net: stmmac: phylink PCS conversion
Subsystems (6)
v2 cover net: stmmac: phylink PCS conversion
This series is radical - it takes the brave step of ripping out much of the existing PCS support code and throwing it all away.
I have discussed the introduction of the STMMAC_FLAG_HAS_INTEGRATED_PCS flag with Bartosz Golaszewski, and the conclusion I came to is that this is to workaround the breakage that I've been going on about concerning the phylink conversion for the last five or six years.
The problem is that the stmmac PCS code manipulates the netif carrier state, which confuses phylink.
There is a way of testing this out on the Jetson Xavier NX platform as the "PCS" code paths can be exercised while in RGMII mode - because RGMII also has in-band status and the status register is shared with SGMII. Testing this out confirms my long held theory: the interrupt handler manipulates the netif carrier state before phylink gets a look-in, which means that the mac_link_up() and mac_link_down() methods are never called, resulting in the device being non-functional.
Moreover, on dwmac4 cores, ethtool reports incorrect information - despite having a full-duplex link, ethtool reports that it is half-dupex.
Thus, this code is completely broken - anyone using it will not have a functional platform, and thus it doesn't deserve to live any longer, especially as it's a thorn in phylink.
Rip all this out, leaving just the bare bones initialisation in place.
However, this is not the last of what's broken. We have this hw->ps integer which is really not descriptive, and the DT property from which it comes from does little to help understand what's going on. Putting all the clues together:
- early configuration of the GMAC configuration register for the speed.
- setting the SGMII rate adapter layer to take its speed from the GMAC configuration register.
Lastly, setting the transmit enable (TE) bit, which is a typo that puts the nail in the coffin of this code. It should be the transmit configuration (TC) bit. Given that when the link comes up, phylink will call mac_link_up() which will overwrite the speed in the GMAC configuration register, the only part of this that is functional is changing where the SGMII rate adapter layer gets its speed from, which is a boolean.
From what I've found so far, everyone who sets the snps,ps-speed property which configures this mode also configures a fixed link, so the pre-configuration is unnecessary - the link will come up anyway.
So, this series rips that out the preconfiguration as well, and replaces hw->ps with a boolean hw->reverse_sgmii_enable flag.
We then move the sole PCS configuration into a phylink_pcs instance, which configures the PCS control register in the same way as is done during the probe function.
Thus, we end up with much easier and simpler conversion to phylink PCS than previous attempts.
Even so, this still results in inband mode always being enabled at the moment in the new .pcs_config() method to reflect what the probe function was doing. The next stage will be to change that to allow phylink to correctly configure the PCS. This needs fixing to allow platform glue maintainers who are currently blocked to progress.
Please note, however, that this has not been tested with any SGMII platform.
I've tried to get as many people into the Cc list with get_maintainers, I hope that's sufficient to get enough eyeballs on this.
Changes since RFC:
- new patch (7) to remove RGMII "pcs" mode
- new patch (8) to move reverse "pcs" mode to stmmac_check_pcs_mode()
- new patch (9) to simplify the code moved in the previous patch
- new patch (10) to rename the confusing hw->ps to something more understandable.
- new patch (11) to shut up inappropriate complaints about "snps,ps-speed" being invalid.
- new patch (13) to add a MAC .pcs_init method, which will only be called when core has PCS present.
- modify patch 14 to use this new pcs_init method.
Despite getting a couple of responses to the RFC series posted in September, I have had nothing testing this on hardware. I have tested this on the Jetson Xavier NX, which included trial runs with enabling the RGMII "pcs" mode, hence the new patches that rip out this mode. I have come to the conclusion that the only way to get stmmac changes tested is to get them merged into net-next, thereby forcing people to have to run with them... and we'll deal with any fallout later.
Changes since v1:
- added Andrew's reviewed-bys
- added additional comments to patch 3, 11 and 14. No code changes.
drivers/net/ethernet/stmicro/stmmac/Makefile | 2 +- drivers/net/ethernet/stmicro/stmmac/common.h | 5 +- .../ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c | 6 +- drivers/net/ethernet/stmicro/stmmac/dwmac1000.h | 6 +- .../net/ethernet/stmicro/stmmac/dwmac1000_core.c | 65 ++------------------- drivers/net/ethernet/stmicro/stmmac/dwmac4.h | 3 +- drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c | 66 ++------------------- .../net/ethernet/stmicro/stmmac/dwxgmac2_core.c | 25 +------- drivers/net/ethernet/stmicro/stmmac/hwif.h | 4 +- drivers/net/ethernet/stmicro/stmmac/stmmac.h | 4 ++ .../net/ethernet/stmicro/stmmac/stmmac_ethtool.c | 68 +--------------------- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 24 ++++---- drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.c | 47 +++++++++++++++ drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h | 23 ++++++-- include/linux/stmmac.h | 1 - 15 files changed, 104 insertions(+), 245 deletions(-)
Signature · 7 lines
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/ FTTP is here! 80Mbps down 10Mbps up. Decent connectivity at last! -- RMK's Patch system: https://www.armlinux.org.uk/developer/patches/ FTTP is here! 80Mbps down 10Mbps up. Decent connectivity at last!
PATCH v2 1/14 net: stmmac: remove broken PCS code
Changing the netif_carrier_*() state behind phylink's back has always been prohibited because it messes up with phylinks state tracking, and means that phylink no longer guarantees to call the mac_link_down() and mac_link_up() methods at the appropriate times. This was later documented in the sfp-phylink network driver conversion guide.
stmmac was converted to phylink in 2019, but nothing was done with the "PCS" code. Since then, apart from the updates as part of phylink development, nothing has happened with stmmac to improve its use of phylink, or even to address this point.
A couple of years ago, a has_integrated_pcs boolean was added by Bart, which later became the STMMAC_FLAG_HAS_INTEGRATED_PCS flag, to avoid manipulating the netif_carrier_*() state. This flag is mis-named, because whenever the stmmac is synthesized for its native SGMII, TBI or RTBI interfaces, it has an "integrated PCS". This boolean/flag actually means "ignore the status from the integrated PCS".
Discussing with Bart, the reasons for this are lost to the winds of time (which is why we should always document the reasons in the commit message.)
RGMII also has in-band status, and the dwmac cores and stmmac code supports this but with one bug that saves the day.
When dwmac cores are synthesised for RGMII only, they do not contain an integrated PCS, and so priv->dma_cap.pcs is clear, which prevents (incorrectly) the "RGMII PCS" being used, meaning we don't read the in-band status. However, a core synthesised for RGMII and also SGMII, TBI or RTBI will have this capability bit set, thus making these code paths reachable.
The Jetson Xavier NX uses RGMII mode to talk to its PHY, and removing the incorrect check for priv->dma_cap.pcs reveals the theortical issue with netif_carrier_*() manipulation is real:
dwc-eth-dwmac 2490000.ethernet eth0: Register MEM_TYPE_PAGE_POOL RxQ-0 dwc-eth-dwmac 2490000.ethernet eth0: PHY [stmmac-0:00] driver [RTL8211F Gigabit Ethernet] (irq=141) dwc-eth-dwmac 2490000.ethernet eth0: No Safety Features support found
dwc-eth-dwmac 2490000.ethernet eth0: IEEE 1588-2008 Advanced Timestamp supported dwc-eth-dwmac 2490000.ethernet eth0: registered PTP clock
dwc-eth-dwmac 2490000.ethernet eth0: configuring for phy/rgmii-id link mode
8021q: adding VLAN 0 to HW filter on device eth0
dwc-eth-dwmac 2490000.ethernet eth0: Adding VLAN ID 0 is not supported
Link is Up - 1000/Full
Link is Down
Link is Up - 1000/Full
This looks good until one realises that the phylink "Link" status messages are missing, even when the RJ45 cable is reconnected. Nothing one can do results in the interface working. The interrupt handler (which prints those "Link is" messages) always wins over phylink's resolve worker, meaning phylink never calls the mac_link_up() nor mac_link_down() methods.
eth0 also sees no traffic received, and is unable to obtain a DHCP address:
3: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group defa ult qlen 1000
link/ether e6:d3:6a:e6:92:de brd ff:ff:ff:ff:ff:ff
RX: bytes packets errors dropped overrun mcast
0 0 0 0 0 0
TX: bytes packets errors dropped carrier collsns
27686 149 0 0 0 0
With the STMMAC_FLAG_HAS_INTEGRATED_PCS flag set, which disables the netif_carrier_*() manipulation then stmmac works normally:
dwc-eth-dwmac 2490000.ethernet eth0: Register MEM_TYPE_PAGE_POOL RxQ-0 dwc-eth-dwmac 2490000.ethernet eth0: PHY [stmmac-0:00] driver [RTL8211F Gigabit Ethernet] (irq=141) dwc-eth-dwmac 2490000.ethernet eth0: No Safety Features support found
dwc-eth-dwmac 2490000.ethernet eth0: IEEE 1588-2008 Advanced Timestamp supported dwc-eth-dwmac 2490000.ethernet eth0: registered PTP clock
dwc-eth-dwmac 2490000.ethernet eth0: configuring for phy/rgmii-id link mode
8021q: adding VLAN 0 to HW filter on device eth0
dwc-eth-dwmac 2490000.ethernet eth0: Adding VLAN ID 0 is not supported
Link is Up - 1000/Full
dwc-eth-dwmac 2490000.ethernet eth0: Link is Up - 1Gbps/Full - flow control rx/tx
and packets can be transferred.
