* [PATCH] compaction: fix build error in CMA && !COMPACTION
@ 2012-12-20 5:25 Minchan Kim
2012-12-20 10:47 ` Mel Gorman
0 siblings, 1 reply; 3+ messages in thread
From: Minchan Kim @ 2012-12-20 5:25 UTC (permalink / raw)
To: Andrew Morton
Cc: linux-mm, linux-kernel, Minchan Kim, Mel Gorman, Marek Szyprowski
isolate_freepages_block and isolate_migratepages_range is used for CMA
as well as compaction so it breaks build for CONFIG_CMA &&
!CONFIG_COMPACTION.
This patch fixes it.
Cc: Mel Gorman <mgorman@suse.de>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Minchan Kim <minchan@kernel.org>
---
mm/compaction.c | 26 ++++++++++++++++++++------
1 file changed, 20 insertions(+), 6 deletions(-)
diff --git a/mm/compaction.c b/mm/compaction.c
index 5ad7f4f..70f4443 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -17,6 +17,21 @@
#include <linux/balloon_compaction.h>
#include "internal.h"
+#ifdef CONFIG_COMPACTION
+static inline void count_compact_event(enum vm_event_item item)
+{
+ count_vm_event(item);
+}
+
+static inline void count_compact_events(enum vm_event_item item, long delta)
+{
+ count_vm_events(item, delta);
+}
+#else
+#define count_compact_event(item)
+#define count_compact_events(item, delta)
+#endif
+
#if defined CONFIG_COMPACTION || defined CONFIG_CMA
#define CREATE_TRACE_POINTS
@@ -303,10 +318,9 @@ static unsigned long isolate_freepages_block(struct compact_control *cc,
if (blockpfn == end_pfn)
update_pageblock_skip(cc, valid_page, total_isolated, false);
- count_vm_events(COMPACTFREE_SCANNED, nr_scanned);
+ count_compact_events(COMPACTFREE_SCANNED, nr_scanned);
if (total_isolated)
- count_vm_events(COMPACTISOLATED, total_isolated);
-
+ count_compact_events(COMPACTISOLATED, total_isolated);
return total_isolated;
}
@@ -613,9 +627,9 @@ next_pageblock:
trace_mm_compaction_isolate_migratepages(nr_scanned, nr_isolated);
- count_vm_events(COMPACTMIGRATE_SCANNED, nr_scanned);
+ count_compact_events(COMPACTMIGRATE_SCANNED, nr_scanned);
if (nr_isolated)
- count_vm_events(COMPACTISOLATED, nr_isolated);
+ count_compact_events(COMPACTISOLATED, nr_isolated);
return low_pfn;
}
@@ -1110,7 +1124,7 @@ unsigned long try_to_compact_pages(struct zonelist *zonelist,
if (!order || !may_enter_fs || !may_perform_io)
return rc;
- count_vm_event(COMPACTSTALL);
+ count_compact_event(COMPACTSTALL);
#ifdef CONFIG_CMA
if (allocflags_to_migratetype(gfp_mask) == MIGRATE_MOVABLE)
--
1.7.9.5
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH] compaction: fix build error in CMA && !COMPACTION
2012-12-20 5:25 [PATCH] compaction: fix build error in CMA && !COMPACTION Minchan Kim
@ 2012-12-20 10:47 ` Mel Gorman
2012-12-21 0:49 ` Minchan Kim
0 siblings, 1 reply; 3+ messages in thread
From: Mel Gorman @ 2012-12-20 10:47 UTC (permalink / raw)
To: Minchan Kim; +Cc: Andrew Morton, linux-mm, linux-kernel, Marek Szyprowski
On Thu, Dec 20, 2012 at 02:25:30PM +0900, Minchan Kim wrote:
> isolate_freepages_block and isolate_migratepages_range is used for CMA
> as well as compaction so it breaks build for CONFIG_CMA &&
> !CONFIG_COMPACTION.
>
> This patch fixes it.
