linux-kernel - [PATCH 3/7] Avoid taking waitqueue lock in dmapool

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Tue,  4 Dec 2007 13:26:04 -0800
From:	Matthew Wilcox <matthew@....cx>
To:	linux-kernel@...r.kernel.org
Cc:	Matthew Wilcox <matthew@....cx>,
	Matthew Wilcox <willy@...ux.intel.com>
Subject: [PATCH 3/7] Avoid taking waitqueue lock in dmapool

With one trivial change (taking the lock slightly earlier on wakeup
from schedule), all uses of the waitq are under the pool lock, so we
can use the locked (or __) versions of the wait queue functions, and
avoid the extra spinlock.

Signed-off-by: Matthew Wilcox <willy@...ux.intel.com>
Acked-by: David S. Miller <davem@...emloft.net>
---
 mm/dmapool.c |    9 +++++----
 1 files changed, 5 insertions(+), 4 deletions(-)

diff --git a/mm/dmapool.c b/mm/dmapool.c
index 92e886d..b5ff9ce 100644
--- a/mm/dmapool.c
+++ b/mm/dmapool.c
@@ -275,8 +275,8 @@ void *dma_pool_alloc(struct dma_pool *pool, gfp_t mem_flags,
 	size_t offset;
 	void *retval;
 
- restart:
 	spin_lock_irqsave(&pool->lock, flags);
+ restart:
 	list_for_each_entry(page, &pool->page_list, page_list) {
 		int i;
 		/* only cachable accesses here ... */
@@ -299,12 +299,13 @@ void *dma_pool_alloc(struct dma_pool *pool, gfp_t mem_flags,
 			DECLARE_WAITQUEUE(wait, current);
 
 			__set_current_state(TASK_INTERRUPTIBLE);
-			add_wait_queue(&pool->waitq, &wait);
+			__add_wait_queue(&pool->waitq, &wait);
 			spin_unlock_irqrestore(&pool->lock, flags);
 
 			schedule_timeout(POOL_TIMEOUT_JIFFIES);
 
-			remove_wait_queue(&pool->waitq, &wait);
+			spin_lock_irqsave(&pool->lock, flags);
+			__remove_wait_queue(&pool->waitq, &wait);
 			goto restart;
 		}
 		retval = NULL;
@@ -406,7 +407,7 @@ void dma_pool_free(struct dma_pool *pool, void *vaddr, dma_addr_t dma)
 	page->in_use--;
 	set_bit(block, &page->bitmap[map]);
 	if (waitqueue_active(&pool->waitq))
-		wake_up(&pool->waitq);
+		wake_up_locked(&pool->waitq);
 	/*
 	 * Resist a temptation to do
 	 *    if (!is_page_busy(bpp, page->bitmap)) pool_free_page(pool, page);
-- 
1.4.4.4
--
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/