From: Suparna Bhattacharya <suparna@in.ibm.com>
To: akpm@osdl.org
Cc: linux-kernel@vger.kernel.org
Subject: Re: [PATCH 2/5] Fix writeback page range to use exact limits
Date: Sun, 1 Aug 2004 13:19:35 +0530 [thread overview]
Message-ID: <20040801074935.GB7327@in.ibm.com> (raw)
In-Reply-To: <20040801074518.GA7310@in.ibm.com>
On Sun, Aug 01, 2004 at 01:15:18PM +0530, Suparna Bhattacharya wrote:
>
> The attached patches (generated against 2.6.8-rc2) enable concurrent
> O_SYNC writers to different parts of the same file by avoiding
> serialising on i_sem across the wait for IO completion.
>
> This is mostly your work, ported to the tagged radix tree VFS changes
> and a few fixes. I have been carrying these patches for sometime now;
> they can be the merged upstream. Please apply.
>
[2] fix-writeback-range.patch
Regards
Suparna
--
Suparna Bhattacharya (suparna@in.ibm.com)
Linux Technology Center
IBM Software Lab, India
---------------------------------------------------------------
wait_on_page_writeback_range shouldn't wait for pages beyond the
specified range. Ideally, the radix-tree-lookup could accept an
end_index parameter so that it doesn't return the extra pages
in the first place, but for now we just add a few extra checks
to skip such pages.
Signed-off-by: Suparna Bhattacharya <suparna@in.ibm.com>
linux-2.6.8-rc2-suparna/mm/filemap.c | 7 ++++++-
1 files changed, 6 insertions(+), 1 deletion(-)
diff -puN mm/filemap.c~fix-writeback-range mm/filemap.c
--- linux-2.6.8-rc2/mm/filemap.c~fix-writeback-range 2004-08-01 12:32:04.000000000 +0530
+++ linux-2.6.8-rc2-suparna/mm/filemap.c 2004-08-01 12:32:04.000000000 +0530
@@ -198,7 +198,8 @@ static int wait_on_page_writeback_range(
pagevec_init(&pvec, 0);
index = start;
- while ((nr_pages = pagevec_lookup_tag(&pvec, mapping, &index,
+ while ((index <= end) &&
+ (nr_pages = pagevec_lookup_tag(&pvec, mapping, &index,
PAGECACHE_TAG_WRITEBACK,
min(end - index, (pgoff_t)PAGEVEC_SIZE-1) + 1)) != 0) {
unsigned i;
@@ -206,6 +207,10 @@ static int wait_on_page_writeback_range(
for (i = 0; i < nr_pages; i++) {
struct page *page = pvec.pages[i];
+ /* until radix tree lookup accepts end_index */
+ if (page->index > end) {
+ continue;
+ }
wait_on_page_writeback(page);
if (PageError(page))
ret = -EIO;
_
next prev parent reply other threads:[~2004-08-01 7:40 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-08-01 7:45 [PATCH 0/5] Concurrent O_SYNC write speedups using radix-tree walks Suparna Bhattacharya
2004-08-01 7:47 ` [PATCH 1/5] Writeback page range hint Suparna Bhattacharya
2004-08-01 7:49 ` Suparna Bhattacharya [this message]
2004-08-01 7:50 ` [PATCH 3/5] mpage writepages range limit fix Suparna Bhattacharya
2004-08-01 7:52 ` [PATCH 4/5] filemap_fdatawrite range interface Suparna Bhattacharya
2004-08-01 7:53 ` [PATCH 5/5] Concurrent O_SYNC write support Suparna Bhattacharya
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=20040801074935.GB7327@in.ibm.com \
--to=suparna@in.ibm.com \
--cc=akpm@osdl.org \
--cc=linux-kernel@vger.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
Powered by JetHome