On Tue, Jul 5, 2022 at 2:11 AM Geert Uytterhoeven [off-list ref] wrote:

Hi Saravana,

On Wed, Jun 1, 2022 at 2:44 PM Saravana Kannan [off-list ref] wrote:

quoted

Now that fw_devlink=on by default and fw_devlink supports interrupt
properties, the execution will never get to the point where
driver_deferred_probe_check_state() is called before the supplier has
probed successfully or before deferred probe timeout has expired.

So, delete the call and replace it with -ENODEV.

Signed-off-by: Saravana Kannan <redacted>

Thanks for your patch, which is now commit f8217275b57aa48d ("net:
mdio: Delete usage of driver_deferred_probe_check_state()") in
driver-core/driver-core-next.

Seems like I missed something when providing my T-b for this series,
sorry for that.

No worries. Appreciate any testing help.

arch/arm/boot/dts/r8a7791-koelsch.dts has:

    &ether {
            pinctrl-0 = <&ether_pins>, <&phy1_pins>;
            pinctrl-names = "default";

            phy-handle = <&phy1>;
            renesas,ether-link-active-low;
            status = "okay";

            phy1: ethernet-phy@1 {
                    compatible = "ethernet-phy-id0022.1537",
                                 "ethernet-phy-ieee802.3-c22";
                    reg = <1>;
                    interrupt-parent = <&irqc0>;
                    interrupts = <0 IRQ_TYPE_LEVEL_LOW>;
                    micrel,led-mode = <1>;
                    reset-gpios = <&gpio5 22 GPIO_ACTIVE_LOW>;
            };
    };

Despite the interrupts property, &ether is now probed before irqc0
(interrupt-controller@e61c0000 in arch/arm/boot/dts/r8a7791.dtsi),
causing the PHY not finding its interrupt, and resorting to polling:

I'd still expect the device link to have been created properly for
this phy device. Could you enable the logging in device_link_add() to
check the link is created between the phy and the IRQ?

My guess is that this probably has something to do with phys being
attached to drivers differently.

    -Micrel KSZ8041RNLI ee700000.ethernet-ffffffff:01: attached PHY
driver (mii_bus:phy_addr=ee700000.ethernet-ffffffff:01, irq=185)
    +Micrel KSZ8041RNLI ee700000.ethernet-ffffffff:01: attached PHY
driver (mii_bus:phy_addr=ee700000.ethernet-ffffffff:01, irq=POLL)

Can you drop a WARN() where this is printed to get the stack trace to
check my hypothesis?

-Saravana

Reverting this commit, and commit 9cbffc7a59561be9 ("driver core:
Delete driver_deferred_probe_check_state()") fixes that.

quoted

--- a/drivers/net/mdio/fwnode_mdio.c
+++ b/drivers/net/mdio/fwnode_mdio.c

@@ -47,9 +47,7 @@ int fwnode_mdiobus_phy_device_register(struct mii_bus *mdio,
         * just fall back to poll mode
         */
        if (rc == -EPROBE_DEFER)
-               rc = driver_deferred_probe_check_state(&phy->mdio.dev);
-       if (rc == -EPROBE_DEFER)
-               return rc;
+               rc = -ENODEV;

        if (rc > 0) {
                phy->irq = rc;

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds