mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Heiner Kallweit <hkallweit1@gmail.com>
To: Fabio Baltieri <fabio.baltieri@gmail.com>,
	nic_swsd@realtek.com, Andrew Lunn <andrew+netdev@lunn.ch>,
	"David S. Miller" <davem@davemloft.net>,
	Eric Dumazet <edumazet@google.com>,
	Jakub Kicinski <kuba@kernel.org>,
	Michael Zimmermann <sigmaepsilon92@gmail.com>,
	Paolo Abeni <pabeni@redhat.com>
Cc: netdev@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2] r8169: add support for RTL8127ATF
Date: Thu, 20 Nov 2025 21:55:26 +0100	[thread overview]
Message-ID: <f263daf0-70c2-46c2-af25-653ff3179cb0@gmail.com> (raw)
In-Reply-To: <20251120195055.3127-1-fabio.baltieri@gmail.com>

On 11/20/2025 8:50 PM, Fabio Baltieri wrote:
> Add support for ATF variant of the RTL8127, this uses an SFP transceiver
> instead of the twisted pair one and only runs at 1Gbps or 10Gbps, only
> 10Gbps is supported with this patch.
> 
> The change is based on the r8127 driver package version 11.015.00
> available on the Realtek website and also on
> https://github.com/openwrt/rtl8127.
> 
> There's no public datasheet for the chip so this is just porting over
> the original vendor code to the API and style of the upstream one.
> 
> Signed-off-by: Fabio Baltieri <fabio.baltieri@gmail.com>
> ---
> 
> Hi, did more tests with 1g mode on the v1 of this patch, the setting
> itself works but triggering it from ethtool reliably seems problematic
> due to some weird behavior of the phy code which end up reporting a
> specific link speed when the link is down, making it impossible to set
> it with ethool as it thinks it's already set.
> 
> Since it seems like supporting 1g properly needs expanding the scope of
> the patch, I'm taking the suggestion from Heiner in v1 review and
> stripped this down so it only supports 10g, which is likely what the
> vast majority of the users need anyway.
> 
> Also tested this on suspend/resume, this is also affected by the wol
> issue but with that bit set it now works correctly.
> 
> v1 -> v2
> - stripped out 1g support
> - moved the sds settings in rtl8169_init_phy() so it get called on
>   resume
> - renamed fiber_mode to sfp_mode to avoid confusion
> 
> Cheers,
> Fabio
> 
>  drivers/net/ethernet/realtek/r8169_main.c | 128 +++++++++++++++++++++-
>  1 file changed, 126 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
> index d18734fe12e..e19518c7b98 100644
> --- a/drivers/net/ethernet/realtek/r8169_main.c
> +++ b/drivers/net/ethernet/realtek/r8169_main.c
> @@ -729,6 +729,7 @@ struct rtl8169_private {
>  	unsigned supports_gmii:1;
>  	unsigned aspm_manageable:1;
>  	unsigned dash_enabled:1;
> +	unsigned sfp_mode:1;
>  	dma_addr_t counters_phys_addr;
>  	struct rtl8169_counters *counters;
>  	struct rtl8169_tc_offsets tc_offset;
> @@ -842,7 +843,8 @@ static bool rtl_supports_eee(struct rtl8169_private *tp)
>  {
>  	return tp->mac_version >= RTL_GIGA_MAC_VER_34 &&
>  	       tp->mac_version != RTL_GIGA_MAC_VER_37 &&
> -	       tp->mac_version != RTL_GIGA_MAC_VER_39;
> +	       tp->mac_version != RTL_GIGA_MAC_VER_39 &&
> +	       !tp->sfp_mode;
>  }
>  
>  static void rtl_read_mac_from_reg(struct rtl8169_private *tp, u8 *mac, int reg)
> @@ -1399,6 +1401,95 @@ DECLARE_RTL_COND(rtl_ocp_tx_cond)
>  	return RTL_R8(tp, IBISR0) & 0x20;
>  }
>  
> +#define R8127_SDS_CMD 0x2348
> +#define R8127_SDS_ADDR 0x234a
> +#define R8127_SDS_DATA_IN 0x234c
> +#define R8127_SDS_DATA_OUT 0x234e
> +
> +#define R8127_MAKE_SDS_ADDR(_index, _page, _reg) \
> +	(((_index) << 11) | ((_page) << 5) | (_reg))
> +
> +#define R8127_SDS_CMD_IN BIT(0)
> +#define R8127_SDS_WE_IN BIT(1)
> +
> +DECLARE_RTL_COND(rtl_sds_cmd_done)
> +{
> +	return RTL_R16(tp, R8127_SDS_CMD) & R8127_SDS_CMD_IN;
> +}
> +
> +static u16 rtl8127_sds_phy_read(struct rtl8169_private *tp,
> +				u16 index, u16 page, u16 reg)
> +{
> +	RTL_W16(tp, R8127_SDS_ADDR, R8127_MAKE_SDS_ADDR(index, page, reg));
> +	RTL_W16(tp, R8127_SDS_CMD, R8127_SDS_CMD_IN);
> +
> +	if (rtl_loop_wait_low(tp, &rtl_sds_cmd_done, 1, 100))
> +		return RTL_R16(tp, R8127_SDS_DATA_OUT);
> +	else
> +		return 0xffff;
> +}
> +
> +static void rtl8127_sds_phy_write(struct rtl8169_private *tp,
> +				  u16 index, u16 page, u16 reg, u16 val)
> +{
> +	RTL_W16(tp, R8127_SDS_DATA_IN, val);
> +	RTL_W16(tp, R8127_SDS_ADDR, R8127_MAKE_SDS_ADDR(index, page, reg));
> +	RTL_W16(tp, R8127_SDS_CMD, R8127_SDS_CMD_IN | R8127_SDS_WE_IN);
> +
> +	rtl_loop_wait_low(tp, &rtl_sds_cmd_done, 1, 100);
> +}
> +
> +static void rtl8127_sds_phy_modify(struct rtl8169_private *tp,
> +				   u16 index, u16 page, u16 addr,
> +				   u16 mask, u16 set)
> +{
> +	u16 val;
> +
> +	val = rtl8127_sds_phy_read(tp, index, page, addr);
> +	val = (val & ~mask) | set;
> +	rtl8127_sds_phy_write(tp, index, page, addr, val);
> +}
> +
> +static void rtl8127_sds_phy_reset(struct rtl8169_private *tp)
> +{
> +	RTL_W8(tp, 0x2350, RTL_R8(tp, 0x2350) & ~BIT(0));
> +	udelay(1);
> +
> +	RTL_W16(tp, 0x233a, 0x801f);
> +	RTL_W8(tp, 0x2350, RTL_R8(tp, 0x2350) | BIT(0));
> +	udelay(10);
> +}
> +
> +static void rtl8127_sds_phy_exit_1g(struct rtl8169_private *tp)
> +{
> +	rtl8127_sds_phy_modify(tp, 0, 1, 31, BIT(3), 0);
> +	rtl8127_sds_phy_modify(tp, 0, 2, 0,
> +			       BIT(13) | BIT(12) | BIT(6),
> +			       BIT(6));
> +
> +	rtl8127_sds_phy_reset(tp);
> +}
> +
> +static void rtl8127_set_sds_phy_caps_10g(struct rtl8169_private *tp)
> +{
> +	u16 val;
> +
> +	RTL_W16(tp, 0x233a, 0x801a);
> +
> +	val = RTL_R16(tp, 0x233e);
> +	val &= BIT(13) | BIT(12) | BIT(1) | BIT(0);
> +	val |= BIT(12);
> +	RTL_W16(tp, 0x233e, val);
> +
> +	r8169_mdio_write(tp, 0xc40a, 0x0);
> +	r8169_mdio_write(tp, 0xc466, 0x3);
> +	r8169_mdio_write(tp, 0xc808, 0x0);
> +	r8169_mdio_write(tp, 0xc80a, 0x0);
> +
This function isn't usable on RTL8127.

