Announcement: Enhanced NUMA scheduling with adaptive affinity

Hi,

This series implements an improved version of NUMA scheduling, based on the review and testing feedback we got.

Like the previous version, this code is driven by working set probing faults (so much of the VM machinery remains) - but the subsequent utilization of those faults and the scheduler policy has changed substantially.

The scheduler's affinity logic has been generalized, and this allowed us to eliminate the 'home node' concept that was needlessly restrictive.

The biggest conceptual addition, beyond the elimination of the home node, is that the scheduler is now able to recognize 'private' versus 'shared' pages, by carefully analyzing the pattern of how CPUs touch the working set pages. The scheduler automatically recognizes tasks that share memory with each other (and make dominant use of that memory) - versus tasks that allocate and use their working set privately.

This new scheduler code is then able to group tasks that are "memory related" via their memory access patterns together: in the NUMA context moving them on the same node if possible, and spreading them amongst nodes if they use private memory.

Note that this adaptive NUMA affinity mechanism integrated into the scheduler is essentially free of heuristics - only the access patterns determine which tasks are related and grouped. As a result this adaptive affinity code is able to move both threads and processes close(r) to each other if they are related - and let them spread if they are not. If a workload changes its characteristics dynamically then its scheduling will adapt dynamically as well.

You can find the finer details in the individual patches. The series is based on commit 02743c9c03f1 you can find in linux-next. Reviews and testing feedback are welcome! (We'll also review some of the other feedback we got in the last 2 weeks that we might not have reacted to yet, please be patient.)

Next we plan to pick up bits from Mel's recent series like his page migration patch.

Thanks,

Peter, Ingo

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

Avoid a few #ifdef's later on.

---
 kernel/sched/sched.h |    6 ++++++
 1 file changed, 6 insertions(+)

Index: linux/kernel/sched/sched.h
===================================================================
@@ -663,6 +663,12 @@ extern struct static_key sched_feat_keys
 #define sched_feat(x) (sysctl_sched_features & (1UL << __SCHED_FEAT_##x))
 #endif /* SCHED_DEBUG && HAVE_JUMP_LABEL */
 
+#ifdef CONFIG_SCHED_NUMA
+#define sched_feat_numa(x) sched_feat(x)
+#else
+#define sched_feat_numa(x) (0)
+#endif
+
 static inline u64 global_rt_period(void)
 {
 	return (u64)sysctl_sched_rt_period * NSEC_PER_USEC;

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

Add THP migration for the NUMA working set scanning fault case.

It uses the page lock to serialize. No migration pte dance is necessary because the pte is already unmapped when we decide to migrate.

---
 mm/huge_memory.c |  133 ++++++++++++++++++++++++++++++++++++++++++-------------
 mm/migrate.c     |    2 
 2 files changed, 104 insertions(+), 31 deletions(-)

Index: linux/mm/huge_memory.c
===================================================================
@@ -743,12 +743,13 @@ void do_huge_pmd_numa_page(struct mm_str
 			   unsigned int flags, pmd_t entry)
 {
 	unsigned long haddr = address & HPAGE_PMD_MASK;
+	struct page *new_page = NULL;
 	struct page *page = NULL;
-	int node;
+	int node, lru;
 
 	spin_lock(&mm->page_table_lock);
 	if (unlikely(!pmd_same(*pmd, entry)))
-		goto out_unlock;
+		goto unlock;
 
 	if (unlikely(pmd_trans_splitting(entry))) {
 		spin_unlock(&mm->page_table_lock);

@@ -756,45 +757,117 @@ void do_huge_pmd_numa_page(struct mm_str
 		return;
 	}
 
-#ifdef CONFIG_NUMA
 	page = pmd_page(entry);
-	VM_BUG_ON(!PageCompound(page) || !PageHead(page));
+	if (page) {
+		VM_BUG_ON(!PageCompound(page) || !PageHead(page));
 
-	get_page(page);
+		get_page(page);
+		node = mpol_misplaced(page, vma, haddr);
+		if (node != -1)
+			goto migrate;
+	}
+
+fixup:
+	/* change back to regular protection */
+	entry = pmd_modify(entry, vma->vm_page_prot);
+	set_pmd_at(mm, haddr, pmd, entry);
+	update_mmu_cache_pmd(vma, address, entry);
+
+unlock:
 	spin_unlock(&mm->page_table_lock);
+	if (page)
+		put_page(page);
 
-	/*
-	 * XXX should we serialize against split_huge_page ?
-	 */
-
-	node = mpol_misplaced(page, vma, haddr);
-	if (node == -1)
-		goto do_fixup;
-
-	/*
-	 * Due to lacking code to migrate thp pages, we'll split
-	 * (which preserves the special PROT_NONE) and re-take the
-	 * fault on the normal pages.
-	 */
-	split_huge_page(page);
-	put_page(page);
 	return;
 
-do_fixup:
+migrate:
+	spin_unlock(&mm->page_table_lock);
+
+	lock_page(page);
 	spin_lock(&mm->page_table_lock);
-	if (unlikely(!pmd_same(*pmd, entry)))
-		goto out_unlock;
-#endif
+	if (unlikely(!pmd_same(*pmd, entry))) {
+		spin_unlock(&mm->page_table_lock);
+		unlock_page(page);
+		put_page(page);
+		return;
+	}
+	spin_unlock(&mm->page_table_lock);
 
-	/* change back to regular protection */
-	entry = pmd_modify(entry, vma->vm_page_prot);
-	if (pmdp_set_access_flags(vma, haddr, pmd, entry, 1))
-		update_mmu_cache_pmd(vma, address, entry);
+	new_page = alloc_pages_node(node,
+	    (GFP_TRANSHUGE | GFP_THISNODE) & ~__GFP_WAIT,
+	    HPAGE_PMD_ORDER);
+
+	if (!new_page)
+		goto alloc_fail;
+
+	lru = PageLRU(page);
+
+	if (lru && isolate_lru_page(page)) /* does an implicit get_page() */
+		goto alloc_fail;
+
+	if (!trylock_page(new_page))
+		BUG();
+
+	/* anon mapping, we can simply copy page->mapping to the new page: */
+	new_page->mapping = page->mapping;
+	new_page->index = page->index;
 
-out_unlock:
+	migrate_page_copy(new_page, page);
+
+	WARN_ON(PageLRU(new_page));
+
+	spin_lock(&mm->page_table_lock);
+	if (unlikely(!pmd_same(*pmd, entry))) {
+		spin_unlock(&mm->page_table_lock);
+		if (lru)
+			putback_lru_page(page);
+
+		unlock_page(new_page);
+		ClearPageActive(new_page);	/* Set by migrate_page_copy() */
+		new_page->mapping = NULL;
+		put_page(new_page);		/* Free it */
+
+		unlock_page(page);
+		put_page(page);			/* Drop the local reference */
+
+		return;
+	}
+
+	entry = mk_pmd(new_page, vma->vm_page_prot);
+	entry = maybe_pmd_mkwrite(pmd_mkdirty(entry), vma);
+	entry = pmd_mkhuge(entry);
+
+	page_add_new_anon_rmap(new_page, vma, haddr);
+
+	set_pmd_at(mm, haddr, pmd, entry);
+	update_mmu_cache_pmd(vma, address, entry);
+	page_remove_rmap(page);
 	spin_unlock(&mm->page_table_lock);
-	if (page)
+
+	put_page(page);			/* Drop the rmap reference */
+
+	if (lru)
+		put_page(page);		/* drop the LRU isolation reference */
+
+	unlock_page(new_page);
+	unlock_page(page);
+	put_page(page);			/* Drop the local reference */
+
+	return;
+
+alloc_fail:
+	if (new_page)
+		put_page(new_page);
+
+	unlock_page(page);
+
+	spin_lock(&mm->page_table_lock);
+	if (unlikely(!pmd_same(*pmd, entry))) {
 		put_page(page);
+		page = NULL;
+		goto unlock;
+	}
+	goto fixup;
 }
 
 int copy_huge_pmd(struct mm_struct *dst_mm, struct mm_struct *src_mm,

Index: linux/mm/migrate.c
===================================================================
@@ -417,7 +417,7 @@ int migrate_huge_page_move_mapping(struc
  */
 void migrate_page_copy(struct page *newpage, struct page *page)
 {
-	if (PageHuge(page))
+	if (PageHuge(page) || PageTransHuge(page))
 		copy_huge_page(newpage, page);
 	else
 		copy_highpage(newpage, page);

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

The NUMA placement code has been rewritten several times, but the basic ideas took a lot of work to develop. The people who put in the work deserve credit for it. Thanks Andrea & Peter :)

[ The Documentation/scheduler/numa-problem.txt file should
probably be rewritten once we figure out the final details of what the NUMA code needs to do, and why. ]

----
This is against tip.git numa/core

---
 CREDITS             |    1 +
 kernel/sched/fair.c |    3 +++
 mm/memory.c         |    2 ++
 3 files changed, 6 insertions(+)

Index: linux/CREDITS
===================================================================
@@ -125,6 +125,7 @@ D: Author of pscan that helps to fix lp/
 D: Author of lil (Linux Interrupt Latency benchmark)
 D: Fixed the shm swap deallocation at swapoff time (try_to_unuse message)
 D: VM hacker
+D: NUMA task placement
 D: Various other kernel hacks
 S: Imola 40026
 S: Italy

Index: linux/kernel/sched/fair.c
===================================================================
@@ -18,6 +18,9 @@
  *
  *  Adaptive scheduling granularity, math enhancements by Peter Zijlstra
  *  Copyright (C) 2007 Red Hat, Inc., Peter Zijlstra <[email protected]>
+ *
+ *  NUMA placement, statistics and algorithm by Andrea Arcangeli,
+ *  CFS balancing changes by Peter Zijlstra. Copyright (C) 2012 Red Hat, Inc.
  */
 
 #include <linux/latencytop.h>

Index: linux/mm/memory.c
===================================================================
@@ -36,6 +36,8 @@
  *		([email protected])
  *
  * Aug/Sep 2004 Changed to four level page tables (Andi Kleen)
+ *
+ * 2012 - NUMA placement page faults (Andrea Arcangeli, Peter Zijlstra)
  */
 
 #include <linux/kernel_stat.h>

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

Introduce a per-page last_cpu field, fold this into the struct page::flags field whenever possible.

The unlikely/rare 32bit NUMA configs will likely grow the page-frame.

[ Completely dropping 32bit support for CONFIG_SCHED_NUMA would simplify
things, but it would also remove the warning if we grow enough 64bit only page-flags to push the last-cpu out. ]

---
 include/linux/mm.h                |   90 ++++++++++++++++++++------------------
 include/linux/mm_types.h          |    9 +++
 include/linux/mmzone.h            |   14 -----
 include/linux/page-flags-layout.h |   83 +++++++++++++++++++++++++++++++++++
 kernel/bounds.c                   |    2 
 mm/huge_memory.c                  |    3 +
 mm/memory.c                       |    4 +
 7 files changed, 151 insertions(+), 54 deletions(-)

Index: linux/include/linux/mm.h
===================================================================
@@ -594,50 +594,11 @@ static inline pte_t maybe_mkwrite(pte_t
  * sets it, so none of the operations on it need to be atomic.
  */
 
-
-/*
- * page->flags layout:
- *
- * There are three possibilities for how page->flags get
- * laid out.  The first is for the normal case, without
- * sparsemem.  The second is for sparsemem when there is
- * plenty of space for node and section.  The last is when
- * we have run out of space and have to fall back to an
- * alternate (slower) way of determining the node.
- *
- * No sparsemem or sparsemem vmemmap: |       NODE     | ZONE | ... | FLAGS |
- * classic sparse with space for node:| SECTION | NODE | ZONE | ... | FLAGS |
- * classic sparse no space for node:  | SECTION |     ZONE    | ... | FLAGS |
- */
-#if defined(CONFIG_SPARSEMEM) && !defined(CONFIG_SPARSEMEM_VMEMMAP)
-#define SECTIONS_WIDTH		SECTIONS_SHIFT
-#else
-#define SECTIONS_WIDTH		0
-#endif
-
-#define ZONES_WIDTH		ZONES_SHIFT
-
-#if SECTIONS_WIDTH+ZONES_WIDTH+NODES_SHIFT <= BITS_PER_LONG - NR_PAGEFLAGS
-#define NODES_WIDTH		NODES_SHIFT
-#else
-#ifdef CONFIG_SPARSEMEM_VMEMMAP
-#error "Vmemmap: No space for nodes field in page flags"
-#endif
-#define NODES_WIDTH		0
-#endif
-
-/* Page flags: | [SECTION] | [NODE] | ZONE | ... | FLAGS | */
+/* Page flags: | [SECTION] | [NODE] | ZONE | [LAST_CPU] | ... | FLAGS | */
 #define SECTIONS_PGOFF		((sizeof(unsigned long)*8) - SECTIONS_WIDTH)
 #define NODES_PGOFF		(SECTIONS_PGOFF - NODES_WIDTH)
 #define ZONES_PGOFF		(NODES_PGOFF - ZONES_WIDTH)
-
-/*
- * We are going to use the flags for the page to node mapping if its in
- * there.  This includes the case where there is no node, so it is implicit.
- */
-#if !(NODES_WIDTH > 0 || NODES_SHIFT == 0)
-#define NODE_NOT_IN_PAGE_FLAGS
-#endif
+#define LAST_CPU_PGOFF		(ZONES_PGOFF - LAST_CPU_WIDTH)
 
 /*
  * Define the bit shifts to access each section.  For non-existent

@@ -647,6 +608,7 @@ static inline pte_t maybe_mkwrite(pte_t
 #define SECTIONS_PGSHIFT	(SECTIONS_PGOFF * (SECTIONS_WIDTH != 0))
 #define NODES_PGSHIFT		(NODES_PGOFF * (NODES_WIDTH != 0))
 #define ZONES_PGSHIFT		(ZONES_PGOFF * (ZONES_WIDTH != 0))
+#define LAST_CPU_PGSHIFT	(LAST_CPU_PGOFF * (LAST_CPU_WIDTH != 0))
 
 /* NODE:ZONE or SECTION:ZONE is used to ID a zone for the buddy allocator */
 #ifdef NODE_NOT_IN_PAGE_FLAGS

@@ -668,6 +630,7 @@ static inline pte_t maybe_mkwrite(pte_t
 #define ZONES_MASK		((1UL << ZONES_WIDTH) - 1)
 #define NODES_MASK		((1UL << NODES_WIDTH) - 1)
 #define SECTIONS_MASK		((1UL << SECTIONS_WIDTH) - 1)
+#define LAST_CPU_MASK		((1UL << LAST_CPU_WIDTH) - 1)
 #define ZONEID_MASK		((1UL << ZONEID_SHIFT) - 1)
 
 static inline enum zone_type page_zonenum(const struct page *page)

@@ -706,6 +669,51 @@ static inline int page_to_nid(const stru
 }
 #endif
 
+#ifdef CONFIG_SCHED_NUMA
+#ifdef LAST_CPU_NOT_IN_PAGE_FLAGS
+static inline int page_xchg_last_cpu(struct page *page, int cpu)
+{
+	return xchg(&page->_last_cpu, cpu);
+}
+
+static inline int page_last_cpu(struct page *page)
+{
+	return page->_last_cpu;
+}
+#else
+static inline int page_xchg_last_cpu(struct page *page, int cpu)
+{
+	unsigned long old_flags, flags;
+	int last_cpu;
+
+	do {
+		old_flags = flags = page->flags;
+		last_cpu = (flags >> LAST_CPU_PGSHIFT) & LAST_CPU_MASK;
+
+		flags &= ~(LAST_CPU_MASK << LAST_CPU_PGSHIFT);
+		flags |= (cpu & LAST_CPU_MASK) << LAST_CPU_PGSHIFT;
+	} while (unlikely(cmpxchg(&page->flags, old_flags, flags) != old_flags));
+
+	return last_cpu;
+}
+
+static inline int page_last_cpu(struct page *page)
+{
+	return (page->flags >> LAST_CPU_PGSHIFT) & LAST_CPU_MASK;
+}
+#endif /* LAST_CPU_NOT_IN_PAGE_FLAGS */
+#else /* CONFIG_SCHED_NUMA */
+static inline int page_xchg_last_cpu(struct page *page, int cpu)
+{
+	return page_to_nid(page);
+}
+
+static inline int page_last_cpu(struct page *page)
+{
+	return page_to_nid(page);
+}
+#endif /* CONFIG_SCHED_NUMA */
+
 static inline struct zone *page_zone(const struct page *page)
 {
 	return &NODE_DATA(page_to_nid(page))->node_zones[page_zonenum(page)];

Index: linux/include/linux/mm_types.h
===================================================================
@@ -12,6 +12,7 @@
 #include <linux/cpumask.h>
 #include <linux/page-debug-flags.h>
 #include <linux/uprobes.h>
+#include <linux/page-flags-layout.h>
 #include <asm/page.h>
 #include <asm/mmu.h>
 

@@ -175,6 +176,10 @@ struct page {
 	 */
 	void *shadow;
 #endif
+
+#ifdef LAST_CPU_NOT_IN_PAGE_FLAGS
+	int _last_cpu;
+#endif
 }
 /*
  * The struct page can be forced to be double word aligned so that atomic ops

@@ -398,6 +403,10 @@ struct mm_struct {
 #ifdef CONFIG_CPUMASK_OFFSTACK
 	struct cpumask cpumask_allocation;
 #endif
+#ifdef CONFIG_SCHED_NUMA
+	unsigned long numa_next_scan;
+	int numa_scan_seq;
+#endif
 	struct uprobes_state uprobes_state;
 };
 

Index: linux/include/linux/mmzone.h
===================================================================
@@ -15,7 +15,7 @@
 #include <linux/seqlock.h>
 #include <linux/nodemask.h>
 #include <linux/pageblock-flags.h>
-#include <generated/bounds.h>
+#include <linux/page-flags-layout.h>
 #include <linux/atomic.h>
 #include <asm/page.h>
 

@@ -318,16 +318,6 @@ enum zone_type {
  * match the requested limits. See gfp_zone() in include/linux/gfp.h
  */
 
-#if MAX_NR_ZONES < 2
-#define ZONES_SHIFT 0
-#elif MAX_NR_ZONES <= 2
-#define ZONES_SHIFT 1
-#elif MAX_NR_ZONES <= 4
-#define ZONES_SHIFT 2
-#else
-#error ZONES_SHIFT -- too many zones configured adjust calculation
-#endif
-
 struct zone {
 	/* Fields commonly accessed by the page allocator */
 

@@ -1030,8 +1020,6 @@ static inline unsigned long early_pfn_to
  * PA_SECTION_SHIFT		physical address to/from section number
  * PFN_SECTION_SHIFT		pfn to/from section number
  */
-#define SECTIONS_SHIFT		(MAX_PHYSMEM_BITS - SECTION_SIZE_BITS)
-
 #define PA_SECTION_SHIFT	(SECTION_SIZE_BITS)
 #define PFN_SECTION_SHIFT	(SECTION_SIZE_BITS - PAGE_SHIFT)
 

Index: linux/kernel/bounds.c
===================================================================
@@ -10,6 +10,7 @@
 #include <linux/mmzone.h>
 #include <linux/kbuild.h>
 #include <linux/page_cgroup.h>
+#include <linux/log2.h>
 
 void foo(void)
 {

@@ -17,5 +18,6 @@ void foo(void)
 	DEFINE(NR_PAGEFLAGS, __NR_PAGEFLAGS);
 	DEFINE(MAX_NR_ZONES, __MAX_NR_ZONES);
 	DEFINE(NR_PCG_FLAGS, __NR_PCG_FLAGS);
+	DEFINE(NR_CPUS_BITS, ilog2(CONFIG_NR_CPUS));
 	/* End of constants */
 }

Index: linux/mm/huge_memory.c
===================================================================
@@ -746,6 +746,7 @@ void do_huge_pmd_numa_page(struct mm_str
 	struct page *new_page = NULL;
 	struct page *page = NULL;
 	int node, lru;
+	int last_cpu;
 
 	spin_lock(&mm->page_table_lock);
 	if (unlikely(!pmd_same(*pmd, entry)))

@@ -760,6 +761,7 @@ void do_huge_pmd_numa_page(struct mm_str
 	page = pmd_page(entry);
 	if (page) {
 		VM_BUG_ON(!PageCompound(page) || !PageHead(page));
+		last_cpu = page_last_cpu(page);
 
 		get_page(page);
 		node = mpol_misplaced(page, vma, haddr);

@@ -1441,6 +1443,7 @@ static void __split_huge_page_refcount(s
 		page_tail->mapping = page->mapping;
 
 		page_tail->index = page->index + i;
+		page_xchg_last_cpu(page, page_last_cpu(page_tail));
 
 		BUG_ON(!PageAnon(page_tail));
 		BUG_ON(!PageUptodate(page_tail));

Index: linux/mm/memory.c
===================================================================
@@ -70,6 +70,10 @@
 
 #include "internal.h"
 
+#ifdef LAST_CPU_NOT_IN_PAGE_FLAGS
+#warning Unfortunate NUMA config, growing page-frame for last_cpu.
+#endif
+
 #ifndef CONFIG_NEED_MULTIPLE_NODES
 /* use the per-pgdat data instead for discontigmem - mbligh */
 unsigned long max_mapnr;

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

A cleanliness side note, this bit does not belong into this patch:

The principal ideas behind this patch are the fundamental difference between shared and privately used memory and the very strong desire to only rely on per-task behavioral state for scheduling decisions.

