Re: [PATCH v7 2/3] PCI: cadence: Add HPA IP debugfs for LTSSM status

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Low] Building PCIE_CADENCE_HOST as built-in and PCIE_CADENCE_DEBUGFS as a module results in a linkage error.
--

commit 20cc8aa0f1ba6b094082684b8473490c278a0462
Author: Hans Zhang <18255117159@163.com>

PCI: cadence: Add HPA IP debugfs for LTSSM status

This commit adds debugfs support for HPA-based Cadence PCIe controllers.
It creates a new ltssm_status file under debugfs to allow users to read
the current LTSSM state as a string and raw value.

> diff --git a/drivers/pci/controller/cadence/Kconfig b/drivers/pci/controller/cadence/Kconfig
> index 9e651d5459735..cb010bc97aad3 100644
> --- a/drivers/pci/controller/cadence/Kconfig
> +++ b/drivers/pci/controller/cadence/Kconfig
> @@ -6,6 +6,15 @@ menu "Cadence-based PCIe controllers"
>  config PCIE_CADENCE
>  	tristate
>  
> +config PCIE_CADENCE_DEBUGFS
> +	tristate "Cadence PCIe debugfs entries"
> +	depends on DEBUG_FS
> +	depends on PCIE_CADENCE_HOST || PCIE_CADENCE_EP

[ ... ]

> diff --git a/drivers/pci/controller/cadence/pcie-cadence.h b/drivers/pci/controller/cadence/pcie-cadence.h
> index 9a464cbaf073d..8de2924057659 100644
> --- a/drivers/pci/controller/cadence/pcie-cadence.h
> +++ b/drivers/pci/controller/cadence/pcie-cadence.h

[ ... ]

> @@ -535,4 +676,16 @@ bool cdns_pcie_hpa_link_up(struct cdns_pcie *pcie);
>  
>  extern const struct dev_pm_ops cdns_pcie_pm_ops;
>  
> +#if IS_ENABLED(CONFIG_PCIE_CADENCE_DEBUGFS)
> +void cdns_pcie_debugfs_deinit(struct cdns_pcie *pci);
> +void cdns_pcie_debugfs_init(struct cdns_pcie *pci);
> +#else
> +static inline void cdns_pcie_debugfs_deinit(struct cdns_pcie *pci)
> +{
> +}
> +static inline void cdns_pcie_debugfs_init(struct cdns_pcie *pci)
> +{
> +}
> +#endif

[Severity: Low]
Does this cause a linkage error if CONFIG_PCIE_CADENCE_HOST is built-in (=y)
and CONFIG_PCIE_CADENCE_DEBUGFS is built as a module (=m)?

When configured this way, IS_ENABLED(CONFIG_PCIE_CADENCE_DEBUGFS) evaluates
to 1, so the preprocessor does not provide the empty inline stubs. Then,
when the built-in host code calls cdns_pcie_debugfs_init(), it results in
an undefined reference at vmlinux link time because the debugfs code is in
a separate module.

Would using IS_REACHABLE() instead of IS_ENABLED(), or restricting the
tristate dependencies in Kconfig, prevent this build failure?

> diff --git a/drivers/pci/controller/cadence/pcie-cadence-host-hpa.c b/drivers/pci/controller/cadence/pcie-cadence-host-hpa.c
> index 0f540bed58e88..89f1d75c2c38d 100644
> --- a/drivers/pci/controller/cadence/pcie-cadence-host-hpa.c
> +++ b/drivers/pci/controller/cadence/pcie-cadence-host-hpa.c

[ ... ]

> @@ -360,7 +373,13 @@ int cdns_pcie_hpa_host_setup(struct cdns_pcie_rc *rc)
>  	if (!bridge->ops)
>  		bridge->ops = &cdns_pcie_hpa_host_ops;
>  
> -	return pci_host_probe(bridge);
> +	ret = pci_host_probe(bridge);
> +	if (ret)
> +		return ret;
> +
> +	cdns_pcie_debugfs_init(pcie);
> +
> +	return 0;
>  }

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260611165933.20071-1-18255117159@163.com?part=2