this_cpu: Implement X86 optimized this_cpu operations
authorChristoph Lameter <cl@linux-foundation.org>
Sat, 3 Oct 2009 10:48:22 +0000 (19:48 +0900)
committerTejun Heo <tj@kernel.org>
Sat, 3 Oct 2009 10:48:22 +0000 (19:48 +0900)
commit30ed1a79f5bf271d33e782afee3323582dcc621e
tree47a242965c507bf5b27275e97cda38f00e03b0e6
parent7340a0b15280c9d902c7dd0608b8e751b5a7c403
this_cpu: Implement X86 optimized this_cpu operations

Basically the existing percpu ops can be used for this_cpu variants that allow
operations also on dynamically allocated percpu data. However, we do not pass a
reference to a percpu variable in. Instead a dynamically or statically
allocated percpu variable is provided.

Preempt, the non preempt and the irqsafe operations generate the same code.
It will always be possible to have the requires per cpu atomicness in a single
RMW instruction with segment override on x86.

64 bit this_cpu operations are not supported on 32 bit.

Signed-off-by: Christoph Lameter <cl@linux-foundation.org>
Signed-off-by: Tejun Heo <tj@kernel.org>
arch/x86/include/asm/percpu.h