We define 'shared memory' as all user memory that is frequently accessed by multiple tasks and conversely 'private memory' is the user memory used predominantly by a single task.

To approximate the above strict definition we recognise that task placement is dominantly per cpu and thus using cpu granular page access state is a natural fit. Thus we introduce page::last_cpu as the cpu that last accessed a page.

Using this, we can construct two per-task node-vectors, 'S_i' and 'P_i' reflecting the amount of shared and privately used pages of this task respectively. Pages for which two consecutive 'hits' are of the same cpu are assumed private and the others are shared.

[ This means that we will start evaluating this state when the
task has not migrated for at least 2 scans, see NUMA_SETTLE ]

Using these vectors we can compute the total number of shared/private pages of this task and determine which dominates.

[ Note that for shared tasks we only see '1/n' the total number
of shared pages for the other tasks will take the other faults; where 'n' is the number of tasks sharing the memory. So for an equal comparison we should divide total private by 'n' as well, but we don't have 'n' so we pick 2. ]

We can also compute which node holds most of our memory, running on this node will be called 'ideal placement' (As per previous patches we will prefer to pull memory towards wherever we run.)

We change the load-balancer to prefer moving tasks in order of:

1) !numa tasks and numa tasks in the direction of more faults
2) allow !ideal tasks getting worse in the direction of faults
3) allow private tasks to get worse
4) allow shared tasks to get worse

This order ensures we prefer increasing memory locality but when we do have to make hard decisions we prefer spreading private over shared, because spreading shared tasks significantly increases the interconnect bandwidth since not all memory can follow.

We also add an extra 'lateral' force to the load balancer that perturbs the state when otherwise 'fairly' balanced. This ensures we don't get 'stuck' in a state which is fair but undesired from a memory location POV (see can_do_numa_run()).

Lastly, we allow shared tasks to defeat the default spreading of tasks such that, when possible, they can aggregate on a single node.

Shared tasks aggregate for the very simple reason that there has to be a single node that holds most of their memory and a second most, etc.. and tasks want to move up the faults ladder.

---
 Documentation/scheduler/numa-problem.txt |   20 
 arch/sh/mm/Kconfig                       |    1 
 include/linux/init_task.h                |    8 
 include/linux/sched.h                    |   44 +
 include/uapi/linux/mempolicy.h           |    1 
 init/Kconfig                             |   14 
 kernel/sched/core.c                      |   68 ++
 kernel/sched/fair.c                      |  983 +++++++++++++++++++++++++------
 kernel/sched/features.h                  |    9 
 kernel/sched/sched.h                     |   32 -
 kernel/sysctl.c                          |   31 
 mm/huge_memory.c                         |    7 
 mm/memory.c                              |    6 
 mm/mempolicy.c                           |   95 ++
 mm/migrate.c                             |    6 
 15 files changed, 1097 insertions(+), 228 deletions(-)

Index: linux/Documentation/scheduler/numa-problem.txt
===================================================================
@@ -133,6 +133,8 @@ XXX properties of this M vs a potential
 
  2b) migrate memory towards 'n_i' using 2 samples.
 
+XXX include the statistical babble on double sampling somewhere near
+
 This separates pages into those that will migrate and those that will not due
 to the two samples not matching. We could consider the first to be of 'p_i'
 (private) and the second to be of 's_i' (shared).

@@ -142,7 +144,17 @@ This interpretation can be motivated by
 's_i' (shared). (here we loose the need for memory limits again, since it
 becomes indistinguishable from shared).
 
-XXX include the statistical babble on double sampling somewhere near
+ 2c) use cpu samples instead of node samples
+
+The problem with sampling on node granularity is that one looses 's_i' for
+the local node, since one cannot distinguish between two accesses from the
+same node.
+
+By increasing the granularity to per-cpu we gain the ability to have both an
+'s_i' and 'p_i' per node. Since we do all task placement per-cpu as well this
+seems like a natural match. The line where we overcommit cpus is where we loose
+granularity again, but when we loose overcommit we naturally spread tasks.
+Therefore it should work out nicely.
 
 This reduces the problem further; we loose 'M' as per 2a, it further reduces
 the 'T_k,l' (interconnect traffic) term to only include shared (since per the

@@ -150,12 +162,6 @@ above all private will be local):
 
   T_k,l = \Sum_i bs_i,l for every n_i = k, l != k
 
-[ more or less matches the state of sched/numa and describes its remaining
-  problems and assumptions. It should work well for tasks without significant
-  shared memory usage between tasks. ]
-
-Possible future directions:
-
 Motivated by the form of 'T_k,l', try and obtain each term of the sum, so we
 can evaluate it;
 

Index: linux/arch/sh/mm/Kconfig
===================================================================
@@ -111,6 +111,7 @@ config VSYSCALL
 config NUMA
 	bool "Non Uniform Memory Access (NUMA) Support"
 	depends on MMU && SYS_SUPPORTS_NUMA && EXPERIMENTAL
+	select EMBEDDED_NUMA
 	default n
 	help
 	  Some SH systems have many various memories scattered around

Index: linux/include/linux/init_task.h
===================================================================
@@ -143,6 +143,13 @@ extern struct task_group root_task_group
 
 #define INIT_TASK_COMM "swapper"
 
+#ifdef CONFIG_SCHED_NUMA
+# define INIT_TASK_NUMA(tsk)						\
+	.numa_shared = -1,
+#else
+# define INIT_TASK_NUMA(tsk)
+#endif
+
 /*
  *  INIT_TASK is used to set up the first task table, touch at
  * your own risk!. Base=0, limit=0x1fffff (=2MB)

@@ -210,6 +217,7 @@ extern struct task_group root_task_group
 	INIT_TRACE_RECURSION						\
 	INIT_TASK_RCU_PREEMPT(tsk)					\
 	INIT_CPUSET_SEQ							\
+	INIT_TASK_NUMA(tsk)						\
 }
 
 

Index: linux/include/uapi/linux/mempolicy.h
===================================================================
@@ -69,6 +69,7 @@ enum mpol_rebind_step {
 #define MPOL_F_LOCAL   (1 << 1)	/* preferred local allocation */
 #define MPOL_F_REBINDING (1 << 2)	/* identify policies in rebinding */
 #define MPOL_F_MOF	(1 << 3) /* this policy wants migrate on fault */
+#define MPOL_F_HOME	(1 << 4) /* this is the home-node policy */
 
 
 #endif /* _UAPI_LINUX_MEMPOLICY_H */

Index: linux/init/Kconfig
===================================================================
@@ -696,6 +696,20 @@ config LOG_BUF_SHIFT
 config HAVE_UNSTABLE_SCHED_CLOCK
 	bool
 
+#
+# For architectures that (ab)use NUMA to represent different memory regions
+# all cpu-local but of different latencies, such as SuperH.
+#
+config EMBEDDED_NUMA
+	bool
+
+config SCHED_NUMA
+	bool "Memory placement aware NUMA scheduler"
+	default n
+	depends on SMP && NUMA && MIGRATION && !EMBEDDED_NUMA
+	help
+	  This option adds support for automatic NUMA aware memory/task placement.
+
 menuconfig CGROUPS
 	boolean "Control Group support"
 	depends on EVENTFD

Index: linux/kernel/sched/features.h
===================================================================
@@ -66,3 +66,12 @@ SCHED_FEAT(TTWU_QUEUE, true)
 SCHED_FEAT(FORCE_SD_OVERLAP, false)
 SCHED_FEAT(RT_RUNTIME_SHARE, true)
 SCHED_FEAT(LB_MIN, false)
+
+#ifdef CONFIG_SCHED_NUMA
+/* Do the working set probing faults: */
+SCHED_FEAT(NUMA,             true)
+SCHED_FEAT(NUMA_FAULTS_UP,   true)
+SCHED_FEAT(NUMA_FAULTS_DOWN, false)
+SCHED_FEAT(NUMA_SETTLE,      true)
+#endif
+

Index: linux/kernel/sched/sched.h
===================================================================
@@ -3,6 +3,7 @@
 #include <linux/mutex.h>
 #include <linux/spinlock.h>
 #include <linux/stop_machine.h>
+#include <linux/slab.h>
 
 #include "cpupri.h"
 

@@ -420,17 +421,29 @@ struct rq {
 	unsigned long cpu_power;
 
 	unsigned char idle_balance;
-	/* For active balancing */
 	int post_schedule;
+
+	/* For active balancing */
 	int active_balance;
-	int push_cpu;
-	struct cpu_stop_work active_balance_work;
+	int ab_dst_cpu;
+	int ab_flags;
+	int ab_failed;
+	int ab_idle;
+	struct cpu_stop_work ab_work;
+
 	/* cpu of this runqueue: */
 	int cpu;
 	int online;
 
 	struct list_head cfs_tasks;
 
+#ifdef CONFIG_SCHED_NUMA
+	unsigned long numa_weight;
+	unsigned long nr_numa_running;
+	unsigned long nr_ideal_running;
+#endif
+	unsigned long nr_shared_running;	/* 0 on non-NUMA */
+
 	u64 rt_avg;
 	u64 age_stamp;
 	u64 idle_stamp;

@@ -501,6 +514,18 @@ DECLARE_PER_CPU(struct rq, runqueues);
 #define cpu_curr(cpu)		(cpu_rq(cpu)->curr)
 #define raw_rq()		(&__raw_get_cpu_var(runqueues))
 
+#ifdef CONFIG_SCHED_NUMA
+extern void sched_setnuma(struct task_struct *p, int node, int shared);
+static inline void task_numa_free(struct task_struct *p)
+{
+	kfree(p->numa_faults);
+}
+#else /* CONFIG_SCHED_NUMA */
+static inline void task_numa_free(struct task_struct *p)
+{
+}
+#endif /* CONFIG_SCHED_NUMA */
+
 #ifdef CONFIG_SMP
 
 #define rcu_dereference_check_sched_domain(p) \

@@ -544,6 +569,7 @@ static inline struct sched_domain *highe
 
 DECLARE_PER_CPU(struct sched_domain *, sd_llc);
 DECLARE_PER_CPU(int, sd_llc_id);
+DECLARE_PER_CPU(struct sched_domain *, sd_node);
 
 extern int group_balance_cpu(struct sched_group *sg);
 

Index: linux/kernel/sysctl.c
===================================================================
@@ -256,9 +256,11 @@ static int min_sched_granularity_ns = 10
 static int max_sched_granularity_ns = NSEC_PER_SEC;	/* 1 second */
 static int min_wakeup_granularity_ns;			/* 0 usecs */
 static int max_wakeup_granularity_ns = NSEC_PER_SEC;	/* 1 second */
+#ifdef CONFIG_SMP
 static int min_sched_tunable_scaling = SCHED_TUNABLESCALING_NONE;
 static int max_sched_tunable_scaling = SCHED_TUNABLESCALING_END-1;
-#endif
+#endif /* CONFIG_SMP */
+#endif /* CONFIG_SCHED_DEBUG */
 
 #ifdef CONFIG_COMPACTION
 static int min_extfrag_threshold;

@@ -301,6 +303,7 @@ static struct ctl_table kern_table[] = {
 		.extra1		= &min_wakeup_granularity_ns,
 		.extra2		= &max_wakeup_granularity_ns,
 	},
+#ifdef CONFIG_SMP
 	{
 		.procname	= "sched_tunable_scaling",
 		.data		= &sysctl_sched_tunable_scaling,

@@ -347,7 +350,31 @@ static struct ctl_table kern_table[] = {
 		.extra1		= &zero,
 		.extra2		= &one,
 	},
-#endif
+#endif /* CONFIG_SMP */
+#ifdef CONFIG_SCHED_NUMA
+	{
+		.procname	= "sched_numa_scan_period_min_ms",
+		.data		= &sysctl_sched_numa_scan_period_min,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+	},
+	{
+		.procname	= "sched_numa_scan_period_max_ms",
+		.data		= &sysctl_sched_numa_scan_period_max,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+	},
+	{
+		.procname	= "sched_numa_settle_count",
+		.data		= &sysctl_sched_numa_settle_count,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+	},
+#endif /* CONFIG_SCHED_NUMA */
+#endif /* CONFIG_SCHED_DEBUG */
 	{
 		.procname	= "sched_rt_period_us",
 		.data		= &sysctl_sched_rt_period,

Index: linux/mm/huge_memory.c
===================================================================
@@ -777,9 +777,10 @@ fixup:
 
 unlock:
 	spin_unlock(&mm->page_table_lock);
-	if (page)
+	if (page) {
+		task_numa_fault(page_to_nid(page), last_cpu, HPAGE_PMD_NR);
 		put_page(page);
-
+	}
 	return;
 
 migrate:

@@ -848,6 +849,8 @@ migrate:
 
 	put_page(page);			/* Drop the rmap reference */
 
+	task_numa_fault(node, last_cpu, HPAGE_PMD_NR);
+
 	if (lru)
 		put_page(page);		/* drop the LRU isolation reference */
 

Index: linux/mm/memory.c
===================================================================
@@ -3484,6 +3484,7 @@ static int do_numa_page(struct mm_struct
 {
 	struct page *page = NULL;
 	int node, page_nid = -1;
+	int last_cpu = -1;
 	spinlock_t *ptl;
 
 	ptl = pte_lockptr(mm, pmd);

@@ -3495,6 +3496,7 @@ static int do_numa_page(struct mm_struct
 	if (page) {
 		get_page(page);
 		page_nid = page_to_nid(page);
+		last_cpu = page_last_cpu(page);
 		node = mpol_misplaced(page, vma, address);
 		if (node != -1)
 			goto migrate;

@@ -3514,8 +3516,10 @@ out_pte_upgrade_unlock:
 out_unlock:
 	pte_unmap_unlock(ptep, ptl);
 out:
-	if (page)
+	if (page) {
+		task_numa_fault(page_nid, last_cpu, 1);
 		put_page(page);
+	}
 
 	return 0;
 

Index: linux/mm/migrate.c
===================================================================
@@ -1427,12 +1427,6 @@ int migrate_misplaced_page(struct page *
 	gfp_t gfp = GFP_HIGHUSER_MOVABLE;
 
 	/*
-	 * Don't migrate pages that are mapped in multiple processes.
-	 */
-	if (page_mapcount(page) != 1)
-		goto out;
-
-	/*
 	 * Never wait for allocations just to migrate on fault, but don't dip
 	 * into reserves. And, only accept pages from the specified node. No
 	 * sense migrating to a different "misplaced" page!

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

"All"? Should that not be "a memory segment that is frequently..."?

The classification is per task? But most tasks have memory areas that are private and other areas where shared accesses occur. Can that be per memory area? Private areas need to be kept with the process. Shared areas may have to be spread across nodes if the memory area is too large.

Guess that is too complicated to determine unless we would be using vmas which may only roughly correlate to the memory regions for which memory policies are currently manually setup.

But then this is rather different from my expectations that I had after reading the intro.

We do useless moves and create additional overhead?

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

Yes, exactly - access patterns are fundamentally and physically per task, as a task can execute only on a single CPU at once. (I say 'task' instead of 'thread' or 'process' because the new code makes no distinction between threads and processes.)

The new code maps out inter-task relationships, statistically.

So we are basically able to (statistically) approximate which task relates to which other task in the system, based on their memory access patterns alone: using a very compact metric and matching scheduler rules and a (lazy) memory placement machinery on the VM side.

Say consider the following 10-task workload, where the scheduler is able to figure out these relationships:

{ A, B, C, D } dominantly share memory X with each other { E, F, G, H } dominantly share memory Y with each other
{ I } uses memory privately
{ J } uses memory privately

and the scheduler rules then try to converge these groups of tasks ideally.

[ The 'role' and grouping of tasks is not static but sampled and
average based - so if a worker thread changes its role, the scheduler will adapt placement to that. ]

[ A 'private task' is basically a special case for sharing
memory: if a task only shares memory with itself. Its placement and spreading is easy. ]

Do you mean per vma, and/or per mm?

How would that work? Consider the above case:

- 12x CPU-intense threads and a large piece of shared memory

- 2x 4 threads are using two large shared memory area to calculate (one area for each group of threads)

- the 4 remaining processes aggregate and sort the results from the 8 threads, in their own dominantly 'private' working set.

how does per vma or per mm describe that properly? The whole workload might be just within a single large vma within a JVM. Or it might be implemented using processes and anonymous shared memory.

If you look at this from a 'per task access pattern and inter-task working set relationship' perspective then the resolution and optimization is natural: the 2x 4 threads should be grouped together modulo capacity constraints, while the remaining 4 'private memory' threads should be spread out over the remaining capacity of the system.

What matters is how tasks relate to each other as they perform processing, not which APIs the workload uses to create tasks and memory areas.

The main constraint from a placement optimization complexity POV is task->CPU placement: for NUMA workloads the main challenge - and 80% of the code and much of the real meat of the feature - is to categorize and place tasks properly.

There might be much discussion about PROT_NONE and memory migration details, but that is because the VM code is 5 times larger than the scheduler code and due to that there's 5 times more VM hackers than scheduler hackers ;-)

In reality the main complexity of this problem [the placement optimization problem portion] is a dominantly CPU/task scheduler feature, and IMO rather fundamentally so: it's not an implementation choice but derives from the Von Neumann model of computing in essence.

