Re: [PATCH v2 2/2] of: Stop naming platform_device using dcr address

From: Benjamin Herrenschmidt
Date: Sat May 24 2014 - 00:10:40 EST


On Fri, 2014-05-23 at 08:36 +0900, Grant Likely wrote:
> There is now a way to ensure all platform devices get a unique name when
> populated from the device tree, and the DCR_NATIVE code path is broken
> anyway. PowerPC Cell (PS3) is the only platform that actually uses this
> path. Most likely nobody will notice if it is killed. Remove the code
> and associated ugly #ifdef.
>
> The user-visible impact of this patch is that any DCR device on Cell
> will get a new name in the /sys/devices hierarchy.

I don't understand. First DCR devices cover more than Cell (and not even
PS3). All 4xx embedded have some form of DCR based devices so I don't
know how much we use as "platform".

What do you mean by "the DCR_NATIVE" code path is broken anyway ?

Changing user visible platform device names has broken stuff in the
past, I'm not saying it is now but I'd like a better justification
for removing this.

Cheers,
Ben.


> Signed-off-by: Grant Likely <grant.likely@xxxxxxxxxx>
> Cc: Rob Herring <robh@xxxxxxxxxx>
> Cc: Benjamin Herrenschmidt <benh@xxxxxxxxxxxxxxxxxxx>
> ---
> arch/powerpc/include/asm/dcr-mmio.h | 4 ----
> arch/powerpc/sysdev/dcr.c | 6 +++---
> drivers/of/platform.c | 24 ------------------------
> 3 files changed, 3 insertions(+), 31 deletions(-)
>
> diff --git a/arch/powerpc/include/asm/dcr-mmio.h b/arch/powerpc/include/asm/dcr-mmio.h
> index acd491dbd45a..93a68b28e695 100644
> --- a/arch/powerpc/include/asm/dcr-mmio.h
> +++ b/arch/powerpc/include/asm/dcr-mmio.h
> @@ -51,10 +51,6 @@ static inline void dcr_write_mmio(dcr_host_mmio_t host,
> out_be32(host.token + ((host.base + dcr_n) * host.stride), value);
> }
>
> -extern u64 of_translate_dcr_address(struct device_node *dev,
> - unsigned int dcr_n,
> - unsigned int *stride);
> -
> #endif /* __KERNEL__ */
> #endif /* _ASM_POWERPC_DCR_MMIO_H */
>
> diff --git a/arch/powerpc/sysdev/dcr.c b/arch/powerpc/sysdev/dcr.c
> index 1bd0eba4d355..e9056e438575 100644
> --- a/arch/powerpc/sysdev/dcr.c
> +++ b/arch/powerpc/sysdev/dcr.c
> @@ -152,9 +152,9 @@ EXPORT_SYMBOL_GPL(dcr_resource_len);
>
> #ifdef CONFIG_PPC_DCR_MMIO
>
> -u64 of_translate_dcr_address(struct device_node *dev,
> - unsigned int dcr_n,
> - unsigned int *out_stride)
> +static u64 of_translate_dcr_address(struct device_node *dev,
> + unsigned int dcr_n,
> + unsigned int *out_stride)
> {
> struct device_node *dp;
> const u32 *p;
> diff --git a/drivers/of/platform.c b/drivers/of/platform.c
> index 95c133a0554b..52780a72d09d 100644
> --- a/drivers/of/platform.c
> +++ b/drivers/of/platform.c
> @@ -51,10 +51,6 @@ struct platform_device *of_find_device_by_node(struct device_node *np)
> }
> EXPORT_SYMBOL(of_find_device_by_node);
>
> -#if defined(CONFIG_PPC_DCR)
> -#include <asm/dcr.h>
> -#endif
> -
> #ifdef CONFIG_OF_ADDRESS
> /*
> * The following routines scan a subtree and registers a device for
> @@ -78,26 +74,6 @@ void of_device_make_bus_id(struct device *dev)
> const __be32 *reg;
> u64 addr;
>
> -#ifdef CONFIG_PPC_DCR
> - /*
> - * If it's a DCR based device, use 'd' for native DCRs
> - * and 'D' for MMIO DCRs.
> - */
> - reg = of_get_property(node, "dcr-reg", NULL);
> - if (reg) {
> -#ifdef CONFIG_PPC_DCR_NATIVE
> - dev_set_name(dev, "d%x.%s", *reg, node->name);
> -#else /* CONFIG_PPC_DCR_NATIVE */
> - u64 addr = of_translate_dcr_address(node, *reg, NULL);
> - if (addr != OF_BAD_ADDR) {
> - dev_set_name(dev, "D%llx.%s",
> - (unsigned long long)addr, node->name);
> - return;
> - }
> -#endif /* !CONFIG_PPC_DCR_NATIVE */
> - }
> -#endif /* CONFIG_PPC_DCR */
> -
> /* Construct the name, using parent nodes if necessary to ensure uniqueness */
> while (node->parent) {
> /*


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/