lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <6c2188ce-a27a-bb76-4b38-5fda5f282f78@ti.com>
Date:   Thu, 13 Feb 2020 11:19:09 +0200
From:   Tomi Valkeinen <tomi.valkeinen@...com>
To:     Yuti Amonkar <yamonkar@...ence.com>,
        <linux-kernel@...r.kernel.org>, <dri-devel@...ts.freedesktop.org>,
        <devicetree@...r.kernel.org>, <robh+dt@...nel.org>,
        <maxime@...no.tech>, <airlied@...ux.ie>, <daniel@...ll.ch>,
        <mark.rutland@....com>, <a.hajda@...sung.com>,
        <narmstrong@...libre.com>, <Laurent.pinchart@...asonboard.com>,
        <jonas@...boo.se>, <jernej.skrabec@...l.net>
CC:     <praneeth@...com>, <jsarha@...com>, <mparab@...ence.com>,
        <sjakhade@...ence.com>
Subject: Re: [PATCH v5 2/3] drm: bridge: Add support for Cadence MHDP DPI/DP
 bridge

On 12/02/2020 06:26, Yuti Amonkar wrote:
> This patch adds new DRM driver for Cadence MHDP DPTX IP used on J721e SoC.
> MHDP DPTX IP is the component that complies with VESA DisplayPort (DP) and
> embedded Display Port (eDP) standards. It integrates uCPU running the
> embedded Firmware(FW) interfaced over APB interface.
> Basically, it takes a DPI stream as input and output it encoded in DP
> format. Currently, it supports only SST mode.
> 
> Signed-off-by: Yuti Amonkar <yamonkar@...ence.com>
> Signed-off-by: Jyri Sarha <jsarha@...com>
> ---
>   drivers/gpu/drm/bridge/Kconfig          |   11 +
>   drivers/gpu/drm/bridge/Makefile         |    3 +
>   drivers/gpu/drm/bridge/cdns-mhdp-core.c | 2206 +++++++++++++++++++++++
>   drivers/gpu/drm/bridge/cdns-mhdp-core.h |  380 ++++
>   4 files changed, 2600 insertions(+)
>   create mode 100644 drivers/gpu/drm/bridge/cdns-mhdp-core.c
>   create mode 100644 drivers/gpu/drm/bridge/cdns-mhdp-core.h
> 
> diff --git a/drivers/gpu/drm/bridge/Kconfig b/drivers/gpu/drm/bridge/Kconfig
> index 8397bf72d2f3..c66f2ef04f71 100644
> --- a/drivers/gpu/drm/bridge/Kconfig
> +++ b/drivers/gpu/drm/bridge/Kconfig
> @@ -27,6 +27,17 @@ config DRM_CDNS_DSI
>   	  Support Cadence DPI to DSI bridge. This is an internal
>   	  bridge and is meant to be directly embedded in a SoC.
>   
> +config DRM_CDNS_MHDP
> +	tristate "Cadence DPI/DP bridge"
> +	select DRM_KMS_HELPER
> +	select DRM_PANEL_BRIDGE
> +	depends on OF
> +	help
> +	  Support Cadence DPI to DP bridge. This is an internal
> +	  bridge and is meant to be directly embedded in a SoC.
> +	  It takes a DPI stream as input and output it encoded
> +	  in DP format.
> +
>   config DRM_DUMB_VGA_DAC
>   	tristate "Dumb VGA DAC Bridge support"
>   	depends on OF
> diff --git a/drivers/gpu/drm/bridge/Makefile b/drivers/gpu/drm/bridge/Makefile
> index 1eb5376c5d68..71019088d257 100644
> --- a/drivers/gpu/drm/bridge/Makefile
> +++ b/drivers/gpu/drm/bridge/Makefile
> @@ -15,6 +15,9 @@ obj-$(CONFIG_DRM_TOSHIBA_TC358767) += tc358767.o
>   obj-$(CONFIG_DRM_I2C_ADV7511) += adv7511/
>   obj-$(CONFIG_DRM_TI_SN65DSI86) += ti-sn65dsi86.o
>   obj-$(CONFIG_DRM_TI_TFP410) += ti-tfp410.o
> +obj-$(CONFIG_DRM_CDNS_MHDP) += cdns-mhdp.o
>   
>   obj-y += analogix/
>   obj-y += synopsys/
> +
> +cdns-mhdp-objs := cdns-mhdp-core.o

I think it's better to keep the two lines added in this change together, instead of having those 
other obj-y lines in between.

> diff --git a/drivers/gpu/drm/bridge/cdns-mhdp-core.c b/drivers/gpu/drm/bridge/cdns-mhdp-core.c
> new file mode 100644
> index 000000000000..51ed9cdee161
> --- /dev/null
> +++ b/drivers/gpu/drm/bridge/cdns-mhdp-core.c
> @@ -0,0 +1,2206 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * Cadence MHDP DP bridge driver.
> + *
> + * Copyright: 2019 Cadence Design Systems, Inc.
> + *
> + * Author: Quentin Schulz <quentin.schulz@...e-electrons.com>

Author is Quentin, but there's no signed-off-by from him? There are also many names in 
MODULE_AUTHOR, without signed-off-bys.