And that is why IMO the task based access pattern metric implementaton is such a good fit in practice as well - and that is why other approaches struggled getting a hold of the NUMA problem.

Thanks,

Ingo

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

That is an intriguing idea. It will be interesting to see how well it works with various workloads.

Unless I am misreading the code (it is a little hard to read in places, more on that further down), the number picked appears to be 4.

This reflects some of the things autonuma does, so I suspect it will work in your code too :)

It is interesting to see how sched/numa has moved from the homenodes-through-syscalls concepts to something so close to what autonuma does.

Yes, they are. I have read this part of the patch several times, and am still not sure what exactly the code is doing, or why.

All rights reversed

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to [email protected].  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

Combined with the fact that we only turn a certain amount of memory into NUMA ptes each second, could this result in a program being classified as a private task one second, and a shared task a few seconds later?

What does the code do to prevent such an oscillating of task classification? (which would have consequences for the way the task's NUMA placement is handled, and might result in the task moving from node to node needlessly)

All rights reversed

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to [email protected].  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

It's a statistical method, like most of scheduling.

It's as prone to oscillation as tasks are already prone to being moved spuriously by the load balancer today, due to the per CPU load average being statistical and them being slightly above or below a critical load average value.

Higher freq oscillation should not happen normally though, we dampen these metrics and have per CPU hysteresis.

( We can also add explicit hysteresis if anyone demonstrates
real oscillation with a real workload - wanted to keep it simple first and change it only as-needed. )

Thanks,

Ingo

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

This heuristic is by no means simple, and there still is no explanation for the serious performance degradations that were seen on a 4 node system running specjbb in 4 node-sized JVMs.

I asked a number of questions on this patch yesterday, and am hoping to get explanations at some point :)

All rights reversed

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to [email protected].  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

Previously, to probe the working set of a task, we'd use a very simple and crude method: mark all of its address space PROT_NONE.

That method has various (obvious) disadvantages:

- it samples the working set at dissimilar rates, giving some tasks a sampling quality advantage over others.

- creates performance problems for tasks with very large working sets

- over-samples processes with large address spaces but which only very rarely execute

Improve that method by keeping a rotating offset into the address space that marks the current position of the scan, and advance it by a constant rate (in a CPU cycles execution proportional manner). If the offset reaches the last mapped address of the mm then it then it starts over at the first address.

The per-task nature of the working set sampling functionality in this tree allows such constant rate, per task, execution-weight proportional sampling of the working set, with an adaptive sampling interval/frequency that goes from once per 100 msecs up to just once per 1.6 seconds. The current sampling volume is 256 MB per interval.

As tasks mature and converge their working set, so does the sampling rate slow down to just a trickle, 256 MB per 1.6 seconds of CPU time executed.

This, beyond being adaptive, also rate-limits rarely executing systems and does not over-sample on overloaded systems.

[ In AutoNUMA speak, this patch deals with the effective sampling
rate of the 'hinting page fault'. AutoNUMA's scanning is currently rate-limited, but it is also fundamentally single-threaded, executing in the knuma_scand kernel thread, so the limit in AutoNUMA is global and does not scale up with the number of CPUs, nor does it scan tasks in an execution proportional manner.

So the idea of rate-limiting the scanning was first implemented in the AutoNUMA tree via a global rate limit. This patch goes beyond that by implementing an execution rate proportional working set sampling rate that is not implemented via a single global scanning daemon. ]

[ Dan Carpenter pointed out a possible NULL pointer dereference in the
first version of this patch. ]

---
 include/linux/mm_types.h |    1 +
 include/linux/sched.h    |    1 +
 kernel/sched/fair.c      |   43 ++++++++++++++++++++++++++++++-------------
 kernel/sysctl.c          |    7 +++++++
 4 files changed, 39 insertions(+), 13 deletions(-)

Index: linux/include/linux/mm_types.h
===================================================================
@@ -405,6 +405,7 @@ struct mm_struct {
 #endif
 #ifdef CONFIG_SCHED_NUMA
 	unsigned long numa_next_scan;
+	unsigned long numa_scan_offset;
 	int numa_scan_seq;
 #endif
 	struct uprobes_state uprobes_state;

Index: linux/include/linux/sched.h
===================================================================
@@ -2047,6 +2047,7 @@ extern enum sched_tunable_scaling sysctl
 
 extern unsigned int sysctl_sched_numa_scan_period_min;
 extern unsigned int sysctl_sched_numa_scan_period_max;
+extern unsigned int sysctl_sched_numa_scan_size;
 extern unsigned int sysctl_sched_numa_settle_count;
 
 #ifdef CONFIG_SCHED_DEBUG

Index: linux/kernel/sysctl.c
===================================================================
@@ -367,6 +367,13 @@ static struct ctl_table kern_table[] = {
 		.proc_handler	= proc_dointvec,
 	},
 	{
+		.procname	= "sched_numa_scan_size_mb",
+		.data		= &sysctl_sched_numa_scan_size,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+	},
+	{
 		.procname	= "sched_numa_settle_count",
 		.data		= &sysctl_sched_numa_settle_count,
 		.maxlen		= sizeof(unsigned int),

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

By accounting against the present PTEs, scanning speed reflects the actual present (mapped) memory.

---
 include/linux/hugetlb.h |    8 ++++++--
 include/linux/mm.h      |    6 +++---
 kernel/sched/fair.c     |   37 +++++++++++++++++++++----------------
 mm/hugetlb.c            |   10 ++++++++--
 mm/mprotect.c           |   41 ++++++++++++++++++++++++++++++-----------
 5 files changed, 68 insertions(+), 34 deletions(-)

Index: linux/include/linux/hugetlb.h
===================================================================
@@ -87,7 +87,7 @@ struct page *follow_huge_pud(struct mm_s
 				pud_t *pud, int write);
 int pmd_huge(pmd_t pmd);
 int pud_huge(pud_t pmd);
-void hugetlb_change_protection(struct vm_area_struct *vma,
+unsigned long hugetlb_change_protection(struct vm_area_struct *vma,
 		unsigned long address, unsigned long end, pgprot_t newprot);
 
 #else /* !CONFIG_HUGETLB_PAGE */

@@ -132,7 +132,11 @@ static inline void copy_huge_page(struct
 {
 }
 
-#define hugetlb_change_protection(vma, address, end, newprot)
+static inline unsigned long hugetlb_change_protection(struct vm_area_struct *vma,
+		unsigned long address, unsigned long end, pgprot_t newprot)
+{
+	return 0;
+}
 
 static inline void __unmap_hugepage_range_final(struct mmu_gather *tlb,
 			struct vm_area_struct *vma, unsigned long start,

Index: linux/include/linux/mm.h
===================================================================
@@ -1099,7 +1099,7 @@ extern unsigned long move_page_tables(st
 extern unsigned long do_mremap(unsigned long addr,
 			       unsigned long old_len, unsigned long new_len,
 			       unsigned long flags, unsigned long new_addr);
-extern void change_protection(struct vm_area_struct *vma, unsigned long start,
+extern unsigned long change_protection(struct vm_area_struct *vma, unsigned long start,
 			      unsigned long end, pgprot_t newprot,
 			      int dirty_accountable);
 extern int mprotect_fixup(struct vm_area_struct *vma,

@@ -1581,10 +1581,10 @@ static inline pgprot_t vma_prot_none(str
 	return pgprot_modify(vma->vm_page_prot, vm_get_page_prot(vmflags));
 }
 
-static inline void
+static inline unsigned long
 change_prot_none(struct vm_area_struct *vma, unsigned long start, unsigned long end)
 {
-	change_protection(vma, start, end, vma_prot_none(vma), 0);
+	return change_protection(vma, start, end, vma_prot_none(vma), 0);
 }
 
 struct vm_area_struct *find_extend_vma(struct mm_struct *, unsigned long addr);

Index: linux/kernel/sched/fair.c
===================================================================
@@ -914,8 +914,8 @@ void task_numa_work(struct callback_head
 	struct task_struct *p = current;
 	struct mm_struct *mm = p->mm;
 	struct vm_area_struct *vma;
-	unsigned long offset, end;
-	long length;
+	unsigned long start, end;
+	long pages;
 
 	WARN_ON_ONCE(p != container_of(work, struct task_struct, numa_work));
 

@@ -942,30 +942,35 @@ void task_numa_work(struct callback_head
 	if (cmpxchg(&mm->numa_next_scan, migrate, next_scan) != migrate)
 		return;
 
-	offset = mm->numa_scan_offset;
-	length = sysctl_sched_numa_scan_size;
-	length <<= 20;
+	start = mm->numa_scan_offset;
+	pages = sysctl_sched_numa_scan_size;
+	pages <<= 20 - PAGE_SHIFT; /* MB in pages */
+	if (!pages)
+		return;
 
 	down_write(&mm->mmap_sem);
-	vma = find_vma(mm, offset);
+	vma = find_vma(mm, start);
 	if (!vma) {
 		ACCESS_ONCE(mm->numa_scan_seq)++;
-		offset = 0;
+		start = 0;
 		vma = mm->mmap;
 	}
-	for (; vma && length > 0; vma = vma->vm_next) {
+	for (; vma; vma = vma->vm_next) {
 		if (!vma_migratable(vma))
 			continue;
 
-		offset = max(offset, vma->vm_start);
-		end = min(ALIGN(offset + length, HPAGE_SIZE), vma->vm_end);
-		length -= end - offset;
-
-		change_prot_none(vma, offset, end);
-
-		offset = end;
+		do {
+			start = max(start, vma->vm_start);
+			end = ALIGN(start + (pages << PAGE_SHIFT), HPAGE_SIZE);
+			end = min(end, vma->vm_end);
+			pages -= change_prot_none(vma, start, end);
+			start = end;
+			if (pages <= 0)
+				goto out;
+		} while (end != vma->vm_end);
 	}
-	mm->numa_scan_offset = offset;
+out:
+	mm->numa_scan_offset = start;
 	up_write(&mm->mmap_sem);
 }
 

Index: linux/mm/hugetlb.c
===================================================================
@@ -3014,7 +3014,7 @@ same_page:
 	return i ? i : -EFAULT;
 }
 
-void hugetlb_change_protection(struct vm_area_struct *vma,
+unsigned long hugetlb_change_protection(struct vm_area_struct *vma,
 		unsigned long address, unsigned long end, pgprot_t newprot)
 {
 	struct mm_struct *mm = vma->vm_mm;

@@ -3022,6 +3022,7 @@ void hugetlb_change_protection(struct vm
 	pte_t *ptep;
 	pte_t pte;
 	struct hstate *h = hstate_vma(vma);
+	unsigned long pages = 0;
 
 	BUG_ON(address >= end);
 	flush_cache_range(vma, address, end);

@@ -3032,12 +3033,15 @@ void hugetlb_change_protection(struct vm
 		ptep = huge_pte_offset(mm, address);
 		if (!ptep)
 			continue;
-		if (huge_pmd_unshare(mm, &address, ptep))
+		if (huge_pmd_unshare(mm, &address, ptep)) {
+			pages++;
 			continue;
+		}
 		if (!huge_pte_none(huge_ptep_get(ptep))) {
 			pte = huge_ptep_get_and_clear(mm, address, ptep);
 			pte = pte_mkhuge(pte_modify(pte, newprot));
 			set_huge_pte_at(mm, address, ptep, pte);
+			pages++;
 		}
 	}
 	spin_unlock(&mm->page_table_lock);

@@ -3049,6 +3053,8 @@ void hugetlb_change_protection(struct vm
 	 */
 	flush_tlb_range(vma, start, end);
 	mutex_unlock(&vma->vm_file->f_mapping->i_mmap_mutex);
+
+	return pages << h->order;
 }
 
 int hugetlb_reserve_pages(struct inode *inode,

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

Add a 1 second delay before starting to scan the working set of a task and starting to balance it amongst nodes.

[ note that before the constant per task WSS sampling rate patch
the initial scan would happen much later still, in effect that patch caused this regression. ]

The theory is that short-run tasks benefit very little from NUMA
placement: they come and go, and they better stick to the node they were started on. As tasks mature and rebalance to other CPUs and nodes, so does their NUMA placement have to change and so does it start to matter more and more.

In practice this change fixes an observable kbuild regression:

# [ a perf stat --null --repeat 10 test of ten bzImage builds to /dev/shm ]

!NUMA:

   45.291088843 seconds time elapsed                                          ( +-  0.40% )
   45.154231752 seconds time elapsed                                          ( +-  0.36% )

+NUMA, no slow start:

   46.172308123 seconds time elapsed                                          ( +-  0.30% )
   46.343168745 seconds time elapsed                                          ( +-  0.25% )

+NUMA, 1 sec slow start:

   45.224189155 seconds time elapsed                                          ( +-  0.25% )
   45.160866532 seconds time elapsed                                          ( +-  0.17% )

and it also fixes an observable perf bench (hackbench) regression:

   # perf stat --null --repeat 10 perf bench sched messaging

-NUMA:

   -NUMA:                  0.246225691 seconds time elapsed                   ( +-  1.31% )
   +NUMA no slow start:    0.252620063 seconds time elapsed                   ( +-  1.13% )

+NUMA 1sec delay: 0.248076230 seconds time elapsed ( +- 1.35% )

The implementation is simple and straightforward, most of the patch deals with adding the /proc/sys/kernel/sched_numa_scan_delay_ms tunable knob.

---
 include/linux/sched.h |    1 +
 kernel/sched/core.c   |    2 +-
 kernel/sched/fair.c   |   16 ++++++++++------
 kernel/sysctl.c       |    7 +++++++
 4 files changed, 19 insertions(+), 7 deletions(-)

Index: linux/include/linux/sched.h
===================================================================
@@ -2045,6 +2045,7 @@ enum sched_tunable_scaling {
 };
 extern enum sched_tunable_scaling sysctl_sched_tunable_scaling;
 
+extern unsigned int sysctl_sched_numa_scan_delay;
 extern unsigned int sysctl_sched_numa_scan_period_min;
 extern unsigned int sysctl_sched_numa_scan_period_max;
 extern unsigned int sysctl_sched_numa_scan_size;

Index: linux/kernel/sched/core.c
===================================================================
@@ -1556,7 +1556,7 @@ static void __sched_fork(struct task_str
 	p->numa_scan_seq = p->mm ? p->mm->numa_scan_seq : 0;
 	p->numa_migrate_seq = 2;
 	p->numa_faults = NULL;
-	p->numa_scan_period = sysctl_sched_numa_scan_period_min;
+	p->numa_scan_period = sysctl_sched_numa_scan_delay;
 	p->numa_work.next = &p->numa_work;
 #endif /* CONFIG_SCHED_NUMA */
 }

Index: linux/kernel/sched/fair.c
===================================================================
@@ -823,11 +823,12 @@ static void account_numa_dequeue(struct
 }
 
 /*
- * numa task sample period in ms: 5s
+ * Scan @scan_size MB every @scan_period after an initial @scan_delay.
  */
-unsigned int sysctl_sched_numa_scan_period_min = 100;
-unsigned int sysctl_sched_numa_scan_period_max = 100*16;
-unsigned int sysctl_sched_numa_scan_size = 256;   /* MB */
+unsigned int sysctl_sched_numa_scan_delay = 1000;	/* ms */
+unsigned int sysctl_sched_numa_scan_period_min = 100;	/* ms */
+unsigned int sysctl_sched_numa_scan_period_max = 100*16;/* ms */
+unsigned int sysctl_sched_numa_scan_size = 256;		/* MB */
 
 /*
  * Wait for the 2-sample stuff to settle before migrating again

@@ -938,10 +939,12 @@ void task_numa_work(struct callback_head
 	if (time_before(now, migrate))
 		return;
 
-	next_scan = now + 2*msecs_to_jiffies(sysctl_sched_numa_scan_period_min);
+	next_scan = now + msecs_to_jiffies(sysctl_sched_numa_scan_period_min);
 	if (cmpxchg(&mm->numa_next_scan, migrate, next_scan) != migrate)
 		return;
 
+	current->numa_scan_period += jiffies_to_msecs(2);
+
 	start = mm->numa_scan_offset;
 	pages = sysctl_sched_numa_scan_size;
 	pages <<= 20 - PAGE_SHIFT; /* MB in pages */

@@ -998,7 +1001,8 @@ void task_tick_numa(struct rq *rq, struc
 	period = (u64)curr->numa_scan_period * NSEC_PER_MSEC;
 
 	if (now - curr->node_stamp > period) {
-		curr->node_stamp = now;
+		curr->node_stamp += period;
+		curr->numa_scan_period = sysctl_sched_numa_scan_period_min;
 
 		/*
 		 * We are comparing runtime to wall clock time here, which

Index: linux/kernel/sysctl.c
===================================================================
@@ -353,6 +353,13 @@ static struct ctl_table kern_table[] = {
 #endif /* CONFIG_SMP */
 #ifdef CONFIG_SCHED_NUMA
 	{
+		.procname	= "sched_numa_scan_delay_ms",
+		.data		= &sysctl_sched_numa_scan_delay,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+	},
+	{
 		.procname	= "sched_numa_scan_period_min_ms",
 		.data		= &sysctl_sched_numa_scan_period_min,
 		.maxlen		= sizeof(unsigned int),

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

Here are some preliminary performance figures, comparing the vanilla kernel against the CONFIG_SCHED_NUMA=y kernel.

Java SPEC benchmark, running on a 4 node, 64 GB, 32-way server system (higher numbers are better):

   v3.7-vanilla:    run #1:    475630
                    run #2:    538271
                    run #3:    533888
                    run #4:    431525
                    ----------------------------------
                       avg:    494828 transactions/sec

   v3.7-NUMA:       run #1:    626692
                    run #2:    622069
                    run #3:    630335
                    run #4:    629817
                    ----------------------------------
                       avg:    627228 transactions/sec    [ +26.7% ]

Beyond the +26.7% performance improvement in throughput, the standard deviation of the results is much lower as well with NUMA scheduling enabled, by about an order of magnitude.

[ That is probably so because memory and task placement is more
balanced with NUMA scheduling enabled - while with the vanilla kernel initial placement of the working set determines the final performance figure. ]

I've also tested Andrea's 'autonumabench' benchmark suite against vanilla and the NUMA kernel, because Mel reported that the CONFIG_SCHED_NUMA=y code regressed. It does not regress anymore:

  #
  # NUMA01
  #
  perf stat --null --repeat 3 ./numa01

v3.7-vanilla: 340.3 seconds ( +/- 0.31% )
v3.7-NUMA: 216.9 seconds [ +56% ] ( +/- 8.32% )
-------------------------------------
v3.7-HARD_BIND: 166.6 seconds

Here the new NUMA code is faster than vanilla by 56% - that is because with the vanilla kernel all memory is allocated on node0, overloading that node's memory bandwidth.

[ Standard deviation on the vanilla kernel is low, because the
autonuma test causes close to the worst-case placement for the vanilla kernel - and there's not much space to deviate away from the worst-case. Despite that, stddev in the NUMA seems a tad high, suggesting further room for improvement. ]

  #
  # NUMA01_THREAD_ALLOC
  #
  perf stat --null --repeat 3 ./numa01_THREAD_ALLOC

   v3.7-vanilla:            425.1 seconds             ( +/- 1.04% )
   v3.7-NUMA:               118.7 seconds  [ +250% ]  ( +/- 0.49% )

-------------------------------------
v3.7-HARD_BIND: 200.56 seconds

Here the NUMA kernel was able to go beyond the (naive) hard-binding result and achieved 3.5x the performance of the vanilla kernel, with a low stddev.

