[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <4BFCCCC4.3030202@monstr.eu>
Date: Wed, 26 May 2010 09:24:52 +0200
From: Michal Simek <monstr@...str.eu>
To: Mike Frysinger <vapier@...too.org>
CC: uclinux-dev@...inux.org, David Howells <dhowells@...hat.com>,
David McCullough <davidm@...pgear.com>,
Greg Ungerer <gerg@...inux.org>,
Paul Mundt <lethal@...ux-sh.org>,
uclinux-dist-devel@...ckfin.uclinux.org,
microblaze-uclinux@...e.uq.edu.au, linux-m32r@...linux-m32r.org,
Hirokazu Takata <takata@...ux-m32r.org>,
linux-kernel@...r.kernel.org,
Yoshinori Sato <ysato@...rs.sourceforge.jp>,
Jie Zhang <jie.zhang@...log.com>
Subject: Re: [PATCH] FLAT: allow arches to declare a larger alignment than
the slab
Mike Frysinger wrote:
> From: Jie Zhang <jie.zhang@...log.com>
>
> The recent commit 1f0ce8b3dd667dca7 which moved the ARCH_SLAB_MINALIGN
> default into the global header inadvertently broke FLAT for a bunch of
> systems. Blackfin systems now fail on any FLAT exec with:
> Unable to read code+data+bss, errno 14
> When your /init is a FLAT binary, obviously this can be annoying ;).
>
> This stems from the alignment usage in the FLAT loader. The behavior
> before was that FLAT would default to ARCH_SLAB_MINALIGN only if it was
> defined, and this was only defined by arches when they wanted a larger
> alignment value. Otherwise it'd default to pointer alignment. Arguably,
> this is kind of hokey that the FLAT is semi-abusing defines it shouldn't.
>
> But let's ignore that and let arches declare a larger FLAT alignment
> specifically anyways as some arches are OK with the default slab alignment
> but need stricter FLAT alignments for shared FLAT libraries.
>
> The nommu arches might want to check to see if they need to declare this
> in their headers as well ...
Microblaze noMMU contains this fault too. I registered it yesterday.
I will use the same fix as you when you find out the correct solution. :-)
Michal
>
> Signed-off-by: Jie Zhang <jie.zhang@...log.com>
> Signed-off-by: Mike Frysinger <vapier@...too.org>
> ---
> arch/blackfin/include/asm/flat.h | 2 ++
> fs/binfmt_flat.c | 11 ++++++++---
> 2 files changed, 10 insertions(+), 3 deletions(-)
>
> diff --git a/arch/blackfin/include/asm/flat.h b/arch/blackfin/include/asm/flat.h
> index c1314c5..cf2a73e 100644
> --- a/arch/blackfin/include/asm/flat.h
> +++ b/arch/blackfin/include/asm/flat.h
> @@ -11,6 +11,8 @@
>
> #include <asm/unaligned.h>
>
> +#define ARCH_FLAT_DATA_ALIGN 0x20
> +
> #define flat_argvp_envp_on_stack() 0
> #define flat_old_ram_flag(flags) (flags)
>
> diff --git a/fs/binfmt_flat.c b/fs/binfmt_flat.c
> index 49566c1..6906170 100644
> --- a/fs/binfmt_flat.c
> +++ b/fs/binfmt_flat.c
> @@ -56,12 +56,17 @@
> #endif
>
> /*
> - * User data (stack, data section and bss) needs to be aligned
> - * for the same reasons as SLAB memory is, and to the same amount.
> + * User data (stack, data section and bss) needs to be aligned.
> + * If ARCH_FLAT_DATA_ALIGN is defined, use it.
> + */
> +#ifdef ARCH_FLAT_DATA_ALIGN
> +#define FLAT_DATA_ALIGN (ARCH_FLAT_DATA_ALIGN)
> +/* Otherwise user data nees to be aligned for the same reasons
> + * as SLAB memory is aligned, and to the same amount.
> * Avoid duplicating architecture specific code by using the same
> * macro as with SLAB allocation:
> */
> -#ifdef ARCH_SLAB_MINALIGN
> +#elif defined(ARCH_SLAB_MINALIGN)
> #define FLAT_DATA_ALIGN (ARCH_SLAB_MINALIGN)
> #else
> #define FLAT_DATA_ALIGN (sizeof(void *))
--
Michal Simek, Ing. (M.Eng)
w: www.monstr.eu p: +42-0-721842854
Maintainer of Linux kernel 2.6 Microblaze Linux - http://www.monstr.eu/fdt/
Microblaze U-BOOT custodian
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists