Skip to contentSkip to conversation

net: stmmac: Fixes for -net

PATCH In review Jose Abreu 2 participants 13 messages Nov 6, 2019 – Nov 7, 2019

Subsystems (4)
lore
Share & tools
lore
author
#7 lore raw

PATCH 7/11 net: stmmac: xgmac: Disable Flow Control when 1 or more queues are in AV

In reply to Jose Abreu

When in AVB mode we need to disable flow control to prevent MAC from pausing in TX side.

Fixes ec6ea8e3eee9 ("net: stmmac: Add CBS support in 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_dma.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c +2 -0
index 2e814aa64a5c..f70ca5300b82 100644
@@ -470,6 +470,7 @@ static void dwxgmac2_enable_tso(void __iomem *ioaddr, bool en, u32 chan)
 static void dwxgmac2_qmode(void __iomem *ioaddr, u32 channel, u8 qmode)
 {
 	u32 value = readl(ioaddr + XGMAC_MTL_TXQ_OPMODE(channel));
+	u32 flow = readl(ioaddr + XGMAC_RX_FLOW_CTRL);
 
 	value &= ~XGMAC_TXQEN;
 	if (qmode != MTL_QUEUE_AVB) {
@@ -477,6 +478,7 @@ static void dwxgmac2_qmode(void __iomem *ioaddr, u32 channel, u8 qmode)
 		writel(0, ioaddr + XGMAC_MTL_TCx_ETS_CONTROL(channel));
 	} else {
 		value |= 0x1 << XGMAC_TXQEN_SHIFT;
+		writel(flow & (~XGMAC_RFE), ioaddr + XGMAC_RX_FLOW_CTRL);
 	}
 
 	writel(value, ioaddr +  XGMAC_MTL_TXQ_OPMODE(channel));
Signature · 1 line
2.7.4
author
#6 lore raw

PATCH 6/11 net: stmmac: xgmac: Fix AV Feature detection

In reply to Jose Abreu

Fix incorrect precedence of operators. For reference: AV implies AV Feature but RAV implies only RX side AV Feature. As we want full AV features we need to check RAV.

Fixes c2b69474d63b ("net: stmmac: xgmac: Correct RAVSEL field interpretation")

---
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_dma.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c +1 -1
index 965cbe3e6f51..2e814aa64a5c 100644
@@ -369,7 +369,7 @@ static void dwxgmac2_get_hw_feature(void __iomem *ioaddr,
 	dma_cap->eee = (hw_cap & XGMAC_HWFEAT_EEESEL) >> 13;
 	dma_cap->atime_stamp = (hw_cap & XGMAC_HWFEAT_TSSEL) >> 12;
 	dma_cap->av = (hw_cap & XGMAC_HWFEAT_AVSEL) >> 11;
-	dma_cap->av &= !(hw_cap & XGMAC_HWFEAT_RAVSEL) >> 10;
+	dma_cap->av &= !((hw_cap & XGMAC_HWFEAT_RAVSEL) >> 10);
 	dma_cap->arpoffsel = (hw_cap & XGMAC_HWFEAT_ARPOFFSEL) >> 9;
 	dma_cap->rmon = (hw_cap & XGMAC_HWFEAT_MMCSEL) >> 8;
 	dma_cap->pmt_magic_frame = (hw_cap & XGMAC_HWFEAT_MGKSEL) >> 7;
Signature · 1 line
2.7.4
author
#5 lore raw

PATCH 5/11 net: stmmac: xgmac: Fix TSA selection

In reply to Jose Abreu

When we change between Transmission Scheduling Algorithms, we need to clear previous values so that the new chosen algorithm is correctly selected.

Fixes ec6ea8e3eee9 ("net: stmmac: Add CBS support in 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_core.c | 1 +
 1 file changed, 1 insertion(+)
drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c +1 -0
index 01075a955c66..070bd7d1ae4c 100644
@@ -224,6 +224,7 @@ static void dwxgmac2_config_cbs(struct mac_device_info *hw,
 	writel(low_credit, ioaddr + XGMAC_MTL_TCx_LOCREDIT(queue));
 
 	value = readl(ioaddr + XGMAC_MTL_TCx_ETS_CONTROL(queue));
+	value &= ~XGMAC_TSA;
 	value |= XGMAC_CC | XGMAC_CBS;
 	writel(value, ioaddr + XGMAC_MTL_TCx_ETS_CONTROL(queue));
 }
Signature · 7 lines
2.7.4


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Jose Abreu

author
#0 lore raw

cover net: stmmac: Fixes for -net

Misc fixes for stmmac.

