* [PATCH v8 phy-next 22/31] drm: bridge: cdns-mhdp8546: use consumer API for getting PHY bus width
From: Vladimir Oltean @ 2026-05-05 10:05 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, Andrzej Hajda, Robert Foss, Laurent Pinchart,
Jonas Karlman, Jernej Skrabec, Maarten Lankhorst, Maxime Ripard,
Thomas Zimmermann, David Airlie, Simona Vetter
In-Reply-To: <20260505100523.1922388-1-vladimir.oltean@nxp.com>
The Cadence MHDP8546 DP bridge driver gets the PHY bus_width attribute
(holding number of lanes) directly, but doing this will no longer be
possible after the definition of struct phy is hidden from consumers.
Use the phy_get_bus_width() API function designed specifically for
consumers.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
---
Cc: Andrzej Hajda <andrzej.hajda@intel.com>
Cc: Neil Armstrong <neil.armstrong@linaro.org>
Cc: Robert Foss <rfoss@kernel.org>
Cc: Laurent Pinchart <Laurent.pinchart@ideasonboard.com>
Cc: Jonas Karlman <jonas@kwiboo.se>
Cc: Jernej Skrabec <jernej.skrabec@gmail.com>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Maxime Ripard <mripard@kernel.org>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: David Airlie <airlied@gmail.com>
Cc: Simona Vetter <simona@ffwll.ch>
v1->v8: none
---
drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c
index 87847978c1fb..76a684d4be26 100644
--- a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c
+++ b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c
@@ -1300,7 +1300,7 @@ static void cdns_mhdp_fill_host_caps(struct cdns_mhdp_device *mhdp)
/* Get source capabilities based on PHY attributes */
- mhdp->host.lanes_cnt = mhdp->phy->attrs.bus_width;
+ mhdp->host.lanes_cnt = phy_get_bus_width(mhdp->phy);
if (!mhdp->host.lanes_cnt)
mhdp->host.lanes_cnt = 4;
--
2.34.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related
* [PATCH v8 phy-next 25/31] pinctrl: tegra-xusb: include PHY provider header
From: Vladimir Oltean @ 2026-05-05 10:05 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, Linus Walleij, Thierry Reding, Jonathan Hunter,
Krzysztof Kozlowski
In-Reply-To: <20260505100523.1922388-1-vladimir.oltean@nxp.com>
The tegra-xusb pinctrl driver is also a PHY provider (calls
devm_phy_create() for PCIe and SATA). However, according to Vinod Koul,
having PHY provider drivers outside of drivers/phy/ is discouraged,
although it would be difficult for me to address a proper movement here.
Include the private provider API header from drivers/phy/, but leave a
FIXME in place. It will have to be moved, eventually.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Linus Walleij <linusw@kernel.org>
---
Cc: Linus Walleij <linusw@kernel.org>
Cc: Thierry Reding <thierry.reding@gmail.com>
Cc: Jonathan Hunter <jonathanh@nvidia.com>
Cc: Krzysztof Kozlowski <krzysztof.kozlowski@oss.qualcomm.com>
v2->v8: none
v1->v2: collect tag
---
drivers/pinctrl/tegra/pinctrl-tegra-xusb.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/pinctrl/tegra/pinctrl-tegra-xusb.c b/drivers/pinctrl/tegra/pinctrl-tegra-xusb.c
index c6a51bb21215..6b609bf685c7 100644
--- a/drivers/pinctrl/tegra/pinctrl-tegra-xusb.c
+++ b/drivers/pinctrl/tegra/pinctrl-tegra-xusb.c
@@ -7,7 +7,6 @@
#include <linux/io.h>
#include <linux/module.h>
#include <linux/of.h>
-#include <linux/phy/phy.h>
#include <linux/platform_device.h>
#include <linux/reset.h>
#include <linux/seq_file.h>
@@ -19,6 +18,7 @@
#include <dt-bindings/pinctrl/pinctrl-tegra-xusb.h>
+#include "../../phy/phy-provider.h" /* FIXME */
#include "../core.h"
#include "../pinctrl-utils.h"
--
2.34.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related
* [PATCH v8 phy-next 21/31] drm/rockchip: dsi: include PHY provider header
From: Vladimir Oltean @ 2026-05-05 10:05 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, Sandy Huang, Heiko Stübner, Andy Yan,
Maarten Lankhorst, Maxime Ripard, Thomas Zimmermann, David Airlie,
Simona Vetter
In-Reply-To: <20260505100523.1922388-1-vladimir.oltean@nxp.com>
The Rockchip DSI controller is a PHY consumer driver, which is also a
PHY provider (calls devm_phy_create()) that lives out of drivers/phy/.
According to Vinod, this is discouraged, although it would be difficult
for me to address a proper movement here.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
---
Cc: Sandy Huang <hjc@rock-chips.com>
Cc: "Heiko Stübner" <heiko@sntech.de>
Cc: Andy Yan <andy.yan@rock-chips.com>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Maxime Ripard <mripard@kernel.org>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: David Airlie <airlied@gmail.com>
Cc: Simona Vetter <simona@ffwll.ch>
v3->v8: none
v2->v3: fix cc list
v1->v2: none
---
drivers/gpu/drm/rockchip/dw-mipi-dsi-rockchip.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/gpu/drm/rockchip/dw-mipi-dsi-rockchip.c b/drivers/gpu/drm/rockchip/dw-mipi-dsi-rockchip.c
index 3547d91b25d3..c0aaa51e0237 100644
--- a/drivers/gpu/drm/rockchip/dw-mipi-dsi-rockchip.c
+++ b/drivers/gpu/drm/rockchip/dw-mipi-dsi-rockchip.c
@@ -27,6 +27,7 @@
#include <drm/drm_print.h>
#include <drm/drm_simple_kms_helper.h>
+#include "../../../phy/phy-provider.h" /* FIXME */
#include "rockchip_drm_drv.h"
#define DSI_PHY_RSTZ 0xa0
--
2.34.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related
* [PATCH v8 phy-next 20/31] phy: introduce phy_get_max_link_rate() helper for consumers
From: Vladimir Oltean @ 2026-05-05 10:05 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, Markus Schneider-Pargmann, Geert Uytterhoeven,
Andrzej Hajda, Robert Foss, Laurent Pinchart, Jonas Karlman,
Jernej Skrabec, Maarten Lankhorst, Maxime Ripard,
Thomas Zimmermann, David Airlie, Simona Vetter, Andy Yan,
Marc Kleine-Budde, Vincent Mailhol, Nicolas Ferre,
Alexandre Belloni, Claudiu Beznea, Magnus Damm
In-Reply-To: <20260505100523.1922388-1-vladimir.oltean@nxp.com>
Consumer drivers shouldn't dereference struct phy, not even to get to
its attributes.
We have phy_get_bus_width() as a precedent for getting the bus_width
attribute, so let's add phy_get_max_link_rate() and use it in DRM and
CAN drivers.
In CAN drivers, the transceiver is acquired through devm_phy_optional_get()
and NULL is given by the API as a non-error case, so the PHY API should
also tolerate NULL coming back to it. This means we can further simplify
the call sites that test for the NULL quality of the transceiver.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Acked-by: Markus Schneider-Pargmann <msp@baylibre.com> # m_can
Acked-by: Geert Uytterhoeven <geert+renesas@glider.be> # rcar_canfd
---
Cc: Andrzej Hajda <andrzej.hajda@intel.com>
Cc: Neil Armstrong <neil.armstrong@linaro.org>
Cc: Robert Foss <rfoss@kernel.org>
Cc: Laurent Pinchart <Laurent.pinchart@ideasonboard.com>
Cc: Jonas Karlman <jonas@kwiboo.se>
Cc: Jernej Skrabec <jernej.skrabec@gmail.com>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Maxime Ripard <mripard@kernel.org>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: David Airlie <airlied@gmail.com>
Cc: Simona Vetter <simona@ffwll.ch>
Cc: Andy Yan <andy.yan@rock-chips.com>
Cc: Marc Kleine-Budde <mkl@pengutronix.de>
Cc: Vincent Mailhol <mailhol@kernel.org>
Cc: Nicolas Ferre <nicolas.ferre@microchip.com>
Cc: Alexandre Belloni <alexandre.belloni@bootlin.com>
Cc: Claudiu Beznea <claudiu.beznea@tuxon.dev>
Cc: Markus Schneider-Pargmann <msp@baylibre.com>
Cc: Geert Uytterhoeven <geert+renesas@glider.be>
Cc: Magnus Damm <magnus.damm@gmail.com>
v7->v8: make "struct phy *" argument const
v3->v7: none
v2->v3: collect tag
v1->v2: make phy_get_bus_width() NULL-tolerant to simplify CAN callers
---
drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c | 4 ++--
drivers/gpu/drm/bridge/synopsys/dw-dp.c | 2 +-
drivers/net/can/at91_can.c | 3 +--
drivers/net/can/flexcan/flexcan-core.c | 3 +--
drivers/net/can/m_can/m_can_platform.c | 3 +--
drivers/net/can/rcar/rcar_canfd.c | 3 +--
drivers/phy/phy-core.c | 9 +++++++++
include/linux/phy/phy.h | 6 ++++++
8 files changed, 22 insertions(+), 11 deletions(-)
diff --git a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c
index d20c0f8ad04c..87847978c1fb 100644
--- a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c
+++ b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c
@@ -1296,7 +1296,7 @@ static u32 cdns_mhdp_get_training_interval_us(struct cdns_mhdp_device *mhdp,
static void cdns_mhdp_fill_host_caps(struct cdns_mhdp_device *mhdp)
{
- unsigned int link_rate;
+ u32 link_rate;
/* Get source capabilities based on PHY attributes */
@@ -1304,7 +1304,7 @@ static void cdns_mhdp_fill_host_caps(struct cdns_mhdp_device *mhdp)
if (!mhdp->host.lanes_cnt)
mhdp->host.lanes_cnt = 4;
- link_rate = mhdp->phy->attrs.max_link_rate;
+ link_rate = phy_get_max_link_rate(mhdp->phy);
if (!link_rate)
link_rate = drm_dp_bw_code_to_link_rate(DP_LINK_BW_8_1);
else
diff --git a/drivers/gpu/drm/bridge/synopsys/dw-dp.c b/drivers/gpu/drm/bridge/synopsys/dw-dp.c
index 45b37885d719..137ea6679732 100644
--- a/drivers/gpu/drm/bridge/synopsys/dw-dp.c
+++ b/drivers/gpu/drm/bridge/synopsys/dw-dp.c
@@ -530,7 +530,7 @@ static int dw_dp_link_parse(struct dw_dp *dp, struct drm_connector *connector)
link->revision = link->dpcd[DP_DPCD_REV];
link->rate = min_t(u32, min(dp->plat_data.max_link_rate,
- dp->phy->attrs.max_link_rate * 100),
+ phy_get_max_link_rate(dp->phy) * 100),
drm_dp_max_link_rate(link->dpcd));
link->lanes = min_t(u8, phy_get_bus_width(dp->phy),
drm_dp_max_lane_count(link->dpcd));
diff --git a/drivers/net/can/at91_can.c b/drivers/net/can/at91_can.c
index 58da323f14d7..7749da0a58f6 100644
--- a/drivers/net/can/at91_can.c
+++ b/drivers/net/can/at91_can.c
@@ -1125,8 +1125,7 @@ static int at91_can_probe(struct platform_device *pdev)
can_rx_offload_add_timestamp(dev, &priv->offload);
- if (transceiver)
- priv->can.bitrate_max = transceiver->attrs.max_link_rate;
+ priv->can.bitrate_max = phy_get_max_link_rate(transceiver);
if (at91_is_sam9263(priv))
dev->sysfs_groups[0] = &at91_sysfs_attr_group;
diff --git a/drivers/net/can/flexcan/flexcan-core.c b/drivers/net/can/flexcan/flexcan-core.c
index f5d22c61503f..093e48b8da58 100644
--- a/drivers/net/can/flexcan/flexcan-core.c
+++ b/drivers/net/can/flexcan/flexcan-core.c
@@ -2210,8 +2210,7 @@ static int flexcan_probe(struct platform_device *pdev)
priv->reg_xceiver = reg_xceiver;
priv->transceiver = transceiver;
- if (transceiver)
- priv->can.bitrate_max = transceiver->attrs.max_link_rate;
+ priv->can.bitrate_max = phy_get_max_link_rate(transceiver);
if (priv->devtype_data.quirks & FLEXCAN_QUIRK_NR_IRQ_3) {
priv->irq_boff = platform_get_irq(pdev, 1);
diff --git a/drivers/net/can/m_can/m_can_platform.c b/drivers/net/can/m_can/m_can_platform.c
index 56da411878af..2a0f163a683a 100644
--- a/drivers/net/can/m_can/m_can_platform.c
+++ b/drivers/net/can/m_can/m_can_platform.c
@@ -131,8 +131,7 @@ static int m_can_plat_probe(struct platform_device *pdev)
goto probe_fail;
}
- if (transceiver)
- mcan_class->can.bitrate_max = transceiver->attrs.max_link_rate;
+ mcan_class->can.bitrate_max = phy_get_max_link_rate(transceiver);
priv->base = addr;
priv->mram_base = mram_addr;
diff --git a/drivers/net/can/rcar/rcar_canfd.c b/drivers/net/can/rcar/rcar_canfd.c
index eaf8cac78038..9062db48d477 100644
--- a/drivers/net/can/rcar/rcar_canfd.c
+++ b/drivers/net/can/rcar/rcar_canfd.c
@@ -1884,8 +1884,7 @@ static int rcar_canfd_channel_probe(struct rcar_canfd_global *gpriv, u32 ch,
priv->transceiver = transceiver;
priv->channel = ch;
priv->gpriv = gpriv;
- if (transceiver)
- priv->can.bitrate_max = transceiver->attrs.max_link_rate;
+ priv->can.bitrate_max = phy_get_max_link_rate(transceiver);
priv->can.clock.freq = fcan_freq;
dev_info(dev, "can_clk rate is %u\n", priv->can.clock.freq);
diff --git a/drivers/phy/phy-core.c b/drivers/phy/phy-core.c
index 6cd33204d2ff..8707111c4666 100644
--- a/drivers/phy/phy-core.c
+++ b/drivers/phy/phy-core.c
@@ -675,6 +675,15 @@ int phy_request_bus_width(struct phy *phy, int bus_width)
}
EXPORT_SYMBOL_GPL(phy_request_bus_width);
+u32 phy_get_max_link_rate(const struct phy *phy)
+{
+ if (!phy)
+ return 0;
+
+ return phy->attrs.max_link_rate;
+}
+EXPORT_SYMBOL_GPL(phy_get_max_link_rate);
+
/**
* _of_phy_get() - lookup and obtain a reference to a phy by phandle
* @np: device_node for which to get the phy
diff --git a/include/linux/phy/phy.h b/include/linux/phy/phy.h
index 99d3d65da614..e65865b4acfe 100644
--- a/include/linux/phy/phy.h
+++ b/include/linux/phy/phy.h
@@ -57,6 +57,7 @@ int phy_notify_disconnect(struct phy *phy, int port);
int phy_notify_state(struct phy *phy, union phy_notify state);
int phy_get_bus_width(struct phy *phy);
int phy_request_bus_width(struct phy *phy, int bus_width);
+u32 phy_get_max_link_rate(const struct phy *phy);
#else
static inline struct phy *phy_get(struct device *dev, const char *string)
{
@@ -260,6 +261,11 @@ static inline int phy_request_bus_width(struct phy *phy, int bus_width)
return -ENOSYS;
}
+
+static inline u32 phy_get_max_link_rate(const struct phy *phy)
+{
+ return 0;
+}
#endif /* IS_ENABLED(CONFIG_GENERIC_PHY) */
#endif /* __PHY_CONSUMER_H */
--
2.34.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related
* [PATCH v8 phy-next 19/31] phy: make phy_get_mode(), phy_get_bus_width() NULL tolerant
From: Vladimir Oltean @ 2026-05-05 10:05 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver
In-Reply-To: <20260505100523.1922388-1-vladimir.oltean@nxp.com>
The PHY API has an optional "get" which returns NULL, so it needs to
accept that NULL coming back in.
