[PATCH V3 0/4] Reducing parameters of alloc_pages* family of functions

[PATCH V3 0/4] Reducing parameters of alloc_pages* family of functions

[Vlastimil Babka]

Hey all,

this is a V3 of attempting something that has been discussed when Minchan
proposed to expand the x86 kernel stack [1], namely the reduction of huge
number of parameters that the alloc_pages* family and get_page_from_freelist()
functions have.

The result is this series, ordered in the subjective importance of the patches.

The series is based on mmotm-2014-12-02-15-55 and I use gcc 4.8.3 20140627 on
openSUSE 13.2. Config includess NUMA and COMPACTION. It also applies to
next-20141212.

Changes since v2:
o Per Linus's suggestion, removed order from alloc_context due to poor
  conceptual fit. Interestingly, this improved the stack/code size a bit.
o Attempted to merge migratetype into alloc_flags, but it was ugly and made
  things only worse. So I didn't even attempt to merge order there.

The core is a new struct alloc_context, which looks like this:

struct alloc_context {
        struct zonelist *zonelist;
        nodemask_t *nodemask;
        struct zone *preferred_zone;
        int classzone_idx;
        int migratetype;
        enum zone_type high_zoneidx;
};

All the contents is mostly constant, except that __alloc_pages_slowpath()
changes preferred_zone, classzone_idx and potentially zonelist. But that's not
a problem in case control returns to retry_cpuset: in __alloc_pages_nodemask(),
those will be reset to initial values again (although it's a bit subtle).
On the other hand, gfp_flags and alloc_info mutate so much that it doesn't
make sense to put them into alloc_context. Still, the result is one parameter
instead of up to 7. This is all in Patch 2.

Patch 3 is a step to expand alloc_context usage out of page_alloc.c itself.
The function try_to_compact_pages() can also much benefit from the parameter
reduction, but it means the struct definition has to be moved to a shared
header.

Patch 1 should IMHO be included even if the rest is deemed not useful enough.
It improves maintainability and also has some code/stack reduction. Patch 4
is OTOH a tiny optimization.

First, let's look at the code size savings by bloat-o-meter, as the patches
stack up:

Patch 1 (vs mmotm):

function                                     old     new   delta
get_page_from_freelist                      2554    2490     -64

Patch 2:

function                                     old     new   delta
__alloc_pages_nodemask                       571    2220   +1649
get_page_from_freelist                      2490    2560     +70
__alloc_pages_direct_compact                 332     302     -30
__alloc_pages_slowpath                      1878       -   -1878

Here gcc decided to inline _slowpath, so let's try comparing with Patch 1
plus forced inline of _slowpath:

add/remove: 0/0 grow/shrink: 1/2 up/down: 70/-428 (-358)
function                                     old     new   delta
get_page_from_freelist                      2490    2560     +70
__alloc_pages_direct_compact                 332     284     -48
__alloc_pages_nodemask                      2618    2194    -424

Looks like get_page_from_freelist() did benefit from getting the parameters
separately, but overal it's a win.

Patch 3:

__alloc_pages_direct_compact                 284     256     -28
try_to_compact_pages                         582     566     -16

A tiny overal win.

Patch 4:

function                                     old     new   delta
__alloc_pages_nodemask                      2194    2191      -3
nr_free_zone_pages                           129     115     -14
get_page_from_freelist                      2560    2505     -55
try_to_compact_pages                         566     579     +13

Small overal win. A few more object files should be also affected,
but were not tested.

Now stack sizes per ./scripts/checkstack.pl:

                        mmotm    P1   P2
__alloc_pages_slowpath    176   176    -
get_page_from_freelist:   160   152  160
__alloc_pages_nodemask    104   104  168 
__alloc_pages_direct_c     32    32   16

Patch 1 saves a bit, Patch 2 result muddled by inlining.
Again, let's use Patch 1 + forced inline as baseline for the rest:

                          P1i    P2    P3    P4
__alloc_pages_nodemask    240   168   168   168
get_page_from_freelist:   152   160   160   160
try_to_compact_pages       64    64    56    56
__alloc_pages_direct_c     32    16     -     -

Again, Patch 2 bloats get_page_from_freelist(), but overal is a win. The
rest almost doesn't affect stack usage.

[1] http://marc.info/?l=linux-mm&m=140142462528257&w=2

Vlastimil Babka (4):
  mm: set page->pfmemalloc in prep_new_page()
  mm, page_alloc: reduce number of alloc_pages* functions' parameters
  mm: reduce try_to_compact_pages parameters
  mm: microoptimize zonelist operations

 include/linux/compaction.h |  15 ++-
 include/linux/mm.h         |  14 +++
 include/linux/mmzone.h     |  13 +--
 mm/compaction.c            |  23 ++---
 mm/mmzone.c                |   4 +-
 mm/page_alloc.c            | 250 +++++++++++++++++++--------------------------
 6 files changed, 144 insertions(+), 175 deletions(-)

-- 
2.1.2

--
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>

[PATCH V3 1/4] mm: set page->pfmemalloc in prep_new_page()

[Vlastimil Babka]

The function prep_new_page() sets almost everything in the struct page of the
page being allocated, except page->pfmemalloc. This is not obvious and has at
least once led to a bug where page->pfmemalloc was forgotten to be set
correctly, see commit 8fb74b9fb2b1 ("mm: compaction: partially revert capture
of suitable high-order page").

This patch moves the pfmemalloc setting to prep_new_page(), which means it
needs to gain alloc_flags parameter. The call to prep_new_page is moved from
buffered_rmqueue() to get_page_from_freelist(), which also leads to simpler
code. An obsolete comment for buffered_rmqueue() is replaced.

In addition to better maintainability there is a small reduction of code and
stack usage for get_page_from_freelist(), which inlines the other functions
involved.

Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Zhang Yanfei <zhangyanfei@cn.fujitsu.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Michal Hocko <mhocko@suse.cz>
---
 mm/page_alloc.c | 37 ++++++++++++++++---------------------
 1 file changed, 16 insertions(+), 21 deletions(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 622929f..bfc00c3 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -970,7 +970,8 @@ static inline int check_new_page(struct page *page)
 	return 0;
 }
 
-static int prep_new_page(struct page *page, unsigned int order, gfp_t gfp_flags)
+static int prep_new_page(struct page *page, unsigned int order, gfp_t gfp_flags,
+								int alloc_flags)
 {
 	int i;
 
@@ -994,6 +995,14 @@ static int prep_new_page(struct page *page, unsigned int order, gfp_t gfp_flags)
 
 	set_page_owner(page, order, gfp_flags);
 
+	/*
+	 * page->pfmemalloc is set when ALLOC_NO_WATERMARKS was necessary to
+	 * allocate the page. The expectation is that the caller is taking
+	 * steps that will free more memory. The caller should avoid the page
+	 * being used for !PFMEMALLOC purposes.
+	 */
+	page->pfmemalloc = !!(alloc_flags & ALLOC_NO_WATERMARKS);
+
 	return 0;
 }
 
@@ -1642,9 +1651,7 @@ int split_free_page(struct page *page)
 }
 
 /*
- * Really, prep_compound_page() should be called from __rmqueue_bulk().  But
- * we cheat by calling it from here, in the order > 0 path.  Saves a branch
- * or two.
+ * Allocate a page from the given zone. Use pcplists for order-0 allocations.
  */
 static inline
 struct page *buffered_rmqueue(struct zone *preferred_zone,
@@ -1655,7 +1662,6 @@ struct page *buffered_rmqueue(struct zone *preferred_zone,
 	struct page *page;
 	bool cold = ((gfp_flags & __GFP_COLD) != 0);
 
-again:
 	if (likely(order == 0)) {
 		struct per_cpu_pages *pcp;
 		struct list_head *list;
@@ -1711,8 +1717,6 @@ again:
 	local_irq_restore(flags);
 
 	VM_BUG_ON_PAGE(bad_range(zone, page), page);
-	if (prep_new_page(page, order, gfp_flags))
-		goto again;
 	return page;
 
 failed:
@@ -2177,25 +2181,16 @@ zonelist_scan:
 try_this_zone:
 		page = buffered_rmqueue(preferred_zone, zone, order,
 						gfp_mask, migratetype);
-		if (page)
-			break;
+		if (page) {
+			if (prep_new_page(page, order, gfp_mask, alloc_flags))
+				goto try_this_zone;
+			return page;
+		}
 this_zone_full:
 		if (IS_ENABLED(CONFIG_NUMA) && zlc_active)
 			zlc_mark_zone_full(zonelist, z);
 	}
 
-	if (page) {
-		/*
-		 * page->pfmemalloc is set when ALLOC_NO_WATERMARKS was
-		 * necessary to allocate the page. The expectation is
-		 * that the caller is taking steps that will free more
-		 * memory. The caller should avoid the page being used
-		 * for !PFMEMALLOC purposes.
-		 */
-		page->pfmemalloc = !!(alloc_flags & ALLOC_NO_WATERMARKS);
-		return page;
-	}
-
 	/*
 	 * The first pass makes sure allocations are spread fairly within the
 	 * local node.  However, the local node might have free pages left
-- 
2.1.2

--
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>

[PATCH V3 2/4] mm, page_alloc: reduce number of alloc_pages* functions' parameters

[Vlastimil Babka]

Introduce struct alloc_context to accumulate the numerous parameters passed
between the alloc_pages* family of functions and get_page_from_freelist().
This excludes gfp_flags and alloc_info, which mutate too much along the way,
and allocation order, which is conceptually different.

The result is shorter function signatures, as well as overal code size and
stack usage reductions.

bloat-o-meter:

add/remove: 0/0 grow/shrink: 1/2 up/down: 70/-472 (-402)                                                                                                                                                           
function                                     old     new   delta
get_page_from_freelist                      2490    2560     +70
__alloc_pages_direct_compact                 332     284     -48
__alloc_pages_nodemask                      2618    2194    -424

checkstack.pl:

function                            old    new
__alloc_pages_nodemask              240    168
get_page_from_freelist              152    160
__alloc_pages_direct_compact         32     16

Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Zhang Yanfei <zhangyanfei@cn.fujitsu.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Michal Hocko <mhocko@suse.cz>
---
 mm/page_alloc.c | 226 +++++++++++++++++++++++++-------------------------------
 1 file changed, 102 insertions(+), 124 deletions(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index bfc00c3..963f9c0 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -232,6 +232,19 @@ EXPORT_SYMBOL(nr_node_ids);
 EXPORT_SYMBOL(nr_online_nodes);
 #endif
 
+/*
+ * Structure for holding the mostly immutable allocation parameters passed
+ * between alloc_pages* family of functions.
+ */
+struct alloc_context {
+	struct zonelist *zonelist;
+	nodemask_t *nodemask;
+	struct zone *preferred_zone;
+	int classzone_idx;
+	int migratetype;
+	enum zone_type high_zoneidx;
+};
+
 int page_group_by_mobility_disabled __read_mostly;
 
 void set_pageblock_migratetype(struct page *page, int migratetype)
@@ -2037,10 +2050,10 @@ static void reset_alloc_batches(struct zone *preferred_zone)
  * a page.
  */
 static struct page *
-get_page_from_freelist(gfp_t gfp_mask, nodemask_t *nodemask, unsigned int order,
-		struct zonelist *zonelist, int high_zoneidx, int alloc_flags,
-		struct zone *preferred_zone, int classzone_idx, int migratetype)
+get_page_from_freelist(gfp_t gfp_mask, unsigned int order, int alloc_flags,
+						const struct alloc_context *ac)
 {
+	struct zonelist *zonelist = ac->zonelist;
 	struct zoneref *z;
 	struct page *page = NULL;
 	struct zone *zone;
@@ -2059,8 +2072,8 @@ zonelist_scan:
 	 * Scan zonelist, looking for a zone with enough free.
 	 * See also __cpuset_node_allowed() comment in kernel/cpuset.c.
 	 */
-	for_each_zone_zonelist_nodemask(zone, z, zonelist,
-						high_zoneidx, nodemask) {
+	for_each_zone_zonelist_nodemask(zone, z, zonelist, ac->high_zoneidx,
+								ac->nodemask) {
 		unsigned long mark;
 
 		if (IS_ENABLED(CONFIG_NUMA) && zlc_active &&
@@ -2077,7 +2090,7 @@ zonelist_scan:
 		 * time the page has in memory before being reclaimed.
 		 */
 		if (alloc_flags & ALLOC_FAIR) {
-			if (!zone_local(preferred_zone, zone))
+			if (!zone_local(ac->preferred_zone, zone))
 				break;
 			if (test_bit(ZONE_FAIR_DEPLETED, &zone->flags)) {
 				nr_fair_skipped++;
@@ -2115,7 +2128,7 @@ zonelist_scan:
 
 		mark = zone->watermark[alloc_flags & ALLOC_WMARK_MASK];
 		if (!zone_watermark_ok(zone, order, mark,
-				       classzone_idx, alloc_flags)) {
+				       ac->classzone_idx, alloc_flags)) {
 			int ret;
 
