linux-kernel - Re: [PATCH v8 3/5] reset: imx8mp-audiomix: Add AudioMix Block Control reset driver

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]
Date: Fri, 21 Jun 2024 20:34:23 +0800
From: Shengjiu Wang <shengjiu.wang@...il.com>
To: Philipp Zabel <p.zabel@...gutronix.de>
Cc: Shengjiu Wang <shengjiu.wang@....com>, abelvesa@...nel.org, peng.fan@....com, 
	mturquette@...libre.com, sboyd@...nel.org, robh@...nel.org, 
	krzk+dt@...nel.org, conor+dt@...nel.org, shawnguo@...nel.org, 
	s.hauer@...gutronix.de, kernel@...gutronix.de, festevam@...il.com, 
	marex@...x.de, linux-clk@...r.kernel.org, imx@...ts.linux.dev, 
	devicetree@...r.kernel.org, linux-arm-kernel@...ts.infradead.org, 
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH v8 3/5] reset: imx8mp-audiomix: Add AudioMix Block Control
 reset driver

On Fri, Jun 21, 2024 at 6:59 PM Philipp Zabel <p.zabel@...gutronix.de> wrote:
>
> Hi,
>
> On Fr, 2024-06-14 at 15:42 +0800, Shengjiu Wang wrote:
> > Add support for the resets on i.MX8MP Audio Block Control module,
> > which includes the EARC PHY software reset and EARC controller
> > software reset. The reset controller is created using the auxiliary
> > device framework and set up in the clock driver.
> >
> > Signed-off-by: Shengjiu Wang <shengjiu.wang@....com>
> > Reviewed-by: Marco Felsch <m.felsch@...gutronix.de>
> > ---
> >  drivers/reset/Kconfig                 |   8 ++
> >  drivers/reset/Makefile                |   1 +
> >  drivers/reset/reset-imx8mp-audiomix.c | 106 ++++++++++++++++++++++++++
> >  3 files changed, 115 insertions(+)
> >  create mode 100644 drivers/reset/reset-imx8mp-audiomix.c
> >
> > diff --git a/drivers/reset/Kconfig b/drivers/reset/Kconfig
> > index 7112f5932609..b3c0e528d08c 100644
> > --- a/drivers/reset/Kconfig
> > +++ b/drivers/reset/Kconfig
> > @@ -91,6 +91,14 @@ config RESET_IMX7
> >       help
> >         This enables the reset controller driver for i.MX7 SoCs.
> >
> > +config RESET_IMX8MP_AUDIOMIX
> > +     tristate "i.MX8MP AudioMix Reset Driver"
> > +     depends on CLK_IMX8MP
>
> I'd like this to be made compile-testable without CLK_IMX8MP being
> enabled.

Ok, will remove the depends.

>
> > +     select AUXILIARY_BUS
> > +     default CLK_IMX8MP
> > +     help
> > +       This enables the reset controller driver for i.MX8MP AudioMix
> > +
> >  config RESET_INTEL_GW
> >       bool "Intel Reset Controller Driver"
> >       depends on X86 || COMPILE_TEST
> > diff --git a/drivers/reset/Makefile b/drivers/reset/Makefile
> > index fd8b49fa46fc..a6796e83900b 100644
> > --- a/drivers/reset/Makefile
> > +++ b/drivers/reset/Makefile
> > @@ -14,6 +14,7 @@ obj-$(CONFIG_RESET_BRCMSTB_RESCAL) += reset-brcmstb-rescal.o
> >  obj-$(CONFIG_RESET_GPIO) += reset-gpio.o
> >  obj-$(CONFIG_RESET_HSDK) += reset-hsdk.o
> >  obj-$(CONFIG_RESET_IMX7) += reset-imx7.o
> > +obj-$(CONFIG_RESET_IMX8MP_AUDIOMIX) += reset-imx8mp-audiomix.o
> >  obj-$(CONFIG_RESET_INTEL_GW) += reset-intel-gw.o
> >  obj-$(CONFIG_RESET_K210) += reset-k210.o
> >  obj-$(CONFIG_RESET_LANTIQ) += reset-lantiq.o
> > diff --git a/drivers/reset/reset-imx8mp-audiomix.c b/drivers/reset/reset-imx8mp-audiomix.c
> > new file mode 100644
> > index 000000000000..1fc984bc08c0
> > --- /dev/null
> > +++ b/drivers/reset/reset-imx8mp-audiomix.c
> > @@ -0,0 +1,106 @@
> > +// SPDX-License-Identifier: GPL-2.0-or-later
> > +/*
> > + * Copyright 2024 NXP
> > + */
> > +
> > +#include <linux/auxiliary_bus.h>
> > +#include <linux/device.h>
> > +#include <linux/io.h>
> > +#include <linux/module.h>
> > +#include <linux/of.h>
> > +#include <linux/of_address.h>
> > +#include <linux/of_platform.h>
> > +#include <linux/platform_device.h>
>
> Still needed?

