[PATCH] ARM: Do not allow unaligned accesses when CONFIG_ALIGNMENT_TRAP

On Thu, 26 May 2011, Will Deacon wrote:

This issue seems to be caused by passing -fconserve-stack to GCC. This
was added in 8f7f5c9f ("kbuild: set -fconserve-stack option for gcc
4.5") and as you can see from the archive:

http://lkml.org/lkml/2009/9/20/39

it was thought to only have an impact on inlining decisions. Looking at
the documentation for GCC 4.6:

-fconserve-stack
          Attempt to minimize stack usage. The compiler will attempt to
use less stack space, even if that makes the program slower. This option
implies setting the ?large-stack-frame? parameter to 100 and the
?large-stack-frame-growth? parameter to 400.

So it sounds like we might not want to enable this blindly across all
architectures. Indeed, on ARM, it encourages the compiler to pack
variables on the stack which leads to the weird and wonderful alignment
situation that has been encountered in this thread.

Can we remove -fconserve-stack from the top-level Makefile (or at least
make it conditional by architecture)?

I think this is an orthogonal issue.

My opinion is that we should use -mno-unaligned-access by default on 
ARM.  The reason is that we've been expecting the compiler not to cause 
unaligned accesses for ages, and letting the compiler, for whatever 
reasons including things like -fconserve-stack, produce unaligned 
accesses behind our back is a change in behavior we might not always be 
prepared for.  Unaligned accesses in the kernel should be rare anyway, 
and allowing the compiler to generate them can be allowed for selected 
files when proven beneficial.

It is possible that -fconserve-stack is still valuable on ARM given that 
it is also used with -mno-unaligned-access for other things than 
structure packing on the stack, and therefore its merits can be debated 
independently from the alignment issue at hand.


Nicolas