Re: [PATCH 3/3] x86: iosf: Make debugfs addition a separate config option

[Ingo Molnar <mingo@kernel.org>]

* David E. Box <david.e.box@linux.intel.com> wrote:

> Signed-off-by: David E. Box <david.e.box@linux.intel.com>
> ---
>  arch/x86/Kconfig           | 6 ++++++
>  arch/x86/kernel/iosf_mbi.c | 8 ++++++--
>  2 files changed, 12 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
> index 2db4b1d..3afcac3 100644
> --- a/arch/x86/Kconfig
> +++ b/arch/x86/Kconfig
> @@ -494,6 +494,12 @@ config IOSF_MBI
>  	  You should say Y if you are running a kernel on one of these
>  	  platforms.
>  
> +config IOSF_MBI_DEBUG
> +	bool "Enable debugfs access to IOSF"
> +	depends on IOSF_MBI && DEBUG_FS
> +	---help---
> +	  Select this option to enable iosf sideband debugging through debugfs.

Please also mention where the debug info is.

> -/********************** debugfs begin ****************************/
> +#if (IS_ENABLED(CONFIG_IOSF_MBI_DEBUG))
>  static u32	dbg_mdr;
>  static u32	dbg_mcr;
>  static u32	dbg_mcrx;
> @@ -257,7 +257,7 @@ static void iosf_sideband_debug_init(void)
>  cleanup:
>  	debugfs_remove_recursive(d);
>  }
> -/********************** debugfs end ****************************/
> +#endif /* CONFIG_IOSF_MBI_DEBUG */
>  
>  static int iosf_mbi_probe(struct pci_dev *pdev,
>  			  const struct pci_device_id *unused)
> @@ -290,13 +290,17 @@ static struct pci_driver iosf_mbi_pci_driver = {
>  
>  static int __init iosf_mbi_init(void)
>  {
> +#if IS_ENABLED(CONFIG_IOSF_MBI_DEBUG)
>  	iosf_sideband_debug_init();
> +#endif
>  	return pci_register_driver(&iosf_mbi_pci_driver);
>  }
>  
>  static void __exit iosf_mbi_exit(void)
>  {
> +#if IS_ENABLED(CONFIG_IOSF_MBI_DEBUG)
>  	debugfs_remove_recursive(iosf_dbg);
> +#endif

The last two #ifdefs are pretty ugly, please make dummy !DEBUG 
versions of those methods, so that the regular code is not 
uglified with debugging details.

Also, because COFIG_IOSF_MBI_DEBUG is a bool, using:

	#ifdef CONFIG_IOSF_MBI_DEBUG

should be enough in general.

Thanks,

	Ingo