 			/* Checked here to keep the fast path fast */
@@ -2136,7 +2149,7 @@ zonelist_scan:
 			}
 
 			if (zone_reclaim_mode == 0 ||
-			    !zone_allows_reclaim(preferred_zone, zone))
+			    !zone_allows_reclaim(ac->preferred_zone, zone))
 				goto this_zone_full;
 
 			/*
@@ -2158,7 +2171,7 @@ zonelist_scan:
 			default:
 				/* did we reclaim enough */
 				if (zone_watermark_ok(zone, order, mark,
-						classzone_idx, alloc_flags))
+						ac->classzone_idx, alloc_flags))
 					goto try_this_zone;
 
 				/*
@@ -2179,8 +2192,8 @@ zonelist_scan:
 		}
 
 try_this_zone:
-		page = buffered_rmqueue(preferred_zone, zone, order,
-						gfp_mask, migratetype);
+		page = buffered_rmqueue(ac->preferred_zone, zone, order,
+						gfp_mask, ac->migratetype);
 		if (page) {
 			if (prep_new_page(page, order, gfp_mask, alloc_flags))
 				goto try_this_zone;
@@ -2203,7 +2216,7 @@ this_zone_full:
 		alloc_flags &= ~ALLOC_FAIR;
 		if (nr_fair_skipped) {
 			zonelist_rescan = true;
-			reset_alloc_batches(preferred_zone);
+			reset_alloc_batches(ac->preferred_zone);
 		}
 		if (nr_online_nodes > 1)
 			zonelist_rescan = true;
@@ -2325,14 +2338,12 @@ should_alloc_retry(gfp_t gfp_mask, unsigned int order,
 
 static inline struct page *
 __alloc_pages_may_oom(gfp_t gfp_mask, unsigned int order,
-	struct zonelist *zonelist, enum zone_type high_zoneidx,
-	nodemask_t *nodemask, struct zone *preferred_zone,
-	int classzone_idx, int migratetype)
+					const struct alloc_context *ac)
 {
 	struct page *page;
 
 	/* Acquire the per-zone oom lock for each zone */
-	if (!oom_zonelist_trylock(zonelist, gfp_mask)) {
+	if (!oom_zonelist_trylock(ac->zonelist, gfp_mask)) {
 		schedule_timeout_uninterruptible(1);
 		return NULL;
 	}
@@ -2342,10 +2353,8 @@ __alloc_pages_may_oom(gfp_t gfp_mask, unsigned int order,
 	 * here, this is only to catch a parallel oom killing, we must fail if
 	 * we're still under heavy pressure.
 	 */
-	page = get_page_from_freelist(gfp_mask|__GFP_HARDWALL, nodemask,
-		order, zonelist, high_zoneidx,
-		ALLOC_WMARK_HIGH|ALLOC_CPUSET,
-		preferred_zone, classzone_idx, migratetype);
+	page = get_page_from_freelist(gfp_mask | __GFP_HARDWALL, order,
+					ALLOC_WMARK_HIGH|ALLOC_CPUSET, ac);
 	if (page)
 		goto out;
 
@@ -2354,7 +2363,7 @@ __alloc_pages_may_oom(gfp_t gfp_mask, unsigned int order,
 		if (order > PAGE_ALLOC_COSTLY_ORDER)
 			goto out;
 		/* The OOM killer does not needlessly kill tasks for lowmem */
-		if (high_zoneidx < ZONE_NORMAL)
+		if (ac->high_zoneidx < ZONE_NORMAL)
 			goto out;
 		/*
 		 * GFP_THISNODE contains __GFP_NORETRY and we never hit this.
@@ -2367,10 +2376,10 @@ __alloc_pages_may_oom(gfp_t gfp_mask, unsigned int order,
 			goto out;
 	}
 	/* Exhausted what can be done so it's blamo time */
-	out_of_memory(zonelist, gfp_mask, order, nodemask, false);
+	out_of_memory(ac->zonelist, gfp_mask, order, ac->nodemask, false);
 
 out:
-	oom_zonelist_unlock(zonelist, gfp_mask);
+	oom_zonelist_unlock(ac->zonelist, gfp_mask);
 	return page;
 }
 
@@ -2378,10 +2387,9 @@ out:
 /* Try memory compaction for high-order allocations before reclaim */
 static struct page *
 __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order,
-	struct zonelist *zonelist, enum zone_type high_zoneidx,
-	nodemask_t *nodemask, int alloc_flags, struct zone *preferred_zone,
-	int classzone_idx, int migratetype, enum migrate_mode mode,
-	int *contended_compaction, bool *deferred_compaction)
+		int alloc_flags, const struct alloc_context *ac,
+		enum migrate_mode mode, int *contended_compaction,
+		bool *deferred_compaction)
 {
 	unsigned long compact_result;
 	struct page *page;
@@ -2390,10 +2398,10 @@ __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order,
 		return NULL;
 
 	current->flags |= PF_MEMALLOC;
-	compact_result = try_to_compact_pages(zonelist, order, gfp_mask,
-						nodemask, mode,
+	compact_result = try_to_compact_pages(ac->zonelist, order, gfp_mask,
+						ac->nodemask, mode,
 						contended_compaction,
-						alloc_flags, classzone_idx);
+						alloc_flags, ac->classzone_idx);
 	current->flags &= ~PF_MEMALLOC;
 
 	switch (compact_result) {
@@ -2412,10 +2420,8 @@ __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order,
 	 */
 	count_vm_event(COMPACTSTALL);
 
