RE: [PATCH v7 2/2] net: stmmac: dwc-qos: Add FSD EQoS support
From: Swathi K S <hidden>
Date: 2025-02-21 11:26:42
Also in:
linux-arm-kernel, linux-devicetree, lkml
-----Original Message----- From: Russell King (Oracle) <linux@armlinux.org.uk> Sent: 20 February 2025 18:10 To: Swathi K S <redacted> Cc: krzk+dt@kernel.org; andrew+netdev@lunn.ch; davem@davemloft.net; edumazet@google.com; kuba@kernel.org; pabeni@redhat.com; robh@kernel.org; conor+dt@kernel.org; richardcochran@gmail.com; mcoquelin.stm32@gmail.com; alexandre.torgue@foss.st.com; netdev@vger.kernel.org; devicetree@vger.kernel.org; linux-stm32@st-md- mailman.stormreply.com; linux-arm-kernel@lists.infradead.org; linux- kernel@vger.kernel.org; pankaj.dubey@samsung.com; ravi.patel@samsung.com; gost.dev@samsung.com Subject: Re: [PATCH v7 2/2] net: stmmac: dwc-qos: Add FSD EQoS support On Thu, Feb 20, 2025 at 10:07:12AM +0530, Swathi K S wrote:quoted
+static int fsd_eqos_probe(struct platform_device *pdev, + struct plat_stmmacenet_data *data, + struct stmmac_resources *res) +{ + struct clk *clk_rx1 = NULL; + struct clk *clk_rx2 = NULL; + + for (int i = 0; i < data->num_clks; i++) { + if (strcmp(data->clks[i].id, "slave_bus") == 0) + data->stmmac_clk = data->clks[i].clk; + else if (strcmp(data->clks[i].id, "eqos_rxclk_mux") == 0) + clk_rx1 = data->clks[i].clk; + else if (strcmp(data->clks[i].id, "eqos_phyrxclk") == 0) + clk_rx2 = data->clks[i].clk; + } + + /* Eth0 RX clock doesn't support MUX */ + if (clk_rx1) + clk_set_parent(clk_rx1, clk_rx2);Isn't there support in DT for automatically setting the clock tree? See https://protect2.fireeye.com/v1/url?k=f0089f78-90ea0225-f0091437- 000babd9f1ba-cf835b8b94ccd94a&q=1&e=4ae794ec-f443-4d77-aee4- 449f53a3a1a4&u=https%3A%2F%2Fgithub.com%2Fdevicetree-org%2Fdt- schema%2Fblob%2Fmain%2Fdtschema%2Fschemas%2Fclock%2Fclock.yaml %23L24 Also, I think a cleanup like the below (sorry, it's on top of other
patches I'm
working on at the moment but could be rebased) would make sense. With both of these, this should mean that your changes amount to: 1. making data->probe optional 2. providing a dwc_eth_dwmac_data structure that has .stmmac_clk_name filled in 3. adding your compatible to the match data with a pointer to the above structure.
