Skip to contentSkip to conversation

net: stmmac: Fixes for -net

PATCH Superseded Jose Abreu 3 participants 12 messages Sep 30, 2019 – Oct 1, 2019

Subsystems (4)

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore

Jose Abreu

author
#0 lore raw

v2 cover net: stmmac: Fixes for -net

Misc fixes for -net tree. More info in commit logs.

v2 is just a rebase of v1 against -net and we added a new patch (09/09) to fix RSS feature.

---
Cc: Giuseppe Cavallaro <[email protected]>
Cc: Alexandre Torgue <[email protected]>
Cc: Jose Abreu <[email protected]>
Cc: "David S. Miller" <[email protected]>
Cc: Maxime Coquelin <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
---

Jose Abreu (9):
net: stmmac: xgmac: Not all Unicast addresses may be available
net: stmmac: xgmac: Detect Hash Table size dinamically
net: stmmac: selftests: Always use max DMA size in Jumbo Test
net: stmmac: dwmac4: Always update the MAC Hash Filter
net: stmmac: Correctly take timestamp for PTPv2
net: stmmac: Do not stop PHY if WoL is enabled
net: stmmac: xgmac: Disable the Timestamp interrupt by default
net: stmmac: xgmac: Fix RSS not writing all Keys to HW
net: stmmac: xgmac: Fix RSS writing wrong keys

 drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c     | 13 +++++++------
 drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h        |  3 ++-
 drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c   |  9 +++++----
 drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c    |  1 +
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c     | 19 +++++++++++++------
 .../net/ethernet/stmicro/stmmac/stmmac_selftests.c    |  4 ----
 6 files changed, 28 insertions(+), 21 deletions(-)
Signature · 1 line
2.7.4
author
#1 lore raw

PATCH v2 1/9 net: stmmac: xgmac: Not all Unicast addresses may be available

Some setups may not have all Unicast addresses filters available. Let's check this before trying to setup filters.

Fixes 0efedbf11f07 ("net: stmmac: xgmac: Fix XGMAC selftests")

---
Cc: Giuseppe Cavallaro <[email protected]>
Cc: Alexandre Torgue <[email protected]>
Cc: Jose Abreu <[email protected]>
Cc: "David S. Miller" <[email protected]>
Cc: Maxime Coquelin <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]

---
 drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c +1 -1
index 2b277b2c586b..6d8ac2ef4fc2 100644
@@ -472,7 +472,7 @@ static void dwxgmac2_set_filter(struct mac_device_info *hw,
 	dwxgmac2_set_mchash(ioaddr, mc_filter, mcbitslog2);
 
 	/* Handle multiple unicast addresses */
-	if (netdev_uc_count(dev) > XGMAC_ADDR_MAX) {
+	if (netdev_uc_count(dev) > hw->unicast_filter_entries) {
 		value |= XGMAC_FILTER_PR;
 	} else {
 		struct netdev_hw_addr *ha;
Signature · 1 line
2.7.4
author
#2 lore raw

PATCH v2 2/9 net: stmmac: xgmac: Detect Hash Table size dinamically

In reply to Jose Abreu

Since commit b8ef7020d6e5 ("net: stmmac: add support for hash table size 128/256 in dwmac4"), we can detect the Hash Table dinamically.

Let's implement this feature in XGMAC cores and fix possible setups that don't support the maximum size for Hash Table.

---
Cc: Giuseppe Cavallaro <[email protected]>
Cc: Alexandre Torgue <[email protected]>
Cc: Jose Abreu <[email protected]>
Cc: "David S. Miller" <[email protected]>
Cc: Maxime Coquelin <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]

---
 drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h     | 1 +
 drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c | 1 +
 2 files changed, 2 insertions(+)
drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h +1 -0
index 5923ca62d793..f7eb06f8fb37 100644
@@ -122,6 +122,7 @@
 #define XGMAC_HWFEAT_GMIISEL		BIT(1)
 #define XGMAC_HW_FEATURE1		0x00000120
 #define XGMAC_HWFEAT_L3L4FNUM		GENMASK(30, 27)
+#define XGMAC_HWFEAT_HASHTBLSZ		GENMASK(25, 24)
 #define XGMAC_HWFEAT_RSSEN		BIT(20)
 #define XGMAC_HWFEAT_TSOEN		BIT(18)
 #define XGMAC_HWFEAT_SPHEN		BIT(17)
drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c +1 -0
index 53c4a40d8386..965cbe3e6f51 100644
@@ -380,6 +380,7 @@ static void dwxgmac2_get_hw_feature(void __iomem *ioaddr,
 	/* MAC HW feature 1 */
 	hw_cap = readl(ioaddr + XGMAC_HW_FEATURE1);
 	dma_cap->l3l4fnum = (hw_cap & XGMAC_HWFEAT_L3L4FNUM) >> 27;
+	dma_cap->hash_tb_sz = (hw_cap & XGMAC_HWFEAT_HASHTBLSZ) >> 24;
 	dma_cap->rssen = (hw_cap & XGMAC_HWFEAT_RSSEN) >> 20;
 	dma_cap->tsoen = (hw_cap & XGMAC_HWFEAT_TSOEN) >> 18;
 	dma_cap->sphen = (hw_cap & XGMAC_HWFEAT_SPHEN) >> 17;
Signature · 1 line
2.7.4
author
#3 lore raw

