Re: [PATCH 2/8] media: imx-pxp: detect PXP version

[Michael Tretter <m.tretter@pengutronix.de>]

On Fri, 06 Jan 2023 14:28:47 +0200, Laurent Pinchart wrote:
> On Fri, Jan 06, 2023 at 01:47:32PM +0200, Laurent Pinchart wrote:
> > Thank you for the patch.
> > 
> > On Thu, Jan 05, 2023 at 02:47:23PM +0100, Michael Tretter wrote:
> > > Different versions of the Pixel Pipeline have different blocks and their
> > > routing may be different. Read the PXP_HW_VERSION register to determine
> > > the version of the PXP and print it to the log for debugging purposes.
> > 
> > Is there a specific reason you chose to read the version register
> > instead of basing the decision on the compatible string ?
> 
> Reading the rest of the series, you use the compatible strings later,
> and never use the hw_version field. I'm tempted to propose dropping this
> patch.

My first try was to use the version register, but it turned out that the
version is the same at least on i.MX6ULL and i.MX7D. I kept it to avoid that
others fall into the same trap.

> 
> > > Signed-off-by: Michael Tretter <m.tretter@pengutronix.de>
> > > ---
> > >  drivers/media/platform/nxp/imx-pxp.c | 18 ++++++++++++++++++
> > >  1 file changed, 18 insertions(+)
> > > 
> > > diff --git a/drivers/media/platform/nxp/imx-pxp.c b/drivers/media/platform/nxp/imx-pxp.c
> > > index 689ae5e6ac62..05abe40558b0 100644
> > > --- a/drivers/media/platform/nxp/imx-pxp.c
> > > +++ b/drivers/media/platform/nxp/imx-pxp.c
> > > @@ -10,6 +10,7 @@
> > >   * Pawel Osciak, <pawel@osciak.com>
> > >   * Marek Szyprowski, <m.szyprowski@samsung.com>
> > >   */
> > > +#include <linux/bitfield.h>
> > >  #include <linux/clk.h>
> > >  #include <linux/delay.h>
> > >  #include <linux/dma-mapping.h>
> > > @@ -52,6 +53,11 @@ MODULE_PARM_DESC(debug, "activates debug info");
> > >  #define MEM2MEM_HFLIP	(1 << 0)
> > >  #define MEM2MEM_VFLIP	(1 << 1)
> > >  
> > > +#define PXP_VERSION_MAJOR(version) \
> > > +	FIELD_GET(BM_PXP_VERSION_MAJOR, version)
> > > +#define PXP_VERSION_MINOR(version) \
> > > +	FIELD_GET(BM_PXP_VERSION_MINOR, version)
> > > +
> > >  #define dprintk(dev, fmt, arg...) \
> > >  	v4l2_dbg(1, debug, &dev->v4l2_dev, "%s: " fmt, __func__, ## arg)
> > >  
> > > @@ -192,6 +198,8 @@ struct pxp_dev {
> > >  	struct clk		*clk;
> > >  	void __iomem		*mmio;
> > >  
> > > +	u32			hw_version;
> > > +
> > >  	atomic_t		num_inst;
> > >  	struct mutex		dev_mutex;
> > >  	spinlock_t		irqlock;
> > > @@ -1660,6 +1668,11 @@ static int pxp_soft_reset(struct pxp_dev *dev)
> > >  	return 0;
> > >  }
> > >  
> > > +static u32 pxp_read_version(struct pxp_dev *dev)
> > > +{
> > > +	return readl(dev->mmio + HW_PXP_VERSION);
> > > +}
> > > +
> > >  static int pxp_probe(struct platform_device *pdev)
> > >  {
> > >  	struct pxp_dev *dev;
> > > @@ -1705,6 +1718,11 @@ static int pxp_probe(struct platform_device *pdev)
> > >  		goto err_clk;
> > >  	}
> > >  
> > > +	dev->hw_version = pxp_read_version(dev);
> > > +	dev_info(&pdev->dev, "PXP Version %d.%d\n",
> > 
> > As the version can't be negative, I'd use %u.%u.
> > 
> > > +		 PXP_VERSION_MAJOR(dev->hw_version),
> > > +		 PXP_VERSION_MINOR(dev->hw_version));
> > > +
> > 
> > The driver now prints two messages at probe time, it would be nice to
> > combine them, or remove the other one. That's a candidate for a future
> > patch though.

I would reduce the level to dev_debug. Then the version is not always printed,
but it can be easily enabled if necessary for the bringup on another platform.

Michael

> > 
> > >  	ret = v4l2_device_register(&pdev->dev, &dev->v4l2_dev);
> > >  	if (ret)
> > >  		goto err_clk;