Patch 1/11 and 2/11, use the correct variable type for bitrev32() calls.

Patch 3/11, fixes the random failures the we were seing when running selftests.

Patch 4/11, prevents a crash that can occur when receiving AVB packets and with SPH feature enabled on XGMAC.

Patch 5/11, fixes the correct settings for CBS on XGMAC.

Patch 6/11, corrects the interpretation of AVB feature on XGMAC.

Patch 7/11, disables Flow Control for AVB enabled queues on XGMAC.

Patch 8/11, disables MMC interrupts on XGMAC, preventing a storm of interrupts.

Patch 9/11, fixes the number of packets that were being taken into account in the RX path cleaning function.

Patch 10/11, fixes an incorrect descriptor setting that could cause IP misbehavior.

Patch 11/11, fixes the IOC generation mechanism when multiple descriptors are used.

---
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 (11):
net: stmmac: gmac4: bitrev32 returns u32
net: stmmac: xgmac: bitrev32 returns u32
net: stmmac: selftests: Prevent false positives in filter tests
net: stmmac: xgmac: Only get SPH header len if available
net: stmmac: xgmac: Fix TSA selection
net: stmmac: xgmac: Fix AV Feature detection
net: stmmac: xgmac: Disable Flow Control when 1 or more queues are in
AV
net: stmmac: xgmac: Disable MMC interrupts by default
net: stmmac: Fix the packet count in stmmac_rx()
net: stmmac: Fix TSO descriptor with Enhanced Addressing
net: stmmac: Fix the TX IOC in xmit path

 drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c  |   2 +-
 .../net/ethernet/stmicro/stmmac/dwxgmac2_core.c    |   3 +-
 .../net/ethernet/stmicro/stmmac/dwxgmac2_descs.c   |   3 +-
 drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c |   4 +-
 drivers/net/ethernet/stmicro/stmmac/mmc_core.c     |   6 +-
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c  |  70 ++++++-----
 .../net/ethernet/stmicro/stmmac/stmmac_selftests.c | 134 +++++++++++++++------
 7 files changed, 144 insertions(+), 78 deletions(-)
Signature · 1 line
2.7.4
author
#2 lore raw

PATCH 2/11 net: stmmac: xgmac: bitrev32 returns u32

The bitrev32 function returns an u32 var, not an int. Fix it.

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 5031398e612c..01075a955c66 100644
@@ -463,7 +463,7 @@ static void dwxgmac2_set_filter(struct mac_device_info *hw,
 		value |= XGMAC_FILTER_HMC;
 
 		netdev_for_each_mc_addr(ha, dev) {
-			int nr = (bitrev32(~crc32_le(~0, ha->addr, 6)) >>
+			u32 nr = (bitrev32(~crc32_le(~0, ha->addr, 6)) >>
 					(32 - mcbitslog2));
 			mc_filter[nr >> 5] |= (1 << (nr & 0x1F));
 		}
Signature · 1 line
2.7.4
author
#11 lore raw

PATCH 11/11 net: stmmac: Fix the TX IOC in xmit path

In reply to Jose Abreu

IOC bit must be only set in the last descriptor. Move the logic up a little bit to make sure it's set in the correct descriptor.

