Re: [PATCH] rtc: stm32: Fix unbalanced clk_disable_unprepare() on probe error path

From: Nobuhiro Iwamatsu
Date: Mon Jun 07 2021 - 21:25:05 EST


Hi,

2021年6月8日(火) 2:38 Martin Fuzzey <martin.fuzzey@flowbird.group>:
>
> The STM32MP1 RTC may have 2 clocks, the pclk and the rtc_ck.
>
> If clk_prepare_enable() fails for the second clock (rtc_ck) we must only
> call clk_disable_unprepare() for the first clock (pclk) but currently we
> call it on both leading to a WARN:
>
> [ 15.629568] WARNING: CPU: 0 PID: 146 at drivers/clk/clk.c:958 clk_core_disable+0xb0/0xc8
> [ 15.637620] ck_rtc already disabled
> [ 15.663322] CPU: 0 PID: 146 Comm: systemd-udevd Not tainted 5.4.77-pknbsp-svn5759-atag-v5.4.77-204-gea4235203137-dirty #2413
> [ 15.674510] Hardware name: STM32 (Device Tree Support)
> [ 15.679658] [<c0111148>] (unwind_backtrace) from [<c010c0b8>] (show_stack+0x10/0x14)
> [ 15.687371] [<c010c0b8>] (show_stack) from [<c0ab3d28>] (dump_stack+0xc0/0xe0)
> [ 15.694574] [<c0ab3d28>] (dump_stack) from [<c012360c>] (__warn+0xc8/0xf0)
> [ 15.701428] [<c012360c>] (__warn) from [<c0123694>] (warn_slowpath_fmt+0x60/0x94)
> [ 15.708894] [<c0123694>] (warn_slowpath_fmt) from [<c053b518>] (clk_core_disable+0xb0/0xc8)
> [ 15.717230] [<c053b518>] (clk_core_disable) from [<c053c190>] (clk_core_disable_lock+0x18/0x24)
> [ 15.725924] [<c053c190>] (clk_core_disable_lock) from [<bf0adc44>] (stm32_rtc_probe+0x124/0x5e4 [rtc_stm32])
> [ 15.735739] [<bf0adc44>] (stm32_rtc_probe [rtc_stm32]) from [<c05f7d4c>] (platform_drv_probe+0x48/0x98)
> [ 15.745095] [<c05f7d4c>] (platform_drv_probe) from [<c05f5cec>] (really_probe+0x1f0/0x458)
> [ 15.753338] [<c05f5cec>] (really_probe) from [<c05f61c4>] (driver_probe_device+0x70/0x1c4)
> [ 15.761584] [<c05f61c4>] (driver_probe_device) from [<c05f6580>] (device_driver_attach+0x58/0x60)
> [ 15.770439] [<c05f6580>] (device_driver_attach) from [<c05f6654>] (__driver_attach+0xcc/0x170)
> [ 15.779032] [<c05f6654>] (__driver_attach) from [<c05f40d8>] (bus_for_each_dev+0x58/0x7c)
> [ 15.787191] [<c05f40d8>] (bus_for_each_dev) from [<c05f4ffc>] (bus_add_driver+0xdc/0x1f8)
> [ 15.795352] [<c05f4ffc>] (bus_add_driver) from [<c05f6ed8>] (driver_register+0x7c/0x110)
> [ 15.803425] [<c05f6ed8>] (driver_register) from [<c01027bc>] (do_one_initcall+0x70/0x1b8)
> [ 15.811588] [<c01027bc>] (do_one_initcall) from [<c01a1094>] (do_init_module+0x58/0x1f8)
> [ 15.819660] [<c01a1094>] (do_init_module) from [<c01a0074>] (load_module+0x1e58/0x23c8)
> [ 15.827646] [<c01a0074>] (load_module) from [<c01a0860>] (sys_finit_module+0xa0/0xd4)
> [ 15.835459] [<c01a0860>] (sys_finit_module) from [<c01011e0>] (__sys_trace_return+0x0/0x20)
>
> Signed-off-by: Martin Fuzzey <martin.fuzzey@flowbird.group>
> Fixes: 4e64350f42e2 ("rtc: add STM32 RTC driver")
> Cc: stable@xxxxxxxxxxxxxxx
> ---
> drivers/rtc/rtc-stm32.c | 6 ++++--
> 1 file changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/rtc/rtc-stm32.c b/drivers/rtc/rtc-stm32.c
> index 75a8924..ac9e228 100644
> --- a/drivers/rtc/rtc-stm32.c
> +++ b/drivers/rtc/rtc-stm32.c
> @@ -754,7 +754,7 @@ static int stm32_rtc_probe(struct platform_device *pdev)
>
> ret = clk_prepare_enable(rtc->rtc_ck);
> if (ret)
> - goto err;
> + goto err_no_rtc_ck;
>
> if (rtc->data->need_dbp)
> regmap_update_bits(rtc->dbp, rtc->dbp_reg,
> @@ -830,10 +830,12 @@ static int stm32_rtc_probe(struct platform_device *pdev)
> }
>
> return 0;
> +
> err:
> + clk_disable_unprepare(rtc->rtc_ck);
> +err_no_rtc_ck:
> if (rtc->data->has_pclk)
> clk_disable_unprepare(rtc->pclk);
> - clk_disable_unprepare(rtc->rtc_ck);
>
> if (rtc->data->need_dbp)
> regmap_update_bits(rtc->dbp, rtc->dbp_reg, rtc->dbp_mask, 0);
> --
> 1.9.1
>
Looks good to me.

Reviewed-by: Nobuhiro Iwamatsu <iwamatsu@xxxxxxxxxxx>

Best regards,
Nobuhiro
--
Nobuhiro Iwamatsu
iwamatsu at {nigauri.org / debian.org / kernel.org}
GPG ID: 40AD1FA6