Re: [PATCH v2 2/3] remoteproc: imx_rproc: mailbox support

From: Mathieu Poirier <mathieu.poirier@linaro.org>
To: nikita.shubin@maquefel.me
Cc: Nikita Shubin <NShubin@topcon.com>,
	Ohad Ben-Cohen <ohad@wizery.com>,
	Bjorn Andersson <bjorn.andersson@linaro.org>,
	Shawn Guo <shawnguo@kernel.org>,
	Sascha Hauer <s.hauer@pengutronix.de>,
	Pengutronix Kernel Team <kernel@pengutronix.de>,
	Fabio Estevam <festevam@gmail.com>,
	NXP Linux Team <linux-imx@nxp.com>,
	linux-remoteproc@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH v2 2/3] remoteproc: imx_rproc: mailbox support
Date: Tue, 14 Apr 2020 11:20:05 -0600	[thread overview]
Message-ID: <20200414172004.GB24061@xps15> (raw)
In-Reply-To: <20200406113310.3041-3-nikita.shubin@maquefel.me>

On Mon, Apr 06, 2020 at 02:33:09PM +0300, nikita.shubin@maquefel.me wrote:
> Add support for mailboxes to imx_rproc
> 
> Signed-off-by: Nikita Shubin <NShubin@topcon.com>
> ---
>  drivers/remoteproc/Kconfig     |   2 +
>  drivers/remoteproc/imx_rproc.c | 142 ++++++++++++++++++++++++++++++++-
>  2 files changed, 143 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/remoteproc/Kconfig b/drivers/remoteproc/Kconfig
> index 94afdde4bc9f..02d23a54c9cf 100644
> --- a/drivers/remoteproc/Kconfig
> +++ b/drivers/remoteproc/Kconfig
> @@ -17,6 +17,8 @@ if REMOTEPROC
>  config IMX_REMOTEPROC
>  	tristate "IMX6/7 remoteproc support"
>  	depends on ARCH_MXC
> +	select MAILBOX
> +	select IMX_MBOX
>  	help
>  	  Say y here to support iMX's remote processors (Cortex M4
>  	  on iMX7D) via the remote processor framework.
> diff --git a/drivers/remoteproc/imx_rproc.c b/drivers/remoteproc/imx_rproc.c
> index bebc58d0f711..d2bede4ccb70 100644
> --- a/drivers/remoteproc/imx_rproc.c
> +++ b/drivers/remoteproc/imx_rproc.c
> @@ -14,6 +14,9 @@
>  #include <linux/platform_device.h>
>  #include <linux/regmap.h>
>  #include <linux/remoteproc.h>
> +#include <linux/mailbox_client.h>
> +
> +#include "remoteproc_internal.h"
>  
>  #define IMX7D_SRC_SCR			0x0C
>  #define IMX7D_ENABLE_M4			BIT(3)
> @@ -47,6 +50,12 @@
>  
>  #define IMX_BOOT_PC			0x4
>  
> +#define IMX_MBOX_NB_VQ			2
> +#define IMX_MBOX_NB_MBX		2

Please align this.

> +
> +#define IMX_MBX_VQ0		"vq0"
> +#define IMX_MBX_VQ1		"vq1"
> +
>  /**
>   * struct imx_rproc_mem - slim internal memory structure
>   * @cpu_addr: MPU virtual address of the memory region
> @@ -80,6 +89,14 @@ struct imx_rproc_dcfg {
>  	size_t				att_size;
>  };
>  
> +struct imx_mbox {
> +	const unsigned char name[10];
> +	struct mbox_chan *chan;
> +	struct mbox_client client;
> +	struct work_struct vq_work;
> +	int vq_id;
> +};
> +
>  struct imx_rproc {
>  	struct device			*dev;
>  	struct regmap			*regmap;
> @@ -88,6 +105,8 @@ struct imx_rproc {
>  	struct imx_rproc_mem		mem[IMX7D_RPROC_MEM_MAX];
>  	struct clk			*clk;
>  	void __iomem			*bootreg;
> +	struct imx_mbox mb[IMX_MBOX_NB_MBX];
> +	struct workqueue_struct *workqueue;
>  };
>  
>  static const struct imx_rproc_att imx_rproc_att_imx7d[] = {
> @@ -251,10 +270,118 @@ static void *imx_rproc_da_to_va(struct rproc *rproc, u64 da, int len)
>  	return va;
>  }
>  
> +static void imx_rproc_mb_vq_work(struct work_struct *work)
> +{
> +	struct imx_mbox *mb = container_of(work, struct imx_mbox, vq_work);
> +	struct rproc *rproc = dev_get_drvdata(mb->client.dev);
> +
> +	if (rproc_vq_interrupt(rproc, mb->vq_id) == IRQ_NONE)
> +		dev_dbg(&rproc->dev, "no message found in vq%d\n", mb->vq_id);
> +}
> +
> +static void imx_rproc_mb_callback(struct mbox_client *cl, void *data)
> +{
> +	struct rproc *rproc = dev_get_drvdata(cl->dev);
> +	struct imx_mbox *mb = container_of(cl, struct imx_mbox, client);
> +	struct imx_rproc *ddata = rproc->priv;
> +
> +	queue_work(ddata->workqueue, &mb->vq_work);
> +}
> +
> +static const struct imx_mbox imx_rproc_mbox[IMX_MBOX_NB_MBX] = {
> +	{
> +		.name = IMX_MBX_VQ0,
> +		.vq_id = 0,
> +		.client = {
> +			.rx_callback = imx_rproc_mb_callback,
> +			.tx_block = false,
> +		},
> +	},
> +	{
> +		.name = IMX_MBX_VQ1,
> +		.vq_id = 1,
> +		.client = {
> +			.rx_callback = imx_rproc_mb_callback,
> +			.tx_block = false,
> +		},
> +	},
> +};
> +
> +static void imx_rproc_request_mbox(struct rproc *rproc)
> +{
> +	struct imx_rproc *ddata = rproc->priv;
> +	struct device *dev = &rproc->dev;
> +	unsigned int i;
> +	const unsigned char *name;
> +	struct mbox_client *cl;
> +
> +	/* Initialise mailbox structure table */
> +	memcpy(ddata->mb, imx_rproc_mbox, sizeof(imx_rproc_mbox));
> +
> +	for (i = 0; i < IMX_MBOX_NB_MBX; i++) {
> +		name = ddata->mb[i].name;
> +
> +		cl = &ddata->mb[i].client;
> +		cl->dev = dev->parent;
> +
> +		ddata->mb[i].chan = mbox_request_channel_byname(cl, name);
> +
> +		dev_dbg(dev, "%s: name=%s, idx=%u\n",
> +			__func__, name, ddata->mb[i].vq_id);
> +
> +		if (IS_ERR(ddata->mb[i].chan)) {
> +			dev_warn(dev, "cannot get %s mbox\n", name);
> +			ddata->mb[i].chan = NULL;