> +	val = r8168_phy_ocp_read(tp, 0xc804);
> +	r8168_phy_ocp_write(tp, 0xc804, (val & ~0x000f) | 0x000c);
> +}
> +
>  static void rtl8168ep_stop_cmac(struct rtl8169_private *tp)
>  {
>  	RTL_W8(tp, IBCR2, RTL_R8(tp, IBCR2) & ~0x01);
> @@ -1512,6 +1603,15 @@ static enum rtl_dash_type rtl_get_dash_type(struct rtl8169_private *tp)
>  	}
>  }
>  
> +static bool rtl_sfp_mode(struct rtl8169_private *tp)
> +{
> +	if (tp->mac_version == RTL_GIGA_MAC_VER_80 &&
> +	    (r8168_mac_ocp_read(tp, 0xd006) & 0xff) == 0x07)
> +		return true;
> +
> +	return false;
> +}
> +
>  static void rtl_set_d3_pll_down(struct rtl8169_private *tp, bool enable)
>  {
>  	if (tp->mac_version >= RTL_GIGA_MAC_VER_25 &&
> @@ -2390,7 +2490,10 @@ static void rtl8125a_config_eee_mac(struct rtl8169_private *tp)
>  
>  static void rtl8125b_config_eee_mac(struct rtl8169_private *tp)
>  {
> -	r8168_mac_ocp_modify(tp, 0xe040, 0, BIT(1) | BIT(0));
> +	if (tp->sfp_mode)
> +		r8168_mac_ocp_modify(tp, 0xe040, BIT(1) | BIT(0), 0);
> +	else
> +		r8168_mac_ocp_modify(tp, 0xe040, 0, BIT(1) | BIT(0));
>  }
>  
>  static void rtl_rar_exgmac_set(struct rtl8169_private *tp, const u8 *addr)
> @@ -2440,6 +2543,25 @@ static void rtl8169_init_phy(struct rtl8169_private *tp)
>  	    tp->pci_dev->subsystem_device == 0xe000)
>  		phy_write_paged(tp->phydev, 0x0001, 0x10, 0xf01b);
>  
> +	if (tp->sfp_mode) {
> +		rtl8127_sds_phy_exit_1g(tp);
> +		rtl8127_set_sds_phy_caps_10g(tp);
> +
> +		phy_remove_link_mode(tp->phydev, ETHTOOL_LINK_MODE_10baseT_Half_BIT);
> +		phy_remove_link_mode(tp->phydev, ETHTOOL_LINK_MODE_10baseT_Full_BIT);
> +		phy_remove_link_mode(tp->phydev, ETHTOOL_LINK_MODE_100baseT_Half_BIT);
> +		phy_remove_link_mode(tp->phydev, ETHTOOL_LINK_MODE_100baseT_Full_BIT);
> +		phy_remove_link_mode(tp->phydev, ETHTOOL_LINK_MODE_1000baseT_Full_BIT);
> +		phy_remove_link_mode(tp->phydev, ETHTOOL_LINK_MODE_2500baseT_Full_BIT);
> +		phy_remove_link_mode(tp->phydev, ETHTOOL_LINK_MODE_5000baseT_Full_BIT);
> +
> +		phy_remove_link_mode(tp->phydev, ETHTOOL_LINK_MODE_Autoneg_BIT);
> +		phy_remove_link_mode(tp->phydev, ETHTOOL_LINK_MODE_Pause_BIT);
> +		phy_remove_link_mode(tp->phydev, ETHTOOL_LINK_MODE_Asym_Pause_BIT);
> +
> +		tp->phydev->autoneg = 0;
> +	}
> +
>  	/* We may have called phy_speed_down before */
>  	phy_speed_up(tp->phydev);
>  
> @@ -5453,6 +5575,8 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
>  	tp->dash_type = rtl_get_dash_type(tp);
>  	tp->dash_enabled = rtl_dash_is_enabled(tp);
>  
> +	tp->sfp_mode = rtl_sfp_mode(tp);
> +
>  	tp->cp_cmd = RTL_R16(tp, CPlusCmd) & CPCMD_MASK;
>  
>  	if (sizeof(dma_addr_t) > 4 && tp->mac_version >= RTL_GIGA_MAC_VER_18 &&

