linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Ric Mason <ric.masonn@gmail.com>
To: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Al Viro <viro@zeniv.linux.org.uk>,
	Hugh Dickins <hughd@google.com>,
	Wu Fengguang <fengguang.wu@intel.com>, Jan Kara <jack@suse.cz>,
	Mel Gorman <mgorman@suse.de>,
	linux-mm@kvack.org, Andi Kleen <ak@linux.intel.com>,
	Matthew Wilcox <matthew.r.wilcox@intel.com>,
	"Kirill A. Shutemov" <kirill@shutemov.name>,
	Hillf Danton <dhillf@gmail.com>,
	linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCHv2, RFC 04/30] radix-tree: implement preload for multiple contiguous elements
Date: Fri, 05 Apr 2013 11:37:23 +0800	[thread overview]
Message-ID: <515E46F3.1060400@gmail.com> (raw)
In-Reply-To: <1363283435-7666-5-git-send-email-kirill.shutemov@linux.intel.com>

Hi Kirill,
On 03/15/2013 01:50 AM, Kirill A. Shutemov wrote:
> From: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
>
> Currently radix_tree_preload() only guarantees enough nodes to insert
> one element. It's a hard limit. You cannot batch a number insert under
> one tree_lock.
>
> This patch introduces radix_tree_preload_count(). It allows to
> preallocate nodes enough to insert a number of *contiguous* elements.
>
> Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
> Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
> ---
>   include/linux/radix-tree.h |    3 +++
>   lib/radix-tree.c           |   32 +++++++++++++++++++++++++-------
>   2 files changed, 28 insertions(+), 7 deletions(-)
>
> diff --git a/include/linux/radix-tree.h b/include/linux/radix-tree.h
> index ffc444c..81318cb 100644
> --- a/include/linux/radix-tree.h
> +++ b/include/linux/radix-tree.h
> @@ -83,6 +83,8 @@ do {									\
>   	(root)->rnode = NULL;						\
>   } while (0)
>   
> +#define RADIX_TREE_PRELOAD_NR		512 /* For THP's benefit */
> +
>   /**
>    * Radix-tree synchronization
>    *
> @@ -231,6 +233,7 @@ unsigned long radix_tree_next_hole(struct radix_tree_root *root,
>   unsigned long radix_tree_prev_hole(struct radix_tree_root *root,
>   				unsigned long index, unsigned long max_scan);
>   int radix_tree_preload(gfp_t gfp_mask);
> +int radix_tree_preload_count(unsigned size, gfp_t gfp_mask);
>   void radix_tree_init(void);
>   void *radix_tree_tag_set(struct radix_tree_root *root,
>   			unsigned long index, unsigned int tag);
> diff --git a/lib/radix-tree.c b/lib/radix-tree.c
> index e796429..9bef0ac 100644
> --- a/lib/radix-tree.c
> +++ b/lib/radix-tree.c
> @@ -81,16 +81,24 @@ static struct kmem_cache *radix_tree_node_cachep;
>    * The worst case is a zero height tree with just a single item at index 0,
>    * and then inserting an item at index ULONG_MAX. This requires 2 new branches
>    * of RADIX_TREE_MAX_PATH size to be created, with only the root node shared.
> + *
> + * Worst case for adding N contiguous items is adding entries at indexes
> + * (ULONG_MAX - N) to ULONG_MAX. It requires nodes to insert single worst-case
> + * item plus extra nodes if you cross the boundary from one node to the next.
> + *

What's the meaning of this comments? Could you explain in details? I 
also don't understand #define RADIX_TREE_PRELOAD_SIZE 
(RADIX_TREE_MAX_PATH * 2 - 1), why RADIX_TREE_MAX_PATH * 2 - 1, I fail 
to understand comments above it.

