From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1760035AbYFGI3j (ORCPT ); Sat, 7 Jun 2008 04:29:39 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1755509AbYFGI27 (ORCPT ); Sat, 7 Jun 2008 04:28:59 -0400 Received: from [78.13.70.189] ([78.13.70.189]:47943 "EHLO linux.localdomain" rhost-flags-FAIL-FAIL-OK-FAIL) by vger.kernel.org with ESMTP id S1754152AbYFGI2y (ORCPT ); Sat, 7 Jun 2008 04:28:54 -0400 From: Andrea Righi To: balbir@linux.vnet.ibm.com, menage@google.com Cc: matt@bluehost.com, roberto@unbit.it, randy.dunlap@oracle.com, akpm@linux-foundation.org, linux-kernel@vger.kernel.org Subject: [PATCH 3/3] i/o accounting and control Date: Sat, 7 Jun 2008 00:27:30 +0200 Message-Id: <1212791250-32320-4-git-send-email-righi.andrea@gmail.com> X-Mailer: git-send-email 1.5.4.3 In-Reply-To: <> References: <> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Apply the io-throttle controller to the opportune kernel functions. Both accounting and throttling functionalities are performed by cgroup_io_account(). Signed-off-by: Andrea Righi --- block/blk-core.c | 2 ++ fs/buffer.c | 11 +++++++++++ fs/direct-io.c | 3 +++ mm/page-writeback.c | 10 ++++++++++ mm/readahead.c | 5 +++++ 5 files changed, 31 insertions(+), 0 deletions(-) diff --git a/block/blk-core.c b/block/blk-core.c index 1905aab..bf56046 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -26,6 +26,7 @@ #include #include #include +#include #include #include #include @@ -1482,6 +1483,7 @@ void submit_bio(int rw, struct bio *bio) count_vm_events(PGPGOUT, count); } else { task_io_account_read(bio->bi_size); + cgroup_io_account(bio->bi_bdev, bio->bi_size); count_vm_events(PGPGIN, count); } diff --git a/fs/buffer.c b/fs/buffer.c index a073f3f..700316d 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -35,6 +35,7 @@ #include #include #include +#include #include #include #include @@ -700,6 +701,9 @@ EXPORT_SYMBOL(mark_buffer_dirty_inode); static int __set_page_dirty(struct page *page, struct address_space *mapping, int warn) { + struct block_device *bdev = NULL; + size_t cgroup_io_acct = 0; + if (unlikely(!mapping)) return !TestSetPageDirty(page); @@ -711,16 +715,23 @@ static int __set_page_dirty(struct page *page, WARN_ON_ONCE(warn && !PageUptodate(page)); if (mapping_cap_account_dirty(mapping)) { + bdev = (mapping->host && + mapping->host->i_sb->s_bdev) ? + mapping->host->i_sb->s_bdev : NULL; + __inc_zone_page_state(page, NR_FILE_DIRTY); __inc_bdi_stat(mapping->backing_dev_info, BDI_RECLAIMABLE); task_io_account_write(PAGE_CACHE_SIZE); + cgroup_io_acct = PAGE_CACHE_SIZE; } radix_tree_tag_set(&mapping->page_tree, page_index(page), PAGECACHE_TAG_DIRTY); } write_unlock_irq(&mapping->tree_lock); __mark_inode_dirty(mapping->host, I_DIRTY_PAGES); + if (cgroup_io_acct) + cgroup_io_account(bdev, cgroup_io_acct); return 1; } diff --git a/fs/direct-io.c b/fs/direct-io.c index 9e81add..9e5d783 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -35,6 +35,7 @@ #include #include #include +#include #include /* @@ -666,6 +667,8 @@ submit_page_section(struct dio *dio, struct page *page, /* * Read accounting is performed in submit_bio() */ + struct block_device *bdev = dio->bio ? dio->bio->bi_bdev : NULL; + cgroup_io_account(bdev, len); task_io_account_write(len); } diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 789b6ad..aa74fba 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include #include @@ -1077,6 +1078,8 @@ int __set_page_dirty_nobuffers(struct page *page) if (!TestSetPageDirty(page)) { struct address_space *mapping = page_mapping(page); struct address_space *mapping2; + struct block_device *bdev = NULL; + size_t cgroup_io_acct = 0; if (!mapping) return 1; @@ -1087,10 +1090,15 @@ int __set_page_dirty_nobuffers(struct page *page) BUG_ON(mapping2 != mapping); WARN_ON_ONCE(!PagePrivate(page) && !PageUptodate(page)); if (mapping_cap_account_dirty(mapping)) { + bdev = (mapping->host && + mapping->host->i_sb->s_bdev) ? + mapping->host->i_sb->s_bdev : NULL; + __inc_zone_page_state(page, NR_FILE_DIRTY); __inc_bdi_stat(mapping->backing_dev_info, BDI_RECLAIMABLE); task_io_account_write(PAGE_CACHE_SIZE); + cgroup_io_acct = PAGE_CACHE_SIZE; } radix_tree_tag_set(&mapping->page_tree, page_index(page), PAGECACHE_TAG_DIRTY); @@ -1100,6 +1108,8 @@ int __set_page_dirty_nobuffers(struct page *page) /* !PageAnon && !swapper_space */ __mark_inode_dirty(mapping->host, I_DIRTY_PAGES); } + if (cgroup_io_acct) + cgroup_io_account(bdev, cgroup_io_acct); return 1; } return 0; diff --git a/mm/readahead.c b/mm/readahead.c index d8723a5..9fdfae8 100644 --- a/mm/readahead.c +++ b/mm/readahead.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #include @@ -58,6 +59,9 @@ int read_cache_pages(struct address_space *mapping, struct list_head *pages, int (*filler)(void *, struct page *), void *data) { struct page *page; + struct block_device *bdev = + (mapping->host && mapping->host->i_sb->s_bdev) ? + mapping->host->i_sb->s_bdev : NULL; int ret = 0; while (!list_empty(pages)) { @@ -76,6 +80,7 @@ int read_cache_pages(struct address_space *mapping, struct list_head *pages, break; } task_io_account_read(PAGE_CACHE_SIZE); + cgroup_io_account(bdev, PAGE_CACHE_SIZE); } return ret; } -- 1.5.4.3