On Thu, Sep 17, 2026 at 11:33:55PM +0100, Mark Brown wrote:
> This doesn't apply against current code, please check and resend.
Sorry for the very late reply, and for the noise.
You are right, and it was not a conflict with another fix: the patch was
generated against a tree predating de27e0cadcce ("ASoC: fsl-asoc-card:
Move bound-component setup to late_probe"), which removed cpu_pdev and
the codec_dev[] array from fsl_asoc_card_probe(). The exit path the
patch was built on no longer exists.
The leak itself does: asrc_pdev, taken by of_find_device_by_node(), is
still never released. I rebased the fix on current code and resent it
on Nov 4:
https://lore.kernel.org/all/[email protected]/
It adds put_device(&asrc_pdev->dev) at asrc_fail, applies cleanly to
310ee4c282a1 (current sound for-next), and builds with arm64 defconfig.
Thanks,
Wentao
