ARM: implement support for read-mostly sections
As our SMP implementation uses MESI protocols. Grouping together data which is mostly only read together means that we avoid unnecessary cache line bouncing when this code shares a cache line with other data. In other words, cache lines associated with read-mostly data are expected to spend most of their time in shared state. Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
This commit is contained in:
@@ -23,4 +23,6 @@
|
|||||||
#define ARCH_SLAB_MINALIGN 8
|
#define ARCH_SLAB_MINALIGN 8
|
||||||
#endif
|
#endif
|
||||||
|
|
||||||
|
#define __read_mostly __attribute__((__section__(".data..read_mostly")))
|
||||||
|
|
||||||
#endif
|
#endif
|
||||||
|
@@ -167,6 +167,7 @@ SECTIONS
|
|||||||
|
|
||||||
NOSAVE_DATA
|
NOSAVE_DATA
|
||||||
CACHELINE_ALIGNED_DATA(32)
|
CACHELINE_ALIGNED_DATA(32)
|
||||||
|
READ_MOSTLY_DATA(32)
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* The exception fixup table (might need resorting at runtime)
|
* The exception fixup table (might need resorting at runtime)
|
||||||
|
Reference in New Issue
Block a user