Re: [PATCH -v2 3/3] ACPI, APEI, report GHES error information via printk

[Andrew Morton <akpm@linux-foundation.org>]

On Tue, 30 Nov 2010 10:51:41 +0800 Huang Ying <ying.huang@intel.com> wrote:

> printk is one of the methods to report hardware errors to user space.
> This patch implements hardware error reporting for GHES via printk.
> 
> Signed-off-by: Huang Ying <ying.huang@intel.com>
> ---
>  drivers/acpi/apei/ghes.c |   21 +++++++++++++++++----
>  1 file changed, 17 insertions(+), 4 deletions(-)
> 
> --- a/drivers/acpi/apei/ghes.c
> +++ b/drivers/acpi/apei/ghes.c
> @@ -255,11 +255,23 @@ static void ghes_do_proc(struct ghes *gh
>  		}
>  #endif
>  	}
> +}
>  
> -	if (!processed && printk_ratelimit())
> -		pr_warning(GHES_PFX
> -		"Unknown error record from generic hardware error source: %d\n",
> -			   ghes->generic->header.source_id);
> +static void ghes_print_estatus(const char *pfx, struct ghes *ghes)
> +{
> +	if (pfx == NULL) {
> +		if (ghes_severity(ghes->estatus->error_severity) <=
> +		    GHES_SEV_CORRECTED)
> +			pfx = KERN_WARNING HW_ERR;
> +		else
> +			pfx = KERN_ERR HW_ERR;
> +	}
> +	if (printk_ratelimit()) {
> +		printk(
> +	"%s""Hardware error from APEI Generic Hardware Error Source: %d\n",
> +	pfx, ghes->generic->header.source_id);
> +		apei_estatus_print(pfx, ghes->estatus);

That code layout is just ghastly.  Please, if it can't be done nicely
in 80-cols then simply exceed the 80 cols.

And please don't use (or retain) printk_ratelimit().  It was a mistake.
A printk_ratelimt() site shares state with all other
printk_ratelimit() states, so if a random firewire driver is doing a lot of
printk_ratelimit() calls, your messages get suppressed!  Use
printk_ratelimited() or __ratelimit().

>
> ...
>