>    * Hence:
>    */
> -#define RADIX_TREE_PRELOAD_SIZE (RADIX_TREE_MAX_PATH * 2 - 1)
> +#define RADIX_TREE_PRELOAD_MIN (RADIX_TREE_MAX_PATH * 2 - 1)
> +#define RADIX_TREE_PRELOAD_MAX \
> +	(RADIX_TREE_PRELOAD_MIN + \
> +	 DIV_ROUND_UP(RADIX_TREE_PRELOAD_NR - 1, RADIX_TREE_MAP_SIZE))
>   
>   /*
>    * Per-cpu pool of preloaded nodes
>    */
>   struct radix_tree_preload {
>   	int nr;
> -	struct radix_tree_node *nodes[RADIX_TREE_PRELOAD_SIZE];
> +	struct radix_tree_node *nodes[RADIX_TREE_PRELOAD_MAX];
>   };
>   static DEFINE_PER_CPU(struct radix_tree_preload, radix_tree_preloads) = { 0, };
>   
> @@ -257,29 +265,34 @@ radix_tree_node_free(struct radix_tree_node *node)
>   
>   /*
>    * Load up this CPU's radix_tree_node buffer with sufficient objects to
> - * ensure that the addition of a single element in the tree cannot fail.  On
> - * success, return zero, with preemption disabled.  On error, return -ENOMEM
> + * ensure that the addition of *contiguous* elements in the tree cannot fail.
> + * On success, return zero, with preemption disabled.  On error, return -ENOMEM
>    * with preemption not disabled.
>    *
>    * To make use of this facility, the radix tree must be initialised without
>    * __GFP_WAIT being passed to INIT_RADIX_TREE().
>    */
> -int radix_tree_preload(gfp_t gfp_mask)
> +int radix_tree_preload_count(unsigned size, gfp_t gfp_mask)
>   {
>   	struct radix_tree_preload *rtp;
>   	struct radix_tree_node *node;
>   	int ret = -ENOMEM;
> +	int alloc = RADIX_TREE_PRELOAD_MIN +
> +		DIV_ROUND_UP(size - 1, RADIX_TREE_MAP_SIZE);
> +
> +	if (size > RADIX_TREE_PRELOAD_NR)
> +		return -ENOMEM;
>   
>   	preempt_disable();
>   	rtp = &__get_cpu_var(radix_tree_preloads);
> -	while (rtp->nr < ARRAY_SIZE(rtp->nodes)) {
> +	while (rtp->nr < alloc) {
>   		preempt_enable();
>   		node = kmem_cache_alloc(radix_tree_node_cachep, gfp_mask);
>   		if (node == NULL)
>   			goto out;
>   		preempt_disable();
>   		rtp = &__get_cpu_var(radix_tree_preloads);
> -		if (rtp->nr < ARRAY_SIZE(rtp->nodes))
> +		if (rtp->nr < alloc)
>   			rtp->nodes[rtp->nr++] = node;
>   		else
>   			kmem_cache_free(radix_tree_node_cachep, node);
> @@ -288,6 +301,11 @@ int radix_tree_preload(gfp_t gfp_mask)
>   out:
>   	return ret;
>   }
> +
> +int radix_tree_preload(gfp_t gfp_mask)
> +{
> +	return radix_tree_preload_count(1, gfp_mask);
> +}
>   EXPORT_SYMBOL(radix_tree_preload);
>   
>   /*

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

  parent reply	other threads:[~2013-04-05  3:37 UTC|newest]