-	page = get_page_from_freelist(gfp_mask, nodemask,
-			order, zonelist, high_zoneidx,
-			alloc_flags & ~ALLOC_NO_WATERMARKS,
-			preferred_zone, classzone_idx, migratetype);
+	page = get_page_from_freelist(gfp_mask, order,
+					alloc_flags & ~ALLOC_NO_WATERMARKS, ac);
 
 	if (page) {
 		struct zone *zone = page_zone(page);
@@ -2439,10 +2445,9 @@ __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order,
 #else
 static inline struct page *
 __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order,
-	struct zonelist *zonelist, enum zone_type high_zoneidx,
-	nodemask_t *nodemask, int alloc_flags, struct zone *preferred_zone,
-	int classzone_idx, int migratetype, enum migrate_mode mode,
-	int *contended_compaction, bool *deferred_compaction)
+		int alloc_flags, const struct alloc_context *ac,
+		enum migrate_mode mode, int *contended_compaction,
+		bool *deferred_compaction)
 {
 	return NULL;
 }
@@ -2450,8 +2455,8 @@ __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order,
 
 /* Perform direct synchronous page reclaim */
 static int
-__perform_reclaim(gfp_t gfp_mask, unsigned int order, struct zonelist *zonelist,
-		  nodemask_t *nodemask)
+__perform_reclaim(gfp_t gfp_mask, unsigned int order,
+					const struct alloc_context *ac)
 {
 	struct reclaim_state reclaim_state;
 	int progress;
@@ -2465,7 +2470,8 @@ __perform_reclaim(gfp_t gfp_mask, unsigned int order, struct zonelist *zonelist,
 	reclaim_state.reclaimed_slab = 0;
 	current->reclaim_state = &reclaim_state;
 
-	progress = try_to_free_pages(zonelist, order, gfp_mask, nodemask);
+	progress = try_to_free_pages(ac->zonelist, order, gfp_mask,
+								ac->nodemask);
 
 	current->reclaim_state = NULL;
 	lockdep_clear_current_reclaim_state();
@@ -2479,28 +2485,23 @@ __perform_reclaim(gfp_t gfp_mask, unsigned int order, struct zonelist *zonelist,
 /* The really slow allocator path where we enter direct reclaim */
 static inline struct page *
 __alloc_pages_direct_reclaim(gfp_t gfp_mask, unsigned int order,
-	struct zonelist *zonelist, enum zone_type high_zoneidx,
-	nodemask_t *nodemask, int alloc_flags, struct zone *preferred_zone,
-	int classzone_idx, int migratetype, unsigned long *did_some_progress)
+		int alloc_flags, const struct alloc_context *ac,
+		unsigned long *did_some_progress)
 {
 	struct page *page = NULL;
 	bool drained = false;
 
-	*did_some_progress = __perform_reclaim(gfp_mask, order, zonelist,
-					       nodemask);
+	*did_some_progress = __perform_reclaim(gfp_mask, order, ac);
 	if (unlikely(!(*did_some_progress)))
 		return NULL;
 
 	/* After successful reclaim, reconsider all zones for allocation */
 	if (IS_ENABLED(CONFIG_NUMA))
-		zlc_clear_zones_full(zonelist);
+		zlc_clear_zones_full(ac->zonelist);
 
 retry:
-	page = get_page_from_freelist(gfp_mask, nodemask, order,
-					zonelist, high_zoneidx,
-					alloc_flags & ~ALLOC_NO_WATERMARKS,
-					preferred_zone, classzone_idx,
-					migratetype);
+	page = get_page_from_freelist(gfp_mask, order,
+					alloc_flags & ~ALLOC_NO_WATERMARKS, ac);
 
 	/*
 	 * If an allocation failed after direct reclaim, it could be because
@@ -2521,36 +2522,30 @@ retry:
  */
 static inline struct page *
 __alloc_pages_high_priority(gfp_t gfp_mask, unsigned int order,
-	struct zonelist *zonelist, enum zone_type high_zoneidx,
-	nodemask_t *nodemask, struct zone *preferred_zone,
-	int classzone_idx, int migratetype)
+				const struct alloc_context *ac)
 {
 	struct page *page;
 
 	do {
-		page = get_page_from_freelist(gfp_mask, nodemask, order,
-			zonelist, high_zoneidx, ALLOC_NO_WATERMARKS,
-			preferred_zone, classzone_idx, migratetype);
+		page = get_page_from_freelist(gfp_mask, order,
+						ALLOC_NO_WATERMARKS, ac);
 
 		if (!page && gfp_mask & __GFP_NOFAIL)
-			wait_iff_congested(preferred_zone, BLK_RW_ASYNC, HZ/50);
+			wait_iff_congested(ac->preferred_zone, BLK_RW_ASYNC,
+									HZ/50);
 	} while (!page && (gfp_mask & __GFP_NOFAIL));
 
 	return page;
 }
 
-static void wake_all_kswapds(unsigned int order,
-			     struct zonelist *zonelist,
-			     enum zone_type high_zoneidx,
-			     struct zone *preferred_zone,
-			     nodemask_t *nodemask)
+static void wake_all_kswapds(unsigned int order, const struct alloc_context *ac)
 {
 	struct zoneref *z;
 	struct zone *zone;
 
-	for_each_zone_zonelist_nodemask(zone, z, zonelist,
-						high_zoneidx, nodemask)
-		wakeup_kswapd(zone, order, zone_idx(preferred_zone));
+	for_each_zone_zonelist_nodemask(zone, z, ac->zonelist,
+						ac->high_zoneidx, ac->nodemask)
+		wakeup_kswapd(zone, order, zone_idx(ac->preferred_zone));
 }
 
 static inline int
@@ -2609,9 +2604,7 @@ bool gfp_pfmemalloc_allowed(gfp_t gfp_mask)
 
 static inline struct page *
 __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
-	struct zonelist *zonelist, enum zone_type high_zoneidx,
-	nodemask_t *nodemask, struct zone *preferred_zone,
-	int classzone_idx, int migratetype)
+						struct alloc_context *ac)
 {
 	const gfp_t wait = gfp_mask & __GFP_WAIT;
 	struct page *page = NULL;
@@ -2647,8 +2640,7 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
 
 restart:
 	if (!(gfp_mask & __GFP_NO_KSWAPD))
-		wake_all_kswapds(order, zonelist, high_zoneidx,
-				preferred_zone, nodemask);
+		wake_all_kswapds(order, ac);
 