Ok, can be removed.

>
> > +#include <linux/reset-controller.h>
> > +
> > +#define EARC                 0x200
> > +#define EARC_RESET_MASK              0x3
> > +
> > +struct imx8mp_audiomix_reset {
> > +     struct reset_controller_dev rcdev;
> > +     void __iomem *base;
> > +};
> > +
> > +static struct imx8mp_audiomix_reset *to_imx8mp_audiomix_reset(struct reset_controller_dev *rcdev)
> > +{
> > +     return container_of(rcdev, struct imx8mp_audiomix_reset, rcdev);
> > +}
> > +
> > +static int imx8mp_audiomix_reset_assert(struct reset_controller_dev *rcdev,
> > +                                     unsigned long id)
> > +{
> > +     struct imx8mp_audiomix_reset *priv = to_imx8mp_audiomix_reset(rcdev);
> > +     void __iomem *reg_addr = priv->base;
> > +     unsigned int mask, reg;
> > +
> > +     if (id >= fls(EARC_RESET_MASK))
> > +             return -EINVAL;
>
> This check is not required.
>
> Since you have nr_resets set to fls(EARC_RESET_MASK), the same is
> already checked in of_reset_simple_xlate, before a reset control is
> even returned.

Ok, will remove it.

>
> > +     mask = BIT(id);
> > +     reg = readl(reg_addr + EARC);
> > +     writel(reg & ~mask, reg_addr + EARC);
>
> There are multiple (well, two) resets in this register, so it would be
> good style to protect the read-modify-write cycle with a spinlock.

Ok, will add spinlock

>
> > +
> > +     return 0;
> > +}
> > +
> > +static int imx8mp_audiomix_reset_deassert(struct reset_controller_dev *rcdev,
> > +                                       unsigned long id)
> > +{
> > +     struct imx8mp_audiomix_reset *priv = to_imx8mp_audiomix_reset(rcdev);
> > +     void __iomem *reg_addr = priv->base;
> > +     unsigned int mask, reg;
> > +
> > +     if (id >= fls(EARC_RESET_MASK))
> > +             return -EINVAL;
> > +
> > +     mask = BIT(id);
> > +     reg = readl(reg_addr + EARC);
> > +     writel(reg | mask, reg_addr + EARC);
> > +
> > +     return 0;
> > +}
> > +
> > +static const struct reset_control_ops imx8mp_audiomix_reset_ops = {
> > +     .assert   = imx8mp_audiomix_reset_assert,
> > +     .deassert = imx8mp_audiomix_reset_deassert,
> > +};
> > +
> > +static int imx8mp_audiomix_reset_probe(struct auxiliary_device *adev,
> > +                                    const struct auxiliary_device_id *id)
> > +{
> > +     struct imx8mp_audiomix_reset *priv;
> > +     struct device *dev = &adev->dev;
> > +
> > +     priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
> > +     if (!priv)
> > +             return -ENOMEM;
> > +
> > +     priv->rcdev.owner     = THIS_MODULE;
> > +     priv->rcdev.nr_resets = fls(EARC_RESET_MASK);
> > +     priv->rcdev.ops       = &imx8mp_audiomix_reset_ops;
> > +     priv->rcdev.of_node   = dev->parent->of_node;
> > +     priv->rcdev.dev       = dev;
> > +     priv->rcdev.of_reset_n_cells = 1;
> > +     priv->base            = of_iomap(dev->parent->of_node, 0);
>
> This is missing the corresponding iounmap().

Ok, will add iounmap.

>
> I wonder why we map at all. If the parent driver already has iomem
> mapped, can't it just pass that in, like JH7110 does?

Because this is a simple reset driver, just two reset bits,
I want to avoid adding a header file for the parent driver and this driver.

Thanks for reviewing.

By the way: shall I just send a new version for this patch only next time?

Best regards
Shengjiu Wang
>
>
> regards
> Philipp