I think the rule is that you should have signed-off-bys for people who have contributed. I'm not 
sure how strict that rule is, and what to do if the people cannot be reached anymore.

<snip>

> +static int mhdp_probe(struct platform_device *pdev)
> +{
> +	const struct of_device_id *match;
> +	struct resource *regs;
> +	struct cdns_mhdp_device *mhdp;
> +	struct clk *clk;
> +	int ret;
> +	unsigned long rate;
> +	int irq;
> +	u32 lanes_prop;
> +	unsigned int link_rate;
> +
> +	mhdp = devm_kzalloc(&pdev->dev, sizeof(struct cdns_mhdp_device),
> +			    GFP_KERNEL);
> +	if (!mhdp)
> +		return -ENOMEM;
> +
> +	clk = devm_clk_get(&pdev->dev, NULL);
> +	if (IS_ERR(clk)) {
> +		dev_err(&pdev->dev, "couldn't get clk: %ld\n", PTR_ERR(clk));
> +		return PTR_ERR(clk);
> +	}
> +
> +	mhdp->clk = clk;
> +	mhdp->dev = &pdev->dev;
> +	mhdp->conn_bus_flags_defaults = DRM_BUS_FLAG_DE_HIGH;
> +	mutex_init(&mhdp->mbox_mutex);
> +	spin_lock_init(&mhdp->start_lock);
> +	dev_set_drvdata(&pdev->dev, mhdp);
> +
> +	drm_dp_aux_init(&mhdp->aux);
> +	mhdp->aux.dev = &pdev->dev;
> +	mhdp->aux.transfer = mhdp_transfer;
> +
> +	regs = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> +	mhdp->regs = devm_ioremap_resource(&pdev->dev, regs);
> +	if (IS_ERR(mhdp->regs))
> +		return PTR_ERR(mhdp->regs);
> +
> +	mhdp->phy = devm_of_phy_get_by_index(&pdev->dev, pdev->dev.of_node, 0);
> +	if (IS_ERR(mhdp->phy)) {
> +		dev_err(&pdev->dev, "no PHY configured\n");
> +		return PTR_ERR(mhdp->phy);
> +	}
> +
> +	platform_set_drvdata(pdev, mhdp);
> +
> +	clk_prepare_enable(clk);
> +
> +	match = of_match_device(mhdp_ids, &pdev->dev);
> +	if (!match)
> +		return -ENODEV;
> +	mhdp->ops = (struct mhdp_platform_ops *)match->data;
> +
> +	pm_runtime_enable(&pdev->dev);
> +	ret = pm_runtime_get_sync(&pdev->dev);
> +	if (ret < 0) {
> +		dev_err(&pdev->dev, "pm_runtime_get_sync failed\n");
> +		pm_runtime_disable(&pdev->dev);
> +		goto clk_disable;
> +	}
> +
> +	if (mhdp->ops && mhdp->ops->init) {
> +		ret = mhdp->ops->init(mhdp);
> +		if (ret != 0) {
> +			dev_err(&pdev->dev, "MHDP platform initialization failed: %d\n",
> +				ret);
> +			goto runtime_put;
> +		}
> +	}
> +
> +	rate = clk_get_rate(clk);
> +	writel(rate % 1000000, mhdp->regs + CDNS_SW_CLK_L);
> +	writel(rate / 1000000, mhdp->regs + CDNS_SW_CLK_H);
> +
> +	dev_dbg(&pdev->dev, "func clk rate %lu Hz\n", rate);
> +
> +	writel(~0, mhdp->regs + CDNS_MB_INT_MASK);
> +	writel(~0, mhdp->regs + CDNS_APB_INT_MASK);
> +
> +	irq = platform_get_irq(pdev, 0);
> +	ret = devm_request_threaded_irq(mhdp->dev, irq, NULL, mhdp_irq_handler,
> +					IRQF_ONESHOT, "mhdp8546", mhdp);
> +	if (ret) {
> +		dev_err(&pdev->dev, "cannot install IRQ %d\n", irq);
> +		ret = -EIO;
> +		goto plat_fini;
> +	}
> +
> +	/* Read source capabilities, based on PHY's device tree properties. */
> +	ret = device_property_read_u32(&mhdp->phy->dev, "cdns,num-lanes",
> +				       &(lanes_prop));
> +	if (ret)
> +		mhdp->host.lanes_cnt = CDNS_LANE_4;
> +	else
> +		mhdp->host.lanes_cnt = lanes_prop;
> +
> +	ret = device_property_read_u32(&mhdp->phy->dev, "cdns,max-bit-rate",
> +				       &(link_rate));
> +	if (ret)
> +		link_rate = drm_dp_bw_code_to_link_rate(DP_LINK_BW_8_1);
> +	else
> +		/* PHY uses Mb/s, DRM uses tens of kb/s. */
> +		link_rate *= 100;