Most PHY functions do this, only the formerly static inline attribute
dereferences did not.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
---
v7->v8: remove phy_set_bus_width() from this change, it has become a
provider function which does not need the protection
v2->v7: none
v1->v2: patch is new
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
---
drivers/phy/phy-core.c | 6 ++++++
include/linux/phy/phy.h | 2 ++
2 files changed, 8 insertions(+)
diff --git a/drivers/phy/phy-core.c b/drivers/phy/phy-core.c
index fe933629286b..6cd33204d2ff 100644
--- a/drivers/phy/phy-core.c
+++ b/drivers/phy/phy-core.c
@@ -624,12 +624,18 @@ EXPORT_SYMBOL_GPL(phy_validate);
enum phy_mode phy_get_mode(struct phy *phy)
{
+ if (!phy)
+ return PHY_MODE_INVALID;
+
return phy->attrs.mode;
}
EXPORT_SYMBOL_GPL(phy_get_mode);
int phy_get_bus_width(struct phy *phy)
{
+ if (!phy)
+ return 0;
+
return phy->attrs.bus_width;
}
EXPORT_SYMBOL_GPL(phy_get_bus_width);
diff --git a/include/linux/phy/phy.h b/include/linux/phy/phy.h
index d716e5e0584c..99d3d65da614 100644
--- a/include/linux/phy/phy.h
+++ b/include/linux/phy/phy.h
@@ -248,6 +248,8 @@ static inline int phy_notify_state(struct phy *phy, union phy_notify state)
static inline int phy_get_bus_width(struct phy *phy)
{
+ if (!phy)
+ return 0;
return -ENOSYS;
}
--
2.34.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related
* [PATCH v8 phy-next 17/31] usb: gadget: tegra-xudc: avoid direct dereference of phy->dev.of_node
From: Vladimir Oltean @ 2026-05-05 10:05 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, Greg Kroah-Hartman, JC Kuo, Johan Hovold,
Jonathan Hunter, Mathias Nyman, Thierry Reding
In-Reply-To: <20260505100523.1922388-1-vladimir.oltean@nxp.com>
In a somewhat similar situation as the Tegra USB host controller driver,
the Tegra XUDC driver for USB gadget mode needs to get to a struct
usb_phy that sits behind the same OF node as the Generic PHY. It does
that directly, which will no longer be possible. The PHY provider is
also the xusb padctl driver.
The rework here is also to implement a parallel OF node lookup path
based on the "phys" phandle and the #phy-cells of the padctl provider.
Some further notes:
- create a local "usbphy" variable to hold the devm_usb_get_phy_by_node()
output. This makes the error checks more obvious (avoids keeping an
error-encoded pointer in xudc->usbphy[i] even temporarily).
- the "if (IS_ERR(utmi_phy)) .. else if (utmi_phy) .. else if (!utmi_phy)"
pattern can be simplified, considering that neither the IS_ERR() nor
the NULL case continue execution in the current block. Therefore, we
can move the case where the "utmi_phy" is a valid pointer outside the
"if" checks, and this reduces the code indentation level.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
---
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: JC Kuo <jckuo@nvidia.com>
Cc: Johan Hovold <johan+linaro@kernel.org>
Cc: Jonathan Hunter <jonathanh@nvidia.com>
Cc: Mathias Nyman <mathias.nyman@intel.com>
Cc: Thierry Reding <thierry.reding@gmail.com>
v7->v8:
- remove duplicated variable "err" which shadows existing one
- stop execution on of_property_match_string() and
of_parse_phandle_with_args() errors, instead of silently continuing
v5->v7: none
v4->v5: patch is new
---
drivers/usb/gadget/udc/tegra-xudc.c | 43 +++++++++++++++++++++--------
1 file changed, 32 insertions(+), 11 deletions(-)
diff --git a/drivers/usb/gadget/udc/tegra-xudc.c b/drivers/usb/gadget/udc/tegra-xudc.c
index e9d33be02866..f93e0a5651ab 100644
--- a/drivers/usb/gadget/udc/tegra-xudc.c
+++ b/drivers/usb/gadget/udc/tegra-xudc.c
@@ -3494,6 +3494,7 @@ static void tegra_xudc_device_params_init(struct tegra_xudc *xudc)
static int tegra_xudc_phy_get(struct tegra_xudc *xudc)
{
+ struct device_node *np = dev_of_node(xudc->dev);
int err = 0, usb3_companion_port;
unsigned int i, j;
@@ -3515,7 +3516,10 @@ static int tegra_xudc_phy_get(struct tegra_xudc *xudc)
xudc->vbus_nb.notifier_call = tegra_xudc_vbus_notify;
for (i = 0; i < xudc->soc->num_phys; i++) {
+ struct of_phandle_args args;
char phy_name[] = "usb.-.";
+ struct usb_phy *usbphy;
+ int index;
/* Get USB2 phy */
snprintf(phy_name, sizeof(phy_name), "usb2-%d", i);
@@ -3525,22 +3529,39 @@ static int tegra_xudc_phy_get(struct tegra_xudc *xudc)
dev_err_probe(xudc->dev, err,
"failed to get PHY for phy-name usb2-%d\n", i);
goto clean_up;
- } else if (xudc->utmi_phy[i]) {
- /* Get usb-phy, if utmi phy is available */
- xudc->usbphy[i] = devm_usb_get_phy_by_node(xudc->dev,
- xudc->utmi_phy[i]->dev.of_node,
- NULL);
- if (IS_ERR(xudc->usbphy[i])) {
- err = PTR_ERR(xudc->usbphy[i]);
- dev_err_probe(xudc->dev, err,
- "failed to get usbphy-%d\n", i);
- goto clean_up;
- }
} else if (!xudc->utmi_phy[i]) {
/* if utmi phy is not available, ignore USB3 phy get */
continue;
}
+ index = of_property_match_string(np, "phy-names", phy_name);
+ if (index < 0) {
+ dev_err(xudc->dev,
+ "no \"phy-names\" entry for \"%s\": %pe\n",
+ phy_name, ERR_PTR(index));
+ goto clean_up;
+ }
+
+ err = of_parse_phandle_with_args(np, "phys", "#phy-cells",
+ index, &args);
+ if (err) {
+ dev_err(xudc->dev,
+ "failed to follow \"%s\" reference\n: %pe\n",
+ phy_name, ERR_PTR(err));
+ goto clean_up;
+ }
+
+ /* Get usb-phy, if utmi phy is available */
+ usbphy = devm_usb_get_phy_by_node(xudc->dev, args.np, NULL);
+ of_node_put(args.np);
+ if (IS_ERR(usbphy)) {
+ err = PTR_ERR(usbphy);
+ dev_err_probe(xudc->dev, err,
+ "failed to get usbphy-%d\n", i);
+ goto clean_up;
+ }
+ xudc->usbphy[i] = usbphy;
+
/* Get USB3 phy */
usb3_companion_port = tegra_xusb_padctl_get_usb3_companion(xudc->padctl, i);
if (usb3_companion_port < 0)
--
2.34.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related
* [PATCH v8 phy-next 16/31] usb: host: tegra: avoid direct dereference of phy->dev.of_node
From: Vladimir Oltean @ 2026-05-05 10:05 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, Greg Kroah-Hartman, JC Kuo, Johan Hovold,
Jonathan Hunter, Mathias Nyman, Thierry Reding
In-Reply-To: <20260505100523.1922388-1-vladimir.oltean@nxp.com>
A piece of context which seems relevant here is that the USB subsystem
is transitioning from struct usb_phy to struct phy (belonging to the
Generic PHY subsystem). Commit 1a229d8690a0 ("Revert "usb: phy: add usb
phy notify port status API"") seems to confirm that this is the case.
In the transition process, some PHY provider drivers register themselves
as both Generic PHY and USB PHY in an attempt to bridge the API gap.
Such is the case with drivers/phy/tegra/xusb.c, accessed here by the
Tegra USB host driver. This USB host expects the PHY device behind the
Generic PHY to also be a USB PHY, and calls
devm_usb_get_phy_by_node(phy->dev.of_node).
The Generic PHY exposes no API to get the OF node from a PHY device, so
the Tegra USB host driver gets it directly. However, "struct phy" will
be made an opaque pointer, to avoid misuse, so this will no longer be
possible.
Considering the fact that the Generic PHY/USB PHY duality is a
transitional state, I am deliberately not planning to make the life of
this driver any easier by providing a helper to get to the OF node
somehow. Instead, implement a parallel lookup path through which the
Tegra USB host driver can continue to get to the OF node provided by the
padctl component, using the 'phys' phandle.
Secondly (minor issue) the driver uses the phy->dev.of_node again to
print using dev_dbg() that a "remote wake" was detected. Just print the
index at which the PHY appears inside the driver's tegra->phys[] array
instead.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
---
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: JC Kuo <jckuo@nvidia.com>
Cc: Johan Hovold <johan+linaro@kernel.org>
Cc: Jonathan Hunter <jonathanh@nvidia.com>
Cc: Mathias Nyman <mathias.nyman@intel.com>
Cc: Thierry Reding <thierry.reding@gmail.com>
v7->v8:
- add missing of_node_put(args.np) in tegra_xusb_init_usb_phy()
- replace %d with %u in tegra_xhci_disable_phy_wake()
Both pointed out by Sashiko.
v5->v7: none
v4->v5: patch is new
---
drivers/usb/host/xhci-tegra.c | 43 ++++++++++++++++++++++-------------
1 file changed, 27 insertions(+), 16 deletions(-)
diff --git a/drivers/usb/host/xhci-tegra.c b/drivers/usb/host/xhci-tegra.c
index d2214d309e96..6724a777ca2a 100644
--- a/drivers/usb/host/xhci-tegra.c
+++ b/drivers/usb/host/xhci-tegra.c
@@ -1491,6 +1491,7 @@ static int tegra_xhci_id_notify(struct notifier_block *nb,
static int tegra_xusb_init_usb_phy(struct tegra_xusb *tegra)
{
+ struct device_node *np = dev_of_node(tegra->dev);
unsigned int i;
tegra->usbphy = devm_kcalloc(tegra->dev, tegra->num_usb_phys,
@@ -1504,23 +1505,34 @@ static int tegra_xusb_init_usb_phy(struct tegra_xusb *tegra)
tegra->otg_usb3_port = -EINVAL;
for (i = 0; i < tegra->num_usb_phys; i++) {
- struct phy *phy = tegra_xusb_get_phy(tegra, "usb2", i);
+ struct of_phandle_args args;
+ struct usb_phy *usbphy;
+ int index, err;
+ char prop[8];
- if (!phy)
+ snprintf(prop, sizeof(prop), "usb2-%d", i);
+
+ /*
+ * usb-phy is optional, continue if it's not available.
+ */
+ index = of_property_match_string(np, "phy-names", prop);
+ if (index < 0)
continue;
- tegra->usbphy[i] = devm_usb_get_phy_by_node(tegra->dev,
- phy->dev.of_node,
- &tegra->id_nb);
- if (!IS_ERR(tegra->usbphy[i])) {
- dev_dbg(tegra->dev, "usbphy-%d registered", i);
- otg_set_host(tegra->usbphy[i]->otg, &tegra->hcd->self);
- } else {
- /*
- * usb-phy is optional, continue if its not available.
- */
- tegra->usbphy[i] = NULL;
- }
+ err = of_parse_phandle_with_args(np, "phys", "#phy-cells",
+ index, &args);
+ if (err)
+ continue;
+
+ usbphy = devm_usb_get_phy_by_node(tegra->dev, args.np,
+ &tegra->id_nb);
+ of_node_put(args.np);
+ if (IS_ERR(usbphy))
+ continue;
+
+ tegra->usbphy[i] = usbphy;
+ dev_dbg(tegra->dev, "usbphy-%d registered", i);
+ otg_set_host(tegra->usbphy[i]->otg, &tegra->hcd->self);
}
return 0;
@@ -2168,8 +2180,7 @@ static void tegra_xhci_disable_phy_wake(struct tegra_xusb *tegra)
continue;
if (tegra_xusb_padctl_remote_wake_detected(padctl, tegra->phys[i]))
- dev_dbg(tegra->dev, "%pOF remote wake detected\n",
- tegra->phys[i]->dev.of_node);
+ dev_dbg(tegra->dev, "PHY %u remote wake detected\n", i);
tegra_xusb_padctl_disable_phy_wake(padctl, tegra->phys[i]);
}
--
2.34.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related
* [PATCH v8 phy-next 14/31] scsi: ufs: qcom: include missing <linux/interrupt.h>
From: Vladimir Oltean @ 2026-05-05 10:05 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, Martin K . Petersen, Can Guo,
James E.J. Bottomley, Dmitry Baryshkov, Nitin Rawat,
Manivannan Sadhasivam
In-Reply-To: <20260505100523.1922388-1-vladimir.oltean@nxp.com>
This driver uses devm_request_irq() without including <linux/interrupt.h>
by itself, which would lead to build failures if the headers providing
this transitively were to stop providing it.
On aarch64, we can see, using KCFLAGS='-H' make drivers/ufs/host/ufs-qcom.o,
that the inclusion path is:
drivers/ufs/host/ufs-qcom.c
-> include/linux/acpi.h
-> arch/arm64/include/asm/acpi.h
-> include/linux/efi.h
-> include/linux/rtc.h
-> include/linux/interrupt.h
Whereas on armv7, the situation is quite different. This architecture
has no CONFIG_ACPI symbol, and therefore on it, <linux/acpi.h> does not
include <asm/acpi.h>, and <linux/interrupt.h> is not provided that way.
It is provided, however, through this "fallback" path:
drivers/ufs/host/ufs-qcom.c
-> include/linux/phy/phy.h
-> include/linux/regulator/consumer.h
-> include/linux/suspend.h
-> include/linux/swap.h
-> include/linux/memcontrol.h
-> include/linux/writeback.h
-> include/linux/interrupt.h
The point is that <linux/phy/phy.h> will stop providing
<linux/regulator/consumer.h>, and this would break the transitive
include chain on armv7.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
---
Cc: Can Guo <quic_cang@quicinc.com>
Cc: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com>
Cc: "Martin K. Petersen" <martin.petersen@oracle.com>
Cc: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Cc: Nitin Rawat <quic_nitirawa@quicinc.com>
Cc: Manivannan Sadhasivam <mani@kernel.org>
v7->v8: none
v6->v7: collect tag from Martin
v5->v6: patch is new
---
drivers/ufs/host/ufs-qcom.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/ufs/host/ufs-qcom.c b/drivers/ufs/host/ufs-qcom.c
index c547d8643480..9a4913e254af 100644
--- a/drivers/ufs/host/ufs-qcom.c
+++ b/drivers/ufs/host/ufs-qcom.c
@@ -10,6 +10,7 @@
#include <linux/devfreq.h>
#include <linux/gpio/consumer.h>
#include <linux/interconnect.h>
+#include <linux/interrupt.h>
#include <linux/module.h>
#include <linux/of.h>
#include <linux/phy/phy.h>
--
2.34.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related
* [PATCH v8 phy-next 15/31] drm/rockchip: dw_hdmi: avoid direct dereference of phy->dev.of_node
From: Vladimir Oltean @ 2026-05-05 10:05 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, Heiko Stueber, Sandy Huang, Andy Yan,
Maarten Lankhorst, Maxime Ripard, Thomas Zimmermann, David Airlie,
Simona Vetter
In-Reply-To: <20260505100523.1922388-1-vladimir.oltean@nxp.com>
The dw_hdmi-rockchip driver validates pixel clock rates against the
HDMI PHY's internal clock provider on certain SoCs like RK3328.
This is currently achieved by dereferencing hdmi->phy->dev.of_node
to obtain the provider node, which violates the Generic PHY API's
encapsulation (the goal is for struct phy to be an opaque pointer
with a hidden definition, to be interacted with only using API
functions or NULL pointer checks, for the case where optional variants
of phy_get() did not find a PHY).
