From: clameter@sgi.com
To: linux-kernel@vger.kernel.org
Cc: Mel Gorman <mel@skynet.ie>,
William Lee Irwin III <wli@holomorphy.com>,
David Chinner <dgc@sgi.com>, Jens Axboe <jens.axboe@oracle.com>,
Badari Pulavarty <pbadari@gmail.com>,
Maxim Levitsky <maximlevitsky@gmail.com>
Subject: [10/17] Variable Order Page Cache: Add clearing and flushing function
Date: Tue, 24 Apr 2007 15:21:15 -0700 [thread overview]
Message-ID: <20070424222306.764423792@sgi.com> (raw)
In-Reply-To: 20070424222105.883597089@sgi.com
[-- Attachment #1: var_pc_flush_zero --]
[-- Type: text/plain, Size: 3803 bytes --]
Add a flushing and clearing function for higher order pages.
These are provisional and will likely have to be optimized.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
---
fs/libfs.c | 4 ++--
include/linux/highmem.h | 12 ++++++++++++
include/linux/pagemap.h | 25 +++++++++++++++++++++++++
mm/filemap.c | 4 ++--
4 files changed, 41 insertions(+), 4 deletions(-)
Index: linux-2.6.21-rc7/include/linux/pagemap.h
===================================================================
--- linux-2.6.21-rc7.orig/include/linux/pagemap.h 2007-04-24 11:41:35.000000000 -0700
+++ linux-2.6.21-rc7/include/linux/pagemap.h 2007-04-24 11:41:51.000000000 -0700
@@ -291,6 +291,31 @@ static inline void wait_on_page_writebac
extern void end_page_writeback(struct page *page);
+/* Support for clearing higher order pages */
+static inline void clear_mapping_page(struct page *page)
+{
+ int nr_pages = compound_pages(page);
+ int i;
+
+ for (i = 0; i < nr_pages; i++)
+ clear_highpage(page + i);
+}
+
+/*
+ * Support for flushing higher order pages.
+ *
+ * A bit stupid: On many platforms flushing the first page
+ * will flush any TLB starting there
+ */
+static inline void flush_mapping_page(struct page *page)
+{
+ int nr_pages = compound_pages(page);
+ int i;
+
+ for (i = 0; i < nr_pages; i++)
+ flush_dcache_page(page + i);
+}
+
/*
* Fault a userspace page into pagetables. Return non-zero on a fault.
*
Index: linux-2.6.21-rc7/fs/libfs.c
===================================================================
--- linux-2.6.21-rc7.orig/fs/libfs.c 2007-04-24 11:41:37.000000000 -0700
+++ linux-2.6.21-rc7/fs/libfs.c 2007-04-24 11:41:51.000000000 -0700
@@ -320,8 +320,8 @@ int simple_rename(struct inode *old_dir,
int simple_readpage(struct file *file, struct page *page)
{
- clear_highpage(page);
- flush_dcache_page(page);
+ clear_mapping_page(page);
+ flush_mapping_page(page);
SetPageUptodate(page);
unlock_page(page);
return 0;
Index: linux-2.6.21-rc7/include/linux/highmem.h
===================================================================
--- linux-2.6.21-rc7.orig/include/linux/highmem.h 2007-04-24 11:42:52.000000000 -0700
+++ linux-2.6.21-rc7/include/linux/highmem.h 2007-04-24 11:59:31.000000000 -0700
@@ -81,6 +81,17 @@ static inline void clear_highpage(struct
kunmap_atomic(kaddr, KM_USER0);
}
+#ifdef CONFIG_LARGE_BLOCKSIZE
+/*
+ * Macro needed because flush_mapping_page definition is not available yet
+ * CONFIG_LARGE_BLOCKSIZE means !HIGHMEM
+ */
+#define memclear_highpage_flush(__page,__offset,__size) \
+{ \
+ memset(page_address(__page), (__offset), (__size)); \
+ flush_mapping_page(__page); \
+}
+#else
/*
* Same but also flushes aliased cache contents to RAM.
