[U-Boot] [PATCH 1/2] x86: move arch-specific asmlinkage to <asm/linkage.h>
Simon Glass
sjg at chromium.org
Sun Dec 7 22:44:21 CET 2014
Hi Masahiro,
On 3 December 2014 at 01:36, Masahiro Yamada <yamada.m at jp.panasonic.com> wrote:
> Commit 65dd74a674d6 (x86: ivybridge: Implement SDRAM init) introduced
> x86-specific asmlinkage into arch/x86/include/asm/config.h.
>
> Commit ed0a2fbf14f7 (x86: Add a definition of asmlinkage) added the
> same macro define again, this time, into include/common.h.
> (Please do not add arch-specific stuff to include/common.h any more;
> it is already too cluttered.)
>
> The generic asmlinkage is defined in <linux/linkage.h>. If you want
> to override it with an arch-specific one, the best way is to add it
> to <asm/linkage.h> like Linux Kernel.
>
> Signed-off-by: Masahiro Yamada <yamada.m at jp.panasonic.com>
> Cc: Simon Glass <sjg at chromium.org>
> ---
>
> arch/x86/include/asm/arch-ivybridge/pei_data.h | 2 ++
> arch/x86/include/asm/config.h | 1 -
> arch/x86/include/asm/linkage.h | 6 ++++++
> arch/x86/lib/bios.c | 1 +
> arch/x86/lib/bios.h | 2 ++
> include/common.h | 3 ---
> include/linux/linkage.h | 2 ++
> 7 files changed, 13 insertions(+), 4 deletions(-)
> create mode 100644 arch/x86/include/asm/linkage.h
>
> diff --git a/arch/x86/include/asm/arch-ivybridge/pei_data.h b/arch/x86/include/asm/arch-ivybridge/pei_data.h
> index 5026c8b..9453336 100644
> --- a/arch/x86/include/asm/arch-ivybridge/pei_data.h
> +++ b/arch/x86/include/asm/arch-ivybridge/pei_data.h
> @@ -7,6 +7,8 @@
> #ifndef ASM_ARCH_PEI_DATA_H
> #define ASM_ARCH_PEI_DATA_H
>
> +#include <linux/linkage.h>
> +
> struct pch_usb3_controller_settings {
> /* 0: Disable, 1: Enable, 2: Auto, 3: Smart Auto */
> uint16_t mode;
> diff --git a/arch/x86/include/asm/config.h b/arch/x86/include/asm/config.h
> index c97d988..ff15828 100644
> --- a/arch/x86/include/asm/config.h
> +++ b/arch/x86/include/asm/config.h
> @@ -10,6 +10,5 @@
> #define CONFIG_SYS_GENERIC_BOARD
> #define CONFIG_LMB
> #define CONFIG_SYS_BOOT_RAMDISK_HIGH
> -#define asmlinkage __attribute__((regparm(0)))
>
> #endif
> diff --git a/arch/x86/include/asm/linkage.h b/arch/x86/include/asm/linkage.h
> new file mode 100644
> index 0000000..bdca72e
> --- /dev/null
> +++ b/arch/x86/include/asm/linkage.h
> @@ -0,0 +1,6 @@
> +#ifndef _ASM_X86_LINKAGE_H
> +#define _ASM_X86_LINKAGE_H
> +
> +#define asmlinkage CPP_ASMLINKAGE __attribute__((regparm(0)))
Why CPP_ASMLINKAGE here?
> +
> +#endif /* _ASM_X86_LINKAGE_H */
> diff --git a/arch/x86/lib/bios.c b/arch/x86/lib/bios.c
> index 298fca6..d1f8933 100644
> --- a/arch/x86/lib/bios.c
> +++ b/arch/x86/lib/bios.c
> @@ -9,6 +9,7 @@
> #include <common.h>
> #include <bios_emul.h>
> #include <vbe.h>
> +#include <linux/linkage.h>
> #include <asm/cache.h>
> #include <asm/processor.h>
> #include <asm/i8259.h>
> diff --git a/arch/x86/lib/bios.h b/arch/x86/lib/bios.h
> index 8491b4a..668f4b5 100644
> --- a/arch/x86/lib/bios.h
> +++ b/arch/x86/lib/bios.h
> @@ -10,6 +10,8 @@
> #ifndef _X86_LIB_BIOS_H
> #define _X86_LIB_BIOS_H
>
> +#include <linux/linkage.h>
> +
> #define REALMODE_BASE 0x600
>
> #ifdef __ASSEMBLY__
> diff --git a/include/common.h b/include/common.h
> index 94c354b..f1ab2cf 100644
> --- a/include/common.h
> +++ b/include/common.h
> @@ -73,9 +73,6 @@ typedef volatile unsigned char vu_char;
> #ifdef CONFIG_ARM
> #define asmlinkage /* nothing */
> #endif
> -#ifdef CONFIG_X86
> -#define asmlinkage __attribute__((regparm(0)))
> -#endif
> #ifdef CONFIG_BLACKFIN
> #include <asm/blackfin.h>
> #endif
> diff --git a/include/linux/linkage.h b/include/linux/linkage.h
> index 7435fcd..5797498 100644
> --- a/include/linux/linkage.h
> +++ b/include/linux/linkage.h
> @@ -17,7 +17,9 @@
> #define CPP_ASMLINKAGE
> #endif
>
> +#ifndef asmlinkage
> #define asmlinkage CPP_ASMLINKAGE
> +#endif
>
> #define SYMBOL_NAME_STR(X) #X
> #define SYMBOL_NAME(X) X
> --
> 1.9.1
>
Tested on chromebook_link:
Tested-by: Simon Glass <sjg at chromium.org>
With the above question answered, I'd like to apply this as it is a
clean-up. Is it OK to so this independently of the ARM patch?
Regards,
Simon
More information about the U-Boot
mailing list