Thread overview: 122+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-03-14 17:50 [PATCHv2, RFC 00/30] Transparent huge page cache Kirill A. Shutemov
2013-03-14 17:50 ` [PATCHv2, RFC 01/30] block: implement add_bdi_stat() Kirill A. Shutemov
2013-03-21 14:46   ` Dave Hansen
2013-03-21 17:19     ` Kirill A. Shutemov
2013-03-14 17:50 ` [PATCHv2, RFC 02/30] mm: implement zero_huge_user_segment and friends Kirill A. Shutemov
2013-03-21 15:23   ` Dave Hansen
2013-03-22  9:21     ` Kirill A. Shutemov
2013-03-14 17:50 ` [PATCHv2, RFC 03/30] mm: drop actor argument of do_generic_file_read() Kirill A. Shutemov
2013-03-15  0:21   ` Hillf Danton
2013-03-15  0:27   ` Hillf Danton
2013-03-15 13:22     ` Kirill A. Shutemov
2013-03-21 15:26       ` Dave Hansen
2013-03-14 17:50 ` [PATCHv2, RFC 04/30] radix-tree: implement preload for multiple contiguous elements Kirill A. Shutemov
2013-03-21 15:56   ` Dave Hansen
2013-03-22  9:47     ` Kirill A. Shutemov
2013-03-22 14:38       ` Dave Hansen
2013-03-25 13:03         ` Kirill A. Shutemov
2013-04-05  3:37   ` Ric Mason [this message]
2013-03-14 17:50 ` [PATCHv2, RFC 05/30] thp, mm: avoid PageUnevictable on active/inactive lru lists Kirill A. Shutemov
2013-03-21 16:15   ` Dave Hansen
2013-03-22 10:11     ` Kirill A. Shutemov
2013-04-05  3:42       ` Ric Mason
2013-03-14 17:50 ` [PATCHv2, RFC 06/30] thp, mm: basic defines for transparent huge page cache Kirill A. Shutemov
2013-03-14 17:50 ` [PATCHv2, RFC 07/30] thp, mm: introduce mapping_can_have_hugepages() predicate Kirill A. Shutemov
2013-03-21 16:21   ` Dave Hansen
2013-03-22 10:12     ` Kirill A. Shutemov
2013-03-22 14:44       ` Dave Hansen
2013-04-02 14:46         ` Kirill A. Shutemov
2013-04-05  3:45   ` Ric Mason
2013-04-05  3:48     ` Ric Mason
2013-03-14 17:50 ` [PATCHv2, RFC 08/30] thp, mm: rewrite add_to_page_cache_locked() to support huge pages Kirill A. Shutemov
2013-03-15  1:30   ` Hillf Danton
2013-03-15 13:23     ` Kirill A. Shutemov
2013-03-15 13:25       ` Hillf Danton
2013-03-15 13:50         ` Kirill A. Shutemov
2013-03-15 13:55           ` Hillf Danton
2013-03-15 15:05             ` Kirill A. Shutemov
2013-03-21 17:11   ` Dave Hansen
2013-03-22 10:34     ` Kirill A. Shutemov
2013-03-22 14:51       ` Dave Hansen
2013-03-14 17:50 ` [PATCHv2, RFC 09/30] thp, mm: rewrite delete_from_page_cache() " Kirill A. Shutemov
2013-03-15  2:25   ` Hillf Danton
2013-03-15 13:23     ` Kirill A. Shutemov
2013-03-14 17:50 ` [PATCHv2, RFC 10/30] thp, mm: locking tail page is a bug Kirill A. Shutemov
2013-03-21 17:20   ` Dave Hansen
2013-03-14 17:50 ` [PATCHv2, RFC 11/30] thp, mm: handle tail pages in page_cache_get_speculative() Kirill A. Shutemov
2013-04-05  4:03   ` Ric Mason
2013-03-14 17:50 ` [PATCHv2, RFC 12/30] thp, mm: add event counters for huge page alloc on write to a file Kirill A. Shutemov
2013-03-21 17:59   ` Dave Hansen
2013-03-26  8:40     ` Kirill A. Shutemov
2013-04-05  4:05       ` Ric Mason
2013-03-14 17:50 ` [PATCHv2, RFC 13/30] thp, mm: implement grab_cache_huge_page_write_begin() Kirill A. Shutemov
2013-03-15  2:34   ` Hillf Danton
2013-03-15 13:24     ` Kirill A. Shutemov
2013-03-15 13:30       ` Hillf Danton
2013-03-15 13:35         ` Kirill A. Shutemov
2013-03-15 13:37           ` Hillf Danton
2013-03-21 18:15   ` Dave Hansen
2013-03-26 10:48     ` Kirill A. Shutemov
2013-03-26 15:40       ` Dave
2013-03-21 18:16   ` Dave Hansen
2013-03-14 17:50 ` [PATCHv2, RFC 14/30] thp, mm: naive support of thp in generic read/write routines Kirill A. Shutemov
2013-03-15  3:11   ` Hillf Danton
2013-03-15 13:27     ` Kirill A. Shutemov
2013-03-22 15:22   ` Dave Hansen
2013-03-28 12:25     ` Kirill A. Shutemov
2013-03-14 17:50 ` [PATCHv2, RFC 15/30] thp, libfs: initial support of thp in simple_read/write_begin/write_end Kirill A. Shutemov
2013-03-22 18:01   ` Dave
2013-03-28 14:29     ` Kirill A. Shutemov
