Re: [PATCHv3] spi: dw-spi: Convert 16bit accesses to 32bit accesses

[Andy Shevchenko <andriy.shevchenko-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>]

On Wed, 2015-03-11 at 14:20 -0500, tthayer-yzvPICuk2ABMcg4IHK0kFoH6Mc4MB0Vx@public.gmane.org wrote:
> From: Thor Thayer <tthayer-yzvPICuk2ABMcg4IHK0kFoH6Mc4MB0Vx@public.gmane.org>

Thanks my couple of comments below. And I will test your next version
tomorrow if everything okay on your side (regarding to my comments)

> Altera's Arria10 SoC interconnect requires a 32 bit write for APB
> peripherals. The current spi-dw driver uses 16bit accesses in
> some locations. A review of the Designware SPI IP databook
> indicates the registers will support 32b read and writes to
> remain consistent with the AHB bus.

Better to exchange this with what you put on cover letter, i.e. cite
documentation here, but describe in common there.

> 
> Request for test with existing platforms. Currently tested
> on Altera CycloneV and Arria10.

> 
> Signed-off-by: Thor Thayer <tthayer-yzvPICuk2ABMcg4IHK0kFoH6Mc4MB0Vx@public.gmane.org>
> ---
> r1: Use function pointers to select 16b or 32b accesses.
> 
> r2: Use 32b version of function pointers for 16b reads and
> writes.
> 
> r3: Convert 16b reads and writes to 32b reads and writes. The
> DW_apb_ssi databook (section 6-1) states:
> All registers in the DW_apb_ssi are addressed at 32-bit boundaries
> to remain consistent with the AHB bus. Where the physical size of
> any register is less than 32-bits wide, the upper unused bits of
> the 32-bit boundary are reserved. Writing to these bits has no
> effect; reading from these bits returns 0.
> ---
>  drivers/spi/spi-dw.c |   34 +++++++++++++++++-----------------
>  drivers/spi/spi-dw.h |   10 ----------
>  2 files changed, 17 insertions(+), 27 deletions(-)
> 
> diff --git a/drivers/spi/spi-dw.c b/drivers/spi/spi-dw.c
> index 0f01069..9451c8a 100644
> --- a/drivers/spi/spi-dw.c
> +++ b/drivers/spi/spi-dw.c
> @@ -157,7 +157,7 @@ static inline u32 tx_max(struct dw_spi *dws)
>  	u32 tx_left, tx_room, rxtx_gap;
>  
>  	tx_left = (dws->tx_end - dws->tx) / dws->n_bytes;
> -	tx_room = dws->fifo_len - dw_readw(dws, DW_SPI_TXFLR);
> +	tx_room = dws->fifo_len - dw_readl(dws, DW_SPI_TXFLR);
>  
>  	/*
>  	 * Another concern is about the tx/rx mismatch, we
> @@ -178,13 +178,13 @@ static inline u32 rx_max(struct dw_spi *dws)
>  {
>  	u32 rx_left = (dws->rx_end - dws->rx) / dws->n_bytes;
>  
> -	return min_t(u32, rx_left, dw_readw(dws, DW_SPI_RXFLR));
> +	return min_t(u32, rx_left, dw_readl(dws, DW_SPI_RXFLR));
>  }
>  
>  static void dw_writer(struct dw_spi *dws)
>  {
>  	u32 max = tx_max(dws);
> -	u16 txw = 0;
> +	u32 txw = 0;

Do we really need to touch types? Can you try without this kind of
changes?