Your patch is a good starting point, however it needs more thoughts / work how to somewhat cleanly
integrate Realtek's design with phylib. E.g. you would want to set 10G and aneg off via ethtool,
but that's not supported by phy_ethtool_ksettings_set().
I'll prepare patches and, if you don't mind, would provide them to you for testing, as I don't
own this hw.
At least you have a working solution for the time being. Thanks!

  reply	other threads:[~2025-11-20 20:55 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-11-20 19:50 Fabio Baltieri
2025-11-20 20:55 ` Heiner Kallweit [this message]
2025-11-21 20:28   ` Fabio Baltieri
2025-11-23 15:58     ` Heiner Kallweit
2025-11-23 16:26       ` Michael Zimmermann
2025-11-23 16:59         ` Heiner Kallweit
2025-11-23 18:41       ` Fabio Baltieri
2025-11-23 22:54         ` Heiner Kallweit
2025-11-24  0:00           ` Fabio Baltieri
2025-12-03 22:11             ` Fabio Baltieri
2025-12-04  6:32               ` Heiner Kallweit
2025-12-29  4:18           ` FUKAUMI Naoki

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=f263daf0-70c2-46c2-af25-653ff3179cb0@gmail.com \
    --to=hkallweit1@gmail.com \
    --cc=andrew+netdev@lunn.ch \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=fabio.baltieri@gmail.com \
    --cc=kuba@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=nic_swsd@realtek.com \
    --cc=pabeni@redhat.com \
    --cc=sigmaepsilon92@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®