> +
> +	mhdp->host.link_rate = link_rate;
> +	mhdp->host.volt_swing = CDNS_VOLT_SWING(3);
> +	mhdp->host.pre_emphasis = CDNS_PRE_EMPHASIS(3);
> +	mhdp->host.pattern_supp = CDNS_SUPPORT_TPS(1) |
> +				  CDNS_SUPPORT_TPS(2) | CDNS_SUPPORT_TPS(3) |
> +				  CDNS_SUPPORT_TPS(4);
> +	mhdp->host.lane_mapping = CDNS_LANE_MAPPING_NORMAL;
> +	mhdp->host.fast_link = false;
> +	mhdp->host.enhanced = true;
> +	mhdp->host.scrambler = true;
> +	mhdp->host.ssc = false;
> +
> +	/* The only currently supported format */
> +	mhdp->display_fmt.y_only = false;
> +	mhdp->display_fmt.color_format = DRM_COLOR_FORMAT_RGB444;
> +	mhdp->display_fmt.bpc = 8;
> +
> +	mhdp->bridge.of_node = pdev->dev.of_node;
> +	mhdp->bridge.funcs = &cdns_mhdp_bridge_funcs;
> +
> +	ret = phy_init(mhdp->phy);
> +	if (ret) {
> +		dev_err(mhdp->dev, "Failed to initialize PHY: %d\n", ret);
> +		goto runtime_put;
> +	}
> +
> +	ret = load_firmware(mhdp);
> +	if (ret)
> +		goto phy_exit;
> +
> +	drm_bridge_add(&mhdp->bridge);
> +
> +	return 0;
> +
> +phy_exit:
> +	phy_exit(mhdp->phy);
> +plat_fini:
> +	if (mhdp->ops && mhdp->ops->exit)
> +		mhdp->ops->exit(mhdp);
> +runtime_put:
> +	pm_runtime_put_sync(&pdev->dev);
> +	pm_runtime_disable(&pdev->dev);
> +clk_disable:
> +	clk_disable_unprepare(mhdp->clk);
> +
> +	return ret;
> +}
> +
> +MODULE_FIRMWARE(FW_NAME);
> +
> +static int mhdp_remove(struct platform_device *pdev)
> +{
> +	struct cdns_mhdp_device *mhdp = dev_get_drvdata(&pdev->dev);
> +	unsigned int timeout = 10;
> +	bool stop_fw = false;
> +	int ret = 0;
> +
> +	if (mhdp->ops && mhdp->ops->exit)
> +		mhdp->ops->exit(mhdp);
> +
> +	drm_bridge_remove(&mhdp->bridge);
> +
> +wait_loading:
> +	spin_lock(&mhdp->start_lock);
> +	if (mhdp->hw_state == MHDP_HW_LOADING && timeout-- > 0) {
> +		spin_unlock(&mhdp->start_lock);
> +		msleep(100);
> +		goto wait_loading;
> +	} else if (mhdp->hw_state == MHDP_HW_READY) {
> +		stop_fw = true;
> +		timeout = 1; /* We were successful even if counter reached 0 */
> +	}
> +	mhdp->hw_state = MHDP_HW_STOPPED;
> +	spin_unlock(&mhdp->start_lock);
> +
> +	if (timeout == 0)
> +		dev_err(mhdp->dev, "%s: Timeout waiting for fw loading\n",
> +			__func__);
> +
> +	if (stop_fw) {
> +		ret = cdns_mhdp_set_firmware_active(mhdp, false);
> +		if (ret)
> +			dev_err(mhdp->dev, "%s: De-activate FW failed: %d\n",
> +				__func__, ret);
> +	}
> +
> +	phy_exit(mhdp->phy);
> +
> +	pm_runtime_put_sync(&pdev->dev);
> +	pm_runtime_disable(&pdev->dev);
> +
> +	clk_disable_unprepare(mhdp->clk);
> +
> +	/* FIXME: check for missing functions */
> +
> +	return ret;
> +}

The sequence here does not match what's done at probe()'s error handling. It probably should.

The FIXME here sounds something that has to be fixed (or removed).

There were also a few FIXMEs in the code, I think think it would be good to review all those, and 
see if they're still valid, and perhaps add a bit more explanation to each.

> +/* mailbox */
> +#define MAILBOX_RETRY_US			1000
> +#define MAILBOX_TIMEOUT_US			5000000

This is 5 seconds. It sounds pretty long time to wait in a blocking function.

  Tomi

-- 
Texas Instruments Finland Oy, Porkkalankatu 22, 00180 Helsinki.
Y-tunnus/Business ID: 0615521-4. Kotipaikka/Domicile: Helsinki

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