[<prev] [next>] [day] [month] [year] [list]
Message-Id: <1219853257-11052-5-git-send-email-righi.andrea@gmail.com>
Date: Wed, 27 Aug 2008 18:07:36 +0200
From: Andrea Righi <righi.andrea@...il.com>
To: Balbir Singh <balbir@...ux.vnet.ibm.com>,
Paul Menage <menage@...gle.com>
Cc: agk@...rceware.org, akpm@...ux-foundation.org, axboe@...nel.dk,
baramsori72@...il.com, Carl Henrik Lunde <chlunde@...g.uio.no>,
dave@...ux.vnet.ibm.com, Divyesh Shah <dpshah@...gle.com>,
eric.rannaud@...il.com, fernando@....ntt.co.jp,
Hirokazu Takahashi <taka@...inux.co.jp>,
Li Zefan <lizf@...fujitsu.com>,
Marco Innocenti <m.innocenti@...eca.it>, matt@...ehost.com,
ngupta@...gle.com, randy.dunlap@...cle.com, roberto@...it.it,
Ryo Tsuruta <ryov@...inux.co.jp>,
Satoshi UCHIDA <s-uchida@...jp.nec.com>,
subrata@...ux.vnet.ibm.com, yoshikawa.takuya@....ntt.co.jp,
containers@...ts.linux-foundation.org,
linux-kernel@...r.kernel.org, Andrea Righi <righi.andrea@...il.com>
Subject: [RFC][PATCH -mm 4/5] i/o controller instrumentation: accounting and throttling
Apply the io-throttle controller to the opportune kernel functions. Both
accounting and throttling functionalities are performed by
cgroup_io_throttle().
Signed-off-by: Andrea Righi <righi.andrea@...il.com>
---
block/blk-core.c | 2 ++
fs/aio.c | 14 ++++++++++++++
include/linux/sched.h | 5 +++++
kernel/fork.c | 6 +++++-
mm/page-writeback.c | 19 +++++++++++++++++++
mm/readahead.c | 5 +++++
6 files changed, 50 insertions(+), 1 deletions(-)
diff --git a/block/blk-core.c b/block/blk-core.c
index fef79cc..a04f230 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -26,6 +26,7 @@
#include <linux/swap.h>
#include <linux/writeback.h>
#include <linux/task_io_accounting_ops.h>
+#include <linux/blk-io-throttle.h>
#include <linux/interrupt.h>
#include <linux/cpu.h>
#include <linux/blktrace_api.h>
@@ -1490,6 +1491,7 @@ void submit_bio(int rw, struct bio *bio)
(unsigned long long)bio->bi_sector,
bdevname(bio->bi_bdev, b));
}
+ cgroup_io_throttle(bio->bi_bdev, bio->bi_size, 1);
}
generic_make_request(bio);
diff --git a/fs/aio.c b/fs/aio.c
index f658441..6937dfb 100644
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -22,6 +22,7 @@
#include <linux/sched.h>
#include <linux/fs.h>
#include <linux/file.h>
+#include <linux/blk-io-throttle.h>
#include <linux/mm.h>
#include <linux/mman.h>
#include <linux/slab.h>
@@ -1558,6 +1559,8 @@ static int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb,
{
struct kiocb *req;
struct file *file;
+ struct block_device *bdev;
+ struct inode *inode;
ssize_t ret;
/* enforce forwards compatibility on users */
@@ -1580,6 +1583,15 @@ static int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb,
if (unlikely(!file))
return -EBADF;
+ /* check if we're exceeding the IO throttling limits */
+ inode = file->f_mapping->host;
+ bdev = inode->i_sb->s_bdev;
+ ret = cgroup_io_throttle(bdev, 0, 0);
+ if (unlikely(ret)) {
+ fput(file);
+ return -EAGAIN;
+ }
+
req = aio_get_req(ctx); /* returns with 2 references to req */
if (unlikely(!req)) {
fput(file);
@@ -1622,12 +1634,14 @@ static int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb,
goto out_put_req;
spin_lock_irq(&ctx->ctx_lock);
+ set_in_aio();
aio_run_iocb(req);
if (!list_empty(&ctx->run_list)) {
/* drain the run list */
while (__aio_run_iocbs(ctx))
;
}
+ unset_in_aio();
spin_unlock_irq(&ctx->ctx_lock);
aio_put_req(req); /* drop extra ref to req */
return 0;
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 1b26ed2..4b1d69e 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1253,6 +1253,11 @@ struct task_struct {
unsigned long ptrace_message;
siginfo_t *last_siginfo; /* For ptrace use. */
struct task_io_accounting ioac;
+#ifdef CONFIG_CGROUP_IO_THROTTLE
+ atomic_t in_aio;
+ unsigned long long io_throttle_cnt;
+ unsigned long long io_throttle_sleep;
+#endif
#if defined(CONFIG_TASK_XACCT)
u64 acct_rss_mem1; /* accumulated rss usage */
u64 acct_vm_mem1; /* accumulated virtual memory usage */
diff --git a/kernel/fork.c b/kernel/fork.c
index 63a87ff..9ee7408 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1011,7 +1011,11 @@ static struct task_struct *copy_process(unsigned long clone_flags,
task_io_accounting_init(&p->ioac);
acct_clear_integrals(p);
-
+#ifdef CONFIG_CGROUP_IO_THROTTLE
+ atomic_set(&p->in_aio, 0);
+ p->io_throttle_cnt = 0;
+ p->io_throttle_sleep = 0;
+#endif
p->it_virt_expires = cputime_zero;
p->it_prof_expires = cputime_zero;
p->it_sched_expires = 0;
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index c6d6088..84b6b9a 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -23,6 +23,7 @@
#include <linux/init.h>
#include <linux/backing-dev.h>
#include <linux/task_io_accounting_ops.h>
+#include <linux/blk-io-throttle.h>
#include <linux/blkdev.h>
#include <linux/mpage.h>
#include <linux/rmap.h>
@@ -556,6 +557,24 @@ void balance_dirty_pages_ratelimited_nr(struct address_space *mapping,
static DEFINE_PER_CPU(unsigned long, ratelimits) = 0;
unsigned long ratelimit;
unsigned long *p;
+ struct block_device *bdev = (mapping->host &&
+ mapping->host->i_sb->s_bdev) ?
+ mapping->host->i_sb->s_bdev : NULL;
+ /*
+ * If we're dirtying pages on IO limited devices, force the current
+ * task to actively flush dirty pages back to the device.
+ *
+ * The throttling will be performed in submit_bio() in the same IO
+ * context of the current task.
+ */
+ if (is_io_throttled(bdev)) {
+ struct backing_dev_info *bdi = mapping->backing_dev_info;
+
+ if (writeback_in_progress(bdi))
+ return;
+ background_writeout(nr_pages_dirtied);
+ return;
+ }
ratelimit = ratelimit_pages;
if (mapping->backing_dev_info->dirty_exceeded)
diff --git a/mm/readahead.c b/mm/readahead.c
index 137bc56..448f065 100644
--- a/mm/readahead.c
+++ b/mm/readahead.c
@@ -14,6 +14,7 @@
#include <linux/blkdev.h>
#include <linux/backing-dev.h>
#include <linux/task_io_accounting_ops.h>
+#include <linux/blk-io-throttle.h>
#include <linux/pagevec.h>
#include <linux/pagemap.h>
@@ -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_throttle(bdev, PAGE_CACHE_SIZE, 1);
}
return ret;
}
--
1.5.4.3
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists