
> +}
> +
> +static const struct spi_offload_ops spi_engine_offload_ops = {
> +	.trigger_enable = spi_engine_trigger_enable,
> +	.trigger_disable = spi_engine_trigger_disable,
> +	.tx_stream_request_dma_chan = spi_engine_tx_stream_request_dma_chan,
> +	.rx_stream_request_dma_chan = spi_engine_rx_stream_request_dma_chan,
> +};
> +
>  static void spi_engine_release_hw(void *p)
>  {
>  	struct spi_engine *spi_engine = p;
> @@ -651,8 +897,7 @@ static int spi_engine_probe(struct platform_device *pdev)
>  	struct spi_engine *spi_engine;
>  	struct spi_controller *host;
>  	unsigned int version;
> -	int irq;
> -	int ret;
> +	int irq, ret;
>  
>  	irq = platform_get_irq(pdev, 0);
>  	if (irq < 0)
> @@ -667,6 +912,46 @@ static int spi_engine_probe(struct platform_device *pdev)
>  	spin_lock_init(&spi_engine->lock);
>  	init_completion(&spi_engine->msg_complete);
>  
> +	/*
> +	 * REVISIT: for now, all SPI Engines only have one offload. In the
> +	 * future, this should be read from a memory mapped register to
> +	 * determine the number of offloads enabled at HDL compile time. For
> +	 * now, we can tell if an offload is present if there is a trigger
> +	 * source wired up to it.
> +	 */
> +	if (device_property_present(&pdev->dev, "trigger-sources")) {
> +		struct spi_engine_offload *priv;
> +
> +		spi_engine->offload =
> +			devm_spi_offload_alloc(&pdev->dev,
> +					       sizeof(struct
> spi_engine_offload));
> +		if (IS_ERR(spi_engine->offload))
> +			return PTR_ERR(spi_engine->offload);
> +
> +		priv = spi_engine->offload->priv;
> +		priv->spi_engine = spi_engine;
> +		priv->offload_num = 0;
> +
> +		spi_engine->offload->ops = &spi_engine_offload_ops;
> +		spi_engine->offload_caps = SPI_OFFLOAD_CAP_TRIGGER;
> +
> +		if (device_property_match_string(&pdev->dev, "dma-names",
> "offload0-rx") >= 0) {
> +			spi_engine->offload_caps |=
> SPI_OFFLOAD_CAP_RX_STREAM_DMA;
> +			spi_engine->offload->xfer_flags |=
> SPI_OFFLOAD_XFER_RX_STREAM;
> +		}
> +
> +		if (device_property_match_string(&pdev->dev, "dma-names",
> "offload0-tx") >= 0) {
> +			spi_engine->offload_caps |=
> SPI_OFFLOAD_CAP_TX_STREAM_DMA;
> +			spi_engine->offload->xfer_flags |=
> SPI_OFFLOAD_XFER_TX_STREAM;
> +		} else {
> +			/*
> +			 * HDL compile option to enable TX DMA stream also
> disables
> +			 * the SDO memory, so can't do both at the same time.
> +			 */
> +			spi_engine->offload_caps |=
> SPI_OFFLOAD_CAP_TX_STATIC_DATA;
> +		}
> +	}
> +
>  	spi_engine->clk = devm_clk_get_enabled(&pdev->dev, "s_axi_aclk");
>  	if (IS_ERR(spi_engine->clk))
>  		return PTR_ERR(spi_engine->clk);
> @@ -688,6 +973,19 @@ static int spi_engine_probe(struct platform_device *pdev)
>  		return -ENODEV;
>  	}
>  
> +	if (ADI_AXI_PCORE_VER_MINOR(version) >= 1) {
> +		unsigned int sizes = readl(spi_engine->base +
> +				SPI_ENGINE_REG_OFFLOAD_MEM_ADDR_WIDTH);
> +
> +		spi_engine->offload_ctrl_mem_size = 1 <<
> +			FIELD_GET(SPI_ENGINE_SPI_OFFLOAD_MEM_WIDTH_CMD,
> sizes);
> +		spi_engine->offload_sdo_mem_size = 1 <<
> +			FIELD_GET(SPI_ENGINE_SPI_OFFLOAD_MEM_WIDTH_SDO,
> sizes);
> +	} else {
> +		spi_engine->offload_ctrl_mem_size =
> SPI_ENGINE_OFFLOAD_CMD_FIFO_SIZE;
> +		spi_engine->offload_sdo_mem_size =
> SPI_ENGINE_OFFLOAD_SDO_FIFO_SIZE;
> +	}
> +
>  	writel_relaxed(0x00, spi_engine->base + SPI_ENGINE_REG_RESET);
>  	writel_relaxed(0xff, spi_engine->base + SPI_ENGINE_REG_INT_PENDING);
>  	writel_relaxed(0x00, spi_engine->base + SPI_ENGINE_REG_INT_ENABLE);
> @@ -709,6 +1007,8 @@ static int spi_engine_probe(struct platform_device *pdev)
>  	host->transfer_one_message = spi_engine_transfer_one_message;
>  	host->optimize_message = spi_engine_optimize_message;
>  	host->unoptimize_message = spi_engine_unoptimize_message;
> +	host->get_offload = spi_engine_get_offload;
> +	host->put_offload = spi_engine_put_offload;
>  	host->num_chipselect = 8;
>  
>  	/* Some features depend of the IP core version. */
>
diff mbox series

