Re: [PATCH v3] rtc: ac100: Fix ac100 determine rate bug

From: Philipp Rossak
Date: Fri Feb 16 2018 - 08:54:03 EST




On 16.02.2018 13:59, Chen-Yu Tsai wrote:
On Fri, Feb 16, 2018 at 8:49 PM, Philipp Rossak <embed3d@xxxxxxxxx> wrote:


On 16.02.2018 05:10, Chen-Yu Tsai wrote:

On Fri, Feb 16, 2018 at 1:53 AM, Philipp Rossak <embed3d@xxxxxxxxx> wrote:



On 15.02.2018 15:11, Maxime Ripard wrote:


On Wed, Feb 14, 2018 at 02:56:12PM +0100, Philipp Rossak wrote:


This patch fixes a bug, that prevents the Allwinner A83T and the A80
from a successful boot.

The bug is there since v4.16-rc1 and appeared after the clk branch was
merged.



Out of curiosity, which patch has introduced this? I couldn't find any
obvious match.


I wasn't also n


To be honest, I'm not sure why this is hitting you and not me.
I have both A83T boards that have assigned-clock-rates set for
the ac100 clock outputs for WiFi. I have them running 4.16-rc1
and have not seen this. The device tree patches that add these
are in 4.15.


Now it is getting curious ... .
I already mentioned that bug in the sunxi-irc and someone else was hitting
that problem also...
I tested it also with the same toolchain you are using (gcc 7.3.0-1 Debian),
but that didn't made any difference.

I don't think that issue is related with the Hardware, but to be on the save
side: Which Hardware version of the BPI-M3 do you have? I have version 1.2.

Can someone else can confirm this bug?

So I might have remembered wrong, as I just realized I have your
patch in my a83t branches. I don't hit this on the A80, which also
has the AC100, but doesn't use assigned-clock-rates in the device
tree.

Could you try rolling back to 4.15 and see if you still hit it?

Clean 4.15, no patches, just cloned before building: No issues.




You can find the shortend trace below:

Unable to handle kernel NULL pointer dereference at virtual address
00000000
pgd = (ptrval)
[00000000] *pgd=00000000
Internal error: Oops: 5 [#1] SMP ARM
Modules linked in:
CPU: 0 PID: 49 Comm: kworker/0:1 Not tainted 4.15.0-10190-gb89e32ccd1be
#2
Hardware name: Allwinner sun8i Family
Workqueue: events deferred_probe_work_func
PC is at clk_hw_get_rate+0x0/0x34
LR is at ac100_clkout_determine_rate+0x48/0x19c

[ ... ]

(clk_hw_get_rate) from (ac100_clkout_determine_rate+0x48/0x19c)
(ac100_clkout_determine_rate) from
(clk_core_set_rate_nolock+0x3c/0x1a0)
(clk_core_set_rate_nolock) from (clk_set_rate+0x30/0x88)
(clk_set_rate) from (of_clk_set_defaults+0x200/0x364)
(of_clk_set_defaults) from (platform_drv_probe+0x18/0xb0)

To fix that bug, we first check if the return of the
clk_hw_get_parent_by_index is non zero. If it is zero we skip that
clock parent.

The BUG report could be found here: https://lkml.org/lkml/2018/2/10/198

Fixes: 04940631b8d2 ("rtc: ac100: Add clk output support")

Signed-off-by: Philipp Rossak <embed3d@xxxxxxxxx>
---

Changes in v3:
* add information when the bug appeared
* make the comment more clear
Changes in v2:
* add tag Fixes: ... to commit message
* add comment to if statement why we are doing this check

drivers/rtc/rtc-ac100.c | 19 ++++++++++++++++++-
1 file changed, 18 insertions(+), 1 deletion(-)

diff --git a/drivers/rtc/rtc-ac100.c b/drivers/rtc/rtc-ac100.c
index 8ff9dc3fe5bf..2412aa2e8399 100644
--- a/drivers/rtc/rtc-ac100.c
+++ b/drivers/rtc/rtc-ac100.c
@@ -183,7 +183,24 @@ static int ac100_clkout_determine_rate(struct
clk_hw
*hw,
for (i = 0; i < num_parents; i++) {
struct clk_hw *parent = clk_hw_get_parent_by_index(hw,
i);
- unsigned long tmp, prate = clk_hw_get_rate(parent);
+ unsigned long tmp, prate;
+
+ /*
+ * The clock has two parents, one is a fixed clock
which
is
+ * internally registered by the ac100 driver. The other
parent
+ * is a clock from the codec side of the chip, which we
+ * properly declare and reference in the devicetree and
is
+ * not implemented in any driver right now.
+ * If the clock core looks for the parent of that
second
+ * missing clock, it can't one that is registered and
+ * returns NULL.
+ * Thus we need to check if the parent exists before
+ * we get the parent rate.
+ */
+ if (!parent)
+ continue;