PATCH v2 3/9 net: stmmac: selftests: Always use max DMA size in Jumbo Test

In reply to Jose Abreu

Although some XGMAC setups support frames larger than DMA size, some of them may not. As we can't know before-hand which ones support let's use the maximum DMA buffer size in the Jumbo Tests.

User can always reconfigure the MTU to achieve larger frames.

Fixes 427849e8c37f ("net: stmmac: selftests: Add Jumbo Frame tests")

---
Cc: Giuseppe Cavallaro <[email protected]>
Cc: Alexandre Torgue <[email protected]>
Cc: Jose Abreu <[email protected]>
Cc: "David S. Miller" <[email protected]>
Cc: Maxime Coquelin <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]

---
 drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c | 4 ----
 1 file changed, 4 deletions(-)
drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c +0 -4
index 5f66f6161629..cc76a42c7466 100644
@@ -1564,10 +1564,6 @@ static int __stmmac_test_jumbo(struct stmmac_priv *priv, u16 queue)
 	struct stmmac_packet_attrs attr = { };
 	int size = priv->dma_buf_sz;
 
-	/* Only XGMAC has SW support for multiple RX descs in same packet */
-	if (priv->plat->has_xgmac)
-		size = priv->dev->max_mtu;
-
 	attr.dst = priv->dev->dev_addr;
 	attr.max_size = size - ETH_FCS_LEN;
 	attr.queue_mapping = queue;
Signature · 1 line
2.7.4
author
#4 lore raw

PATCH v2 4/9 net: stmmac: dwmac4: Always update the MAC Hash Filter

In reply to Jose Abreu

We need to always update the MAC Hash Filter so that previous entries are invalidated.

Found out while running stmmac selftests.

Fixes b8ef7020d6e5 ("net: stmmac: add support for hash table size 128/256 in dwmac4")

---
Cc: Giuseppe Cavallaro <[email protected]>
Cc: Alexandre Torgue <[email protected]>
Cc: Jose Abreu <[email protected]>
Cc: "David S. Miller" <[email protected]>
Cc: Maxime Coquelin <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]

---
 drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)
drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +7 -6
index 9b4b5f69fc02..2cb9c53f93b8 100644
@@ -401,8 +401,11 @@ static void dwmac4_set_filter(struct mac_device_info *hw,
 	int numhashregs = (hw->multicast_filter_bins >> 5);
 	int mcbitslog2 = hw->mcast_bits_log2;
 	unsigned int value;
+	u32 mc_filter[8];
 	int i;
 
+	memset(mc_filter, 0, sizeof(mc_filter));
+
 	value = readl(ioaddr + GMAC_PACKET_FILTER);
 	value &= ~GMAC_PACKET_FILTER_HMC;
 	value &= ~GMAC_PACKET_FILTER_HPF;
@@ -416,16 +419,13 @@ static void dwmac4_set_filter(struct mac_device_info *hw,
 		/* Pass all multi */
 		value |= GMAC_PACKET_FILTER_PM;
 		/* Set all the bits of the HASH tab */
-		for (i = 0; i < numhashregs; i++)
-			writel(0xffffffff, ioaddr + GMAC_HASH_TAB(i));
+		memset(mc_filter, 0xff, sizeof(mc_filter));
 	} else if (!netdev_mc_empty(dev)) {
 		struct netdev_hw_addr *ha;
-		u32 mc_filter[8];
 
 		/* Hash filter for multicast */
 		value |= GMAC_PACKET_FILTER_HMC;
 
-		memset(mc_filter, 0, sizeof(mc_filter));
 		netdev_for_each_mc_addr(ha, dev) {
 			/* The upper n bits of the calculated CRC are used to
 			 * index the contents of the hash table. The number of
@@ -440,10 +440,11 @@ static void dwmac4_set_filter(struct mac_device_info *hw,
 			 */
 			mc_filter[bit_nr >> 5] |= (1 << (bit_nr & 0x1f));
 		}
-		for (i = 0; i < numhashregs; i++)
-			writel(mc_filter[i], ioaddr + GMAC_HASH_TAB(i));
 	}
 
+	for (i = 0; i < numhashregs; i++)
+		writel(mc_filter[i], ioaddr + GMAC_HASH_TAB(i));
+
 	value |= GMAC_PACKET_FILTER_HPF;
 
 	/* Handle multiple unicast addresses */
Signature · 1 line
2.7.4
author
#5 lore raw

PATCH v2 5/9 net: stmmac: Correctly take timestamp for PTPv2

In reply to Jose Abreu

The case for PTPV2_EVENT requires event packets to be captured so add this setting to the list of enabled captures.