This clearly shows that when priv->hw->pcs is set, but STMMAC_FLAG_HAS_INTEGRATED_PCS is clear, the driver reliably fails.
Discovering whether a platform falls into this is impossible as parsing all the dtsi and dts files to find out which use the stmmac driver, whether any of them use RGMII or SGMII and also depends whether an external interface is being used. The kernel likely doesn't contain all dts files either.
The only driver that sets this flag uses the qcom,sa8775p-ethqos compatible, and uses SGMII or 2500BASE-X.
but these are saved from this problem by the incorrect check for priv->dma_cap.pcs.
So, we have to assume that for every other platform that uses SGMII with stmmac is using an external PCS.
Moreover, ethtool output can be incorrect. With the full-duplex link negotiated, ethtool reports:
Speed: 1000Mb/s
Duplex: Halfbecause with dwmac4, the full-duplex bit is in bit 16 of the status, priv->xstats.pcs_duplex becomes BIT(16) for full duplex, but the ethtool ksettings duplex member is u8 - so becomes zero. Moreover, the supported, advertised and link partner modes are all "not reported".
Finally, ksettings_set() won't be able to set the advertisement on a PHY if this PCS code is activated, which is incorrect when SGMII is used with a PHY.
Thus, remove:
1. the incorrect netif_carrier_*() manipulation.
2. the broken ethtool ksettings code.
Given that all uses of STMMAC_FLAG_HAS_INTEGRATED_PCS are now gone, remove the flag from stmmac.h and dwmac-qcom-ethqos.c.
--- .../stmicro/stmmac/dwmac-qcom-ethqos.c | 4 -- .../ethernet/stmicro/stmmac/stmmac_ethtool.c | 55 ------------------- .../net/ethernet/stmicro/stmmac/stmmac_main.c | 9 --- include/linux/stmmac.h | 1 - 4 files changed, 69 deletions(-)
drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c +0 -4
@@ -96,7 +96,6 @@ struct ethqos_emac_driver_data {bool rgmii_config_loopback_en;bool has_emac_ge_3;const char *link_clk_name; -bool has_integrated_pcs;u32 dma_addr_width;struct dwmac4_addrs dwmac4_addrs;bool needs_sgmii_loopback;
@@ -282,7 +281,6 @@ static const struct ethqos_emac_driver_data emac_v4_0_0_data = { .rgmii_config_loopback_en = false, .has_emac_ge_3 = true, .link_clk_name = "phyaux", - .has_integrated_pcs = true, .needs_sgmii_loopback = true, .dma_addr_width = 36, .dwmac4_addrs = {
@@ -856,8 +854,6 @@ static int qcom_ethqos_probe(struct platform_device *pdev) plat_dat->flags |=STMMAC_FLAG_TSO_EN ;if (of_device_is_compatible (np,"qcom,qcs404-ethqos" )) plat_dat->flags |=STMMAC_FLAG_RX_CLK_RUNS_IN_LPI ; -if (data->has_integrated_pcs) - plat_dat->flags |=STMMAC_FLAG_HAS_INTEGRATED_PCS ;if (data->dma_addr_width) plat_dat->host_dma_width = data->dma_addr_width;
drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c +0 -55
67 lines not on this page · raw message
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +0 -9
@@ -6000,15 +6000,6 @@ static void stmmac_common_interrupt(struct stmmac_priv *priv)for (queue =0 ; queue < queues_count; queue++)stmmac_host_mtl_irq_status (priv, priv->hw, queue); -/* PCS link status */ -if (priv->hw->pcs && - !(priv->plat->flags &STMMAC_FLAG_HAS_INTEGRATED_PCS )) { -if (priv->xstats.pcs_link) -netif_carrier_on (priv->dev); -else -netif_carrier_off (priv->dev); - } -stmmac_timestamp_interrupt (priv, priv); } }
include/linux/stmmac.h +0 -1
@@ -171,7 +171,6 @@ struct dwmac4_addrs {u32 mtl_low_cred_offset; }; -#define STMMAC_FLAG_HAS_INTEGRATED_PCS BIT(0)#define STMMAC_FLAG_SPH_DISABLE BIT(1)#define STMMAC_FLAG_USE_PHY_WOL BIT(2)#define STMMAC_FLAG_HAS_SUN8I BIT(3)
Signature · 1 line
2.47.3
PATCH v2 2/14 net: stmmac: remove xstats.pcs_* members
In reply to Russell King (Oracle)As a result of the previous commit, the pcs_link, pcs_duplex and pcs_speed members are not used outside of the interrupt handling code, and are only used to print their status using the misleading "Link is" messages that bear no relation to the actual status of the link.
Remove the printing of these messages, these members, and the code that decodes them from the hardware.
--- drivers/net/ethernet/stmicro/stmmac/common.h | 3 -- .../ethernet/stmicro/stmmac/dwmac1000_core.c | 28 +------------------ .../net/ethernet/stmicro/stmmac/dwmac4_core.c | 28 +------------------ 3 files changed, 2 insertions(+), 57 deletions(-)
drivers/net/ethernet/stmicro/stmmac/common.h +0 -3
@@ -192,9 +192,6 @@ struct stmmac_extra_stats {unsigned long irq_pcs_ane_n;unsigned long irq_pcs_link_n;unsigned long irq_rgmii_n; -unsigned long pcs_link; -unsigned long pcs_duplex; -unsigned long pcs_speed;/* debug register */ unsigned long mtl_tx_status_fifo_full;unsigned long mtl_tx_fifo_not_empty;
drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c +1 -27
@@ -266,34 +266,8 @@ static void dwmac1000_pmt(struct mac_device_info *hw, unsigned long mode)/* RGMII or SMII interface */ static void dwmac1000_rgsmii (void __iomem *ioaddr,struct stmmac_extra_stats *x) { -u32 status; - - status =readl (ioaddr +GMAC_RGSMIIIS ); +readl (ioaddr +GMAC_RGSMIIIS ); x->irq_rgmii_n++; - -/* Check the link status */ -if (status &GMAC_RGSMIIIS_LNKSTS ) { -int speed_value; - - x->pcs_link =1 ; - - speed_value = ((status &GMAC_RGSMIIIS_SPEED ) >> -GMAC_RGSMIIIS_SPEED_SHIFT ); -if (speed_value ==GMAC_RGSMIIIS_SPEED_125 ) - x->pcs_speed =SPEED_1000 ; -else if (speed_value ==GMAC_RGSMIIIS_SPEED_25 ) - x->pcs_speed =SPEED_100 ; -else - x->pcs_speed =SPEED_10 ; - - x->pcs_duplex = (status &GMAC_RGSMIIIS_LNKMOD_MASK ); - -pr_info ("Link is Up - %d/%s\n" , (int )x->pcs_speed, - x->pcs_duplex ?"Full" :"Half" ); - }else { - x->pcs_link =0 ; -pr_info ("Link is Down\n" ); - } } static int dwmac1000_irq_status(struct mac_device_info *hw,
drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +1 -27
@@ -592,34 +592,8 @@ static void dwmac4_ctrl_ane(struct stmmac_priv *priv, bool ane, bool srgmi_ral,/* RGMII or SMII interface */ static void dwmac4_phystatus (void __iomem *ioaddr,struct stmmac_extra_stats *x) { -u32 status; - - status =readl (ioaddr +GMAC_PHYIF_CONTROL_STATUS ); +readl (ioaddr +GMAC_PHYIF_CONTROL_STATUS ); x->irq_rgmii_n++; - -/* Check the link status */ -if (status &GMAC_PHYIF_CTRLSTATUS_LNKSTS ) { -int speed_value; - - x->pcs_link =1 ; - - speed_value = ((status &GMAC_PHYIF_CTRLSTATUS_SPEED ) >> -GMAC_PHYIF_CTRLSTATUS_SPEED_SHIFT ); -if (speed_value ==GMAC_PHYIF_CTRLSTATUS_SPEED_125 ) - x->pcs_speed =SPEED_1000 ; -else if (speed_value ==GMAC_PHYIF_CTRLSTATUS_SPEED_25 ) - x->pcs_speed =SPEED_100 ; -else - x->pcs_speed =SPEED_10 ; - - x->pcs_duplex = (status &GMAC_PHYIF_CTRLSTATUS_LNKMOD ); - -pr_info ("Link is Up - %d/%s\n" , (int )x->pcs_speed, - x->pcs_duplex ?"Full" :"Half" ); - }else { - x->pcs_link =0 ; -pr_info ("Link is Down\n" ); - } } static int dwmac4_irq_mtl_status(struct stmmac_priv *priv,
Signature · 1 line
2.47.3
PATCH v2 3/14 net: stmmac: remove SGMII/RGMII/SMII interrupt handling
In reply to Russell King (Oracle)Now that the only use for the interrupt is to clear it and increment a statistic counter (which is not that relevant anymore) remove all this code and ensure that the interrupt remains disabled to avoid a stuck interrupt.
