arm64: mm: use bit ops rather than arithmetic in pa/va translations
authorArd Biesheuvel <ard.biesheuvel@linaro.org>
Mon, 22 Feb 2016 17:46:04 +0000 (18:46 +0100)
committerCatalin Marinas <catalin.marinas@arm.com>
Tue, 23 Feb 2016 17:43:25 +0000 (17:43 +0000)
commit8439e62a15614e8fcd43835d57b7245cd9870dc5
tree8ace79af5dced3e6f18b5000d6ae4e9c5e6fa5c6
parenta92405f082d43267575444a6927085e4c8a69e4e
arm64: mm: use bit ops rather than arithmetic in pa/va translations

Since PAGE_OFFSET is chosen such that it cuts the kernel VA space right
in half, and since the size of the kernel VA space itself is always a
power of 2, we can treat PAGE_OFFSET as a bitmask and replace the
additions/subtractions with 'or' and 'and-not' operations.

For the comparison against PAGE_OFFSET, a mov/cmp/branch sequence ends
up getting replaced with a single tbz instruction. For the additions and
subtractions, we save a mov instruction since the mask is folded into the
instruction's immediate field.

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
arch/arm64/include/asm/memory.h