Fixes 891434b18ec0 ("stmmac: add IEEE PTPv1 and PTPv2 support.")

---
Cc: Giuseppe Cavallaro <[email protected]>
Cc: Alexandre Torgue <[email protected]>
Cc: Jose Abreu <[email protected]>
Cc: "David S. Miller" <[email protected]>
Cc: Maxime Coquelin <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]

---
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 1 +
 1 file changed, 1 insertion(+)
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +1 -0
index d3232738fb25..31a237ec73bc 100644
@@ -629,6 +629,7 @@ static int stmmac_hwtstamp_set(struct net_device *dev, struct ifreq *ifr)
 			config.rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT;
 			ptp_v2 = PTP_TCR_TSVER2ENA;
 			snap_type_sel = PTP_TCR_SNAPTYPSEL_1;
+			ts_event_en = PTP_TCR_TSEVNTENA;
 			ptp_over_ipv4_udp = PTP_TCR_TSIPV4ENA;
 			ptp_over_ipv6_udp = PTP_TCR_TSIPV6ENA;
 			ptp_over_ethernet = PTP_TCR_TSIPENA;
Signature · 1 line
2.7.4
author
#6 lore raw

PATCH v2 6/9 net: stmmac: Do not stop PHY if WoL is enabled

In reply to Jose Abreu

If WoL is enabled we can't really stop the PHY, otherwise we will not receive the WoL packet. Fix this by telling phylink that only the MAC is down and only stop the PHY if WoL is not enabled.

Fixes 74371272f97f ("net: stmmac: Convert to phylink and remove phylib logic")

---
Cc: Giuseppe Cavallaro <[email protected]>
Cc: Alexandre Torgue <[email protected]>
Cc: Jose Abreu <[email protected]>
Cc: "David S. Miller" <[email protected]>
Cc: Maxime Coquelin <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]

---
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +12 -6
index 31a237ec73bc..843d53e084b7 100644
@@ -4718,9 +4718,7 @@ int stmmac_suspend(struct device *dev)
 
 	mutex_lock(&priv->lock);
 
-	rtnl_lock();
-	phylink_stop(priv->phylink);
-	rtnl_unlock();
+	phylink_mac_change(priv->phylink, false);
 
 	netif_device_detach(ndev);
 	stmmac_stop_all_queues(priv);
@@ -4735,6 +4733,10 @@ int stmmac_suspend(struct device *dev)
 		stmmac_pmt(priv, priv->hw, priv->wolopts);
 		priv->irq_wake = 1;
 	} else {
+		rtnl_lock();
+		phylink_stop(priv->phylink);
+		rtnl_unlock();
+
 		stmmac_mac_set(priv, priv->ioaddr, false);
 		pinctrl_pm_select_sleep_state(priv->device);
 		/* Disable clock in case of PWM is off */
@@ -4825,9 +4827,13 @@ int stmmac_resume(struct device *dev)
 
 	stmmac_start_all_queues(priv);
 
-	rtnl_lock();
-	phylink_start(priv->phylink);
-	rtnl_unlock();
+	if (!device_may_wakeup(priv->device)) {
+		rtnl_lock();
+		phylink_start(priv->phylink);
+		rtnl_unlock();
+	}
+
+	phylink_mac_change(priv->phylink, true);
 
 	mutex_unlock(&priv->lock);
 
Signature · 1 line
2.7.4
author
#7 lore raw

PATCH v2 7/9 net: stmmac: xgmac: Disable the Timestamp interrupt by default

In reply to Jose Abreu

We don't use it anyway as XGMAC only supports polling for timestamp (in current SW implementation). This greatly reduces the system load by reducing the number of interrupts.

Fixes 2142754f8b9c ("net: stmmac: Add MAC related callbacks for XGMAC2")

---
Cc: Giuseppe Cavallaro <[email protected]>
Cc: Alexandre Torgue <[email protected]>
Cc: Jose Abreu <[email protected]>
Cc: "David S. Miller" <[email protected]>
Cc: Maxime Coquelin <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]

