From: Cristian Ciocaltea <cristian.ciocaltea@gmail.com>
To: Manivannan Sadhasivam <mani@kernel.org>
Cc: "Rob Herring" <robh+dt@kernel.org>,
"Andreas Färber" <afaerber@suse.de>,
devicetree@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
linux-actions@lists.infradead.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2 3/6] soc: actions: Add Actions Semi Owl socinfo driver
Date: Fri, 2 Apr 2021 22:36:08 +0300 [thread overview]
Message-ID: <20210402193608.GB2052516@BV030612LT> (raw)
In-Reply-To: <20210402181624.GC31906@thinkpad>
On Fri, Apr 02, 2021 at 11:46:24PM +0530, Manivannan Sadhasivam wrote:
> On Tue, Mar 30, 2021 at 04:48:18PM +0300, Cristian Ciocaltea wrote:
> > The driver provides information about the Action Semi Owl family of
> > SoCs (S500, S700 and S900) to user space via sysfs: machine, family,
> > soc_id, serial_number.
> >
> > Note the serial number is currently provided only for the S500 SoC
> > variant.
> >
> > Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@gmail.com>
> > ---
> > drivers/soc/actions/Kconfig | 8 +
> > drivers/soc/actions/Makefile | 1 +
> > drivers/soc/actions/owl-socinfo.c | 152 ++++++++++++++++++
> > include/linux/soc/actions/owl-serial-number.h | 20 +++
> > 4 files changed, 181 insertions(+)
> > create mode 100644 drivers/soc/actions/owl-socinfo.c
> > create mode 100644 include/linux/soc/actions/owl-serial-number.h
> >
> > diff --git a/drivers/soc/actions/Kconfig b/drivers/soc/actions/Kconfig
> > index 1aca2058a40c..15faade9282d 100644
> > --- a/drivers/soc/actions/Kconfig
> > +++ b/drivers/soc/actions/Kconfig
> > @@ -14,4 +14,12 @@ config OWL_PM_DOMAINS
> > power-gating on Actions Semiconductor S500, S700 and S900 SoCs.
> > If unsure, say 'n'.
> >
> > +config OWL_SOCINFO
> > + bool "Actions Semi Owl SoC info driver"
> > + default ARCH_ACTIONS
> > + select SOC_BUS
> > + help
> > + Say 'y' here to support the Action Semiconductor Owl socinfo
>
> Actions Semi
>
> > + driver, providing information about the SoC to user space.
> > +
> > endif
> > diff --git a/drivers/soc/actions/Makefile b/drivers/soc/actions/Makefile
> > index 4db9e7b050e5..4b2591d3089f 100644
> > --- a/drivers/soc/actions/Makefile
> > +++ b/drivers/soc/actions/Makefile
> > @@ -2,3 +2,4 @@
> >
> > obj-$(CONFIG_OWL_PM_DOMAINS_HELPER) += owl-sps-helper.o
> > obj-$(CONFIG_OWL_PM_DOMAINS) += owl-sps.o
> > +obj-$(CONFIG_OWL_SOCINFO) += owl-socinfo.o
> > diff --git a/drivers/soc/actions/owl-socinfo.c b/drivers/soc/actions/owl-socinfo.c
> > new file mode 100644
> > index 000000000000..f28eafac3792
> > --- /dev/null
> > +++ b/drivers/soc/actions/owl-socinfo.c
> > @@ -0,0 +1,152 @@
>
> [...]
>
> > + * Access SoC's serial number stored by the bootloader in DDR memory.
> > + */
> > +static int owl_socinfo_read_serial_rmem(struct device *dev)
> > +{
> > + struct reserved_mem *rmem;
> > + struct device_node *np;
> > + int ret = 0;
> > +
> > + np = of_find_compatible_node(NULL, NULL, "actions,owl-soc-serial");
> > + if (!np)
> > + return -ENXIO;
> > +
> > + rmem = of_reserved_mem_lookup(np);
>
> If you do this correctly, you could use "pdev->dev.of_node" here instead of
> using "of_find_compatible_node()" for getting np.
Right, as previously explained, my intention was to avoid binding the
driver on the serial number node in order to keep it generic enough.
> > + if (!rmem) {
> > + dev_err(dev, "failed to acquire reserved memory region\n");
> > + ret = -EINVAL;
> > + goto out_put;
> > + }
> > +
>
> [...]
>
> > +static const struct of_device_id owl_socinfo_of_match[] = {
> > + { .compatible = "actions,s500-soc", .data = &s500_soc_info, },
> > + { .compatible = "actions,s700-soc", .data = &s700_soc_info, },
> > + { .compatible = "actions,s900-soc", .data = &s900_soc_info, },
>
> Please don't add S700/S900 for now.
I assumed we can use the driver for S700/S900, even though the serial
number is not provided for them.