>  
>  	while (max--) {
>  		/* Set the tx word if the transfer's original "tx" is not null */
> @@ -194,18 +194,17 @@ static void dw_writer(struct dw_spi *dws)
>  			else
>  				txw = *(u16 *)(dws->tx);
>  		}
> -		dw_writew(dws, DW_SPI_DR, txw);
> +		dw_writel(dws, DW_SPI_DR, txw);
>  		dws->tx += dws->n_bytes;
>  	}
>  }
>  
>  static void dw_reader(struct dw_spi *dws)
>  {
> -	u32 max = rx_max(dws);
> -	u16 rxw;
> +	u32 rxw, max = rx_max(dws);
>  
>  	while (max--) {
> -		rxw = dw_readw(dws, DW_SPI_DR);
> +		rxw = dw_readl(dws, DW_SPI_DR);
>  		/* Care rx only if the transfer's original "rx" is not null */
>  		if (dws->rx_end - dws->len) {
>  			if (dws->n_bytes == 1)
> @@ -228,11 +227,11 @@ static void int_error_stop(struct dw_spi *dws, const char *msg)
>  
>  static irqreturn_t interrupt_transfer(struct dw_spi *dws)
>  {
> -	u16 irq_status = dw_readw(dws, DW_SPI_ISR);
> +	u32 irq_status = dw_readl(dws, DW_SPI_ISR);
>  
>  	/* Error handling */
>  	if (irq_status & (SPI_INT_TXOI | SPI_INT_RXOI | SPI_INT_RXUI)) {
> -		dw_readw(dws, DW_SPI_ICR);
> +		dw_readl(dws, DW_SPI_ICR);
>  		int_error_stop(dws, "interrupt_transfer: fifo overrun/underrun");
>  		return IRQ_HANDLED;
>  	}
> @@ -257,7 +256,7 @@ static irqreturn_t dw_spi_irq(int irq, void *dev_id)
>  {
>  	struct spi_master *master = dev_id;
>  	struct dw_spi *dws = spi_master_get_devdata(master);
> -	u16 irq_status = dw_readw(dws, DW_SPI_ISR) & 0x3f;
> +	u32 irq_status = dw_readl(dws, DW_SPI_ISR) & 0x3f;
>  
>  	if (!irq_status)
>  		return IRQ_NONE;
> @@ -288,7 +287,7 @@ static int dw_spi_transfer_one(struct spi_master *master,
>  	struct dw_spi *dws = spi_master_get_devdata(master);
>  	struct chip_data *chip = spi_get_ctldata(spi);
>  	u8 imask = 0;
> -	u16 txlevel = 0;
> +	u32 txlevel = 0;
>  	u16 clk_div = 0;
>  	u32 speed = 0;
>  	u32 cr0 = 0;
> @@ -354,7 +353,7 @@ static int dw_spi_transfer_one(struct spi_master *master,
>  		cr0 |= (chip->tmode << SPI_TMOD_OFFSET);
>  	}
>  
> -	dw_writew(dws, DW_SPI_CTRL0, cr0);
> +	dw_writel(dws, DW_SPI_CTRL0, cr0);
>  
>  	/* Check if current transfer is a DMA transaction */
>  	if (master->can_dma && master->can_dma(master, spi, transfer))
> @@ -374,8 +373,9 @@ static int dw_spi_transfer_one(struct spi_master *master,
>  			return ret;
>  		}
>  	} else if (!chip->poll_mode) {
> -		txlevel = min_t(u16, dws->fifo_len / 2, dws->len / dws->n_bytes);
> -		dw_writew(dws, DW_SPI_TXFLTR, txlevel);
> +		txlevel = min_t(u32, dws->fifo_len / 2,
> +				dws->len / dws->n_bytes);
> +		dw_writel(dws, DW_SPI_TXFLTR, txlevel);
>  
>  		/* Set the interrupt mask */
>  		imask |= SPI_INT_TXEI | SPI_INT_TXOI |
> @@ -499,11 +499,11 @@ static void spi_hw_init(struct device *dev, struct dw_spi *dws)
>  		u32 fifo;
>  
>  		for (fifo = 1; fifo < 256; fifo++) {
> -			dw_writew(dws, DW_SPI_TXFLTR, fifo);
> -			if (fifo != dw_readw(dws, DW_SPI_TXFLTR))
> +			dw_writel(dws, DW_SPI_TXFLTR, fifo);
> +			if (fifo != dw_readl(dws, DW_SPI_TXFLTR))
>  				break;
>  		}
> -		dw_writew(dws, DW_SPI_TXFLTR, 0);
> +		dw_writel(dws, DW_SPI_TXFLTR, 0);
>  
>  		dws->fifo_len = (fifo == 1) ? 0 : fifo;
>  		dev_dbg(dev, "Detected FIFO size: %u bytes\n", dws->fifo_len);
> diff --git a/drivers/spi/spi-dw.h b/drivers/spi/spi-dw.h
> index 41f77e2..6c91391 100644
> --- a/drivers/spi/spi-dw.h
> +++ b/drivers/spi/spi-dw.h
> @@ -150,16 +150,6 @@ static inline void dw_writel(struct dw_spi *dws, u32 offset, u32 val)
>  	__raw_writel(val, dws->regs + offset);
>  }
>  
> -static inline u16 dw_readw(struct dw_spi *dws, u32 offset)
> -{
> -	return __raw_readw(dws->regs + offset);
> -}
> -
> -static inline void dw_writew(struct dw_spi *dws, u32 offset, u16 val)
> -{
> -	__raw_writew(val, dws->regs + offset);
> -}
> -
>  static inline void spi_enable_chip(struct dw_spi *dws, int enable)
>  {
>  	dw_writel(dws, DW_SPI_SSIENR, (enable ? 1 : 0));


-- 
Andy Shevchenko <andriy.shevchenko-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
Intel Finland Oy

--
To unsubscribe from this list: send the line "unsubscribe linux-spi" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html