All of lore.kernel.org
 help / color / mirror / Atom feed
From: Robert Jennings <rcj@linux.vnet.ibm.com>
To: Nitin Gupta <ngupta@vflare.org>
Cc: Greg Kroah-Hartman <gregkh@suse.de>,
	Pekka Enberg <penberg@cs.helsinki.fi>,
	devel@driverdev.osuosl.org, linux-kernel@vger.kernel.org
Subject: [PATCH 3/7] zram: Speed insertion of new pages with cached idx
Date: Wed, 26 Jan 2011 11:23:55 -0600	[thread overview]
Message-ID: <20110126172354.GE8446@linux.vnet.ibm.com> (raw)
In-Reply-To: <20110126172148.GB8446@linux.vnet.ibm.com>

Calculate the first- and second-level indices for new page when the pool
is initialized rather than calculating them on each insertion.

Signed-off-by: Robert Jennings <rcj@linux.vnet.ibm.com>
---
 drivers/staging/zram/xvmalloc.c     |   13 +++++++++++--
 drivers/staging/zram/xvmalloc_int.h |    4 ++++
 2 files changed, 15 insertions(+), 2 deletions(-)

diff --git a/drivers/staging/zram/xvmalloc.c b/drivers/staging/zram/xvmalloc.c
index 3fdbb8a..a507f95 100644
--- a/drivers/staging/zram/xvmalloc.c
+++ b/drivers/staging/zram/xvmalloc.c
@@ -184,8 +184,13 @@ static void insert_block(struct xv_pool *pool, struct page *page, u32 offset,
 	u32 flindex, slindex;
 	struct block_header *nextblock;
 
-	slindex = get_index_for_insert(block->size);
-	flindex = slindex / BITS_PER_LONG;
+	if (block->size >= (PAGE_SIZE - XV_ALIGN)) {
+		slindex = pagesize_slindex;
+		flindex = pagesize_flindex;
+	} else {
+		slindex = get_index_for_insert(block->size);
+		flindex = slindex / BITS_PER_LONG;
+	}
 
 	block->link.prev_page = 0;
 	block->link.prev_offset = 0;
@@ -316,6 +321,10 @@ struct xv_pool *xv_create_pool(void)
 	if (!pool)
 		return NULL;
 
+	/* cache the first/second-level indices for PAGE_SIZE allocations */
+	pagesize_slindex = get_index_for_insert(PAGE_SIZE);
+	pagesize_flindex = pagesize_slindex / BITS_PER_LONG;
+
 	spin_lock_init(&pool->lock);
 
 	return pool;
diff --git a/drivers/staging/zram/xvmalloc_int.h b/drivers/staging/zram/xvmalloc_int.h
index 051a49b..68db384 100644
--- a/drivers/staging/zram/xvmalloc_int.h
+++ b/drivers/staging/zram/xvmalloc_int.h
@@ -52,6 +52,10 @@
 
 /* End of user params */
 
+/* Cache of the first/second-level indices for PAGE_SIZE allocations */
+static u32 pagesize_slindex;
+static u32 pagesize_flindex;
+
 enum blockflags {
 	BLOCK_FREE,
 	PREV_FREE,
-- 
1.6.0.2


  parent reply	other threads:[~2011-01-26 17:24 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-01-26 17:21 [PATCH 0/7] zram/xvmalloc: 64K page fixes and optimizations Robert Jennings
2011-01-26 17:22 ` [PATCH 1/7] zram/vmalloc: Correct tunings to enable use with 64K pages Robert Jennings
2011-01-26 17:48   ` Pekka Enberg
2011-01-26 18:50     ` Robert Jennings
2011-01-26 17:23 ` [PATCH 2/7] zram: Prevent overflow in logical block size Robert Jennings
2011-01-26 17:42   ` Pekka Enberg
2011-01-26 19:09     ` Robert Jennings
2011-01-26 19:14       ` Pekka Enberg
2011-01-26 17:23 ` Robert Jennings [this message]
2011-01-26 17:53   ` [PATCH 3/7] zram: Speed insertion of new pages with cached idx Pekka Enberg
2011-01-26 19:50     ` Robert Jennings
2011-01-26 17:25 ` [PATCH 4/7] zram/xvmalloc: free bit block insertion optimization Robert Jennings
2011-01-26 17:55   ` Pekka Enberg
2011-01-26 18:40     ` Robert Jennings
2011-01-26 18:51       ` Pekka Enberg
2011-01-26 18:44   ` Nitin Gupta
2011-01-26 17:26 ` [PATCH 5/7] zram/xvmalloc: wrap debug code in #ifdef DEBUG Robert Jennings
2011-01-26 17:44   ` Pekka Enberg
2011-01-26 18:27     ` Robert Jennings
2011-01-26 17:27 ` [PATCH 6/7] zram/xvmalloc: Close 32byte hole on 64-bit CPUs Robert Jennings
2011-01-26 17:43   ` Pekka Enberg
2011-01-26 17:27 ` [PATCH 7/7] zram: mark the disk as non-rotating media Robert Jennings
2011-01-26 17:44   ` Pekka Enberg
2011-01-26 17:48   ` Dan Carpenter
2011-01-26 17:52   ` Robert Jennings
2011-01-26 18:26     ` Robert Jennings

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=20110126172354.GE8446@linux.vnet.ibm.com \
    --to=rcj@linux.vnet.ibm.com \
    --cc=devel@driverdev.osuosl.org \
    --cc=gregkh@suse.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=ngupta@vflare.org \
    --cc=penberg@cs.helsinki.fi \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.