Skip to contentSkip to conversation

net: bcmgenet: Clean up after ACPI enablement

PATCH In review Andy Shevchenko 4 participants 10 messages Apr 20, 2020 – Apr 21, 2020

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

v2 cover net: bcmgenet: Clean up after ACPI enablement

ACPI enablement series had missed some clean ups that would have been done at the same time. Here are these bits.

In v2:
- return dev_dbg() calls to avoid spamming logs when probe is deferred (Florian)
- added Ack (Florian)
- combined two, earlier sent, series together
- added couple more patches

Andy Shevchenko (5):
net: bcmgenet: Drop ACPI_PTR() to avoid compiler warning
net: bcmgenet: Drop useless OF code
net: bcmgenet: Use devm_clk_get_optional() to get the clocks
net: bcmgenet: Use get_unligned_beXX() and put_unaligned_beXX()
net: bcmgenet: Drop too many parentheses in bcmgenet_probe()

 .../net/ethernet/broadcom/genet/bcmgenet.c    | 49 +++++++------------
 1 file changed, 18 insertions(+), 31 deletions(-)
Signature · 1 line
2.26.1
#1 lore raw

PATCH v2 1/5 net: bcmgenet: Drop ACPI_PTR() to avoid compiler warning

When compiled with CONFIG_ACPI=n, ACPI_PTR() will be no-op, and thus genet_acpi_match table defined, but not used. Compiler is not happy about such data. Drop ACPI_PTR() for good.

---
 drivers/net/ethernet/broadcom/genet/bcmgenet.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/broadcom/genet/bcmgenet.c +1 -1
index d975338bf78d..9f2f0e681656 100644
@@ -3740,7 +3740,7 @@ static struct platform_driver bcmgenet_driver = {
 		.name	= "bcmgenet",
 		.of_match_table = bcmgenet_match,
 		.pm	= &bcmgenet_pm_ops,
-		.acpi_match_table = ACPI_PTR(genet_acpi_match),
+		.acpi_match_table = genet_acpi_match,
 	},
 };
 module_platform_driver(bcmgenet_driver);
Signature · 1 line
2.26.1
#2 lore raw

PATCH v2 2/5 net: bcmgenet: Drop useless OF code

In reply to Andy Shevchenko

There is nothing which needs a set of OF headers, followed by redundant OF node ID check. Drop them for good.

---
 drivers/net/ethernet/broadcom/genet/bcmgenet.c | 13 -------------
 1 file changed, 13 deletions(-)
drivers/net/ethernet/broadcom/genet/bcmgenet.c +0 -13
index 9f2f0e681656..ef275db018f7 100644
@@ -23,11 +23,6 @@
 #include <linux/dma-mapping.h>
 #include <linux/pm.h>
 #include <linux/clk.h>
-#include <linux/of.h>
-#include <linux/of_address.h>
-#include <linux/of_irq.h>
-#include <linux/of_net.h>
-#include <linux/of_platform.h>
 #include <net/arp.h>
 
 #include <linux/mii.h>
@@ -3417,8 +3412,6 @@ MODULE_DEVICE_TABLE(of, bcmgenet_match);
 static int bcmgenet_probe(struct platform_device *pdev)
 {
 	struct bcmgenet_platform_data *pd = pdev->dev.platform_data;
-	struct device_node *dn = pdev->dev.of_node;
-	const struct of_device_id *of_id = NULL;
 	const struct bcmgenet_plat_data *pdata;
 	struct bcmgenet_priv *priv;
 	struct net_device *dev;
@@ -3433,12 +3426,6 @@ static int bcmgenet_probe(struct platform_device *pdev)
 		return -ENOMEM;
 	}
 
-	if (dn) {
-		of_id = of_match_node(bcmgenet_match, dn);
-		if (!of_id)
-			return -EINVAL;
-	}
-
 	priv = netdev_priv(dev);
 	priv->irq0 = platform_get_irq(pdev, 0);
 	if (priv->irq0 < 0) {
Signature · 1 line
2.26.1
#4 lore raw

PATCH v2 3/5 net: bcmgenet: Use devm_clk_get_optional() to get the clocks

In reply to Andy Shevchenko

Conversion to devm_clk_get_optional() makes it explicit that clocks are optional. This change allows to handle deferred probe in case clocks are defined, but not yet probed. Due to above changes bail out in error case.

While here, check potential error when enable main clock.

---
 .../net/ethernet/broadcom/genet/bcmgenet.c    | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)
drivers/net/ethernet/broadcom/genet/bcmgenet.c +12 -7
index ef275db018f7..86666e9ab3e7 100644
@@ -3487,13 +3487,16 @@ static int bcmgenet_probe(struct platform_device *pdev)
 		priv->dma_max_burst_length = DMA_MAX_BURST_LENGTH;
 	}
 
-	priv->clk = devm_clk_get(&priv->pdev->dev, "enet");
+	priv->clk = devm_clk_get_optional(&priv->pdev->dev, "enet");
 	if (IS_ERR(priv->clk)) {
 		dev_dbg(&priv->pdev->dev, "failed to get enet clock\n");
-		priv->clk = NULL;
+		err = PTR_ERR(priv->clk);
+		goto err;
 	}
 
-	clk_prepare_enable(priv->clk);
+	err = clk_prepare_enable(priv->clk);
+	if (err)
+		goto err;
 
 	bcmgenet_set_hw_params(priv);
 
@@ -3511,16 +3514,18 @@ static int bcmgenet_probe(struct platform_device *pdev)
 	priv->rx_buf_len = RX_BUF_LENGTH;
 	INIT_WORK(&priv->bcmgenet_irq_work, bcmgenet_irq_task);
 
