From:   Keith Busch <kbusch@meta.com>
To:     <linux-mm@kvack.org>, <linux-kernel@vger.kernel.org>,
        <willy@infradead.org>, <hch@lst.de>, <tonyb@cybernetics.com>,
        <akpm@linux-foundation.org>
CC:     <kernel-team@meta.com>, Keith Busch <kbusch@kernel.org>
Subject: [PATCHv4 10/12] dmapool: don't memset on free twice
Date:   Thu, 26 Jan 2023 13:51:23 -0800
Message-ID: <20230126215125.4069751-11-kbusch@meta.com>
In-Reply-To: <20230126215125.4069751-1-kbusch@meta.com>
References: <20230126215125.4069751-1-kbusch@meta.com>
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain
Precedence: bulk

From: Keith Busch <kbusch@kernel.org>

If debug is enabled, dmapool will poison the range, so no need to clear
it to 0 immediately before writing over it.

Signed-off-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
---
 mm/dmapool.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/mm/dmapool.c b/mm/dmapool.c
index 4dea2a0dbd336..21e6d362c7264 100644
--- a/mm/dmapool.c
+++ b/mm/dmapool.c
@@ -160,6 +160,8 @@ static void pool_check_block(struct dma_pool *pool, v=
oid *retval,
 static bool pool_page_err(struct dma_pool *pool, struct dma_page *page,
 			  void *vaddr, dma_addr_t dma)
 {
+	if (want_init_on_free())
+		memset(vaddr, 0, pool->size);
 	return false;
 }
=20
@@ -441,8 +443,6 @@ void dma_pool_free(struct dma_pool *pool, void *vaddr=
, dma_addr_t dma)
 		return;
 	}
=20
-	if (want_init_on_free())
-		memset(vaddr, 0, pool->size);
 	if (pool_page_err(pool, page, vaddr, dma)) {
 		spin_unlock_irqrestore(&pool->lock, flags);
 		return;
--=20
2.30.2