public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Badari Pulavarty <pbadari@us.ibm.com>
To: linux-kernel@vger.kernel.org
Cc: akpm@osdl.org, hch@lst.de, miklos@szeredi.hu, nathans@sgi.com,
	reiser@namesys.com, swhiteho@redhat.com, vs@namesys.com
Subject: [PATCH -mm only] Pass IO size to batch_write() address space operation
Date: Mon, 24 Jul 2006 10:50:29 -0700	[thread overview]
Message-ID: <44C50865.3000403@us.ibm.com> (raw)
In-Reply-To: <200607091208.k69C8umv021827@shell0.pdx.osdl.net>

[-- Attachment #1: Type: text/plain, Size: 281 bytes --]

Hi,

Here is the patch to pass size of the remaining IO to batch_write() 
interface.
I would like to use it ext3/ext4 to allocate in chunks. Currently its 
passing
only the size of the current buffer (in the vector entry).

Comments ? Flames ? Looks reasonable ?

Thanks,
Badari



[-- Attachment #2: pass_io_size_to_batch_write.patch --]
[-- Type: text/plain, Size: 1831 bytes --]

Pass remaining size of this IO to batch_write(). This
way filesystems could choose to allocate for the entire IO,
instead of current buffer size.

I would like to use this for setting ext3 reservation window
or allocating entire extent (ext4).


Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com>

---
 include/linux/fs.h |    2 ++
 mm/filemap.c       |    1 +
 2 files changed, 3 insertions(+)

Index: linux-2.6.18-rc1/include/linux/fs.h
===================================================================
--- linux-2.6.18-rc1.orig/include/linux/fs.h	2006-07-24 10:12:18.000000000 -0700
+++ linux-2.6.18-rc1/include/linux/fs.h	2006-07-24 10:35:51.000000000 -0700
@@ -356,6 +356,7 @@ struct writeback_control;
  * struct write_descriptor - set of write arguments
  * @pos: offset from the start of the file to write to
  * @count: number of bytes to write
+ * @iosize: remaining number of bytes in the IO
  * @buf: pointer to data to be written
  * @lru_pvec: multipage container to batch adding pages to LRU list
  * @cached_page: allocated but not used on previous call
@@ -366,6 +367,7 @@ struct writeback_control;
 struct write_descriptor {
 	loff_t pos;
 	size_t count;
+	size_t iosize;
 	char __user *buf;
 	struct page *cached_page;
 	struct pagevec *lru_pvec;
Index: linux-2.6.18-rc1/mm/filemap.c
===================================================================
--- linux-2.6.18-rc1.orig/mm/filemap.c	2006-07-24 10:12:20.000000000 -0700
+++ linux-2.6.18-rc1/mm/filemap.c	2006-07-24 10:36:44.000000000 -0700
@@ -2278,6 +2278,7 @@ generic_file_buffered_write(struct kiocb
 	do {
 		/* do not walk over current segment */
 		desc.buf = cur_iov->iov_base + iov_base;
+		desc.iosize = count;
 		desc.count = min(count, cur_iov->iov_len - iov_base);
 		if (desc.count > 0)
 			status = (*batch_write)(file, &desc, &copied);

           reply	other threads:[~2006-07-24 17:50 UTC|newest]

Thread overview: expand[flat|nested]  mbox.gz  Atom feed
 [parent not found: <200607091208.k69C8umv021827@shell0.pdx.osdl.net>]

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=44C50865.3000403@us.ibm.com \
    --to=pbadari@us.ibm.com \
    --cc=akpm@osdl.org \
    --cc=hch@lst.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=miklos@szeredi.hu \
    --cc=nathans@sgi.com \
    --cc=reiser@namesys.com \
    --cc=swhiteho@redhat.com \
    --cc=vs@namesys.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