Refactor dw_hdmi_rockchip_bind() to perform a manual phandle lookup
on the "hdmi" PHY index within the controller's DT node. This provides
a parallel path to the clock provider's OF node without relying on the
internal structure of the struct phy handle.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Heiko Stueber <heiko@sntech.de>
---
Cc: Sandy Huang <hjc@rock-chips.com>
Cc: "Heiko Stübner" <heiko@sntech.de>
Cc: Andy Yan <andy.yan@rock-chips.com>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Maxime Ripard <mripard@kernel.org>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: David Airlie <airlied@gmail.com>
Cc: Simona Vetter <simona@ffwll.ch>
v4->v8: none
v3->v4: add commit message clarification of what is understood by
"opaque pointer"
v1->v3: none
---
drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c | 25 ++++++++++++---------
1 file changed, 15 insertions(+), 10 deletions(-)
diff --git a/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c b/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c
index 0dc1eb5d2ae3..7abb42e486c0 100644
--- a/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c
+++ b/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c
@@ -537,21 +537,22 @@ static int dw_hdmi_rockchip_bind(struct device *dev, struct device *master,
void *data)
{
struct platform_device *pdev = to_platform_device(dev);
+ struct device_node *np = dev_of_node(dev);
struct dw_hdmi_plat_data *plat_data;
const struct of_device_id *match;
struct drm_device *drm = data;
struct drm_encoder *encoder;
struct rockchip_hdmi *hdmi;
- int ret;
+ int ret, index;
- if (!pdev->dev.of_node)
+ if (!np)
return -ENODEV;
hdmi = devm_kzalloc(&pdev->dev, sizeof(*hdmi), GFP_KERNEL);
if (!hdmi)
return -ENOMEM;
- match = of_match_node(dw_hdmi_rockchip_dt_ids, pdev->dev.of_node);
+ match = of_match_node(dw_hdmi_rockchip_dt_ids, np);
plat_data = devm_kmemdup(&pdev->dev, match->data,
sizeof(*plat_data), GFP_KERNEL);
if (!plat_data)
@@ -564,9 +565,9 @@ static int dw_hdmi_rockchip_bind(struct device *dev, struct device *master,
plat_data->priv_data = hdmi;
encoder = &hdmi->encoder.encoder;
- encoder->possible_crtcs = drm_of_find_possible_crtcs(drm, dev->of_node);
+ encoder->possible_crtcs = drm_of_find_possible_crtcs(drm, np);
rockchip_drm_encoder_set_crtc_endpoint_id(&hdmi->encoder,
- dev->of_node, 0, 0);
+ np, 0, 0);
/*
* If we failed to find the CRTC(s) which this encoder is
@@ -588,13 +589,17 @@ static int dw_hdmi_rockchip_bind(struct device *dev, struct device *master,
return dev_err_probe(hdmi->dev, ret, "failed to get phy\n");
}
- if (hdmi->phy) {
+ index = of_property_match_string(np, "phy-names", "hdmi");
+ if (index >= 0) {
struct of_phandle_args clkspec;
- clkspec.np = hdmi->phy->dev.of_node;
- hdmi->hdmiphy_clk = of_clk_get_from_provider(&clkspec);
- if (IS_ERR(hdmi->hdmiphy_clk))
- hdmi->hdmiphy_clk = NULL;
+ if (!of_parse_phandle_with_args(np, "phys", "#phy-cells", index,
+ &clkspec)) {
+ hdmi->hdmiphy_clk = of_clk_get_from_provider(&clkspec);
+ of_node_put(clkspec.np);
+ if (IS_ERR(hdmi->hdmiphy_clk))
+ hdmi->hdmiphy_clk = NULL;
+ }
}
if (hdmi->chip_data == &rk3568_chip_data) {
--
2.34.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related
* [PATCH v8 phy-next 11/31] scsi: ufs: qcom: call phy_init() before phy_power_on()
From: Vladimir Oltean @ 2026-05-05 10:05 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, James E.J. Bottomley, Martin K. Petersen,
Dmitry Baryshkov, Nitin Rawat, Manivannan Sadhasivam
In-Reply-To: <20260505100523.1922388-1-vladimir.oltean@nxp.com>
The Qualcomm UFS host controller interacts with the QMP PHY in a way
which violates the Generic PHY API expectation, documented in section
"Order of API calls" from Documentation/driver-api/phy/phy.rst, and then
tries to hide it.
Namely, calls must be made in the phy_init() -> phy_power_on() ->
phy_power_off() -> phy_exit() sequence.
What we actually have is:
ufshcd_init()
-> ufshcd_hba_init()
-> ufshcd_setup_clocks(hba, true)
-> ufshcd_vops_setup_clocks(hba, true, POST_CHANGE)
-> ufs_qcom_setup_clocks(hba, true, POST_CHANGE)
-> ufs_qcom_init() has not run, simply ignore
-> ufshcd_variant_hba_init()
-> ufs_qcom_init()
-> ufs_qcom_setup_clocks(hba, true, POST_CHANGE)
-> phy_power_on(phy)
-> ufshcd_hba_enable()
-> ufshcd_vops_hce_enable_notify()
-> ufs_qcom_hce_enable_notify()
-> ufs_qcom_power_up_sequence()
-> if (phy->power_count) phy_power_off(phy)
-> phy_init(phy)
This "works" because the way that the "phy_power_on was called before
phy_init\n" warning condition is detected in phy-core.c is if the
power_count is positive at the phy_init() call time.
By having that "if (phy->power_count) phy_power_off(phy)" logic, the
ufs-qcom.c technically sidesteps the test, but actually violates the
Generic PHY API even more (calls phy_power_on() *and* phy_power_off()
before phy_init()).
The reason why I stumbled upon this was that I was trying to remove
dereferences of phy->power_count (a PHY internal field) from consumer
drivers.
phy_init(), implemented as qmp_ufs_phy_init(), calls qmp->ufs_reset =
devm_reset_control_get_exclusive(), so my understanding is that it needs
to be called:
- no earlier than ufs_qcom_init() -> devm_reset_controller_register()
which makes qmp->ufs_reset available
- no later than ufs_qcom_power_up_sequence() -> phy_calibrate() ->
qmp_ufs_phy_calibrate() where the qmp->ufs_reset is needed; although
phy_init() should be the first PHY API call made.
The only mystery is why is the current phy_init() placement so late, in
ufs_qcom_power_up_sequence(), but I guess the answer is that the
placement is vestigial. After the incremental work of commit
c9b589791fc1 ("phy: qcom: Utilize UFS reset controller") from
Evan Green and commit cbfd6c124f27 ("phy: qcom-qmp-ufs: Refactor
phy_power_on and phy_calibrate callbacks") from Nitin Rawat, the entire
multi-stage PHY init procedure was moved to phy_power_on(), but nobody
bothered to move phy_init() anywhere else more natural.
So hopefully if the calculations are right, any placement within that
bounding box should be good, and I'm picking the new phy_init() location
to be in ufs_qcom_init().
Even with phy_init() out of the way, ufs_qcom_power_up_sequence() ->
phy_power_off() is still needed, for a separate reason which will be
dealt with separately.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
---
Cc: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com>
Cc: "Martin K. Petersen" <martin.petersen@oracle.com>
Cc: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Cc: Nitin Rawat <quic_nitirawa@quicinc.com>
Cc: Manivannan Sadhasivam <mani@kernel.org>
v7->v8: patch is new
Commit was previously posted here but did not get any testing.
https://lore.kernel.org/linux-phy/20260327112858.r5lpqygtvsane2vf@skbuf/
---
drivers/ufs/host/ufs-qcom.c | 22 ++++++++--------------
1 file changed, 8 insertions(+), 14 deletions(-)
diff --git a/drivers/ufs/host/ufs-qcom.c b/drivers/ufs/host/ufs-qcom.c
index bc037db46624..9039b087bf21 100644
--- a/drivers/ufs/host/ufs-qcom.c
+++ b/drivers/ufs/host/ufs-qcom.c
@@ -513,13 +513,6 @@ static int ufs_qcom_power_up_sequence(struct ufs_hba *hba)
/* phy initialization - calibrate the phy */
- ret = phy_init(phy);
- if (ret) {
- dev_err(hba->dev, "%s: phy init failed, ret = %d\n",
- __func__, ret);
- return ret;
- }
-
ret = phy_set_mode_ext(phy, mode, host->phy_gear);
if (ret)
goto out_disable_phy;
@@ -529,23 +522,18 @@ static int ufs_qcom_power_up_sequence(struct ufs_hba *hba)
if (ret) {
dev_err(hba->dev, "%s: phy power on failed, ret = %d\n",
__func__, ret);
- goto out_disable_phy;
+ return ret;
}
ret = phy_calibrate(phy);
if (ret) {
dev_err(hba->dev, "Failed to calibrate PHY: %d\n", ret);
- goto out_disable_phy;
+ return ret;
}
ufs_qcom_select_unipro_mode(host);
return 0;
-
-out_disable_phy:
- phy_exit(phy);
-
- return ret;
}
/*
@@ -1625,6 +1613,12 @@ static int ufs_qcom_init(struct ufs_hba *hba)
if (err)
goto out_variant_clear;
+ err = phy_init(host->generic_phy);
+ if (err) {
+ dev_err(hba->dev, "phy_init failed: %pe\n", ERR_PTR(err));
+ goto out_variant_clear;
+ }
+
ufs_qcom_setup_clocks(hba, true, POST_CHANGE);
ufs_qcom_get_default_testbus_cfg(host);
--
2.34.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related
* [PATCH v8 phy-next 13/31] scsi: ufs: qcom: keep separate track of PHY power state
From: Vladimir Oltean @ 2026-05-05 10:05 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, Can Guo, James E.J. Bottomley, Martin K. Petersen,
Dmitry Baryshkov, Nitin Rawat, Manivannan Sadhasivam
In-Reply-To: <20260505100523.1922388-1-vladimir.oltean@nxp.com>
The Generic PHY API needs the following call order:
phy_init() -> phy_power_on() -> phy_power_off() -> phy_exit()
with a balanced number of phy_init() <-> phy_exit() and
phy_power_on() <-> phy_power_off() calls.
The UFS framework is not exactly great in helping out with obeying these
requirements. For example, the Qualcomm UFS HCD driver insists pairing
the PHY power to the clock setup operations. But during driver removal,
we have:
ufshcd_hba_exit()
-> ufshcd_variant_hba_exit()
-> ufs_qcom_exit()
-> ufshcd_setup_clocks(hba, false)
which means that we will underflow the PHY power_count.
Adding a "bool power_count" to the driver and checking it before each
call helps avoid this issue.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
---
Cc: Can Guo <quic_cang@quicinc.com>
Cc: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com>
Cc: "Martin K. Petersen" <martin.petersen@oracle.com>
Cc: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Cc: Nitin Rawat <quic_nitirawa@quicinc.com>
Cc: Manivannan Sadhasivam <mani@kernel.org>
v7->v8: patch is new to fix an issue reported by Sashiko
https://sashiko.dev/#/patchset/20260430110652.558622-1-vladimir.oltean@nxp.com
---
drivers/ufs/host/ufs-qcom.c | 53 ++++++++++++++++++++++++++++---------
drivers/ufs/host/ufs-qcom.h | 1 +
2 files changed, 41 insertions(+), 13 deletions(-)
diff --git a/drivers/ufs/host/ufs-qcom.c b/drivers/ufs/host/ufs-qcom.c
index e28edcfd13a1..c547d8643480 100644
--- a/drivers/ufs/host/ufs-qcom.c
+++ b/drivers/ufs/host/ufs-qcom.c
@@ -485,6 +485,42 @@ static u32 ufs_qcom_get_hs_gear(struct ufs_hba *hba)
return UFS_HS_G3;
}
+static int ufs_qcom_phy_power_on(struct ufs_qcom_host *host)
+{
+ int ret;
+
+ if (host->phy_powered_on)
+ return 0;
+
+ ret = phy_power_on(host->generic_phy);
+ if (ret) {
+ dev_err(host->hba->dev, "Failed to power on PHY: %pe\n",
+ ERR_PTR(ret));
+ return ret;
+ }
+
+ host->phy_powered_on = true;
+
+ return 0;
+}
+
+static void ufs_qcom_phy_power_off(struct ufs_qcom_host *host)
+{
+ int ret;
+
+ if (!host->phy_powered_on)
+ return;
+
+ ret = phy_power_off(host->generic_phy);
+ if (ret) {
+ dev_warn(host->hba->dev, "Failed to power off PHY: %pe\n",
+ ERR_PTR(ret));
+ return;
+ }
+
+ host->phy_powered_on = false;
+}
+
static int ufs_qcom_phy_change_mode(struct ufs_hba *hba)
{
struct ufs_qcom_host *host = ufshcd_get_variant(hba);
@@ -1390,7 +1426,6 @@ static int ufs_qcom_setup_clocks(struct ufs_hba *hba, bool on,
enum ufs_notify_change_status status)
{
struct ufs_qcom_host *host = ufshcd_get_variant(hba);
- struct phy *phy;
int err;
/*
@@ -1401,8 +1436,6 @@ static int ufs_qcom_setup_clocks(struct ufs_hba *hba, bool on,
if (!host)
return 0;
- phy = host->generic_phy;
-
switch (status) {
case PRE_CHANGE:
if (on) {
@@ -1420,20 +1453,14 @@ static int ufs_qcom_setup_clocks(struct ufs_hba *hba, bool on,
ufs_qcom_dev_ref_clk_ctrl(host, false);
}
- err = phy_power_off(phy);
- if (err) {
- dev_err(hba->dev, "phy power off failed, ret=%d\n", err);
- return err;
- }
+ ufs_qcom_phy_power_off(host);
}
break;
case POST_CHANGE:
if (on) {
- err = phy_power_on(phy);
- if (err) {
- dev_err(hba->dev, "phy power on failed, ret = %d\n", err);
+ err = ufs_qcom_phy_power_on(host);
+ if (err)
return err;
- }
/* enable the device ref clock for HS mode*/
if (ufshcd_is_hs_mode(&hba->pwr_info))
@@ -1629,7 +1656,7 @@ static void ufs_qcom_exit(struct ufs_hba *hba)
struct ufs_qcom_host *host = ufshcd_get_variant(hba);
ufs_qcom_disable_lane_clks(host);
- phy_power_off(host->generic_phy);
+ ufs_qcom_phy_power_off(host);
phy_exit(host->generic_phy);
}
diff --git a/drivers/ufs/host/ufs-qcom.h b/drivers/ufs/host/ufs-qcom.h
index 5d083331a7f4..6eafba3c203b 100644
--- a/drivers/ufs/host/ufs-qcom.h
+++ b/drivers/ufs/host/ufs-qcom.h
@@ -322,6 +322,7 @@ struct ufs_qcom_host {
struct clk_bulk_data *clks;
u32 num_clks;
bool is_lane_clks_enabled;
+ bool phy_powered_on;
struct icc_path *icc_ddr;
struct icc_path *icc_cpu;
--
2.34.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related
* [PATCH v8 phy-next 12/31] scsi: ufs: qcom: make use of QMP PHY dynamic gear switching ability
From: Vladimir Oltean @ 2026-05-05 10:05 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, Can Guo, James E.J. Bottomley, Martin K. Petersen,
Dmitry Baryshkov, Nitin Rawat, Manivannan Sadhasivam
In-Reply-To: <20260505100523.1922388-1-vladimir.oltean@nxp.com>
Currently, phy_set_mode_ext() on the QMP UFS PHY makes no change to the
hardware state, instead it is mandatory that phy_power_on() followed by
phy_calibrate() be run afterwards, for the new mode to be picked up.
By absorbing the phy_power_off() -> ... -> phy_power_on() ->
phy_calibrate() surrounding sequence into phy_set_mode_ext(), the UFS
HCD consumer driver can be greatly simplified, and we also have a proper
self-standing phy_set_mode_ext() implementation which does not rely on
other calls to do its job.
So simplify ufs_qcom_power_up_sequence() to only call phy_set_mode_ext()
and let PHY power management be handled just by ufs_qcom_setup_clocks().
Actually, after this change, ufs_qcom_power_up_sequence() becomes an
inadequate name, since from the consumer perspective the powering up is
invisible. So change it to ufs_qcom_phy_change_mode().
