ARM: cache-l2x0: avoid taking spinlock for every iteration
authorRussell King <rmk+kernel@arm.linux.org.uk>
Thu, 19 Nov 2009 11:12:15 +0000 (11:12 +0000)
committerRussell King <rmk+kernel@arm.linux.org.uk>
Mon, 14 Dec 2009 13:34:58 +0000 (13:34 +0000)
commit0eb948dd7f7c3cec37440c16a6c738c8e75efcda
tree96ecc9b1a3b5326f7de82cd2055b448a7e43c9f8
parent22763c5cf3690a681551162c15d34d935308c8d7
ARM: cache-l2x0: avoid taking spinlock for every iteration

Taking the spinlock for every iteration is very expensive; instead,
batch iterations up into 4K blocks, releasing and reacquiring the
spinlock between each block.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
arch/arm/mm/cache-l2x0.c