x86/espfix: Init espfix on the boot CPU side
authorZhu Guihua <zhugh.fnst@cn.fujitsu.com>
Fri, 3 Jul 2015 09:37:19 +0000 (17:37 +0800)
committerIngo Molnar <mingo@kernel.org>
Mon, 6 Jul 2015 13:00:34 +0000 (15:00 +0200)
commit20d5e4a9cd453991e2590a4c25230a99b42dee0c
tree2471151b639d5de698f1d3d9a36fb49a9b07189a
parent1db875631f8d5bbf497f67e47f254eece888d51d
x86/espfix: Init espfix on the boot CPU side

As we alloc pages with GFP_KERNEL in init_espfix_ap() which is
called before we enable local irqs, so the lockdep sub-system
would (correctly) trigger a warning about the potentially
blocking API.

So we allocate them on the boot CPU side when the secondary CPU is
brought up by the boot CPU, and hand them over to the secondary
CPU.

And we use alloc_pages_node() with the secondary CPU's node, to
make sure the espfix stack is NUMA-local to the CPU that is
going to use it.

Signed-off-by: Zhu Guihua <zhugh.fnst@cn.fujitsu.com>
Cc: <bp@alien8.de>
Cc: <luto@amacapital.net>
Cc: <luto@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/c97add2670e9abebb90095369f0cfc172373ac94.1435824469.git.zhugh.fnst@cn.fujitsu.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
arch/x86/kernel/espfix_64.c
arch/x86/kernel/smpboot.c