2013-03-14 17:50 ` [PATCHv2, RFC 16/30] thp: handle file pages in split_huge_page() Kirill A. Shutemov
2013-03-15  6:15   ` Hillf Danton
2013-03-15 13:26     ` Kirill A. Shutemov
2013-03-15 13:33       ` Hillf Danton
2013-03-22 18:18   ` Dave
2013-03-28 14:32     ` Kirill A. Shutemov
2013-03-14 17:50 ` [PATCHv2, RFC 17/30] thp: wait_split_huge_page(): serialize over i_mmap_mutex too Kirill A. Shutemov
2013-03-22 18:22   ` Dave
2013-03-28 15:08     ` Kirill A. Shutemov
2013-03-14 17:50 ` [PATCHv2, RFC 18/30] thp, mm: truncate support for transparent huge page cache Kirill A. Shutemov
2013-03-22 18:29   ` Dave
2013-03-28 15:31     ` Kirill A. Shutemov
2013-03-14 17:50 ` [PATCHv2, RFC 19/30] thp, mm: split huge page on mmap file page Kirill A. Shutemov
2013-03-15  6:58   ` Hillf Danton
2013-03-15 13:29     ` Kirill A. Shutemov
2013-03-15 13:35       ` Hillf Danton
2013-03-15 13:45         ` Kirill A. Shutemov
2013-03-14 17:50 ` [PATCHv2, RFC 20/30] ramfs: enable transparent huge page cache Kirill A. Shutemov
2013-04-02 16:28   ` Kirill A. Shutemov
2013-04-02 22:15     ` Hugh Dickins
2013-04-03  1:11       ` Minchan Kim
2013-04-05  6:47         ` Simon Jeons
2013-04-05  8:01           ` Minchan Kim
2013-04-05  8:22             ` Wanpeng Li
     [not found]             ` <515e89d2.e725320a.3a74.7fe7SMTPIN_ADDED_BROKEN@mx.google.com>
2013-04-05  8:31               ` Minchan Kim
2013-04-05  8:35                 ` Wanpeng Li
2013-04-05 13:46                 ` Christoph Lameter
2013-04-03 13:53       ` Christoph Lameter
2013-03-14 17:50 ` [PATCHv2, RFC 21/30] x86-64, mm: proper alignment mappings with hugepages Kirill A. Shutemov
2013-03-22 18:37   ` Dave
2013-03-14 17:50 ` [PATCHv2, RFC 22/30] mm: add huge_fault() callback to vm_operations_struct Kirill A. Shutemov
2013-03-14 17:50 ` [PATCHv2, RFC 23/30] thp: prepare zap_huge_pmd() to uncharge file pages Kirill A. Shutemov
2013-03-15  7:09   ` Hillf Danton
2013-03-15 13:30     ` Kirill A. Shutemov
2013-03-14 17:50 ` [PATCHv2, RFC 24/30] thp: move maybe_pmd_mkwrite() out of mk_huge_pmd() Kirill A. Shutemov
2013-03-15  7:31   ` Hillf Danton
2013-03-14 17:50 ` [PATCHv2, RFC 25/30] thp, mm: basic huge_fault implementation for generic_file_vm_ops Kirill A. Shutemov
2013-03-15  7:44   ` Hillf Danton
2013-03-15 13:30     ` Kirill A. Shutemov
2013-03-14 17:50 ` [PATCHv2, RFC 26/30] thp: extract fallback path from do_huge_pmd_anonymous_page() to a function Kirill A. Shutemov
2013-03-14 17:50 ` [PATCHv2, RFC 27/30] thp: initial implementation of do_huge_linear_fault() Kirill A. Shutemov
2013-03-14 17:50 ` [PATCHv2, RFC 28/30] thp: handle write-protect exception to file-backed huge pages Kirill A. Shutemov
2013-03-14 17:50 ` [PATCHv2, RFC 29/30] thp: call __vma_adjust_trans_huge() for file-backed VMA Kirill A. Shutemov
2013-03-14 17:50 ` [PATCHv2, RFC 30/30] thp: map file-backed huge pages on fault Kirill A. Shutemov
2013-03-15  0:33 ` [PATCHv2, RFC 00/30] Transparent huge page cache Hillf Danton
2013-03-15 13:33   ` Kirill A. Shutemov
2013-03-18  4:03 ` Simon Jeons
2013-03-18  5:23   ` Simon Jeons
2013-03-18 11:19     ` Kirill A. Shutemov
2013-03-18 11:29       ` Simon Jeons
2013-03-18 11:42         ` Kirill A. Shutemov
2013-03-18 11:42           ` Ric Mason
2013-03-20  1:09 ` Simon Jeons

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=515E46F3.1060400@gmail.com \
    --to=ric.masonn@gmail.com \
    --cc=aarcange@redhat.com \
    --cc=ak@linux.intel.com \
    --cc=akpm@linux-foundation.org \
    --cc=dhillf@gmail.com \
    --cc=fengguang.wu@intel.com \
    --cc=hughd@google.com \
    --cc=jack@suse.cz \
    --cc=kirill.shutemov@linux.intel.com \
    --cc=kirill@shutemov.name \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=matthew.r.wilcox@intel.com \
    --cc=mgorman@suse.de \
    --cc=viro@zeniv.linux.org.uk \
    /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).