[net-next] net: fec: Allow reception of frames bigger than 1522 bytes

Message ID 1501436165-20259-1-git-send-email-andrew@lunn.ch
State Accepted
Delegated to: David Miller
Headers show

Commit Message

Andrew Lunn July 30, 2017, 5:36 p.m.
The FEC Receive Control Register has a 14 bit field indicating the
longest frame that my be received. It is being set to 1522. Frames
longer than this are discarded, but counted as being in error.

When using DSA, frames from the switch has an additional header,
either 4 or 8 bytes if a Marvell switch is used. Thus a full MTU frame
of 1522 bytes received by the switch on a port becomes 1530 bytes when
passed to the host via the FEC interface.

Change the maximum receive size to 2048 - 64, where 64 is the maximum
rx_alignment applied on the receive buffer for AVB capable FEC
cores. Use this value also for the maximum receive buffer size. The
driver is already allocating a receive SKB of 2048 bytes, so this
change should not have any significant effects.

Tested on imx51, imx6, vf610.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
---
 drivers/net/ethernet/freescale/fec_main.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

Comments

Andy Duan July 31, 2017, 1:41 a.m. | #1
From: Andrew Lunn <andrew@lunn.ch> Sent: Monday, July 31, 2017 1:36 AM
>The FEC Receive Control Register has a 14 bit field indicating the longest frame
>that my be received. It is being set to 1522. Frames longer than this are

My -> may

>discarded, but counted as being in error.
>
>When using DSA, frames from the switch has an additional header, either 4 or
>8 bytes if a Marvell switch is used. Thus a full MTU frame of 1522 bytes
>received by the switch on a port becomes 1530 bytes when passed to the host
>via the FEC interface.
>
>Change the maximum receive size to 2048 - 64, where 64 is the maximum
>rx_alignment applied on the receive buffer for AVB capable FEC cores. Use this
>value also for the maximum receive buffer size. The driver is already allocating
>a receive SKB of 2048 bytes, so this change should not have any significant
>effects.
>
>Tested on imx51, imx6, vf610.
>
>Signed-off-by: Andrew Lunn <andrew@lunn.ch>
>---
> drivers/net/ethernet/freescale/fec_main.c | 8 +++++---
> 1 file changed, 5 insertions(+), 3 deletions(-)
>
>diff --git a/drivers/net/ethernet/freescale/fec_main.c
>b/drivers/net/ethernet/freescale/fec_main.c
>index a6e323f15637..47ee74a17a9f 100644
>--- a/drivers/net/ethernet/freescale/fec_main.c
>+++ b/drivers/net/ethernet/freescale/fec_main.c
>@@ -173,10 +173,12 @@ MODULE_PARM_DESC(macaddr, "FEC Ethernet
>MAC address");  #endif /* CONFIG_M5272 */
>
> /* The FEC stores dest/src/type/vlan, data, and checksum for receive packets.
>+ *
>+ * 2048 byte skbufs are allocated. However, alignment requirements
>+ * varies between FEC variants. Worst case is 64, so round down by 64.
>  */
>-#define PKT_MAXBUF_SIZE		1522
>+#define PKT_MAXBUF_SIZE		(round_down(2048 - 64, 64))
> #define PKT_MINBUF_SIZE		64
>-#define PKT_MAXBLR_SIZE		1536
>
> /* FEC receive acceleration */
> #define FEC_RACC_IPDIS		(1 << 1)
>@@ -851,7 +853,7 @@ static void fec_enet_enable_ring(struct net_device
>*ndev)
> 	for (i = 0; i < fep->num_rx_queues; i++) {
> 		rxq = fep->rx_queue[i];
> 		writel(rxq->bd.dma, fep->hwp + FEC_R_DES_START(i));
>-		writel(PKT_MAXBLR_SIZE, fep->hwp + FEC_R_BUFF_SIZE(i));
>+		writel(PKT_MAXBUF_SIZE, fep->hwp + FEC_R_BUFF_SIZE(i));
>
> 		/* enable DMA1/2 */
> 		if (i)
>--
>2.13.2
David Miller July 31, 2017, 2:26 a.m. | #2
From: Andrew Lunn <andrew@lunn.ch>
Date: Sun, 30 Jul 2017 19:36:05 +0200