---
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 | 64 ++++++++++++-----------
 1 file changed, 34 insertions(+), 30 deletions(-)
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +34 -30
index b0a16d7c6e3d..f826365c979d 100644
@@ -3024,6 +3024,19 @@ static netdev_tx_t stmmac_tso_xmit(struct sk_buff *skb, struct net_device *dev)
 	/* Only the last descriptor gets to point to the skb. */
 	tx_q->tx_skbuff[tx_q->cur_tx] = skb;
 
+	/* Manage tx mitigation */
+	tx_q->tx_count_frames += nfrags + 1;
+	if (likely(priv->tx_coal_frames > tx_q->tx_count_frames) &&
+	    !((skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) &&
+	      priv->hwts_tx_en)) {
+		stmmac_tx_timer_arm(priv, queue);
+	} else {
+		desc = &tx_q->dma_tx[tx_q->cur_tx];
+		tx_q->tx_count_frames = 0;
+		stmmac_set_tx_ic(priv, desc);
+		priv->xstats.tx_set_ic_bit++;
+	}
+
 	/* We've used all descriptors we need for this skb, however,
 	 * advance cur_tx so that it references a fresh descriptor.
 	 * ndo_start_xmit will fill this descriptor the next time it's
@@ -3041,19 +3054,6 @@ static netdev_tx_t stmmac_tso_xmit(struct sk_buff *skb, struct net_device *dev)
 	priv->xstats.tx_tso_frames++;
 	priv->xstats.tx_tso_nfrags += nfrags;
 
-	/* Manage tx mitigation */
-	tx_q->tx_count_frames += nfrags + 1;
-	if (likely(priv->tx_coal_frames > tx_q->tx_count_frames) &&
-	    !(priv->synopsys_id >= DWMAC_CORE_4_00 &&
-	    (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) &&
-	    priv->hwts_tx_en)) {
-		stmmac_tx_timer_arm(priv, queue);
-	} else {
-		tx_q->tx_count_frames = 0;
-		stmmac_set_tx_ic(priv, desc);
-		priv->xstats.tx_set_ic_bit++;
-	}
-
 	if (priv->sarc_type)
 		stmmac_set_desc_sarc(priv, first, priv->sarc_type);
 
@@ -3225,6 +3225,27 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 	/* Only the last descriptor gets to point to the skb. */
 	tx_q->tx_skbuff[entry] = skb;
 
+	/* According to the coalesce parameter the IC bit for the latest
+	 * segment is reset and the timer re-started to clean the tx status.
+	 * This approach takes care about the fragments: desc is the first
+	 * element in case of no SG.
+	 */
+	tx_q->tx_count_frames += nfrags + 1;
+	if (likely(priv->tx_coal_frames > tx_q->tx_count_frames) &&
+	    !((skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) &&
+	      priv->hwts_tx_en)) {
+		stmmac_tx_timer_arm(priv, queue);
+	} else {
+		if (likely(priv->extend_desc))
+			desc = &tx_q->dma_etx[entry].basic;
+		else
+			desc = &tx_q->dma_tx[entry];
+
+		tx_q->tx_count_frames = 0;
+		stmmac_set_tx_ic(priv, desc);
+		priv->xstats.tx_set_ic_bit++;
+	}
+
 	/* We've used all descriptors we need for this skb, however,
 	 * advance cur_tx so that it references a fresh descriptor.
 	 * ndo_start_xmit will fill this descriptor the next time it's
@@ -3260,23 +3281,6 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	dev->stats.tx_bytes += skb->len;
 
-	/* According to the coalesce parameter the IC bit for the latest
-	 * segment is reset and the timer re-started to clean the tx status.
-	 * This approach takes care about the fragments: desc is the first
-	 * element in case of no SG.
-	 */
-	tx_q->tx_count_frames += nfrags + 1;
-	if (likely(priv->tx_coal_frames > tx_q->tx_count_frames) &&
-	    !(priv->synopsys_id >= DWMAC_CORE_4_00 &&
-	    (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) &&
-	    priv->hwts_tx_en)) {
-		stmmac_tx_timer_arm(priv, queue);
-	} else {
-		tx_q->tx_count_frames = 0;
-		stmmac_set_tx_ic(priv, desc);
-		priv->xstats.tx_set_ic_bit++;
-	}
-
 	if (priv->sarc_type)
 		stmmac_set_desc_sarc(priv, first, priv->sarc_type);
 
Signature · 1 line
2.7.4
author
#3 lore raw

PATCH 3/11 net: stmmac: selftests: Prevent false positives in filter tests

In reply to Jose Abreu

In L2 tests that filter packets by destination MAC address we need to prevent false positives that can occur if we add an address that collides with the existing ones.