Patch

diff --git a/drivers/spi/Kconfig b/drivers/spi/Kconfig
index 2cfc14be869790f5226130428bb7cb40aadfb031..f496ab127ef011d092f66063e05772725ab89771 100644
--- a/drivers/spi/Kconfig
+++ b/drivers/spi/Kconfig
@@ -179,6 +179,7 @@  config SPI_AU1550
 config SPI_AXI_SPI_ENGINE
 	tristate "Analog Devices AXI SPI Engine controller"
 	depends on HAS_IOMEM
+	select SPI_OFFLOAD
 	help
 	  This enables support for the Analog Devices AXI SPI Engine SPI controller.
 	  It is part of the SPI Engine framework that is used in some Analog Devices
diff --git a/drivers/spi/spi-axi-spi-engine.c b/drivers/spi/spi-axi-spi-engine.c
index 7c252126b33ea83fe6a6e80c6cb87499243069f5..dd6077d3ff7b8d29b0ca2e803a5930c4cedf2e93 100644
--- a/drivers/spi/spi-axi-spi-engine.c
+++ b/drivers/spi/spi-axi-spi-engine.c
@@ -2,11 +2,14 @@ 
 /*
  * SPI-Engine SPI controller driver
  * Copyright 2015 Analog Devices Inc.
+ * Copyright 2024 BayLibre, SAS
  *  Author: Lars-Peter Clausen <lars@metafoo.de>
  */
 
+#include <linux/bitops.h>
 #include <linux/clk.h>
 #include <linux/completion.h>
+#include <linux/dmaengine.h>
 #include <linux/fpga/adi-axi-common.h>
 #include <linux/interrupt.h>
 #include <linux/io.h>
@@ -14,9 +17,11 @@ 
 #include <linux/module.h>
 #include <linux/overflow.h>
 #include <linux/platform_device.h>
+#include <linux/spi/offload/provider.h>
 #include <linux/spi/spi.h>
 #include <trace/events/spi.h>
 
+#define SPI_ENGINE_REG_OFFLOAD_MEM_ADDR_WIDTH	0x10
 #define SPI_ENGINE_REG_RESET			0x40
 
 #define SPI_ENGINE_REG_INT_ENABLE		0x80
@@ -24,6 +29,7 @@ 
 #define SPI_ENGINE_REG_INT_SOURCE		0x88
 
 #define SPI_ENGINE_REG_SYNC_ID			0xc0
+#define SPI_ENGINE_REG_OFFLOAD_SYNC_ID		0xc4
 