*/
@@ -95,6 +106,7 @@ static inline void memclear_highpage_flu
flush_dcache_page(page);
kunmap_atomic(kaddr, KM_USER0);
}
+#endif
#ifndef __HAVE_ARCH_COPY_USER_HIGHPAGE
Index: linux-2.6.21-rc7/mm/filemap.c
===================================================================
--- linux-2.6.21-rc7.orig/mm/filemap.c 2007-04-24 11:41:37.000000000 -0700
+++ linux-2.6.21-rc7/mm/filemap.c 2007-04-24 11:59:36.000000000 -0700
@@ -925,7 +925,7 @@ page_ok:
* before reading the page on the kernel side.
*/
if (mapping_writably_mapped(mapping))
- flush_dcache_page(page);
+ flush_mapping_page(page);
/*
* When (part of) the same page is read multiple times
@@ -2139,7 +2139,7 @@ generic_file_buffered_write(struct kiocb
else
copied = filemap_copy_from_user_iovec(page, offset,
cur_iov, iov_base, bytes);
- flush_dcache_page(page);
+ flush_mapping_page(page);
status = a_ops->commit_write(file, page, offset, offset+bytes);
if (status == AOP_TRUNCATED_PAGE) {
page_cache_release(page);
--
next prev parent reply other threads:[~2007-04-24 22:23 UTC|newest]
Thread overview: 235+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-04-24 22:21 [00/17] Large Blocksize Support V3 clameter
2007-04-24 22:21 ` [01/17] Remove open coded implementation of memclear_highpage flush clameter
2007-04-24 22:21 ` [02/17] Fix page allocation flags in grow_dev_page() clameter
2007-04-24 22:21 ` [03/17] Fix: find_or_create_page does not spread memory clameter
2007-04-24 22:21 ` [04/17] Free up page->private for compound pages clameter
2007-04-24 22:21 ` [05/17] More compound page features clameter
2007-04-24 22:21 ` [06/17] Fix up handling of Compound head pages clameter
2007-04-24 22:21 ` [07/17] vmstat.c: Support accounting for compound pages clameter
2007-04-24 22:21 ` [08/17] Define functions for page cache handling clameter
2007-04-24 23:00 ` Eric Dumazet
2007-04-25 6:27 ` Christoph Lameter
2007-04-24 22:21 ` [09/17] Convert PAGE_CACHE_xxx -> page_cache_xxx function calls clameter
2007-04-24 22:21 ` clameter [this message]
2007-04-26 7:02 ` [10/17] Variable Order Page Cache: Add clearing and flushing function Christoph Lameter
2007-04-26 8:14 ` David Chinner
2007-04-24 22:21 ` [11/17] Readahead support for the variable order page cache clameter
2007-04-24 22:21 ` [12/17] Variable Page Cache Size: Fix up reclaim counters clameter
2007-04-24 22:21 ` [13/17] set_blocksize: Allow to set a larger block size than PAGE_SIZE clameter
2007-04-24 22:21 ` [14/17] Add VM_BUG_ONs to check for correct page order clameter
2007-04-24 22:21 ` [15/17] ramfs: Variable order page cache support clameter
2007-04-24 22:21 ` [16/17] ext2: " clameter