To fix this, lets manually check if the new address to be added is already present in the NIC and use a different one if so. For Hash filtering this also envolves converting the address to the hash.

Fixes 091810dbded9 ("net: stmmac: Introduce selftests 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]

---
 .../net/ethernet/stmicro/stmmac/stmmac_selftests.c | 134 +++++++++++++++------
 1 file changed, 94 insertions(+), 40 deletions(-)
drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c +94 -40
index e4ac3c401432..ac3f658105c0 100644
@@ -6,7 +6,9 @@
  * Author: Jose Abreu <[email protected]>
  */
 
+#include <linux/bitrev.h>
 #include <linux/completion.h>
+#include <linux/crc32.h>
 #include <linux/ethtool.h>
 #include <linux/ip.h>
 #include <linux/phy.h>
@@ -485,12 +487,48 @@ static int stmmac_filter_check(struct stmmac_priv *priv)
 	return -EOPNOTSUPP;
 }
 
+static bool stmmac_hash_check(struct stmmac_priv *priv, unsigned char *addr)
+{
+	int mc_offset = 32 - priv->hw->mcast_bits_log2;
+	struct netdev_hw_addr *ha;
+	u32 hash, hash_nr;
+
+	/* First compute the hash for desired addr */
+	hash = bitrev32(~crc32_le(~0, addr, 6)) >> mc_offset;
+	hash_nr = hash >> 5;
+	hash = 1 << (hash & 0x1f);
+
+	/* Now, check if it collides with any existing one */
+	netdev_for_each_mc_addr(ha, priv->dev) {
+		u32 nr = bitrev32(~crc32_le(~0, ha->addr, ETH_ALEN)) >> mc_offset;
+		if (((nr >> 5) == hash_nr) && ((1 << (nr & 0x1f)) == hash))
+			return false;
+	}
+
+	/* No collisions, address is good to go */
+	return true;
+}
+
+static bool stmmac_perfect_check(struct stmmac_priv *priv, unsigned char *addr)
+{
+	struct netdev_hw_addr *ha;
+
+	/* Check if it collides with any existing one */
+	netdev_for_each_uc_addr(ha, priv->dev) {
+		if (!memcmp(ha->addr, addr, ETH_ALEN))
+			return false;
+	}
+
+	/* No collisions, address is good to go */
+	return true;
+}
+
 static int stmmac_test_hfilt(struct stmmac_priv *priv)
 {
-	unsigned char gd_addr[ETH_ALEN] = {0x01, 0xee, 0xdd, 0xcc, 0xbb, 0xaa};
-	unsigned char bd_addr[ETH_ALEN] = {0x01, 0x01, 0x02, 0x03, 0x04, 0x05};
+	unsigned char gd_addr[ETH_ALEN] = {0xf1, 0xee, 0xdd, 0xcc, 0xbb, 0xaa};
+	unsigned char bd_addr[ETH_ALEN] = {0xf1, 0xff, 0xff, 0xff, 0xff, 0xff};
 	struct stmmac_packet_attrs attr = { };
-	int ret;
+	int ret, tries = 256;
 
 	ret = stmmac_filter_check(priv);
 	if (ret)
@@ -499,6 +537,16 @@ static int stmmac_test_hfilt(struct stmmac_priv *priv)
 	if (netdev_mc_count(priv->dev) >= priv->hw->multicast_filter_bins)
 		return -EOPNOTSUPP;
 
+	while (--tries) {
+		/* We only need to check the bd_addr for collisions */
+		bd_addr[ETH_ALEN - 1] = tries;
+		if (stmmac_hash_check(priv, bd_addr))
+			break;
+	}
+
+	if (!tries)
+		return -EOPNOTSUPP;
+
 	ret = dev_mc_add(priv->dev, gd_addr);
 	if (ret)
 		return ret;
@@ -523,13 +571,25 @@ static int stmmac_test_hfilt(struct stmmac_priv *priv)
 
