Regression in probing some AMBA devices possibly devlink related

[Linus Walleij]

On Sun, Feb 26, 2023 at 10:55 PM Linus Walleij <linus.walleij at linaro.org> wrote:
> On Sun, Feb 26, 2023 at 12:56 AM Saravana Kannan <saravanak at google.com> wrote:
> > On Sat, Feb 25, 2023 at 6:29 AM Linus Walleij <linus.walleij at linaro.org> wrote:
> > >
> > > Hi Saravana,
> > >
> > > I have a boot regression for Ux500 on mainline, but bisecting mainline
> > > isn't quite working for misc reasons :/
> > >
> > > I'm not sure about this regression, but it smells like devlink-related.
> > >
> > > Ux500 have a bunch of normal and some AMBA devices. After
> > > boot this happens and we hang waiting for rootfs (which is on eMMC):
> >
> > Hmmm... my recent changes were definitely tested on systems with AMBA
> > devices and it worked.
>
> Some of them work, such as the TWD, interrupt controller GIC
> and the CoreSight stuff.
>
> > > [   31.849586] amba 80126000.mmc: deferred probe pending
> > > [   31.854801] amba 80118000.mmc: deferred probe pending
> > > [   31.859895] amba 80005000.mmc: deferred probe pending
> > > [   31.870297] amba 80120000.uart: deferred probe pending
> > > [   31.875472] amba 80121000.uart: deferred probe pending
> > > [   31.880689] amba 80004000.i2c: deferred probe pending
> > > [   31.885799] amba 80128000.i2c: deferred probe pending
> > > [   31.890932] amba 80110000.i2c: deferred probe pending
> > > [   51.688361] vmem_3v3: disabling
> >
> > What does /debug/devices_deferred say about these? That should tell us
> > exactly what's blocking it.
>
> Can't get to check that sadly, because the root fs is on eMMC and
> that one is deferring indefinitely.
>
> I tried to compile in an initramfs but that became too big for this
> target. I'll check if I can use that on another target which I
> can boot directly from RAM.
>
> > Also if you convert all the pr_debug and dev_dbg in
> > drivers/base/core.c that should give us enough of an idea of what's
> > happening. Can you do that too and send it as an attachment (I logs
> > are hard to read when they get word wrapped)?
>
> This I could do!
>
> See attachment.
>
> It seems these devices are waiting for the PRCC reset-controller
> which is in
> drivers/clk/ux500/reset-prcc.c
>
> Very well! I went into
> arch/arm/boot/dts/ste-dbx5x0.dtsi
> and commented out all "resets = < &prcc....>; statements,
> and indeed then the system comes up fine! (The resets are
> not vital to boot the system.)
>
> But this is weird because the driver for the resets is spawn from
> the same device node that is spawning the clocks...
> All resources (clocks and resets) are spawn from here:
> drivers/clk/ux500/u8500_of_clk.c
>
> From these nodes in the DTS
> arch/arm/boot/dts/ste-dbx5x0.dtsi:
>
>                 clocks {
>                         compatible = "stericsson,u8500-clks";
>                         /*
>                          * Registers for the CLKRST block on peripheral
>                          * groups 1, 2, 3, 5, 6,
>                          */
>                         reg = <0x8012f000 0x1000>, <0x8011f000 0x1000>,
>                             <0x8000f000 0x1000>, <0xa03ff000 0x1000>,
>                             <0xa03cf000 0x1000>;
>
>                         prcmu_clk: prcmu-clock {
>                                 #clock-cells = <1>;
>                         };
>
>                         prcc_pclk: prcc-periph-clock {
>                                 #clock-cells = <2>;
>                         };
>
>                         prcc_kclk: prcc-kernel-clock {
>                                 #clock-cells = <2>;
>                         };
>
>                         prcc_reset: prcc-reset-controller {
>                                 #reset-cells = <2>;
>                         };
>
>                         rtc_clk: rtc32k-clock {
>                                 #clock-cells = <0>;
>                         };
>
>                         smp_twd_clk: smp-twd-clock {
>                                 #clock-cells = <0>;
>                         };
>
>                         clkout_clk: clkout-clock {
>                                 /* Cell 1 id, cell 2 source, cell 3 div */
>                                 #clock-cells = <3>;
>                         };
>                 };
>
> So it seems like the device core things that the
> resets are never coming online. But they are! I put some
> debug prints into the PRCC reset driver to make sure.
>
> diff --git a/drivers/clk/ux500/reset-prcc.c b/drivers/clk/ux500/reset-prcc.c
> index f7e48941fbc7..b1ed9a48cdfb 100644
> --- a/drivers/clk/ux500/reset-prcc.c
> +++ b/drivers/clk/ux500/reset-prcc.c
> @@ -178,4 +178,6 @@ void u8500_prcc_reset_init(struct device_node *np,
> struct u8500_prcc_reset *ur)
>         ret = reset_controller_register(rcdev);
>         if (ret)
>                 pr_err("PRCC failed to register reset controller\n");
> +
> +       pr_info("PRCC reset controller registered\n");
>  }
>
> [    0.000000] PRCC reset controller registered
>
> It is registered as early as the clocks.
>
> So it seems something is fishy with the reset resource resolution?
>
> > > The last regulator (vmem_3v3) is something the eMMC that didn't
> > > probe was supposed to use.
> > >
> > > All the failing drivers are AMBA PrimeCell devices:
> > > drivers/mmc/host/mmci.c
> > > drivers/tty/serial/amba-pl011.c
> > > drivers/i2c/busses/i2c-nomadik.c
> > >
> > > This makes me suspect something was done for ordinary (platform)
> > > devices that didn't happen for AMBA devices?
> > >
> > > This is the main portion of the device tree containing these
> > > devices and their resources:
> > > arch/arm/boot/dts/ste-dbx5x0.dtsi
> >
> > I'll take a closer look on Monday. Btw, I always prefer the dts file
> > because there's always some property that adds a dependency and that
> > might be the clue to whatever is broken. But I'll take a look at this.
>
> The top level DTS for this system I'm testing on (source for the
> log) is
> arch/arm/boot/dts/ste-ux500-samsung-golden.dts
>
> > It's unlikely the patch attached might fix it, but can you give it a shot?
>
> Sadly does not help!

Did this mail go through with the attachment?

I have pastebin:ed it here as well:
https://pastebin.com/Fwc7E1Sm

Yours,
Linus Walleij