> > + { }
> > +};
> > +
> > +static struct platform_driver owl_socinfo_platform_driver = {
> > + .probe = owl_socinfo_probe,
> > + .driver = {
> > + .name = "owl-socinfo",
> > + .of_match_table = owl_socinfo_of_match,
> > + },
> > +};
> > +
> > +static int __init owl_socinfo_init(void)
> > +{
> > + return platform_driver_register(&owl_socinfo_platform_driver);
> > +}
> > +subsys_initcall(owl_socinfo_init);
> > diff --git a/include/linux/soc/actions/owl-serial-number.h b/include/linux/soc/actions/owl-serial-number.h
> > new file mode 100644
> > index 000000000000..f8595417668f
> > --- /dev/null
> > +++ b/include/linux/soc/actions/owl-serial-number.h
> > @@ -0,0 +1,20 @@
> > +/* SPDX-License-Identifier: GPL-2.0+ */
> > +/*
> > + * Copyright (c) 2021 Cristian Ciocaltea <cristian.ciocaltea@gmail.com>
> > + */
> > +
> > +#ifndef __SOC_ACTIONS_OWL_SERIAL_NUMBER_H__
> > +#define __SOC_ACTIONS_OWL_SERIAL_NUMBER_H__
> > +
> > +#if IS_ENABLED(CONFIG_OWL_SOCINFO)
> > +u32 owl_get_soc_serial_low(void);
> > +u32 owl_get_soc_serial_high(void);
>
> Where are these APIs used?
This will be used by the Ethernet MAC driver to generate a stable
address. Initially I used the global 'system_serial_{low,high}'
variables to pass this information outside the driver, but that is
not portable, i.e. only supported on ARM32, not ARM64.
Thanks for the review,
Cristi
> Thanks,
> Mani
>
> > +#else
> > +static inline u32 owl_get_soc_serial_low(void)
> > +{ return 0; }
> > +
> > +static inline u32 owl_get_soc_serial_high(void)
> > +{ return 0; }
> > +#endif /* CONFIG_OWL_SOCINFO */
> > +
> > +#endif /* __SOC_ACTIONS_OWL_SERIAL_NUMBER_H__ */
> > --
> > 2.31.1
> >
next prev parent reply other threads:[~2021-04-02 19:36 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-30 13:48 [PATCH v2 0/6] Add support for Actions Semi Owl socinfo Cristian Ciocaltea
2021-03-30 13:48 ` [PATCH v2 1/6] dt-bindings: reserved-memory: Add Owl SoC serial number binding Cristian Ciocaltea
2021-04-01 17:07 ` Rob Herring
2021-04-01 17:40 ` Cristian Ciocaltea
2021-04-02 18:06 ` Manivannan Sadhasivam
2021-04-02 19:44 ` Cristian Ciocaltea
2021-03-30 13:48 ` [PATCH v2 2/6] dt-bindings: soc: actions: Add Actions Semi Owl socinfo binding Cristian Ciocaltea
2021-04-01 17:08 ` Rob Herring
2021-04-01 17:57 ` Cristian Ciocaltea
2021-04-02 18:04 ` Manivannan Sadhasivam
2021-04-02 19:25 ` Cristian Ciocaltea
2021-03-30 13:48 ` [PATCH v2 3/6] soc: actions: Add Actions Semi Owl socinfo driver Cristian Ciocaltea
2021-04-02 18:16 ` Manivannan Sadhasivam
2021-04-02 19:36 ` Cristian Ciocaltea [this message]
2021-03-30 13:48 ` [PATCH v2 4/6] arm: dts: owl-s500: Add reserved-memory range for Owl SoC serial number Cristian Ciocaltea
2021-03-30 13:48 ` [PATCH v2 5/6] arm: dts: owl-s500: Add socinfo support Cristian Ciocaltea
2021-03-30 13:48 ` [PATCH v2 6/6] MAINTAINERS: Add entries for Owl reserved-memory and socinfo bindings Cristian Ciocaltea
2021-04-01 5:24 ` [PATCH v2 0/6] Add support for Actions Semi Owl socinfo Manivannan Sadhasivam
2021-04-01 9:40 ` Cristian Ciocaltea
2021-04-01 10:27 ` Manivannan Sadhasivam
2021-04-01 10:49 ` Andreas Färber
2021-04-01 10:58 ` Cristian Ciocaltea
2021-04-01 11:07 ` Manivannan Sadhasivam
2021-04-01 13:19 ` Cristian Ciocaltea
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210402193608.GB2052516@BV030612LT \
--to=cristian.ciocaltea@gmail.com \
--cc=afaerber@suse.de \
--cc=devicetree@vger.kernel.org \
--cc=linux-actions@lists.infradead.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mani@kernel.org \
--cc=robh+dt@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).