From: Ioana Radulescu <ruxandra.radulescu@nxp.com>
To: <gregkh@linuxfoundation.org>
CC: <devel@driverdev.osuosl.org>, <linux-kernel@vger.kernel.org>,
        <bogdan.purcareata@nxp.com>
Subject: [PATCH 6/6] staging: fsl-dpaa2/eth: Make Tx buffer alignment optional
Date: Fri, 8 Dec 2017 06:47:58 -0600
Message-ID: <20171208124758.2732-7-ruxandra.radulescu@nxp.com>
In-Reply-To: <20171208124758.2732-1-ruxandra.radulescu@nxp.com>
References: <20171208124758.2732-1-ruxandra.radulescu@nxp.com>
MIME-Version: 1.0
Content-Type: text/plain
SpamDiagnosticOutput: 1:99
SpamDiagnosticMetadata: NSPM
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 08 Dec 2017 12:48:08.3246
 (UTC)
X-MS-Exchange-CrossTenant-Network-Message-Id: 898f6cdc-e4a3-44b2-8580-08d53e39eff1
X-MS-Exchange-CrossTenant-Id: 5afe0b00-7697-4969-b663-5eab37d5f47e
X-MS-Exchange-CrossTenant-OriginalAttributedTenantConnectingIp: TenantId=5afe0b00-7697-4969-b663-5eab37d5f47e;Ip=[192.88.168.50];Helo=[tx30smr01.am.freescale.net]
X-MS-Exchange-CrossTenant-FromEntityHeader: HybridOnPrem
X-MS-Exchange-Transport-CrossTenantHeadersStamped: DM5PR03MB2697
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org
Content-Length: 2215
Lines: 58

Aligning the Tx buffers at 64B is a performance optimization
recommendation, not a hard requirement.

Make optional the alignment of Tx FD buffers, without enforcing
a reallocation in case there is not enough headroom for it.

On Rx, we keep allocating buffers with enough headroom to allow
Tx alignment of forwarded frames.

Signed-off-by: Ioana Radulescu <ruxandra.radulescu@nxp.com>
---
 drivers/staging/fsl-dpaa2/ethernet/dpaa2-eth.c | 14 ++++++++++----
 drivers/staging/fsl-dpaa2/ethernet/dpaa2-eth.h |  2 +-
 2 files changed, 11 insertions(+), 5 deletions(-)

diff --git a/drivers/staging/fsl-dpaa2/ethernet/dpaa2-eth.c b/drivers/staging/fsl-dpaa2/ethernet/dpaa2-eth.c
index 8edb7c1..7f3e4fa 100644
--- a/drivers/staging/fsl-dpaa2/ethernet/dpaa2-eth.c
+++ b/drivers/staging/fsl-dpaa2/ethernet/dpaa2-eth.c
@@ -433,13 +433,19 @@ static int build_single_fd(struct dpaa2_eth_priv *priv,
 			   struct dpaa2_fd *fd)
 {
 	struct device *dev = priv->net_dev->dev.parent;
-	u8 *buffer_start;
+	u8 *buffer_start, *aligned_start;
 	struct sk_buff **skbh;
 	dma_addr_t addr;
 
-	buffer_start = PTR_ALIGN(skb->data -
-				 dpaa2_eth_needed_headroom(priv, skb),
-				 DPAA2_ETH_TX_BUF_ALIGN);
+	buffer_start = skb->data - dpaa2_eth_needed_headroom(priv, skb);
+
+	/* If there's enough room to align the FD address, do it.
+	 * It will help hardware optimize accesses.
+	 */
+	aligned_start = PTR_ALIGN(buffer_start - DPAA2_ETH_TX_BUF_ALIGN,
+				  DPAA2_ETH_TX_BUF_ALIGN);
+	if (aligned_start >= skb->head)
+		buffer_start = aligned_start;
 
 	/* Store a backpointer to the skb at the beginning of the buffer
 	 * (in the private data area) such that we can release it
diff --git a/drivers/staging/fsl-dpaa2/ethernet/dpaa2-eth.h b/drivers/staging/fsl-dpaa2/ethernet/dpaa2-eth.h
index 4ea41d8..d68ac38 100644
--- a/drivers/staging/fsl-dpaa2/ethernet/dpaa2-eth.h
+++ b/drivers/staging/fsl-dpaa2/ethernet/dpaa2-eth.h
@@ -370,7 +370,7 @@ unsigned int dpaa2_eth_needed_headroom(struct dpaa2_eth_priv *priv,
 	if (skb_is_nonlinear(skb))
 		return 0;
 
-	return DPAA2_ETH_SWA_SIZE + DPAA2_ETH_TX_BUF_ALIGN;
+	return DPAA2_ETH_SWA_SIZE;
 }
 
 /* Extra headroom space requested to hardware, in order to make sure there's
-- 
2.7.4