The consumer and the provider are modified at once because ufs-qcom.c
already calls phy_set_mode_ext() while the QMP PHY is powered on, so
introducing the extra logic in qmp_ufs_set_mode() would cause a
potentially breaking second QMP PHY power sequence until the consumer is
patched to remove its own calls.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
---
Cc: Can Guo <quic_cang@quicinc.com>
Cc: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com>
Cc: "Martin K. Petersen" <martin.petersen@oracle.com>
Cc: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Cc: Nitin Rawat <quic_nitirawa@quicinc.com>
Cc: Manivannan Sadhasivam <mani@kernel.org>
v7->v8: patch is new
Commit was previously posted here but did not get any testing.
https://lore.kernel.org/linux-phy/20260327112858.r5lpqygtvsane2vf@skbuf/
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
---
drivers/phy/qualcomm/phy-qcom-qmp-ufs.c | 6 ++++++
drivers/ufs/host/ufs-qcom.c | 25 +++++--------------------
2 files changed, 11 insertions(+), 20 deletions(-)
diff --git a/drivers/phy/qualcomm/phy-qcom-qmp-ufs.c b/drivers/phy/qualcomm/phy-qcom-qmp-ufs.c
index 771bc7c2ab50..e4e7966eb39a 100644
--- a/drivers/phy/qualcomm/phy-qcom-qmp-ufs.c
+++ b/drivers/phy/qualcomm/phy-qcom-qmp-ufs.c
@@ -2012,6 +2012,12 @@ static int qmp_ufs_set_mode(struct phy *phy, enum phy_mode mode, int submode)
qmp->mode = mode;
qmp->submode = submode;
+ if (phy->power_count) {
+ qmp_ufs_power_off(phy);
+ qmp_ufs_power_on(phy);
+ return qmp_ufs_phy_calibrate(phy);
+ }
+
return 0;
}
diff --git a/drivers/ufs/host/ufs-qcom.c b/drivers/ufs/host/ufs-qcom.c
index 9039b087bf21..e28edcfd13a1 100644
--- a/drivers/ufs/host/ufs-qcom.c
+++ b/drivers/ufs/host/ufs-qcom.c
@@ -485,7 +485,7 @@ static u32 ufs_qcom_get_hs_gear(struct ufs_hba *hba)
return UFS_HS_G3;
}
-static int ufs_qcom_power_up_sequence(struct ufs_hba *hba)
+static int ufs_qcom_phy_change_mode(struct ufs_hba *hba)
{
struct ufs_qcom_host *host = ufshcd_get_variant(hba);
struct ufs_host_params *host_params = &host->host_params;
@@ -508,26 +508,11 @@ static int ufs_qcom_power_up_sequence(struct ufs_hba *hba)
if (ret)
return ret;
- if (phy->power_count)
- phy_power_off(phy);
-
-
- /* phy initialization - calibrate the phy */
ret = phy_set_mode_ext(phy, mode, host->phy_gear);
- if (ret)
- goto out_disable_phy;
-
- /* power on phy - start serdes and phy's power and clocks */
- ret = phy_power_on(phy);
- if (ret) {
- dev_err(hba->dev, "%s: phy power on failed, ret = %d\n",
- __func__, ret);
- return ret;
- }
-
- ret = phy_calibrate(phy);
if (ret) {
- dev_err(hba->dev, "Failed to calibrate PHY: %d\n", ret);
+ dev_err(hba->dev,
+ "Failed to change PHY mode %d submode %d: %pe\n",
+ mode, host->phy_gear, ERR_PTR(ret));
return ret;
}
@@ -582,7 +567,7 @@ static int ufs_qcom_hce_enable_notify(struct ufs_hba *hba,
switch (status) {
case PRE_CHANGE:
- err = ufs_qcom_power_up_sequence(hba);
+ err = ufs_qcom_phy_change_mode(hba);
if (err)
return err;
--
2.34.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related
* [PATCH v8 phy-next 10/31] scsi: ufs: exynos: use dedicated API for updating PHY bus width
From: Vladimir Oltean @ 2026-05-05 10:05 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, Alim Akhtar, Bart Van Assche, Peter Griffin,
James E.J. Bottomley, Martin K. Petersen, Krzysztof Kozlowski,
Chanho Park
In-Reply-To: <20260505100523.1922388-1-vladimir.oltean@nxp.com>
I am trying to get rid of code instances where PHY consumers (like the
Exynos UFS HCD) poke inside struct phy fields, in order to further turn
struct phy into an opaque data structure.
The ufs-exynos.c driver interacts with phy-samsung-ufs.c in order to
power it on and to update the lane count. For the later purpose, it
(ab)uses phy_set_bus_width().
The phy_set_bus_width() function is a PHY provider function, not a
consumer one, and I am calling its use from ufs-exynos.c an abuse
because
(1) commit 8feed347d33b ("phy: add phy_get_bus_width()/phy_set_bus_width()
calls") clearly states the intended use.
(2) phy_set_bus_width() only alters phy->attrs.bus_width, and does not
call into phy_ops at all. So a consumer that makes a call to
phy_set_bus_width() can not possibly produce any hardware change in
the provider at all.
This is where the Exynos UFS HCD driver decided to be creative and
hijacked phy_init() to pick up the change of the bus_width attribute.
This requires a very careful dance where the PHY consumer needs to
simultaneously juggle multiple requirements:
- the UFS PHY needs to pick up the updated lane count in its
samsung_ufs_phy_init() handler for the phy_init() call
- phy_init() calls need to be balanced with phy_exit(), otherwise
subsequent phy_init() calls don't make it into samsung_ufs_phy_init()
and just leave the PHY with an elevated init_count
- phy_power_on() can't be called without phy_init()
The difficulty to observe all requirements is why the following bug fix
commits exist:
3d73b200f989 ("scsi: ufs: ufs-exynos: Change ufs phy control sequence")
7f05fd9a3b6f ("scsi: ufs: exynos: Ensure consistent phy reference counts")
Currently the UFS HCD driver tries to keep the PHY init_count and
power_count in tight lockstep, but even this is error-prone. For
example, if exynos_ufs_suspend() runs and then exynos_ufs_exit(),
the PHY power_count will underflow.
If we address the root issue first (phy_init() abused to pick up new
lane count) by introducing a new PHY consumer method which actually does
call into the PHY provider driver, then we are able to absorb the entire
UFS HCD dance and update the lane count without altering the PHY
init_count or power_count.
This allows more consumer flexibility to call phy_init() from other
places, and same goes for phy_power_on().
It is common practice to call phy_init() only once, right after
phy_get(), and doing so will naturally balance it with phy_exit().
We can also leave the phy_power_on() call to be on demand, placed inside
exynos_ufs_pre_link(). But the UFS core (specifically ufshcd_link_startup())
may call the variant operation exynos_ufs_pre_link() -> exynos_ufs_phy_init()
multiple times if the link startup fails and needs to be retried.
For this reason we need a consumer-specific "bool phy_powered_on" which
ensures that we call phy_power_on() at most once, and that
exynos_ufs_exit() only calls phy_power_off() if phy_power_on() was
previously called. Using the phy->power_count for this purpose is
undesirable because it is going away, and the PHY API is not offering a
helper for it (would be a foot gun, because multiple consumers of the
same provider shouldn't interfere with each other; each should only undo
the effects of what it did itself)
Inside the new samsung_ufs_phy_request_bus_width(), I've sanity checked
that the bus width is either 1 or 2 lanes. This coincides with
samsung_ufs_phy_config() which only configures LANE_0 and LANE_1.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
---
Cc: Alim Akhtar <alim.akhtar@samsung.com>
Cc: Bart Van Assche <bvanassche@acm.org>
Cc: Peter Griffin <peter.griffin@linaro.org>
Cc: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com>
Cc: "Martin K. Petersen" <martin.petersen@oracle.com>
Cc: Krzysztof Kozlowski <krzk@kernel.org>
Cc: Chanho Park <chanho61.park@samsung.com>
v7->v8:
- rewrote commit after Sashiko pointed out the new handling is still
not correct:
https://sashiko.dev/#/patchset/20260430110652.558622-1-vladimir.oltean@nxp.com
- removed Reviewed-by, Tested-by and Acked-by tags from Alim, Bart and
Peter
v6->v7: collect tags from Martin and Peter
v5->v6: collect tags from Alim Akhtar
v4->v5: collect tag, add "scsi: " prefix to commit title
v3->v4: none
v2->v3:
- add Cc Chanho Park, author of commit 3d73b200f989 ("scsi: ufs:
ufs-exynos: Change ufs phy control sequence")
v1->v2:
- add better ufs->phy_powered_on handling in exynos_ufs_exit(),
exynos_ufs_suspend() and exynos_ufs_resume() which ensures we won't
enter a phy->power_count underrun condition
---
drivers/phy/phy-core.c | 35 ++++++++
drivers/phy/samsung/phy-samsung-ufs.c | 27 +++++-
drivers/ufs/host/ufs-exynos.c | 114 ++++++++++++++++++--------
drivers/ufs/host/ufs-exynos.h | 1 +
include/linux/phy/phy.h | 11 +++
5 files changed, 151 insertions(+), 37 deletions(-)
diff --git a/drivers/phy/phy-core.c b/drivers/phy/phy-core.c
index 21aaf2f76e53..634ac07fd1f3 100644
--- a/drivers/phy/phy-core.c
+++ b/drivers/phy/phy-core.c
@@ -606,6 +606,41 @@ int phy_validate(struct phy *phy, enum phy_mode mode, int submode,
}
EXPORT_SYMBOL_GPL(phy_validate);
+/**
+ * phy_request_bus_width() - request PHY to change its bus width
+ * @phy: the phy returned by phy_get()
+ * @bus_width: new bus width
+ *
+ * Consumers can use this method to request the PHY to update itself to a new
+ * bus width (typically meaning lane count). Can be called from any init state
+ * and power state. PHY is expected to use the new lane count as soon as this
+ * method returns.
+ *
+ * Returns: 0 if successful or if operating on an optional and absent PHY,
+ * -EOPNOTSUPP if the operation is not implemented, -EINVAL if the
+ * requested bus width is not supported, other negative error codes for
+ * driver-specific failures.
+ */
+int phy_request_bus_width(struct phy *phy, int bus_width)
+{
+ int ret;
+
+ if (!phy)
+ return 0;
+
+ if (!phy->ops->request_bus_width)
+ return -EOPNOTSUPP;
+
+ mutex_lock(&phy->mutex);
+ ret = phy->ops->request_bus_width(phy, bus_width);
+ if (!ret)
+ phy_set_bus_width(phy, bus_width);
+ mutex_unlock(&phy->mutex);
+
+ return ret;
+}
+EXPORT_SYMBOL_GPL(phy_request_bus_width);
+
/**
* _of_phy_get() - lookup and obtain a reference to a phy by phandle
* @np: device_node for which to get the phy
diff --git a/drivers/phy/samsung/phy-samsung-ufs.c b/drivers/phy/samsung/phy-samsung-ufs.c
index ee665f26c236..b09a35ab6acd 100644
--- a/drivers/phy/samsung/phy-samsung-ufs.c
+++ b/drivers/phy/samsung/phy-samsung-ufs.c
@@ -165,7 +165,6 @@ static int samsung_ufs_phy_init(struct phy *phy)
{
struct samsung_ufs_phy *ss_phy = get_samsung_ufs_phy(phy);
- ss_phy->lane_cnt = phy->attrs.bus_width;
ss_phy->ufs_phy_state = CFG_PRE_INIT;
return 0;
@@ -204,6 +203,31 @@ static int samsung_ufs_phy_power_off(struct phy *phy)
return 0;
}
+static int samsung_ufs_phy_request_bus_width(struct phy *phy, int bus_width)
+{
+ struct samsung_ufs_phy *ss_phy = get_samsung_ufs_phy(phy);
+ u8 old_lane_cnt = ss_phy->lane_cnt;
+ int err = 0;
+
+ if (bus_width != 1 && bus_width != 2)
+ return -EINVAL;
+
+ ss_phy->lane_cnt = bus_width;
+
+ if (phy->init_count)
+ samsung_ufs_phy_init(phy);
+
+ /* If the init_count is 0, the power_count should also be 0 */
+ if (phy->power_count) {
+ samsung_ufs_phy_power_off(phy);
+ err = samsung_ufs_phy_power_on(phy);
+ if (err)
+ ss_phy->lane_cnt = old_lane_cnt;
+ }
+
+ return err;
+}
+
static int samsung_ufs_phy_set_mode(struct phy *generic_phy,
enum phy_mode mode, int submode)
{
@@ -272,6 +296,7 @@ static const struct phy_ops samsung_ufs_phy_ops = {
.calibrate = samsung_ufs_phy_calibrate,
.set_mode = samsung_ufs_phy_set_mode,
.notify_phystate = samsung_ufs_phy_notify_state,
+ .request_bus_width = samsung_ufs_phy_request_bus_width,
.owner = THIS_MODULE,
};
diff --git a/drivers/ufs/host/ufs-exynos.c b/drivers/ufs/host/ufs-exynos.c
index 77a6c8e44485..b90876b268db 100644
--- a/drivers/ufs/host/ufs-exynos.c
+++ b/drivers/ufs/host/ufs-exynos.c
@@ -931,11 +931,73 @@ static void exynos_ufs_specify_nexus_t_tm_req(struct ufs_hba *hba,
}
}
-static int exynos_ufs_phy_init(struct exynos_ufs *ufs)
+static int exynos_ufs_phy_init(struct device *dev, struct exynos_ufs *ufs)
+{
+ struct phy *generic_phy;
+ int ret;
+
+ generic_phy = devm_phy_get(dev, "ufs-phy");
+ if (IS_ERR(generic_phy)) {
+ ret = PTR_ERR(generic_phy);
+ dev_err(dev, "failed to get ufs-phy: %pe\n", ERR_PTR(ret));
+ return ret;
+ }
+
+ ret = phy_init(generic_phy);
+ if (ret) {
+ dev_err(dev, "phy init failed: %pe\n", ERR_PTR(ret));
+ return ret;
+ }
+
+ ufs->phy = generic_phy;
+
+ return ret;
+}
+