 static int stmmac_test_pfilt(struct stmmac_priv *priv)
 {
-	unsigned char gd_addr[ETH_ALEN] = {0x00, 0x01, 0x44, 0x55, 0x66, 0x77};
-	unsigned char bd_addr[ETH_ALEN] = {0x08, 0x00, 0x22, 0x33, 0x44, 0x55};
+	unsigned char gd_addr[ETH_ALEN] = {0xf0, 0x01, 0x44, 0x55, 0x66, 0x77};
+	unsigned char bd_addr[ETH_ALEN] = {0xf0, 0xff, 0xff, 0xff, 0xff, 0xff};
 	struct stmmac_packet_attrs attr = { };
-	int ret;
+	int ret, tries = 256;
 
 	if (stmmac_filter_check(priv))
 		return -EOPNOTSUPP;
+	if (netdev_uc_count(priv->dev) >= priv->hw->unicast_filter_entries)
+		return -EOPNOTSUPP;
+
+	while (--tries) {
+		/* We only need to check the bd_addr for collisions */
+		bd_addr[ETH_ALEN - 1] = tries;
+		if (stmmac_perfect_check(priv, bd_addr))
+			break;
+	}
+
+	if (!tries)
+		return -EOPNOTSUPP;
 
 	ret = dev_uc_add(priv->dev, gd_addr);
 	if (ret)
@@ -553,39 +613,31 @@ static int stmmac_test_pfilt(struct stmmac_priv *priv)
 	return ret;
 }
 
-static int stmmac_dummy_sync(struct net_device *netdev, const u8 *addr)
-{
-	return 0;
-}
-
-static void stmmac_test_set_rx_mode(struct net_device *netdev)
-{
-	/* As we are in test mode of ethtool we already own the rtnl lock
-	 * so no address will change from user. We can just call the
-	 * ndo_set_rx_mode() callback directly */
-	if (netdev->netdev_ops->ndo_set_rx_mode)
-		netdev->netdev_ops->ndo_set_rx_mode(netdev);
-}
-
 static int stmmac_test_mcfilt(struct stmmac_priv *priv)
 {
-	unsigned char uc_addr[ETH_ALEN] = {0x00, 0x01, 0x44, 0x55, 0x66, 0x77};
-	unsigned char mc_addr[ETH_ALEN] = {0x01, 0x01, 0x44, 0x55, 0x66, 0x77};
+	unsigned char uc_addr[ETH_ALEN] = {0xf0, 0xff, 0xff, 0xff, 0xff, 0xff};
+	unsigned char mc_addr[ETH_ALEN] = {0xf1, 0xff, 0xff, 0xff, 0xff, 0xff};
 	struct stmmac_packet_attrs attr = { };
-	int ret;
+	int ret, tries = 256;
 
 	if (stmmac_filter_check(priv))
 		return -EOPNOTSUPP;
-	if (!priv->hw->multicast_filter_bins)
+	if (netdev_uc_count(priv->dev) >= priv->hw->unicast_filter_entries)
 		return -EOPNOTSUPP;
 
-	/* Remove all MC addresses */
-	__dev_mc_unsync(priv->dev, NULL);
-	stmmac_test_set_rx_mode(priv->dev);
+	while (--tries) {
+		/* We only need to check the mc_addr for collisions */
+		mc_addr[ETH_ALEN - 1] = tries;
+		if (stmmac_hash_check(priv, mc_addr))
+			break;
+	}
+
+	if (!tries)
+		return -EOPNOTSUPP;
 
 	ret = dev_uc_add(priv->dev, uc_addr);
 	if (ret)
-		goto cleanup;
+		return ret;
 
 	attr.dst = uc_addr;
 
