From: Rik van Riel <riel@surriel.com>
To: linux-mm@kvack.org
Cc: akpm@linux-foundation.org, aarcange@redhat.com,
peterz@infradead.org, minchan@gmail.com,
kosaki.motohiro@gmail.com, andi@firstfloor.org,
hannes@cmpxchg.org, mel@csn.ul.ie, linux-kernel@vger.kernel.org,
Rik van Riel <riel@surriel.com>, Rik van Riel <riel@redhat.com>
Subject: [PATCH -mm v2 02/11] mm: rearrange vm_area_struct for fewer cache misses
Date: Thu, 21 Jun 2012 17:57:06 -0400 [thread overview]
Message-ID: <1340315835-28571-3-git-send-email-riel@surriel.com> (raw)
In-Reply-To: <1340315835-28571-1-git-send-email-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: 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 9fc0291..23bd1e2 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -199,7 +199,8 @@ 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. */
@@ -207,9 +208,6 @@ 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;
/*
@@ -220,6 +218,12 @@ struct vm_area_struct {
*/
unsigned long free_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 prio tree, or
--
1.7.7.6
--
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>
next prev parent reply other threads:[~2012-06-21 21:57 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-06-21 21:57 [PATCH -mm v2 00/11] mm: scalable and unified arch_get_unmapped_area Rik van Riel
2012-06-21 21:57 ` [PATCH -mm v2 01/11] mm: track free size between VMAs in VMA rbtree Rik van Riel
2012-06-22 9:57 ` Peter Zijlstra
2012-06-22 9:58 ` Peter Zijlstra
2012-06-22 14:11 ` Rik van Riel
2012-06-22 14:13 ` Peter Zijlstra
2012-06-22 14:25 ` Rik van Riel
2012-06-22 14:37 ` Peter Zijlstra
2012-06-22 15:41 ` Rik van Riel
2012-06-25 19:29 ` Peter Zijlstra
2012-06-25 21:52 ` Rik van Riel
2012-06-26 8:31 ` Peter Zijlstra
2012-06-26 13:05 ` Rik van Riel
2012-06-26 13:45 ` Peter Zijlstra
2012-06-26 15:49 ` Rik van Riel
2012-06-27 12:27 ` Peter Zijlstra
2012-06-26 8:37 ` Peter Zijlstra
2012-06-22 10:02 ` Peter Zijlstra
2012-06-29 23:46 ` Michel Lespinasse
2012-07-03 21:37 ` Rik van Riel
2012-07-03 23:16 ` Michel Lespinasse
2012-07-04 10:12 ` Peter Zijlstra
2012-06-21 21:57 ` Rik van Riel [this message]
2012-06-21 21:57 ` [PATCH -mm v2 03/11] mm: vma_adjust: only call adjust_free_gap when needed Rik van Riel
2012-06-21 21:57 ` [PATCH -mm v2 04/11] rbtree: add helpers to find nearest uncle node Rik van Riel
2012-06-22 9:49 ` Peter Zijlstra
2012-06-21 21:57 ` [PATCH -mm v2 05/11] mm: get unmapped area from VMA tree Rik van Riel
2012-06-30 1:33 ` Michel Lespinasse
2012-07-03 0:23 ` Michel Lespinasse
2012-06-30 2:42 ` Michel Lespinasse
2012-06-21 21:57 ` [PATCH -mm v2 06/11] mm: arbitrary address ranges for arch_get_unmapped_area Rik van Riel
2012-06-21 21:57 ` [PATCH -mm v2 07/11] mm: make cache alignment code generic Rik van Riel
2012-06-30 2:22 ` Michel Lespinasse
2012-06-21 21:57 ` [PATCH -mm v2 08/11] mm: remove x86 arch_get_unmapped_area(_topdown) Rik van Riel
2012-06-21 21:57 ` [PATCH -mm v2 09/11] mm: remove MIPS arch_get_unmapped_area code Rik van Riel
2012-06-21 21:57 ` [PATCH -mm v2 10/11] mm: remove ARM arch_get_unmapped_area functions Rik van Riel
2012-06-22 22:27 ` Russell King - ARM Linux
2012-06-23 17:50 ` Johannes Weiner
2012-06-21 21:57 ` [PATCH -mm v2 11/11] mm: remove SH " Rik van Riel
2012-06-25 2:11 ` Paul Mundt
2012-06-22 14:24 ` [PATCH -mm v2 00/11] mm: scalable and unified arch_get_unmapped_area John Stoffel
2012-06-22 21:47 ` Andrew Morton
2012-06-23 16:03 ` John Stoffel
2012-06-22 15:01 ` Johannes Weiner
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=1340315835-28571-3-git-send-email-riel@surriel.com \
--to=riel@surriel.com \
--cc=aarcange@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=andi@firstfloor.org \
--cc=hannes@cmpxchg.org \
--cc=kosaki.motohiro@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mel@csn.ul.ie \
--cc=minchan@gmail.com \
--cc=peterz@infradead.org \
--cc=riel@redhat.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).