x86-64, NUMA: Remove custom phys_to_nid() implementation
authorTejun Heo <tj@kernel.org>
Fri, 1 Apr 2011 09:15:12 +0000 (11:15 +0200)
committerTejun Heo <tj@kernel.org>
Fri, 1 Apr 2011 09:15:12 +0000 (11:15 +0200)
commit052936080c8fb2f791103995b21bd4018c8df886
treea195ff3279760f867b3b9120ed22f441f9e49f61
parent9d42a53e0f46505b39494041d514372235817e15
x86-64, NUMA: Remove custom phys_to_nid() implementation

phys_to_nid() maps physical address to NUMA node id.  This is
implemented by building perfect hash in compute_hash_shift() during
initialization.

However, with SPARSE memory model, the nid is encoded in page flags.
The perfect hash implementation was for DISCONTIG memory model which
got removed years ago by b263295dbf (x86: 64-bit, make sparsemem
vmemmap the only memory model).

So, the perfect hash ends up being used only during initialization
when the core SPARSE code already provides perfectly acceptable
generic early_pfn_to_nid() implementation.

Drop phys_to_nid() and use the generic ealry_pfn_to_nid() instead.

Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Christoph Lameter <cl@linux-foundation.org>
Acked-by: Yinghai Lu <yinghai@kernel.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
arch/x86/Kconfig
arch/x86/include/asm/mmzone_64.h
arch/x86/mm/numa_64.c