  #
  # NUMA02
  #
  perf stat --null --repeat 3 ./numa02

   v3.7-vanilla:           56.1 seconds               ( +/- 0.72% )
   v3.7-NUMA:              17.0 seconds    [ +230% ]  ( +/- 0.18% )

-------------------------------------
v3.7-HARD_BIND: 14.9 seconds

Here the NUMA kernel runs the test much (3.3x) faster than the vanilla kernel. The workload is able to converge very quickly and approximate the hard-binding ideal number very closely. If runtime was a bit longer it would approximate it even closer.

Standard deviation is also 3 times lower than vanilla, suggesting stable NUMA convergence.

  #
  # NUMA02_SMT
  #
  perf stat --null --repeat 3 ./numa02_SMT
   v3.7-vanilla:            56.1 seconds                 ( +- 0.42% )
   v3.7-NUMA:               17.3 seconds     [ +220% ]   ( +- 0.88% )
   -------------------------------------
   v3.7-HARD_BIND:          14.6 seconds

In this test too the NUMA kernel outperforms the vanilla kernel, by a factor of 3.2x. It comes very close to the ideal hard-binding convergence result. Standard deviation is a bit high.

I have also created a new perf benchmarking and workload generation tool: 'perf bench numa' (I'll post it later in a separate reply).

Via 'perf bench numa' we can generate arbitrary process and thread layouts, with arbitrary memory sharing arrangements between them.

Here are various comparisons to the vanilla kernel (higher numbers are better):

#
# 4 processes with 4 threads per process, sharing 4x 1GB of

  # process-wide memory:
  #
  # perf bench numa mem -l 100 -zZ0 -p 4 -t 4 -P 1024 -T    0
  #
           v3.7-vanilla:       14.8 GB/sec
           v3.7-NUMA:          32.9 GB/sec    [ +122.3% ]

2.2 times faster.

#
# 4 processes with 4 threads per process, sharing 4x 1GB of

  # process-wide memory:
  #
  # perf bench numa mem -l 100 -zZ0 -p 4 -t 4 -P    0 -T 1024
  #

           v3.7-vanilla:        17.0 GB/sec
           v3.7-NUMA:           36.3 GB/sec    [ +113.5% ]

2.1 times faster.

So it's a nice improvement all around. With this version the regressions that Mel Gorman reported a week ago appear to be fixed as well.

Thanks,

Ingo

ps. If anyone is curious about further details, let me know.
The base kernel I used for measurement was commit 02743c9c03f1 + the 8 patches Peter sent out.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

Ok, I used a 4-node, 64G, 48-way server system. We have different CPUs but the same number of nodes. In case it makes a difference each of my machines nodes are the same size.

I did not see the same results. I used 3.7-rc4 as a baseline as it's what I'm developing against currently. For your patches I pulled tip/sched/core and then applied the patches you posted to the mailing list on top. It means my tree looks different to yours but it was necessary if I was going to do a like-with-like comparison. I also rebased Andrea'a autonuma28fast branch from his git tree onto 3.7-rc4 (some mess, but nothing very serious).

As before, I'm cutting this report short

SPECJBB BOPS

                          3.7.0                 3.7.0                 3.7.0
                rc4-stats-v2r34    rc4-schednuma-v2r3  rc4-autonuma-v28fast
Mean   1      25034.25 (  0.00%)     20598.50 (-17.72%)     25192.25 (  0.63%)
Mean   2      53176.00 (  0.00%)     43906.50 (-17.43%)     55508.25 (  4.39%)
Mean   3      77350.50 (  0.00%)     60342.75 (-21.99%)     82122.50 (  6.17%)
Mean   4      99919.50 (  0.00%)     80781.75 (-19.15%)    107233.25 (  7.32%)
Mean   5     119797.00 (  0.00%)     97870.00 (-18.30%)    131016.00 (  9.37%)
Mean   6     135858.00 (  0.00%)    123912.50 ( -8.79%)    152444.75 ( 12.21%)
Mean   7     136074.00 (  0.00%)    126574.25 ( -6.98%)    157372.75 ( 15.65%)
Mean   8     132426.25 (  0.00%)    121766.00 ( -8.05%)    161655.25 ( 22.07%)
Mean   9     129432.75 (  0.00%)    114224.25 (-11.75%)    160530.50 ( 24.03%)
Mean   10    118399.75 (  0.00%)    109040.50 ( -7.90%)    158692.00 ( 34.03%)
Mean   11    119604.00 (  0.00%)    105566.50 (-11.74%)    154462.00 ( 29.14%)
Mean   12    112742.25 (  0.00%)    101728.75 ( -9.77%)    149546.00 ( 32.64%)
Mean   13    109480.75 (  0.00%)    103737.50 ( -5.25%)    144929.25 ( 32.38%)
Mean   14    109724.00 (  0.00%)    103516.00 ( -5.66%)    143804.50 ( 31.06%)
Mean   15    109111.75 (  0.00%)    100817.00 ( -7.60%)    141878.00 ( 30.03%)
Mean   16    105385.75 (  0.00%)     99327.25 ( -5.75%)    140156.75 ( 32.99%)
Mean   17    101903.50 (  0.00%)     96464.50 ( -5.34%)    138402.00 ( 35.82%)
Mean   18    103632.50 (  0.00%)     95632.50 ( -7.72%)    137781.50 ( 32.95%)
Stddev 1       1195.76 (  0.00%)       358.07 ( 70.06%)       861.97 ( 27.91%)
Stddev 2        883.39 (  0.00%)      1203.29 (-36.21%)       855.08 (  3.20%)
Stddev 3        997.25 (  0.00%)      3755.67 (-276.60%)       545.50 ( 45.30%)
Stddev 4       1115.16 (  0.00%)      6390.65 (-473.07%)      1183.49 ( -6.13%)
Stddev 5       1367.09 (  0.00%)      9710.70 (-610.32%)      1022.09 ( 25.24%)
Stddev 6       1125.22 (  0.00%)      1097.83 (  2.43%)      1013.52 (  9.93%)
Stddev 7       3211.72 (  0.00%)      1533.62 ( 52.25%)       512.61 ( 84.04%)
Stddev 8       4194.96 (  0.00%)      1518.26 ( 63.81%)       493.64 ( 88.23%)
Stddev 9       6175.10 (  0.00%)      2648.75 ( 57.11%)      2109.83 ( 65.83%)
Stddev 10      4754.87 (  0.00%)      1941.47 ( 59.17%)      2948.98 ( 37.98%)
Stddev 11      2706.18 (  0.00%)      1247.95 ( 53.89%)      5907.16 (-118.28%)
Stddev 12      3607.76 (  0.00%)       663.63 ( 81.61%)      9063.28 (-151.22%)
Stddev 13      2771.67 (  0.00%)      1447.87 ( 47.76%)      8716.51 (-214.49%)
Stddev 14      2522.18 (  0.00%)      1510.28 ( 40.12%)      9286.98 (-268.21%)
Stddev 15      2711.16 (  0.00%)      1719.54 ( 36.58%)      9895.88 (-265.01%)
Stddev 16      2797.21 (  0.00%)       983.63 ( 64.84%)      9302.92 (-232.58%)
Stddev 17      4019.85 (  0.00%)      1927.25 ( 52.06%)      9998.34 (-148.72%)
Stddev 18      3332.20 (  0.00%)      1401.68 ( 57.94%)     12056.08 (-261.80%)
TPut   1     100137.00 (  0.00%)     82394.00 (-17.72%)    100769.00 (  0.63%)
TPut   2     212704.00 (  0.00%)    175626.00 (-17.43%)    222033.00 (  4.39%)
TPut   3     309402.00 (  0.00%)    241371.00 (-21.99%)    328490.00 (  6.17%)
TPut   4     399678.00 (  0.00%)    323127.00 (-19.15%)    428933.00 (  7.32%)
TPut   5     479188.00 (  0.00%)    391480.00 (-18.30%)    524064.00 (  9.37%)
TPut   6     543432.00 (  0.00%)    495650.00 ( -8.79%)    609779.00 ( 12.21%)
TPut   7     544296.00 (  0.00%)    506297.00 ( -6.98%)    629491.00 ( 15.65%)
TPut   8     529705.00 (  0.00%)    487064.00 ( -8.05%)    646621.00 ( 22.07%)
TPut   9     517731.00 (  0.00%)    456897.00 (-11.75%)    642122.00 ( 24.03%)
TPut   10    473599.00 (  0.00%)    436162.00 ( -7.90%)    634768.00 ( 34.03%)
TPut   11    478416.00 (  0.00%)    422266.00 (-11.74%)    617848.00 ( 29.14%)
TPut   12    450969.00 (  0.00%)    406915.00 ( -9.77%)    598184.00 ( 32.64%)
TPut   13    437923.00 (  0.00%)    414950.00 ( -5.25%)    579717.00 ( 32.38%)
TPut   14    438896.00 (  0.00%)    414064.00 ( -5.66%)    575218.00 ( 31.06%)
TPut   15    436447.00 (  0.00%)    403268.00 ( -7.60%)    567512.00 ( 30.03%)
TPut   16    421543.00 (  0.00%)    397309.00 ( -5.75%)    560627.00 ( 32.99%)
TPut   17    407614.00 (  0.00%)    385858.00 ( -5.34%)    553608.00 ( 35.82%)
TPut   18    414530.00 (  0.00%)    382530.00 ( -7.72%)    551126.00 ( 32.95%)

It is important to know how this was configured. I was running one JVM per node and the JVMs were sized that they should fit in the node. This is a semi-ideal configuration because it could also be hard-bound for best performance on the vanilla kernel. You did not say if you ran with a single JVM or multiple JVMs and it's important.

The mean values are based on the individual throughput figures reported by each JVM. schednuma regresses against mainline quite badly. For low numbers of warehouses it also deviates more but it's much steadier for higher numbers of warehouses. In terms of overall throughput though, it's worse.

autonuma deviates a *lot* with massive variances between the JVMs. However, the average and total throughput is very high.

SPECJBB PEAKS
3.7.0 3.7.0 3.7.0
rc4-stats-v2r34 rc4-schednuma-v2r3 rc4-autonuma-v28fast

 Expctd Warehouse                   12.00 (  0.00%)                   12.00 (  0.00%)                   12.00 (  0.00%)
 Expctd Peak Bops               450969.00 (  0.00%)               406915.00 ( -9.77%)               598184.00 ( 32.64%)
 Actual Warehouse                    7.00 (  0.00%)                    7.00 (  0.00%)                    8.00 ( 14.29%)
 Actual Peak Bops               544296.00 (  0.00%)               506297.00 ( -6.98%)               646621.00 ( 18.80%)

There is no major difference in terms of scalability. They peak at around the 7 warehouse mark. autonuma peaked at 8 but you can see from the figures that it was not by a whole lot. autonumas actual peak operations was very high (18% gain) where schednuma regressed by close to 7%.

MMTests Statistics: duration

               3.7.0       3.7.0       3.7.0
        rc4-stats-v2r34rc4-schednuma-v2r3rc4-autonuma-v28fast

User 101949.84 86817.79 101748.80
System 66.05 13094.99 191.40
Elapsed 2456.35 2459.16 2451.96

system CPU time is high for schednuma. autonuma reports low system CPU usage but as it is using kernel threads for much of its work, it cannot be considered reliable as it would not be captured here.

For machines with more than 2 nodes, numa01 is an adverse workload.

With this benchark, I'm generally seeing very good results in terms of elapsed time.

AUTONUMA BENCH

                                          3.7.0                 3.7.0                 3.7.0
                                rc4-stats-v2r34    rc4-schednuma-v2r3  rc4-autonuma-v28fast
User    NUMA01               67351.66 (  0.00%)    47146.57 ( 30.00%)    30273.64 ( 55.05%)
User    NUMA01_THEADLOCAL    54788.28 (  0.00%)    17198.99 ( 68.61%)    17039.73 ( 68.90%)
User    NUMA02                7179.87 (  0.00%)     2096.07 ( 70.81%)     2099.85 ( 70.75%)
User    NUMA02_SMT            3028.11 (  0.00%)      998.22 ( 67.03%)     1052.97 ( 65.23%)
System  NUMA01                  45.68 (  0.00%)     3531.04 (-7629.95%)      423.91 (-828.00%)
System  NUMA01_THEADLOCAL       40.92 (  0.00%)      926.72 (-2164.71%)      188.15 (-359.80%)
System  NUMA02                   1.72 (  0.00%)       23.64 (-1274.42%)       27.37 (-1491.28%)
System  NUMA02_SMT               0.92 (  0.00%)        8.18 (-789.13%)       18.43 (-1903.26%)
Elapsed NUMA01                1514.61 (  0.00%)     1122.78 ( 25.87%)      722.66 ( 52.29%)
Elapsed NUMA01_THEADLOCAL     1264.08 (  0.00%)      393.79 ( 68.85%)      391.48 ( 69.03%)
Elapsed NUMA02                 181.88 (  0.00%)       49.44 ( 72.82%)       61.55 ( 66.16%)
Elapsed NUMA02_SMT             168.41 (  0.00%)       47.49 ( 71.80%)       54.72 ( 67.51%)
CPU     NUMA01                4449.00 (  0.00%)     4513.00 ( -1.44%)     4247.00 (  4.54%)
CPU     NUMA01_THEADLOCAL     4337.00 (  0.00%)     4602.00 ( -6.11%)     4400.00 ( -1.45%)
CPU     NUMA02                3948.00 (  0.00%)     4287.00 ( -8.59%)     3455.00 ( 12.49%)
CPU     NUMA02_SMT            1798.00 (  0.00%)     2118.00 (-17.80%)     1957.00 ( -8.84%)

On NUMA01, I'm seeing a large gain for schednuma. The test was not run multiple times so I do not know how much it deviates by on each run. However, the system CPU usage was again very high.

NUMA01_THEADLOCAL figures were comparable with autonuma. The system CPU usage was high. As before, autonumas looks low but with the kernel threads we cannot be sure.

schednuma was a clear winner on NUMA02 and NUMA02_SMT.

So for the synthetic benchmarks, schednuma looks good in terms of elapsed time. On specjbb though, it is not looking great and this may be due to differences in how we configured the JVMs.

I would have some comparison data with my own stuff but unfortunately the machine crashed when running tests with schednuma. That said, I expect the figures to be bad if they had run. With V2, the CPU-follows placement policy is broken as is PMD handling. In my current tree I'm expecting the system CPU usage to be also high but I won't know for sure until later today.

The machine was meant to test all this overnight but unfortunately when running a kernel build benchmark on the schednuma patches the machine hung while downloading the tarball with this

[   73.863226] BUG: unable to handle kernel NULL pointer dereference at           (null)
[   73.871062] IP: [<ffffffff8146feaa>] skb_gro_receive+0xaa/0x590
[   73.876983] PGD 0 
[   73.878998] Oops: 0002 [#1] PREEMPT SMP 
[   73.882938] Modules linked in: af_packet mperf kvm_intel coretemp kvm crc32c_intel ghash_clmulni_intel aesni_intel ablk_helper cryptd sr_mod lrw cdrom aes_x86_64 ses pcspkr xts i7core_edac ata_piix enclosure lpc_ich dcdbas sg gf128mul mfd_core bnx2 edac_core wmi acpi_power_meter button serio_raw joydev microcode autofs4 processor thermal_sys scsi_dh_rdac scsi_dh_hp_sw scsi_dh_alua scsi_dh_emc scsi_dh ata_generic megaraid_sas pata_atiixp [last unloaded: oprofile]
[   73.924659] CPU 0 
[   73.926493] Pid: 0, comm: swapper/0 Not tainted 3.7.0-rc4-schednuma-v2r3 #1 Dell Inc. PowerEdge R810/0TT6JF
[   73.936380] RIP: 0010:[<ffffffff8146feaa>]  [<ffffffff8146feaa>] skb_gro_receive+0xaa/0x590
[   73.944714] RSP: 0018:ffff88047f803b50  EFLAGS: 00010282
[   73.950004] RAX: 0000000000000000 RBX: ffff88046c2bdbc0 RCX: 0000000000000900
[   73.957113] RDX: 00000000000005a8 RSI: ffff88046c2bdbc0 RDI: ffff88046eadb800
[   73.964221] RBP: ffff88047f803bb0 R08: 00000000000005dc R09: ffff88046ddeccc0
[   73.971328] R10: ffff88086d795d78 R11: 0000000000000001 R12: ffff880462b282c0
[   73.978436] R13: 0000000000000034 R14: 00000000000005a8 R15: ffff88046eadbec0
[   73.985543] FS:  0000000000000000(0000) GS:ffff88047f800000(0000) knlGS:0000000000000000
[   73.993602] CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
[   73.999326] CR2: 0000000000000000 CR3: 0000000001a0c000 CR4: 00000000000007f0
[   74.006435] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[   74.013543] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
[   74.020651] Process swapper/0 (pid: 0, threadinfo ffffffff81a00000, task ffffffff81a14420)
[   74.028883] Stack:
[   74.030885]  0000000000000060 ffff880462b282c0 ffff88086d795d78 ffffffff000005dc
[   74.038300]  ffff88046e5f46c0 000000606a275ec0 0000000000000000 ffff88046c2bdbc0
[   74.045715]  00000000000005a8 ffff88086d795d78 00000000000005a8 000000006c001080
[   74.053131] Call Trace:
[   74.055567]  <IRQ> 
[   74.057486]  [<ffffffff814b9573>] tcp_gro_receive+0x213/0x2b0
[   74.063419]  [<ffffffff814cff49>] tcp4_gro_receive+0x99/0x110
[   74.069150]  [<ffffffff814e096d>] inet_gro_receive+0x1cd/0x200
[   74.074965]  [<ffffffff8147b30a>] dev_gro_receive+0x1ba/0x2b0
[   74.080691]  [<ffffffff8147b6e3>] napi_gro_receive+0xe3/0x130
[   74.086426]  [<ffffffffa009fda8>] bnx2_rx_int+0x3e8/0xf10 [bnx2]
[   74.092416]  [<ffffffffa00a0cbd>] bnx2_poll_work+0x3ed/0x450 [bnx2]
[   74.098666]  [<ffffffffa00a0d5e>] bnx2_poll_msix+0x3e/0xc0 [bnx2]
[   74.104739]  [<ffffffff8147b969>] net_rx_action+0x159/0x290
[   74.110298]  [<ffffffff8104d148>] __do_softirq+0xc8/0x250
[   74.115682]  [<ffffffff8107bf9e>] ? sched_clock_idle_wakeup_event+0x1e/0x20
[   74.122625]  [<ffffffff81577c9c>] call_softirq+0x1c/0x30
[   74.127922]  [<ffffffff8100470d>] do_softirq+0x6d/0xa0
[   74.133041]  [<ffffffff8104d44d>] irq_exit+0xad/0xc0
[   74.137996]  [<ffffffff8107779d>] scheduler_ipi+0x5d/0x110
[   74.143469]  [<ffffffff8102b7a4>] ? native_apic_msr_eoi_write+0x14/0x20
[   74.150060]  [<ffffffff810257d5>] smp_reschedule_interrupt+0x25/0x30
[   74.156394]  [<ffffffff8157785d>] reschedule_interrupt+0x6d/0x80
[   74.162376]  <EOI> 
[   74.164295]  [<ffffffff81316798>] ? intel_idle+0xe8/0x150
[   74.169875]  [<ffffffff81316779>] ? intel_idle+0xc9/0x150
[   74.175259]  [<ffffffff8143de99>] cpuidle_enter+0x19/0x20
[   74.180642]  [<ffffffff8143e522>] cpuidle_idle_call+0xa2/0x340
[   74.186458]  [<ffffffff8100baca>] cpu_idle+0x7a/0xf0
[   74.191410]  [<ffffffff8154b44b>] rest_init+0x7b/0x80
[   74.196447]  [<ffffffff81ac3be2>] start_kernel+0x38f/0x39c
[   74.201913]  [<ffffffff81ac3652>] ? repair_env_string+0x5e/0x5e
[   74.207815]  [<ffffffff81ac3335>] x86_64_start_reservations+0x131/0x135
[   74.214407]  [<ffffffff81ac3439>] x86_64_start_kernel+0x100/0x10f
[   74.220475] Code: 8b e8 00 00 00 0f 87 86 00 00 00 8b 53 68 8b 43 6c 44 29 ea 39 d0 89 53 68 0f 87 c7 04 00 00 4c 01 ab e0 00 00 00 49 8b 44 24 08 <48> 89 18 49 89 5c 24 08 0f b6 43 7c a8 10 0f 85 ac 04 00 00 83 
[   74.240051] RIP  [<ffffffff8146feaa>] skb_gro_receive+0xaa/0x590
[   74.246046]  RSP <ffff88047f803b50>
[   74.249518] CR2: 0000000000000000
[   74.252821] ---[ end trace 97cb529523f52c9b ]---
[   74.258895] Kernel panic - not syncing: Fatal exception in interrupt

-- 0:console -- time-stamp -- Nov/15/12 3:09:06 --

I've no idea if it is directly related to your patches and I didn't try to reproduce it yet.

