Re: [PATCH v1] scsi: ufs: core: Hold a clock reference across the probe

[Naomi Chu <naomi.chu@mediatek.com>]

On Wed, Sep 09, 2026 at 09:25:36AM +0000, sashiko-bot@kernel.org wrote:
> Pre-existing issues:
> - [Critical] Missing scsi_remove_host() on the ufshcd_init() error path
> leaves dangling sysfs references leading to use-after-free.
[...]
> If ufshcd_init() fails after successfully calling ufshcd_add_scsi_host()
> (for example, if ufshcd_link_startup() or ufshcd_verify_dev_init() fails
> and jumps to out_release), the function returns an error without calling
> scsi_remove_host().

Thanks for the report. As the bot itself notes, this is a pre-existing
issue, and this patch neither introduces nor changes it.

The new out_release label only drops the clock reference taken earlier in
ufshcd_init() and then falls through to out_disable. Every goto that this
patch retargets used to jump to out_disable directly, so the set of
failures that leaves ufshcd_init() without calling scsi_remove_host() is
exactly the same before and after this patch.

Fixing that error path is independent of the clock reference problem
addressed here, so it should be fixed by a completely separate patch
rather than folded into this one.

Thanks,
Naomi