[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20251117101921.1862427-5-wei.fang@nxp.com>
Date: Mon, 17 Nov 2025 18:19:20 +0800
From: Wei Fang <wei.fang@....com>
To: shenwei.wang@....com,
xiaoning.wang@....com,
andrew+netdev@...n.ch,
davem@...emloft.net,
edumazet@...gle.com,
kuba@...nel.org,
pabeni@...hat.com,
eric@...int.com,
Frank.Li@....com
Cc: imx@...ts.linux.dev,
netdev@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: [PATCH v2 net-next 4/5] net: fec: remove rx_align from fec_enet_private
The rx_align was introduced by the commit 41ef84ce4c72 ("net: fec: change
FEC alignment according to i.mx6 sx requirement"). Because the i.MX6 SX
requires RX buffer must be 64 bytes alignment.
Since the commit 95698ff6177b ("net: fec: using page pool to manage RX
buffers"), the address of the RX buffer is always the page address plus
FEC_ENET_XDP_HEADROOM which is 256 bytes, so the RX buffer is always
64-byte aligned. Therefore, rx_align has no effect since that commit,
and we can safely remove it.
In addition, to prevent future modifications to FEC_ENET_XDP_HEADROOM,
a BUILD_BUG_ON() test has been added to the driver, which ensures that
FEC_ENET_XDP_HEADROOM provides the required alignment.
Signed-off-by: Wei Fang <wei.fang@....com>
---
drivers/net/ethernet/freescale/fec.h | 1 -
drivers/net/ethernet/freescale/fec_main.c | 19 ++++++++++++++-----
2 files changed, 14 insertions(+), 6 deletions(-)
diff --git a/drivers/net/ethernet/freescale/fec.h b/drivers/net/ethernet/freescale/fec.h
index c5bbc2c16a4f..a25dca9c7d71 100644
--- a/drivers/net/ethernet/freescale/fec.h
+++ b/drivers/net/ethernet/freescale/fec.h
@@ -660,7 +660,6 @@ struct fec_enet_private {
struct pm_qos_request pm_qos_req;
unsigned int tx_align;
- unsigned int rx_align;
/* hw interrupt coalesce */
unsigned int rx_pkts_itr;
diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c
index 1408e3e6650a..785bae8e1699 100644
--- a/drivers/net/ethernet/freescale/fec_main.c
+++ b/drivers/net/ethernet/freescale/fec_main.c
@@ -3435,6 +3435,19 @@ fec_enet_alloc_rxq_buffers(struct net_device *ndev, unsigned int queue)
return err;
}
+ /* Some platforms require the RX buffer must be 64 bytes alignment.
+ * Some platforms require 16 bytes alignment. And some platforms
+ * require 4 bytes alignment. But since the page pool have been
+ * introduced into the driver, the address of RX buffer is always
+ * the page address plus FEC_ENET_XDP_HEADROOM, and
+ * FEC_ENET_XDP_HEADROOM is 256 bytes. Therefore, this address can
+ * satisfy all platforms. To prevent future modifications to
+ * FEC_ENET_XDP_HEADROOM from ignoring this hardware limitation, a
+ * BUILD_BUG_ON() test has been added, which ensures that
+ * FEC_ENET_XDP_HEADROOM provides the required alignment.
+ */
+ BUILD_BUG_ON(FEC_ENET_XDP_HEADROOM & 0x3f);
+
for (i = 0; i < rxq->bd.ring_size; i++) {
page = page_pool_dev_alloc_pages(rxq->page_pool);
if (!page)
@@ -4069,10 +4082,8 @@ static int fec_enet_init(struct net_device *ndev)
WARN_ON(dsize != (1 << dsize_log2));
#if defined(CONFIG_ARM) || defined(CONFIG_ARM64)
- fep->rx_align = 0xf;
fep->tx_align = 0xf;
#else
- fep->rx_align = 0x3;
fep->tx_align = 0x3;
#endif
fep->rx_pkts_itr = FEC_ITR_ICFT_DEFAULT;
@@ -4161,10 +4172,8 @@ static int fec_enet_init(struct net_device *ndev)
fep->csum_flags |= FLAG_RX_CSUM_ENABLED;
}
- if (fep->quirks & FEC_QUIRK_HAS_MULTI_QUEUES) {
+ if (fep->quirks & FEC_QUIRK_HAS_MULTI_QUEUES)
fep->tx_align = 0;
- fep->rx_align = 0x3f;
- }
ndev->hw_features = ndev->features;
--
2.34.1
Powered by blists - more mailing lists