Hi Russell, Thanks for your input. Will implement this in v8. But I could not find your patch 'net: stmmac: clean up clock initialisation' in mailing list Could you point me to that? Or do you want me to integrate the below changes into my patch series and post? Please let me know - Swathi
In other words, support for your device becomes just a matter of adding
data
structures rather than a chunk of extra code. Thanks. 8<==== From: "Russell King (Oracle)" <redacted> Subject: [PATCH net-next] net: stmmac: clean up clock initialisation Clean up the clock initialisation by providing a helper to find a named
clock in
quoted hunk ↗ jump to hunk
the bulk clocks, and provide the name of the stmmac clock in match data so we can locate the stmmac clock in generic code. Signed-off-by: Russell King (Oracle) <redacted> --- .../stmicro/stmmac/dwmac-dwc-qos-eth.c | 32 +++++++++++-------- 1 file changed, 18 insertions(+), 14 deletions(-)diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.cb/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c index 581c0b40db57..8e343ab7a7e2 100644--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c@@ -34,6 +34,16 @@ struct tegra_eqos { struct gpio_desc *reset; }; +static struct clk *dwc_eth_find_clk(struct plat_stmmacenet_data
*plat_dat,
quoted hunk ↗ jump to hunk
+ const char *name) +{ + for (int i = 0; i < plat_dat->num_clks; i++) + if (strcmp(plat_dat->clks[i].id, name) == 0) + return plat_dat->clks[i].clk; + + return 0; +} + static int dwc_eth_dwmac_config_dt(struct platform_device *pdev, struct plat_stmmacenet_data *plat_dat) {@@ -120,12 +130,7 @@ static int dwc_qos_probe(struct platform_device*pdev, struct plat_stmmacenet_data *plat_dat, struct stmmac_resources *stmmac_res) { - for (int i = 0; i < plat_dat->num_clks; i++) { - if (strcmp(plat_dat->clks[i].id, "apb_pclk") == 0) - plat_dat->stmmac_clk = plat_dat->clks[i].clk; - else if (strcmp(plat_dat->clks[i].id, "phy_ref_clk") == 0) - plat_dat->pclk = plat_dat->clks[i].clk; - } + plat_dat->pclk = dwc_eth_find_clk(plat_dat, "phy_ref_clk"); return 0; }@@ -230,18 +235,12 @@ static int tegra_eqos_probe(struct platform_device*pdev, eqos->dev = &pdev->dev; eqos->regs = res->addr; + eqos->clk_slave = data->stmmac_clk; if (!is_of_node(dev->fwnode)) goto bypass_clk_reset_gpio; - for (int i = 0; i < data->num_clks; i++) { - if (strcmp(data->clks[i].id, "slave_bus") == 0) { - eqos->clk_slave = data->clks[i].clk; - data->stmmac_clk = eqos->clk_slave; - } else if (strcmp(data->clks[i].id, "tx") == 0) { - data->clk_tx_i = data->clks[i].clk; - } - } + data->clk_tx_i = dwc_eth_find_clk(data, "tx"); eqos->reset = devm_gpiod_get(&pdev->dev, "phy-reset", GPIOD_OUT_HIGH); if (IS_ERR(eqos->reset)) {@@ -306,15 +305,18 @@ struct dwc_eth_dwmac_data { struct plat_stmmacenet_data *data, struct stmmac_resources *res); void (*remove)(struct platform_device *pdev); + const char *stmmac_clk_name; }; static const struct dwc_eth_dwmac_data dwc_qos_data = { .probe = dwc_qos_probe, + .stmmac_clk_name = "apb_pclk", }; static const struct dwc_eth_dwmac_data tegra_eqos_data = { .probe = tegra_eqos_probe, .remove = tegra_eqos_remove, + .stmmac_clk_name = "slave_bus", }; static int dwc_eth_dwmac_probe(struct platform_device *pdev) @@ -354,6 +356,8 @@ static int dwc_eth_dwmac_probe(struct platform_device *pdev) if (ret) return dev_err_probe(&pdev->dev, ret, "Failed to enableclocks\n"); + data->stmmac_clk = dwc_eth_find_clk(plat_dat, data-quoted
stmmac_clk_name);+ ret = data->probe(pdev, plat_dat, &stmmac_res); if (ret < 0) { dev_err_probe(&pdev->dev, ret, "failed to probe subdriver\n"); -- 2.30.2 -- RMK's Patch system: https://protect2.fireeye.com/v1/url?k=bd23955d- ddc10800-bd221e12-000babd9f1ba-890d84739eaced1c&q=1&e=4ae794ec- f443-4d77-aee4- 449f53a3a1a4&u=https%3A%2F%2Fwww.armlinux.org.uk%2Fdeveloper%2F patches%2F FTTP is here! 80Mbps down 10Mbps up. Decent connectivity at last!