If the mailbox isn't ready this driver will fail without a chance of recovery.
Since most of the code in this patch is a carbon copy of the implementation
found in stm32_proc.c, I suggest you do the same as they did in
stm32_rproc_request_mbox() and privision for cases where requesting a channel
returns -EPROBE_DEFER.

> +		}
> +
> +		if (ddata->mb[i].vq_id >= 0)
> +			INIT_WORK(&ddata->mb[i].vq_work, imx_rproc_mb_vq_work);
> +	}
> +}
> +
> +static void imx_rproc_free_mbox(struct rproc *rproc)
> +{
> +	struct imx_rproc *ddata = rproc->priv;
> +	unsigned int i;
> +
> +	dev_dbg(&rproc->dev, "%s: %d boxes\n",
> +		__func__, ARRAY_SIZE(ddata->mb));
> +
> +	for (i = 0; i < ARRAY_SIZE(ddata->mb); i++) {
> +		if (ddata->mb[i].chan)
> +			mbox_free_channel(ddata->mb[i].chan);
> +		ddata->mb[i].chan = NULL;
> +	}
> +}
> +
> +static void imx_rproc_kick(struct rproc *rproc, int vqid)
> +{
> +	struct imx_rproc *ddata = rproc->priv;
> +	unsigned int i;
> +	int err;
> +
> +	if (WARN_ON(vqid >= IMX_MBOX_NB_VQ))
> +		return;
> +
> +	for (i = 0; i < IMX_MBOX_NB_MBX; i++) {
> +		if (vqid != ddata->mb[i].vq_id)
> +			continue;
> +		if (!ddata->mb[i].chan)
> +			return;
> +		dev_dbg(&rproc->dev, "sending message : vqid = %d\n", vqid);
> +		err = mbox_send_message(ddata->mb[i].chan, &vqid);
> +		if (err < 0)
> +			dev_err(&rproc->dev, "%s: failed (%s, err:%d)\n",
> +					__func__, ddata->mb[i].name, err);
> +			return;
> +	}
> +}
> +
>  static const struct rproc_ops imx_rproc_ops = {
>  	.start		= imx_rproc_start,
>  	.stop		= imx_rproc_stop,
>  	.da_to_va	= imx_rproc_da_to_va,
> +	.kick		= imx_rproc_kick,
>  	.get_boot_addr	= rproc_elf_get_boot_addr,
>  };
>  
> @@ -384,14 +511,26 @@ static int imx_rproc_probe(struct platform_device *pdev)
>  		goto err_put_rproc;
>  	}
>  
> +	priv->workqueue = create_workqueue(dev_name(dev));
> +	if (!priv->workqueue) {
> +		dev_err(dev, "cannot create workqueue\n");
> +		ret = -ENOMEM;
> +		goto err_put_clk;
> +	}
> +
> +	imx_rproc_request_mbox(rproc);
> +
>  	ret = rproc_add(rproc);
>  	if (ret) {
>  		dev_err(dev, "rproc_add failed\n");
> -		goto err_put_clk;
> +		goto err_free_mb;
>  	}
>  
>  	return 0;
>  
> +err_free_mb:
> +	imx_rproc_free_mbox(rproc);
> +	destroy_workqueue(priv->workqueue);
>  err_put_clk:
>  	clk_disable_unprepare(priv->clk);
>  err_put_rproc:
> @@ -407,6 +546,7 @@ static int imx_rproc_remove(struct platform_device *pdev)
>  
>  	clk_disable_unprepare(priv->clk);
>  	rproc_del(rproc);
> +	imx_rproc_free_mbox(rproc);

I have no issues with people reusing code already found in the kernel - in fact
I encourage it because it makes reviewing patches much easier.  On the flip side
you have to give credit where it is due.  Here adding a line in the changelog
that mentions where you took your inspiration from will be much appreciated.

Thanks,
Mathieu

>  	rproc_free(rproc);
>  
>  	return 0;
> -- 
> 2.25.1
> 