+static void exynos_ufs_phy_exit(struct exynos_ufs *ufs)
+{
+ phy_exit(ufs->phy);
+}
+
+static int exynos_ufs_phy_power_on(struct exynos_ufs *ufs)
+{
+ int ret;
+
+ if (ufs->phy_powered_on)
+ return 0;
+
+ ret = phy_power_on(ufs->phy);
+ if (ret) {
+ dev_err(ufs->hba->dev, "Failed to power on PHY: %pe\n",
+ ERR_PTR(ret));
+ return ret;
+ }
+
+ ufs->phy_powered_on = true;
+
+ return 0;
+}
+
+static void exynos_ufs_phy_power_off(struct exynos_ufs *ufs)
+{
+ int ret;
+
+ if (!ufs->phy_powered_on)
+ return;
+
+ ret = phy_power_off(ufs->phy);
+ if (ret)
+ dev_warn(ufs->hba->dev, "Failed to power off PHY: %pe\n",
+ ERR_PTR(ret));
+
+ ufs->phy_powered_on = false;
+}
+
+static int exynos_ufs_phy_update_bus_width(struct exynos_ufs *ufs)
{
struct ufs_hba *hba = ufs->hba;
struct phy *generic_phy = ufs->phy;
- int ret = 0;
+ int ret;
if (ufs->avail_ln_rx == 0 || ufs->avail_ln_tx == 0) {
ufshcd_dme_get(hba, UIC_ARG_MIB(PA_AVAILRXDATALANES),
@@ -947,30 +1009,11 @@ static int exynos_ufs_phy_init(struct exynos_ufs *ufs)
ufs->avail_ln_rx, ufs->avail_ln_tx);
}
- phy_set_bus_width(generic_phy, ufs->avail_ln_rx);
-
- if (generic_phy->power_count) {
- phy_power_off(generic_phy);
- phy_exit(generic_phy);
- }
-
- ret = phy_init(generic_phy);
- if (ret) {
- dev_err(hba->dev, "%s: phy init failed, ret = %d\n",
- __func__, ret);
- return ret;
- }
-
- ret = phy_power_on(generic_phy);
+ ret = phy_request_bus_width(generic_phy, ufs->avail_ln_rx);
if (ret)
- goto out_exit_phy;
-
- return 0;
-
-out_exit_phy:
- phy_exit(generic_phy);
+ return ret;
- return ret;
+ return exynos_ufs_phy_power_on(ufs);
}
static void exynos_ufs_config_unipro(struct exynos_ufs *ufs)
@@ -1055,7 +1098,7 @@ static int exynos_ufs_pre_link(struct ufs_hba *hba)
ufs->drv_data->pre_link(ufs);
/* m-phy */
- exynos_ufs_phy_init(ufs);
+ exynos_ufs_phy_update_bus_width(ufs);
if (!(ufs->opts & EXYNOS_UFS_OPT_SKIP_CONFIG_PHY_ATTR)) {
exynos_ufs_config_phy_time_attr(ufs);
exynos_ufs_config_phy_cap_attr(ufs);
@@ -1475,12 +1518,9 @@ static int exynos_ufs_init(struct ufs_hba *hba)
goto out;
}
- ufs->phy = devm_phy_get(dev, "ufs-phy");
- if (IS_ERR(ufs->phy)) {
- ret = PTR_ERR(ufs->phy);
- dev_err(dev, "failed to get ufs-phy\n");
+ ret = exynos_ufs_phy_init(dev, ufs);
+ if (ret)
goto out;
- }
exynos_ufs_priv_init(hba, ufs);
@@ -1490,13 +1530,13 @@ static int exynos_ufs_init(struct ufs_hba *hba)
ret = ufs->drv_data->drv_init(ufs);
if (ret) {
dev_err(dev, "failed to init drv-data\n");
- goto out;
+ goto out_phy_exit;
}
}
ret = exynos_ufs_get_clk_info(ufs);
if (ret)
- goto out;
+ goto out_phy_exit;
exynos_ufs_specify_phy_time_attr(ufs);
exynos_ufs_config_smu(ufs);
@@ -1504,6 +1544,8 @@ static int exynos_ufs_init(struct ufs_hba *hba)
hba->host->dma_alignment = DATA_UNIT_SIZE - 1;
return 0;
+out_phy_exit:
+ exynos_ufs_phy_exit(ufs);
out:
hba->priv = NULL;
return ret;
@@ -1513,8 +1555,8 @@ static void exynos_ufs_exit(struct ufs_hba *hba)
{
struct exynos_ufs *ufs = ufshcd_get_variant(hba);
- phy_power_off(ufs->phy);
- phy_exit(ufs->phy);
+ exynos_ufs_phy_power_off(ufs);
+ exynos_ufs_phy_exit(ufs);
}
static int exynos_ufs_host_reset(struct ufs_hba *hba)
@@ -1728,7 +1770,7 @@ static int exynos_ufs_suspend(struct ufs_hba *hba, enum ufs_pm_op pm_op,
ufs->drv_data->suspend(ufs);
if (!ufshcd_is_link_active(hba))
- phy_power_off(ufs->phy);
+ exynos_ufs_phy_power_off(ufs);
return 0;
}
@@ -1738,7 +1780,7 @@ static int exynos_ufs_resume(struct ufs_hba *hba, enum ufs_pm_op pm_op)
struct exynos_ufs *ufs = ufshcd_get_variant(hba);
if (!ufshcd_is_link_active(hba))
- phy_power_on(ufs->phy);
+ exynos_ufs_phy_power_on(ufs);
exynos_ufs_config_smu(ufs);
exynos_ufs_fmp_resume(hba);
diff --git a/drivers/ufs/host/ufs-exynos.h b/drivers/ufs/host/ufs-exynos.h
index abe7e472759e..683b9150e2ba 100644
--- a/drivers/ufs/host/ufs-exynos.h
+++ b/drivers/ufs/host/ufs-exynos.h
@@ -227,6 +227,7 @@ struct exynos_ufs {
int avail_ln_rx;
int avail_ln_tx;
int rx_sel_idx;
+ bool phy_powered_on;
struct ufs_pa_layer_attr dev_req_params;
struct ufs_phy_time_cfg t_cfg;
ktime_t entry_hibern8_t;
diff --git a/include/linux/phy/phy.h b/include/linux/phy/phy.h
index ea47975e288a..01af84f97608 100644
--- a/include/linux/phy/phy.h
+++ b/include/linux/phy/phy.h
@@ -93,6 +93,7 @@ union phy_configure_opts {
* @reset: resetting the phy
* @calibrate: calibrate the phy
* @notify_phystate: notify and configure the phy for a particular state
+ * @request_bus_width: request a different bus width for the phy
* @release: ops to be performed while the consumer relinquishes the PHY
* @owner: the module owner containing the ops
*/
@@ -143,6 +144,7 @@ struct phy_ops {
int (*disconnect)(struct phy *phy, int port);
int (*notify_phystate)(struct phy *phy, union phy_notify state);
+ int (*request_bus_width)(struct phy *phy, int bus_width);
void (*release)(struct phy *phy);
struct module *owner;
};
@@ -275,6 +277,7 @@ static inline void phy_set_bus_width(struct phy *phy, int bus_width)
{
phy->attrs.bus_width = bus_width;
}
+int phy_request_bus_width(struct phy *phy, int bus_width);
struct phy *phy_get(struct device *dev, const char *string);
struct phy *devm_phy_get(struct device *dev, const char *string);
struct phy *devm_phy_optional_get(struct device *dev, const char *string);
@@ -456,6 +459,14 @@ static inline void phy_set_bus_width(struct phy *phy, int bus_width)
return;
}
+static inline int phy_request_bus_width(struct phy *phy, int bus_width)
+{
+ if (!phy)
+ return 0;
+
+ return -ENOSYS;
+}
+
static inline struct phy *phy_get(struct device *dev, const char *string)
{
return ERR_PTR(-ENOSYS);
--
2.34.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related
* [PATCH v8 phy-next 08/31] net: lan969x: include missing <linux/of.h>
From: Vladimir Oltean @ 2026-05-05 10:05 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, Daniel Machon, Andrew Lunn, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Steen Hegelund
In-Reply-To: <20260505100523.1922388-1-vladimir.oltean@nxp.com>
This file is calling of_property_read_u32() without including the proper
header for it. It is provided by <linux/phy/phy.h>, which wants to get
rid of it.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Acked-by: Daniel Machon <daniel.machon@microchip.com>
---
Cc: Daniel Machon <daniel.machon@microchip.com>
Cc: Andrew Lunn <andrew+netdev@lunn.ch>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Jakub Kicinski <kuba@kernel.org>
Cc: Paolo Abeni <pabeni@redhat.com>
Cc: Steen Hegelund <Steen.Hegelund@microchip.com>
v2->v8: none
v1->v2: collect tag
---
drivers/net/ethernet/microchip/sparx5/lan969x/lan969x_rgmii.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/ethernet/microchip/sparx5/lan969x/lan969x_rgmii.c b/drivers/net/ethernet/microchip/sparx5/lan969x/lan969x_rgmii.c
index 4e422ca50828..249114b40c42 100644
--- a/drivers/net/ethernet/microchip/sparx5/lan969x/lan969x_rgmii.c
+++ b/drivers/net/ethernet/microchip/sparx5/lan969x/lan969x_rgmii.c
@@ -4,6 +4,7 @@
* Copyright (c) 2024 Microchip Technology Inc. and its subsidiaries.
*/
+#include <linux/of.h>
#include "lan969x.h"
/* Tx clock selectors */
--
2.34.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related
* [PATCH v8 phy-next 09/31] PCI: Remove device links to PHY
From: Vladimir Oltean @ 2026-05-05 10:05 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, Bjorn Helgaas, Manivannan Sadhasivam,
Lorenzo Pieralisi, Krzysztof Wilczyński, Rob Herring,
Vignesh Raghavendra, Siddharth Vadapalli
In-Reply-To: <20260505100523.1922388-1-vladimir.oltean@nxp.com>
This is practically a full revert of commit
7a4db656a635 ("PCI: dra7xx: Create functional dependency between PCIe and PHY")
and a partial revert of the device link pieces from commits
dfb80534692d ("PCI: cadence: Add generic PHY support to host and EP drivers")
49229238ab47 ("PCI: keystone: Cleanup PHY handling")
The trouble with these commits is that they dereference fields inside
struct phy from a consumer driver, which will become no longer possible.
Since commit 987351e1ea77 ("phy: core: Add consumer device link
support") from 2019, the PHY core also adds a device link to order PHY
provider and consumer suspend/resume operations. All reverted commits
are from 2017-2018, and what they do should actually be redundant now.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Manivannan Sadhasivam <mani@kernel.org>
---
Cc: Lorenzo Pieralisi <lpieralisi@kernel.org>
Cc: "Krzysztof Wilczyński" <kwilczynski@kernel.org>
Cc: Manivannan Sadhasivam <mani@kernel.org>
Cc: Rob Herring <robh@kernel.org>
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: Vignesh Raghavendra <vigneshr@ti.com>
Cc: Siddharth Vadapalli <s-vadapalli@ti.com>
v7->v8: only context change - "return 0" became "return ret" for
cdns_plat_pcie_probe()
v6->v7: none
v5->v6: collect tag from Manivannan Sadhasivam
v3->v5: none
v2->v3:
- remove dangling set but unused phy_count local variable in
cdns_plat_pcie_probe()
v1->v2:
- fully remove struct device link **link from struct cdns_pcie and from
cdns_plat_pcie_probe() error path
- collect tag from Bjorn Helgaas
- adjust commit title
---
.../controller/cadence/pcie-cadence-plat.c | 4 ---
drivers/pci/controller/cadence/pcie-cadence.c | 16 +---------
drivers/pci/controller/cadence/pcie-cadence.h | 2 --
drivers/pci/controller/dwc/pci-dra7xx.c | 16 ----------
drivers/pci/controller/dwc/pci-keystone.c | 31 +++----------------
5 files changed, 5 insertions(+), 64 deletions(-)
diff --git a/drivers/pci/controller/cadence/pcie-cadence-plat.c b/drivers/pci/controller/cadence/pcie-cadence-plat.c
index 8b12a46b5601..6fd16e09e244 100644
--- a/drivers/pci/controller/cadence/pcie-cadence-plat.c
+++ b/drivers/pci/controller/cadence/pcie-cadence-plat.c
@@ -41,7 +41,6 @@ static int cdns_plat_pcie_probe(struct platform_device *pdev)
struct pci_host_bridge *bridge;
struct cdns_pcie_ep *ep;
struct cdns_pcie_rc *rc;
- int phy_count;
bool is_rc;
int ret;
@@ -122,9 +121,6 @@ static int cdns_plat_pcie_probe(struct platform_device *pdev)
pm_runtime_put_sync(dev);
pm_runtime_disable(dev);
cdns_pcie_disable_phy(cdns_plat_pcie->pcie);
- phy_count = cdns_plat_pcie->pcie->phy_count;
- while (phy_count--)
- device_link_del(cdns_plat_pcie->pcie->link[phy_count]);
return ret;
}
diff --git a/drivers/pci/controller/cadence/pcie-cadence.c b/drivers/pci/controller/cadence/pcie-cadence.c
index a1eada56edba..0ac980249941 100644
--- a/drivers/pci/controller/cadence/pcie-cadence.c
+++ b/drivers/pci/controller/cadence/pcie-cadence.c
@@ -222,7 +222,6 @@ int cdns_pcie_init_phy(struct device *dev, struct cdns_pcie *pcie)
struct device_node *np = dev->of_node;
int phy_count;
struct phy **phy;
- struct device_link **link;
int i;
int ret;
const char *name;
@@ -238,10 +237,6 @@ int cdns_pcie_init_phy(struct device *dev, struct cdns_pcie *pcie)
if (!phy)
return -ENOMEM;
- link = devm_kcalloc(dev, phy_count, sizeof(*link), GFP_KERNEL);
- if (!link)
- return -ENOMEM;
-
for (i = 0; i < phy_count; i++) {
of_property_read_string_index(np, "phy-names", i, &name);
phy[i] = devm_phy_get(dev, name);
@@ -249,17 +244,10 @@ int cdns_pcie_init_phy(struct device *dev, struct cdns_pcie *pcie)
ret = PTR_ERR(phy[i]);
goto err_phy;
}
- link[i] = device_link_add(dev, &phy[i]->dev, DL_FLAG_STATELESS);
- if (!link[i]) {
- devm_phy_put(dev, phy[i]);
- ret = -EINVAL;
- goto err_phy;
- }
}
pcie->phy_count = phy_count;
pcie->phy = phy;
- pcie->link = link;
ret = cdns_pcie_enable_phy(pcie);
if (ret)
@@ -268,10 +256,8 @@ int cdns_pcie_init_phy(struct device *dev, struct cdns_pcie *pcie)
return 0;
err_phy:
- while (--i >= 0) {
- device_link_del(link[i]);
+ while (--i >= 0)
devm_phy_put(dev, phy[i]);
- }
return ret;
}
diff --git a/drivers/pci/controller/cadence/pcie-cadence.h b/drivers/pci/controller/cadence/pcie-cadence.h
index 574e9cf4d003..232b504b5cf9 100644
--- a/drivers/pci/controller/cadence/pcie-cadence.h
+++ b/drivers/pci/controller/cadence/pcie-cadence.h
@@ -82,7 +82,6 @@ struct cdns_plat_pcie_of_data {
* @is_rc: tell whether the PCIe controller mode is Root Complex or Endpoint.
