Commit
b9a50f74905a ("ARM: 7450/1: dcache: select DCACHE_WORD_ACCESS for
little-endian ARMv6+ CPUs") added support for word-at-time path
comparisons, relying on the ability to perform unaligned loads with
negligible performance impact in hardware.
For nommu configurations without MPU support, this is unpredictable and
so we should fall back to the byte-by-byte routines.
Acked-by: Nicolas Pitre <nico@linaro.org>
Tested-by: Jonathan Austin <jonathan.austin@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
select ARCH_WANT_IPC_PARSE_VERSION
select BUILDTIME_EXTABLE_SORT if MMU
select CPU_PM if (SUSPEND || CPU_IDLE)
- select DCACHE_WORD_ACCESS if (CPU_V6 || CPU_V6K || CPU_V7) && !CPU_BIG_ENDIAN
+ select DCACHE_WORD_ACCESS if (CPU_V6 || CPU_V6K || CPU_V7) && !CPU_BIG_ENDIAN && MMU
select GENERIC_ATOMIC64 if (CPU_V6 || !CPU_32v6K || !AEABI)
select GENERIC_CLOCKEVENTS_BROADCAST if SMP
select GENERIC_IRQ_PROBE