linux-kernel - Re: [PATCH net-next v3 4/4] net: phy: dp83td510: Add support for the DP83TD510 Ethernet PHY

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: <20201031091825.uucn2ax2cjzzuy2e@skbuf>
Date:   Sat, 31 Oct 2020 11:18:25 +0200
From:   Ioana Ciornei <ciorneiioana@...il.com>
To:     Dan Murphy <dmurphy@...com>
Cc:     davem@...emloft.net, andrew@...n.ch, f.fainelli@...il.com,
        hkallweit1@...il.com, robh@...nel.org, devicetree@...r.kernel.org,
        netdev@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH net-next v3 4/4] net: phy: dp83td510: Add support for the
 DP83TD510 Ethernet PHY

On Fri, Oct 30, 2020 at 12:29:50PM -0500, Dan Murphy wrote:
> The DP83TD510E is an ultra-low power Ethernet physical layer transceiver
> that supports 10M single pair cable.
> 
> The device supports both 2.4-V p2p and 1-V p2p output voltage as defined
> by IEEE 802.3cg 10Base-T1L specfications. These modes can be forced via
> the device tree or the device is defaulted to auto negotiation to
> determine the proper p2p voltage.
> 
> Signed-off-by: Dan Murphy <dmurphy@...com>
> ---
>  drivers/net/phy/Kconfig     |   6 +
>  drivers/net/phy/Makefile    |   1 +
>  drivers/net/phy/dp83td510.c | 681 ++++++++++++++++++++++++++++++++++++
>  3 files changed, 688 insertions(+)
>  create mode 100644 drivers/net/phy/dp83td510.c
>

(...)

> +static int dp83td510_ack_interrupt(struct phy_device *phydev)
> +{
> +	int ret;
> +
> +	ret = phy_read(phydev, DP83TD510_INT_REG1);
> +	if (ret < 0)
> +		return ret;
> +
> +	ret = phy_read(phydev, DP83TD510_INT_REG2);
> +	if (ret < 0)
> +		return ret;
> +
> +	return 0;
> +}
> +
> +static int dp83td510_config_intr(struct phy_device *phydev)
> +{
> +	int int_status;
> +	int gen_cfg_val;
> +	int ret;
> +
> +	if (phydev->interrupts == PHY_INTERRUPT_ENABLED) {
> +		int_status = phy_read(phydev, DP83TD510_INT_REG1);
> +		if (int_status < 0)
> +			return int_status;
> +
> +		int_status = (DP83TD510_INT1_ESD_EN | DP83TD510_INT1_LINK_EN |
> +			      DP83TD510_INT1_RHF_EN);
> +
> +		ret = phy_write(phydev, DP83TD510_INT_REG1, int_status);
> +		if (ret)
> +			return ret;
> +
> +		int_status = phy_read(phydev, DP83TD510_INT_REG2);
> +		if (int_status < 0)
> +			return int_status;
> +
> +		int_status = (DP83TD510_INT2_POR | DP83TD510_INT2_POL |
> +				DP83TD510_INT2_PAGE);
> +
> +		ret = phy_write(phydev, DP83TD510_INT_REG2, int_status);
> +		if (ret)
> +			return ret;
> +
> +		gen_cfg_val = phy_read(phydev, DP83TD510_GEN_CFG);
> +		if (gen_cfg_val < 0)
> +			return gen_cfg_val;
> +
> +		gen_cfg_val |= DP83TD510_INT_OE | DP83TD510_INT_EN;
> +
> +	} else {
> +		ret = phy_write(phydev, DP83TD510_INT_REG1, 0);
> +		if (ret)
> +			return ret;
> +
> +		ret = phy_write(phydev, DP83TD510_INT_REG2, 0);
> +		if (ret)
> +			return ret;
> +
> +		gen_cfg_val = phy_read(phydev, DP83TD510_GEN_CFG);
> +		if (gen_cfg_val < 0)
> +			return gen_cfg_val;
> +
> +		gen_cfg_val &= ~DP83TD510_INT_EN;
> +	}
> +
> +	return phy_write(phydev, DP83TD510_GEN_CFG, gen_cfg_val);
> +}
> +

I am not really sure if the shared-IRQ work in the below linked patch
set will go through, but I think it would be cleaner just to ack any
pending interrupts after you disable them.

https://lore.kernel.org/netdev/20201029100741.462818-1-ciorneiioana@gmail.com/

I see that you are reading the INT_REG1 and INT_REG2 registers
(basically servicing any pending interrupts) before enabling the IRQ.
The same reads should be done after the IRQ has been disabled.

> +static struct phy_driver dp83td510_driver[] = {
> +	{
> +		PHY_ID_MATCH_MODEL(DP83TD510E_PHY_ID),
> +		.name		= "TI DP83TD510E",
> +		.probe          = dp83td510_probe,
> +		.config_init	= dp83td510_config_init,
> +		.soft_reset	= dp83td510_phy_reset,
> +
> +		/* IRQ related */
> +		.ack_interrupt	= dp83td510_ack_interrupt,
> +		.config_intr	= dp83td510_config_intr,

I think the PHY maintainers could comment on this more, but maybe it
would help if the driver implements the .handle_interrupt() callback
just so that I wouldn't have to touch a driver that was just added to
rework it for the shared-IRQ transition.

Ioana