On Thu, 21. Nov 14:47, Vitalii Mordan wrote:
> If the clocks priv->iclk and priv->fclk were not enabled in ehci_hcd_sh_probe,
> they should not be disabled in any path.
>
> Conversely, if they was enabled in ehci_hcd_sh_probe, they must be disabled
> in all error paths to ensure proper cleanup.
>
> Found by Linux Verification Center (linuxtesting.org) with Klever.
>
> Fixes: 63c845522263 ("usb: ehci-hcd: Add support for SuperH EHCI.")
> Cc: stable@xxxxxxxxxxxxxxx # ff30bd6a6618: sh: clk: Fix clk_enable() to return 0 on NULL clk
> Signed-off-by: Vitalii Mordan <mordan@xxxxxxxxx>
> ---
The current patch was added to stable kernels lacking the necessary
prerequisite ff30bd6a6618 ("sh: clk: Fix clk_enable() to return 0 on
NULL clk") which is specified in Cc:stable tag of the commit description
per stable kernels documentation [1].
[1]: https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html#option-1
Could you please cherry-pick ff30bd6a6618 ("sh: clk: Fix clk_enable() to
return 0 on NULL clk") to 6.1.y, 5.15.y, 5.10.y and 5.4.y ? It applies
cleanly.
Thanks!
> drivers/usb/host/ehci-sh.c | 9 +++++++--
> 1 file changed, 7 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/usb/host/ehci-sh.c b/drivers/usb/host/ehci-sh.c
> index d31d9506e41a..77460aac6dbd 100644
> --- a/drivers/usb/host/ehci-sh.c
> +++ b/drivers/usb/host/ehci-sh.c
> @@ -119,8 +119,12 @@ static int ehci_hcd_sh_probe(struct platform_device *pdev)
> if (IS_ERR(priv->iclk))
> priv->iclk = NULL;
>
> - clk_enable(priv->fclk);
> - clk_enable(priv->iclk);
> + ret = clk_enable(priv->fclk);
> + if (ret)
> + goto fail_request_resource;
> + ret = clk_enable(priv->iclk);
> + if (ret)
> + goto fail_iclk;
>
> ret = usb_add_hcd(hcd, irq, IRQF_SHARED);
> if (ret != 0) {
> @@ -136,6 +140,7 @@ static int ehci_hcd_sh_probe(struct platform_device *pdev)
>
> fail_add_hcd:
> clk_disable(priv->iclk);
> +fail_iclk:
> clk_disable(priv->fclk);
>
> fail_request_resource:
> --
> 2.25.1
>