 #define SPI_ENGINE_REG_CMD_FIFO_ROOM		0xd0
 #define SPI_ENGINE_REG_SDO_FIFO_ROOM		0xd4
@@ -34,10 +40,24 @@ 
 #define SPI_ENGINE_REG_SDI_DATA_FIFO		0xe8
 #define SPI_ENGINE_REG_SDI_DATA_FIFO_PEEK	0xec
 
+#define SPI_ENGINE_MAX_NUM_OFFLOADS		32
+
+#define SPI_ENGINE_REG_OFFLOAD_CTRL(x)		(0x100 + SPI_ENGINE_MAX_NUM_OFFLOADS * (x))
+#define SPI_ENGINE_REG_OFFLOAD_STATUS(x)	(0x104 + SPI_ENGINE_MAX_NUM_OFFLOADS * (x))
+#define SPI_ENGINE_REG_OFFLOAD_RESET(x)		(0x108 + SPI_ENGINE_MAX_NUM_OFFLOADS * (x))
+#define SPI_ENGINE_REG_OFFLOAD_CMD_FIFO(x)	(0x110 + SPI_ENGINE_MAX_NUM_OFFLOADS * (x))
+#define SPI_ENGINE_REG_OFFLOAD_SDO_FIFO(x)	(0x114 + SPI_ENGINE_MAX_NUM_OFFLOADS * (x))
+
+#define SPI_ENGINE_SPI_OFFLOAD_MEM_WIDTH_SDO	GENMASK(15, 8)
+#define SPI_ENGINE_SPI_OFFLOAD_MEM_WIDTH_CMD	GENMASK(7, 0)
+
 #define SPI_ENGINE_INT_CMD_ALMOST_EMPTY		BIT(0)
 #define SPI_ENGINE_INT_SDO_ALMOST_EMPTY		BIT(1)
 #define SPI_ENGINE_INT_SDI_ALMOST_FULL		BIT(2)
 #define SPI_ENGINE_INT_SYNC			BIT(3)
+#define SPI_ENGINE_INT_OFFLOAD_SYNC		BIT(4)
+
+#define SPI_ENGINE_OFFLOAD_CTRL_ENABLE		BIT(0)
 
 #define SPI_ENGINE_CONFIG_CPHA			BIT(0)
 #define SPI_ENGINE_CONFIG_CPOL			BIT(1)
@@ -79,6 +99,10 @@ 
 #define SPI_ENGINE_CMD_CS_INV(flags) \
 	SPI_ENGINE_CMD(SPI_ENGINE_INST_CS_INV, 0, (flags))
 
+/* default sizes - can be changed when SPI Engine firmware is compiled */
+#define SPI_ENGINE_OFFLOAD_CMD_FIFO_SIZE	16
+#define SPI_ENGINE_OFFLOAD_SDO_FIFO_SIZE	16
+
 struct spi_engine_program {
 	unsigned int length;
 	uint16_t instructions[] __counted_by(length);
@@ -106,6 +130,17 @@  struct spi_engine_message_state {
 	uint8_t *rx_buf;
 };
 
+enum {
+	SPI_ENGINE_OFFLOAD_FLAG_ASSIGNED,
+	SPI_ENGINE_OFFLOAD_FLAG_PREPARED,
+};
+
+struct spi_engine_offload {
+	struct spi_engine *spi_engine;
+	unsigned long flags;
+	unsigned int offload_num;
+};
+
 struct spi_engine {
 	struct clk *clk;
 	struct clk *ref_clk;
@@ -118,6 +153,11 @@  struct spi_engine {
 	unsigned int int_enable;
 	/* shadows hardware CS inversion flag state */
 	u8 cs_inv;
+
+	unsigned int offload_ctrl_mem_size;
+	unsigned int offload_sdo_mem_size;
+	struct spi_offload *offload;
+	u32 offload_caps;
 };
 
 static void spi_engine_program_add_cmd(struct spi_engine_program *p,
@@ -163,9 +203,9 @@  static void spi_engine_gen_xfer(struct spi_engine_program *p, bool dry,
 		unsigned int n = min(len, 256U);
 		unsigned int flags = 0;
 
