BUILD_BUG_ON(): fix it and a couple of bogus uses of it
authorJan Beulich <JBeulich@novell.com>
Tue, 22 Sep 2009 23:43:52 +0000 (16:43 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Wed, 23 Sep 2009 14:39:29 +0000 (07:39 -0700)
commit8c87df457cb58fe75b9b893007917cf8095660a0
tree32446b329e4b83ae6158fa1505c36634a75dbfe8
parent1fe72eaa0f46a0fa4cdcd8f3f7853b6d39469784
BUILD_BUG_ON(): fix it and a couple of bogus uses of it

gcc permitting variable length arrays makes the current construct used for
BUILD_BUG_ON() useless, as that doesn't produce any diagnostic if the
controlling expression isn't really constant.  Instead, this patch makes
it so that a bit field gets used here.  Consequently, those uses where the
condition isn't really constant now also need fixing.

Note that in the gfp.h, kmemcheck.h, and virtio_config.h cases
MAYBE_BUILD_BUG_ON() really just serves documentation purposes - even if
the expression is compile time constant (__builtin_constant_p() yields
true), the array is still deemed of variable length by gcc, and hence the
whole expression doesn't have the intended effect.

[akpm@linux-foundation.org: make arch/sparc/include/asm/vio.h compile]
[akpm@linux-foundation.org: more nonsensical assertions in tpm.c..]
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Rajiv Andrade <srajiv@linux.vnet.ibm.com>
Cc: Mimi Zohar <zohar@us.ibm.com>
Cc: James Morris <jmorris@namei.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
arch/sparc/include/asm/vio.h
drivers/char/tpm/tpm.c
drivers/net/niu.c
include/linux/gfp.h
include/linux/kernel.h
include/linux/kmemcheck.h
include/linux/virtio_config.h