On Wed, 28 Nov 2007 13:34:02 +0100 (CET) Geert Uytterhoeven <[email protected]> wrote: > On Wed, 28 Nov 2007, Pierre Ossman wrote: > > > > Is there no directive we can stick in there that forces a reasonable alignment (e.g. alignment == sizeof(type)) independently of arch? > > We could use something like is used for compat_*. > E.g. compare compat_s64 in <asm/compat.h> for x86 and powerpc. > Yeah, that could work. Have a header with stuff like this: typedef u16 __attribute__((aligned(2))) aligned_u16; typedef u32 __attribute__((aligned(4))) aligned_u32; and let all structures in mod_devicetable.h use those types. Now does anyone have the time to code and test this? Rgds Pierre
Attachment:
signature.asc
Description: PGP signature
- Follow-Ups:
- Correct types for mod_devicetable.h (was: Re: m68k build failure)
- From: Geert Uytterhoeven <[email protected]>
- Correct types for mod_devicetable.h (was: Re: m68k build failure)
- References:
- m68k build failure
- From: Andrew Morton <[email protected]>
- Re: m68k build failure
- From: Pierre Ossman <[email protected]>
- Re: m68k build failure
- From: Andrew Morton <[email protected]>
- Re: m68k build failure
- From: Al Viro <[email protected]>
- Re: m68k build failure
- From: Pierre Ossman <[email protected]>
- Re: m68k build failure
- From: Geert Uytterhoeven <[email protected]>
- m68k build failure
- Prev by Date: Re: [BUG] Strange 1-second pauses during Resume-from-RAM
- Next by Date: Re: WARNING: at kernel/resource.c:189 __release_resource
- Previous by thread: Re: m68k build failure
- Next by thread: Correct types for mod_devicetable.h (was: Re: m68k build failure)
- Index(es):
 |