Re: [PATCH 13/19] EDAC, mc: Determine mci pointer from the error descriptor

[Mauro Carvalho Chehab <mchehab+samsung@kernel.org>]

Em Thu, 10 Oct 2019 20:25:31 +0000
Robert Richter <rrichter@marvell.com> escreveu:

> Each struct mci has its own error descriptor. Create a function
> error_desc_to_mci() to determine the corresponding mci from an error
> descriptor. This eases the parameter list of edac_raw_mc_handle_
> error() as the mci pointer do not need to be passed any longer.
> 
> While at it, reorder parameters of edac_raw_mc_handle_error().
> 
> Signed-off-by: Robert Richter <rrichter@marvell.com>

Reviewed-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>

> ---
>  drivers/edac/edac_mc.c   | 13 +++++++++----
>  drivers/edac/edac_mc.h   |  8 +++-----
>  drivers/edac/ghes_edac.c |  2 +-
>  3 files changed, 13 insertions(+), 10 deletions(-)
> 
> diff --git a/drivers/edac/edac_mc.c b/drivers/edac/edac_mc.c
> index ca206854b8ee..9e8c5716a8c0 100644
> --- a/drivers/edac/edac_mc.c
> +++ b/drivers/edac/edac_mc.c
> @@ -1040,10 +1040,15 @@ static void edac_ue_error(struct mem_ctl_info *mci,
>  	edac_inc_ue_error(mci, dimm, error_count);
>  }
>  
> -void edac_raw_mc_handle_error(struct mem_ctl_info *mci,
> -			      struct dimm_info *dimm,
> -			      struct edac_raw_error_desc *e)
> +static struct mem_ctl_info *error_desc_to_mci(struct edac_raw_error_desc *e)
> +{
> +	return container_of(e, struct mem_ctl_info, error_desc);
> +}
> +
> +void edac_raw_mc_handle_error(struct edac_raw_error_desc *e,
> +			      struct dimm_info *dimm)
>  {
> +	struct mem_ctl_info *mci = error_desc_to_mci(e);
>  	char detail[80];
>  
>  	/* Memory type dependent details about the error */
> @@ -1243,6 +1248,6 @@ void edac_mc_handle_error(const enum hw_event_mc_err_type type,
>  
>  	dimm = edac_get_dimm(mci, top_layer, mid_layer, low_layer);
>  
> -	edac_raw_mc_handle_error(mci, dimm, e);
> +	edac_raw_mc_handle_error(e, dimm);
>  }
>  EXPORT_SYMBOL_GPL(edac_mc_handle_error);
> diff --git a/drivers/edac/edac_mc.h b/drivers/edac/edac_mc.h
> index a8f1b5b5e873..3b01d5d9d7bc 100644
> --- a/drivers/edac/edac_mc.h
> +++ b/drivers/edac/edac_mc.h
> @@ -212,17 +212,15 @@ extern int edac_mc_find_csrow_by_page(struct mem_ctl_info *mci,
>   * edac_raw_mc_handle_error() - Reports a memory event to userspace without
>   *	doing anything to discover the error location.
>   *
> - * @mci:		a struct mem_ctl_info pointer
> - * @dimm:		a struct dimm_info pointer
>   * @e:			error description
> + * @dimm:		a struct dimm_info pointer
>   *
>   * This raw function is used internally by edac_mc_handle_error(). It should
>   * only be called directly when the hardware error come directly from BIOS,
>   * like in the case of APEI GHES driver.
>   */
> -void edac_raw_mc_handle_error(struct mem_ctl_info *mci,
> -			      struct dimm_info *dimm,
> -			      struct edac_raw_error_desc *e);
> +void edac_raw_mc_handle_error(struct edac_raw_error_desc *e,
> +			      struct dimm_info *dimm);
>  
>  /**
>   * edac_mc_handle_error() - Reports a memory event to userspace.
> diff --git a/drivers/edac/ghes_edac.c b/drivers/edac/ghes_edac.c
> index 1db1c012bed9..8078d4ec9631 100644
> --- a/drivers/edac/ghes_edac.c
> +++ b/drivers/edac/ghes_edac.c
> @@ -439,7 +439,7 @@ void ghes_edac_report_mem_error(int sev, struct cper_sec_mem_err *mem_err)
>  
>  	dimm = edac_get_dimm_by_index(mci, e->top_layer);
>  
> -	edac_raw_mc_handle_error(mci, dimm, e);
> +	edac_raw_mc_handle_error(e, dimm);
>  
>  	spin_unlock_irqrestore(&ghes_lock, flags);
>  }



Thanks,
Mauro