On Mon, Dec 12, 2011 at 03:41:04PM +0100, Michal Nazarewicz wrote:

On Mon, 12 Dec 2011 15:29:07 +0100, Mel Gorman [off-list ref] wrote:

quoted

On Fri, Nov 18, 2011 at 05:43:11PM +0100, Marek Szyprowski wrote:

quoted

From: Michal Nazarewicz <redacted>

This commit exports some of the functions from compaction.c file
outside of it adding their declaration into internal.h header
file so that other mm related code can use them.

This forced compaction.c to always be compiled (as opposed to being
compiled only if CONFIG_COMPACTION is defined) but as to avoid
introducing code that user did not ask for, part of the compaction.c
is now wrapped in on #ifdef.

Signed-off-by: Michal Nazarewicz <redacted>
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
---
mm/Makefile     |    3 +-
mm/compaction.c |  112 +++++++++++++++++++++++--------------------------------
mm/internal.h   |   35 +++++++++++++++++
3 files changed, 83 insertions(+), 67 deletions(-)

diff --git a/mm/Makefile b/mm/Makefile
index 50ec00e..24ed801 100644
--- a/mm/Makefile
+++ b/mm/Makefile

@@ -13,7 +13,7 @@ obj-y			:= filemap.o mempool.o oom_kill.o fadvise.o \

			   readahead.o swap.o truncate.o vmscan.o shmem.o \
			   prio_tree.o util.o mmzone.o vmstat.o backing-dev.o \
			   page_isolation.o mm_init.o mmu_context.o percpu.o \
-			   $(mmu-y)
+			   $(mmu-y) compaction.o

That should be

compaction.o $(mmu-y)

for consistency.

Overall, this patch implies that CMA is always compiled in.

Not really.  But yes, it produces some bloat when neither CMA nor
compaction are compiled.  I assume that linker will be able to deal
with that (since the functions are not EXPORT_SYMBOL'ed).

The bloat exists either way. I don't believe the linker strips it out so
overall it would make more sense to depend on compaction to keep the
vmstat counters for debugging reasons if nothing else. It's not
something I feel very strongly about though.

-- 
Mel Gorman
SUSE Labs