Skip to contentSkip to conversation

board: freescale: ls1012a: fix RGMII TX-delay issue

PATCH New Calvin Johnson 2 participants 3 messages Oct 26, 2017 – Feb 27, 2018 Network drivers

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 board: freescale: ls1012a: fix RGMII TX-delay issue

Recently logic to enable RGMII TX-delay was changed by commit 05b29aa0cb68 ("net: phy: realtek: fix enabling of the TX-delay for RTL8211F")

Based on this, here we are enabling the TX-delay again using PHY_INTERFACE_MODE_RGMII_TXID.

---

 board/freescale/ls1012ardb/eth.c | 2 +-
 drivers/net/pfe_eth/pfe_eth.c    | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
board/freescale/ls1012ardb/eth.c +1 -1
index 286bc8a..c0f6306 100644
@@ -62,7 +62,7 @@ int board_eth_init(bd_t *bis)
 	/* MAC2 */
 	ls1012a_set_mdio(1, miiphy_get_dev_by_name(DEFAULT_PFE_MDIO_NAME));
 	ls1012a_set_phy_address_mode(1, EMAC2_PHY_ADDR,
-				     PHY_INTERFACE_MODE_RGMII);
+				     PHY_INTERFACE_MODE_RGMII_TXID);
 
 	cpu_eth_init(bis);
 #endif
drivers/net/pfe_eth/pfe_eth.c +1 -1
index 8d8de40..4db823f 100644
@@ -27,7 +27,7 @@ struct gemac_s gem_info[] = {
 
 		/* phy iface */
 		.phy_address = EMAC2_PHY_ADDR,
-		.phy_mode = PHY_INTERFACE_MODE_RGMII,
+		.phy_mode = PHY_INTERFACE_MODE_RGMII_TXID,
 	},
 };
 
Signature · 1 line
2.7.4
#1 lore raw

PATCH 1/1 board: freescale: ls1012a: fix RGMII TX-delay issue

drivers/net/pfe_eth/pfe_eth.c

Quoted 32 lines · Calvin Johnson

Recently logic to enable RGMII TX-delay was changed by commit 05b29aa0cb68 ("net: phy: realtek: fix enabling of the TX-delay for RTL8211F")

Based on this, here we are enabling the TX-delay again using PHY_INTERFACE_MODE_RGMII_TXID.

---

 board/freescale/ls1012ardb/eth.c | 2 +-
 drivers/net/pfe_eth/pfe_eth.c    | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
board/freescale/ls1012ardb/eth.c +1 -1
index 286bc8a..c0f6306 100644
@@ -62,7 +62,7 @@ int board_eth_init(bd_t *bis)
        /* MAC2 */
        ls1012a_set_mdio(1, miiphy_get_dev_by_name(DEFAULT_PFE_MDIO_NAME));
        ls1012a_set_phy_address_mode(1, EMAC2_PHY_ADDR,
-                                    PHY_INTERFACE_MODE_RGMII);
+                                    PHY_INTERFACE_MODE_RGMII_TXID);
 
        cpu_eth_init(bis);
 #endif
drivers/net/pfe_eth/pfe_eth.c +0 -0
index 8d8de40..4db823f 100644

Is this still needed with v2 of your patch?

Quoted 16 lines
drivers/net/pfe_eth/pfe_eth.c +1 -2
@@ -27,7 +27,7 @@ struct gemac_s gem_info[] = {
 
                /* phy iface */
                .phy_address = EMAC2_PHY_ADDR,
-               .phy_mode = PHY_INTERFACE_MODE_RGMII,
+               .phy_mode = PHY_INTERFACE_MODE_RGMII_TXID,
        },
 };
 
--

2.7.4

_______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de
https://lists.denx.de/listinfo/u-boot

#2 lore raw

PATCH 1/1 board: freescale: ls1012a: fix RGMII TX-delay issue

Quoted 38 lines

-----Original Message-----
From: Joe Hershberger [mailto:joe.hershberger at ni.com]
Sent: Tuesday, February 27, 2018 9:28 PM
To: Calvin Johnson <[email protected]>
Cc: u-boot <[email protected]>; Joe Hershberger <[email protected]>; Anji Jagarlmudi <[email protected]>
Subject: Re: [U-Boot] [PATCH] board: freescale: ls1012a: fix RGMII TX-delay issue

Quoted 9 lines · Calvin Johnson

Recently logic to enable RGMII TX-delay was changed by commit 05b29aa0cb68 ("net: phy: realtek: fix enabling of the TX-delay for RTL8211F")

Based on this, here we are enabling the TX-delay again using PHY_INTERFACE_MODE_RGMII_TXID.

Review

Quoted 7 lines

---

 board/freescale/ls1012ardb/eth.c | 2 +-
 drivers/net/pfe_eth/pfe_eth.c    | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
(quoted diff) +0 -0

b/board/freescale/ls1012ardb/eth.c

board/freescale/ls1012ardb/eth.c int board_eth_init(bd_t *bis) · line 63

Quoted 6 lines

index 286bc8a..c0f6306 100644

board/freescale/ls1012ardb/eth.c +0 -0
@@ -62,7 +62,7 @@ int board_eth_init(bd_t *bis)
        /* MAC2 */
        ls1012a_set_mdio(1,

miiphy_get_dev_by_name(DEFAULT_PFE_MDIO_NAME));

Quoted 4 lines
board/freescale/ls1012ardb/eth.c +1 -1
        ls1012a_set_phy_address_mode(1, EMAC2_PHY_ADDR,
-                                    PHY_INTERFACE_MODE_RGMII);
+                                    PHY_INTERFACE_MODE_RGMII_TXID);
 
Quoted 8 lines

drivers/net/pfe_eth/pfe_eth.c

Quoted 6 lines
board/freescale/ls1012ardb/eth.c +0 -0
        cpu_eth_init(bis);
 #endif
drivers/net/pfe_eth/pfe_eth.c +0 -0
index 8d8de40..4db823f 100644

Is this still needed with v2 of your patch?

No. It is already squashed into v2.

Along with this, following patches also were squashed as you and York suggested. [U-Boot,1/3] drivers: net: pfe_eth: add pfe_rx_done to clear bd after packet processing [U-Boot,2/3] drivers: net: pfe_eth: use writel/readl to access hw bds [U-Boot,3/3] drivers: net: pfe_eth: cleanup typos and indent
[U-Boot,1/2] drivers: net: pfe_eth: remove unused code under CONFIG_UTIL_PE_DISABLED [U-Boot,2/2] drivers: net: pfe_eth: remove unused code under CONFIG_PFE_WARN_WA

Thanks
Calvin