[1/1] rtc: pcf85063: add support for fixed clock

TQ-Systems' TQMa8Mx module (SoM) uses a pcf85063 as RTC. The default output
is 32768Hz. This is to provide the i.MX8M CKIL clock. Once the RTC driver
is probed, the clock is disabled and all i.MX8M functionality depending on
the 32 KHz clock will halt. In our case the whole system halts and a power
cycle is required.

Referencing the pcf85063 directly results in a deadlock. The kernel
will see, that i.MX8M system clock needs the RTC clock and do probe
deferral. But the i.MX8M I2C module never becomes usable without the
i.MX8M CKIL clock and thus the RTC's clock will not be probed. So
from the kernel's perspective this is a chicken-and-egg problem.

Technically everything is fine by not touching anything, since
the RTC clock correctly enables the clock on reset (i.e. on
battery backup power loss).

A workaround for this issue is describing the square wave pin
as fixed-clock, which is registered early and basically how
this pin is used on the i.MX8M.

This addresses the exact same issue as in commit f765e349c3e1 ("rtc:
m41t80: add support for fixed clock").

Signed-off-by: Alexander Stein <alexander.stein@ew.tq-group.com>
---
As mentioned in the commit message this addesses the same problem as on
Congatec's board, although we use a different RTC.
I also shamelessly reused most of the commit message.

 .../devicetree/bindings/rtc/nxp,pcf85063.txt         |  9 +++++++++
 drivers/rtc/rtc-pcf85063.c                           | 12 ++++++++++++
 2 files changed, 21 insertions(+)

Hi Alexander,

Thank you for the patch! Perhaps something to improve:

[auto build test WARNING on abelloni/rtc-next]
[also build test WARNING on v5.15-rc5 next-20211013]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch]

url:    https://github.com/0day-ci/linux/commits/Alexander-Stein/rtc-pcf85063-add-support-for-fixed-clock/20211013-155115
base:   https://git.kernel.org/pub/scm/linux/kernel/git/abelloni/linux.git rtc-next
config: x86_64-randconfig-s022-20211013 (attached as .config)
compiler: gcc-9 (Debian 9.3.0-22) 9.3.0
reproduce:
        # apt-get install sparse
        # sparse version: v0.6.4-dirty
        # https://github.com/0day-ci/linux/commit/9f6ceaac6d87d4d5a2a35d209d3e630ff2bdeb4b
        git remote add linux-review https://github.com/0day-ci/linux
        git fetch --no-tags linux-review Alexander-Stein/rtc-pcf85063-add-support-for-fixed-clock/20211013-155115
        git checkout 9f6ceaac6d87d4d5a2a35d209d3e630ff2bdeb4b
        # save the attached .config to linux build tree
        make W=1 C=1 CF='-fdiagnostic-prefix -D__CHECK_ENDIAN__' O=build_dir ARCH=x86_64 SHELL=/bin/bash

If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>


sparse warnings: (new ones prefixed by >>)
>> drivers/rtc/rtc-pcf85063.c:492:24: sparse: sparse: Using plain integer as NULL pointer

vim +492 drivers/rtc/rtc-pcf85063.c

   476	
   477	static struct clk *pcf85063_clkout_register_clk(struct pcf85063 *pcf85063)
   478	{
   479		struct clk *clk;
   480		struct clk_init_data init;
   481		struct device_node *node = pcf85063->rtc->dev.parent->of_node;
   482		struct device_node *fixed_clock;
   483	
   484		fixed_clock = of_get_child_by_name(node, "clock");
   485		if (fixed_clock) {
   486			/*
   487			 * skip registering square wave clock when a fixed
   488			 * clock has been registered. The fixed clock is
   489			 * registered automatically when being referenced.
   490			 */
   491			of_node_put(fixed_clock);
 > 492			return 0;
   493		}
   494	
   495		init.name = "pcf85063-clkout";
   496		init.ops = &pcf85063_clkout_ops;
   497		init.flags = 0;
   498		init.parent_names = NULL;
   499		init.num_parents = 0;
   500		pcf85063->clkout_hw.init = &init;
   501	
   502		/* optional override of the clockname */
   503		of_property_read_string(node, "clock-output-names", &init.name);
   504	
   505		/* register the clock */
   506		clk = devm_clk_register(&pcf85063->rtc->dev, &pcf85063->clkout_hw);
   507	
   508		if (!IS_ERR(clk))
   509			of_clk_add_provider(node, of_clk_src_simple_get, clk);
   510	
   511		return clk;
   512	}
   513	#endif
   514	

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org

On Wed, 13 Oct 2021 09:49:54 +0200, Alexander Stein wrote:
> TQ-Systems' TQMa8Mx module (SoM) uses a pcf85063 as RTC. The default output
> is 32768Hz. This is to provide the i.MX8M CKIL clock. Once the RTC driver
> is probed, the clock is disabled and all i.MX8M functionality depending on
> the 32 KHz clock will halt. In our case the whole system halts and a power
> cycle is required.
> 
> Referencing the pcf85063 directly results in a deadlock. The kernel
> will see, that i.MX8M system clock needs the RTC clock and do probe
> deferral. But the i.MX8M I2C module never becomes usable without the
> i.MX8M CKIL clock and thus the RTC's clock will not be probed. So
> from the kernel's perspective this is a chicken-and-egg problem.
> 
> [...]

Applied, thanks!

[1/1] rtc: pcf85063: add support for fixed clock
      commit: 4c8a7b80d5f3c924fbe08b24634fb67a97f96465

Best regards,