Received-SPF: pass (google.com: domain of linux-kernel+bounces-135744-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.48.161 as permitted sender) client-ip=147.75.48.161;
From: =?UTF-8?q?Ilpo=20J=C3=A4rvinen?= <ilpo.jarvinen@linux.intel.com>
Date: Mon, 8 Apr 2024 20:22:57 +0300 (EEST)
To: "David E. Box" <david.e.box@linux.intel.com>
cc: Hans de Goede <hdegoede@redhat.com>, platform-driver-x86@vger.kernel.org, 
    LKML <linux-kernel@vger.kernel.org>, 
    sathyanarayanan.kuppuswamy@linux.intel.com
Subject: Re: [PATCH V3 8/9] platform/x86/intel/sdsi: Simplify ascii
 printing
In-Reply-To: <20240405032507.2637311-9-david.e.box@linux.intel.com>
Message-ID: <6dd8b2c0-1c89-43f0-0426-df8d483a7b1e@linux.intel.com>
References: <20240405032507.2637311-1-david.e.box@linux.intel.com> <20240405032507.2637311-9-david.e.box@linux.intel.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII

On Thu, 4 Apr 2024, David E. Box wrote:

> Use printf width specifier to set the display length of encoded feature
> names.
> 
> Signed-off-by: David E. Box <david.e.box@linux.intel.com>
> Reviewed-by: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@linux.intel.com>
> ---
> 
> V3 - Add FEAT_LEN #def
> 
> V2 - Split of V1 patch 7
> 
>  tools/arch/x86/intel_sdsi/intel_sdsi.c | 13 ++++++-------
>  1 file changed, 6 insertions(+), 7 deletions(-)
> 
> diff --git a/tools/arch/x86/intel_sdsi/intel_sdsi.c b/tools/arch/x86/intel_sdsi/intel_sdsi.c
> index 45bc69e6718e..0c9670ba1f15 100644
> --- a/tools/arch/x86/intel_sdsi/intel_sdsi.c
> +++ b/tools/arch/x86/intel_sdsi/intel_sdsi.c
> @@ -43,6 +43,7 @@
>  #define METER_CERT_MAX_SIZE	4096
>  #define STATE_MAX_NUM_LICENSES	16
>  #define STATE_MAX_NUM_IN_BUNDLE	(uint32_t)8
> +#define FEAT_LEN		4
>  
>  #define __round_mask(x, y) ((__typeof__(x))((y) - 1))
>  #define round_up(x, y) ((((x) - 1) | __round_mask(x, y)) + 1)
> @@ -409,11 +410,10 @@ static int sdsi_meter_cert_show(struct sdsi_dev *s)
>  
>  	printf("Number of Feature Counters:   %ld\n", BUNDLE_COUNT(mc->bundle_length));
>  	while (count < BUNDLE_COUNT(mc->bundle_length)) {
> -		char feature[5];
> +		char feature[FEAT_LEN];
>  
> -		feature[4] = '\0';
>  		get_feature(bec[count].encoding, feature);
> -		printf("    %s:          %d\n", feature, bec[count].counter);
> +		printf("    %.4s:          %d\n", feature, bec[count].counter);
>  		++count;
>  	}
>  
> @@ -494,7 +494,7 @@ static int sdsi_state_cert_show(struct sdsi_dev *s)
>  			sizeof(*lki) +			// size of the license key info
>  			offset;				// offset to this blob content
>  		struct bundle_encoding *bundle = (void *)(lbc) + sizeof(*lbc);
> -		char feature[5];
> +		char feature[FEAT_LEN];
>  		uint32_t i;
>  
>  		printf("     Blob %d:\n", count - 1);
> @@ -507,11 +507,9 @@ static int sdsi_state_cert_show(struct sdsi_dev *s)
>  		printf("        Blob revision ID:           %u\n", lbc->rev_id);
>  		printf("        Number of Features:         %u\n", lbc->num_bundles);
>  
> -		feature[4] = '\0';
> -
>  		for (i = 0; i < min(lbc->num_bundles, STATE_MAX_NUM_IN_BUNDLE); i++) {
>  			get_feature(bundle[i].encoding, feature);
> -			printf("                 Feature %d:         %s\n", i, feature);
> +			printf("                 Feature %d:         %.4s\n", i, feature);
>  		}
>  
>  		if (lbc->num_bundles > STATE_MAX_NUM_IN_BUNDLE)
> 

Hi,

After staring this for a while, I cannot get rid of the feeling that the 
removal of NUL termination is a step into wrong direction. But IMO, 
instead of the caller side, the NUL termination could be added inside 
get_feature().

-- 
 i.