dwmac-sun8i still uses this statistic counter, so it is inappropriate for this patch to remove it.
--
v2: add comment about why the statistic counter needs to stay.
--- drivers/net/ethernet/stmicro/stmmac/dwmac1000.h | 6 +++--- drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c | 10 ---------- drivers/net/ethernet/stmicro/stmmac/dwmac4.h | 3 +-- drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c | 9 --------- 4 files changed, 4 insertions(+), 24 deletions(-)
drivers/net/ethernet/stmicro/stmmac/dwmac1000.h +3 -3
@@ -38,10 +38,10 @@#define GMAC_INT_DISABLE_PCSAN BIT(2)#define GMAC_INT_DISABLE_PMT BIT(3)#define GMAC_INT_DISABLE_TIMESTAMP BIT(9) -#define GMAC_INT_DISABLE_PCS (GMAC_INT_DISABLE_RGMII | \ - GMAC_INT_DISABLE_PCSLINK | \ +#define GMAC_INT_DISABLE_PCS (GMAC_INT_DISABLE_PCSLINK | \ GMAC_INT_DISABLE_PCSAN) -#define GMAC_INT_DEFAULT_MASK (GMAC_INT_DISABLE_TIMESTAMP | \ +#define GMAC_INT_DEFAULT_MASK (GMAC_INT_DISABLE_RGMII | \ + GMAC_INT_DISABLE_TIMESTAMP | \ GMAC_INT_DISABLE_PCS)/* PMT Control and Status */
drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c +0 -10
@@ -263,13 +263,6 @@ static void dwmac1000_pmt(struct mac_device_info *hw, unsigned long mode)writel (pmt, ioaddr +GMAC_PMT ); } -/* RGMII or SMII interface */ -static void dwmac1000_rgsmii (void __iomem *ioaddr,struct stmmac_extra_stats *x) -{ -readl (ioaddr +GMAC_RGSMIIIS ); - x->irq_rgmii_n++; -} -static int dwmac1000_irq_status (struct mac_device_info *hw,struct stmmac_extra_stats *x) {
@@ -311,9 +304,6 @@ static int dwmac1000_irq_status(struct mac_device_info *hw,dwmac_pcs_isr (ioaddr,GMAC_PCS_BASE , intr_status, x); -if (intr_status &PCS_RGSMIIIS_IRQ ) -dwmac1000_rgsmii (ioaddr, x); -return ret; }
drivers/net/ethernet/stmicro/stmmac/dwmac4.h +1 -2
@@ -106,8 +106,7 @@#define GMAC_INT_LPI_EN BIT(5)#define GMAC_INT_TSIE BIT(12) -#define GMAC_PCS_IRQ_DEFAULT (GMAC_INT_RGSMIIS | GMAC_INT_PCS_LINK | \ - GMAC_INT_PCS_ANE) +#define GMAC_PCS_IRQ_DEFAULT (GMAC_INT_PCS_LINK | GMAC_INT_PCS_ANE)#define GMAC_INT_DEFAULT_ENABLE (GMAC_INT_PMT_EN | GMAC_INT_LPI_EN | \ GMAC_INT_TSIE)
drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +0 -9
@@ -589,13 +589,6 @@ static void dwmac4_ctrl_ane(struct stmmac_priv *priv, bool ane, bool srgmi_ral,dwmac_ctrl_ane (priv->ioaddr,GMAC_PCS_BASE , ane, srgmi_ral, loopback); } -/* RGMII or SMII interface */ -static void dwmac4_phystatus (void __iomem *ioaddr,struct stmmac_extra_stats *x) -{ -readl (ioaddr +GMAC_PHYIF_CONTROL_STATUS ); - x->irq_rgmii_n++; -} -static int dwmac4_irq_mtl_status (struct stmmac_priv *priv,struct mac_device_info *hw,u32 chan) {
@@ -667,8 +660,6 @@ static int dwmac4_irq_status(struct mac_device_info *hw, }dwmac_pcs_isr (ioaddr,GMAC_PCS_BASE , intr_status, x); -if (intr_status &PCS_RGSMIIIS_IRQ ) -dwmac4_phystatus (ioaddr, x);return ret; }
Signature · 1 line
2.47.3
PATCH v2 4/14 net: stmmac: remove PCS "mode" pause handling
In reply to Russell King (Oracle)Remove the "we always autoneg pause" forcing when the stmmac driver decides that a "PCS" is present, which blocks passing the ethtool pause calls to phylink when using SGMII mode.
This prevents the pause results being reported when a PHY is attached using SGMII mode, or the pause settings being changed in SGMII mode. There is no reason to prevent this.
--- .../net/ethernet/stmicro/stmmac/stmmac_ethtool.c | 13 ++----------- 1 file changed, 2 insertions(+), 11 deletions(-)
drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c +2 -11
@@ -424,11 +424,7 @@ stmmac_get_pauseparam(struct net_device *netdev, {struct stmmac_priv *priv =netdev_priv (netdev); -if (priv->hw->pcs) { - pause->autoneg =1 ; - }else { -phylink_ethtool_get_pauseparam (priv->phylink, pause); - } +phylink_ethtool_get_pauseparam (priv->phylink, pause); } static int
@@ -437,12 +433,7 @@ stmmac_set_pauseparam(struct net_device *netdev, {struct stmmac_priv *priv =netdev_priv (netdev); -if (priv->hw->pcs) { - pause->autoneg =1 ; -return 0 ; - }else { -return phylink_ethtool_set_pauseparam (priv->phylink, pause); - } +return phylink_ethtool_set_pauseparam (priv->phylink, pause); }static u64 stmmac_get_rx_normal_irq_n (struct stmmac_priv *priv,int q)
Signature · 1 line
2.47.3
PATCH v2 5/14 net: stmmac: remove unused PCS loopback support
In reply to Russell King (Oracle)Nothing calls stmmac_pcs_ctrl_ane() with the "loopback" argument set to anything except zero, so this serves no useful purpose. Remove the argument to reduce the code complexity.
--- drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c | 2 +- drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c | 4 ++-- drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c | 5 ++--- drivers/net/ethernet/stmicro/stmmac/hwif.h | 4 ++-- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 2 +- drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h | 6 +----- 6 files changed, 9 insertions(+), 14 deletions(-)
drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c +1 -1
@@ -622,7 +622,7 @@ static void ethqos_set_serdes_speed(struct qcom_ethqos *ethqos, int speed)static void ethqos_pcs_set_inband (struct stmmac_priv *priv,bool enable) { -stmmac_pcs_ctrl_ane (priv, enable,0 ,0 ); +stmmac_pcs_ctrl_ane (priv, enable,0 ); }/* On interface toggle MAC registers gets reset.
drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c +2 -2
@@ -358,9 +358,9 @@ static void dwmac1000_set_eee_timer(struct mac_device_info *hw, int ls, int tw) }static void dwmac1000_ctrl_ane (struct stmmac_priv *priv,bool ane, -bool srgmi_ral,bool loopback) +bool srgmi_ral) { -dwmac_ctrl_ane (priv->ioaddr,GMAC_PCS_BASE , ane, srgmi_ral, loopback); +dwmac_ctrl_ane (priv->ioaddr,GMAC_PCS_BASE , ane, srgmi_ral); } static void dwmac1000_debug(struct stmmac_priv *priv, void __iomem *ioaddr,
drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +2 -3
@@ -583,10 +583,9 @@ static void dwmac4_flow_ctrl(struct mac_device_info *hw, unsigned int duplex, } } -static void dwmac4_ctrl_ane (struct stmmac_priv *priv,bool ane,bool srgmi_ral, -bool loopback) +static void dwmac4_ctrl_ane (struct stmmac_priv *priv,bool ane,bool srgmi_ral) { -dwmac_ctrl_ane (priv->ioaddr,GMAC_PCS_BASE , ane, srgmi_ral, loopback); +dwmac_ctrl_ane (priv->ioaddr,GMAC_PCS_BASE , ane, srgmi_ral); } static int dwmac4_irq_mtl_status(struct stmmac_priv *priv,
drivers/net/ethernet/stmicro/stmmac/hwif.h +2 -2
@@ -374,8 +374,8 @@ struct stmmac_ops { struct stmmac_extra_stats *x, u32 rx_queues, u32 tx_queues);/* PCS calls */ -void (*pcs_ctrl_ane)(struct stmmac_priv *priv,bool ane,bool srgmi_ral, -bool loopback); +void (*pcs_ctrl_ane)(struct stmmac_priv *priv,bool ane, +bool srgmi_ral);/* Safety Features */ int (*safety_feat_config)(void __iomem *ioaddr,unsigned int asp,struct stmmac_safety_feature_cfg *safety_cfg);
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +1 -1
@@ -3493,7 +3493,7 @@ static int stmmac_hw_setup(struct net_device *dev) }if (priv->hw->pcs) -stmmac_pcs_ctrl_ane (priv,1 , priv->hw->ps,0 ); +stmmac_pcs_ctrl_ane (priv,1 , priv->hw->ps);/* set TX and RX rings length */ stmmac_set_rings_length (priv);
drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h +1 -5
@@ -82,13 +82,12 @@ static inline void dwmac_pcs_isr(void __iomem *ioaddr, u32 reg,* @reg: Base address of the AN Control Register. * @ane: to enable the auto-negotiation * @srgmi_ral: to manage MAC-2-MAC SGMII connections. -* @loopback: to cause the PHY to loopback tx data into rx path. * Description: this is the main function to configure the AN control register * and init the ANE, select loopback (usually for debugging purpose) and * configure SGMII RAL. */ static inline void dwmac_ctrl_ane (void __iomem *ioaddr,u32 reg,bool ane, -bool srgmi_ral,bool loopback) +bool srgmi_ral) {u32 value =readl (ioaddr +GMAC_AN_CTRL (reg));
@@ -104,9 +103,6 @@ static inline void dwmac_ctrl_ane(void __iomem *ioaddr, u32 reg, bool ane,if (srgmi_ral) value |=GMAC_AN_CTRL_SGMRAL ; -if (loopback) - value |=GMAC_AN_CTRL_ELE ; - writel(value, ioaddr + GMAC_AN_CTRL(reg)); }#endif /* __STMMAC_PCS_H__ */
Signature · 1 line
2.47.3
PATCH v2 6/14 net: stmmac: remove hw->ps xxx_core_init() hardware setup
In reply to Russell King (Oracle)After a lot of digging, it seems that the oddly named hw->ps member is all about configuring the core for reverse SGMII. This member is set to one of 0, SPEED_10, SPEED_100 or SPEED_1000 depending on priv->plat->mac_port_sel_speed. On DT systems, this comes from the "snps,ps-speed" DT property.
When set to a non-zero value, it:
1. Configures the MAC at initialisation time to operate at a specific speed. However, this will be overwritten by mac_link_up() when the link comes up (e.g. with the fixed-link parameters.)
Note that dwxgmac2 wants to also support SPEED_2500 and SPEED_10000, but both these values are impossible.
2. It _incorrectly_ enables the transmitter (GMAC_CONFIG_TE) which makes no sense, rather than enabling the "transmit configuration" bit (GMAC_CONFIG_TC). Likely a typo.
3. It configures the SGMII rate adapter layer to retrieve its speed setting from the MAC configuration register rather than the PHY.
There are two ways forward here:
a) fixing (2) so that we set GMAC_CONFIG_TC. However, we have platform that set the "snps,ps-speed" property and that work today. Fixing this will cause the RGMII, SGMII or SMII inband configuration to be transmitted, which will be a functional change which could cause a regression.
b) ripping out (1) and (2) as they are ineffective. This also has the possibility of regressions, but the patch author believes this risk is much lower than (a).
Therefore, this commit takes the approach in (b).
--- .../ethernet/stmicro/stmmac/dwmac1000_core.c | 23 +++-------------- .../net/ethernet/stmicro/stmmac/dwmac4_core.c | 24 +++--------------- .../ethernet/stmicro/stmmac/dwxgmac2_core.c | 25 ++----------------- 3 files changed, 8 insertions(+), 64 deletions(-)
drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c +3 -20
@@ -26,35 +26,18 @@ static void dwmac1000_core_init(struct mac_device_info *hw, struct net_device *dev) {void __iomem *ioaddr = hw->pcsr; -u32 value =readl (ioaddr +GMAC_CONTROL );int mtu = dev->mtu; +u32 value;/* Configure GMAC core */ - value |=GMAC_CORE_INIT ; + value =readl (ioaddr +GMAC_CONTROL );if (mtu >1500 ) value |=GMAC_CONTROL_2K ;if (mtu >2000 ) value |=GMAC_CONTROL_JE ; -if (hw->ps) { - value |=GMAC_CONTROL_TE ; - - value &= ~hw->link.speed_mask; -switch (hw->ps) { -case SPEED_1000 : - value |= hw->link.speed1000; -break ; -case SPEED_100 : - value |= hw->link.speed100; -break ; -case SPEED_10 : - value |= hw->link.speed10; -break ; - } - } - -writel (value, ioaddr +GMAC_CONTROL ); +writel (value |GMAC_CORE_INIT , ioaddr +GMAC_CONTROL );/* Mask GMAC interrupts */ value =GMAC_INT_DEFAULT_MASK ;
drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +3 -21
@@ -27,29 +27,11 @@ static void dwmac4_core_init(struct mac_device_info *hw, {struct stmmac_priv *priv =netdev_priv (dev);void __iomem *ioaddr = hw->pcsr; -u32 value =readl (ioaddr +GMAC_CONFIG );unsigned long clk_rate; +u32 value; - value |=GMAC_CORE_INIT ; - -if (hw->ps) { - value |=GMAC_CONFIG_TE ; - - value &= hw->link.speed_mask; -switch (hw->ps) { -case SPEED_1000 : - value |= hw->link.speed1000; -break ; -case SPEED_100 : - value |= hw->link.speed100; -break ; -case SPEED_10 : - value |= hw->link.speed10; -break ; - } - } - -writel (value, ioaddr +GMAC_CONFIG ); + value =readl (ioaddr +GMAC_CONFIG ); +writel (value |GMAC_CORE_INIT , ioaddr +GMAC_CONFIG );/* Configure LPI 1us counter to number of CSR clock ticks in 1us - 1 */ clk_rate =clk_get_rate (priv->plat->stmmac_clk);
drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c +2 -23
@@ -23,29 +23,8 @@ static void dwxgmac2_core_init(struct mac_device_info *hw, tx =readl (ioaddr +XGMAC_TX_CONFIG ); rx =readl (ioaddr +XGMAC_RX_CONFIG ); - tx |=XGMAC_CORE_INIT_TX ; - rx |=XGMAC_CORE_INIT_RX ; - -if (hw->ps) { - tx |=XGMAC_CONFIG_TE ; - tx &= ~hw->link.speed_mask; - -switch (hw->ps) { -case SPEED_10000 : - tx |= hw->link.xgmii.speed10000; -break ; -case SPEED_2500 : - tx |= hw->link.speed2500; -break ; -case SPEED_1000 : -default : - tx |= hw->link.speed1000; -break ; - } - } - -writel (tx, ioaddr +XGMAC_TX_CONFIG ); -writel (rx, ioaddr +XGMAC_RX_CONFIG ); +writel (tx |XGMAC_CORE_INIT_TX , ioaddr +XGMAC_TX_CONFIG ); +writel (rx |XGMAC_CORE_INIT_RX , ioaddr +XGMAC_RX_CONFIG );writel (XGMAC_INT_DEFAULT_EN , ioaddr +XGMAC_INT_EN ); }
Signature · 1 line
2.47.3
PATCH v2 7/14 net: stmmac: remove RGMII "pcs" mode
In reply to Russell King (Oracle)Remove the RGMII "pcs" code in stmmac_check_pcs_mode() due to:
1) This should never have been conditional on a PCS being present, as when a core is synthesised using only RGMII, the PCS won't be present and priv->dma_cap.pcs will be false. Only multi-interface cores which have a PCS present would have detected RGMII.