That is really cool.

Unfortunately I cannot concur. I'm still seeing high system CPU usage in places and the specjbb figures are rather unfortunate.

Mel Gorman
SUSE Labs

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to [email protected].  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

Davem, Eric -

this oops may be related to the numa patches, but quite frankly, I
don't see why/how it should be. And there's been some GRO work since 3.6, and we had an earlier oops case, so I thought I'd forward this.

The code decodes to

14: 39 d0 cmp %edx,%eax
16: 89 53 68 mov %edx,0x68(%rbx)
19: 0f 87 c7 04 00 00 ja 0x4e6
1f: 4c 01 ab e0 00 00 00 add %r13,0xe0(%rbx)
26: 49 8b 44 24 08 mov 0x8(%r12),%rax
2b:* 48 89 18 mov %rbx,(%rax) <-- trapping instruction
2e: 49 89 5c 24 08 mov %rbx,0x8(%r12)
33: 0f b6 43 7c movzbl 0x7c(%rbx),%eax
37: a8 10 test $0x10,%al

and if I read the disassembly right (which is not guaranteed), it's the line

        p->prev->next = skb;

in the "merge:" case in skb_gro_receive() (just after the __skb_pull()
- the "ja" and "add" above the trapping instruction is the BUG_ON() plus the "skb->data += len" part of the inlined __skb_pull()).

Linus

Mel, do you have info on exactly what model system you were running these tests on?

Obviously your results are very different from the ones that Ingo saw. It would be most helpful if we could find a similar system in one of the Red Hat labs, so Ingo can play around with it and see what's going on :)

... on Ingo's system:

... on Mel's system:

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

Dell PowerEdge R810
CPU Intel(R) Xeon(R) CPU E7- 4807 @ 1.87GHz
RAM 64G
Single disk

4 JVMs, one per node
SpecJBB configured to run in multi JVM configuration No special binding
JVM switches -Xmx12882m

All run through an unreleased version of MMTests. I'll make a release of mmtests either tomorrow or Monday when I get the chance.

Also compare how the benchmark is actually configured and which figures he's reporting. I'm posting up the throughput for each warehouse and the peak throughput.

It is possible Ingo's figures are based on other patches in the tip tree that have not been identified. If that's the case it's interesting in itself.

Mel Gorman
SUSE Labs

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to [email protected].  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

Ugh.

According to these numbers, the latest sched-numa actually regresses against mainline on Specjbb.

No way is this even close to ready for merging in the 3.8 timeframe.

I would ask the invilved people to please come up with a set of initial patches that people agree on, so that we can at least start merging some of the infrastructure, and see how far we can get on at least getting *started*. As I mentioned to Andrew and Mel separately, nobody seems to disagree with the TLB optimization patches. What else? Is Mel's set of early patches still considered a reasonable starting point for everybody?

Ingo? Andrea? With the understanding that we're not going to merge the actual full schednuma/autonuma, what are the initial parts we can *agree* on?

Linus

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

Mel's infrastructure patches, 1-14 and 17 out of his latest series, could be a great starting point.

Ingo is trying to get the mm/ code in his tree to be mostly the same to Mel's code anyway, so that is the infrastructure everybody wants.

At that point, we can focus our discussions on just the policy side, which could help us zoom in on the issues.

It would also make it possible for us to do apple to apple comparisons between the various policy decisions, allowing us to reach a decision based on data, not just gut feel.

As long as each tree has its own basic infrastructure, we cannot do apples to apples comparisons; this has frustrated the discussion for months.

Having all that basic infrastructure upstream should short-circuit that part of the discussion.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

V3 increased a lot in size due to rate-limiting of migration which was yanked out of autonuma. The rate limiting has two obvious purposes. One, during periods of fast convergency it will prevent the memory bus being saturated with traffic and causing stalls. As a side effect it should decrease system CPU usage in some cases. Two, if the placement policy completely breaks down, it will help contain the damage. If we added a vmstat that increments when the rate limiting kicked in then users could report broken policies by checking if the migration and rate-limited counter are increasing. If they are both increasing rapidly then the placement policy is broken. I think identifying when it's broken is just as important as identifying when it's working.

The equivalent numbered patches in the new series to match what Rik suggests above are Patches 1-17, 19. I'll swap patches 19 and 18 to avoid this mess. The TLB patches are 33-35 but are not contested. I am going to move them to the start of the series.

With some shuffling the question on what to consider for merging becomes

1. TLB optimisation patches 1-3? Patches 1-3
2. Stats for migration? Patches 4-6
3. Common NUMA infrastructure? Patches 7-21
4. Basic fault-driven policy, stats, ratelimits Patches 22-35

Patches 36-43 are complete cabbage and should not be considered at this stage. It should be possible to build the placement policies and the scheduling decisions from schednuma, autonuma, some combination of the above or something completely different on top of patches 1-35.

Peter, Ingo, Andrea?

I know that other common patches that should exist but they are optimisations to the policies and not a fundamental design choice.

Preferably yes and we'd have a comparison points of mainline and the most basic of placement policies to work with that should be bisectable as a last resort.

Agreed.

Mel Gorman
SUSE Labs

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to [email protected].  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

Hi,

I assume you mean simply reshuffling 33-35 as 1-3.

The patches 1-35 looks a great foundation so I think they'd be an ideal candidate for a first upstream inclusion.

Thanks,
Andrea

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

Yes.

Thanks.

Mel Gorman
SUSE Labs

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to [email protected].  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

That would definitely be a step forward.

My suggestion for a 'foundation' would be all the non-policy bits in numa/core:

c740b1cccdcb x86/mm: Completely drop the TLB flush from ptep_set_access_flags() 02743c9c03f1 mm/mpol: Use special PROT_NONE to migrate pages b33467764d8a mm/migrate: Introduce migrate_misplaced_page() db4aa58db59a numa, mm: Support NUMA hinting page faults from gup/gup_fast ca2ea0747a5b mm/mpol: Add MPOL_MF_LAZY
f05ea0948708 mm/mpol: Create special PROT_NONE infrastructure 37081a3de2bf mm/mpol: Check for misplaced page
cd203e33c39d mm/mpol: Add MPOL_MF_NOOP
88f4670789e3 mm/mpol: Make MPOL_LOCAL a real policy
83babc0d2944 mm/pgprot: Move the pgprot_modify() fallback definition to mm.h 536165ead34b sched, numa, mm, MIPS/thp: Add pmd_pgprot() implementation 6fe64360a759 mm: Only flush the TLB when clearing an accessible pte e9df40bfeb25 x86/mm: Introduce pte_accessible()
3f2b613771ec mm/thp: Preserve pgprot across huge page split a5a608d83e0e sched, numa, mm, s390/thp: Implement pmd_pgprot() for s390 995334a2ee83 sched, numa, mm: Describe the NUMA scheduling problem formally 7ee9d9209c57 sched, numa, mm: Make find_busiest_queue() a method 4fd98847ba5c x86/mm: Only do a local tlb flush in ptep_set_access_flags() d24fc0571afb mm/generic: Only flush the local TLB in ptep_set_access_flags()

Which I've pushed out into the separate numa/base tree:

git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git numa/base

These are just the minimal set of patches needed to get to be able to concentrate on the real details.

AFAICS Mel started going in this design direction as well in his latest patches, so there should be no real technical objections to this other than any details I might have missed: and I'll rebase this tree if the mm/ folks have any other suggestions for improvement, as that seems the be the preferred mm workflow.

Andrea, Mel?

Getting this out of the way would be a big help.

Thanks,

Ingo

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

That is not what I tested: as I described it in the mail I tested 32 warehouses: i.e. spanning the whole system.

You tested 4 parallel JVMs running one per node, right?

Thanks,

Ingo

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

Good (sortof) because that's my preferred explanation as to why we are seeing different results. Different machines and different kernels would be a lot more problematic.

4 parallel JVMs sized so they they could fit one-per-node. However, I did *not* bind them to nodes because that would be completely pointless for this type of test.

I've queued up another set of tests and added a single-JVM configuration to the mix. The kernels will have debugging, lockstat enabled and will be running two passes with the second pass running profiling so the results will not be directly comparable. However, I'll keep a close eye on the Single vs Multi JVM results.

Thanks.

Mel Gorman
SUSE Labs

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to [email protected].  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

Not just 'sortof' good but it appears it's unconditionally good:
meanwhile other testers have reproduced the single-JVM speedup with the latest numa/core code as well, so the speedup is not just on my system.

Please post your kernel .config so I can check why the 4x JVM test does not perform so well on your system. Maybe there's something special to your system.

Thanks,

Ingo

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

I was referring to the very strong likelihood that this was our major source of difference. I would not go as far as "unconditionally good" but I'm very happy that this was the major difference so it was "sortof good" :) The alternatives completely sucked.

Good. Did they also test kernel building because that showed around a 50% regression for me as posted elsewhere. It's for catching things like this I wanted the logical progression so we can catch exactly where problems got introduced and figure out what was missed at each time.

I do not store configs from tests but this is the config that should have been generated for the test.

#
# Automatically generated file; DO NOT EDIT.
# Linux/x86_64 3.7.0-rc4 Kernel Configuration
#
CONFIG_64BIT=y
CONFIG_X86_64=y
CONFIG_X86=y
CONFIG_INSTRUCTION_DECODER=y
CONFIG_OUTPUT_FORMAT="elf64-x86-64"
CONFIG_ARCH_DEFCONFIG="arch/x86/configs/x86_64_defconfig"
CONFIG_LOCKDEP_SUPPORT=y
CONFIG_STACKTRACE_SUPPORT=y
CONFIG_HAVE_LATENCYTOP_SUPPORT=y
CONFIG_MMU=y
CONFIG_NEED_DMA_MAP_STATE=y
CONFIG_NEED_SG_DMA_LENGTH=y
CONFIG_GENERIC_ISA_DMA=y
CONFIG_GENERIC_BUG=y
CONFIG_GENERIC_BUG_RELATIVE_POINTERS=y
CONFIG_GENERIC_HWEIGHT=y
CONFIG_GENERIC_GPIO=y
CONFIG_ARCH_MAY_HAVE_PC_FDC=y
CONFIG_RWSEM_XCHGADD_ALGORITHM=y
CONFIG_GENERIC_CALIBRATE_DELAY=y
CONFIG_ARCH_HAS_CPU_RELAX=y
CONFIG_ARCH_HAS_DEFAULT_IDLE=y
CONFIG_ARCH_HAS_CACHE_LINE_SIZE=y
CONFIG_ARCH_HAS_CPU_AUTOPROBE=y
CONFIG_HAVE_SETUP_PER_CPU_AREA=y
CONFIG_NEED_PER_CPU_EMBED_FIRST_CHUNK=y
CONFIG_NEED_PER_CPU_PAGE_FIRST_CHUNK=y
CONFIG_ARCH_HIBERNATION_POSSIBLE=y
CONFIG_ARCH_SUSPEND_POSSIBLE=y
CONFIG_ZONE_DMA32=y
CONFIG_AUDIT_ARCH=y
CONFIG_ARCH_SUPPORTS_OPTIMIZED_INLINING=y
CONFIG_ARCH_SUPPORTS_DEBUG_PAGEALLOC=y
CONFIG_HAVE_INTEL_TXT=y
CONFIG_X86_64_SMP=y
CONFIG_X86_HT=y
CONFIG_ARCH_HWEIGHT_CFLAGS="-fcall-saved-rdi -fcall-saved-rsi -fcall-saved-rdx -fcall-saved-rcx -fcall-saved-r8 -fcall-saved-r9 -fcall-saved-r10 -fcall-saved-r11" CONFIG_ARCH_CPU_PROBE_RELEASE=y
CONFIG_ARCH_SUPPORTS_UPROBES=y
CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
CONFIG_HAVE_IRQ_WORK=y
CONFIG_IRQ_WORK=y
CONFIG_BUILDTIME_EXTABLE_SORT=y

#
# General setup
#
CONFIG_EXPERIMENTAL=y
CONFIG_INIT_ENV_ARG_LIMIT=32
CONFIG_CROSS_COMPILE=""
CONFIG_LOCALVERSION=""
# CONFIG_LOCALVERSION_AUTO is not set
CONFIG_HAVE_KERNEL_GZIP=y
CONFIG_HAVE_KERNEL_BZIP2=y
CONFIG_HAVE_KERNEL_LZMA=y
CONFIG_HAVE_KERNEL_XZ=y
CONFIG_HAVE_KERNEL_LZO=y
CONFIG_KERNEL_GZIP=y
# CONFIG_KERNEL_BZIP2 is not set
# CONFIG_KERNEL_LZMA is not set
# CONFIG_KERNEL_XZ is not set
# CONFIG_KERNEL_LZO is not set
CONFIG_DEFAULT_HOSTNAME="(none)"
CONFIG_SWAP=y
CONFIG_SYSVIPC=y
CONFIG_SYSVIPC_SYSCTL=y
CONFIG_POSIX_MQUEUE=y
CONFIG_POSIX_MQUEUE_SYSCTL=y
CONFIG_FHANDLE=y
CONFIG_AUDIT=y
CONFIG_AUDITSYSCALL=y
CONFIG_AUDIT_WATCH=y
CONFIG_AUDIT_TREE=y
# CONFIG_AUDIT_LOGINUID_IMMUTABLE is not set
CONFIG_HAVE_GENERIC_HARDIRQS=y

#
# IRQ subsystem
#
CONFIG_GENERIC_HARDIRQS=y
CONFIG_GENERIC_IRQ_PROBE=y
CONFIG_GENERIC_IRQ_SHOW=y
CONFIG_GENERIC_PENDING_IRQ=y
CONFIG_GENERIC_IRQ_CHIP=y
CONFIG_IRQ_DOMAIN=y
# CONFIG_IRQ_DOMAIN_DEBUG is not set
CONFIG_IRQ_FORCED_THREADING=y
CONFIG_SPARSE_IRQ=y
CONFIG_CLOCKSOURCE_WATCHDOG=y
CONFIG_ARCH_CLOCKSOURCE_DATA=y
CONFIG_GENERIC_TIME_VSYSCALL=y
CONFIG_GENERIC_CLOCKEVENTS=y
CONFIG_GENERIC_CLOCKEVENTS_BUILD=y
CONFIG_GENERIC_CLOCKEVENTS_BROADCAST=y
CONFIG_GENERIC_CLOCKEVENTS_MIN_ADJUST=y
CONFIG_GENERIC_CMOS_UPDATE=y

#
# Timers subsystem
#
CONFIG_TICK_ONESHOT=y
CONFIG_NO_HZ=y
CONFIG_HIGH_RES_TIMERS=y

#
# CPU/Task time and stats accounting
#
CONFIG_TICK_CPU_ACCOUNTING=y
# CONFIG_IRQ_TIME_ACCOUNTING is not set
CONFIG_BSD_PROCESS_ACCT=y
CONFIG_BSD_PROCESS_ACCT_V3=y
CONFIG_TASKSTATS=y
CONFIG_TASK_DELAY_ACCT=y
CONFIG_TASK_XACCT=y
CONFIG_TASK_IO_ACCOUNTING=y

#
# RCU Subsystem
#
CONFIG_TREE_PREEMPT_RCU=y
CONFIG_PREEMPT_RCU=y
# CONFIG_RCU_USER_QS is not set
CONFIG_RCU_FANOUT=64
CONFIG_RCU_FANOUT_LEAF=16
# CONFIG_RCU_FANOUT_EXACT is not set
CONFIG_RCU_FAST_NO_HZ=y
# CONFIG_TREE_RCU_TRACE is not set
CONFIG_RCU_BOOST=y
CONFIG_RCU_BOOST_PRIO=1
CONFIG_RCU_BOOST_DELAY=500
CONFIG_IKCONFIG=y
CONFIG_IKCONFIG_PROC=y
CONFIG_LOG_BUF_SHIFT=18
CONFIG_HAVE_UNSTABLE_SCHED_CLOCK=y
CONFIG_ARCH_SUPPORTS_NUMA_BALANCING=y
CONFIG_NUMA_BALANCING=y
CONFIG_CGROUPS=y
# CONFIG_CGROUP_DEBUG is not set
CONFIG_CGROUP_FREEZER=y
CONFIG_CGROUP_DEVICE=y
CONFIG_CPUSETS=y
CONFIG_PROC_PID_CPUSET=y
CONFIG_CGROUP_CPUACCT=y
CONFIG_RESOURCE_COUNTERS=y
# CONFIG_MEMCG is not set
# CONFIG_CGROUP_HUGETLB is not set
CONFIG_CGROUP_PERF=y
CONFIG_CGROUP_SCHED=y
CONFIG_FAIR_GROUP_SCHED=y
CONFIG_CFS_BANDWIDTH=y
CONFIG_RT_GROUP_SCHED=y
CONFIG_BLK_CGROUP=y
# CONFIG_DEBUG_BLK_CGROUP is not set
# CONFIG_CHECKPOINT_RESTORE is not set
CONFIG_NAMESPACES=y
CONFIG_UTS_NS=y
CONFIG_IPC_NS=y
CONFIG_PID_NS=y
CONFIG_NET_NS=y
CONFIG_SCHED_AUTOGROUP=y
# CONFIG_SYSFS_DEPRECATED is not set
CONFIG_RELAY=y
CONFIG_BLK_DEV_INITRD=y
CONFIG_INITRAMFS_SOURCE=""
CONFIG_RD_GZIP=y
CONFIG_RD_BZIP2=y
CONFIG_RD_LZMA=y
CONFIG_RD_XZ=y
CONFIG_RD_LZO=y
# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
CONFIG_SYSCTL=y
CONFIG_ANON_INODES=y
# CONFIG_EXPERT is not set
CONFIG_HAVE_UID16=y
CONFIG_UID16=y
# CONFIG_SYSCTL_SYSCALL is not set
CONFIG_SYSCTL_EXCEPTION_TRACE=y
CONFIG_KALLSYMS=y
CONFIG_KALLSYMS_ALL=y
CONFIG_HOTPLUG=y
CONFIG_PRINTK=y
CONFIG_BUG=y
CONFIG_ELF_CORE=y
CONFIG_PCSPKR_PLATFORM=y
CONFIG_HAVE_PCSPKR_PLATFORM=y
CONFIG_BASE_FULL=y
CONFIG_FUTEX=y
CONFIG_EPOLL=y
CONFIG_SIGNALFD=y
CONFIG_TIMERFD=y
CONFIG_EVENTFD=y
CONFIG_SHMEM=y
CONFIG_AIO=y
# CONFIG_EMBEDDED is not set
CONFIG_HAVE_PERF_EVENTS=y