* @phy_count: number of supported PHY devices
* @phy: list of pointers to specific PHY control blocks
- * @link: list of pointers to corresponding device link representations
* @ops: Platform-specific ops to control various inputs from Cadence PCIe
* wrapper
* @cdns_pcie_reg_offsets: Register bank offsets for different SoC
@@ -95,7 +94,6 @@ struct cdns_pcie {
bool is_rc;
int phy_count;
struct phy **phy;
- struct device_link **link;
const struct cdns_pcie_ops *ops;
const struct cdns_plat_pcie_of_data *cdns_pcie_reg_offsets;
};
diff --git a/drivers/pci/controller/dwc/pci-dra7xx.c b/drivers/pci/controller/dwc/pci-dra7xx.c
index cd904659c321..9c1aef4d0bd1 100644
--- a/drivers/pci/controller/dwc/pci-dra7xx.c
+++ b/drivers/pci/controller/dwc/pci-dra7xx.c
@@ -9,7 +9,6 @@
#include <linux/clk.h>
#include <linux/delay.h>
-#include <linux/device.h>
#include <linux/err.h>
#include <linux/interrupt.h>
#include <linux/irq.h>
@@ -679,7 +678,6 @@ static int dra7xx_pcie_probe(struct platform_device *pdev)
int i;
int phy_count;
struct phy **phy;
- struct device_link **link;
void __iomem *base;
struct dw_pcie *pci;
struct dra7xx_pcie *dra7xx;
@@ -727,10 +725,6 @@ static int dra7xx_pcie_probe(struct platform_device *pdev)
if (!phy)
return -ENOMEM;
- link = devm_kcalloc(dev, phy_count, sizeof(*link), GFP_KERNEL);
- if (!link)
- return -ENOMEM;
-
dra7xx->clk = devm_clk_get_optional(dev, NULL);
if (IS_ERR(dra7xx->clk))
return dev_err_probe(dev, PTR_ERR(dra7xx->clk),
@@ -745,12 +739,6 @@ static int dra7xx_pcie_probe(struct platform_device *pdev)
phy[i] = devm_phy_get(dev, name);
if (IS_ERR(phy[i]))
return PTR_ERR(phy[i]);
-
- link[i] = device_link_add(dev, &phy[i]->dev, DL_FLAG_STATELESS);
- if (!link[i]) {
- ret = -EINVAL;
- goto err_link;
- }
}
dra7xx->base = base;
@@ -852,10 +840,6 @@ static int dra7xx_pcie_probe(struct platform_device *pdev)
pm_runtime_disable(dev);
dra7xx_pcie_disable_phy(dra7xx);
-err_link:
- while (--i >= 0)
- device_link_del(link[i]);
-
return ret;
}
diff --git a/drivers/pci/controller/dwc/pci-keystone.c b/drivers/pci/controller/dwc/pci-keystone.c
index fe4836925c4e..cab9fad212ad 100644
--- a/drivers/pci/controller/dwc/pci-keystone.c
+++ b/drivers/pci/controller/dwc/pci-keystone.c
@@ -130,7 +130,6 @@ struct keystone_pcie {
int num_lanes;
u32 num_viewport;
struct phy **phy;
- struct device_link **link;
struct device_node *msi_intc_np;
struct irq_domain *intx_irq_domain;
struct device_node *np;
@@ -1130,7 +1129,6 @@ static int ks_pcie_probe(struct platform_device *pdev)
enum dw_pcie_device_mode mode;
struct dw_pcie *pci;
struct keystone_pcie *ks_pcie;
- struct device_link **link;
struct gpio_desc *gpiod;
struct resource *res;
void __iomem *base;
@@ -1201,31 +1199,17 @@ static int ks_pcie_probe(struct platform_device *pdev)
if (!phy)
return -ENOMEM;
- link = devm_kcalloc(dev, num_lanes, sizeof(*link), GFP_KERNEL);
- if (!link)
- return -ENOMEM;
-
for (i = 0; i < num_lanes; i++) {
snprintf(name, sizeof(name), "pcie-phy%d", i);
phy[i] = devm_phy_optional_get(dev, name);
if (IS_ERR(phy[i])) {
ret = PTR_ERR(phy[i]);
- goto err_link;
- }
-
- if (!phy[i])
- continue;
-
- link[i] = device_link_add(dev, &phy[i]->dev, DL_FLAG_STATELESS);
- if (!link[i]) {
- ret = -EINVAL;
- goto err_link;
+ goto err;
}
}
ks_pcie->np = np;
ks_pcie->pci = pci;
- ks_pcie->link = link;
ks_pcie->num_lanes = num_lanes;
ks_pcie->phy = phy;
@@ -1235,7 +1219,7 @@ static int ks_pcie_probe(struct platform_device *pdev)
ret = PTR_ERR(gpiod);
if (ret != -EPROBE_DEFER)
dev_err(dev, "Failed to get reset GPIO\n");
- goto err_link;
+ goto err;
}
/* Obtain references to the PHYs */
@@ -1250,7 +1234,7 @@ static int ks_pcie_probe(struct platform_device *pdev)
if (ret) {
dev_err(dev, "failed to enable phy\n");
- goto err_link;
+ goto err;
}
platform_set_drvdata(pdev, ks_pcie);
@@ -1337,25 +1321,18 @@ static int ks_pcie_probe(struct platform_device *pdev)
pm_runtime_disable(dev);
ks_pcie_disable_phy(ks_pcie);
-err_link:
- while (--i >= 0 && link[i])
- device_link_del(link[i]);
-
+err:
return ret;
}
static void ks_pcie_remove(struct platform_device *pdev)
{
struct keystone_pcie *ks_pcie = platform_get_drvdata(pdev);
- struct device_link **link = ks_pcie->link;
- int num_lanes = ks_pcie->num_lanes;
struct device *dev = &pdev->dev;
pm_runtime_put(dev);
pm_runtime_disable(dev);
ks_pcie_disable_phy(ks_pcie);
- while (num_lanes--)
- device_link_del(link[num_lanes]);
}
static struct platform_driver ks_pcie_driver = {
--
2.34.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related
* [PATCH v8 phy-next 07/31] phy: spacemit: include missing <linux/phy/phy.h>
From: Vladimir Oltean @ 2026-05-05 10:04 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, Yixun Lan
In-Reply-To: <20260505100523.1922388-1-vladimir.oltean@nxp.com>
This driver relies on a transitive inclusion of the PHY API header
through the USB headers.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Yixun Lan <dlan@kernel.org>
---
Cc: Yixun Lan <dlan@kernel.org>
v5->v8: none
v4->v5: collect tag
v1->v4: none
---
drivers/phy/spacemit/phy-k1-usb2.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/phy/spacemit/phy-k1-usb2.c b/drivers/phy/spacemit/phy-k1-usb2.c
index 9215d0b223b2..01af310d6683 100644
--- a/drivers/phy/spacemit/phy-k1-usb2.c
+++ b/drivers/phy/spacemit/phy-k1-usb2.c
@@ -9,6 +9,7 @@
#include <linux/bitfield.h>
#include <linux/clk.h>
#include <linux/iopoll.h>
+#include <linux/phy/phy.h>
#include <linux/platform_device.h>
#include <linux/regmap.h>
#include <linux/usb/of.h>
--
2.34.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related
* [PATCH v8 phy-next 06/31] phy: add <linux/pm_runtime.h> where missing
From: Vladimir Oltean @ 2026-05-05 10:04 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, Geert Uytterhoeven, André Draszik,
Peter Griffin, Tudor Ambarus, Magnus Damm, Heiko Stuebner
In-Reply-To: <20260505100523.1922388-1-vladimir.oltean@nxp.com>
It appears that a number of PHY provider drivers call runtime PM
operations without including the proper header.
This was provided by <linux/phy/phy.h> but no function exported by this
header directly needs it. So we need to drop it from there, and fix up
drivers that used to depend on that.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be> # renesas
Reviewed-by: André Draszik <andre.draszik@linaro.org> # google
---
Cc: Peter Griffin <peter.griffin@linaro.org>
Cc: "André Draszik" <andre.draszik@linaro.org>
Cc: Tudor Ambarus <tudor.ambarus@linaro.org>
Cc: Geert Uytterhoeven <geert+renesas@glider.be>
Cc: Magnus Damm <magnus.damm@gmail.com>
Cc: Heiko Stuebner <heiko@sntech.de>
v7->v8:
- stop listing the PHY drivers which need to be changed. The list
was incomplete; not listing 12 drivers. Flagged by Sashiko:
https://sashiko.dev/#/patchset/20260430110652.558622-1-vladimir.oltean@nxp.com
v2->v7: none
v1->v2: collect tags
---
drivers/phy/motorola/phy-mapphone-mdm6600.c | 1 +
drivers/phy/phy-google-usb.c | 1 +
drivers/phy/qualcomm/phy-qcom-qmp-combo.c | 1 +
drivers/phy/qualcomm/phy-qcom-qmp-usb-legacy.c | 1 +
drivers/phy/qualcomm/phy-qcom-qmp-usb.c | 1 +
drivers/phy/qualcomm/phy-qcom-qmp-usbc.c | 1 +
drivers/phy/qualcomm/phy-qcom-qusb2.c | 1 +
drivers/phy/qualcomm/phy-qcom-snps-femto-v2.c | 1 +
drivers/phy/renesas/phy-rcar-gen3-pcie.c | 1 +
drivers/phy/renesas/r8a779f0-ether-serdes.c | 1 +
drivers/phy/rockchip/phy-rockchip-typec.c | 1 +
drivers/phy/ti/phy-tusb1210.c | 1 +
12 files changed, 12 insertions(+)
diff --git a/drivers/phy/motorola/phy-mapphone-mdm6600.c b/drivers/phy/motorola/phy-mapphone-mdm6600.c
index fd0e0cd1c1cf..ce1dad8c438d 100644
--- a/drivers/phy/motorola/phy-mapphone-mdm6600.c
+++ b/drivers/phy/motorola/phy-mapphone-mdm6600.c
@@ -17,6 +17,7 @@
#include <linux/of_platform.h>
#include <linux/phy/phy.h>
#include <linux/pinctrl/consumer.h>
+#include <linux/pm_runtime.h>
#define PHY_MDM6600_PHY_DELAY_MS 4000 /* PHY enable 2.2s to 3.5s */
#define PHY_MDM6600_ENABLED_DELAY_MS 8000 /* 8s more total for MDM6600 */
diff --git a/drivers/phy/phy-google-usb.c b/drivers/phy/phy-google-usb.c
index ab20bc20f19e..48cfa2e28347 100644
--- a/drivers/phy/phy-google-usb.c
+++ b/drivers/phy/phy-google-usb.c
@@ -16,6 +16,7 @@
#include <linux/of.h>
#include <linux/phy/phy.h>
#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
#include <linux/regmap.h>
#include <linux/reset.h>
#include <linux/usb/typec_mux.h>
diff --git a/drivers/phy/qualcomm/phy-qcom-qmp-combo.c b/drivers/phy/qualcomm/phy-qcom-qmp-combo.c
index 93f1aa10d400..b9ea7d058e93 100644
--- a/drivers/phy/qualcomm/phy-qcom-qmp-combo.c
+++ b/drivers/phy/qualcomm/phy-qcom-qmp-combo.c
@@ -16,6 +16,7 @@
#include <linux/of_graph.h>
#include <linux/phy/phy.h>
#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
#include <linux/regulator/consumer.h>
#include <linux/reset.h>
#include <linux/slab.h>
diff --git a/drivers/phy/qualcomm/phy-qcom-qmp-usb-legacy.c b/drivers/phy/qualcomm/phy-qcom-qmp-usb-legacy.c
index 8bf951b0490c..2bd5862c5ba8 100644
--- a/drivers/phy/qualcomm/phy-qcom-qmp-usb-legacy.c
+++ b/drivers/phy/qualcomm/phy-qcom-qmp-usb-legacy.c
@@ -16,6 +16,7 @@
#include <linux/of_address.h>
#include <linux/phy/phy.h>
#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
#include <linux/regulator/consumer.h>
#include <linux/reset.h>
#include <linux/slab.h>
diff --git a/drivers/phy/qualcomm/phy-qcom-qmp-usb.c b/drivers/phy/qualcomm/phy-qcom-qmp-usb.c
index b0ecd5ba2464..d88b8a415e85 100644
--- a/drivers/phy/qualcomm/phy-qcom-qmp-usb.c
+++ b/drivers/phy/qualcomm/phy-qcom-qmp-usb.c
@@ -15,6 +15,7 @@
#include <linux/of_address.h>
#include <linux/phy/phy.h>
#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
#include <linux/regulator/consumer.h>
#include <linux/reset.h>
#include <linux/slab.h>
diff --git a/drivers/phy/qualcomm/phy-qcom-qmp-usbc.c b/drivers/phy/qualcomm/phy-qcom-qmp-usbc.c
index c342479a3798..f62e1f6ecc07 100644
--- a/drivers/phy/qualcomm/phy-qcom-qmp-usbc.c
+++ b/drivers/phy/qualcomm/phy-qcom-qmp-usbc.c
@@ -16,6 +16,7 @@
#include <linux/of_address.h>
#include <linux/phy/phy.h>
#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
#include <linux/regmap.h>
#include <linux/regulator/consumer.h>
#include <linux/reset.h>
diff --git a/drivers/phy/qualcomm/phy-qcom-qusb2.c b/drivers/phy/qualcomm/phy-qcom-qusb2.c
index eb93015be841..191040f6d60f 100644
--- a/drivers/phy/qualcomm/phy-qcom-qusb2.c
+++ b/drivers/phy/qualcomm/phy-qcom-qusb2.c
@@ -14,6 +14,7 @@
#include <linux/of.h>
#include <linux/phy/phy.h>
#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
#include <linux/regmap.h>
#include <linux/regulator/consumer.h>
#include <linux/reset.h>
diff --git a/drivers/phy/qualcomm/phy-qcom-snps-femto-v2.c b/drivers/phy/qualcomm/phy-qcom-snps-femto-v2.c
index eb0b0f61d98e..8915fa250e81 100644
--- a/drivers/phy/qualcomm/phy-qcom-snps-femto-v2.c
+++ b/drivers/phy/qualcomm/phy-qcom-snps-femto-v2.c
@@ -12,6 +12,7 @@
#include <linux/of.h>
#include <linux/phy/phy.h>
#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
#include <linux/regmap.h>
#include <linux/regulator/consumer.h>
#include <linux/reset.h>
diff --git a/drivers/phy/renesas/phy-rcar-gen3-pcie.c b/drivers/phy/renesas/phy-rcar-gen3-pcie.c
index c0e5a4ac82de..3e2cf59ad480 100644
--- a/drivers/phy/renesas/phy-rcar-gen3-pcie.c
+++ b/drivers/phy/renesas/phy-rcar-gen3-pcie.c
@@ -11,6 +11,7 @@
#include <linux/of.h>
#include <linux/phy/phy.h>
#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
#include <linux/spinlock.h>
#define PHY_CTRL 0x4000 /* R8A77980 only */
diff --git a/drivers/phy/renesas/r8a779f0-ether-serdes.c b/drivers/phy/renesas/r8a779f0-ether-serdes.c
index 8a6b6f366fe3..c34427ac4fdb 100644
--- a/drivers/phy/renesas/r8a779f0-ether-serdes.c
+++ b/drivers/phy/renesas/r8a779f0-ether-serdes.c
@@ -12,6 +12,7 @@
#include <linux/phy.h>
#include <linux/phy/phy.h>
#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
#include <linux/reset.h>
#define R8A779F0_ETH_SERDES_NUM 3
diff --git a/drivers/phy/rockchip/phy-rockchip-typec.c b/drivers/phy/rockchip/phy-rockchip-typec.c
index d9701b6106d5..0a318ccf1bbf 100644
--- a/drivers/phy/rockchip/phy-rockchip-typec.c
+++ b/drivers/phy/rockchip/phy-rockchip-typec.c
@@ -49,6 +49,7 @@
#include <linux/of_address.h>
#include <linux/of_platform.h>
#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
#include <linux/regmap.h>
#include <linux/reset.h>
diff --git a/drivers/phy/ti/phy-tusb1210.c b/drivers/phy/ti/phy-tusb1210.c
index c3ae9d7948d7..b7080403e649 100644
--- a/drivers/phy/ti/phy-tusb1210.c
+++ b/drivers/phy/ti/phy-tusb1210.c
@@ -13,6 +13,7 @@
#include <linux/ulpi/regs.h>
#include <linux/gpio/consumer.h>
#include <linux/phy/ulpi_phy.h>
+#include <linux/pm_runtime.h>
#include <linux/power_supply.h>
#include <linux/property.h>
#include <linux/workqueue.h>
--
2.34.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related
* [PATCH v8 phy-next 05/31] drm: add <linux/pm_runtime.h> where missing
From: Vladimir Oltean @ 2026-05-05 10:04 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, Andrzej Hajda, Robert Foss, Laurent Pinchart,
Jonas Karlman, Jernej Skrabec, Maarten Lankhorst, Maxime Ripard,
Thomas Zimmermann, David Airlie, Simona Vetter, Inki Dae,
Jagan Teki, Marek Szyprowski, Rob Clark, Dmitry Baryshkov
In-Reply-To: <20260505100523.1922388-1-vladimir.oltean@nxp.com>
Multiple DRM bridge drivers use runtime PM operations without
including the proper header, instead relying on transitive inclusion
by <linux/phy/phy.h>.
The PHY subsystem wants to get rid of headers it provides for no reason,
so modify these drivers to include what they need directly.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
---
Cc: Andrzej Hajda <andrzej.hajda@intel.com>
Cc: Neil Armstrong <neil.armstrong@linaro.org>
Cc: Robert Foss <rfoss@kernel.org>
Cc: Laurent Pinchart <Laurent.pinchart@ideasonboard.com>
Cc: Jonas Karlman <jonas@kwiboo.se>
Cc: Jernej Skrabec <jernej.skrabec@gmail.com>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Maxime Ripard <mripard@kernel.org>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: David Airlie <airlied@gmail.com>
Cc: Simona Vetter <simona@ffwll.ch>
Cc: Inki Dae <inki.dae@samsung.com>
Cc: Jagan Teki <jagan@amarulasolutions.com>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Cc: Rob Clark <robin.clark@oss.qualcomm.com>
Cc: Dmitry Baryshkov <lumag@kernel.org>
v1->v8: none
---
drivers/gpu/drm/bridge/analogix/analogix_dp_core.c | 1 +
drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c | 1 +
drivers/gpu/drm/bridge/nwl-dsi.c | 1 +
drivers/gpu/drm/bridge/samsung-dsim.c | 1 +
drivers/gpu/drm/msm/dp/dp_aux.c | 1 +
drivers/gpu/drm/rockchip/cdn-dp-core.c | 1 +
6 files changed, 6 insertions(+)
diff --git a/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c b/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c
index 8dee5f2fbde5..4ee08663e626 100644
--- a/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c
+++ b/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c
@@ -18,6 +18,7 @@
#include <linux/of.h>
#include <linux/phy/phy.h>
#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
#include <drm/bridge/analogix_dp.h>
#include <drm/drm_atomic.h>
diff --git a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c
index 064c6915c896..d20c0f8ad04c 100644
--- a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c
+++ b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c
@@ -32,6 +32,7 @@
#include <linux/phy/phy.h>
#include <linux/phy/phy-dp.h>
#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
#include <linux/slab.h>
#include <linux/wait.h>
diff --git a/drivers/gpu/drm/bridge/nwl-dsi.c b/drivers/gpu/drm/bridge/nwl-dsi.c
index 2f7429b24fc2..9ac8796ae91e 100644
--- a/drivers/gpu/drm/bridge/nwl-dsi.c
+++ b/drivers/gpu/drm/bridge/nwl-dsi.c
@@ -18,6 +18,7 @@
#include <linux/of.h>
#include <linux/phy/phy.h>
#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
#include <linux/regmap.h>
#include <linux/reset.h>
#include <linux/sys_soc.h>
diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c
index c3eb437ef1b0..4244434747af 100644
--- a/drivers/gpu/drm/bridge/samsung-dsim.c
+++ b/drivers/gpu/drm/bridge/samsung-dsim.c
@@ -20,6 +20,7 @@
#include <linux/of.h>
#include <linux/phy/phy.h>
#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
#include <linux/units.h>
#include <video/mipi_display.h>
diff --git a/drivers/gpu/drm/msm/dp/dp_aux.c b/drivers/gpu/drm/msm/dp/dp_aux.c
index 3825a2fb48e2..5ee22f88bd28 100644
--- a/drivers/gpu/drm/msm/dp/dp_aux.c
+++ b/drivers/gpu/drm/msm/dp/dp_aux.c
@@ -6,6 +6,7 @@
#include <linux/delay.h>
#include <linux/iopoll.h>
#include <linux/phy/phy.h>
+#include <linux/pm_runtime.h>
#include <drm/drm_print.h>
#include "dp_reg.h"
diff --git a/drivers/gpu/drm/rockchip/cdn-dp-core.c b/drivers/gpu/drm/rockchip/cdn-dp-core.c
index 177e30445ee8..68556daa54ae 100644
--- a/drivers/gpu/drm/rockchip/cdn-dp-core.c
+++ b/drivers/gpu/drm/rockchip/cdn-dp-core.c
@@ -10,6 +10,7 @@
#include <linux/firmware.h>
#include <linux/mfd/syscon.h>
#include <linux/phy/phy.h>
+#include <linux/pm_runtime.h>
#include <linux/regmap.h>
#include <linux/reset.h>
--
2.34.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related
* [PATCH v8 phy-next 04/31] usb: add missing headers transitively included by <linux/phy/phy.h>
From: Vladimir Oltean @ 2026-05-05 10:04 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, Thinh Nguyen, Greg Kroah-Hartman, Peter Chen,
Frank Li, Sascha Hauer, Pengutronix Kernel Team, Fabio Estevam,
Xu Yang
In-Reply-To: <20260505100523.1922388-1-vladimir.oltean@nxp.com>
The chipidea ci_hdrc_imx driver uses regulator consumer API like
regulator_enable() but does not include <linux/regulator/consumer.h>.