 	/*
 	 * OK, we're below the kswapd watermark and have kicked background
@@ -2661,18 +2653,17 @@ restart:
 	 * Find the true preferred zone if the allocation is unconstrained by
 	 * cpusets.
 	 */
-	if (!(alloc_flags & ALLOC_CPUSET) && !nodemask) {
+	if (!(alloc_flags & ALLOC_CPUSET) && !ac->nodemask) {
 		struct zoneref *preferred_zoneref;
-		preferred_zoneref = first_zones_zonelist(zonelist, high_zoneidx,
-				NULL, &preferred_zone);
-		classzone_idx = zonelist_zone_idx(preferred_zoneref);
+		preferred_zoneref = first_zones_zonelist(ac->zonelist,
+				ac->high_zoneidx, NULL, &ac->preferred_zone);
+		ac->classzone_idx = zonelist_zone_idx(preferred_zoneref);
 	}
 
 rebalance:
 	/* This is the last chance, in general, before the goto nopage. */
-	page = get_page_from_freelist(gfp_mask, nodemask, order, zonelist,
-			high_zoneidx, alloc_flags & ~ALLOC_NO_WATERMARKS,
-			preferred_zone, classzone_idx, migratetype);
+	page = get_page_from_freelist(gfp_mask, order,
+				alloc_flags & ~ALLOC_NO_WATERMARKS, ac);
 	if (page)
 		goto got_pg;
 
@@ -2683,11 +2674,10 @@ rebalance:
 		 * the allocation is high priority and these type of
 		 * allocations are system rather than user orientated
 		 */
-		zonelist = node_zonelist(numa_node_id(), gfp_mask);
+		ac->zonelist = node_zonelist(numa_node_id(), gfp_mask);
+
+		page = __alloc_pages_high_priority(gfp_mask, order, ac);
 
-		page = __alloc_pages_high_priority(gfp_mask, order,
-				zonelist, high_zoneidx, nodemask,
-				preferred_zone, classzone_idx, migratetype);
 		if (page) {
 			goto got_pg;
 		}
@@ -2716,11 +2706,9 @@ rebalance:
 	 * Try direct compaction. The first pass is asynchronous. Subsequent
 	 * attempts after direct reclaim are synchronous
 	 */
-	page = __alloc_pages_direct_compact(gfp_mask, order, zonelist,
-					high_zoneidx, nodemask, alloc_flags,
-					preferred_zone,
-					classzone_idx, migratetype,
-					migration_mode, &contended_compaction,
+	page = __alloc_pages_direct_compact(gfp_mask, order, alloc_flags, ac,
+					migration_mode,
+					&contended_compaction,
 					&deferred_compaction);
 	if (page)
 		goto got_pg;
@@ -2766,12 +2754,8 @@ rebalance:
 		migration_mode = MIGRATE_SYNC_LIGHT;
 
 	/* Try direct reclaim and then allocating */
-	page = __alloc_pages_direct_reclaim(gfp_mask, order,
-					zonelist, high_zoneidx,
-					nodemask,
-					alloc_flags, preferred_zone,
-					classzone_idx, migratetype,
-					&did_some_progress);
+	page = __alloc_pages_direct_reclaim(gfp_mask, order, alloc_flags, ac,
+							&did_some_progress);
 	if (page)
 		goto got_pg;
 
@@ -2787,10 +2771,7 @@ rebalance:
 			if ((current->flags & PF_DUMPCORE) &&
 			    !(gfp_mask & __GFP_NOFAIL))
 				goto nopage;
-			page = __alloc_pages_may_oom(gfp_mask, order,
-					zonelist, high_zoneidx,
-					nodemask, preferred_zone,
-					classzone_idx, migratetype);
+			page = __alloc_pages_may_oom(gfp_mask, order, ac);
 			if (page)
 				goto got_pg;
 
@@ -2808,7 +2789,7 @@ rebalance:
 				 * allocations to prevent needlessly killing
 				 * innocent tasks.
 				 */
-				if (high_zoneidx < ZONE_NORMAL)
+				if (ac->high_zoneidx < ZONE_NORMAL)
 					goto nopage;
 			}
 
@@ -2821,7 +2802,7 @@ rebalance:
 	if (should_alloc_retry(gfp_mask, order, did_some_progress,
 						pages_reclaimed)) {
 		/* Wait for some write requests to complete then retry */
-		wait_iff_congested(preferred_zone, BLK_RW_ASYNC, HZ/50);
+		wait_iff_congested(ac->preferred_zone, BLK_RW_ASYNC, HZ/50);
 		goto rebalance;
 	} else {
 		/*
@@ -2829,11 +2810,9 @@ rebalance:
 		 * direct reclaim and reclaim/compaction depends on compaction
 		 * being called after reclaim so call directly if necessary
 		 */
-		page = __alloc_pages_direct_compact(gfp_mask, order, zonelist,
-					high_zoneidx, nodemask, alloc_flags,
-					preferred_zone,
-					classzone_idx, migratetype,
-					migration_mode, &contended_compaction,
+		page = __alloc_pages_direct_compact(gfp_mask, order,
+					alloc_flags, ac, migration_mode,
+					&contended_compaction,
 					&deferred_compaction);
 		if (page)
 			goto got_pg;
@@ -2856,14 +2835,15 @@ struct page *
 __alloc_pages_nodemask(gfp_t gfp_mask, unsigned int order,
 			struct zonelist *zonelist, nodemask_t *nodemask)
 {
-	enum zone_type high_zoneidx = gfp_zone(gfp_mask);
-	struct zone *preferred_zone;
 	struct zoneref *preferred_zoneref;
 	struct page *page = NULL;
-	int migratetype = gfpflags_to_migratetype(gfp_mask);
 	unsigned int cpuset_mems_cookie;
 	int alloc_flags = ALLOC_WMARK_LOW|ALLOC_CPUSET|ALLOC_FAIR;
-	int classzone_idx;
+	struct alloc_context ac = {
+		.high_zoneidx = gfp_zone(gfp_mask),
+		.nodemask = nodemask,
+		.migratetype = gfpflags_to_migratetype(gfp_mask),
+	};
 
 	gfp_mask &= gfp_allowed_mask;
 