-		if (xfer->tx_buf)
+		if (xfer->tx_buf || (xfer->offload_flags & SPI_OFFLOAD_XFER_TX_STREAM))
 			flags |= SPI_ENGINE_TRANSFER_WRITE;
-		if (xfer->rx_buf)
+		if (xfer->rx_buf || (xfer->offload_flags & SPI_OFFLOAD_XFER_RX_STREAM))
 			flags |= SPI_ENGINE_TRANSFER_READ;
 
 		spi_engine_program_add_cmd(p, dry,
@@ -217,16 +257,24 @@  static void spi_engine_gen_cs(struct spi_engine_program *p, bool dry,
  *
  * NB: This is separate from spi_engine_compile_message() because the latter
  * is called twice and would otherwise result in double-evaluation.
+ *
+ * Returns 0 on success, -EINVAL on failure.
  */
-static void spi_engine_precompile_message(struct spi_message *msg)
+static int spi_engine_precompile_message(struct spi_message *msg)
 {
 	unsigned int clk_div, max_hz = msg->spi->controller->max_speed_hz;
 	struct spi_transfer *xfer;
 
 	list_for_each_entry(xfer, &msg->transfers, transfer_list) {
+		/* If we have an offload transfer, we can't rx to buffer */
+		if (msg->offload && xfer->rx_buf)
+			return -EINVAL;
+
 		clk_div = DIV_ROUND_UP(max_hz, xfer->speed_hz);
 		xfer->effective_speed_hz = max_hz / min(clk_div, 256U);
 	}
+
+	return 0;
 }
 
 static void spi_engine_compile_message(struct spi_message *msg, bool dry,
@@ -521,11 +569,105 @@  static irqreturn_t spi_engine_irq(int irq, void *devid)
 	return IRQ_HANDLED;
 }
 
+static int spi_engine_offload_prepare(struct spi_message *msg)
+{
+	struct spi_controller *host = msg->spi->controller;
+	struct spi_engine *spi_engine = spi_controller_get_devdata(host);
+	struct spi_engine_program *p = msg->opt_state;
+	struct spi_engine_offload *priv = msg->offload->priv;
+	struct spi_transfer *xfer;
+	void __iomem *cmd_addr;
+	void __iomem *sdo_addr;
+	size_t tx_word_count = 0;
+	unsigned int i;
+
+	if (p->length > spi_engine->offload_ctrl_mem_size)
+		return -EINVAL;
+
+	/* count total number of tx words in message */
+	list_for_each_entry(xfer, &msg->transfers, transfer_list) {
+		/* no support for reading to rx_buf */
+		if (xfer->rx_buf)
+			return -EINVAL;
+
+		if (!xfer->tx_buf)
+			continue;
+
+		if (xfer->bits_per_word <= 8)
+			tx_word_count += xfer->len;
+		else if (xfer->bits_per_word <= 16)
+			tx_word_count += xfer->len / 2;
+		else
+			tx_word_count += xfer->len / 4;
+	}
+
+	if (tx_word_count && !(spi_engine->offload_caps & SPI_OFFLOAD_CAP_TX_STATIC_DATA))
+		return -EINVAL;
+
+	if (tx_word_count > spi_engine->offload_sdo_mem_size)
+		return -EINVAL;
+
+	/*
+	 * This protects against calling spi_optimize_message() with an offload
+	 * that has already been prepared with a different message.
+	 */
+	if (test_and_set_bit_lock(SPI_ENGINE_OFFLOAD_FLAG_PREPARED, &priv->flags))
+		return -EBUSY;
+
+	cmd_addr = spi_engine->base +
+		   SPI_ENGINE_REG_OFFLOAD_CMD_FIFO(priv->offload_num);
+	sdo_addr = spi_engine->base +
+		   SPI_ENGINE_REG_OFFLOAD_SDO_FIFO(priv->offload_num);
+
+	list_for_each_entry(xfer, &msg->transfers, transfer_list) {
+		if (!xfer->tx_buf)
+			continue;
+
+		if (xfer->bits_per_word <= 8) {
+			const u8 *buf = xfer->tx_buf;
+
+			for (i = 0; i < xfer->len; i++)
+				writel_relaxed(buf[i], sdo_addr);
+		} else if (xfer->bits_per_word <= 16) {
+			const u16 *buf = xfer->tx_buf;
+
+			for (i = 0; i < xfer->len / 2; i++)
+				writel_relaxed(buf[i], sdo_addr);
+		} else {
+			const u32 *buf = xfer->tx_buf;
+
+			for (i = 0; i < xfer->len / 4; i++)
+				writel_relaxed(buf[i], sdo_addr);
+		}
+	}
+
+	for (i = 0; i < p->length; i++)
+		writel_relaxed(p->instructions[i], cmd_addr);
+
+	return 0;
+}
+
+static void spi_engine_offload_unprepare(struct spi_offload *offload)
+{
+	struct spi_engine_offload *priv = offload->priv;
+	struct spi_engine *spi_engine = priv->spi_engine;
+
+	writel_relaxed(1, spi_engine->base +
+			  SPI_ENGINE_REG_OFFLOAD_RESET(priv->offload_num));
+	writel_relaxed(0, spi_engine->base +
+			  SPI_ENGINE_REG_OFFLOAD_RESET(priv->offload_num));
+
+	clear_bit_unlock(SPI_ENGINE_OFFLOAD_FLAG_PREPARED, &priv->flags);
+}
+
 static int spi_engine_optimize_message(struct spi_message *msg)
 {
 	struct spi_engine_program p_dry, *p;
+	int ret;
 
