Re: [PATCH] Optimize generic get_unaligned / put_unaligned implementations.

[Ralf Baechle <ralf@linux-mips.org>]

On Thu, Feb 15, 2007 at 05:27:20PM -0800, Andrew Morton wrote:

> No, icc surely supports attribute(packed).  My point is that we shouldn't
> rely upon the gcc info file for this, because other compilers can (or
> could) be used to build the kernel.
> 
> So it would be safer if the C spec said (or could be interpreted to say)
> "members of packed structures are always copied bytewise".  So then we
> can be reasonably confident that this change won't break the use of
> those compilers.
> 
> But then, I don't even know if any C standard says anything about packing.

Memory layout and alignment of structures and members are implementation
defined according to the C standard; the standard provides no means to
influence these.  So it takes a compiler extension such as gcc's
__attribute__().

> Ho hum.  Why are we talking about this, anyway?  Does the patch make the
> code faster?  Or just nicer?

Smaller binary and from looking at the disassembly a tad faster also.

  Ralf