#
# Kernel Performance Events And Counters
#
CONFIG_PERF_EVENTS=y
# CONFIG_DEBUG_PERF_USE_VMALLOC is not set
CONFIG_VM_EVENT_COUNTERS=y
CONFIG_PCI_QUIRKS=y
# CONFIG_COMPAT_BRK is not set
CONFIG_SLAB=y
# CONFIG_SLUB is not set
CONFIG_PROFILING=y
CONFIG_TRACEPOINTS=y
CONFIG_OPROFILE=m
# CONFIG_OPROFILE_EVENT_MULTIPLEX is not set
CONFIG_HAVE_OPROFILE=y
CONFIG_OPROFILE_NMI_TIMER=y
CONFIG_KPROBES=y
CONFIG_JUMP_LABEL=y
CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS=y
CONFIG_KRETPROBES=y
CONFIG_USER_RETURN_NOTIFIER=y
CONFIG_HAVE_IOREMAP_PROT=y
CONFIG_HAVE_KPROBES=y
CONFIG_HAVE_KRETPROBES=y
CONFIG_HAVE_OPTPROBES=y
CONFIG_HAVE_ARCH_TRACEHOOK=y
CONFIG_HAVE_DMA_ATTRS=y
CONFIG_USE_GENERIC_SMP_HELPERS=y
CONFIG_GENERIC_SMP_IDLE_THREAD=y
CONFIG_HAVE_REGS_AND_STACK_ACCESS_API=y
CONFIG_HAVE_DMA_API_DEBUG=y
CONFIG_HAVE_HW_BREAKPOINT=y
CONFIG_HAVE_MIXED_BREAKPOINTS_REGS=y
CONFIG_HAVE_USER_RETURN_NOTIFIER=y
CONFIG_HAVE_PERF_EVENTS_NMI=y
CONFIG_HAVE_PERF_REGS=y
CONFIG_HAVE_PERF_USER_STACK_DUMP=y
CONFIG_HAVE_ARCH_JUMP_LABEL=y
CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG=y
CONFIG_HAVE_CMPXCHG_LOCAL=y
CONFIG_HAVE_CMPXCHG_DOUBLE=y
CONFIG_ARCH_WANT_COMPAT_IPC_PARSE_VERSION=y
CONFIG_ARCH_WANT_OLD_COMPAT_IPC=y
CONFIG_GENERIC_KERNEL_THREAD=y
CONFIG_GENERIC_KERNEL_EXECVE=y
CONFIG_HAVE_ARCH_SECCOMP_FILTER=y
CONFIG_SECCOMP_FILTER=y
CONFIG_HAVE_RCU_USER_QS=y
CONFIG_HAVE_IRQ_TIME_ACCOUNTING=y
CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE=y
CONFIG_MODULES_USE_ELF_RELA=y

#
# GCOV-based kernel profiling
#
# CONFIG_GCOV_KERNEL is not set
# CONFIG_HAVE_GENERIC_DMA_COHERENT is not set
CONFIG_SLABINFO=y
CONFIG_RT_MUTEXES=y
CONFIG_BASE_SMALL=0
CONFIG_MODULES=y
CONFIG_MODULE_FORCE_LOAD=y
CONFIG_MODULE_UNLOAD=y
CONFIG_MODULE_FORCE_UNLOAD=y
CONFIG_MODVERSIONS=y
CONFIG_MODULE_SRCVERSION_ALL=y
# CONFIG_MODULE_SIG is not set
CONFIG_STOP_MACHINE=y
CONFIG_BLOCK=y
CONFIG_BLK_DEV_BSG=y
CONFIG_BLK_DEV_BSGLIB=y
CONFIG_BLK_DEV_INTEGRITY=y
CONFIG_BLK_DEV_THROTTLING=y

#
# Partition Types
#
CONFIG_PARTITION_ADVANCED=y
# CONFIG_ACORN_PARTITION is not set
CONFIG_OSF_PARTITION=y
# CONFIG_AMIGA_PARTITION is not set
CONFIG_ATARI_PARTITION=y
CONFIG_MAC_PARTITION=y
CONFIG_MSDOS_PARTITION=y
CONFIG_BSD_DISKLABEL=y
# CONFIG_MINIX_SUBPARTITION is not set
CONFIG_SOLARIS_X86_PARTITION=y
CONFIG_UNIXWARE_DISKLABEL=y
CONFIG_LDM_PARTITION=y
# CONFIG_LDM_DEBUG is not set
CONFIG_SGI_PARTITION=y
CONFIG_ULTRIX_PARTITION=y
CONFIG_SUN_PARTITION=y
CONFIG_KARMA_PARTITION=y
CONFIG_EFI_PARTITION=y
CONFIG_SYSV68_PARTITION=y
CONFIG_BLOCK_COMPAT=y

#
# IO Schedulers
#
CONFIG_IOSCHED_NOOP=y
CONFIG_IOSCHED_DEADLINE=y
CONFIG_IOSCHED_CFQ=y
CONFIG_CFQ_GROUP_IOSCHED=y
# CONFIG_DEFAULT_DEADLINE is not set
CONFIG_DEFAULT_CFQ=y
# CONFIG_DEFAULT_NOOP is not set
CONFIG_DEFAULT_IOSCHED="cfq"
CONFIG_PREEMPT_NOTIFIERS=y
CONFIG_PADATA=y
CONFIG_UNINLINE_SPIN_UNLOCK=y
CONFIG_MUTEX_SPIN_ON_OWNER=y
CONFIG_FREEZER=y

#
# Processor type and features
#
CONFIG_ZONE_DMA=y
CONFIG_SMP=y
CONFIG_X86_X2APIC=y
CONFIG_X86_MPPARSE=y CONFIG_X86_EXTENDED_PLATFORM=y
# CONFIG_X86_NUMACHIP is not set
# CONFIG_X86_VSMP is not set
CONFIG_X86_UV=y CONFIG_X86_SUPPORTS_MEMORY_FAILURE=y CONFIG_SCHED_OMIT_FRAME_POINTER=y CONFIG_PARAVIRT_GUEST=y
# CONFIG_PARAVIRT_TIME_ACCOUNTING is not set # CONFIG_XEN is not set
# CONFIG_XEN_PRIVILEGED_GUEST is not set CONFIG_KVM_GUEST=y
CONFIG_PARAVIRT=y
# CONFIG_PARAVIRT_SPINLOCKS is not set CONFIG_PARAVIRT_CLOCK=y
# CONFIG_PARAVIRT_DEBUG is not set CONFIG_NO_BOOTMEM=y
CONFIG_MEMTEST=y
# CONFIG_MK8 is not set
# CONFIG_MPSC is not set
# CONFIG_MCORE2 is not set
# CONFIG_MATOM is not set CONFIG_GENERIC_CPU=y CONFIG_X86_INTERNODE_CACHE_SHIFT=6 CONFIG_X86_CMPXCHG=y CONFIG_X86_L1_CACHE_SHIFT=6 CONFIG_X86_XADD=y
CONFIG_X86_WP_WORKS_OK=y
CONFIG_X86_TSC=y
CONFIG_X86_CMPXCHG64=y
CONFIG_X86_CMOV=y CONFIG_X86_MINIMUM_CPU_FAMILY=64 CONFIG_X86_DEBUGCTLMSR=y CONFIG_CPU_SUP_INTEL=y
CONFIG_CPU_SUP_AMD=y CONFIG_CPU_SUP_CENTAUR=y
CONFIG_HPET_TIMER=y CONFIG_HPET_EMULATE_RTC=y
CONFIG_DMI=y
CONFIG_GART_IOMMU=y
CONFIG_CALGARY_IOMMU=y
# CONFIG_CALGARY_IOMMU_ENABLED_BY_DEFAULT is not set CONFIG_SWIOTLB=y
CONFIG_IOMMU_HELPER=y
# CONFIG_MAXSMP is not set CONFIG_NR_CPUS=512
CONFIG_SCHED_SMT=y
CONFIG_SCHED_MC=y
# CONFIG_PREEMPT_NONE is not set
# CONFIG_PREEMPT_VOLUNTARY is not set CONFIG_PREEMPT=y
CONFIG_PREEMPT_COUNT=y CONFIG_X86_LOCAL_APIC=y
CONFIG_X86_IO_APIC=y CONFIG_X86_REROUTE_FOR_BROKEN_BOOT_IRQS=y CONFIG_X86_MCE=y
CONFIG_X86_MCE_INTEL=y
CONFIG_X86_MCE_AMD=y CONFIG_X86_MCE_THRESHOLD=y CONFIG_X86_MCE_INJECT=m CONFIG_X86_THERMAL_VECTOR=y
CONFIG_I8K=m
CONFIG_MICROCODE=m
CONFIG_MICROCODE_INTEL=y CONFIG_MICROCODE_AMD=y CONFIG_MICROCODE_OLD_INTERFACE=y CONFIG_X86_MSR=y
CONFIG_X86_CPUID=m CONFIG_ARCH_PHYS_ADDR_T_64BIT=y CONFIG_ARCH_DMA_ADDR_T_64BIT=y CONFIG_DIRECT_GBPAGES=y
CONFIG_NUMA=y
CONFIG_AMD_NUMA=y
CONFIG_X86_64_ACPI_NUMA=y CONFIG_NODES_SPAN_OTHER_NODES=y CONFIG_NUMA_EMU=y
CONFIG_NODES_SHIFT=9 CONFIG_ARCH_SPARSEMEM_ENABLE=y CONFIG_ARCH_SPARSEMEM_DEFAULT=y CONFIG_ARCH_SELECT_MEMORY_MODEL=y CONFIG_ARCH_MEMORY_PROBE=y CONFIG_ARCH_PROC_KCORE_TEXT=y CONFIG_ILLEGAL_POINTER_VALUE=0xdead000000000000 CONFIG_SELECT_MEMORY_MODEL=y CONFIG_SPARSEMEM_MANUAL=y
CONFIG_SPARSEMEM=y CONFIG_NEED_MULTIPLE_NODES=y CONFIG_HAVE_MEMORY_PRESENT=y CONFIG_SPARSEMEM_EXTREME=y CONFIG_SPARSEMEM_VMEMMAP_ENABLE=y CONFIG_SPARSEMEM_ALLOC_MEM_MAP_TOGETHER=y CONFIG_SPARSEMEM_VMEMMAP=y CONFIG_HAVE_MEMBLOCK=y CONFIG_HAVE_MEMBLOCK_NODE_MAP=y CONFIG_ARCH_DISCARD_MEMBLOCK=y CONFIG_MEMORY_ISOLATION=y CONFIG_MEMORY_HOTPLUG=y CONFIG_MEMORY_HOTPLUG_SPARSE=y CONFIG_MEMORY_HOTREMOVE=y CONFIG_PAGEFLAGS_EXTENDED=y CONFIG_SPLIT_PTLOCK_CPUS=4 CONFIG_COMPACTION=y
CONFIG_MIGRATION=y CONFIG_PHYS_ADDR_T_64BIT=y CONFIG_ZONE_DMA_FLAG=1
CONFIG_BOUNCE=y
CONFIG_VIRT_TO_BUS=y
CONFIG_MMU_NOTIFIER=y
CONFIG_KSM=y CONFIG_DEFAULT_MMAP_MIN_ADDR=65536 CONFIG_ARCH_SUPPORTS_MEMORY_FAILURE=y CONFIG_MEMORY_FAILURE=y CONFIG_HWPOISON_INJECT=m CONFIG_TRANSPARENT_HUGEPAGE=y CONFIG_TRANSPARENT_HUGEPAGE_ALWAYS=y
# CONFIG_TRANSPARENT_HUGEPAGE_MADVISE is not set CONFIG_CROSS_MEMORY_ATTACH=y CONFIG_CLEANCACHE=y
# CONFIG_FRONTSWAP is not set CONFIG_X86_CHECK_BIOS_CORRUPTION=y CONFIG_X86_BOOTPARAM_MEMORY_CORRUPTION_CHECK=y CONFIG_X86_RESERVE_LOW=64
CONFIG_MTRR=y
CONFIG_MTRR_SANITIZER=y CONFIG_MTRR_SANITIZER_ENABLE_DEFAULT=0 CONFIG_MTRR_SANITIZER_SPARE_REG_NR_DEFAULT=1 CONFIG_X86_PAT=y CONFIG_ARCH_USES_PG_UNCACHED=y CONFIG_ARCH_RANDOM=y
CONFIG_X86_SMAP=y
CONFIG_EFI=y
CONFIG_EFI_STUB=y
CONFIG_SECCOMP=y
# CONFIG_CC_STACKPROTECTOR is not set
# CONFIG_HZ_100 is not set
# CONFIG_HZ_250 is not set
# CONFIG_HZ_300 is not set
CONFIG_HZ_1000=y
CONFIG_HZ=1000
CONFIG_SCHED_HRTICK=y
CONFIG_KEXEC=y
CONFIG_CRASH_DUMP=y
# CONFIG_KEXEC_JUMP is not set CONFIG_PHYSICAL_START=0x200000 CONFIG_RELOCATABLE=y CONFIG_PHYSICAL_ALIGN=0x1000000 CONFIG_HOTPLUG_CPU=y
# CONFIG_COMPAT_VDSO is not set
# CONFIG_CMDLINE_BOOL is not set CONFIG_ARCH_ENABLE_MEMORY_HOTPLUG=y CONFIG_ARCH_ENABLE_MEMORY_HOTREMOVE=y CONFIG_USE_PERCPU_NUMA_NODE_ID=y

#
# Power management and ACPI options
#
CONFIG_ARCH_HIBERNATION_HEADER=y
CONFIG_SUSPEND=y
CONFIG_SUSPEND_FREEZER=y
CONFIG_HIBERNATE_CALLBACKS=y
CONFIG_HIBERNATION=y
CONFIG_PM_STD_PARTITION=""
CONFIG_PM_SLEEP=y
CONFIG_PM_SLEEP_SMP=y
# CONFIG_PM_AUTOSLEEP is not set
# CONFIG_PM_WAKELOCKS is not set
CONFIG_PM_RUNTIME=y
CONFIG_PM=y
CONFIG_PM_DEBUG=y
CONFIG_PM_ADVANCED_DEBUG=y
# CONFIG_PM_TEST_SUSPEND is not set
CONFIG_PM_SLEEP_DEBUG=y
CONFIG_PM_TRACE=y
CONFIG_PM_TRACE_RTC=y
CONFIG_ACPI=y
CONFIG_ACPI_SLEEP=y
CONFIG_ACPI_PROCFS=y
CONFIG_ACPI_PROCFS_POWER=y
CONFIG_ACPI_EC_DEBUGFS=m
CONFIG_ACPI_PROC_EVENT=y
CONFIG_ACPI_AC=m
CONFIG_ACPI_BATTERY=m
CONFIG_ACPI_BUTTON=m
CONFIG_ACPI_VIDEO=m
CONFIG_ACPI_FAN=m
CONFIG_ACPI_DOCK=y
CONFIG_ACPI_PROCESSOR=m
CONFIG_ACPI_IPMI=m
CONFIG_ACPI_HOTPLUG_CPU=y
CONFIG_ACPI_PROCESSOR_AGGREGATOR=m
CONFIG_ACPI_THERMAL=m
CONFIG_ACPI_NUMA=y
CONFIG_ACPI_CUSTOM_DSDT_FILE=""
# CONFIG_ACPI_CUSTOM_DSDT is not set
CONFIG_ACPI_BLACKLIST_YEAR=0
CONFIG_ACPI_DEBUG=y
# CONFIG_ACPI_DEBUG_FUNC_TRACE is not set
CONFIG_ACPI_PCI_SLOT=m
CONFIG_X86_PM_TIMER=y
CONFIG_ACPI_CONTAINER=m
CONFIG_ACPI_HOTPLUG_MEMORY=m
CONFIG_ACPI_SBS=m
CONFIG_ACPI_HED=y
# CONFIG_ACPI_CUSTOM_METHOD is not set
# CONFIG_ACPI_BGRT is not set
CONFIG_ACPI_APEI=y
CONFIG_ACPI_APEI_GHES=y
CONFIG_ACPI_APEI_PCIEAER=y
CONFIG_ACPI_APEI_MEMORY_FAILURE=y
CONFIG_ACPI_APEI_EINJ=m
# CONFIG_ACPI_APEI_ERST_DEBUG is not set
# CONFIG_SFI is not set

#
# CPU Frequency scaling
#
CONFIG_CPU_FREQ=y
CONFIG_CPU_FREQ_TABLE=y CONFIG_CPU_FREQ_STAT=m CONFIG_CPU_FREQ_STAT_DETAILS=y
# CONFIG_CPU_FREQ_DEFAULT_GOV_PERFORMANCE is not set # CONFIG_CPU_FREQ_DEFAULT_GOV_USERSPACE is not set CONFIG_CPU_FREQ_DEFAULT_GOV_ONDEMAND=y
# CONFIG_CPU_FREQ_DEFAULT_GOV_CONSERVATIVE is not set CONFIG_CPU_FREQ_GOV_PERFORMANCE=y CONFIG_CPU_FREQ_GOV_POWERSAVE=m CONFIG_CPU_FREQ_GOV_USERSPACE=m CONFIG_CPU_FREQ_GOV_ONDEMAND=y CONFIG_CPU_FREQ_GOV_CONSERVATIVE=m

#

# x86 CPU frequency scaling drivers
#
# CONFIG_X86_PCC_CPUFREQ is not set
CONFIG_X86_ACPI_CPUFREQ=m
CONFIG_X86_ACPI_CPUFREQ_CPB=y
CONFIG_X86_POWERNOW_K8=m
# CONFIG_X86_SPEEDSTEP_CENTRINO is not set
# CONFIG_X86_P4_CLOCKMOD is not set

#

# shared options
#
# CONFIG_X86_SPEEDSTEP_LIB is not set
CONFIG_CPU_IDLE=y
CONFIG_CPU_IDLE_GOV_LADDER=y
CONFIG_CPU_IDLE_GOV_MENU=y
# CONFIG_ARCH_NEEDS_CPU_IDLE_COUPLED is not set
CONFIG_INTEL_IDLE=y