---
 drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h +1 -1
index f7eb06f8fb37..99037386080a 100644
@@ -84,7 +84,7 @@
 #define XGMAC_TSIE			BIT(12)
 #define XGMAC_LPIIE			BIT(5)
 #define XGMAC_PMTIE			BIT(4)
-#define XGMAC_INT_DEFAULT_EN		(XGMAC_LPIIE | XGMAC_PMTIE | XGMAC_TSIE)
+#define XGMAC_INT_DEFAULT_EN		(XGMAC_LPIIE | XGMAC_PMTIE)
 #define XGMAC_Qx_TX_FLOW_CTRL(x)	(0x00000070 + (x) * 4)
 #define XGMAC_PT			GENMASK(31, 16)
 #define XGMAC_PT_SHIFT			16
Signature · 1 line
2.7.4
author
#8 lore raw

PATCH v2 8/9 net: stmmac: xgmac: Fix RSS not writing all Keys to HW

In reply to Jose Abreu

The sizeof(cfg->key) is != ARRAY_SIZE(cfg->key). Fix it. This warning is triggered when running with cc flag -Wsizeof-array-div.

Fixes 76067459c686 ("net: stmmac: Implement RSS and enable it in XGMAC core")

---
Cc: Giuseppe Cavallaro <[email protected]>
Cc: Alexandre Torgue <[email protected]>
Cc: Jose Abreu <[email protected]>
Cc: "David S. Miller" <[email protected]>
Cc: Maxime Coquelin <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: Nick Desaulniers <[email protected]>
Cc: Nathan Chancellor <[email protected]>

---
 drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c +1 -1
index 6d8ac2ef4fc2..4a1f52474dbc 100644
@@ -533,7 +533,7 @@ static int dwxgmac2_rss_configure(struct mac_device_info *hw,
 		return 0;
 	}
 
-	for (i = 0; i < (sizeof(cfg->key) / sizeof(u32)); i++) {
+	for (i = 0; i < (ARRAY_SIZE(cfg->key) / sizeof(u32)); i++) {
 		ret = dwxgmac2_rss_write_reg(ioaddr, true, i, cfg->key[i]);
 		if (ret)
 			return ret;
Signature · 1 line
2.7.4
author
#9 lore raw

PATCH v2 9/9 net: stmmac: xgmac: Fix RSS writing wrong keys

In reply to Jose Abreu

Commit b6b6cc9acd7b, changed the call to dwxgmac2_rss_write_reg() passing it the variable cfg->key[i].

As key is an u8 but we write 32 bits at a time we need to cast it into an u32 so that the correct key values are written. Notice that the for loop already takes this into account so we don't try to write past the keys size.

Fixes b6b6cc9acd7b ("net: stmmac: selftest: avoid large stack usage")

---
Cc: Giuseppe Cavallaro <[email protected]>
Cc: Alexandre Torgue <[email protected]>
Cc: Jose Abreu <[email protected]>
Cc: "David S. Miller" <[email protected]>
Cc: Maxime Coquelin <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: Arnd Bergmann <[email protected]>

---
 drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c +3 -2
index 4a1f52474dbc..5031398e612c 100644
@@ -523,8 +523,8 @@ static int dwxgmac2_rss_configure(struct mac_device_info *hw,
 				  struct stmmac_rss *cfg, u32 num_rxq)
 {
 	void __iomem *ioaddr = hw->pcsr;
+	u32 value, *key;
 	int i, ret;
-	u32 value;
 
 	value = readl(ioaddr + XGMAC_RSS_CTRL);
 	if (!cfg || !cfg->enable) {
@@ -533,8 +533,9 @@ static int dwxgmac2_rss_configure(struct mac_device_info *hw,
 		return 0;
 	}
 
+	key = (u32 *)cfg->key;
 	for (i = 0; i < (ARRAY_SIZE(cfg->key) / sizeof(u32)); i++) {
-		ret = dwxgmac2_rss_write_reg(ioaddr, true, i, cfg->key[i]);
+		ret = dwxgmac2_rss_write_reg(ioaddr, true, i, key[i]);
 		if (ret)
 			return ret;
 	}
Signature · 1 line
2.7.4
Reviewed-by on v2 9/9
#10 lore raw

Re: [PATCH v2 net 9/9] net: stmmac: xgmac: Fix RSS writing wrong keys

Quoted 8 lines · Jose Abreu

Commit b6b6cc9acd7b, changed the call to dwxgmac2_rss_write_reg() passing it the variable cfg->key[i].

As key is an u8 but we write 32 bits at a time we need to cast it into an u32 so that the correct key values are written. Notice that the for loop already takes this into account so we don't try to write past the keys size.

Right, sorry about my mistake.

Quoted 2 lines
Fixes b6b6cc9acd7b ("net: stmmac: selftest: avoid large stack usage")
maintainer
#11 lore raw
In reply to Jose Abreu

From: Jose Abreu <[email protected]>
Date: Mon, 30 Sep 2019 10:19:04 +0200

Quoted 4 lines

Misc fixes for -net tree. More info in commit logs.

v2 is just a rebase of v1 against -net and we added a new patch (09/09) to fix RSS feature.

Series applied, thanks.