From: James Hogan
> On 03/02/14 10:35, David Laight wrote:
> > From: James Hogan
> >> Combining __packed with __aligned(2) does the trick though (__packed
> >> alone sets __aligned(1) which is obviously going to be suboptimal).
...
>
> Meta is also one of those arches, and according to my quick tests,
> __packed alone does correctly make it fall back to byte loads/stores,
> but with __packed __aligned(2) it uses 16bit loads/stores. I've also
> confirmed that with an ARM toolchain (see below for example).
I would either:
1a) Add explicit padding to the relevant structures so that they are
multiple of 4 bytes.
or:
1b) #define some token to "__packed __aligned(2)" before all the structures
that require changing, and use that in there definitions.
This lets you comment on WHY you are doing it.
and:
2) Add a compile-time assert that the structures are the correct size.
Clearly you don't want to mark anything that contains a 32bit value
with __packed __aligned(2).
I'm not at all clear whether you are sometimes using a different compiler.
David
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