#
# Memory power savings
#
CONFIG_I7300_IDLE_IOAT_CHANNEL=y
CONFIG_I7300_IDLE=m

#
# Bus options (PCI etc.)
#
CONFIG_PCI=y
CONFIG_PCI_DIRECT=y
CONFIG_PCI_MMCONFIG=y
CONFIG_PCI_DOMAINS=y
CONFIG_PCIEPORTBUS=y
CONFIG_HOTPLUG_PCI_PCIE=m
CONFIG_PCIEAER=y
# CONFIG_PCIE_ECRC is not set
CONFIG_PCIEAER_INJECT=m
CONFIG_PCIEASPM=y
# CONFIG_PCIEASPM_DEBUG is not set
CONFIG_PCIEASPM_DEFAULT=y
# CONFIG_PCIEASPM_POWERSAVE is not set
# CONFIG_PCIEASPM_PERFORMANCE is not set
CONFIG_PCIE_PME=y
CONFIG_ARCH_SUPPORTS_MSI=y
CONFIG_PCI_MSI=y
# CONFIG_PCI_DEBUG is not set
# CONFIG_PCI_REALLOC_ENABLE_AUTO is not set
CONFIG_PCI_STUB=y
CONFIG_HT_IRQ=y
CONFIG_PCI_ATS=y
CONFIG_PCI_IOV=y
CONFIG_PCI_PRI=y
CONFIG_PCI_PASID=y
CONFIG_PCI_IOAPIC=m
CONFIG_PCI_LABEL=y
CONFIG_ISA_DMA_API=y
CONFIG_AMD_NB=y
# CONFIG_PCCARD is not set
CONFIG_HOTPLUG_PCI=m
CONFIG_HOTPLUG_PCI_ACPI=m
CONFIG_HOTPLUG_PCI_ACPI_IBM=m
CONFIG_HOTPLUG_PCI_CPCI=y
CONFIG_HOTPLUG_PCI_CPCI_ZT5550=m
CONFIG_HOTPLUG_PCI_CPCI_GENERIC=m
CONFIG_HOTPLUG_PCI_SHPC=m
# CONFIG_RAPIDIO is not set

#
# Executable file formats / Emulations
#
CONFIG_BINFMT_ELF=y
CONFIG_COMPAT_BINFMT_ELF=y
CONFIG_ARCH_BINFMT_ELF_RANDOMIZE_PIE=y
CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS=y
# CONFIG_HAVE_AOUT is not set
CONFIG_BINFMT_MISC=m
CONFIG_COREDUMP=y
CONFIG_IA32_EMULATION=y
CONFIG_IA32_AOUT=m
CONFIG_X86_X32=y
CONFIG_COMPAT=y
CONFIG_COMPAT_FOR_U64_ALIGNMENT=y
CONFIG_SYSVIPC_COMPAT=y
CONFIG_KEYS_COMPAT=y
CONFIG_HAVE_TEXT_POKE_SMP=y
CONFIG_X86_DEV_DMA_OPS=y
CONFIG_NET=y

#
# Networking options
#
CONFIG_PACKET=m
# CONFIG_PACKET_DIAG is not set
CONFIG_UNIX=y
CONFIG_UNIX_DIAG=m
CONFIG_XFRM=y
CONFIG_XFRM_ALGO=m
CONFIG_XFRM_USER=m
CONFIG_XFRM_SUB_POLICY=y
CONFIG_XFRM_MIGRATE=y
# CONFIG_XFRM_STATISTICS is not set
CONFIG_XFRM_IPCOMP=m
CONFIG_NET_KEY=m
CONFIG_NET_KEY_MIGRATE=y
CONFIG_INET=y
CONFIG_IP_MULTICAST=y
CONFIG_IP_ADVANCED_ROUTER=y
# CONFIG_IP_FIB_TRIE_STATS is not set
CONFIG_IP_MULTIPLE_TABLES=y
CONFIG_IP_ROUTE_MULTIPATH=y
CONFIG_IP_ROUTE_VERBOSE=y
CONFIG_IP_ROUTE_CLASSID=y
CONFIG_IP_PNP=y
CONFIG_IP_PNP_DHCP=y
CONFIG_IP_PNP_BOOTP=y
CONFIG_IP_PNP_RARP=y
CONFIG_NET_IPIP=m
CONFIG_NET_IPGRE_DEMUX=m
CONFIG_NET_IPGRE=m
CONFIG_NET_IPGRE_BROADCAST=y
CONFIG_IP_MROUTE=y
CONFIG_IP_MROUTE_MULTIPLE_TABLES=y
CONFIG_IP_PIMSM_V1=y
CONFIG_IP_PIMSM_V2=y
# CONFIG_ARPD is not set
CONFIG_SYN_COOKIES=y
# CONFIG_NET_IPVTI is not set
CONFIG_INET_AH=m
CONFIG_INET_ESP=m
CONFIG_INET_IPCOMP=m
CONFIG_INET_XFRM_TUNNEL=m
CONFIG_INET_TUNNEL=m
CONFIG_INET_XFRM_MODE_TRANSPORT=m
CONFIG_INET_XFRM_MODE_TUNNEL=m
CONFIG_INET_XFRM_MODE_BEET=m
CONFIG_INET_LRO=y
CONFIG_INET_DIAG=m
CONFIG_INET_TCP_DIAG=m
CONFIG_INET_UDP_DIAG=m
CONFIG_TCP_CONG_ADVANCED=y
CONFIG_TCP_CONG_BIC=m
CONFIG_TCP_CONG_CUBIC=y
CONFIG_TCP_CONG_WESTWOOD=m
CONFIG_TCP_CONG_HTCP=m
CONFIG_TCP_CONG_HSTCP=m
CONFIG_TCP_CONG_HYBLA=m
CONFIG_TCP_CONG_VEGAS=m
CONFIG_TCP_CONG_SCALABLE=m
CONFIG_TCP_CONG_LP=m
CONFIG_TCP_CONG_VENO=m
CONFIG_TCP_CONG_YEAH=m
CONFIG_TCP_CONG_ILLINOIS=m
CONFIG_DEFAULT_CUBIC=y
# CONFIG_DEFAULT_RENO is not set
CONFIG_DEFAULT_TCP_CONG="cubic"
# CONFIG_TCP_MD5SIG is not set
CONFIG_IPV6=y
CONFIG_IPV6_PRIVACY=y
CONFIG_IPV6_ROUTER_PREF=y
CONFIG_IPV6_ROUTE_INFO=y
# CONFIG_IPV6_OPTIMISTIC_DAD is not set
CONFIG_INET6_AH=m
CONFIG_INET6_ESP=m
CONFIG_INET6_IPCOMP=m
CONFIG_IPV6_MIP6=m
CONFIG_INET6_XFRM_TUNNEL=m
CONFIG_INET6_TUNNEL=m
CONFIG_INET6_XFRM_MODE_TRANSPORT=m
CONFIG_INET6_XFRM_MODE_TUNNEL=m
CONFIG_INET6_XFRM_MODE_BEET=m
CONFIG_INET6_XFRM_MODE_ROUTEOPTIMIZATION=m
CONFIG_IPV6_SIT=m
# CONFIG_IPV6_SIT_6RD is not set
CONFIG_IPV6_NDISC_NODETYPE=y
CONFIG_IPV6_TUNNEL=m
# CONFIG_IPV6_GRE is not set
CONFIG_IPV6_MULTIPLE_TABLES=y
CONFIG_IPV6_SUBTREES=y
# CONFIG_IPV6_MROUTE is not set
CONFIG_NETLABEL=y
CONFIG_NETWORK_SECMARK=y
# CONFIG_NETWORK_PHY_TIMESTAMPING is not set
CONFIG_NETFILTER=y
# CONFIG_NETFILTER_DEBUG is not set
CONFIG_NETFILTER_ADVANCED=y
CONFIG_BRIDGE_NETFILTER=y

#
# Core Netfilter Configuration
#
CONFIG_NETFILTER_NETLINK=m
CONFIG_NETFILTER_NETLINK_ACCT=m
CONFIG_NETFILTER_NETLINK_QUEUE=m
CONFIG_NETFILTER_NETLINK_LOG=m
CONFIG_NF_CONNTRACK=m
CONFIG_NF_CONNTRACK_MARK=y
CONFIG_NF_CONNTRACK_SECMARK=y
CONFIG_NF_CONNTRACK_ZONES=y
CONFIG_NF_CONNTRACK_PROCFS=y
CONFIG_NF_CONNTRACK_EVENTS=y
CONFIG_NF_CONNTRACK_TIMEOUT=y
CONFIG_NF_CONNTRACK_TIMESTAMP=y
CONFIG_NF_CT_PROTO_DCCP=m
CONFIG_NF_CT_PROTO_GRE=m
CONFIG_NF_CT_PROTO_SCTP=m
CONFIG_NF_CT_PROTO_UDPLITE=m
CONFIG_NF_CONNTRACK_AMANDA=m
CONFIG_NF_CONNTRACK_FTP=m
CONFIG_NF_CONNTRACK_H323=m
CONFIG_NF_CONNTRACK_IRC=m
CONFIG_NF_CONNTRACK_BROADCAST=m
CONFIG_NF_CONNTRACK_NETBIOS_NS=m
CONFIG_NF_CONNTRACK_SNMP=m
CONFIG_NF_CONNTRACK_PPTP=m
CONFIG_NF_CONNTRACK_SANE=m
CONFIG_NF_CONNTRACK_SIP=m
CONFIG_NF_CONNTRACK_TFTP=m
CONFIG_NF_CT_NETLINK=m
CONFIG_NF_CT_NETLINK_TIMEOUT=m
# CONFIG_NETFILTER_NETLINK_QUEUE_CT is not set
CONFIG_NETFILTER_TPROXY=m
CONFIG_NETFILTER_XTABLES=m

#
# Xtables combined modules
#
CONFIG_NETFILTER_XT_MARK=m
CONFIG_NETFILTER_XT_CONNMARK=m
CONFIG_NETFILTER_XT_SET=m

#
# Xtables targets
#
CONFIG_NETFILTER_XT_TARGET_AUDIT=m
CONFIG_NETFILTER_XT_TARGET_CHECKSUM=m
CONFIG_NETFILTER_XT_TARGET_CLASSIFY=m
CONFIG_NETFILTER_XT_TARGET_CONNMARK=m
CONFIG_NETFILTER_XT_TARGET_CONNSECMARK=m
CONFIG_NETFILTER_XT_TARGET_CT=m
CONFIG_NETFILTER_XT_TARGET_DSCP=m
CONFIG_NETFILTER_XT_TARGET_HL=m
# CONFIG_NETFILTER_XT_TARGET_HMARK is not set
CONFIG_NETFILTER_XT_TARGET_IDLETIMER=m
CONFIG_NETFILTER_XT_TARGET_LED=m
CONFIG_NETFILTER_XT_TARGET_LOG=m
CONFIG_NETFILTER_XT_TARGET_MARK=m
CONFIG_NETFILTER_XT_TARGET_NFLOG=m
CONFIG_NETFILTER_XT_TARGET_NFQUEUE=m
CONFIG_NETFILTER_XT_TARGET_RATEEST=m
CONFIG_NETFILTER_XT_TARGET_TEE=m
CONFIG_NETFILTER_XT_TARGET_TPROXY=m
CONFIG_NETFILTER_XT_TARGET_TRACE=m
CONFIG_NETFILTER_XT_TARGET_SECMARK=m
CONFIG_NETFILTER_XT_TARGET_TCPMSS=m
CONFIG_NETFILTER_XT_TARGET_TCPOPTSTRIP=m

#
# Xtables matches
#
CONFIG_NETFILTER_XT_MATCH_ADDRTYPE=m
CONFIG_NETFILTER_XT_MATCH_CLUSTER=m
CONFIG_NETFILTER_XT_MATCH_COMMENT=m
CONFIG_NETFILTER_XT_MATCH_CONNBYTES=m
CONFIG_NETFILTER_XT_MATCH_CONNLIMIT=m
CONFIG_NETFILTER_XT_MATCH_CONNMARK=m
CONFIG_NETFILTER_XT_MATCH_CONNTRACK=m
CONFIG_NETFILTER_XT_MATCH_CPU=m
CONFIG_NETFILTER_XT_MATCH_DCCP=m
CONFIG_NETFILTER_XT_MATCH_DEVGROUP=m
CONFIG_NETFILTER_XT_MATCH_DSCP=m
CONFIG_NETFILTER_XT_MATCH_ECN=m
CONFIG_NETFILTER_XT_MATCH_ESP=m
CONFIG_NETFILTER_XT_MATCH_HASHLIMIT=m
CONFIG_NETFILTER_XT_MATCH_HELPER=m
CONFIG_NETFILTER_XT_MATCH_HL=m
CONFIG_NETFILTER_XT_MATCH_IPRANGE=m
CONFIG_NETFILTER_XT_MATCH_IPVS=m
CONFIG_NETFILTER_XT_MATCH_LENGTH=m
CONFIG_NETFILTER_XT_MATCH_LIMIT=m
CONFIG_NETFILTER_XT_MATCH_MAC=m
CONFIG_NETFILTER_XT_MATCH_MARK=m
CONFIG_NETFILTER_XT_MATCH_MULTIPORT=m
CONFIG_NETFILTER_XT_MATCH_NFACCT=m
CONFIG_NETFILTER_XT_MATCH_OSF=m
CONFIG_NETFILTER_XT_MATCH_OWNER=m
CONFIG_NETFILTER_XT_MATCH_POLICY=m
CONFIG_NETFILTER_XT_MATCH_PHYSDEV=m
CONFIG_NETFILTER_XT_MATCH_PKTTYPE=m
CONFIG_NETFILTER_XT_MATCH_QUOTA=m
CONFIG_NETFILTER_XT_MATCH_RATEEST=m
CONFIG_NETFILTER_XT_MATCH_REALM=m
CONFIG_NETFILTER_XT_MATCH_RECENT=m
CONFIG_NETFILTER_XT_MATCH_SCTP=m
CONFIG_NETFILTER_XT_MATCH_SOCKET=m
CONFIG_NETFILTER_XT_MATCH_STATE=m
CONFIG_NETFILTER_XT_MATCH_STATISTIC=m
CONFIG_NETFILTER_XT_MATCH_STRING=m
CONFIG_NETFILTER_XT_MATCH_TCPMSS=m
CONFIG_NETFILTER_XT_MATCH_TIME=m
CONFIG_NETFILTER_XT_MATCH_U32=m
CONFIG_IP_SET=m
CONFIG_IP_SET_MAX=256
CONFIG_IP_SET_BITMAP_IP=m
CONFIG_IP_SET_BITMAP_IPMAC=m
CONFIG_IP_SET_BITMAP_PORT=m
CONFIG_IP_SET_HASH_IP=m
CONFIG_IP_SET_HASH_IPPORT=m
CONFIG_IP_SET_HASH_IPPORTIP=m
CONFIG_IP_SET_HASH_IPPORTNET=m
CONFIG_IP_SET_HASH_NET=m
CONFIG_IP_SET_HASH_NETPORT=m
CONFIG_IP_SET_HASH_NETIFACE=m
CONFIG_IP_SET_LIST_SET=m
CONFIG_IP_VS=m
CONFIG_IP_VS_IPV6=y
# CONFIG_IP_VS_DEBUG is not set
CONFIG_IP_VS_TAB_BITS=12

#
# IPVS transport protocol load balancing support
#
CONFIG_IP_VS_PROTO_TCP=y
CONFIG_IP_VS_PROTO_UDP=y
CONFIG_IP_VS_PROTO_AH_ESP=y
CONFIG_IP_VS_PROTO_ESP=y
CONFIG_IP_VS_PROTO_AH=y
CONFIG_IP_VS_PROTO_SCTP=y

#
# IPVS scheduler
#
CONFIG_IP_VS_RR=m
CONFIG_IP_VS_WRR=m
CONFIG_IP_VS_LC=m
CONFIG_IP_VS_WLC=m
CONFIG_IP_VS_LBLC=m
CONFIG_IP_VS_LBLCR=m
CONFIG_IP_VS_DH=m
CONFIG_IP_VS_SH=m
CONFIG_IP_VS_SED=m
CONFIG_IP_VS_NQ=m

#
# IPVS SH scheduler
#
CONFIG_IP_VS_SH_TAB_BITS=8

#
# IPVS application helper
#
CONFIG_IP_VS_NFCT=y
CONFIG_IP_VS_PE_SIP=m

#
# IP: Netfilter Configuration
#
CONFIG_NF_DEFRAG_IPV4=m
CONFIG_NF_CONNTRACK_IPV4=m
# CONFIG_NF_CONNTRACK_PROC_COMPAT is not set
CONFIG_IP_NF_QUEUE=m
CONFIG_IP_NF_IPTABLES=m
CONFIG_IP_NF_MATCH_AH=m
CONFIG_IP_NF_MATCH_ECN=m
CONFIG_IP_NF_MATCH_RPFILTER=m
CONFIG_IP_NF_MATCH_TTL=m
CONFIG_IP_NF_FILTER=m
CONFIG_IP_NF_TARGET_REJECT=m
CONFIG_IP_NF_TARGET_ULOG=m
# CONFIG_NF_NAT_IPV4 is not set
CONFIG_IP_NF_MANGLE=m
CONFIG_IP_NF_TARGET_CLUSTERIP=m
CONFIG_IP_NF_TARGET_ECN=m
CONFIG_IP_NF_TARGET_TTL=m
CONFIG_IP_NF_RAW=m
CONFIG_IP_NF_SECURITY=m
CONFIG_IP_NF_ARPTABLES=m
CONFIG_IP_NF_ARPFILTER=m
CONFIG_IP_NF_ARP_MANGLE=m

#
# IPv6: Netfilter Configuration
#
CONFIG_NF_DEFRAG_IPV6=m
CONFIG_NF_CONNTRACK_IPV6=m
CONFIG_IP6_NF_IPTABLES=m
CONFIG_IP6_NF_MATCH_AH=m
CONFIG_IP6_NF_MATCH_EUI64=m
CONFIG_IP6_NF_MATCH_FRAG=m
CONFIG_IP6_NF_MATCH_OPTS=m
CONFIG_IP6_NF_MATCH_HL=m
CONFIG_IP6_NF_MATCH_IPV6HEADER=m
CONFIG_IP6_NF_MATCH_MH=m
CONFIG_IP6_NF_MATCH_RPFILTER=m
CONFIG_IP6_NF_MATCH_RT=m
CONFIG_IP6_NF_TARGET_HL=m
CONFIG_IP6_NF_FILTER=m
CONFIG_IP6_NF_TARGET_REJECT=m
CONFIG_IP6_NF_MANGLE=m
CONFIG_IP6_NF_RAW=m
CONFIG_IP6_NF_SECURITY=m
# CONFIG_NF_NAT_IPV6 is not set
CONFIG_BRIDGE_NF_EBTABLES=m
CONFIG_BRIDGE_EBT_BROUTE=m
CONFIG_BRIDGE_EBT_T_FILTER=m
CONFIG_BRIDGE_EBT_T_NAT=m
CONFIG_BRIDGE_EBT_802_3=m
CONFIG_BRIDGE_EBT_AMONG=m
CONFIG_BRIDGE_EBT_ARP=m
CONFIG_BRIDGE_EBT_IP=m
CONFIG_BRIDGE_EBT_IP6=m
CONFIG_BRIDGE_EBT_LIMIT=m
CONFIG_BRIDGE_EBT_MARK=m
CONFIG_BRIDGE_EBT_PKTTYPE=m
CONFIG_BRIDGE_EBT_STP=m
CONFIG_BRIDGE_EBT_VLAN=m
CONFIG_BRIDGE_EBT_ARPREPLY=m
CONFIG_BRIDGE_EBT_DNAT=m
CONFIG_BRIDGE_EBT_MARK_T=m
CONFIG_BRIDGE_EBT_REDIRECT=m
CONFIG_BRIDGE_EBT_SNAT=m
CONFIG_BRIDGE_EBT_LOG=m
CONFIG_BRIDGE_EBT_ULOG=m
CONFIG_BRIDGE_EBT_NFLOG=m
CONFIG_IP_DCCP=m
CONFIG_INET_DCCP_DIAG=m