2) STMMAC_PCS_RGMII has no effect since the broken netif_carrier and ethtool code was removed.
--- drivers/net/ethernet/stmicro/stmmac/common.h | 1 - drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 14 +++----------- 2 files changed, 3 insertions(+), 12 deletions(-)
drivers/net/ethernet/stmicro/stmmac/common.h +0 -1
@@ -270,7 +270,6 @@ struct stmmac_safety_stats {#define FLOW_AUTO (FLOW_TX | FLOW_RX)/* PCS defines */ -#define STMMAC_PCS_RGMII (1 << 0)#define STMMAC_PCS_SGMII (1 << 1)#define SF_DMA_MODE 1/* DMA STORE-AND-FORWARD Operation Mode */
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +3 -11
@@ -1087,17 +1087,9 @@ static void stmmac_check_pcs_mode(struct stmmac_priv *priv) {int interface = priv->plat->phy_interface; -if (priv->dma_cap.pcs) { -if ((interface ==PHY_INTERFACE_MODE_RGMII ) || - (interface ==PHY_INTERFACE_MODE_RGMII_ID ) || - (interface ==PHY_INTERFACE_MODE_RGMII_RXID ) || - (interface ==PHY_INTERFACE_MODE_RGMII_TXID )) { -netdev_dbg (priv->dev,"PCS RGMII support enabled\n" ); - priv->hw->pcs =STMMAC_PCS_RGMII ; - }else if (interface ==PHY_INTERFACE_MODE_SGMII ) { -netdev_dbg (priv->dev,"PCS SGMII support enabled\n" ); - priv->hw->pcs =STMMAC_PCS_SGMII ; - } +if (priv->dma_cap.pcs && interface ==PHY_INTERFACE_MODE_SGMII ) { +netdev_dbg (priv->dev,"PCS SGMII support enabled\n" ); + priv->hw->pcs =STMMAC_PCS_SGMII ; } }
Signature · 1 line
2.47.3
PATCH v2 8/14 net: stmmac: move reverse-"pcs" mode setup to stmmac_check_pcs_mode()
In reply to Russell King (Oracle)The broken reverse-mode, selected by snps,ps-speed, is configured when the platform provides a valid port speed and a PCS is being used.
Both these remain constant after the driver has probed, so the software state doesn't need to be re-initialised each time stmmac_hw_setup() is called (which is called at open and resume time.)
Move the software setup of reverse-mode to stmmac_check_pcs_mode() which is called from the driver probe function.
--- .../net/ethernet/stmicro/stmmac/stmmac_main.c | 26 +++++++++---------- 1 file changed, 13 insertions(+), 13 deletions(-)
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +13 -13
@@ -1091,6 +1091,19 @@ static void stmmac_check_pcs_mode(struct stmmac_priv *priv)netdev_dbg (priv->dev,"PCS SGMII support enabled\n" ); priv->hw->pcs =STMMAC_PCS_SGMII ; } + +/* PS and related bits will be programmed according to the speed */ +if (priv->hw->pcs) { +int speed = priv->plat->mac_port_sel_speed; + +if ((speed ==SPEED_10 ) || (speed ==SPEED_100 ) || + (speed ==SPEED_1000 )) { + priv->hw->ps = speed; + }else { +dev_warn (priv->device,"invalid port speed\n" ); + priv->hw->ps =0 ; + } + } }/**
@@ -3435,19 +3448,6 @@ static int stmmac_hw_setup(struct net_device *dev)stmmac_set_umac_addr (priv, priv->hw, dev->dev_addr,0 );phylink_rx_clk_stop_unblock (priv->phylink); -/* PS and related bits will be programmed according to the speed */ -if (priv->hw->pcs) { -int speed = priv->plat->mac_port_sel_speed; - -if ((speed ==SPEED_10 ) || (speed ==SPEED_100 ) || - (speed ==SPEED_1000 )) { - priv->hw->ps = speed; - }else { -dev_warn (priv->device,"invalid port speed\n" ); - priv->hw->ps =0 ; - } - } -/* Initialize the MAC Core */ stmmac_core_init (priv, priv->hw, dev);
Signature · 1 line
2.47.3
PATCH v2 11/14 net: stmmac: do not require snps,ps-speed for SGMII
In reply to Russell King (Oracle)SGMII mode does not require port-speed to be specified; this only switches SGMII to use the MAC configuration register speed settings and the actual value is irrelevant when the link comes up.
As it seems the intention was to support "reverse SGMII" with this setting, but the code didn't actually configure that due to a typo, the warning and bad DT binding documentation has led people to specify snps,ps-speed in their DT files inappropriately.
If mac_port_sel_speed is zero, then don't complain that the speed is invalid, as this means we're using "normal" SGMII.
This does _not_ obsolete snps,ps-speed, nor does it change the behaviour of that property, with the exception of not making people mistakenly believe that they need to specify this option to use normal SGMII. There is no need to modify the binding.
---
v2: add comment about why there's no requirement to change the DT binding.
--- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +2 -0
@@ -1101,6 +1101,8 @@ static void stmmac_check_pcs_mode(struct stmmac_priv *priv)default :dev_warn (priv->device,"invalid port speed\n" ); + fallthrough; +case 0 : priv->hw->reverse_sgmii_enable = false;break ; }
Signature · 1 line
2.47.3
PATCH v2 13/14 net: stmmac: provide PCS initialisation hook
In reply to Russell King (Oracle)dwmac cores provide a feature bit to indicate when the PCS block is present, but features are only read after the core's setup() function has been called, meaning we can't decide whether to initialise the integrated PCS in the setup function. Provide a new MAC core hook for PCS initialisation, which will be called after the feature registers have been read.
--- drivers/net/ethernet/stmicro/stmmac/hwif.h | 4 ++++ drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 7 +++++++ 2 files changed, 11 insertions(+)
drivers/net/ethernet/stmicro/stmmac/hwif.h +4 -0
@@ -313,6 +313,8 @@ enum stmmac_lpi_mode {/* Helpers to program the MAC core */ struct stmmac_ops { +/* Initialise any PCS instances */ +int (*pcs_init)(struct stmmac_priv *priv);/* MAC core initialization */ void (*core_init)(struct mac_device_info *hw,struct net_device *dev);/* Update MAC capabilities */
@@ -413,6 +415,8 @@ struct stmmac_ops { u32 pclass); }; +#define stmmac_mac_pcs_init (__priv) \ + stmmac_do_callback(__priv, mac, pcs_init, __priv) #define stmmac_core_init(__priv, __args...) \ stmmac_do_void_callback(__priv, mac, core_init, __args)#define stmmac_mac_update_caps (__priv) \
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +7 -0
@@ -7238,6 +7238,13 @@ static int stmmac_hw_init(struct stmmac_priv *priv) "Enable RX Mitigation via HW Watchdog Timer\n"); } +/* Unimplemented PCS init (as indicated by stmmac_do_callback() +* perversely returning -EINVAL) is non-fatal. +*/ + ret =stmmac_mac_pcs_init (priv); +if (ret != -EINVAL ) +return ret; +return 0 ; }
Signature · 1 line
2.47.3
PATCH v2 14/14 net: stmmac: convert to phylink PCS support
In reply to Russell King (Oracle)Now that stmmac's PCS support is much more simple - just a matter of configuring the control register - the basic conversion to phylink PCS support becomes straight forward.
Create the infrastructure to setup a phylink_pcs instance for the integrated PCS:
- add a struct stmmac_pcs to encapsulate the phylink_pcs structure, pointer to stmmac_priv, and the core-specific base address of the PCS registers.
- modify stmmac_priv and stmmac_mac_select_pcs() to return the embedded phylink_pcs structure when setup and STMMAC_PCS_SGMII is in use, and move the comment from stmmac_hw_setup() to here.
- create stmmac_pcs.c, which contains the phylink_pcs_ops structure, a dummy .pcs_get_state() method which always reports link-down, and .pcs_config() method, moving the call to stmmac_pcs_ctrl_ane() here, but without indirecting through the dwmac specific core code. The link-down behaviour mentioned above maintains the current behaviour when phylink is used with inband but without a PCS.
This will ensure that the PCS control register is configured to the same settings as before, but will now happen when the netdev is opened or reusmed rather than only during probe time. However, this will be before the .fix_mac_speed() method is called, which is critical for the behaviour in dwmac-qcom-ethqos's ethqos_configure_sgmii() function to be maintained.