The core USB HCD driver calls invalidate_kernel_vmap_range() and
flush_kernel_vmap_range(), but does not include <linux/highmem.h>.
The DWC3 gadget driver calls:
- device_property_present()
- device_property_count_u8()
- device_property_read_u8_array()
but does not include <linux/property.h>
Similarly, dwc3-imx uses device_property_read_bool() without including
<linux/property.h>.
The dwc3-generic-plat driver uses of_device_get_match_data() but does
not include <linux/of.h>.
In all these cases, the necessary includes were still provided somehow,
directly or indirectly, through <linux/phy/phy.h>. I found the following
command to be quite helpful in figuring out the include chain:
$ make KCFLAGS="-H" drivers/usb/dwc3/dwc3-imx.o
Since <linux/phy/phy.h> wants to drop the unnecessary includes, fill in
the required headers to avoid any breakage.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Acked-by: Thinh Nguyen <Thinh.Nguyen@synopsys.com> # dwc3
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
Cc: Peter Chen <peter.chen@kernel.org>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Frank Li <Frank.Li@nxp.com>
Cc: Sascha Hauer <s.hauer@pengutronix.de>
Cc: Pengutronix Kernel Team <kernel@pengutronix.de>
Cc: Fabio Estevam <festevam@gmail.com>
Cc: Thinh Nguyen <Thinh.Nguyen@synopsys.com>
Cc: Xu Yang <xu.yang_2@nxp.com>
v7->v8: none
v6->v7:
- add drivers/usb/dwc3/dwc3-imx.c to the list of patched files
- collect tag from Greg, keeping it despite the new addition because the
change is minor and in the same spirit as the rest
(https://lore.kernel.org/linux-phy/2026033028-squint-yield-4c23@gregkh/)
v2->v6: none
v1->v2: collect tag
---
drivers/usb/chipidea/ci_hdrc_imx.c | 1 +
drivers/usb/core/hcd.c | 1 +
drivers/usb/dwc3/dwc3-generic-plat.c | 1 +
drivers/usb/dwc3/dwc3-imx.c | 1 +
drivers/usb/dwc3/gadget.c | 1 +
5 files changed, 5 insertions(+)
diff --git a/drivers/usb/chipidea/ci_hdrc_imx.c b/drivers/usb/chipidea/ci_hdrc_imx.c
index 56d2ba824a0b..0a21d7cc5f5a 100644
--- a/drivers/usb/chipidea/ci_hdrc_imx.c
+++ b/drivers/usb/chipidea/ci_hdrc_imx.c
@@ -17,6 +17,7 @@
#include <linux/clk.h>
#include <linux/pinctrl/consumer.h>
#include <linux/pm_qos.h>
+#include <linux/regulator/consumer.h>
#include "ci.h"
#include "ci_hdrc_imx.h"
diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
index 89221f1ce769..b3826ebcbe98 100644
--- a/drivers/usb/core/hcd.c
+++ b/drivers/usb/core/hcd.c
@@ -10,6 +10,7 @@
*/
#include <linux/bcd.h>
+#include <linux/highmem.h>
#include <linux/module.h>
#include <linux/version.h>
#include <linux/kernel.h>
diff --git a/drivers/usb/dwc3/dwc3-generic-plat.c b/drivers/usb/dwc3/dwc3-generic-plat.c
index ca69ac0eb07c..2f2ae6f4704f 100644
--- a/drivers/usb/dwc3/dwc3-generic-plat.c
+++ b/drivers/usb/dwc3/dwc3-generic-plat.c
@@ -8,6 +8,7 @@
*/
#include <linux/clk.h>
+#include <linux/of.h>
#include <linux/platform_device.h>
#include <linux/reset.h>
#include <linux/regmap.h>
diff --git a/drivers/usb/dwc3/dwc3-imx.c b/drivers/usb/dwc3/dwc3-imx.c
index 973a486b544d..6e122674edaf 100644
--- a/drivers/usb/dwc3/dwc3-imx.c
+++ b/drivers/usb/dwc3/dwc3-imx.c
@@ -13,6 +13,7 @@
#include <linux/of_platform.h>
#include <linux/platform_device.h>
#include <linux/pm_runtime.h>
+#include <linux/property.h>
#include "core.h"
#include "glue.h"
diff --git a/drivers/usb/dwc3/gadget.c b/drivers/usb/dwc3/gadget.c
index 3d4ca68e584c..b5a6fd2899f1 100644
--- a/drivers/usb/dwc3/gadget.c
+++ b/drivers/usb/dwc3/gadget.c
@@ -10,6 +10,7 @@
#include <linux/kernel.h>
#include <linux/delay.h>
+#include <linux/property.h>
#include <linux/slab.h>
#include <linux/spinlock.h>
#include <linux/platform_device.h>
--
2.34.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related
* [PATCH v8 phy-next 03/31] PCI: Add missing headers transitively included by <linux/phy/phy.h>
From: Vladimir Oltean @ 2026-05-05 10:04 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, Bjorn Helgaas, Lorenzo Pieralisi,
Krzysztof Wilczyński, Manivannan Sadhasivam, Rob Herring,
Heiko Stuebner, Shawn Guo, Yixun Lan, Thierry Reding,
Jonathan Hunter, Shawn Lin, Kevin Xie
In-Reply-To: <20260505100523.1922388-1-vladimir.oltean@nxp.com>
The tegra as well as a few dwc PCI controller drivers uses PM runtime
operations without including the required <linux/pm_runtime.h> header.
Similarly, pcie-rockchip-host, pcie-starfive as well as a few dwc PCI
controllers use the regulator consumer API without including
<linux/regulator/consumer.h>.
pcie-spacemit-k1.c uses of_get_next_available_child() and of_node_put()
without including <linux/of.h>.
It seems these function prototypes were indirectly provided by
<linux/phy/phy.h>, mostly by mistake (none of the functions it exports
need it).
Before the PHY header can drop the unnecessary includes, make sure the
PCI controller drivers include what they use.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
---
Cc: Lorenzo Pieralisi <lpieralisi@kernel.org>
Cc: "Krzysztof Wilczyński" <kwilczynski@kernel.org>
Cc: Manivannan Sadhasivam <mani@kernel.org>
Cc: Rob Herring <robh@kernel.org>
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: Heiko Stuebner <heiko@sntech.de>
Cc: Shawn Guo <shawn.guo@linaro.org>
Cc: Yixun Lan <dlan@kernel.org>
Cc: Thierry Reding <thierry.reding@gmail.com>
Cc: Jonathan Hunter <jonathanh@nvidia.com>
Cc: Shawn Lin <shawn.lin@rock-chips.com>
Cc: Kevin Xie <kevin.xie@starfivetech.com>
v5->v8: none
v4->v5: fix pcie-spacemit-k1 driver, previously missed due to limited
build coverage
v2->v4: none
v1->v2: collect tag, adjust commit title
---
drivers/pci/controller/dwc/pci-keystone.c | 1 +
drivers/pci/controller/dwc/pcie-dw-rockchip.c | 1 +
drivers/pci/controller/dwc/pcie-histb.c | 1 +
drivers/pci/controller/dwc/pcie-qcom-ep.c | 1 +
drivers/pci/controller/dwc/pcie-spacemit-k1.c | 3 +++
drivers/pci/controller/dwc/pcie-tegra194.c | 1 +
drivers/pci/controller/pci-tegra.c | 1 +
drivers/pci/controller/pcie-rockchip-host.c | 1 +
drivers/pci/controller/plda/pcie-starfive.c | 1 +
9 files changed, 11 insertions(+)
diff --git a/drivers/pci/controller/dwc/pci-keystone.c b/drivers/pci/controller/dwc/pci-keystone.c
index 278d2dba1db0..fe4836925c4e 100644
--- a/drivers/pci/controller/dwc/pci-keystone.c
+++ b/drivers/pci/controller/dwc/pci-keystone.c
@@ -24,6 +24,7 @@
#include <linux/of_pci.h>
#include <linux/phy/phy.h>
#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
#include <linux/regmap.h>
#include <linux/resource.h>
#include <linux/signal.h>
diff --git a/drivers/pci/controller/dwc/pcie-dw-rockchip.c b/drivers/pci/controller/dwc/pcie-dw-rockchip.c
index 731d93663cca..ae27ce05247c 100644
--- a/drivers/pci/controller/dwc/pcie-dw-rockchip.c
+++ b/drivers/pci/controller/dwc/pcie-dw-rockchip.c
@@ -21,6 +21,7 @@
#include <linux/phy/phy.h>
#include <linux/platform_device.h>
#include <linux/regmap.h>
+#include <linux/regulator/consumer.h>
#include <linux/reset.h>
#include <linux/workqueue.h>
#include <trace/events/pci_controller.h>
diff --git a/drivers/pci/controller/dwc/pcie-histb.c b/drivers/pci/controller/dwc/pcie-histb.c
index a52071589377..432a54c5bfce 100644
--- a/drivers/pci/controller/dwc/pcie-histb.c
+++ b/drivers/pci/controller/dwc/pcie-histb.c
@@ -18,6 +18,7 @@
#include <linux/pci.h>
#include <linux/phy/phy.h>
#include <linux/platform_device.h>
+#include <linux/regulator/consumer.h>
#include <linux/resource.h>
#include <linux/reset.h>
diff --git a/drivers/pci/controller/dwc/pcie-qcom-ep.c b/drivers/pci/controller/dwc/pcie-qcom-ep.c
index 257c2bcb5f76..35a297923e7f 100644
--- a/drivers/pci/controller/dwc/pcie-qcom-ep.c
+++ b/drivers/pci/controller/dwc/pcie-qcom-ep.c
@@ -19,6 +19,7 @@
#include <linux/phy/phy.h>
#include <linux/platform_device.h>
#include <linux/pm_domain.h>
+#include <linux/pm_runtime.h>
#include <linux/regmap.h>
#include <linux/reset.h>
#include <linux/module.h>
diff --git a/drivers/pci/controller/dwc/pcie-spacemit-k1.c b/drivers/pci/controller/dwc/pcie-spacemit-k1.c
index be20a520255b..41316aa54106 100644
--- a/drivers/pci/controller/dwc/pcie-spacemit-k1.c
+++ b/drivers/pci/controller/dwc/pcie-spacemit-k1.c
@@ -13,9 +13,12 @@
#include <linux/gfp.h>
#include <linux/mfd/syscon.h>
#include <linux/mod_devicetable.h>
+#include <linux/of.h>
#include <linux/phy/phy.h>
#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
#include <linux/regmap.h>
+#include <linux/regulator/consumer.h>
#include <linux/reset.h>
#include <linux/types.h>
diff --git a/drivers/pci/controller/dwc/pcie-tegra194.c b/drivers/pci/controller/dwc/pcie-tegra194.c
index 9dcfa194050e..f1d08814a73c 100644
--- a/drivers/pci/controller/dwc/pcie-tegra194.c
+++ b/drivers/pci/controller/dwc/pcie-tegra194.c
@@ -27,6 +27,7 @@
#include <linux/platform_device.h>
#include <linux/pm_runtime.h>
#include <linux/random.h>
+#include <linux/regulator/consumer.h>
#include <linux/reset.h>
#include <linux/resource.h>
#include <linux/types.h>
diff --git a/drivers/pci/controller/pci-tegra.c b/drivers/pci/controller/pci-tegra.c
index 512309763d1f..a2c1662b6e81 100644
--- a/drivers/pci/controller/pci-tegra.c
+++ b/drivers/pci/controller/pci-tegra.c
@@ -36,6 +36,7 @@
#include <linux/phy/phy.h>
#include <linux/pinctrl/consumer.h>
#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
#include <linux/reset.h>
#include <linux/sizes.h>
#include <linux/slab.h>
diff --git a/drivers/pci/controller/pcie-rockchip-host.c b/drivers/pci/controller/pcie-rockchip-host.c
index ee1822ca01db..46adb4582fcc 100644
--- a/drivers/pci/controller/pcie-rockchip-host.c
+++ b/drivers/pci/controller/pcie-rockchip-host.c
@@ -24,6 +24,7 @@
#include <linux/of_pci.h>
#include <linux/phy/phy.h>
#include <linux/platform_device.h>
+#include <linux/regulator/consumer.h>
#include "../pci.h"
#include "pcie-rockchip.h"
diff --git a/drivers/pci/controller/plda/pcie-starfive.c b/drivers/pci/controller/plda/pcie-starfive.c
index 298036c3e7f9..22344cca167b 100644
--- a/drivers/pci/controller/plda/pcie-starfive.c
+++ b/drivers/pci/controller/plda/pcie-starfive.c
@@ -21,6 +21,7 @@
#include <linux/platform_device.h>
#include <linux/pm_runtime.h>
#include <linux/regmap.h>
+#include <linux/regulator/consumer.h>
#include <linux/reset.h>
#include "../../pci.h"
--
2.34.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related
* [PATCH v8 phy-next 02/31] ata: add <linux/pm_runtime.h> where missing
From: Vladimir Oltean @ 2026-05-05 10:04 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, Damien Le Moal, Niklas Cassel
In-Reply-To: <20260505100523.1922388-1-vladimir.oltean@nxp.com>
It appears that libahci.c, ahci.c as well as the ahci_brcm, ahci_ceva
and ahci_qoriq drivers are using runtime PM operations without including
<linux/pm_runtime.h>. This header is somehow being indirectly provided
by <linux/phy/phy.h>, which would like to drop it (none of the functions
it exports need it).
