[PATCH] mm,numa: Remove remaining traces of rate-limiting.

[PATCH] mm,numa: Remove remaining traces of rate-limiting.

[Srikar Dronamraju]

With Commit efaffc5e40ae ("mm, sched/numa: Remove rate-limiting of automatic
NUMA balancing migration"), we no more require migrate lock and its
initialization. Its redundant. Hence remove it.

Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
---
 include/linux/mmzone.h |  4 ----
 mm/page_alloc.c        | 10 ----------
 2 files changed, 14 deletions(-)

diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 3f4c0b1..d4b0c79 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -668,10 +668,6 @@ typedef struct pglist_data {
 	wait_queue_head_t kcompactd_wait;
 	struct task_struct *kcompactd;
 #endif
-#ifdef CONFIG_NUMA_BALANCING
-	/* Lock serializing the migrate rate limiting window */
-	spinlock_t numabalancing_migrate_lock;
-#endif
 	/*
 	 * This is a per-node reserve of pages that are not available
 	 * to userspace allocations.
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 706a738..e2ef1c1 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -6193,15 +6193,6 @@ static unsigned long __init calc_memmap_size(unsigned long spanned_pages,
 	return PAGE_ALIGN(pages * sizeof(struct page)) >> PAGE_SHIFT;
 }
 
-#ifdef CONFIG_NUMA_BALANCING
-static void pgdat_init_numabalancing(struct pglist_data *pgdat)
-{
-	spin_lock_init(&pgdat->numabalancing_migrate_lock);
-}
-#else
-static void pgdat_init_numabalancing(struct pglist_data *pgdat) {}
-#endif
-
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
 static void pgdat_init_split_queue(struct pglist_data *pgdat)
 {
@@ -6226,7 +6217,6 @@ static void __meminit pgdat_init_internals(struct pglist_data *pgdat)
 {
 	pgdat_resize_init(pgdat);
 
-	pgdat_init_numabalancing(pgdat);
 	pgdat_init_split_queue(pgdat);
 	pgdat_init_kcompactd(pgdat);
 
-- 
2.7.4

Re: [PATCH] mm,numa: Remove remaining traces of rate-limiting.

[Mel Gorman]

On Sat, Oct 06, 2018 at 04:53:19PM +0530, Srikar Dronamraju wrote:
> With Commit efaffc5e40ae ("mm, sched/numa: Remove rate-limiting of automatic
> NUMA balancing migration"), we no more require migrate lock and its
> initialization. Its redundant. Hence remove it.
> 
> Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>

Acked-by: Mel Gorman <mgorman@techsingularity.net>

-- 
Mel Gorman
SUSE Labs

Re: [PATCH] mm,numa: Remove remaining traces of rate-limiting.

[Mel Gorman]

On Sat, Oct 06, 2018 at 04:53:19PM +0530, Srikar Dronamraju wrote:
> With Commit efaffc5e40ae ("mm, sched/numa: Remove rate-limiting of automatic
> NUMA balancing migration"), we no more require migrate lock and its
> initialization. Its redundant. Hence remove it.
> 
> Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>

Hi Ingo, 

Can this be sent with the rest of the patches that got merged for 4.19-rc7
so they are more or less together? It's functionally harmless to delay
until the 4.20 merge window but it's a bit untidy. The mistake was mine
switching between a backport and mainline versions of the original patch.

Thanks

-- 
Mel Gorman
SUSE Labs

Re: [PATCH] mm,numa: Remove remaining traces of rate-limiting.

[Ingo Molnar]

* Mel Gorman <mgorman@techsingularity.net> wrote:

> On Sat, Oct 06, 2018 at 04:53:19PM +0530, Srikar Dronamraju wrote:
> > With Commit efaffc5e40ae ("mm, sched/numa: Remove rate-limiting of automatic
> > NUMA balancing migration"), we no more require migrate lock and its
> > initialization. Its redundant. Hence remove it.
> > 
> > Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
> 
> Hi Ingo, 
> 
> Can this be sent with the rest of the patches that got merged for 4.19-rc7
> so they are more or less together? It's functionally harmless to delay
> until the 4.20 merge window but it's a bit untidy. The mistake was mine
> switching between a backport and mainline versions of the original patch.
> 
> Thanks

Ok, agreed and done - I queued it up in sched/urgent.

Thanks,

	Ingo

[tip:sched/urgent] mm, sched/numa: Remove remaining traces of NUMA rate-limiting

[tip-bot for Srikar Dronamraju]

Commit-ID:  e054637597ba36d3729ba6a3a3dd7aad8e2a3003
Gitweb:     https://git.kernel.org/tip/e054637597ba36d3729ba6a3a3dd7aad8e2a3003
Author:     Srikar Dronamraju <srikar@linux.vnet.ibm.com>
AuthorDate: Sat, 6 Oct 2018 16:53:19 +0530
Committer:  Ingo Molnar <mingo@kernel.org>
CommitDate: Tue, 9 Oct 2018 08:30:51 +0200

mm, sched/numa: Remove remaining traces of NUMA rate-limiting

Remove the leftover pglist_data::numabalancing_migrate_lock and its
initialization, we stopped using this lock with:

  efaffc5e40ae ("mm, sched/numa: Remove rate-limiting of automatic NUMA balancing migration")

[ mingo: Rewrote the changelog. ]

Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Linux-MM <linux-mm@kvack.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Rik van Riel <riel@surriel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1538824999-31230-1-git-send-email-srikar@linux.vnet.ibm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
 include/linux/mmzone.h |  4 ----
 mm/page_alloc.c        | 10 ----------
 2 files changed, 14 deletions(-)

diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 3f4c0b167333..d4b0c79d2924 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -667,10 +667,6 @@ typedef struct pglist_data {
 	enum zone_type kcompactd_classzone_idx;
 	wait_queue_head_t kcompactd_wait;
 	struct task_struct *kcompactd;
-#endif
-#ifdef CONFIG_NUMA_BALANCING
-	/* Lock serializing the migrate rate limiting window */
-	spinlock_t numabalancing_migrate_lock;
 #endif
 	/*
 	 * This is a per-node reserve of pages that are not available
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 706a738c0aee..e2ef1c17942f 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -6193,15 +6193,6 @@ static unsigned long __init calc_memmap_size(unsigned long spanned_pages,
 	return PAGE_ALIGN(pages * sizeof(struct page)) >> PAGE_SHIFT;
 }
 
-#ifdef CONFIG_NUMA_BALANCING
-static void pgdat_init_numabalancing(struct pglist_data *pgdat)
-{
-	spin_lock_init(&pgdat->numabalancing_migrate_lock);
-}
-#else
-static void pgdat_init_numabalancing(struct pglist_data *pgdat) {}
-#endif
-
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
 static void pgdat_init_split_queue(struct pglist_data *pgdat)
 {
@@ -6226,7 +6217,6 @@ static void __meminit pgdat_init_internals(struct pglist_data *pgdat)
 {
 	pgdat_resize_init(pgdat);
 
-	pgdat_init_numabalancing(pgdat);
 	pgdat_init_split_queue(pgdat);
 	pgdat_init_kcompactd(pgdat);