-	spi_engine_precompile_message(msg);
+	ret = spi_engine_precompile_message(msg);
+	if (ret)
+		return ret;
 
 	p_dry.length = 0;
 	spi_engine_compile_message(msg, true, &p_dry);
@@ -537,20 +679,61 @@  static int spi_engine_optimize_message(struct spi_message *msg)
 	spi_engine_compile_message(msg, false, p);
 
 	spi_engine_program_add_cmd(p, false, SPI_ENGINE_CMD_SYNC(
-						AXI_SPI_ENGINE_CUR_MSG_SYNC_ID));
+		msg->offload ? 0 : AXI_SPI_ENGINE_CUR_MSG_SYNC_ID));
 
 	msg->opt_state = p;
 
+	if (msg->offload) {
+		ret = spi_engine_offload_prepare(msg);
+		if (ret) {
+			msg->opt_state = NULL;
+			kfree(p);
+			return ret;
+		}
+	}
+
 	return 0;
 }
 
 static int spi_engine_unoptimize_message(struct spi_message *msg)
 {
+	if (msg->offload)
+		spi_engine_offload_unprepare(msg->offload);
+
 	kfree(msg->opt_state);
 
 	return 0;
 }
 
+static struct spi_offload
+*spi_engine_get_offload(struct spi_device *spi,
+			const struct spi_offload_config *config)
+{
+	struct spi_controller *host = spi->controller;
+	struct spi_engine *spi_engine = spi_controller_get_devdata(host);
+	struct spi_engine_offload *priv;
+
+	if (!spi_engine->offload)
+		return ERR_PTR(-ENODEV);
+
+	if (config->capability_flags & ~spi_engine->offload_caps)
+		return ERR_PTR(-EINVAL);
+
+	priv = spi_engine->offload->priv;
+
+	if (test_and_set_bit_lock(SPI_ENGINE_OFFLOAD_FLAG_ASSIGNED, &priv->flags))
+		return ERR_PTR(-EBUSY);
+
+	return spi_engine->offload;
+}
+
+static void spi_engine_put_offload(struct spi_offload *offload)
+{
+	struct spi_engine_offload *priv = offload->priv;
+
+	clear_bit_unlock(SPI_ENGINE_OFFLOAD_FLAG_ASSIGNED, &priv->flags);
+}
+
 static int spi_engine_setup(struct spi_device *device)
 {
 	struct spi_controller *host = device->controller;
@@ -583,6 +766,12 @@  static int spi_engine_transfer_one_message(struct spi_controller *host,
 	unsigned int int_enable = 0;
 	unsigned long flags;
 