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Acked-by: Damien Le Moal <dlemoal@kernel.org>
---
Cc: Damien Le Moal <dlemoal@kernel.org>
Cc: Niklas Cassel <cassel@kernel.org>
v2->v8: none
v1->v2: collect tag
---
drivers/ata/ahci.c | 1 +
drivers/ata/ahci_brcm.c | 1 +
drivers/ata/ahci_ceva.c | 1 +
drivers/ata/ahci_qoriq.c | 1 +
drivers/ata/libahci.c | 1 +
5 files changed, 5 insertions(+)
diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 1d73a53370cf..1396a53bd6df 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -26,6 +26,7 @@
#include <linux/device.h>
#include <linux/dmi.h>
#include <linux/gfp.h>
+#include <linux/pm_runtime.h>
#include <scsi/scsi_host.h>
#include <scsi/scsi_cmnd.h>
#include <linux/libata.h>
diff --git a/drivers/ata/ahci_brcm.c b/drivers/ata/ahci_brcm.c
index 29be74fedcf0..48460e515722 100644
--- a/drivers/ata/ahci_brcm.c
+++ b/drivers/ata/ahci_brcm.c
@@ -16,6 +16,7 @@
#include <linux/module.h>
#include <linux/of.h>
#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
#include <linux/reset.h>
#include <linux/string.h>
diff --git a/drivers/ata/ahci_ceva.c b/drivers/ata/ahci_ceva.c
index 2d6a08c23d6a..3938bf378341 100644
--- a/drivers/ata/ahci_ceva.c
+++ b/drivers/ata/ahci_ceva.c
@@ -12,6 +12,7 @@
#include <linux/module.h>
#include <linux/of.h>
#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
#include <linux/reset.h>
#include "ahci.h"
diff --git a/drivers/ata/ahci_qoriq.c b/drivers/ata/ahci_qoriq.c
index 0dec1a17e5b1..409152bfefb6 100644
--- a/drivers/ata/ahci_qoriq.c
+++ b/drivers/ata/ahci_qoriq.c
@@ -14,6 +14,7 @@
#include <linux/device.h>
#include <linux/of.h>
#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
#include <linux/libata.h>
#include "ahci.h"
diff --git a/drivers/ata/libahci.c b/drivers/ata/libahci.c
index c79abdfcd7a9..e0de4703a4f2 100644
--- a/drivers/ata/libahci.c
+++ b/drivers/ata/libahci.c
@@ -30,6 +30,7 @@
#include <scsi/scsi_cmnd.h>
#include <linux/libata.h>
#include <linux/pci.h>
+#include <linux/pm_runtime.h>
#include "ahci.h"
#include "libata.h"
--
2.34.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related
* [PATCH v8 phy-next 01/31] PCI: cadence: Preserve all error codes in cdns_plat_pcie_probe()
From: Vladimir Oltean @ 2026-05-05 10:04 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, Bjorn Helgaas, Krzysztof Wilczyński,
Lorenzo Pieralisi, Manikandan K Pillai, Manivannan Sadhasivam,
Rob Herring
In-Reply-To: <20260505100523.1922388-1-vladimir.oltean@nxp.com>
The blamed commit functionally changed the error path of
cdns_pcie_host_probe(), now cdns_plat_pcie_probe().
When the old code path executed "goto err_get_sync", the PCIe controller
probe function propagated the pm_runtime_get_sync() error code. The new
code doesn't, and returns 0.
Similarly for the "goto err_init" previously triggered by
cdns_pcie_host_init() errors, and now triggered by
cdns_pcie_host_setup() and cdns_pcie_ep_setup() errors. These are not
propagated and will result in probing success, which is incorrect.
Fixes: bd22885aa188 ("PCI: cadence: Refactor driver to use as a core library")
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
---
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: "Krzysztof Wilczyński" <kwilczynski@kernel.org>
Cc: Lorenzo Pieralisi <lpieralisi@kernel.org>
Cc: Manikandan K Pillai <mpillai@cadence.com>
Cc: Manivannan Sadhasivam <mani@kernel.org>
Cc: Rob Herring <robh@kernel.org>
v7->v8: patch is new, issue was flagged by Sashiko
https://sashiko.dev/#/patchset/20260430110652.558622-1-vladimir.oltean@nxp.com
---
drivers/pci/controller/cadence/pcie-cadence-plat.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/pci/controller/cadence/pcie-cadence-plat.c b/drivers/pci/controller/cadence/pcie-cadence-plat.c
index b067a3296dd3..8b12a46b5601 100644
--- a/drivers/pci/controller/cadence/pcie-cadence-plat.c
+++ b/drivers/pci/controller/cadence/pcie-cadence-plat.c
@@ -126,7 +126,7 @@ static int cdns_plat_pcie_probe(struct platform_device *pdev)
while (phy_count--)
device_link_del(cdns_plat_pcie->pcie->link[phy_count]);
- return 0;
+ return ret;
}
static void cdns_plat_pcie_shutdown(struct platform_device *pdev)
--
2.34.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related
* Re: [PATCH v7 phy-next 13/27] usb: host: tegra: avoid direct dereference of phy->dev.of_node
From: Vladimir Oltean @ 2026-05-05 9:39 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, Greg Kroah-Hartman, JC Kuo, Johan Hovold,
Jonathan Hunter, Mathias Nyman, Thierry Reding
In-Reply-To: <20260430110652.558622-14-vladimir.oltean@nxp.com>
Sashiko says:
> commit 037a329bd1e70f26b8d861fb06f3bbc539fb2c49
> Author: Vladimir Oltean <vladimir.oltean@nxp.com>
> usb: host: tegra: avoid direct dereference of phy->dev.of_node
>
> This commit updates the Tegra USB host driver to retrieve the PHY OF node
> using the "phys" phandle rather than directly accessing phy->dev.of_node.
> This addresses an upcoming change where struct phy will become an opaque
> pointer. It also simplifies a debug message to log the array index.
>
> > diff --git a/drivers/usb/host/xhci-tegra.c b/drivers/usb/host/xhci-tegra.c
> > index d2214d309e96..99b0ac9763f9 100644
> > --- a/drivers/usb/host/xhci-tegra.c
> > +++ b/drivers/usb/host/xhci-tegra.c
> > @@ -1491,6 +1491,7 @@ static int tegra_xhci_id_notify(struct notifier_block *nb,
> >
> > static int tegra_xusb_init_usb_phy(struct tegra_xusb *tegra)
> > {
> > + struct device_node *np = dev_of_node(tegra->dev);
> > unsigned int i;
> >
> > tegra->usbphy = devm_kcalloc(tegra->dev, tegra->num_usb_phys,
> > @@ -1504,23 +1505,33 @@ static int tegra_xusb_init_usb_phy(struct tegra_xusb *tegra)
> > tegra->otg_usb3_port = -EINVAL;
> >
> > for (i = 0; i < tegra->num_usb_phys; i++) {
> > - struct phy *phy = tegra_xusb_get_phy(tegra, "usb2", i);
> > + struct of_phandle_args args;
> > + struct usb_phy *usbphy;
> > + int index, err;
> > + char prop[8];
> >
> > - if (!phy)
> > + snprintf(prop, sizeof(prop), "usb2-%d", i);
> > +
> > + /*
> > + * usb-phy is optional, continue if it's not available.
> > + */
> > + index = of_property_match_string(np, "phy-names", prop);
> > + if (index < 0)
> > continue;
> >
> > - tegra->usbphy[i] = devm_usb_get_phy_by_node(tegra->dev,
> > - phy->dev.of_node,
> > - &tegra->id_nb);
> > - if (!IS_ERR(tegra->usbphy[i])) {
> > - dev_dbg(tegra->dev, "usbphy-%d registered", i);
> > - otg_set_host(tegra->usbphy[i]->otg, &tegra->hcd->self);
> > - } else {
> > - /*
> > - * usb-phy is optional, continue if its not available.
> > - */
> > - tegra->usbphy[i] = NULL;
> > - }
> > + err = of_parse_phandle_with_args(np, "phys", "#phy-cells",
> > + index, &args);
> > + if (err)
> > + continue;
>
> The call to of_parse_phandle_with_args() increments the reference count
> of the device tree node and populates it in args.np. The new code passes
> args.np to devm_usb_get_phy_by_node(), but does not appear to call
> of_node_put() to drop the reference when it is done. Does this leak the
> device node reference on both the success and error paths of the loop?
I've added an of_node_put(args.np) after devm_usb_get_phy_by_node() for v8.
> > +
> > + usbphy = devm_usb_get_phy_by_node(tegra->dev, args.np,
> > + &tegra->id_nb);
>
> Also in tegra_xusb_init_usb_phy(), the single notifier block instance
> tegra->id_nb is passed to devm_usb_get_phy_by_node() for every USB PHY
> being processed. Because the struct notifier_block API uses an embedded next
> pointer for its singly-linked list, does registering the exact same object
> to multiple independent PHY notifier chains overwrite its next pointer and
> corrupt the lists?
If this is a problem, it is a pre-existing and completely unrelated
problem which I don't have knowledge how to handle, so I will leave it
alone.
> > + if (IS_ERR(usbphy))
> > + continue;
> > +
> > + tegra->usbphy[i] = usbphy;
> > + dev_dbg(tegra->dev, "usbphy-%d registered", i);
> > + otg_set_host(tegra->usbphy[i]->otg, &tegra->hcd->self);
> > }
> >
> > return 0;
> [...]
> > drivers/usb/host/xhci-tegra.c | 42 ++++++++++++++++++++++-------------
> > 1 file changed, 26 insertions(+), 16 deletions(-)
> >
> > @@ -2168,8 +2179,7 @@ static void tegra_xhci_disable_phy_wake(struct tegra_xusb *tegra)
> > continue;
> >
> > if (tegra_xusb_padctl_remote_wake_detected(padctl, tegra->phys[i]))
> > - dev_dbg(tegra->dev, "%pOF remote wake detected\n",
> > - tegra->phys[i]->dev.of_node);
> > + dev_dbg(tegra->dev, "PHY %d remote wake detected\n", i);
> >
> > tegra_xusb_padctl_disable_phy_wake(padctl, tegra->phys[i]);
> > }
>
> In tegra_xhci_disable_phy_wake(), the loop counter i is declared as an
> unsigned int. The updated dev_dbg() message formats this variable using
> the %d specifier. This isn't a bug since the values are bounded by
> tegra->num_phys, but should it use %u to match the unsigned integer type?
I've changed to %u printing for v8.
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply
* Re: [PATCH v7 phy-next 14/27] usb: gadget: tegra-xudc: avoid direct dereference of phy->dev.of_node
From: Vladimir Oltean @ 2026-05-05 9:34 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, Greg Kroah-Hartman, JC Kuo, Johan Hovold,
Jonathan Hunter, Mathias Nyman, Thierry Reding
In-Reply-To: <20260430110652.558622-15-vladimir.oltean@nxp.com>
Sashiko says:
> commit 1ca88ec2e954b3a331eb5afb8081406d9200c8f6
> Author: Vladimir Oltean <vladimir.oltean@nxp.com>
> usb: gadget: tegra-xudc: avoid direct dereference of phy->dev.of_node
>
> This commit replaces direct dereferences of the device tree node with a
> parallel lookup using the phys phandle and the padctl provider. It also
> introduces a local usbphy variable to hold the output and simplify error
> checks.
>
> > diff --git a/drivers/usb/gadget/udc/tegra-xudc.c b/drivers/usb/gadget/udc/tegra-xudc.c
> > index e9d33be028664..cf4e6c87e44da 100644
> > --- a/drivers/usb/gadget/udc/tegra-xudc.c
> > +++ b/drivers/usb/gadget/udc/tegra-xudc.c
> [ ... ]
> > @@ -3515,7 +3516,10 @@ static int tegra_xudc_phy_get(struct tegra_xudc *xudc)
> > xudc->vbus_nb.notifier_call = tegra_xudc_vbus_notify;
> >
> > for (i = 0; i < xudc->soc->num_phys; i++) {
> > + struct of_phandle_args args;
> > char phy_name[] = "usb.-.";
> > + struct usb_phy *usbphy;
> > + int index, err;
>
> This declares a local err variable inside the loop, shadowing the outer
> err variable from the function scope.
> Later in this loop, if devm_phy_optional_get() or devm_usb_get_phy_by_node()
> fails, the code assigns the error code to this inner err and uses goto
> clean_up.
> Since the inner err is destroyed when leaving the loop block, will the
> function incorrectly return the outer err which is still 0?
This is correct, I've removed the duplicated "err" variable declaration
for v8.
> >
> > /* Get USB2 phy */
> > snprintf(phy_name, sizeof(phy_name), "usb2-%d", i);
> [ ... ]
> > @@ -3525,22 +3529,31 @@ static int tegra_xudc_phy_get(struct tegra_xudc *xudc)
> > dev_err_probe(xudc->dev, err,
> > "failed to get PHY for phy-name usb2-%d\n", i);
> > goto clean_up;
> > - } else if (xudc->utmi_phy[i]) {
> [ ... ]
> > } else if (!xudc->utmi_phy[i]) {
> > /* if utmi phy is not available, ignore USB3 phy get */
> > continue;
> > }
> >
> > + index = of_property_match_string(np, "phy-names", phy_name);
> > + if (index < 0)
> > + continue;
> > +
> > + err = of_parse_phandle_with_args(np, "phys", "#phy-cells",
> > + index, &args);
> > + if (err)
> > + continue;
>
> If either of these device tree parsing steps fail, the code uses continue
> to skip to the next iteration.
> Does this unintentionally skip the entire USB3 PHY initialization block
> that follows later in this loop for the current port?
This is correct, although theoretically of_property_match_string() and
of_parse_phandle_with_args() should not fail as long as devm_phy_optional_get()
succeeded for that very same phy_name starting from the very same OF node.
But nonetheless, if this happens due to bugs, I've instead treated the
condition as fatal and printed error messages for v8.
> > +
> > + /* Get usb-phy, if utmi phy is available */
> > + usbphy = devm_usb_get_phy_by_node(xudc->dev, args.np, NULL);
> > + of_node_put(args.np);
> [ ... ]
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply
* Re: [PATCH v7 phy-next 21/27] net: renesas: rswitch: include PHY provider header
From: Vladimir Oltean @ 2026-05-05 9:24 UTC (permalink / raw)
To: linux-phy
Cc: Vinod Koul, Neil Armstrong, dri-devel, freedreno,
linux-arm-kernel, linux-arm-msm, linux-can, linux-gpio, linux-ide,
linux-kernel, linux-media, linux-pci, linux-renesas-soc,
linux-riscv, linux-rockchip, linux-samsung-soc, linux-scsi,
linux-sunxi, linux-tegra, linux-usb, netdev, spacemit,
UNGLinuxDriver, Geert Uytterhoeven, Yoshihiro Shimoda,
Michael Dege, Andrew Lunn, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, Russell King (Oracle)
In-Reply-To: <20260430110652.558622-22-vladimir.oltean@nxp.com>
Sashiko says:
> commit 4e15b6920cd2b2dfce685835fa09c66664407579
> Author: Vladimir Oltean <vladimir.oltean@nxp.com>
> net: renesas: rswitch: include PHY provider header
>
> This commit includes the PHY provider header in the rswitch driver as a
> workaround to avoid breaking the build when struct phy becomes opaque.
> It allows the driver to continue accessing internal PHY fields while
> acknowledging that the driver's PHY API call sequence needs a complete
> redesign.
>
> > diff --git a/drivers/net/ethernet/renesas/rswitch_main.c b/drivers/net/ethernet/renesas/rswitch_main.c
> > index 6fe964816322a..132be5f15073c 100644
> > --- a/drivers/net/ethernet/renesas/rswitch_main.c
> > +++ b/drivers/net/ethernet/renesas/rswitch_main.c
> > @@ -27,6 +27,7 @@
> > #include <linux/spinlock.h>
> > #include <linux/sys_soc.h>
> >
> > +#include "../../../phy/phy-provider.h" /* FIXME */
>
> Does preserving the ability to check the provider's global reference counts
> (power_count and init_count) via this internal header leave the driver
> vulnerable to data races?
>
> Since these fields are protected by the PHY provider's internal phy->mutex,
> reading them locklessly can create a time-of-check to time-of-use
> vulnerability. The count could be altered by a concurrent operation before
> the mutex is acquired inside phy_power_off() or phy_exit().
>
> Additionally, if the PHY is shared across multiple ports (like in QSGMII
> mode), would checking the global reference count instead of tracking the
> consumer's local vote cause a port to observe a non-zero count from
> another port's vote? This could lead to an incorrect phy_power_off()
> call, prematurely decrementing the shared refcount and powering down the
> PHY while it is still in active use by other ports.
>
> While the commit message notes this is a known issue preserved as a
> workaround until a complete redesign, could retaining these unlocked
> reads cause a regression in power vote tracking in the interim?
>
> > #include "rswitch.h"
> > #include "rswitch_l2.h"
I will ignore this comment. Inaction will leave the driver in exactly
the same state as before. Not great, true, but I don't see how not
changing anything could cause a regression.
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox