From: Matthew Wilcox <willy@linux.intel.com>
To: linux-kernel@vger.kernel.org, Andrew Morton <akpm@linux-foundation.org>
Cc: Matthew Wilcox <willy@linux.intel.com>,
Ross Zwisler <ross.zwisler@linux.intel.com>,
linux-mm@kvack.org, linux-fsdevel@vger.kernel.org,
Konstantin Khlebnikov <koct9i@gmail.com>,
Kirill Shutemov <kirill.shutemov@linux.intel.com>,
Jan Kara <jack@suse.com>, Neil Brown <neilb@suse.de>
Subject: [PATCH 08/30] Introduce CONFIG_RADIX_TREE_MULTIORDER
Date: Wed, 6 Apr 2016 17:21:17 -0400 [thread overview]
Message-ID: <1459977699-2349-9-git-send-email-willy@linux.intel.com> (raw)
In-Reply-To: <1459977699-2349-1-git-send-email-willy@linux.intel.com>
I've been receiving increasingly concerned notes from 0day about how much
my recent changes have been bloating the radix tree. Make it happier by
only including multiorder support if CONFIG_TRANSPARENT_HUGEPAGES is set.
This is an independent Kconfig option, so other radix tree users can
also set it if they have a need.
Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Reviewed-by: Ross Zwisler <ross.zwisler@linux.intel.com>
---
lib/Kconfig | 3 +++
lib/radix-tree.c | 25 +++++++++++++++++--------
mm/Kconfig | 1 +
tools/testing/radix-tree/linux/kernel.h | 1 +
4 files changed, 22 insertions(+), 8 deletions(-)
diff --git a/lib/Kconfig b/lib/Kconfig
index 3cca1222578e..56403a6fba41 100644
--- a/lib/Kconfig
+++ b/lib/Kconfig
@@ -362,6 +362,9 @@ config INTERVAL_TREE
for more information.
+config RADIX_TREE_MULTIORDER
+ bool
+
config ASSOCIATIVE_ARRAY
bool
help
diff --git a/lib/radix-tree.c b/lib/radix-tree.c
index 1624c4117961..ae53ad56e01e 100644
--- a/lib/radix-tree.c
+++ b/lib/radix-tree.c
@@ -484,6 +484,7 @@ int __radix_tree_create(struct radix_tree_root *root, unsigned long index,
slot = node->slots[offset];
}
+#ifdef CONFIG_RADIX_TREE_MULTIORDER
/* Insert pointers to the canonical entry */
if ((shift - order) > 0) {
int i, n = 1 << (shift - order);
@@ -499,6 +500,7 @@ int __radix_tree_create(struct radix_tree_root *root, unsigned long index,
node->count++;
}
}
+#endif
if (nodep)
*nodep = node;
@@ -1469,6 +1471,19 @@ bool __radix_tree_delete_node(struct radix_tree_root *root,
return deleted;
}
+static inline void delete_sibling_entries(struct radix_tree_node *node,
+ void *ptr, unsigned offset)
+{
+#ifdef CONFIG_RADIX_TREE_MULTIORDER
+ int i;
+ for (i = 1; offset + i < RADIX_TREE_MAP_SIZE; i++) {
+ if (node->slots[offset + i] != ptr)
+ break;
+ node->slots[offset + i] = NULL;
+ node->count--;
+ }
+#endif
+}
/**
* radix_tree_delete_item - delete an item from a radix tree
* @root: radix tree root
@@ -1484,7 +1499,7 @@ void *radix_tree_delete_item(struct radix_tree_root *root,
unsigned long index, void *item)
{
struct radix_tree_node *node;
- unsigned int offset, i;
+ unsigned int offset;
void **slot;
void *entry;
int tag;
@@ -1513,13 +1528,7 @@ void *radix_tree_delete_item(struct radix_tree_root *root,
radix_tree_tag_clear(root, index, tag);
}
- /* Delete any sibling slots pointing to this slot */
- for (i = 1; offset + i < RADIX_TREE_MAP_SIZE; i++) {
- if (node->slots[offset + i] != ptr_to_indirect(slot))
- break;
- node->slots[offset + i] = NULL;
- node->count--;
- }
+ delete_sibling_entries(node, ptr_to_indirect(slot), offset);
node->slots[offset] = NULL;
node->count--;
diff --git a/mm/Kconfig b/mm/Kconfig
index 989f8f3d77e0..6d5b39e6d326 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -393,6 +393,7 @@ config TRANSPARENT_HUGEPAGE
bool "Transparent Hugepage Support"
depends on HAVE_ARCH_TRANSPARENT_HUGEPAGE
select COMPACTION
+ select RADIX_TREE_MULTIORDER
help
Transparent Hugepages allows the kernel to use huge pages and
huge tlb transparently to the applications whenever possible.
diff --git a/tools/testing/radix-tree/linux/kernel.h b/tools/testing/radix-tree/linux/kernel.h
index 31fe2c77d7ae..8ea0ed450810 100644
--- a/tools/testing/radix-tree/linux/kernel.h
+++ b/tools/testing/radix-tree/linux/kernel.h
@@ -9,6 +9,7 @@
#include "../../include/linux/compiler.h"
+#define CONFIG_RADIX_TREE_MULTIORDER
#define CONFIG_SHMEM
#define CONFIG_SWAP
--
2.8.0.rc3
next prev parent reply other threads:[~2016-04-06 21:24 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-04-06 21:21 [PATCH 00/30] Radix tree multiorder fixes Matthew Wilcox
2016-04-06 21:21 ` [PATCH 01/30] radix-tree: Introduce radix_tree_empty Matthew Wilcox
2016-04-06 21:21 ` [PATCH 02/30] radix tree test suite: Fix build Matthew Wilcox
2016-04-06 21:21 ` [PATCH 03/30] radix tree test suite: Add tests for radix_tree_locate_item() Matthew Wilcox
2016-04-06 21:21 ` [PATCH 04/30] radix tree test suite: Allow testing other fan-out values Matthew Wilcox
2016-04-06 21:21 ` [PATCH 05/30] radix tree test suite: keep regression test runs short Matthew Wilcox
2016-04-06 21:21 ` [PATCH 06/30] radix tree test suite: rebuild when headers change Matthew Wilcox
2016-04-06 21:21 ` [PATCH 07/30] radix-tree: remove unused looping macros Matthew Wilcox
2016-04-06 21:21 ` Matthew Wilcox [this message]
2016-04-06 21:21 ` [PATCH 09/30] radix-tree: Add missing sibling entry functionality Matthew Wilcox
2016-04-06 21:21 ` [PATCH 10/30] radix-tree: Fix sibling entry insertion Matthew Wilcox
2016-04-06 21:21 ` [PATCH 11/30] radix-tree: Fix deleting a multi-order entry through an alias Matthew Wilcox
2016-04-06 21:21 ` [PATCH 12/30] radix-tree: Remove restriction on multi-order entries Matthew Wilcox
2016-04-06 21:21 ` [PATCH 13/30] radix-tree: Introduce radix_tree_load_root() Matthew Wilcox
2016-04-06 21:21 ` [PATCH 14/30] radix-tree: Fix extending the tree for multi-order entries at offset 0 Matthew Wilcox
2016-04-06 21:21 ` [PATCH 15/30] radix-tree: Fix several shrinking bugs with multiorder entries Matthew Wilcox
2016-04-06 21:21 ` [PATCH 16/30] radix tree test suite: Start adding multiorder tests Matthew Wilcox
2016-04-06 21:21 ` [PATCH 17/30] radix-tree: Rewrite __radix_tree_lookup Matthew Wilcox
2016-04-06 21:21 ` [PATCH 18/30] radix-tree: Fix multiorder BUG_ON in radix_tree_insert Matthew Wilcox
2016-04-06 21:21 ` [PATCH 19/30] radix-tree: add support for multi-order iterating Matthew Wilcox
2016-04-06 21:21 ` [PATCH 20/30] radix tree test suite: multi-order iteration test Matthew Wilcox
2016-04-06 21:21 ` [PATCH 21/30] radix tree test suite: Add multiorder shrinking test Matthew Wilcox
2016-04-06 21:21 ` [PATCH 22/30] radix-tree: Rewrite radix_tree_tag_set Matthew Wilcox
2016-04-06 21:21 ` [PATCH 23/30] radix-tree: Rewrite radix_tree_tag_clear Matthew Wilcox
2016-04-06 21:21 ` [PATCH 24/30] radix-tree: Rewrite radix_tree_tag_get Matthew Wilcox
2016-04-06 21:21 ` [PATCH 25/30] radix-tree test suite: add multi-order tag test Matthew Wilcox
2016-04-06 21:21 ` [PATCH 26/30] radix-tree: Fix radix_tree_create for sibling entries Matthew Wilcox
2016-04-06 21:21 ` [PATCH 27/30] radix-tree: Rewrite radix_tree_locate_item Matthew Wilcox
2016-04-06 21:21 ` [PATCH 28/30] radix-tree: Fix two bugs in radix_tree_range_tag_if_tagged() Matthew Wilcox
2016-04-06 21:21 ` [PATCH 29/30] radix-tree: Fix radix_tree_dump() for multi-order entries Matthew Wilcox
2016-04-06 21:21 ` [PATCH 30/30] radix-tree: Add copyright statements Matthew Wilcox
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1459977699-2349-9-git-send-email-willy@linux.intel.com \
--to=willy@linux.intel.com \
--cc=akpm@linux-foundation.org \
--cc=jack@suse.com \
--cc=kirill.shutemov@linux.intel.com \
--cc=koct9i@gmail.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=neilb@suse.de \
--cc=ross.zwisler@linux.intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).