>
> Cc: Mel Gorman <mgorman@suse.de>
> Cc: Marek Szyprowski <m.szyprowski@samsung.com>
> Signed-off-by: Minchan Kim <minchan@kernel.org>
> ---
> mm/compaction.c | 26 ++++++++++++++++++++------
> 1 file changed, 20 insertions(+), 6 deletions(-)
>
> diff --git a/mm/compaction.c b/mm/compaction.c
> index 5ad7f4f..70f4443 100644
> --- a/mm/compaction.c
> +++ b/mm/compaction.c
> @@ -17,6 +17,21 @@
> #include <linux/balloon_compaction.h>
> #include "internal.h"
>
> +#ifdef CONFIG_COMPACTION
> +static inline void count_compact_event(enum vm_event_item item)
> +{
> + count_vm_event(item);
> +}
> +
> +static inline void count_compact_events(enum vm_event_item item, long delta)
> +{
> + count_vm_events(item, delta);
> +}
> +#else
> +#define count_compact_event(item)
> +#define count_compact_events(item, delta)
> +#endif
> +
That should be
do {} while (0)
otherwise a block like this
if (foo)
count_compact_event(COMPACTFREE_SCANNED)
bar;
will get parsed as
if (foo)
bar;
which is wrong.
Now that I look at the do {} while (0) thing it is also strictly speaking
wrong for count_vm_numa_events() too because it would do the wrong thing for
count_compact_events(COMPACTFREE_SCANNED, foo++);
There happens to be no examples where we depend on such side-effects but
I've taken a TODO item to fix it up in the New Year.
--
Mel Gorman
SUSE Labs
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] compaction: fix build error in CMA && !COMPACTION
2012-12-20 10:47 ` Mel Gorman
@ 2012-12-21 0:49 ` Minchan Kim
0 siblings, 0 replies; 3+ messages in thread
From: Minchan Kim @ 2012-12-21 0:49 UTC (permalink / raw)
To: Mel Gorman; +Cc: Andrew Morton, linux-mm, linux-kernel, Marek Szyprowski
On Thu, Dec 20, 2012 at 10:47:07AM +0000, Mel Gorman wrote:
> On Thu, Dec 20, 2012 at 02:25:30PM +0900, Minchan Kim wrote:
> > isolate_freepages_block and isolate_migratepages_range is used for CMA
> > as well as compaction so it breaks build for CONFIG_CMA &&
> > !CONFIG_COMPACTION.
> >
> > This patch fixes it.
> >
> > Cc: Mel Gorman <mgorman@suse.de>
> > Cc: Marek Szyprowski <m.szyprowski@samsung.com>
> > Signed-off-by: Minchan Kim <minchan@kernel.org>
> > ---
> > mm/compaction.c | 26 ++++++++++++++++++++------
> > 1 file changed, 20 insertions(+), 6 deletions(-)
> >
> > diff --git a/mm/compaction.c b/mm/compaction.c
> > index 5ad7f4f..70f4443 100644
> > --- a/mm/compaction.c
> > +++ b/mm/compaction.c
> > @@ -17,6 +17,21 @@
> > #include <linux/balloon_compaction.h>
> > #include "internal.h"
> >
> > +#ifdef CONFIG_COMPACTION
> > +static inline void count_compact_event(enum vm_event_item item)
> > +{
> > + count_vm_event(item);
> > +}
> > +
> > +static inline void count_compact_events(enum vm_event_item item, long delta)
> > +{
> > + count_vm_events(item, delta);
> > +}
> > +#else
> > +#define count_compact_event(item)
> > +#define count_compact_events(item, delta)
> > +#endif
> > +
>
> That should be
>
> do {} while (0)
>
> otherwise a block like this
>
> if (foo)
> count_compact_event(COMPACTFREE_SCANNED)
> bar;
>
> will get parsed as
>
> if (foo)
> bar;
>
> which is wrong.
Indeed.
>
> Now that I look at the do {} while (0) thing it is also strictly speaking
> wrong for count_vm_numa_events() too because it would do the wrong thing for
If I noticed it, I should named it as count_vm_compact_events for the consistency.
>
> count_compact_events(COMPACTFREE_SCANNED, foo++);
>
> There happens to be no examples where we depend on such side-effects but
Agreed.
> I've taken a TODO item to fix it up in the New Year.
Andrew picked up the patch with your fix. Please correct it for New Year's
present. :)
Thanks, Mel.
>
> --
> Mel Gorman
> SUSE Labs
>
> --
> To unsubscribe, send a message with 'unsubscribe linux-mm' in
> the body to majordomo@kvack.org. For more info on Linux MM,
> see: http://www.linux-mm.org/ .
> Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
--
Kind regards,
Minchan Kim
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2012-12-21 0:49 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-12-20 5:25 [PATCH] compaction: fix build error in CMA && !COMPACTION Minchan Kim
2012-12-20 10:47 ` Mel Gorman
2012-12-21 0:49 ` Minchan Kim
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).