Re: [PATCH] ARM: Add support for Realtek SOC

From: Arnd Bergmann
Date: Thu Sep 05 2019 - 04:31:20 EST


On Thu, Sep 5, 2019 at 7:48 AM <jamestai.sky@xxxxxxxxx> wrote:
>
> From: "james.tai" <james.tai@xxxxxxxxxxx>
>
> This patch adds the basic machine file for
> the Realtek RTD16XX platform.
>
> Signed-off-by: james.tai <james.tai@xxxxxxxxxxx>

Hi James,

Thanks a lot for your submission! I'm glad to see interest in upstream
support for this SoC family. I have a few small comments on
details, mostly where I would either like to see an explanation
in the patch description, or things that looks like they can be
left out from the patch.

> index 33b00579beff..c7c9a3662eb7 100644
> --- a/arch/arm/Kconfig
> +++ b/arch/arm/Kconfig
> @@ -836,6 +836,8 @@ source "arch/arm/mach-zx/Kconfig"
>
> source "arch/arm/mach-zynq/Kconfig"
>
> +source "arch/arm/mach-realtek/Kconfig"
> +
> diff --git a/arch/arm/mach-realtek/Kconfig b/arch/arm/mach-realtek/Kconfig
> @@ -225,6 +226,7 @@ machine-$(CONFIG_ARCH_VT8500) += vt8500
> machine-$(CONFIG_ARCH_W90X900) += w90x900
> machine-$(CONFIG_ARCH_ZX) += zx
> machine-$(CONFIG_ARCH_ZYNQ) += zynq
> +machine-$(CONFIG_ARCH_REALTEK) += realtek
> machine-$(CONFIG_PLAT_SPEAR) += spear
>
> # Platform directory name. This list is sorted alphanumerically

Please keep these lists in alphabetical order.

> # ARMv7-M architecture
> config ARCH_EFM32
> bool "Energy Micro efm32"
> diff --git a/arch/arm/Makefile b/arch/arm/Makefile
> index c3624ca6c0bc..1f0926449d47 100644
> --- a/arch/arm/Makefile
> +++ b/arch/arm/Makefile
> @@ -148,6 +148,7 @@ endif
> textofs-$(CONFIG_ARCH_MSM8X60) := 0x00208000
> textofs-$(CONFIG_ARCH_MSM8960) := 0x00208000
> textofs-$(CONFIG_ARCH_MESON) := 0x00208000
> +textofs-$(CONFIG_ARCH_REALTEK) := 0x00208000
> textofs-$(CONFIG_ARCH_AXXIA) := 0x00308000

Can you explain why this is needed for your platform?

> # Machine directory name. This list is sorted alphanumerically
> new file mode 100644
> index 000000000000..a8269964dbdb
> --- /dev/null
> +++ b/arch/arm/mach-realtek/Kconfig
> @@ -0,0 +1,32 @@
> +# SPDX-License-Identifier: GPL-2.0-only
> +menuconfig ARCH_REALTEK
> + bool "Realtek SoCs"

Please add "depends on ARCH_MULTI_V7" to avoid
compile time issues when selecting it on an earlier
architecture.

> + select ARM_GLOBAL_TIMER
> + select CLKDEV_LOOKUP
> + select HAVE_SMP
> + select HAVE_MACH_CLKDEV
> + select GENERIC_CLOCKEVENTS
> + select HAVE_SCHED_CLOCK
> + select ARCH_HAS_CPUFREQ
> + select CLKSRC_OF
> + select ARCH_REQUIRE_GPIOLIB
> + select GENERIC_IRQ_CHIP
> + select IRQ_DOMAIN
> + select PINCTRL
> + select COMMON_CLK
> + select ARCH_HAS_BARRIERS
> + select SPARSE_IRQ
> + select PM_OPP
> + select ARM_HAS_SG_CHAIN
> + select ARM_PATCH_PHYS_VIRT
> + select AUTO_ZRELADDR
> + select MIGHT_HAVE_PCI
> + select MULTI_IRQ_HANDLER
> + select PCI_DOMAINS if PCI
> + select USE_OF

Almost all of the symbols above are implied by
ARCH_MULTI_V7 and should not be selected
separately.

> +config ARCH_RTD16XX
> + bool "Enable support for RTD1619"
> + depends on ARCH_REALTEK
> + select ARM_GIC_V3
> + select ARM_PSCI

As I understand, this chip uses a Cortex-A55. What is the reason
for adding support only to the 32-bit ARM architecture rather than
64-bit?

Most 64-bit SoCs are only supported with arch/arm64, but generally
speaking that is not a requirement. My rule of thumb is that on
systems with 1GB of RAM or more, one would want to run a 64-bit
kernel, while systems with less than that are better off with a 32-bit
one, but that is clearly not the only reason for picking one over the
other.

> +
> +static int rtk_boot_secondary(unsigned int cpu, struct task_struct *idle)
> +{
> + unsigned long entry_pa = __pa_symbol(secondary_startup);
> +
> + writel_relaxed(entry_pa | (cpu << CPUID), cpu_release_virt);
> +
> + arch_send_wakeup_ipi_mask(cpumask_of(cpu));
> +
> + return 0;
> +}

It's very unusual to see custom smp operations on an ARMv8
system, as we normally use PSCI here. Can you explain what
is going on here? Are you able to use a boot wrapper that implements
these in psci instead?
> +
> +#include "platsmp.h"
> +
> +#define RBUS_BASE_PHYS (0x98000000)
> +#define RBUS_BASE_VIRT (0xfe000000)
> +#define RBUS_BASE_SIZE (0x00100000)
> +
> +static struct map_desc rtk_io_desc[] __initdata = {
> + {
> + .virtual = (unsigned long) IOMEM(RBUS_BASE_VIRT),
> + .pfn = __phys_to_pfn(RBUS_BASE_PHYS),
> + .length = RBUS_BASE_SIZE,
> + .type = MT_DEVICE,
> + },
> +};

This needs a comment: Why do you require a static mapping for
"RBUS_BASE_PHYS"? Normally device drivers should just use
ioremap() for mapping whichever registers they want to access.

> +static void __init rtk_dt_init(void)
> +{
> + of_platform_populate(NULL, of_default_bus_match_table, NULL, NULL);
> +}

This should be taken care of by the
of_platform_default_populate_init() and can be dropped.

> +static void __init rtk_timer_init(void)
> +{
> +#ifdef CONFIG_COMMON_CLK
> + of_clk_init(NULL);
> +#endif

COMMON_CLK is implied by ARCH_MULTI_V7, so the
#ifdef can be dropped.

> + timer_probe();
> + tick_setup_hrtimer_broadcast();
> +}

What do you need tick_setup_hrtimer_broadcast() for? I don't
see any other platform calling this.

> +bool __init rtk_smp_init_ops(void)
> +{
> + smp_set_ops(smp_ops(rtk_smp_ops));
> +
> + return true;
> +}

I think this can also be dropped, as you set the smp_ops in the
machine descriptor.


Arnd