@@ -602,30 +654,34 @@ static int stmmac_test_mcfilt(struct stmmac_priv *priv)
 
 cleanup:
 	dev_uc_del(priv->dev, uc_addr);
-	__dev_mc_sync(priv->dev, stmmac_dummy_sync, NULL);
-	stmmac_test_set_rx_mode(priv->dev);
 	return ret;
 }
 
 static int stmmac_test_ucfilt(struct stmmac_priv *priv)
 {
-	unsigned char uc_addr[ETH_ALEN] = {0x00, 0x01, 0x44, 0x55, 0x66, 0x77};
-	unsigned char mc_addr[ETH_ALEN] = {0x01, 0x01, 0x44, 0x55, 0x66, 0x77};
+	unsigned char uc_addr[ETH_ALEN] = {0xf0, 0xff, 0xff, 0xff, 0xff, 0xff};
+	unsigned char mc_addr[ETH_ALEN] = {0xf1, 0xff, 0xff, 0xff, 0xff, 0xff};
 	struct stmmac_packet_attrs attr = { };
-	int ret;
+	int ret, tries = 256;
 
 	if (stmmac_filter_check(priv))
 		return -EOPNOTSUPP;
-	if (!priv->hw->multicast_filter_bins)
+	if (netdev_mc_count(priv->dev) >= priv->hw->multicast_filter_bins)
 		return -EOPNOTSUPP;
 
-	/* Remove all UC addresses */
-	__dev_uc_unsync(priv->dev, NULL);
-	stmmac_test_set_rx_mode(priv->dev);
+	while (--tries) {
+		/* We only need to check the uc_addr for collisions */
+		uc_addr[ETH_ALEN - 1] = tries;
+		if (stmmac_perfect_check(priv, uc_addr))
+			break;
+	}
+
+	if (!tries)
+		return -EOPNOTSUPP;
 
 	ret = dev_mc_add(priv->dev, mc_addr);
 	if (ret)
-		goto cleanup;
+		return ret;
 
 	attr.dst = mc_addr;
 
@@ -642,8 +698,6 @@ static int stmmac_test_ucfilt(struct stmmac_priv *priv)
 
 cleanup:
 	dev_mc_del(priv->dev, mc_addr);
-	__dev_uc_sync(priv->dev, stmmac_dummy_sync, NULL);
-	stmmac_test_set_rx_mode(priv->dev);
 	return ret;
 }
 
Signature · 1 line
2.7.4
author
#10 lore raw

PATCH 10/11 net: stmmac: Fix TSO descriptor with Enhanced Addressing

In reply to Jose Abreu

When using addressing > 32 bits the TSO first descriptor only has the header so we can't set the payload field for this descriptor. Let's reset the variable so that buffer 2 value is zero.

Fixes a993db88d17d ("net: stmmac: Enable support for > 32 Bits addressing in XGMAC")

---
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 1ab1eea1556a..b0a16d7c6e3d 100644
@@ -2996,6 +2996,7 @@ static netdev_tx_t stmmac_tso_xmit(struct sk_buff *skb, struct net_device *dev)
 		stmmac_set_desc_addr(priv, first, des);
 		tmp_pay_len = pay_len;
 		des += proto_hdr_len;
+		pay_len = 0;
 	}
 
 	stmmac_tso_allocator(priv, des, tmp_pay_len, (nfrags == 0), queue);
Signature · 1 line
2.7.4
author
#9 lore raw

PATCH 9/11 net: stmmac: Fix the packet count in stmmac_rx()

In reply to Jose Abreu

Currently, stmmac_rx() is counting the number of descriptors but it should count the number of packets as specified by the NAPI limit.

Fix this.

Fixes ec222003bd94 ("net: stmmac: Prepare to add Split Header 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 | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +3 -2
index 4e9c848c67cc..1ab1eea1556a 100644
@@ -3506,8 +3506,6 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit, u32 queue)
 		if (unlikely(status & dma_own))
 			break;
 
-		count++;
-
 		rx_q->cur_rx = STMMAC_GET_ENTRY(rx_q->cur_rx, DMA_RX_SIZE);
 		next_entry = rx_q->cur_rx;
 
@@ -3534,6 +3532,7 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit, u32 queue)
 			goto read_again;
 		if (unlikely(error)) {
 			dev_kfree_skb(skb);
+			count++;
 			continue;
 		}
 
@@ -3573,6 +3572,7 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit, u32 queue)
 			skb = napi_alloc_skb(&ch->rx_napi, len);
 			if (!skb) {
 				priv->dev->stats.rx_dropped++;
+				count++;
 				continue;
 			}
 
@@ -3638,6 +3638,7 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit, u32 queue)
 
 		priv->dev->stats.rx_packets++;
 		priv->dev->stats.rx_bytes += len;
+		count++;
 	}
 
 	if (status & rx_not_ls) {
Signature · 1 line
2.7.4
author
#1 lore raw

PATCH 1/11 net: stmmac: gmac4: bitrev32 returns u32

In reply to Jose Abreu

The bitrev32 function returns an u32 var, not an int. Fix it.