> The FEC Receive Control Register has a 14 bit field indicating the
> longest frame that my be received. It is being set to 1522. Frames
> longer than this are discarded, but counted as being in error.
> 
> When using DSA, frames from the switch has an additional header,
> either 4 or 8 bytes if a Marvell switch is used. Thus a full MTU frame
> of 1522 bytes received by the switch on a port becomes 1530 bytes when
> passed to the host via the FEC interface.
> 
> Change the maximum receive size to 2048 - 64, where 64 is the maximum
> rx_alignment applied on the receive buffer for AVB capable FEC
> cores. Use this value also for the maximum receive buffer size. The
> driver is already allocating a receive SKB of 2048 bytes, so this
> change should not have any significant effects.
> 
> Tested on imx51, imx6, vf610.
> 
> Signed-off-by: Andrew Lunn <andrew@lunn.ch>

Applied with commit log message typo fixed.
Andrew Lunn July 31, 2017, 2:27 a.m. | #3
On Sun, Jul 30, 2017 at 07:26:29PM -0700, David Miller wrote:
> From: Andrew Lunn <andrew@lunn.ch>
> Date: Sun, 30 Jul 2017 19:36:05 +0200
> 
> > The FEC Receive Control Register has a 14 bit field indicating the
> > longest frame that my be received. It is being set to 1522. Frames
> > longer than this are discarded, but counted as being in error.
> > 
> > When using DSA, frames from the switch has an additional header,
> > either 4 or 8 bytes if a Marvell switch is used. Thus a full MTU frame
> > of 1522 bytes received by the switch on a port becomes 1530 bytes when
> > passed to the host via the FEC interface.
> > 
> > Change the maximum receive size to 2048 - 64, where 64 is the maximum
> > rx_alignment applied on the receive buffer for AVB capable FEC
> > cores. Use this value also for the maximum receive buffer size. The
> > driver is already allocating a receive SKB of 2048 bytes, so this
> > change should not have any significant effects.
> > 
> > Tested on imx51, imx6, vf610.
> > 
> > Signed-off-by: Andrew Lunn <andrew@lunn.ch>
> 
> Applied with commit log message typo fixed.

Hi David

Thanks for fixing the typo.

       Andrew

Patch

diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c
index a6e323f15637..47ee74a17a9f 100644
--- a/drivers/net/ethernet/freescale/fec_main.c
+++ b/drivers/net/ethernet/freescale/fec_main.c
@@ -173,10 +173,12 @@  MODULE_PARM_DESC(macaddr, "FEC Ethernet MAC address");
 #endif /* CONFIG_M5272 */
 
 /* The FEC stores dest/src/type/vlan, data, and checksum for receive packets.
+ *
+ * 2048 byte skbufs are allocated. However, alignment requirements
+ * varies between FEC variants. Worst case is 64, so round down by 64.
  */
-#define PKT_MAXBUF_SIZE		1522
+#define PKT_MAXBUF_SIZE		(round_down(2048 - 64, 64))
 #define PKT_MINBUF_SIZE		64
-#define PKT_MAXBLR_SIZE		1536
 
 /* FEC receive acceleration */
 #define FEC_RACC_IPDIS		(1 << 1)
@@ -851,7 +853,7 @@  static void fec_enet_enable_ring(struct net_device *ndev)
 	for (i = 0; i < fep->num_rx_queues; i++) {
 		rxq = fep->rx_queue[i];
 		writel(rxq->bd.dma, fep->hwp + FEC_R_DES_START(i));
-		writel(PKT_MAXBLR_SIZE, fep->hwp + FEC_R_BUFF_SIZE(i));
+		writel(PKT_MAXBUF_SIZE, fep->hwp + FEC_R_BUFF_SIZE(i));
 
 		/* enable DMA1/2 */
 		if (i)