Re: [PATCH v6 14/15] ASoC: msm8916-wcd-analog: add pm8953 codec

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:

New issues:
- [High] Missing NULL check for the result of `of_device_get_match_data()` before dereferencing it.
- [Low] Initialization sequence improperly writes to a status register (`CDC_A_RX_EAR_STATUS`).

Pre-existing issues:
- [High] Interrupt handlers dereference a NULL `component` pointer if triggered before the ALSA SoC component finishes initialization.
--

--- Patch [14]: [PATCH v6 14/15] ASoC: msm8916-wcd-analog: add pm8953 codec ---
commit 20f92545eb7f1109d22b0cad48d91169236a8d8d
Author: Vladimir Lypak <vladimir.lypak@gmail.com>
ASoC: msm8916-wcd-analog: add pm8953 codec

This commit adds register overrides for the PM8953 codec to the msm8916
wcd analog driver.

(Note: The standard inline patch quotation format has been altered to a
free-form summary in this report to comply with strict text recitation
policies. The issues below are presented without quoting the diff.)

[Severity: Low]
In the newly added wcd_reg_defaults_pm8953 array, there is a write of 0x10
to CDC_A_RX_EAR_STATUS. Is this intended? Since this is typically a read-only
status register, could this be a typo for a control register like
CDC_A_RX_EAR_CTL, which might lead to missing configuration?

[Severity: High]
In pm8916_wcd_analog_probe(), the return value of of_device_get_match_data()
is assigned to wcd_reg_init_data and immediately dereferenced in the loop
condition without validation. If the driver is instantiated without a matching
device tree node (for example, via a sysfs bind by name), could this lead to
a NULL pointer dereference and a kernel panic?

[Severity: High]
This is a pre-existing issue, but in the interrupt handlers
mbhc_btn_press_irq_handler() and pm8916_mbhc_switch_irq_handler(), there
is a read of priv->component followed by a call to snd_soc_component_read()
without checking if the component is fully initialized. Since the IRQs are
requested in the platform driver probe before the ALSA component probe
assigns the pointer, could a pending interrupt from the bootloader fire
immediately and cause a NULL pointer dereference?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260730174353.108023-1-mailingradian@gmail.com?part=14