---
v2: add comment about why the link-down function is added.
--- drivers/net/ethernet/stmicro/stmmac/Makefile | 2 +- .../ethernet/stmicro/stmmac/dwmac1000_core.c | 9 ++++ .../net/ethernet/stmicro/stmmac/dwmac4_core.c | 11 +++++ drivers/net/ethernet/stmicro/stmmac/stmmac.h | 4 ++ .../net/ethernet/stmicro/stmmac/stmmac_main.c | 15 +++--- .../net/ethernet/stmicro/stmmac/stmmac_pcs.c | 47 +++++++++++++++++++ .../net/ethernet/stmicro/stmmac/stmmac_pcs.h | 17 +++++++ 7 files changed, 97 insertions(+), 8 deletions(-) create mode 100644 drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.c
drivers/net/ethernet/stmicro/stmmac/Makefile +1 -1
@@ -7,7 +7,7 @@ stmmac-objs:= stmmac_main.o stmmac_ethtool.o stmmac_mdio.o ring_mode.o \ dwmac4_dma.o dwmac4_lib.o dwmac4_core.o dwmac5.o hwif.o \ stmmac_tc.o dwxgmac2_core.o dwxgmac2_dma.o dwxgmac2_descs.o \ stmmac_xdp.o stmmac_est.o stmmac_fpe.o stmmac_vlan.o \ - $(stmmac-y) + stmmac_pcs.o $(stmmac-y) stmmac-$(CONFIG_STMMAC_SELFTESTS) += stmmac_selftests.o
drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c +9 -0
@@ -22,6 +22,14 @@#include "stmmac_ptp.h" #include "dwmac1000.h" +static int dwmac1000_pcs_init (struct stmmac_priv *priv) +{ +if (!priv->dma_cap.pcs) +return 0 ; + +return stmmac_integrated_pcs_init (priv,GMAC_PCS_BASE ); +} +static void dwmac1000_core_init (struct mac_device_info *hw,struct net_device *dev) {
@@ -435,6 +443,7 @@ static void dwmac1000_set_mac_loopback(void __iomem *ioaddr, bool enable) }const struct stmmac_ops dwmac1000_ops = { + .pcs_init = dwmac1000_pcs_init, .core_init = dwmac1000_core_init, .set_mac = stmmac_set_mac, .rx_ipc = dwmac1000_rx_ipc_enable,
drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +11 -0
@@ -22,6 +22,14 @@#include "dwmac4.h" #include "dwmac5.h" +static int dwmac4_pcs_init (struct stmmac_priv *priv) +{ +if (!priv->dma_cap.pcs) +return 0 ; + +return stmmac_integrated_pcs_init (priv,GMAC_PCS_BASE ); +} +static void dwmac4_core_init (struct mac_device_info *hw,struct net_device *dev) {
@@ -875,6 +883,7 @@ static int dwmac4_config_l4_filter(struct mac_device_info *hw, u32 filter_no, }const struct stmmac_ops dwmac4_ops = { + .pcs_init = dwmac4_pcs_init, .core_init = dwmac4_core_init, .update_caps = dwmac4_update_caps, .set_mac = stmmac_set_mac,
@@ -909,6 +918,7 @@ const struct stmmac_ops dwmac4_ops = { };const struct stmmac_ops dwmac410_ops = { + .pcs_init = dwmac4_pcs_init, .core_init = dwmac4_core_init, .update_caps = dwmac4_update_caps, .set_mac = stmmac_dwmac4_set_mac,
@@ -945,6 +955,7 @@ const struct stmmac_ops dwmac410_ops = { };const struct stmmac_ops dwmac510_ops = { + .pcs_init = dwmac4_pcs_init, .core_init = dwmac4_core_init, .update_caps = dwmac4_update_caps, .set_mac = stmmac_dwmac4_set_mac,
drivers/net/ethernet/stmicro/stmmac/stmmac.h +4 -0
@@ -25,6 +25,8 @@#include <net/xdp.h> #include <uapi/linux/bpf.h> +struct stmmac_pcs ; +struct stmmac_resources {void __iomem *addr;u8 mac[ETH_ALEN ];
@@ -273,6 +275,8 @@ struct stmmac_priv {unsigned int pause_time;struct mii_bus *mii; +struct stmmac_pcs *integrated_pcs; +struct phylink_config phylink_config;struct phylink *phylink;
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +8 -7
@@ -46,6 +46,7 @@#include "stmmac_ptp.h" #include "stmmac_fpe.h" #include "stmmac.h" +#include "stmmac_pcs.h" #include "stmmac_xdp.h" #include <linux/reset.h> #include <linux/of_mdio.h>
@@ -850,6 +851,13 @@ static struct phylink_pcs *stmmac_mac_select_pcs(struct phylink_config *config,return pcs; } +/* The PCS control register is only relevant for SGMII, TBI and RTBI +* modes. We no longer support TBI or RTBI, so only configure this +* register when operating in SGMII mode with the integrated PCS. +*/ +if (priv->hw->pcs &STMMAC_PCS_SGMII && priv->integrated_pcs) +return &priv->integrated_pcs->pcs; +return NULL ; }
@@ -3487,13 +3495,6 @@ static int stmmac_hw_setup(struct net_device *dev) } } -/* The PCS control register is only relevant for SGMII, TBI and RTBI -* modes. We no longer support TBI or RTBI, so only configure this -* register when operating in SGMII mode with the integrated PCS. -*/ -if (priv->hw->pcs &STMMAC_PCS_SGMII ) -stmmac_pcs_ctrl_ane (priv,1 , priv->hw->reverse_sgmii_enable); -/* set TX and RX rings length */ stmmac_set_rings_length (priv);
drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.c new file +47 -0
47 lines not on this page · raw message
drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h +17 -0
@@ -9,6 +9,7 @@#ifndef __STMMAC_PCS_H__#define __STMMAC_PCS_H__ +#include <linux/phylink.h> #include <linux/slab.h> #include <linux/io.h> #include "common.h"
@@ -46,6 +47,22 @@#define GMAC_ANE_RFE_SHIFT 12#define GMAC_ANE_ACK BIT(14) +struct stmmac_priv ; + +struct stmmac_pcs { +struct stmmac_priv *priv; +void __iomem *base; +struct phylink_pcs pcs; +}; + +static inline struct stmmac_pcs * +phylink_pcs_to_stmmac_pcs (struct phylink_pcs *pcs) +{ + return container_of(pcs, struct stmmac_pcs, pcs); +} + +int stmmac_integrated_pcs_init (struct stmmac_priv *priv,unsigned int offset); +/** * dwmac_pcs_isr - TBI, RTBI, or SGMII PHY ISR * @ioaddr: IO registers pointer
Signature · 1 line
2.47.3
PATCH v2 10/14 net: stmmac: hw->ps becomes hw->reverse_sgmii_enable
In reply to Russell King (Oracle)After a lot of digging, it seems that the oddly named hw->ps member is all about setting the core into reverse SGMII speed. When set to a non-zero value, it:
1. Configures the MAC at initialisation time to operate at a specific speed.
2. It _incorrectly_ enables the transmitter (GMAC_CONFIG_TE) which makes no sense, rather than enabling the "transmit configuration" bit (GMAC_CONFIG_TC).
3. It configures the SGMII rate adapter layer to retrieve its speed setting from the MAC configuration register rather than the PHY.
In the previous commit, we removed (1) and (2) as phylink overwrites the configuration set at that step.
Thus, the only functional aspect is (3), which is a boolean operation. This means there is no need to store the actual speed, and just have a boolean flag.
Convert the priv->ps member to a boolean, and rename it to priv->reverse_sgmii_enable to make it more understandable.
--- drivers/net/ethernet/stmicro/stmmac/common.h | 2 +- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 6 +++--- 2 files changed, 4 insertions(+), 4 deletions(-)
drivers/net/ethernet/stmicro/stmmac/common.h +1 -1
@@ -599,13 +599,13 @@ struct mac_device_info {unsigned int mcast_bits_log2;unsigned int rx_csum;unsigned int pcs; -unsigned int ps;unsigned int xlgmac;unsigned int num_vlan;u32 vlan_filter[32 ];bool vlan_fail_q_en;u8 vlan_fail_q;bool hw_vlan_en; +bool reverse_sgmii_enable; };struct stmmac_rx_routing {
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +3 -3
@@ -1096,12 +1096,12 @@ static void stmmac_check_pcs_mode(struct stmmac_priv *priv)case SPEED_10 :case SPEED_100 :case SPEED_1000 : - priv->hw->ps = speed; + priv->hw->reverse_sgmii_enable = true;break ;default :dev_warn (priv->device,"invalid port speed\n" ); - priv->hw->ps =0 ; + priv->hw->reverse_sgmii_enable = false;break ; } }
@@ -3486,7 +3486,7 @@ static int stmmac_hw_setup(struct net_device *dev) }if (priv->hw->pcs) -stmmac_pcs_ctrl_ane (priv,1 , priv->hw->ps); +stmmac_pcs_ctrl_ane (priv,1 , priv->hw->reverse_sgmii_enable);/* set TX and RX rings length */ stmmac_set_rings_length (priv);
Signature · 1 line
2.47.3
PATCH v2 12/14 net: stmmac: only call stmmac_pcs_ctrl_ane() for integrated SGMII PCS
In reply to Russell King (Oracle)The internal PCS registers only exist if the core is synthesized with SGMII, TBI or RTBI support. They have no relevance for RGMII.