-	priv->clk_wol = devm_clk_get(&priv->pdev->dev, "enet-wol");
+	priv->clk_wol = devm_clk_get_optional(&priv->pdev->dev, "enet-wol");
 	if (IS_ERR(priv->clk_wol)) {
 		dev_dbg(&priv->pdev->dev, "failed to get enet-wol clock\n");
-		priv->clk_wol = NULL;
+		err = PTR_ERR(priv->clk_wol);
+		goto err;
 	}
 
-	priv->clk_eee = devm_clk_get(&priv->pdev->dev, "enet-eee");
+	priv->clk_eee = devm_clk_get_optional(&priv->pdev->dev, "enet-eee");
 	if (IS_ERR(priv->clk_eee)) {
 		dev_dbg(&priv->pdev->dev, "failed to get enet-eee clock\n");
-		priv->clk_eee = NULL;
+		err = PTR_ERR(priv->clk_eee);
+		goto err;
 	}
 
 	/* If this is an internal GPHY, power it on now, before UniMAC is
Signature · 1 line
2.26.1
#5 lore raw

PATCH v2 4/5 net: bcmgenet: Use get_unligned_beXX() and put_unaligned_beXX()

In reply to Andy Shevchenko

It's convenient to use get_unligned_beXX() and put_unaligned_beXX() helpers to get or set MAC instead of open-coded variants.

---
 drivers/net/ethernet/broadcom/genet/bcmgenet.c | 13 ++++---------
 1 file changed, 4 insertions(+), 9 deletions(-)
drivers/net/ethernet/broadcom/genet/bcmgenet.c +4 -9
index 86666e9ab3e7..2c9881032a24 100644
@@ -2702,9 +2702,8 @@ static void bcmgenet_umac_reset(struct bcmgenet_priv *priv)
 static void bcmgenet_set_hw_addr(struct bcmgenet_priv *priv,
 				 unsigned char *addr)
 {
-	bcmgenet_umac_writel(priv, (addr[0] << 24) | (addr[1] << 16) |
-			(addr[2] << 8) | addr[3], UMAC_MAC0);
-	bcmgenet_umac_writel(priv, (addr[4] << 8) | addr[5], UMAC_MAC1);
+	bcmgenet_umac_writel(priv, get_unaligned_be32(&addr[0]), UMAC_MAC0);
+	bcmgenet_umac_writel(priv, get_unaligned_be16(&addr[4]), UMAC_MAC1);
 }
 
 static void bcmgenet_get_hw_addr(struct bcmgenet_priv *priv,
@@ -2713,13 +2712,9 @@ static void bcmgenet_get_hw_addr(struct bcmgenet_priv *priv,
 	u32 addr_tmp;
 
 	addr_tmp = bcmgenet_umac_readl(priv, UMAC_MAC0);
-	addr[0] = addr_tmp >> 24;
-	addr[1] = (addr_tmp >> 16) & 0xff;
-	addr[2] = (addr_tmp >>	8) & 0xff;
-	addr[3] = addr_tmp & 0xff;
+	put_unaligned_be32(addr_tmp, &addr[0]);
 	addr_tmp = bcmgenet_umac_readl(priv, UMAC_MAC1);
-	addr[4] = (addr_tmp >> 8) & 0xff;
-	addr[5] = addr_tmp & 0xff;
+	put_unaligned_be16(addr_tmp, &addr[4]);
 }
 
 /* Returns a reusable dma control register value */
Signature · 1 line
2.26.1
#6 lore raw

PATCH v2 5/5 net: bcmgenet: Drop too many parentheses in bcmgenet_probe()

In reply to Andy Shevchenko

No need to have parentheses around plain pointer variable or negation operator. Drop them for good.

---
 drivers/net/ethernet/broadcom/genet/bcmgenet.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/broadcom/genet/bcmgenet.c +1 -1
index 2c9881032a24..20aba79becce 100644
@@ -3529,7 +3529,7 @@ static int bcmgenet_probe(struct platform_device *pdev)
 	if (device_get_phy_mode(&pdev->dev) == PHY_INTERFACE_MODE_INTERNAL)
 		bcmgenet_power_up(priv, GENET_POWER_PASSIVE);
 
-	if ((pd) && (!IS_ERR_OR_NULL(pd->mac_address)))
+	if (pd && !IS_ERR_OR_NULL(pd->mac_address))
 		ether_addr_copy(dev->dev_addr, pd->mac_address);
 	else
 		if (!device_get_mac_address(&pdev->dev, dev->dev_addr, ETH_ALEN))
Signature · 1 line
2.26.1
maintainer
#8 lore raw
In reply to Andy Shevchenko

From: Andy Shevchenko <[email protected]>
Date: Tue, 21 Apr 2020 00:51:16 +0300

Quoted 8 lines

ACPI enablement series had missed some clean ups that would have been done at the same time. Here are these bits.

In v2:
- return dev_dbg() calls to avoid spamming logs when probe is deferred (Florian)
- added Ack (Florian)
- combined two, earlier sent, series together
- added couple more patches

Series applied to net-next, thanks.

Tested-by
#9 lore raw
Quoted 6 lines · David Miller

From: Andy Shevchenko <[email protected]>
Date: Tue, 21 Apr 2020 00:51:16 +0300

Quoted 3 lines

ACPI enablement series had missed some clean ups that would have been done at the same time. Here are these bits.

Quoted 8 lines
Quoted 5 lines

In v2:
- return dev_dbg() calls to avoid spamming logs when probe is deferred (Florian)
- added Ack (Florian)
- combined two, earlier sent, series together
- added couple more patches

Series applied to net-next, thanks.

Well, you guys are two fast for me.. <chuckle> Or rather the couple hours it took to build the kernel on the rpi4.

Anyway, the changes looks like fine, and boots fine in ACPI mode.

Its probably to late for this but anyway:

Thanks,