+	if (msg->offload) {
+		dev_err(&host->dev, "Single transfer offload not supported\n");
+		msg->status = -EOPNOTSUPP;
+		goto out;
+	}
+
 	/* reinitialize message state for this transfer */
 	memset(st, 0, sizeof(*st));
 	st->cmd_buf = p->instructions;
@@ -632,11 +821,68 @@  static int spi_engine_transfer_one_message(struct spi_controller *host,
 			trace_spi_transfer_stop(msg, xfer);
 	}
 
+out:
 	spi_finalize_current_message(host);
 
 	return msg->status;
 }
 
+static int spi_engine_trigger_enable(struct spi_offload *offload)
+{
+	struct spi_engine_offload *priv = offload->priv;
+	struct spi_engine *spi_engine = priv->spi_engine;
+	unsigned int reg;
+
+	reg = readl_relaxed(spi_engine->base +
+			    SPI_ENGINE_REG_OFFLOAD_CTRL(priv->offload_num));
+	reg |= SPI_ENGINE_OFFLOAD_CTRL_ENABLE;
+	writel_relaxed(reg, spi_engine->base +
+			    SPI_ENGINE_REG_OFFLOAD_CTRL(priv->offload_num));
+	return 0;
+}
+
+static void spi_engine_trigger_disable(struct spi_offload *offload)
+{
+	struct spi_engine_offload *priv = offload->priv;
+	struct spi_engine *spi_engine = priv->spi_engine;
+	unsigned int reg;
+
+	reg = readl_relaxed(spi_engine->base +
+			    SPI_ENGINE_REG_OFFLOAD_CTRL(priv->offload_num));
+	reg &= ~SPI_ENGINE_OFFLOAD_CTRL_ENABLE;
+	writel_relaxed(reg, spi_engine->base +
+			    SPI_ENGINE_REG_OFFLOAD_CTRL(priv->offload_num));
+}
+
+static struct dma_chan
+*spi_engine_tx_stream_request_dma_chan(struct spi_offload *offload)
+{
+	struct spi_engine_offload *priv = offload->priv;
+	char name[16];
+
+	snprintf(name, sizeof(name), "offload%u-tx", priv->offload_num);
+
+	return dma_request_chan(offload->provider_dev, name);
+}
+
+static struct dma_chan
+*spi_engine_rx_stream_request_dma_chan(struct spi_offload *offload)
+{
+	struct spi_engine_offload *priv = offload->priv;
+	char name[16];
+
+	snprintf(name, sizeof(name), "offload%u-rx", priv->offload_num);
+
+	return dma_request_chan(offload->provider_dev, name);
+}
+
+static const struct spi_offload_ops spi_engine_offload_ops = {
+	.trigger_enable = spi_engine_trigger_enable,
+	.trigger_disable = spi_engine_trigger_disable,
+	.tx_stream_request_dma_chan = spi_engine_tx_stream_request_dma_chan,
+	.rx_stream_request_dma_chan = spi_engine_rx_stream_request_dma_chan,
+};
+
 static void spi_engine_release_hw(void *p)
 {
 	struct spi_engine *spi_engine = p;
@@ -651,8 +897,7 @@  static int spi_engine_probe(struct platform_device *pdev)
 	struct spi_engine *spi_engine;
 	struct spi_controller *host;
 	unsigned int version;
-	int irq;
-	int ret;
+	int irq, ret;
 
 	irq = platform_get_irq(pdev, 0);
 	if (irq < 0)
@@ -667,6 +912,46 @@  static int spi_engine_probe(struct platform_device *pdev)
 	spin_lock_init(&spi_engine->lock);
 	init_completion(&spi_engine->msg_complete);
 
