On Mon, Dec 04, 2023 at 05:37:18AM +0100, Christoph Hellwig wrote:
> On Thu, Nov 30, 2023 at 06:06:58PM -0800, Darrick J. Wong wrote:
> > I copy-pasta'd the whole mess from compiler_types.h and build_bug.h into
> > include/xfs.h. It works, but it might be kinda egregious though.
>
> Oh. I actually have a local patch to simply switch to static_assert
> as that completly relies on the compiler and gives better output. I
> haven't even written a proper commit log, but this is it:
>
> diff --git a/fs/xfs/xfs_ondisk.h b/fs/xfs/xfs_ondisk.h
> index 2f24bd42ac1dd7..3a5581ecb36d4c 100644
> --- a/fs/xfs/xfs_ondisk.h
> +++ b/fs/xfs/xfs_ondisk.h
> @@ -7,16 +7,16 @@
> #define __XFS_ONDISK_H
>
> #define XFS_CHECK_STRUCT_SIZE(structname, size) \
> - BUILD_BUG_ON_MSG(sizeof(structname) != (size), "XFS: sizeof(" \
> - #structname ") is wrong, expected " #size)
> + static_assert(sizeof(structname) == (size), \
> + "XFS: sizeof(" #structname ") is wrong, expected " #size)
>
> #define XFS_CHECK_OFFSET(structname, member, off) \
> - BUILD_BUG_ON_MSG(offsetof(structname, member) != (off), \
> + static_assert(offsetof(structname, member) == (off), \
> "XFS: offsetof(" #structname ", " #member ") is wrong, " \
> "expected " #off)
>
> #define XFS_CHECK_VALUE(value, expected) \
> - BUILD_BUG_ON_MSG((value) != (expected), \
> + static_assert((value) == (expected), \
HAH LOL that's much better. I think I even see that kernel code is
using it now, and wonder why BUG_BUILD_ON still exists.
Going back for another cup of koolaid now,
Reviewed-by: Darrick J. Wong <djwong@xxxxxxxxxx>
--D
> "XFS: value of " #value " is wrong, expected " #expected)
>
> static inline void __init
>
