[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20170517141146.11063-3-richard.weiyang@gmail.com>
Date: Wed, 17 May 2017 22:11:42 +0800
From: Wei Yang <richard.weiyang@...il.com>
To: cl@...ux.com, penberg@...nel.org, rientjes@...gle.com,
akpm@...ux-foundation.org
Cc: linux-mm@...ck.org, linux-kernel@...r.kernel.org,
Wei Yang <richard.weiyang@...il.com>
Subject: [PATCH 2/6] mm/slub: not include cpu_partial data in cpu_slabs sysfs
There are four level slabs:
CPU
CPU_PARTIAL
PARTIAL
FULL
In current implementation, cpu_slabs sysfs would give statistics including
the first two levels. While there is another sysfs entry cpu_partial_slabs
gives details on the second level slab statistics. Since each cpu has one
slab for the first level, the current cpu_slabs output is easy to be
calculated from cpu_partial_slabs.
This patch removes the cpu_partial data in cpu_slabs for more specific slab
statistics and leave room to retrieve objects and total objects on CPU
level in the future.
Signed-off-by: Wei Yang <richard.weiyang@...il.com>
---
mm/slub.c | 13 -------------
1 file changed, 13 deletions(-)
diff --git a/mm/slub.c b/mm/slub.c
index 1100d2e75870..c7dddf22829d 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -4762,19 +4762,6 @@ static ssize_t show_slab_objects(struct kmem_cache *s,
total += x;
nodes[node] += x;
-
- page = slub_percpu_partial_read_once(c);
- if (page) {
- node = page_to_nid(page);
- if (flags & SO_TOTAL)
- WARN_ON_ONCE(1);
- else if (flags & SO_OBJECTS)
- WARN_ON_ONCE(1);
- else
- x = page->pages;
- total += x;
- nodes[node] += x;
- }
}
}
--
2.11.0
Powered by blists - more mailing lists