Re: [PATCH v5 3/4] remoteproc: imx_dsp_rproc: Add remoteproc driver for DSP on i.MX

From: Shengjiu Wang
Date: Thu Oct 07 2021 - 21:53:35 EST


Hi Mathieu

On Thu, Oct 7, 2021 at 12:25 AM Mathieu Poirier
<mathieu.poirier@xxxxxxxxxx> wrote:
>
> Hi Shengjiu,
>
> This pachset doesn't apply to rproc-next, which is now located here[1]. The
> change is in linux-next but not in mainline yet.
>
> https://git.kernel.org/pub/scm/linux/kernel/git/remoteproc/linux.git/log/?h=rproc-next

Ok, I will double check it and fix it.

>
> On Sun, Sep 26, 2021 at 11:07:09AM +0800, Shengjiu Wang wrote:
> > Provide a basic driver to control DSP processor found on NXP i.MX8QM,
> > i.MX8QXP, i.MX8MP and i.MX8ULP.
> >
> > Currently it is able to resolve addresses between DSP and main CPU,
> > start and stop the processor, suspend and resume.
> >
> > The communication between DSP and main CPU is based on mailbox, there
> > are three mailbox channels (tx, rx, rxdb).
> >
> > This driver was tested on NXP i.MX8QM, i.MX8QXP, i.MX8MP and i.MX8ULP.
> >
> > Signed-off-by: Shengjiu Wang <shengjiu.wang@xxxxxxx>
> > ---
> > drivers/remoteproc/Kconfig | 11 +
> > drivers/remoteproc/Makefile | 1 +
> > drivers/remoteproc/imx_dsp_rproc.c | 1206 ++++++++++++++++++++++++++++
> > 3 files changed, 1218 insertions(+)
> > create mode 100644 drivers/remoteproc/imx_dsp_rproc.c
> >
>
> [...]
>
> > +
> > +/**
> > + * imx_dsp_attach_pm_domains() - attach the power domains
> > + * @priv: private data pointer
> > + *
> > + * On i.MX8QM and i.MX8QXP there is multiple power domains
> > + * required, so need to link them.
> > + */
> > +static int imx_dsp_attach_pm_domains(struct imx_dsp_rproc *priv)
> > +{
> > + struct device *dev = priv->rproc->dev.parent;
> > + int ret, i;
> > +
> > + priv->num_domains = of_count_phandle_with_args(dev->of_node,
> > + "power-domains",
> > + "#power-domain-cells");
> > +
> > + /* If only one domain, then no need to link the device */
> > + if (priv->num_domains <= 1)
> > + return 0;
> > +
> > + priv->pd_dev = devm_kmalloc_array(dev, priv->num_domains,
> > + sizeof(*priv->pd_dev),
> > + GFP_KERNEL);
> > + if (!priv->pd_dev)
> > + return -ENOMEM;
> > +
> > + priv->pd_dev_link = devm_kmalloc_array(dev, priv->num_domains,
> > + sizeof(*priv->pd_dev_link),
> > + GFP_KERNEL);
> > + if (!priv->pd_dev_link)
> > + return -ENOMEM;
> > +
> > + for (i = 0; i < priv->num_domains; i++) {
> > + priv->pd_dev[i] = dev_pm_domain_attach_by_id(dev, i);
> > + if (IS_ERR(priv->pd_dev[i])) {
> > + ret = PTR_ERR(priv->pd_dev[i]);
> > + goto detach_pm;
> > + }
>
> I have pointed a problem with the error handling in the above during the
> previous review and it was not addressed.

I have considered your comments. Actually when
dev_pm_domain_attach_by_id() return NULL, the device_link_add()
will break, I have added comments below, so above error handling
for dev_pm_domain_attach_by_id() is enough.

Best regards
Wang Shengjiu
>
> > +
> > + /*
> > + * device_link_add will check priv->pd_dev[i], if it is
> > + * NULL, then will break.
> > + */
> > + priv->pd_dev_link[i] = device_link_add(dev,
> > + priv->pd_dev[i],
> > + DL_FLAG_STATELESS |
> > + DL_FLAG_PM_RUNTIME);
> > + if (!priv->pd_dev_link[i]) {
> > + dev_pm_domain_detach(priv->pd_dev[i], false);
> > + ret = -EINVAL;
> > + goto detach_pm;
> > + }
> > + }
> > +
> > + return 0;
> > +
> > +detach_pm:
> > + while (--i >= 0) {
> > + device_link_del(priv->pd_dev_link[i]);
> > + dev_pm_domain_detach(priv->pd_dev[i], false);
> > + }
> > +
> > + return ret;
> > +}
> > +