Re: [PATCH] Fix big endian ext2 bmap in 2.2.

From: Alexander Viro (viro@math.psu.edu)
Date: Tue Sep 19 2000 - 18:01:46 EST

Next message: Alexander Viro: "[PATCH] abuse of macros in swab.h"
Previous message: David S. Miller: "Re: [PATCH] Fix big endian ext2 bmap in 2.2."
In reply to: Andi Kleen: "[PATCH] Fix big endian ext2 bmap in 2.2."
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

On Wed, 20 Sep 2000, Andi Kleen wrote:

>
> This patch fixes an obvious bug introduced with the ext2 changes in 2.2.18pre
> (look up the definition of le32_to_cpu on BE machines without a special
> assembler version for it and on machines that have it)
>
> Patch against 2.2.18pre9

Wrong fix. swab32() is the guilty party here - it should be

static inline __u32 swab32(__u32 x)
{
        return
                ((x & (__u32)0x000000ffUL) << 24) |
                ((x & (__u32)0x0000ff00UL) << 8) |
                ((x & (__u32)0x00ff0000UL) >> 8) |
                ((x & (__u32)0xff000000UL) >> 24) );
}

Ditto for other stuff in linux/byteorder/swab.h

My fault - I didn't realize that the bloody thing was a macro (it
obviously shouldn't be, exactly due to problems with side effects).

Alan, what do you prefer? I think that fixing the macros in swab.h is the
Right Thing(tm).

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

Next message: Alexander Viro: "[PATCH] abuse of macros in swab.h"
Previous message: David S. Miller: "Re: [PATCH] Fix big endian ext2 bmap in 2.2."
In reply to: Andi Kleen: "[PATCH] Fix big endian ext2 bmap in 2.2."
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

This archive was generated by hypermail 2b29 : Sat Sep 23 2000 - 21:00:21 EST