linux-kernel - Re: [PATCH net-next 07/12] net: dsa: rzn1-a5psw: add statistics support

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <YlirO7VrfyUH33rV@lunn.ch>
Date:   Fri, 15 Apr 2022 01:16:11 +0200
From:   Andrew Lunn <andrew@...n.ch>
To:     Clément Léger <clement.leger@...tlin.com>
Cc:     Vivien Didelot <vivien.didelot@...il.com>,
        Florian Fainelli <f.fainelli@...il.com>,
        Vladimir Oltean <olteanv@...il.com>,
        "David S . Miller" <davem@...emloft.net>,
        Jakub Kicinski <kuba@...nel.org>,
        Paolo Abeni <pabeni@...hat.com>,
        Rob Herring <robh+dt@...nel.org>,
        Krzysztof Kozlowski <krzk+dt@...nel.org>,
        Geert Uytterhoeven <geert+renesas@...der.be>,
        Magnus Damm <magnus.damm@...il.com>,
        Heiner Kallweit <hkallweit1@...il.com>,
        Russell King <linux@...linux.org.uk>,
        Thomas Petazzoni <thomas.petazzoni@...tlin.com>,
        Herve Codina <herve.codina@...tlin.com>,
        Miquèl Raynal <miquel.raynal@...tlin.com>,
        Milan Stevanovic <milan.stevanovic@...com>,
        Jimmy Lalande <jimmy.lalande@...com>,
        linux-kernel@...r.kernel.org, devicetree@...r.kernel.org,
        linux-renesas-soc@...r.kernel.org, netdev@...r.kernel.org
Subject: Re: [PATCH net-next 07/12] net: dsa: rzn1-a5psw: add statistics
 support

On Thu, Apr 14, 2022 at 02:22:45PM +0200, Clément Léger wrote:
> Add per-port statistics. This support requries to add a stat lock since
> statistics are stored in two 32 bits registers, the hi part one being
> global and latched when accessing the lo part.
> 
> Signed-off-by: Clément Léger <clement.leger@...tlin.com>
> ---
>  drivers/net/dsa/rzn1_a5psw.c | 101 +++++++++++++++++++++++++++++++++++
>  drivers/net/dsa/rzn1_a5psw.h |   2 +
>  2 files changed, 103 insertions(+)
> 
> diff --git a/drivers/net/dsa/rzn1_a5psw.c b/drivers/net/dsa/rzn1_a5psw.c
> index 5bee999f7050..7ab7d9054427 100644
> --- a/drivers/net/dsa/rzn1_a5psw.c
> +++ b/drivers/net/dsa/rzn1_a5psw.c
> @@ -16,6 +16,59 @@
>  
>  #include "rzn1_a5psw.h"
>  
> +struct a5psw_stats {
> +	u16 offset;
> +	const char *name;
> +};
> +
> +#define STAT_DESC(_offset, _name) {.offset = _offset, .name = _name}
> +
> +static const struct a5psw_stats a5psw_stats[] = {
> +	STAT_DESC(0x868, "aFrameTransmitted"),
> +	STAT_DESC(0x86C, "aFrameReceived"),
> +	STAT_DESC(0x8BC, "etherStatsetherStatsOversizePktsDropEvents"),

> +};


> +static void a5psw_get_strings(struct dsa_switch *ds, int port, u32 stringset,
> +			      uint8_t *data)
> +{
> +	unsigned int u;
> +
> +	if (stringset != ETH_SS_STATS)
> +		return;
> +
> +	for (u = 0; u < ARRAY_SIZE(a5psw_stats); u++) {
> +		strncpy(data + u * ETH_GSTRING_LEN, a5psw_stats[u].name,
> +			ETH_GSTRING_LEN);
> +	}

The kernel strncpy() is like the user space one. It does not add a
NULL if the string is longer than ETH_GSTRING_LEN and it needs to
truncate. So there is a danger here.

What you find most drivers do is

struct a5psw_stats {
	u16 offset;
	const char name[ETH_GSTRING_LEN];
};

You should then get a compiler warning/error if you string is ever
longer than allowed. And use memcpy() rather than strcpy(), which is
faster anyway. But you do use up a bit more memory.

> +static void a5psw_get_ethtool_stats(struct dsa_switch *ds, int port,
> +				    uint64_t *data)
> +{
> +	struct a5psw *a5psw = ds->priv;
> +	u32 reg_lo, reg_hi;
> +	unsigned int u;
> +
> +	for (u = 0; u < ARRAY_SIZE(a5psw_stats); u++) {
> +		/* A5PSW_STATS_HIWORD is global and thus, access must be
> +		 * exclusive
> +		 */

Could you explain that a bit more. The RTNL lock will prevent two
parallel calls to this function.

> +		spin_lock(&a5psw->stats_lock);
> +		reg_lo = a5psw_reg_readl(a5psw, a5psw_stats[u].offset +
> +					 A5PSW_PORT_OFFSET(port));
> +		/* A5PSW_STATS_HIWORD is latched on stat read */
> +		reg_hi = a5psw_reg_readl(a5psw, A5PSW_STATS_HIWORD);
> +
> +		data[u] = ((u64)reg_hi << 32) | reg_lo;
> +		spin_unlock(&a5psw->stats_lock);
> +	}
> +}

  Andrew