Fixes 477286b53f55 ("stmmac: add GMAC4 core 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/dwmac4_core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +1 -1
index 5a7b0aca1d31..66e60c7e9850 100644
@@ -432,7 +432,7 @@ static void dwmac4_set_filter(struct mac_device_info *hw,
 			 * bits used depends on the hardware configuration
 			 * selected at core configuration time.
 			 */
-			int bit_nr = bitrev32(~crc32_le(~0, ha->addr,
+			u32 bit_nr = bitrev32(~crc32_le(~0, ha->addr,
 					ETH_ALEN)) >> (32 - mcbitslog2);
 			/* The most significant bit determines the register to
 			 * use (H/L) while the other 5 bits determine the bit
Signature · 1 line
2.7.4
author
#8 lore raw

PATCH 8/11 net: stmmac: xgmac: Disable MMC interrupts by default

In reply to Jose Abreu

MMC interrupts were being enabled, which is not what we want because it will lead to a storm of interrupts that are not handled at all. Fix it by disabling all MMC interrupts for XGMAC.

Fixes b6cdf09f51c2 ("net: stmmac: xgmac: Implement MMC counters")

---
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/mmc_core.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/net/ethernet/stmicro/stmmac/mmc_core.c +4 -2
index a223584f5f9a..252cf48c5816 100644
@@ -176,6 +176,7 @@
 #define MMC_XGMAC_RX_PKT_SMD_ERR	0x22c
 #define MMC_XGMAC_RX_PKT_ASSEMBLY_OK	0x230
 #define MMC_XGMAC_RX_FPE_FRAG		0x234
+#define MMC_XGMAC_RX_IPC_INTR_MASK	0x25c
 
 static void dwmac_mmc_ctrl(void __iomem *mmcaddr, unsigned int mode)
 {
@@ -333,8 +334,9 @@ static void dwxgmac_mmc_ctrl(void __iomem *mmcaddr, unsigned int mode)
 
 static void dwxgmac_mmc_intr_all_mask(void __iomem *mmcaddr)
 {
-	writel(MMC_DEFAULT_MASK, mmcaddr + MMC_RX_INTR_MASK);
-	writel(MMC_DEFAULT_MASK, mmcaddr + MMC_TX_INTR_MASK);
+	writel(0x0, mmcaddr + MMC_RX_INTR_MASK);
+	writel(0x0, mmcaddr + MMC_TX_INTR_MASK);
+	writel(MMC_DEFAULT_MASK, mmcaddr + MMC_XGMAC_RX_IPC_INTR_MASK);
 }
 
 static void dwxgmac_read_mmc_reg(void __iomem *addr, u32 reg, u32 *dest)
Signature · 1 line
2.7.4
author
#4 lore raw

PATCH 4/11 net: stmmac: xgmac: Only get SPH header len if available

In reply to Jose Abreu

Split Header length is only available when L34T == 0. Fix this by correctly checking if L34T is zero before trying to get Header length.

Fixes 67afd6d1cfdf ("net: stmmac: Add Split Header support and enable it in XGMAC cores")

---
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_descs.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/ethernet/stmicro/stmmac/dwxgmac2_descs.c +2 -1
index ae48154f933c..bd5838ce1e8a 100644
@@ -288,7 +288,8 @@ static int dwxgmac2_get_rx_hash(struct dma_desc *p, u32 *hash,
 
 static int dwxgmac2_get_rx_header_len(struct dma_desc *p, unsigned int *len)
 {
-	*len = le32_to_cpu(p->des2) & XGMAC_RDES2_HL;
+	if (le32_to_cpu(p->des3) & XGMAC_RDES3_L34T)
+		*len = le32_to_cpu(p->des2) & XGMAC_RDES2_HL;
 	return 0;
 }
 
Signature · 1 line
2.7.4