On Wed, Jul 29, 2026 at 05:05:15PM +0300, Dmitry Baryshkov wrote:
> On Tue, Jul 28, 2026 at 11:26:14AM +0000, [email protected] wrote:
> > Thank you for your contribution! Sashiko AI review found 12 potential
> > issue(s) to consider:
> > - [Critical] Recursive locking deadlock in `lt9611c_hdmi_audio_prepare`.
Will fix it in v9
> > - [High] The IRQ is enabled before the hardware reset is completed during
> > resume.
Will fix it in v9 by changing the order.
> > - [High] Premature manual teardown in remove() conflicts with devm_* APIs,
> > leading to UAF and timeouts.
Will fix it in v9.
> > - [High] The IRQ handler incorrectly clears unhandled interrupts by writing
> > an inverted bitmask.
This is ok, as right now only hpd interrupt is there in firmware.
> > - [High] Early registration of the DRM bridge exposes it before resource
> > initialization is complete, risking use-after-free on probe failure.
Will fix it in v9.
> > - [Medium] The return value of devm_drm_bridge_add() is ignored.
Will fix it in v9.
> > - [Medium] Selecting REGMAP_I2C without a dependency on I2C causes a
> > Kconfig warning and build failure.
Will fix it in v9.
> > - [Medium] request_firmware() is called while holding the ocm_lock mutex.
Will fix it in v9.
> > - [Medium] Missing `regmap_read()` error handling causes uninitialized
> > memory usage and broken firmware upgrade logic.
Will fix it in v9.
> > - [Medium] Incomplete error recovery in suspend callback leaves the device
> > in a permanently broken state.
Will fix it in v9.
> > - [Medium] Synchronous firmware request during probe may permanently fail
> > if the root filesystem is not yet mounted.
This is ok as fw_version == 0 only occurs on factory-blank chips and
synchronous firmware is better.
> > - [Low] Firmware version is not updated in the driver state after a
> > successful sysfs upgrade.
Will fix it in v9.
> > --
>
> Please respond to Sashiko findings.
>
>
> --
> With best wishes
> Dmitry
