[U-Boot] [Patch v2, batch 4 25/29] powerpc/p5040: fix mdio mux for 10G port

York Sun yorksun at freescale.com
Mon Mar 25 18:40:21 CET 2013


From: Shaohui Xie <Shaohui.Xie at freescale.com>

Current driver of p5040 assumes 10G port follows 1G port DTSEC5 in
eth port enum structure, it will assign mdio mux depend on this assumption.
This is not true with Fman V3, which added more 1G ports after port DTSEC5
in eth port enum structure, then 10G ports on p5040 will have wrong mdio mux.
So we use dynamic index for 10G ports instead of hardcoded enum value
when doing mdio mux for 10G ports.

Signed-off-by: Shaohui Xie <Shaohui.Xie at freescale.com>
---
 board/freescale/corenet_ds/eth_superhydra.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/board/freescale/corenet_ds/eth_superhydra.c b/board/freescale/corenet_ds/eth_superhydra.c
index ef9de25..ae07073 100644
--- a/board/freescale/corenet_ds/eth_superhydra.c
+++ b/board/freescale/corenet_ds/eth_superhydra.c
@@ -605,8 +605,8 @@ int board_eth_init(bd_t *bis)
 	lane = serdes_get_first_lane(XAUI_FM1);
 	if (lane >= 0) {
 		debug("FM1 at TGEC1 expects XAUI in slot %u\n", lane_to_slot[lane]);
-		mdio_mux[FM1_10GEC1].mask = BRDCFG1_EMI2_SEL_MASK;
-		mdio_mux[FM1_10GEC1].val = BRDCFG1_EMI2_SEL_SLOT2;
+		mdio_mux[i].mask = BRDCFG1_EMI2_SEL_MASK;
+		mdio_mux[i].val = BRDCFG1_EMI2_SEL_SLOT2;
 		super_hydra_mdio_set_mux("SUPER_HYDRA_FM1_TGEC_MDIO",
 					mdio_mux[i].mask, mdio_mux[i].val);
 	}
@@ -704,8 +704,8 @@ int board_eth_init(bd_t *bis)
 	lane = serdes_get_first_lane(XAUI_FM2);
 	if (lane >= 0) {
 		debug("FM2 at TGEC1 expects XAUI in slot %u\n", lane_to_slot[lane]);
-		mdio_mux[FM2_10GEC1].mask = BRDCFG1_EMI2_SEL_MASK;
-		mdio_mux[FM2_10GEC1].val = BRDCFG1_EMI2_SEL_SLOT1;
+		mdio_mux[i].mask = BRDCFG1_EMI2_SEL_MASK;
+		mdio_mux[i].val = BRDCFG1_EMI2_SEL_SLOT1;
 		super_hydra_mdio_set_mux("SUPER_HYDRA_FM2_TGEC_MDIO",
 					mdio_mux[i].mask, mdio_mux[i].val);
 	}
-- 
1.7.9.5




More information about the U-Boot mailing list