2007-04-24 22:21 ` [17/17] xfs: " clameter
2007-04-25 0:46 ` [00/17] Large Blocksize Support V3 Jörn Engel
2007-04-25 0:47 ` H. Peter Anvin
2007-04-25 3:11 ` William Lee Irwin III
2007-04-25 11:35 ` Jens Axboe
2007-04-25 15:36 ` Christoph Lameter
2007-04-25 17:53 ` Jens Axboe
2007-04-25 18:03 ` Christoph Lameter
2007-04-25 18:05 ` Jens Axboe
2007-04-25 18:14 ` Christoph Lameter
2007-04-25 18:16 ` Jens Axboe
2007-04-25 13:28 ` Mel Gorman
2007-04-25 15:23 ` Christoph Lameter
2007-04-25 22:46 ` Badari Pulavarty
2007-04-26 1:14 ` David Chinner
2007-04-26 1:17 ` David Chinner
2007-04-26 4:51 ` Eric W. Biederman
2007-04-26 5:05 ` Christoph Lameter
2007-04-26 5:44 ` Eric W. Biederman
2007-04-26 6:37 ` Christoph Lameter
2007-04-26 9:16 ` Mel Gorman
2007-04-26 6:38 ` Nick Piggin
2007-04-26 6:46 ` Christoph Lameter
2007-04-26 6:57 ` Nick Piggin
2007-04-26 7:10 ` Christoph Lameter
2007-04-26 7:22 ` Nick Piggin
2007-04-26 7:34 ` Christoph Lameter
2007-04-26 7:48 ` Nick Piggin
2007-04-26 9:20 ` David Chinner
2007-04-26 13:53 ` Avi Kivity
2007-04-26 14:33 ` David Chinner
2007-04-26 14:56 ` Avi Kivity
2007-04-26 15:20 ` Nick Piggin
2007-04-26 17:42 ` Jens Axboe
2007-04-26 18:59 ` Eric W. Biederman
2007-04-26 16:07 ` Christoph Hellwig
2007-04-27 10:05 ` Nick Piggin
2007-04-27 13:06 ` Mel Gorman
2007-04-26 13:50 ` William Lee Irwin III
2007-04-26 18:09 ` Eric W. Biederman
2007-04-26 23:34 ` William Lee Irwin III
2007-04-26 7:48 ` Questions on printk and console_drivers gshan
2007-04-26 10:06 ` [00/17] Large Blocksize Support V3 Mel Gorman
2007-04-26 14:47 ` Nick Piggin
2007-04-26 15:58 ` Christoph Hellwig
2007-04-26 16:05 ` Jens Axboe
2007-04-26 16:16 ` Christoph Hellwig
2007-04-26 13:28 ` Alan Cox
2007-04-26 13:30 ` Jens Axboe
2007-04-29 14:12 ` Matt Mackall
2007-04-28 10:55 ` Pierre Ossman
2007-04-28 15:39 ` Eric W. Biederman
2007-04-26 5:37 ` Nick Piggin
2007-04-26 6:38 ` David Chinner
2007-04-26 6:50 ` Nick Piggin
2007-04-26 8:40 ` Mel Gorman
2007-04-26 8:55 ` Nick Piggin
2007-04-26 10:30 ` Mel Gorman
2007-04-26 10:54 ` Eric W. Biederman
2007-04-26 12:23 ` Mel Gorman
2007-04-26 17:58 ` Christoph Lameter
2007-04-26 18:02 ` Jens Axboe
2007-04-26 16:11 ` Christoph Hellwig
2007-04-26 17:49 ` Eric W. Biederman
2007-04-26 18:03 ` Christoph Lameter
2007-04-26 18:03 ` Jens Axboe
2007-04-26 18:09 ` Christoph Hellwig
2007-04-26 18:12 ` Jens Axboe
2007-04-26 18:24 ` Christoph Hellwig
2007-04-26 18:24 ` Jens Axboe
2007-04-26 18:28 ` Christoph Lameter
2007-04-26 18:29 ` Jens Axboe
2007-04-26 18:35 ` Christoph Lameter
2007-04-26 18:39 ` Jens Axboe
2007-04-26 19:35 ` Eric W. Biederman