I'm sorry, but I still don't get it. When you register that clock, you
will give it two parents. Why would that change during the life of the
clock?

This really looks like a workaround rather than an actual fix.

Maxime

I agree this is more a workaround!
A proper solution/fix would be to define the devicetree correct like
this:

diff --git a/arch/arm/boot/dts/sun8i-a83t-bananapi-m3.dts
b/arch/arm/boot/dts/sun8i-a83t-bananapi-m3.dts
index 6550bf0e594b..6f56d429f17e 100644
--- a/arch/arm/boot/dts/sun8i-a83t-bananapi-m3.dts
+++ b/arch/arm/boot/dts/sun8i-a83t-bananapi-m3.dts
@@ -175,11 +175,18 @@
compatible = "x-powers,ac100-rtc";
interrupt-parent = <&r_intc>;
interrupts = <0 IRQ_TYPE_LEVEL_LOW>;
- clocks = <&ac100_codec>;
+ clocks = <&ac100_rtc_32k>;
#clock-cells = <1>;
clock-output-names = "cko1_rtc",
"cko2_rtc",
"cko3_rtc";
+
+ ac100_rtc_32k: rtc-32k-oscillator {
+ compatible = "fixed-clock";
+ #clock-cells = <0>;
+ clock-frequency = <32768>;
+ clock-output-names = "ac100-rtc-32k";
+ };
};
};
};

What do you think about that solution?


That's not quite right either. As I mentioned before, the
RTC block has two clock inputs, one 4MHz signal from the
codec block, and one 32.768 kHz signal from an external
crystal. The original device tree binding describes the
first one, and the 32.768 kHz clock was registered by
the RTC driver internally.

If you're going to add the crystal clock, you still need
to keep the codec one. Note that this does not fix what
Maxime is asking you. I've already provided an explanation:

The clock core allows registering clocks with not-yet-existing
clock parents. Parents are matches by string names. If no
clock by that name is registered yet, the clock core simply
orphans the new clock if the unregistered parent is its
current parent or simply ignores that parent if its not the
current parent. This is entirely valid and is what we are
counting on here, as we haven't implemented the codec-side
driver.

Note we also sort of depend on this behavior for sunxi-ng
clocks, where in some cases we get circular dependencies
between clock blocks.

BTW, since this is mostly related to the clk subsystem,
please CC the clk maintainers as well.


Ok, I added also now the clk mailing list and maintainers.
As reference this here is the boot log [1].

Im not sure if this in relation to this bug:
But, already with the 4.15 kernel I get from time to time some oops during
runtime on the a31s (bpi-m2) and all are clock related and causes a crash of
the system.

Are you referring to the A31 CLK_OUT bug? That is an entirely separate
bug in the A31/A31s CCU driver. The bug was there from the beginning,
but did not cause a crash as the incompatible data structures had lined
up in a way that the embedded data structure that was used by the incorrect
clk ops was actually compatible. This changed with 4.16-rc1, as a new field
was added to some of the clock types, breaking the alignment. I made some
changes but have yet to write the commit log for it. Also I don't have any
A31/A31s boards to test it on right now. I can send you the patch later.

No I'm talking about an other bug. But, right now I'm not sure if this is related to the CLK_OUT bug. My first plan was to wait until you submitted the CLK_OUT patch, to clarify that. But I thought it would be good to mention.

That one I'm talking about is very strange.
So you can run your hardware for hours and nothing happens. Then when it appears the system crashs and reliable booting afterwards is not possible (for a few minutes). But that is not related to the thermal behavior of the soc. I'm still trying to find out under which conditions it crashes...
I also ran some test with cpuburn and cpufreq-ljt-stress-test but everything looks good, related to those tests.


Philipp