lkml.org 
[lkml]   [2014]   [Sep]   [11]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
From
Subject[PATCH] mm: dmapool: add/remove sysfs file outside of the pool lock
Date
cat /sys/…/pools followed by removal the device leads to:

|======================================================
|[ INFO: possible circular locking dependency detected ]
|3.17.0-rc4+ #1498 Not tainted
|-------------------------------------------------------
|rmmod/2505 is trying to acquire lock:
| (s_active#28){++++.+}, at: [<c017f754>] kernfs_remove_by_name_ns+0x3c/0x88
|
|but task is already holding lock:
| (pools_lock){+.+.+.}, at: [<c011494c>] dma_pool_destroy+0x18/0x17c
|
|which lock already depends on the new lock.

The problem is the lock order of pools_lock and kernfs_mutex in
dma_pool_destroy() vs show_pools().

This patch breaks out the creation of the sysfs file outside of the
pools_lock mutex.
In theory we would have to create the link in the error path of
device_create_file() in case the dev->dma_pools list is not empty. In
reality I doubt that there will be a single device creating dma-pools in
parallel where it would matter.

Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
---
mm/dmapool.c | 29 +++++++++++++++++++++--------
1 file changed, 21 insertions(+), 8 deletions(-)

diff --git a/mm/dmapool.c b/mm/dmapool.c
index 306baa594f95..0cad8ee7891f 100644
--- a/mm/dmapool.c
+++ b/mm/dmapool.c
@@ -132,6 +132,7 @@ struct dma_pool *dma_pool_create(const char *name, struct device *dev,
{
struct dma_pool *retval;
size_t allocation;
+ bool empty = false;

if (align == 0) {
align = 1;
@@ -173,14 +174,22 @@ struct dma_pool *dma_pool_create(const char *name, struct device *dev,
INIT_LIST_HEAD(&retval->pools);

mutex_lock(&pools_lock);
- if (list_empty(&dev->dma_pools) &&
- device_create_file(dev, &dev_attr_pools)) {
- kfree(retval);
- return NULL;
- } else
- list_add(&retval->pools, &dev->dma_pools);
+ if (list_empty(&dev->dma_pools))
+ empty = true;
+ list_add(&retval->pools, &dev->dma_pools);
mutex_unlock(&pools_lock);
-
+ if (empty) {
+ int err;
+
+ err = device_create_file(dev, &dev_attr_pools);
+ if (err) {
+ mutex_lock(&pools_lock);
+ list_del(&retval->pools);
+ mutex_unlock(&pools_lock);
+ kfree(retval);
+ return NULL;
+ }
+ }
return retval;
}
EXPORT_SYMBOL(dma_pool_create);
@@ -251,11 +260,15 @@ static void pool_free_page(struct dma_pool *pool, struct dma_page *page)
*/
void dma_pool_destroy(struct dma_pool *pool)
{
+ bool empty = false;
+
mutex_lock(&pools_lock);
list_del(&pool->pools);
if (pool->dev && list_empty(&pool->dev->dma_pools))
- device_remove_file(pool->dev, &dev_attr_pools);
+ empty = true;
mutex_unlock(&pools_lock);
+ if (empty)
+ device_remove_file(pool->dev, &dev_attr_pools);

while (!list_empty(&pool->page_list)) {
struct dma_page *page;
--
2.1.0
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2014-09-11 22:01    [W:0.112 / U:0.536 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site