#
# DCCP CCIDs Configuration (EXPERIMENTAL)
#
# CONFIG_IP_DCCP_CCID2_DEBUG is not set
CONFIG_IP_DCCP_CCID3=y
# CONFIG_IP_DCCP_CCID3_DEBUG is not set
CONFIG_IP_DCCP_TFRC_LIB=y

#
# DCCP Kernel Hacking
#
# CONFIG_IP_DCCP_DEBUG is not set
# CONFIG_NET_DCCPPROBE is not set
CONFIG_IP_SCTP=m
CONFIG_NET_SCTPPROBE=m
# CONFIG_SCTP_DBG_MSG is not set
# CONFIG_SCTP_DBG_OBJCNT is not set
# CONFIG_SCTP_HMAC_NONE is not set
# CONFIG_SCTP_HMAC_SHA1 is not set
CONFIG_SCTP_HMAC_MD5=y
CONFIG_RDS=m
CONFIG_RDS_TCP=m
# CONFIG_RDS_DEBUG is not set
# CONFIG_TIPC is not set
CONFIG_ATM=m
CONFIG_ATM_CLIP=m
# CONFIG_ATM_CLIP_NO_ICMP is not set
CONFIG_ATM_LANE=m
CONFIG_ATM_MPOA=m
CONFIG_ATM_BR2684=m
# CONFIG_ATM_BR2684_IPFILTER is not set
# CONFIG_L2TP is not set
CONFIG_STP=m
CONFIG_GARP=m
CONFIG_BRIDGE=m
CONFIG_BRIDGE_IGMP_SNOOPING=y
CONFIG_NET_DSA=y
# CONFIG_NET_DSA_TAG_DSA is not set
# CONFIG_NET_DSA_TAG_EDSA is not set
# CONFIG_NET_DSA_TAG_TRAILER is not set
CONFIG_VLAN_8021Q=m
CONFIG_VLAN_8021Q_GVRP=y
# CONFIG_DECNET is not set
CONFIG_LLC=m
CONFIG_LLC2=m
# CONFIG_IPX is not set
# CONFIG_ATALK is not set
# CONFIG_X25 is not set
# CONFIG_LAPB is not set
# CONFIG_WAN_ROUTER is not set
# CONFIG_PHONET is not set
CONFIG_IEEE802154=m
CONFIG_IEEE802154_6LOWPAN=m
# CONFIG_MAC802154 is not set
CONFIG_NET_SCHED=y

#
# Queueing/Scheduling
#
CONFIG_NET_SCH_CBQ=m
CONFIG_NET_SCH_HTB=m
CONFIG_NET_SCH_HFSC=m
CONFIG_NET_SCH_ATM=m
CONFIG_NET_SCH_PRIO=m
CONFIG_NET_SCH_MULTIQ=m
CONFIG_NET_SCH_RED=m
CONFIG_NET_SCH_SFB=m
CONFIG_NET_SCH_SFQ=m
CONFIG_NET_SCH_TEQL=m
CONFIG_NET_SCH_TBF=m
CONFIG_NET_SCH_GRED=m
CONFIG_NET_SCH_DSMARK=m
CONFIG_NET_SCH_NETEM=m
CONFIG_NET_SCH_DRR=m
CONFIG_NET_SCH_MQPRIO=m
CONFIG_NET_SCH_CHOKE=m
CONFIG_NET_SCH_QFQ=m
# CONFIG_NET_SCH_CODEL is not set
# CONFIG_NET_SCH_FQ_CODEL is not set
CONFIG_NET_SCH_INGRESS=m
CONFIG_NET_SCH_PLUG=m

#
# Classification
#
CONFIG_NET_CLS=y
CONFIG_NET_CLS_BASIC=m
CONFIG_NET_CLS_TCINDEX=m
CONFIG_NET_CLS_ROUTE4=m
CONFIG_NET_CLS_FW=m
CONFIG_NET_CLS_U32=m
CONFIG_CLS_U32_PERF=y
CONFIG_CLS_U32_MARK=y
CONFIG_NET_CLS_RSVP=m
CONFIG_NET_CLS_RSVP6=m
CONFIG_NET_CLS_FLOW=m
CONFIG_NET_CLS_CGROUP=y
CONFIG_NET_EMATCH=y
CONFIG_NET_EMATCH_STACK=32
CONFIG_NET_EMATCH_CMP=m
CONFIG_NET_EMATCH_NBYTE=m
CONFIG_NET_EMATCH_U32=m
CONFIG_NET_EMATCH_META=m
CONFIG_NET_EMATCH_TEXT=m
# CONFIG_NET_EMATCH_IPSET is not set
CONFIG_NET_CLS_ACT=y
CONFIG_NET_ACT_POLICE=m
CONFIG_NET_ACT_GACT=m
CONFIG_GACT_PROB=y
CONFIG_NET_ACT_MIRRED=m
CONFIG_NET_ACT_IPT=m
CONFIG_NET_ACT_NAT=m
CONFIG_NET_ACT_PEDIT=m
CONFIG_NET_ACT_SIMP=m
CONFIG_NET_ACT_SKBEDIT=m
CONFIG_NET_ACT_CSUM=m
CONFIG_NET_CLS_IND=y
CONFIG_NET_SCH_FIFO=y
CONFIG_DCB=y
CONFIG_DNS_RESOLVER=y
# CONFIG_BATMAN_ADV is not set
CONFIG_OPENVSWITCH=m
CONFIG_RPS=y
CONFIG_RFS_ACCEL=y
CONFIG_XPS=y
CONFIG_NETPRIO_CGROUP=m
CONFIG_BQL=y
CONFIG_BPF_JIT=y

#
# Network testing
#
CONFIG_NET_PKTGEN=m
CONFIG_NET_TCPPROBE=m
# CONFIG_NET_DROP_MONITOR is not set
# CONFIG_HAMRADIO is not set
# CONFIG_CAN is not set
# CONFIG_IRDA is not set
# CONFIG_BT is not set
# CONFIG_AF_RXRPC is not set
CONFIG_FIB_RULES=y
# CONFIG_WIRELESS is not set
# CONFIG_WIMAX is not set
CONFIG_RFKILL=m
CONFIG_RFKILL_LEDS=y
CONFIG_RFKILL_INPUT=y
# CONFIG_NET_9P is not set
# CONFIG_CAIF is not set
CONFIG_CEPH_LIB=m
CONFIG_CEPH_LIB_PRETTYDEBUG=y
# CONFIG_CEPH_LIB_USE_DNS_RESOLVER is not set
CONFIG_NFC=m
CONFIG_NFC_NCI=m
# CONFIG_NFC_HCI is not set
CONFIG_NFC_LLCP=y

#
# Near Field Communication (NFC) devices
#
CONFIG_NFC_PN533=m
CONFIG_NFC_WILINK=m
CONFIG_HAVE_BPF_JIT=y

#
# Device Drivers
#

#
# Generic Driver Options
#
CONFIG_UEVENT_HELPER_PATH=""
CONFIG_DEVTMPFS=y
CONFIG_DEVTMPFS_MOUNT=y
# CONFIG_STANDALONE is not set
CONFIG_PREVENT_FIRMWARE_BUILD=y
CONFIG_FW_LOADER=y
# CONFIG_FIRMWARE_IN_KERNEL is not set
CONFIG_EXTRA_FIRMWARE=""
# CONFIG_DEBUG_DRIVER is not set
# CONFIG_DEBUG_DEVRES is not set
# CONFIG_SYS_HYPERVISOR is not set
# CONFIG_GENERIC_CPU_DEVICES is not set
CONFIG_REGMAP=y
CONFIG_REGMAP_I2C=m
CONFIG_DMA_SHARED_BUFFER=y

#
# Bus devices
#
# CONFIG_OMAP_OCP2SCP is not set
CONFIG_CONNECTOR=y
CONFIG_PROC_EVENTS=y
# CONFIG_MTD is not set
# CONFIG_PARPORT is not set
CONFIG_PNP=y
# CONFIG_PNP_DEBUG_MESSAGES is not set

#
# Protocols
#
CONFIG_PNPACPI=y
CONFIG_BLK_DEV=y
CONFIG_BLK_DEV_FD=m
CONFIG_BLK_DEV_PCIESSD_MTIP32XX=m
# CONFIG_BLK_CPQ_DA is not set
# CONFIG_BLK_CPQ_CISS_DA is not set
# CONFIG_BLK_DEV_DAC960 is not set
# CONFIG_BLK_DEV_UMEM is not set
# CONFIG_BLK_DEV_COW_COMMON is not set
CONFIG_BLK_DEV_LOOP=m
CONFIG_BLK_DEV_LOOP_MIN_COUNT=8
CONFIG_BLK_DEV_CRYPTOLOOP=m
# CONFIG_BLK_DEV_DRBD is not set
CONFIG_BLK_DEV_NBD=m
# CONFIG_BLK_DEV_NVME is not set
# CONFIG_BLK_DEV_OSD is not set
CONFIG_BLK_DEV_SX8=m
CONFIG_BLK_DEV_RAM=m
CONFIG_BLK_DEV_RAM_COUNT=16
CONFIG_BLK_DEV_RAM_SIZE=131072
CONFIG_BLK_DEV_XIP=y
CONFIG_CDROM_PKTCDVD=m
CONFIG_CDROM_PKTCDVD_BUFFERS=8
CONFIG_CDROM_PKTCDVD_WCACHE=y
CONFIG_ATA_OVER_ETH=m
CONFIG_VIRTIO_BLK=m
# CONFIG_BLK_DEV_HD is not set
CONFIG_BLK_DEV_RBD=m

#
# Misc devices
#
CONFIG_SENSORS_LIS3LV02D=m
CONFIG_AD525X_DPOT=m
CONFIG_AD525X_DPOT_I2C=m
CONFIG_AD525X_DPOT_SPI=m
CONFIG_IBM_ASM=m
CONFIG_PHANTOM=m
# CONFIG_INTEL_MID_PTI is not set
CONFIG_SGI_IOC4=m
CONFIG_TIFM_CORE=m
CONFIG_TIFM_7XX1=m
CONFIG_ICS932S401=m
CONFIG_ENCLOSURE_SERVICES=m
CONFIG_SGI_XP=m
CONFIG_HP_ILO=m
CONFIG_SGI_GRU=m
# CONFIG_SGI_GRU_DEBUG is not set
# CONFIG_APDS9802ALS is not set
# CONFIG_ISL29003 is not set
CONFIG_ISL29020=m
CONFIG_SENSORS_TSL2550=m
CONFIG_SENSORS_BH1780=m
CONFIG_SENSORS_BH1770=m
CONFIG_SENSORS_APDS990X=m
CONFIG_HMC6352=m
CONFIG_DS1682=m
CONFIG_TI_DAC7512=m
CONFIG_VMWARE_BALLOON=m
# CONFIG_BMP085_I2C is not set
# CONFIG_BMP085_SPI is not set
CONFIG_PCH_PHUB=m
CONFIG_USB_SWITCH_FSA9480=m
CONFIG_C2PORT=m
CONFIG_C2PORT_DURAMAR_2150=m

#
# EEPROM support
#
CONFIG_EEPROM_AT24=m
CONFIG_EEPROM_AT25=m
CONFIG_EEPROM_LEGACY=m
CONFIG_EEPROM_MAX6875=m
CONFIG_EEPROM_93CX6=m
CONFIG_EEPROM_93XX46=m
CONFIG_CB710_CORE=m
# CONFIG_CB710_DEBUG is not set
CONFIG_CB710_DEBUG_ASSUMPTIONS=y

#
# Texas Instruments shared transport line discipline #
CONFIG_TI_ST=m
CONFIG_SENSORS_LIS3_I2C=m

#
# Altera FPGA firmware download module
#
CONFIG_ALTERA_STAPL=m
CONFIG_HAVE_IDE=y
# CONFIG_IDE is not set

#
# SCSI device support
#
CONFIG_SCSI_MOD=y
CONFIG_RAID_ATTRS=m
CONFIG_SCSI=y
CONFIG_SCSI_DMA=y
CONFIG_SCSI_TGT=m
CONFIG_SCSI_NETLINK=y
CONFIG_SCSI_PROC_FS=y

#
# SCSI support type (disk, tape, CD-ROM)
#
CONFIG_BLK_DEV_SD=y
CONFIG_CHR_DEV_ST=m
CONFIG_CHR_DEV_OSST=m
CONFIG_BLK_DEV_SR=m
# CONFIG_BLK_DEV_SR_VENDOR is not set
CONFIG_CHR_DEV_SG=m
CONFIG_CHR_DEV_SCH=m
CONFIG_SCSI_ENCLOSURE=m
CONFIG_SCSI_MULTI_LUN=y
CONFIG_SCSI_CONSTANTS=y
CONFIG_SCSI_LOGGING=y
# CONFIG_SCSI_SCAN_ASYNC is not set

#
# SCSI Transports
#
CONFIG_SCSI_SPI_ATTRS=m
CONFIG_SCSI_FC_ATTRS=m
CONFIG_SCSI_FC_TGT_ATTRS=y
CONFIG_SCSI_ISCSI_ATTRS=m
CONFIG_SCSI_SAS_ATTRS=m
CONFIG_SCSI_SAS_LIBSAS=m
CONFIG_SCSI_SAS_ATA=y
CONFIG_SCSI_SAS_HOST_SMP=y
CONFIG_SCSI_SRP_ATTRS=m
CONFIG_SCSI_SRP_TGT_ATTRS=y
CONFIG_SCSI_LOWLEVEL=y
CONFIG_ISCSI_TCP=m
CONFIG_ISCSI_BOOT_SYSFS=m
CONFIG_SCSI_CXGB3_ISCSI=m
CONFIG_SCSI_CXGB4_ISCSI=m
CONFIG_SCSI_BNX2_ISCSI=m
CONFIG_SCSI_BNX2X_FCOE=m
CONFIG_BE2ISCSI=m
CONFIG_BLK_DEV_3W_XXXX_RAID=m
CONFIG_SCSI_HPSA=m
CONFIG_SCSI_3W_9XXX=m
CONFIG_SCSI_3W_SAS=m
CONFIG_SCSI_ACARD=m
CONFIG_SCSI_AACRAID=m
CONFIG_SCSI_AIC7XXX=m
CONFIG_AIC7XXX_CMDS_PER_DEVICE=32
CONFIG_AIC7XXX_RESET_DELAY_MS=15000
# CONFIG_AIC7XXX_DEBUG_ENABLE is not set
CONFIG_AIC7XXX_DEBUG_MASK=0
CONFIG_AIC7XXX_REG_PRETTY_PRINT=y
CONFIG_SCSI_AIC7XXX_OLD=m
CONFIG_SCSI_AIC79XX=m
CONFIG_AIC79XX_CMDS_PER_DEVICE=32
CONFIG_AIC79XX_RESET_DELAY_MS=5000
# CONFIG_AIC79XX_DEBUG_ENABLE is not set
CONFIG_AIC79XX_DEBUG_MASK=0
CONFIG_AIC79XX_REG_PRETTY_PRINT=y
CONFIG_SCSI_AIC94XX=m
# CONFIG_AIC94XX_DEBUG is not set
CONFIG_SCSI_MVSAS=m
# CONFIG_SCSI_MVSAS_DEBUG is not set
CONFIG_SCSI_MVSAS_TASKLET=y
CONFIG_SCSI_MVUMI=m
CONFIG_SCSI_DPT_I2O=m
CONFIG_SCSI_ADVANSYS=m
CONFIG_SCSI_ARCMSR=m
CONFIG_MEGARAID_NEWGEN=y
CONFIG_MEGARAID_MM=m
CONFIG_MEGARAID_MAILBOX=m
CONFIG_MEGARAID_LEGACY=m
CONFIG_MEGARAID_SAS=m
CONFIG_SCSI_MPT2SAS=m
CONFIG_SCSI_MPT2SAS_MAX_SGE=128
# CONFIG_SCSI_MPT2SAS_LOGGING is not set
CONFIG_SCSI_UFSHCD=m
CONFIG_SCSI_HPTIOP=m
CONFIG_SCSI_BUSLOGIC=m
CONFIG_VMWARE_PVSCSI=m
CONFIG_HYPERV_STORAGE=m
CONFIG_LIBFC=m
CONFIG_LIBFCOE=m
CONFIG_FCOE=m
CONFIG_FCOE_FNIC=m
CONFIG_SCSI_DMX3191D=m
CONFIG_SCSI_EATA=m
CONFIG_SCSI_EATA_TAGGED_QUEUE=y
CONFIG_SCSI_EATA_LINKED_COMMANDS=y
CONFIG_SCSI_EATA_MAX_TAGS=16
CONFIG_SCSI_FUTURE_DOMAIN=m
CONFIG_SCSI_GDTH=m
CONFIG_SCSI_ISCI=m
CONFIG_SCSI_IPS=m
CONFIG_SCSI_INITIO=m
CONFIG_SCSI_INIA100=m
CONFIG_SCSI_STEX=m
CONFIG_SCSI_SYM53C8XX_2=m
CONFIG_SCSI_SYM53C8XX_DMA_ADDRESSING_MODE=1
CONFIG_SCSI_SYM53C8XX_DEFAULT_TAGS=16
CONFIG_SCSI_SYM53C8XX_MAX_TAGS=64
CONFIG_SCSI_SYM53C8XX_MMIO=y
CONFIG_SCSI_IPR=m
CONFIG_SCSI_IPR_TRACE=y
CONFIG_SCSI_IPR_DUMP=y
CONFIG_SCSI_QLOGIC_1280=m
CONFIG_SCSI_QLA_FC=m
# CONFIG_TCM_QLA2XXX is not set
CONFIG_SCSI_QLA_ISCSI=m
CONFIG_SCSI_LPFC=m
# CONFIG_SCSI_LPFC_DEBUG_FS is not set
CONFIG_SCSI_DC395x=m
CONFIG_SCSI_DC390T=m
CONFIG_SCSI_DEBUG=m
CONFIG_SCSI_PMCRAID=m
CONFIG_SCSI_PM8001=m
CONFIG_SCSI_SRP=m
CONFIG_SCSI_BFA_FC=m
CONFIG_SCSI_VIRTIO=m
CONFIG_SCSI_DH=m
CONFIG_SCSI_DH_RDAC=m
CONFIG_SCSI_DH_HP_SW=m
CONFIG_SCSI_DH_EMC=m
CONFIG_SCSI_DH_ALUA=m
CONFIG_SCSI_OSD_INITIATOR=m
CONFIG_SCSI_OSD_ULD=m
CONFIG_SCSI_OSD_DPRINT_SENSE=1
# CONFIG_SCSI_OSD_DEBUG is not set
CONFIG_ATA=y
# CONFIG_ATA_NONSTANDARD is not set
CONFIG_ATA_VERBOSE_ERROR=y
CONFIG_ATA_ACPI=y
CONFIG_SATA_PMP=y