2007-04-26 19:42 ` Jens Axboe
2007-04-27 4:05 ` Eric W. Biederman
2007-04-27 10:26 ` Nick Piggin
2007-04-27 13:51 ` Eric W. Biederman
2007-04-26 20:22 ` Mel Gorman
2007-04-27 0:21 ` William Lee Irwin III
2007-04-27 5:16 ` Jens Axboe
2007-04-27 10:38 ` Nick Piggin
2007-04-26 10:10 ` Eric W. Biederman
2007-04-26 13:50 ` David Chinner
2007-04-26 14:40 ` William Lee Irwin III
2007-04-26 15:38 ` Nick Piggin
2007-04-26 15:58 ` William Lee Irwin III
2007-04-27 9:46 ` Nick Piggin
2007-04-27 0:19 ` Jeremy Higdon
2007-04-26 18:07 ` Christoph Lameter
2007-04-26 18:45 ` Eric W. Biederman
2007-04-26 18:59 ` Christoph Lameter
2007-04-26 19:21 ` Eric W. Biederman
2007-04-26 6:40 ` Christoph Lameter
2007-04-26 6:53 ` Nick Piggin
2007-04-26 7:04 ` David Chinner
2007-04-26 7:07 ` Nick Piggin
2007-04-26 7:11 ` Christoph Lameter
2007-04-26 7:17 ` Nick Piggin
2007-04-26 7:28 ` Christoph Lameter
2007-04-26 7:45 ` Nick Piggin
2007-04-26 18:10 ` Christoph Lameter
2007-04-27 10:08 ` Nick Piggin
2007-04-26 7:07 ` Christoph Lameter
2007-04-26 7:15 ` Nick Piggin
2007-04-26 7:22 ` Christoph Lameter
2007-04-26 7:42 ` Nick Piggin
2007-04-26 10:48 ` Mel Gorman
2007-04-26 12:37 ` Andy Whitcroft
2007-04-26 14:18 ` David Chinner
2007-04-26 15:08 ` Nick Piggin
2007-04-26 15:19 ` William Lee Irwin III
2007-04-26 15:28 ` David Chinner
2007-04-26 14:53 ` William Lee Irwin III
2007-04-26 18:16 ` Christoph Lameter
2007-04-26 18:21 ` Eric W. Biederman
2007-04-27 0:32 ` William Lee Irwin III
2007-04-27 10:22 ` Nick Piggin
2007-04-27 12:58 ` William Lee Irwin III
2007-04-27 13:06 ` Nick Piggin
2007-04-27 14:49 ` William Lee Irwin III
2007-04-26 18:13 ` Christoph Lameter
2007-04-27 10:15 ` Nick Piggin
2007-04-26 14:49 ` William Lee Irwin III
2007-04-26 18:50 ` Maxim Levitsky
2007-04-27 2:04 ` Andrew Morton
2007-04-27 2:27 ` David Chinner
2007-04-27 2:53 ` Andrew Morton
2007-04-27 3:47 ` [00/17] Large Blocksize Support V3 (mmap conceptual discussion) Christoph Lameter
2007-04-27 4:20 ` [00/17] Large Blocksize Support V3 David Chinner
2007-04-27 5:15 ` Andrew Morton
2007-04-27 5:49 ` Christoph Lameter
2007-04-27 6:55 ` Andrew Morton
2007-04-27 7:19 ` Christoph Lameter
2007-04-27 7:26 ` Andrew Morton
2007-04-27 8:37 ` David Chinner
2007-04-27 12:01 ` Christoph Lameter
2007-04-27 16:36 ` David Chinner
2007-04-27 17:34 ` David Chinner
2007-04-27 19:11 ` Andrew Morton
2007-04-28 1:43 ` Nick Piggin
2007-04-28 8:04 ` Peter Zijlstra
2007-04-28 8:22 ` Andrew Morton
2007-04-28 8:32 ` Peter Zijlstra
2007-04-28 8:55 ` Andrew Morton
2007-04-28 9:36 ` Peter Zijlstra
2007-04-28 14:09 ` William Lee Irwin III
2007-04-28 18:26 ` Andrew Morton