@@ -2882,38 +2862,36 @@ __alloc_pages_nodemask(gfp_t gfp_mask, unsigned int order,
 	if (unlikely(!zonelist->_zonerefs->zone))
 		return NULL;
 
-	if (IS_ENABLED(CONFIG_CMA) && migratetype == MIGRATE_MOVABLE)
+	if (IS_ENABLED(CONFIG_CMA) && ac.migratetype == MIGRATE_MOVABLE)
 		alloc_flags |= ALLOC_CMA;
 
 retry_cpuset:
 	cpuset_mems_cookie = read_mems_allowed_begin();
 
+	/* We set it here, as __alloc_pages_slowpath might have changed it */
+	ac.zonelist = zonelist;
 	/* The preferred zone is used for statistics later */
-	preferred_zoneref = first_zones_zonelist(zonelist, high_zoneidx,
-				nodemask ? : &cpuset_current_mems_allowed,
-				&preferred_zone);
-	if (!preferred_zone)
+	preferred_zoneref = first_zones_zonelist(ac.zonelist, ac.high_zoneidx,
+				ac.nodemask ? : &cpuset_current_mems_allowed,
+				&ac.preferred_zone);
+	if (!ac.preferred_zone)
 		goto out;
-	classzone_idx = zonelist_zone_idx(preferred_zoneref);
+	ac.classzone_idx = zonelist_zone_idx(preferred_zoneref);
 
 	/* First allocation attempt */
-	page = get_page_from_freelist(gfp_mask|__GFP_HARDWALL, nodemask, order,
-			zonelist, high_zoneidx, alloc_flags,
-			preferred_zone, classzone_idx, migratetype);
+	page = get_page_from_freelist(gfp_mask|__GFP_HARDWALL, order,
+							alloc_flags, &ac);
 	if (unlikely(!page)) {
 		/*
 		 * Runtime PM, block IO and its error handling path
 		 * can deadlock because I/O on the device might not
 		 * complete.
 		 */
-		gfp_t mask = memalloc_noio_flags(gfp_mask);
-
-		page = __alloc_pages_slowpath(mask, order,
-				zonelist, high_zoneidx, nodemask,
-				preferred_zone, classzone_idx, migratetype);
+		gfp_mask = memalloc_noio_flags(gfp_mask);
+		page = __alloc_pages_slowpath(gfp_mask, order, &ac);
 	}
 
-	trace_mm_page_alloc(page, order, gfp_mask, migratetype);
+	trace_mm_page_alloc(page, order, gfp_mask, ac.migratetype);
 
 out:
 	/*
-- 
2.1.2

--
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>

[PATCH V3 3/4] mm: reduce try_to_compact_pages parameters

[Vlastimil Babka]

Expand the usage of the struct alloc_context introduced in the previous patch
also for calling try_to_compact_pages(), to reduce the number of its
parameters. Since the function is in different compilation module, the
definition of the alloc_context struct is moved to mm.h.

With this change we get simpler code and small savings of code size and stack
usage.

Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Zhang Yanfei <zhangyanfei@cn.fujitsu.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Michal Hocko <mhocko@suse.cz>
---
 include/linux/compaction.h | 15 +++++++--------
 include/linux/mm.h         | 14 ++++++++++++++
 mm/compaction.c            | 23 +++++++++++------------
 mm/page_alloc.c            | 19 ++-----------------
 4 files changed, 34 insertions(+), 37 deletions(-)

diff --git a/include/linux/compaction.h b/include/linux/compaction.h
index 3238ffa..44478a2 100644
--- a/include/linux/compaction.h
+++ b/include/linux/compaction.h
@@ -30,10 +30,9 @@ extern int sysctl_extfrag_handler(struct ctl_table *table, int write,
 			void __user *buffer, size_t *length, loff_t *ppos);
 
 extern int fragmentation_index(struct zone *zone, unsigned int order);
-extern unsigned long try_to_compact_pages(struct zonelist *zonelist,
-			int order, gfp_t gfp_mask, nodemask_t *mask,
-			enum migrate_mode mode, int *contended,
-			int alloc_flags, int classzone_idx);
+extern unsigned long try_to_compact_pages(gfp_t gfp_mask, unsigned int order,
+			int alloc_flags, const struct alloc_context *ac,
+			enum migrate_mode mode, int *contended);
 extern void compact_pgdat(pg_data_t *pgdat, int order);
 extern void reset_isolation_suitable(pg_data_t *pgdat);
 extern unsigned long compaction_suitable(struct zone *zone, int order,
@@ -101,10 +100,10 @@ static inline bool compaction_restarting(struct zone *zone, int order)
 }
 
 #else
-static inline unsigned long try_to_compact_pages(struct zonelist *zonelist,
-			int order, gfp_t gfp_mask, nodemask_t *nodemask,
-			enum migrate_mode mode, int *contended,
-			int alloc_flags, int classzone_idx)
+static inline unsigned long try_to_compact_pages(gfp_t gfp_mask,
+			unsigned int order, int alloc_flags,
+			const struct alloc_context *ac,
+			enum migrate_mode mode, int *contended)
 {
 	return COMPACT_CONTINUE;
 }
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 38cf1d6..b3a9244 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -296,6 +296,20 @@ static inline int get_freepage_migratetype(struct page *page)
 }
 
 /*
+ * Structure for holding the mostly immutable allocation parameters passed
+ * between functions involved in allocations, including the alloc_pages*
+ * family of functions.
+ */
+struct alloc_context {
+	struct zonelist *zonelist;
+	nodemask_t *nodemask;
+	struct zone *preferred_zone;
+	int classzone_idx;
+	int migratetype;
+	enum zone_type high_zoneidx;
+};
+
+/*
  * FIXME: take this include out, include page-flags.h in
  * files which need it (119 of them)
  */
diff --git a/mm/compaction.c b/mm/compaction.c
index 546e571..9c7e690 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -1335,22 +1335,20 @@ int sysctl_extfrag_threshold = 500;
 