However, priv->hw->pcs contains a STMMAC_PCS_RGMII flag, which is set if a PCS has been synthesized but we are operating in RGMII mode. As the register has no effect for RGMII, there is no point calling stmmac_pcs_ctrl_ane() in this case. Add a comment describing this and make it conditional on STMMAC_PCS_SGMII.
--- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +5 -1
@@ -3487,7 +3487,11 @@ static int stmmac_hw_setup(struct net_device *dev) } } -if (priv->hw->pcs) +/* The PCS control register is only relevant for SGMII, TBI and RTBI +* modes. We no longer support TBI or RTBI, so only configure this +* register when operating in SGMII mode with the integrated PCS. +*/ +if (priv->hw->pcs &STMMAC_PCS_SGMII )stmmac_pcs_ctrl_ane (priv,1 , priv->hw->reverse_sgmii_enable);/* set TX and RX rings length */
Signature · 1 line
2.47.3
PATCH v2 9/14 net: stmmac: simplify stmmac_check_pcs_mode()
In reply to Russell King (Oracle)Now that we only support one mode, simplify stmmac_check_pcs_mode().
--- .../net/ethernet/stmicro/stmmac/stmmac_main.c | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-)
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +9 -8
@@ -1086,22 +1086,23 @@ static const struct phylink_mac_ops stmmac_phylink_mac_ops = {static void stmmac_check_pcs_mode (struct stmmac_priv *priv) {int interface = priv->plat->phy_interface; +int speed = priv->plat->mac_port_sel_speed;if (priv->dma_cap.pcs && interface ==PHY_INTERFACE_MODE_SGMII ) {netdev_dbg (priv->dev,"PCS SGMII support enabled\n" ); priv->hw->pcs =STMMAC_PCS_SGMII ; - } - -/* PS and related bits will be programmed according to the speed */ -if (priv->hw->pcs) { -int speed = priv->plat->mac_port_sel_speed; -if ((speed ==SPEED_10 ) || (speed ==SPEED_100 ) || - (speed ==SPEED_1000 )) { +switch (speed) { +case SPEED_10 : +case SPEED_100 : +case SPEED_1000 : priv->hw->ps = speed; - }else { +break ; + +default :dev_warn (priv->device,"invalid port speed\n" ); priv->hw->ps =0 ; +break ; } } }
Signature · 1 line
2.47.3
Hello,
Quoted 86 lines · Russell King (Oracle)
This series is radical - it takes the brave step of ripping out much of the existing PCS support code and throwing it all away.
I have discussed the introduction of the
STMMAC_FLAG_HAS_INTEGRATED_PCSflag with Bartosz Golaszewski, and the conclusion I came to is that this is to workaround the breakage that I've been going on about concerning the phylink conversion for the last five or six years.The problem is that the stmmac PCS code manipulates the netif carrier state, which confuses phylink.
There is a way of testing this out on the Jetson Xavier NX platform as the "PCS" code paths can be exercised while in RGMII mode - because RGMII also has in-band status and the status register is shared with SGMII. Testing this out confirms my long held theory: the interrupt handler manipulates the netif carrier state before phylink gets a look-in, which means that the
mac_link_up()andmac_link_down()methods are never called, resulting in the device being non-functional.Moreover, on dwmac4 cores, ethtool reports incorrect information - despite having a full-duplex link, ethtool reports that it is half-dupex.
Thus, this code is completely broken - anyone using it will not have a functional platform, and thus it doesn't deserve to live any longer, especially as it's a thorn in phylink.
Rip all this out, leaving just the bare bones initialisation in place.
However, this is not the last of what's broken. We have this
hw->psinteger which is really not descriptive, and the DT property from which it comes from does little to help understand what's going on. Putting all the clues together:- early configuration of the GMAC configuration register for the speed.
- setting the SGMII rate adapter layer to take its speed from the GMAC configuration register.Lastly, setting the transmit enable (TE) bit, which is a typo that puts the nail in the coffin of this code. It should be the transmit configuration (TC) bit. Given that when the link comes up, phylink will call
mac_link_up()which will overwrite the speed in the GMAC configuration register, the only part of this that is functional is changing where the SGMII rate adapter layer gets its speed from, which is a boolean.From what I've found so far, everyone who sets the snps,ps-speed property which configures this mode also configures a fixed link, so the pre-configuration is unnecessary - the link will come up anyway.
So, this series rips that out the preconfiguration as well, and replaces
hw->pswith a booleanhw->reverse_sgmii_enableflag.We then move the sole PCS configuration into a
phylink_pcsinstance, which configures the PCS control register in the same way as is done during the probe function.Thus, we end up with much easier and simpler conversion to phylink PCS than previous attempts.
Even so, this still results in inband mode always being enabled at the moment in the new .pcs_config() method to reflect what the probe function was doing. The next stage will be to change that to allow phylink to correctly configure the PCS. This needs fixing to allow platform glue maintainers who are currently blocked to progress.
Please note, however, that this has not been tested with any SGMII platform.
I've tried to get as many people into the Cc list with
get_maintainers, I hope that's sufficient to get enough eyeballs on this.Changes since RFC:
- new patch (7) to remove RGMII "pcs" mode
- new patch (8) to move reverse "pcs" mode tostmmac_check_pcs_mode()
- new patch (9) to simplify the code moved in the previous patch
- new patch (10) to rename the confusinghw->psto something more understandable.
- new patch (11) to shut up inappropriate complaints about "snps,ps-speed" being invalid.
- new patch (13) to add a MAC .pcs_init method, which will only be called when core has PCS present.
- modify patch 14 to use this newpcs_initmethod.
Quoted 7 lines
Despite getting a couple of responses to the RFC series posted in September, I have had nothing testing this on hardware. I have tested this on the Jetson Xavier NX, which included trial runs with enabling the RGMII "pcs" mode, hence the new patches that rip out this mode. I have come to the conclusion that the only way to get stmmac changes tested is to get them merged into net-next, thereby forcing people to have to run with them... and we'll deal with any fallout later.
I have tested this on :
- dwmac-socfpga w/ Lynx, SGMII and 1000BaseX mode
- dwmac-socfpga w/ RGMII mode
- stm32-dwmac (on stm32mp157a), w/ RGMII mode
I don't have any device available with stmmac using the internal PCS implementation, but at least the 2 platforms above don't regress with this series.
Maxime
Hi,
Quoted 140 lines
From: Russell King <[email protected]>
Sent: 16 October 2025 15:35
To: Andrew Lunn <[email protected]>; Heiner Kallweit <[email protected]>
Cc: Abhishek Chauhan <[email protected]>; Alexandre Torgue <[email protected]>; Alexis Lothore <[email protected]>; Andrew Lunn <[email protected]>; Boon Khai Ng <[email protected]>; Choong Yong Liang <[email protected]>; Daniel Machon <[email protected]>; David S. Miller <[email protected]>; Drew Fustini <[email protected]>; Emil Renner Berthing <[email protected]>; Eric Dumazet <[email protected]>; Faizal Rahim <[email protected]>; Furong Xu <[email protected]>; Inochi Amaoto <[email protected]>; Jacob Keller <[email protected]>; Jakub Kicinski <[email protected]>; Jan Petrous (OSS) <[email protected]>; Jisheng Zhang <[email protected]>; Kees Cook <[email protected]>; Kunihiko Hayashi <[email protected]>; Prabhakar Mahadev Lad <[email protected]>; Ley Foon Tan <[email protected]>; [email protected]; [email protected]; [email protected]; Matthew Gerlach <[email protected]>; Maxime Chevallier <[email protected]>; Maxime Coquelin <[email protected]>; Michal Swiatkowski <[email protected]>; [email protected]; Oleksij Rempel <[email protected]>; Paolo Abeni <[email protected]>; Rohan G Thomas <[email protected]>; Shenwei Wang <[email protected]>; Simon Horman <[email protected]>; Song Yoong Siang <[email protected]>; Swathi K S <[email protected]>; Tiezhu Yang <[email protected]>; Vinod Koul <[email protected]>; Vladimir Oltean <[email protected]>; Vladimir Oltean <[email protected]>; Yu- Chun Lin <[email protected]>
Subject: [PATCH net-next v2 00/14] net: stmmac: phylink PCS conversionThis series is radical - it takes the brave step of ripping out much of the existing PCS support code and throwing it all away.