2007-04-28 19:19 ` William Lee Irwin III
2007-04-28 21:28 ` Andrew Morton
2007-04-28 3:17 ` David Chinner
2007-04-28 3:49 ` Christoph Lameter
2007-04-28 4:56 ` Andrew Morton
2007-04-28 5:08 ` Christoph Lameter
2007-04-28 5:36 ` Andrew Morton
2007-04-28 6:24 ` Christoph Lameter
2007-04-28 6:52 ` Andrew Morton
2007-04-30 5:30 ` Christoph Lameter
2007-04-28 9:43 ` Alan Cox
2007-04-28 9:58 ` Andrew Morton
2007-04-28 10:21 ` Alan Cox
2007-04-28 10:25 ` Andrew Morton
2007-04-28 11:29 ` Alan Cox
2007-04-28 14:37 ` William Lee Irwin III
2007-04-27 7:22 ` Christoph Lameter
2007-04-27 7:29 ` Andrew Morton
2007-04-27 7:35 ` Christoph Lameter
2007-04-27 7:43 ` Andrew Morton
2007-04-27 11:05 ` Paul Mackerras
2007-04-27 11:41 ` Nick Piggin
2007-04-27 12:12 ` Christoph Lameter
2007-04-27 12:25 ` Nick Piggin
2007-04-27 13:39 ` Christoph Hellwig
2007-04-28 2:27 ` Nick Piggin
2007-04-28 2:39 ` William Lee Irwin III
2007-04-28 2:50 ` Nick Piggin
2007-04-28 3:16 ` William Lee Irwin III
2007-04-28 8:16 ` Christoph Hellwig
2007-04-27 16:48 ` Christoph Lameter
2007-04-27 13:37 ` Christoph Hellwig
2007-04-27 12:14 ` Paul Mackerras
2007-04-27 12:36 ` Nick Piggin
2007-04-27 13:42 ` Christoph Hellwig
2007-04-27 11:58 ` Christoph Lameter
2007-04-27 13:44 ` William Lee Irwin III
2007-04-27 19:15 ` Andrew Morton
2007-04-28 2:21 ` William Lee Irwin III
2007-04-27 6:09 ` David Chinner
2007-04-27 7:04 ` Andrew Morton
2007-04-27 8:03 ` David Chinner
2007-04-27 8:48 ` Andrew Morton
2007-04-27 16:45 ` Theodore Tso
2007-05-04 13:33 ` Eric W. Biederman
2007-05-07 4:29 ` David Chinner
2007-05-07 4:48 ` Eric W. Biederman
2007-05-07 5:27 ` David Chinner
2007-05-07 6:43 ` Eric W. Biederman
2007-05-07 6:49 ` William Lee Irwin III
2007-05-07 7:06 ` William Lee Irwin III
2007-05-08 8:49 ` William Lee Irwin III
2007-05-07 16:06 ` Christoph Lameter
2007-05-07 17:29 ` William Lee Irwin III
2007-05-04 12:57 ` Eric W. Biederman
2007-05-04 13:31 ` Eric W. Biederman
2007-05-04 16:11 ` Christoph Lameter
2007-05-07 4:58 ` David Chinner
2007-05-07 6:56 ` Eric W. Biederman
2007-05-07 15:17 ` Weigert, Daniel
2007-04-27 16:55 ` Theodore Tso
2007-04-27 17:32 ` Nicholas Miell
2007-04-27 18:12 ` William Lee Irwin III
2007-04-28 16:39 ` Maxim Levitsky
2007-04-30 5:23 ` Christoph Lameter
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=20070424222306.764423792@sgi.com \
--to=clameter@sgi.com \
--cc=dgc@sgi.com \
--cc=jens.axboe@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=maximlevitsky@gmail.com \
--cc=mel@skynet.ie \
--cc=pbadari@gmail.com \
--cc=wli@holomorphy.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