+	/*
+	 * REVISIT: for now, all SPI Engines only have one offload. In the
+	 * future, this should be read from a memory mapped register to
+	 * determine the number of offloads enabled at HDL compile time. For
+	 * now, we can tell if an offload is present if there is a trigger
+	 * source wired up to it.
+	 */
+	if (device_property_present(&pdev->dev, "trigger-sources")) {
+		struct spi_engine_offload *priv;
+
+		spi_engine->offload =
+			devm_spi_offload_alloc(&pdev->dev,
+					       sizeof(struct spi_engine_offload));
+		if (IS_ERR(spi_engine->offload))
+			return PTR_ERR(spi_engine->offload);
+
+		priv = spi_engine->offload->priv;
+		priv->spi_engine = spi_engine;
+		priv->offload_num = 0;
+
+		spi_engine->offload->ops = &spi_engine_offload_ops;
+		spi_engine->offload_caps = SPI_OFFLOAD_CAP_TRIGGER;
+
+		if (device_property_match_string(&pdev->dev, "dma-names", "offload0-rx") >= 0) {
+			spi_engine->offload_caps |= SPI_OFFLOAD_CAP_RX_STREAM_DMA;
+			spi_engine->offload->xfer_flags |= SPI_OFFLOAD_XFER_RX_STREAM;
+		}
+
+		if (device_property_match_string(&pdev->dev, "dma-names", "offload0-tx") >= 0) {
+			spi_engine->offload_caps |= SPI_OFFLOAD_CAP_TX_STREAM_DMA;
+			spi_engine->offload->xfer_flags |= SPI_OFFLOAD_XFER_TX_STREAM;
+		} else {
+			/*
+			 * HDL compile option to enable TX DMA stream also disables
+			 * the SDO memory, so can't do both at the same time.
+			 */
+			spi_engine->offload_caps |= SPI_OFFLOAD_CAP_TX_STATIC_DATA;
+		}
+	}
+
 	spi_engine->clk = devm_clk_get_enabled(&pdev->dev, "s_axi_aclk");
 	if (IS_ERR(spi_engine->clk))
 		return PTR_ERR(spi_engine->clk);
@@ -688,6 +973,19 @@  static int spi_engine_probe(struct platform_device *pdev)
 		return -ENODEV;
 	}
 
+	if (ADI_AXI_PCORE_VER_MINOR(version) >= 1) {
+		unsigned int sizes = readl(spi_engine->base +
+				SPI_ENGINE_REG_OFFLOAD_MEM_ADDR_WIDTH);
+
+		spi_engine->offload_ctrl_mem_size = 1 <<
+			FIELD_GET(SPI_ENGINE_SPI_OFFLOAD_MEM_WIDTH_CMD, sizes);
+		spi_engine->offload_sdo_mem_size = 1 <<
+			FIELD_GET(SPI_ENGINE_SPI_OFFLOAD_MEM_WIDTH_SDO, sizes);
+	} else {
+		spi_engine->offload_ctrl_mem_size = SPI_ENGINE_OFFLOAD_CMD_FIFO_SIZE;
+		spi_engine->offload_sdo_mem_size = SPI_ENGINE_OFFLOAD_SDO_FIFO_SIZE;
+	}
+
 	writel_relaxed(0x00, spi_engine->base + SPI_ENGINE_REG_RESET);
 	writel_relaxed(0xff, spi_engine->base + SPI_ENGINE_REG_INT_PENDING);
 	writel_relaxed(0x00, spi_engine->base + SPI_ENGINE_REG_INT_ENABLE);
@@ -709,6 +1007,8 @@  static int spi_engine_probe(struct platform_device *pdev)
 	host->transfer_one_message = spi_engine_transfer_one_message;
 	host->optimize_message = spi_engine_optimize_message;
 	host->unoptimize_message = spi_engine_unoptimize_message;
+	host->get_offload = spi_engine_get_offload;
+	host->put_offload = spi_engine_put_offload;
 	host->num_chipselect = 8;
 
 	/* Some features depend of the IP core version. */