-tip: block, mm: Fix build error in mm/memcontrol.c

-tip: block, mm: Fix build error in mm/memcontrol.c

[Ingo Molnar]

Today's upstream tree build (x86, 64-bit, allyesconfig) failed with 
this build error:

    mm/memcontrol.c: In function ‘mem_cgroup_force_empty’:
    mm/memcontrol.c:1976: error: ‘BLK_RW_ASYNC’ undeclared (first use in this function)
    mm/memcontrol.c:1976: error: (Each undeclared identifier is reported only once
    mm/memcontrol.c:1976: error: for each function it appears in.)

Due to commit:

  8aa7e84: Fix congestion_wait() sync/async vs read/write confusion

Missing to include a required header file for the BLK_RW_ASYNC.

Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 mm/memcontrol.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index e2fa20d..708103d 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -25,6 +25,7 @@
 #include <linux/smp.h>
 #include <linux/page-flags.h>
 #include <linux/backing-dev.h>
+#include <linux/blkdev.h>
 #include <linux/bit_spinlock.h>
 #include <linux/rcupdate.h>
 #include <linux/limits.h>

Re: -tip: block, mm: Fix build error in mm/memcontrol.c

[Linus Torvalds]

On Sat, 11 Jul 2009, Ingo Molnar wrote:
> 
> Today's upstream tree build (x86, 64-bit, allyesconfig) failed with 
> this build error:

I merged a better fix from Trond, which should make all of this 
unnecessary (just move BLK_RW_[A]SYNC to the header file that people need 
anyway).

		Linus

Re: -tip: block, mm: Fix build error in mm/memcontrol.c

[Ingo Molnar]

* Linus Torvalds <torvalds@linux-foundation.org> wrote:

> On Sat, 11 Jul 2009, Ingo Molnar wrote:
> > 
> > Today's upstream tree build (x86, 64-bit, allyesconfig) failed 
> > with this build error:
> 
> I merged a better fix from Trond, which should make all of this 
> unnecessary (just move BLK_RW_[A]SYNC to the header file that 
> people need anyway).

ok, that's better. I think if in addition to that we also did what i 
suggested:

> Fix it the simplest way for now, by moving the definitions early 
> in blkdev.h (this cannot break anything), but the real fix would 
> be to split up blkdev.h into blkdev-types.h and blkdev-api.h and 
> make only the API definitions/declarations dependent on 
> CONFIG_BLOCK.

would be one step cleaner.

We could then include blkdev-types.h in backing-dev.h so that it 
picks up the constants. Right now having BLK_RW_[A]SYNC in 
backing-dev.h looks a bit weird. No strong opinion though.

	Ingo