6.6-stable review patch. If anyone has any objections, please let me know.
------------------
From: Andrew Davis afd@ti.com
[ Upstream commit 0f40d5099cd6d828fd7de6227d3eabe86016724c ]
When the node for this phy selector is a child node of a syscon node then the property 'reg' is used as an offset into the parent regmap. When the node is standalone and gets its own regmap this offset is pre-applied. So we need to track which method was used to get the regmap and not apply the offset in the standalone case.
Fixes: 1fdfa7cccd35 ("phy: ti: gmii-sel: Allow parent to not be syscon node") Signed-off-by: Andrew Davis afd@ti.com Reviewed-by: Roger Quadros rogerq@kernel.org Link: https://lore.kernel.org/r/20231025143302.1265633-1-afd@ti.com Signed-off-by: Vinod Koul vkoul@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/phy/ti/phy-gmii-sel.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/phy/ti/phy-gmii-sel.c b/drivers/phy/ti/phy-gmii-sel.c index 555b323f45da1..bc847d3879f79 100644 --- a/drivers/phy/ti/phy-gmii-sel.c +++ b/drivers/phy/ti/phy-gmii-sel.c @@ -64,6 +64,7 @@ struct phy_gmii_sel_priv { u32 num_ports; u32 reg_offset; u32 qsgmii_main_ports; + bool no_offset; };
static int phy_gmii_sel_mode(struct phy *phy, enum phy_mode mode, int submode) @@ -402,7 +403,8 @@ static int phy_gmii_sel_init_ports(struct phy_gmii_sel_priv *priv) priv->num_ports = size / sizeof(u32); if (!priv->num_ports) return -EINVAL; - priv->reg_offset = __be32_to_cpu(*offset); + if (!priv->no_offset) + priv->reg_offset = __be32_to_cpu(*offset); }
if_phys = devm_kcalloc(dev, priv->num_ports, @@ -471,6 +473,7 @@ static int phy_gmii_sel_probe(struct platform_device *pdev) dev_err(dev, "Failed to get syscon %d\n", ret); return ret; } + priv->no_offset = true; }
ret = phy_gmii_sel_init_ports(priv);