On Thu, Nov 12, 2015 at 10:57 AM, Matias Bjørling <m@xxxxxxxxxxx> wrote:
>
> Grah, I need to be more careful. I sent the wrong patch after I had fixed it
> to unsigned short.
Actually, I think "unsigned int" was better.
You're not saving any space with "unsigned short" (the size of the
structure will be rounded up to the alignment of it anyway), and we
should generally strive to avoid 16-bit accesses unless there is some
real reason for them, because they are often slower than either "char"
or "int". Several architectures have weak support for 16-bit accesses
(eg alpha), and even on x86 you end up having operand size overrides
etc.
So unless there is a clear *reason* to use "short" - just don't.
Linus
--
To unsubscribe from this list: send the line "unsubscribe linux-block" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