 /**
  * try_to_compact_pages - Direct compact to satisfy a high-order allocation
- * @zonelist: The zonelist used for the current allocation
- * @order: The order of the current allocation
  * @gfp_mask: The GFP mask of the current allocation
- * @nodemask: The allowed nodes to allocate from
+ * @order: The order of the current allocation
+ * @alloc_flags: The allocation flags of the current allocation
+ * @ac: The context of current allocation
  * @mode: The migration mode for async, sync light, or sync migration
  * @contended: Return value that determines if compaction was aborted due to
  *	       need_resched() or lock contention
  *
  * This is the main entry point for direct page compaction.
  */
-unsigned long try_to_compact_pages(struct zonelist *zonelist,
-			int order, gfp_t gfp_mask, nodemask_t *nodemask,
-			enum migrate_mode mode, int *contended,
-			int alloc_flags, int classzone_idx)
+unsigned long try_to_compact_pages(gfp_t gfp_mask, unsigned int order,
+			int alloc_flags, const struct alloc_context *ac,
+			enum migrate_mode mode, int *contended)
 {
-	enum zone_type high_zoneidx = gfp_zone(gfp_mask);
 	int may_enter_fs = gfp_mask & __GFP_FS;
 	int may_perform_io = gfp_mask & __GFP_IO;
 	struct zoneref *z;
@@ -1365,8 +1363,8 @@ unsigned long try_to_compact_pages(struct zonelist *zonelist,
 		return COMPACT_SKIPPED;
 
 	/* Compact each zone in the list */
-	for_each_zone_zonelist_nodemask(zone, z, zonelist, high_zoneidx,
-								nodemask) {
+	for_each_zone_zonelist_nodemask(zone, z, ac->zonelist, ac->high_zoneidx,
+								ac->nodemask) {
 		int status;
 		int zone_contended;
 
@@ -1374,7 +1372,8 @@ unsigned long try_to_compact_pages(struct zonelist *zonelist,
 			continue;
 
 		status = compact_zone_order(zone, order, gfp_mask, mode,
-				&zone_contended, alloc_flags, classzone_idx);
+				&zone_contended, alloc_flags,
+				ac->classzone_idx);
 		rc = max(status, rc);
 		/*
 		 * It takes at least one zone that wasn't lock contended
@@ -1384,7 +1383,7 @@ unsigned long try_to_compact_pages(struct zonelist *zonelist,
 
 		/* If a normal allocation would succeed, stop compacting */
 		if (zone_watermark_ok(zone, order, low_wmark_pages(zone),
-					classzone_idx, alloc_flags)) {
+					ac->classzone_idx, alloc_flags)) {
 			/*
 			 * We think the allocation will succeed in this zone,
 			 * but it is not certain, hence the false. The caller
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 963f9c0..9416ed2 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -232,19 +232,6 @@ EXPORT_SYMBOL(nr_node_ids);
 EXPORT_SYMBOL(nr_online_nodes);
 #endif
 
-/*
- * Structure for holding the mostly immutable allocation parameters passed
- * between alloc_pages* family of functions.
- */
-struct alloc_context {
-	struct zonelist *zonelist;
-	nodemask_t *nodemask;
-	struct zone *preferred_zone;
-	int classzone_idx;
-	int migratetype;
-	enum zone_type high_zoneidx;
-};
-
 int page_group_by_mobility_disabled __read_mostly;
 
 void set_pageblock_migratetype(struct page *page, int migratetype)
@@ -2398,10 +2385,8 @@ __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order,
 		return NULL;
 
 	current->flags |= PF_MEMALLOC;
-	compact_result = try_to_compact_pages(ac->zonelist, order, gfp_mask,
-						ac->nodemask, mode,
-						contended_compaction,
-						alloc_flags, ac->classzone_idx);
+	compact_result = try_to_compact_pages(gfp_mask, order, alloc_flags, ac,
+						mode, contended_compaction);
 	current->flags &= ~PF_MEMALLOC;
 
 	switch (compact_result) {
-- 
2.1.2

--
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>

[PATCH V3 4/4] mm: microoptimize zonelist operations

[Vlastimil Babka]

The function next_zones_zonelist() returns zoneref pointer, as well as zone
pointer via extra parameter. Since the latter can be trivially obtained by
dereferencing the former, the overhead of the extra parameter is unjustified.

This patch thus removes the zone parameter from next_zones_zonelist(). Both
callers happen to be in the same header file, so it's simple to add the
zoneref dereference inline. We save some bytes of code size.

Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Zhang Yanfei <zhangyanfei@cn.fujitsu.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Michal Hocko <mhocko@suse.cz>
---
 include/linux/mmzone.h | 13 +++++++------
 mm/mmzone.c            |  4 +---
 2 files changed, 8 insertions(+), 9 deletions(-)

diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 2f0856d..a2884ef 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -970,7 +970,6 @@ static inline int zonelist_node_idx(struct zoneref *zoneref)
  * @z - The cursor used as a starting point for the search
  * @highest_zoneidx - The zone index of the highest zone to return
  * @nodes - An optional nodemask to filter the zonelist with
- * @zone - The first suitable zone found is returned via this parameter
  *
  * This function returns the next zone at or below a given zone index that is
  * within the allowed nodemask using a cursor as the starting point for the
@@ -980,8 +979,7 @@ static inline int zonelist_node_idx(struct zoneref *zoneref)
  */
 struct zoneref *next_zones_zonelist(struct zoneref *z,
 					enum zone_type highest_zoneidx,
-					nodemask_t *nodes,
-					struct zone **zone);
+					nodemask_t *nodes);
 
 /**
  * first_zones_zonelist - Returns the first zone at or below highest_zoneidx within the allowed nodemask in a zonelist
@@ -1000,8 +998,10 @@ static inline struct zoneref *first_zones_zonelist(struct zonelist *zonelist,
 					nodemask_t *nodes,
 					struct zone **zone)
 {
-	return next_zones_zonelist(zonelist->_zonerefs, highest_zoneidx, nodes,
-								zone);
+	struct zoneref *z = next_zones_zonelist(zonelist->_zonerefs,
+							highest_zoneidx, nodes);
+	*zone = zonelist_zone(z);
+	return z;
 }
 
