linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Michel Lespinasse <walken@google.com>
To: Andrew Morton <akpm@linux-foundation.org>,
	Rik van Riel <riel@redhat.com>, Hugh Dickins <hughd@google.com>,
	linux-kernel@vger.kernel.org,
	Russell King <linux@arm.linux.org.uk>,
	Ralf Baechle <ralf@linux-mips.org>,
	Paul Mundt <lethal@linux-sh.org>,
	"David S. Miller" <davem@davemloft.net>,
	Chris Metcalf <cmetcalf@tilera.com>,
	x86@kernel.org, William Irwin <wli@holomorphy.com>
Cc: linux-mm@kvack.org, linux-arm-kernel@lists.infradead.org,
	linux-mips@linux-mips.org, linux-sh@vger.kernel.org,
	sparclinux@vger.kernel.org, Rik van Riel <riel@surriel.com>
Subject: [PATCH 04/16] mm: rearrange vm_area_struct for fewer cache misses
Date: Mon,  5 Nov 2012 14:47:01 -0800	[thread overview]
Message-ID: <1352155633-8648-5-git-send-email-walken@google.com> (raw)
In-Reply-To: <1352155633-8648-1-git-send-email-walken@google.com>

From: Rik van Riel <riel@surriel.com>

The kernel walks the VMA rbtree in various places, including
the page fault path.  However, the vm_rb node spanned two
cache lines, on 64 bit systems with 64 byte cache lines (most
x86 systems).

Rearrange vm_area_struct a little, so all the information we
need to do a VMA tree walk is in the first cache line.

Signed-off-by: Michel Lespinasse <walken@google.com>
Signed-off-by: Rik van Riel <riel@redhat.com>

---
 include/linux/mm_types.h |   12 ++++++++----
 1 files changed, 8 insertions(+), 4 deletions(-)

diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index 94fa52b28ee8..528da4abf8ee 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -224,7 +224,8 @@ struct vm_region {
  * library, the executable area etc).
  */
 struct vm_area_struct {
-	struct mm_struct * vm_mm;	/* The address space we belong to. */
+	/* The first cache line has the info for VMA tree walking. */
+
 	unsigned long vm_start;		/* Our start address within vm_mm. */
 	unsigned long vm_end;		/* The first byte after our end address
 					   within vm_mm. */
@@ -232,9 +233,6 @@ struct vm_area_struct {
 	/* linked list of VM areas per task, sorted by address */
 	struct vm_area_struct *vm_next, *vm_prev;
 
-	pgprot_t vm_page_prot;		/* Access permissions of this VMA. */
-	unsigned long vm_flags;		/* Flags, see mm.h. */
-
 	struct rb_node vm_rb;
 
 	/*
@@ -245,6 +243,12 @@ struct vm_area_struct {
 	 */
 	unsigned long rb_subtree_gap;
 
+	/* Second cache line starts here. */
+
+	struct mm_struct * vm_mm;	/* The address space we belong to. */
+	pgprot_t vm_page_prot;		/* Access permissions of this VMA. */
+	unsigned long vm_flags;		/* Flags, see mm.h. */
+
 	/*
 	 * For areas with an address space and backing store,
 	 * linkage into the address_space->i_mmap interval tree, or
-- 
1.7.7.3

--
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:[~2012-11-05 22:47 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-11-05 22:46 [PATCH 00/16] mm: use augmented rbtrees for finding unmapped areas Michel Lespinasse
2012-11-05 22:46 ` [PATCH 01/16] mm: add anon_vma_lock to validate_mm() Michel Lespinasse
2012-11-05 23:06   ` Rik van Riel
2012-11-06  8:00     ` Michel Lespinasse
2012-11-05 22:46 ` [PATCH 02/16] mm: augment vma rbtree with rb_subtree_gap Michel Lespinasse
2012-11-05 22:47 ` [PATCH 03/16] mm: check rb_subtree_gap correctness Michel Lespinasse
2012-11-06 22:38   ` Andrew Morton
2012-11-09 14:13   ` Sasha Levin
2012-11-09 20:06     ` Hugh Dickins
2012-11-12 11:55     ` Michel Lespinasse
2012-11-05 22:47 ` Michel Lespinasse [this message]
2012-11-05 22:47 ` [PATCH 05/16] mm: vm_unmapped_area() lookup function Michel Lespinasse
2012-11-05 22:47 ` [PATCH 06/16] mm: use vm_unmapped_area() on x86_64 architecture Michel Lespinasse
2012-11-05 22:47 ` [PATCH 07/16] mm: fix cache coloring " Michel Lespinasse
2012-11-06 22:38   ` Andrew Morton
2012-11-05 22:47 ` [PATCH 08/16] mm: use vm_unmapped_area() in hugetlbfs Michel Lespinasse
2012-11-05 23:33   ` Rik van Riel
2012-11-05 22:47 ` [PATCH 09/16] mm: use vm_unmapped_area() in hugetlbfs on i386 architecture Michel Lespinasse
2012-11-05 23:07   ` Rik van Riel
2012-11-06 22:38   ` Andrew Morton
2012-11-06 22:48     ` Rik van Riel
2012-11-05 22:47 ` [PATCH 10/16] mm: use vm_unmapped_area() on mips architecture Michel Lespinasse
2012-11-05 23:34   ` Rik van Riel
2012-11-06 22:38   ` Andrew Morton
2012-11-05 22:47 ` [PATCH 11/16] mm: use vm_unmapped_area() on arm architecture Michel Lespinasse
2012-11-05 23:35   ` Rik van Riel
2012-11-06 22:38   ` Andrew Morton
2012-11-05 22:47 ` [PATCH 12/16] mm: use vm_unmapped_area() on sh architecture Michel Lespinasse
2012-11-05 23:35   ` Rik van Riel
2012-11-05 22:47 ` [PATCH 13/16] mm: use vm_unmapped_area() on sparc64 architecture Michel Lespinasse
2012-11-05 23:36   ` Rik van Riel
2012-11-05 22:47 ` [PATCH 14/16] mm: use vm_unmapped_area() in hugetlbfs " Michel Lespinasse
2012-11-05 23:36   ` Rik van Riel
2012-11-05 22:47 ` [PATCH 15/16] mm: use vm_unmapped_area() on sparc32 architecture Michel Lespinasse
2012-11-05 23:37   ` Rik van Riel
2012-11-06  1:25   ` David Miller
2012-11-06  3:13     ` Michel Lespinasse
2012-11-06  7:30     ` Rik van Riel
2012-11-06 17:41       ` David Miller
2012-11-05 22:47 ` [PATCH 16/16] mm: use vm_unmapped_area() in hugetlbfs on tile architecture Michel Lespinasse
2012-11-05 23:37   ` Rik van Riel
2012-11-06 22:11 ` [PATCH 00/16] mm: use augmented rbtrees for finding unmapped areas Andrew Morton

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=1352155633-8648-5-git-send-email-walken@google.com \
    --to=walken@google.com \
    --cc=akpm@linux-foundation.org \
    --cc=cmetcalf@tilera.com \
    --cc=davem@davemloft.net \
    --cc=hughd@google.com \
    --cc=lethal@linux-sh.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mips@linux-mips.org \
    --cc=linux-mm@kvack.org \
    --cc=linux-sh@vger.kernel.org \
    --cc=linux@arm.linux.org.uk \
    --cc=ralf@linux-mips.org \
    --cc=riel@redhat.com \
    --cc=riel@surriel.com \
    --cc=sparclinux@vger.kernel.org \
    --cc=wli@holomorphy.com \
    --cc=x86@kernel.org \
    /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).