I have discussed the introduction of the
STMMAC_FLAG_HAS_INTEGRATED_PCSflag with Bartosz Golaszewski, and the conclusion I came to is that this is to workaround the breakage that I've been going on about concerning the phylink conversion for the last five or six years.The problem is that the stmmac PCS code manipulates the netif carrier state, which confuses phylink.
There is a way of testing this out on the Jetson Xavier NX platform as the "PCS" code paths can be exercised while in RGMII mode - because RGMII also has in-band status and the status register is shared with SGMII. Testing this out confirms my long held theory: the interrupt handler manipulates the netif carrier state before phylink gets a look-in, which means that the
mac_link_up()andmac_link_down()methods are never called, resulting in the device being non-functional.Moreover, on dwmac4 cores, ethtool reports incorrect information - despite having a full-duplex link, ethtool reports that it is half-dupex.
Thus, this code is completely broken - anyone using it will not have a functional platform, and thus it doesn't deserve to live any longer, especially as it's a thorn in phylink.
Rip all this out, leaving just the bare bones initialisation in place.
However, this is not the last of what's broken. We have this
hw->psinteger which is really not descriptive, and the DT property from which it comes from does little to help understand what's going on. Putting all the clues together:- early configuration of the GMAC configuration register for the speed.
- setting the SGMII rate adapter layer to take its speed from the GMAC configuration register.Lastly, setting the transmit enable (TE) bit, which is a typo that puts the nail in the coffin of this code. It should be the transmit configuration (TC) bit. Given that when the link comes up, phylink will call
mac_link_up()which will overwrite the speed in the GMAC configuration register, the only part of this that is functional is changing where the SGMII rate adapter layer gets its speed from, which is a boolean.From what I've found so far, everyone who sets the snps,ps-speed property which configures this mode also configures a fixed link, so the pre- configuration is unnecessary - the link will come up anyway.
So, this series rips that out the preconfiguration as well, and replaces
hw->pswith a booleanhw->reverse_sgmii_enableflag.We then move the sole PCS configuration into a
phylink_pcsinstance, which configures the PCS control register in the same way as is done during the probe function.Thus, we end up with much easier and simpler conversion to phylink PCS than previous attempts.
Even so, this still results in inband mode always being enabled at the moment in the new .pcs_config() method to reflect what the probe function was doing. The next stage will be to change that to allow phylink to correctly configure the PCS. This needs fixing to allow platform glue maintainers who are currently blocked to progress.
Please note, however, that this has not been tested with any SGMII platform.
I've tried to get as many people into the Cc list with
get_maintainers, I hope that's sufficient to get enough eyeballs on this.Changes since RFC:
- new patch (7) to remove RGMII "pcs" mode
- new patch (8) to move reverse "pcs" mode tostmmac_check_pcs_mode()
- new patch (9) to simplify the code moved in the previous patch
- new patch (10) to rename the confusinghw->psto something more understandable.
- new patch (11) to shut up inappropriate complaints about "snps,ps-speed" being invalid.
- new patch (13) to add a MAC .pcs_init method, which will only be called when core has PCS present.
- modify patch 14 to use this newpcs_initmethod.Despite getting a couple of responses to the RFC series posted in September, I have had nothing testing this on hardware. I have tested this on the Jetson Xavier NX, which included trial runs with enabling the RGMII "pcs" mode, hence the new patches that rip out this mode. I have come to the conclusion that the only way to get stmmac changes tested is to get them merged into net-next, thereby forcing people to have to run with them... and we'll deal with any fallout later.
Changes since v1:
- added Andrew's reviewed-bys
- added additional comments to patch 3, 11 and 14. No code changes.drivers/net/ethernet/stmicro/stmmac/Makefile | 2 +- drivers/net/ethernet/stmicro/stmmac/common.h | 5 +- .../ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c | 6 +- drivers/net/ethernet/stmicro/stmmac/dwmac1000.h | 6 +- .../net/ethernet/stmicro/stmmac/dwmac1000_core.c | 65 ++--------------- ---- drivers/net/ethernet/stmicro/stmmac/dwmac4.h | 3 +- drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c | 66 ++--------------- ---- .../net/ethernet/stmicro/stmmac/dwxgmac2_core.c | 25 +------- drivers/net/ethernet/stmicro/stmmac/hwif.h | 4 +-
Quoted 9 lines
drivers/net/ethernet/stmicro/stmmac/stmmac.h | 4 ++ .../net/ethernet/stmicro/stmmac/stmmac_ethtool.c | 68 +---------------- ----- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 24 ++++---- drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.c | 47 +++++++++++++++ drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h | 23 ++++++-- include/linux/stmmac.h | 1 - 15 files changed, 104 insertions(+), 245 deletions(-)
Although RZ/V2H doesn't have PCS, I tested this on Renesas RZ/V2H EVK and found no regressions.
Cheers,
Prabhakar
Quoted 8 lines
--
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 80Mbps down 10Mbps up. Decent connectivity at last!--
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 80Mbps down 10Mbps up. Decent connectivity at last!
This series was applied to netdev/net-next.git (main) by Jakub Kicinski <[email protected]>:
Hello:
This series was applied to netdev/net-next.git (main) by Jakub Kicinski <[email protected]>:
Quoted 9 lines · On Thu, 16 Oct 2025 15:35:22 +0100 you wrote:
This series is radical - it takes the brave step of ripping out much of the existing PCS support code and throwing it all away.
I have discussed the introduction of the
STMMAC_FLAG_HAS_INTEGRATED_PCSflag with Bartosz Golaszewski, and the conclusion I came to is that this is to workaround the breakage that I've been going on about concerning the phylink conversion for the last five or six years.[...]
Here is the summary with links:
- [net-next,v2,01/14] net: stmmac: remove broken PCS code
https://git.kernel.org/netdev/net-next/c/813882ae2275 - [net-next,v2,02/14] net: stmmac: remove xstats.pcs_* members https://git.kernel.org/netdev/net-next/c/14f74bc6dc69 - [net-next,v2,03/14] net: stmmac: remove SGMII/RGMII/SMII interrupt handling https://git.kernel.org/netdev/net-next/c/2e2c878a3141 - [net-next,v2,04/14] net: stmmac: remove PCS "mode" pause handling https://git.kernel.org/netdev/net-next/c/ebc5d656b78c - [net-next,v2,05/14] net: stmmac: remove unused PCS loopback support https://git.kernel.org/netdev/net-next/c/19064a58bd3c - [net-next,v2,06/14] net: stmmac: remove hw->ps xxx_core_init() hardware setup https://git.kernel.org/netdev/net-next/c/aa1b6775aef7 - [net-next,v2,07/14] net: stmmac: remove RGMII "pcs" mode https://git.kernel.org/netdev/net-next/c/70589b05a03e - [net-next,v2,08/14] net: stmmac: move reverse-"pcs" mode setup to stmmac_check_pcs_mode() https://git.kernel.org/netdev/net-next/c/c7b0d7874de0 - [net-next,v2,09/14] net: stmmac: simplify stmmac_check_pcs_mode() https://git.kernel.org/netdev/net-next/c/412d5f32cb36 - [net-next,v2,10/14] net: stmmac: hw->ps becomes hw->reverse_sgmii_enable https://git.kernel.org/netdev/net-next/c/5d1e7621f869 - [net-next,v2,11/14] net: stmmac: do not require snps,ps-speed for SGMII https://git.kernel.org/netdev/net-next/c/5c61db08d9ae - [net-next,v2,12/14] net: stmmac: only call stmmac_pcs_ctrl_ane() for integrated SGMII PCS https://git.kernel.org/netdev/net-next/c/045d7e5727c4 - [net-next,v2,13/14] net: stmmac: provide PCS initialisation hook https://git.kernel.org/netdev/net-next/c/237e54caeaef - [net-next,v2,14/14] net: stmmac: convert to phylink PCS support https://git.kernel.org/netdev/net-next/c/2c81f3357136
You are awesome, thank you!
Signature · 2 lines
Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/patchwork/pwbot.html