 /**
@@ -1018,7 +1018,8 @@ static inline struct zoneref *first_zones_zonelist(struct zonelist *zonelist,
 #define for_each_zone_zonelist_nodemask(zone, z, zlist, highidx, nodemask) \
 	for (z = first_zones_zonelist(zlist, highidx, nodemask, &zone);	\
 		zone;							\
-		z = next_zones_zonelist(++z, highidx, nodemask, &zone))	\
+		z = next_zones_zonelist(++z, highidx, nodemask),	\
+			zone = zonelist_zone(z))			\
 
 /**
  * for_each_zone_zonelist - helper macro to iterate over valid zones in a zonelist at or below a given zone index
diff --git a/mm/mmzone.c b/mm/mmzone.c
index bf34fb8..7d87ebb 100644
--- a/mm/mmzone.c
+++ b/mm/mmzone.c
@@ -54,8 +54,7 @@ static inline int zref_in_nodemask(struct zoneref *zref, nodemask_t *nodes)
 /* Returns the next zone at or below highest_zoneidx in a zonelist */
 struct zoneref *next_zones_zonelist(struct zoneref *z,
 					enum zone_type highest_zoneidx,
-					nodemask_t *nodes,
-					struct zone **zone)
+					nodemask_t *nodes)
 {
 	/*
 	 * Find the next suitable zone to use for the allocation.
@@ -69,7 +68,6 @@ struct zoneref *next_zones_zonelist(struct zoneref *z,
 				(z->zone && !zref_in_nodemask(z, nodes)))
 			z++;
 
-	*zone = zonelist_zone(z);
 	return z;
 }
 
-- 
2.1.2

--
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>

Re: [PATCH V3 0/4] Reducing parameters of alloc_pages* family of functions

[Andrew Morton]

On Fri, 12 Dec 2014 17:13:21 +0100 Vlastimil Babka <vbabka@suse.cz> wrote:

> Vlastimil Babka (4):
>   mm: set page->pfmemalloc in prep_new_page()
>   mm, page_alloc: reduce number of alloc_pages* functions' parameters
>   mm: reduce try_to_compact_pages parameters
>   mm: microoptimize zonelist operations

That all looks pretty straightforward.  It would be nice to have a
summary of the code-size and stack-usage changes for the whole
patchset.

Can we move `struct alloc_context' into mm/internal.h?

I pity the poor schmuck who has to maintain this patchset for 2 months.
[2/4] already throws a large pile of rejects against page_alloc.c so
can you please refresh/retest/resend?


--
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>

Re: [PATCH V3 0/4] Reducing parameters of alloc_pages* family of functions

[Vlastimil Babka]

On 12/18/2014 10:26 PM, Andrew Morton wrote:
> On Fri, 12 Dec 2014 17:13:21 +0100 Vlastimil Babka <vbabka@suse.cz> wrote:
> 
>> Vlastimil Babka (4):
>>   mm: set page->pfmemalloc in prep_new_page()
>>   mm, page_alloc: reduce number of alloc_pages* functions' parameters
>>   mm: reduce try_to_compact_pages parameters
>>   mm: microoptimize zonelist operations
> 
> That all looks pretty straightforward.  It would be nice to have a
> summary of the code-size and stack-usage changes for the whole
> patchset.

OK

> Can we move `struct alloc_context' into mm/internal.h?

Only if we moved also try_to_compact_pages() declaration from
include/linux/compaction.h to mm/internal.h. I guess it's not a bad idea, as
it's a MM-only function and mm/internal.h already contains compaction stuff.

> I pity the poor schmuck who has to maintain this patchset for 2 months.
> [2/4] already throws a large pile of rejects against page_alloc.c so
> can you please refresh/retest/resend?

Right :)

Vlastimil


--
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>

Re: [PATCH V3 0/4] Reducing parameters of alloc_pages* family of functions

[Vlastimil Babka]

On 01/05/2015 03:22 PM, Vlastimil Babka wrote:
> On 12/18/2014 10:26 PM, Andrew Morton wrote:
>> On Fri, 12 Dec 2014 17:13:21 +0100 Vlastimil Babka <vbabka@suse.cz> wrote:
>> 
>>> Vlastimil Babka (4):
>>>   mm: set page->pfmemalloc in prep_new_page()
>>>   mm, page_alloc: reduce number of alloc_pages* functions' parameters
>>>   mm: reduce try_to_compact_pages parameters
>>>   mm: microoptimize zonelist operations
>> 
>> That all looks pretty straightforward.  It would be nice to have a
>> summary of the code-size and stack-usage changes for the whole
>> patchset.
> 
> OK
> 
>> Can we move `struct alloc_context' into mm/internal.h?
> 
> Only if we moved also try_to_compact_pages() declaration from
> include/linux/compaction.h to mm/internal.h. I guess it's not a bad idea, as
> it's a MM-only function and mm/internal.h already contains compaction stuff.

Hm, nope. The !CONFIG_COMPACTION variant of try_to_compact_pages() is static
inline that returns COMPACT_CONTINUE, which is defined in compaction.h.
Another solution is to add a "forward" declaration (not actually followed later
by a full definition) of struct alloc_context into compaction.h. Seems to work
here, but I'm not sure if such thing is allowed?

>> I pity the poor schmuck who has to maintain this patchset for 2 months.
>> [2/4] already throws a large pile of rejects against page_alloc.c so
>> can you please refresh/retest/resend?
> 
> Right :)
> 
> Vlastimil
> 
> 
> --
> 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>
> 

--
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>

Re: [PATCH V3 0/4] Reducing parameters of alloc_pages* family of functions

[Linus Torvalds]

On Mon, Jan 5, 2015 at 8:01 AM, Vlastimil Babka <vbabka@suse.cz> wrote:
>
> Hm, nope. The !CONFIG_COMPACTION variant of try_to_compact_pages() is static
> inline that returns COMPACT_CONTINUE, which is defined in compaction.h.
> Another solution is to add a "forward" declaration (not actually followed later
> by a full definition) of struct alloc_context into compaction.h. Seems to work
> here, but I'm not sure if such thing is allowed?

We do forward struct declarations quite often (well, _relatively_
often) in order to avoid nasty circular header includes, and sometimes
just to avoid unnecessarily many header includes.

See for example

    git grep '\<struct [a-zA-Z_0-9]*;'

it's not exactly rare.

So it's fine.

                        Linus

--
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>