Re: [PATCHv12 3/3] ARM:drm ivip Intel FPGA Video and Image Processing Suite

From: Sam Ravnborg
Date: Mon Feb 11 2019 - 01:56:07 EST


Hi Hean-Loong, Ong

Patch looks good to me, but there is a few trivial
things I spotted while browsing the code.
See below.

Sam

> +++ b/drivers/gpu/drm/ivip/Makefile
> @@ -0,0 +1,7 @@
> +#
> +# Makefile for the drm device driver. This driver provides support for the
> +# Direct Rendering Infrastructure (DRI) in XFree86 4.1.0 and higher.
> +
> +obj-$(CONFIG_DRM_IVIP) += ivip.o
> +ivip-objs := intel_vip_of.o intel_vip_core.o \
> + intel_vip_conn.o

You could use:
ivip-y := intel_vip_of.o intel_vip_core.o
ivip-y += intel_vip_conn.o


Using "ivip-y" is the recommend syntax today.
And using "+=" you get rid of the ugly "\" to continue a line.
(Some people prefer "\" in makefiles, but there are not needed)

> +++ b/drivers/gpu/drm/ivip/intel_vip_core.c
> @@ -0,0 +1,189 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * Copyright (C) 2019 Intel Corporation.
> + *
> + * intel_vip_core.c -- Intel Video and Image Processing(VIP)
> + * Frame Buffer II driver
> + *
> + * This driver supports the Intel VIP Frame Reader component.
> + * More info on the hardware can be found in the Intel Video
> + * and Image Processing Suite User Guide at this address
> + * http://www.altera.com/literature/ug/ug_vip.pdf.
> + *
> + * Authors:
> + * Walter Goossens <waltergoossens@xxxxxxx>
> + * Thomas Chou <thomas@xxxxxxxxxxxxx>
> + * Chris Rauer <crauer@xxxxxxxxxx>
> + * Ong, Hean-Loong <hean.loong.ong@xxxxxxxxx>
> + *
> + */
> +
> +#include <drm/drmP.h>
Please do not use drmP.h in new drivers, we try to get rid of this file.

> +#include <drm/drm_atomic.h>
> +#include <drm/drm_atomic_helper.h>
> +#include <drm/drm_crtc_helper.h>
> +#include <drm/drm_fb_helper.h>
> +#include <drm/drm_fb_cma_helper.h>
> +#include <drm/drm_gem_cma_helper.h>
> +#include <drm/drm_plane_helper.h>
> +#include <drm/drm_simple_kms_helper.h>
> +#include <drm/drm_gem_framebuffer_helper.h>
Sort the list of include files.
(Looks like this was properly done before, and only one file is out of palce)


> +static void intelvipfb_enable(struct drm_simple_display_pipe *pipe,
> + struct drm_crtc_state *crtc_state, struct drm_plane_state *
> + plane_state)
Fix indent. The parameters on second line and following should be aligned
below the opening paranthesis.
Use tab(s) + spaces to align properly.

> +void intelvipfb_display_pipe_update(struct drm_simple_display_pipe *pipe,
> + struct drm_plane_state *old_state)
Align parameter

> +}
> +EXPORT_SYMBOL(intelvipfb_display_pipe_update);
> +
> +static struct drm_simple_display_pipe_funcs fbpriv_funcs = {
> + .prepare_fb = drm_gem_fb_simple_display_pipe_prepare_fb,
> + .update = intelvipfb_display_pipe_update,
> + .enable = intelvipfb_enable,
> + .disable = intelvipfb_disable,
> +};
> +
> +
> +int intelvipfb_probe(struct device *dev)
> +{
> + int retval;
> + struct drm_device *drm;
> + struct intelvipfb_priv *fbpriv = dev_get_drvdata(dev);
> +
> + struct drm_connector *connector;
> + u32 formats[] = {DRM_FORMAT_XRGB8888};
> +
> + drm = fbpriv->drm;
> +
> + drm->dev_private = fbpriv;

It would be simpler to just pass fbpriv as a parameter.
There is only one user of intelvipfb_probe() so no need
to avoid it.

Also it would be more logical to set drm = fbpriv->drm; where memory are allocated.

> +
> + intelvipfb_setup_mode_config(drm);
> +
> + connector = intelvipfb_conn_setup(drm);
> + if (!connector) {
> + dev_err(drm->dev, "Connector setup failed\n");
> + goto err_mode_config;
> + }
> +
> + retval = drm_simple_display_pipe_init(drm,
> + &fbpriv->pipe,
> + &fbpriv_funcs,
> + formats,
> + ARRAY_SIZE(formats),
> + NULL, connector);
Consider indent, where subsequent parameters are aligned right after the
opening '('.

> +
> + if (retval < 0) {
> + dev_err(drm->dev, "Cannot setup simple display pipe\n");
> + goto err_mode_config;
> + }
> +
> + drm_mode_config_reset(drm);
> +
> + drm_dev_register(drm, 0);
> +
> + drm_fbdev_generic_setup(drm, 32);
> +
> + dev_info(drm->dev, "ivip: Successfully created fb\n");
> +
> + return retval;
> +
> +err_mode_config:
> +
> + drm_mode_config_cleanup(drm);
> + return -ENODEV;
> +}
> +

> diff --git a/drivers/gpu/drm/ivip/intel_vip_of.c b/drivers/gpu/drm/ivip/intel_vip_of.c
> new file mode 100644
> index 0000000..c899e30
> --- /dev/null
> +++ b/drivers/gpu/drm/ivip/intel_vip_of.c
> @@ -0,0 +1,181 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * Copyright (C) 2019 Intel Corporation.
> + *
> + * intel_vip_of.c -- Intel Video and Image Processing(VIP)
> + * Frame Buffer II driver

The need for this file confuses me.
It does not include only "of" related stuff, but also generic device driver stuff.

Maybe merge with intel_vip_core.c?
And rename that file to intel_vip_drv.c?

> + *
> + * This driver supports the Intel VIP Frame Reader component.
> + * More info on the hardware can be found in the Intel Video
> + * and Image Processing Suite User Guide at this address
> + * http://www.altera.com/literature/ug/ug_vip.pdf.
> + *
> + * Authors:
> + * Ong, Hean-Loong <hean.loong.ong@xxxxxxxxx>
> + *
> + */
> +#include <drm/drmP.h>
Drop use of drmP.h

> +static int intelvipfb_of_probe(struct platform_device *pdev)
> +{
> + int retval;
> + struct resource *reg_res;
> + struct intelvipfb_priv *fbpriv;
> + struct device *dev = &pdev->dev;
> + struct drm_device *drm;
> +
> + fbpriv = devm_kzalloc(dev, sizeof(*fbpriv), GFP_KERNEL);
> + if (!fbpriv)
> + return -ENOMEM;
> +
> + /*setup DRM */
Space before "setup"

> +
> +static const struct of_device_id intelvipfb_of_match[] = {
> + { .compatible = "altr,vip-frame-buffer-2.0" },
> + {},
Maybe add "/* sentinel */" comment?