netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Ansuel Smith <ansuelsmth@gmail.com>
To: Florian Fainelli <f.fainelli@gmail.com>
Cc: Ansuel Smith <ansuelsmth@gmail.com>, Andrew Lunn <andrew@lunn.ch>,
	Vivien Didelot <vivien.didelot@gmail.com>,
	Vladimir Oltean <olteanv@gmail.com>,
	"David S. Miller" <davem@davemloft.net>,
	Jakub Kicinski <kuba@kernel.org>,
	Russell King <linux@armlinux.org.uk>,
	netdev@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [RFC PATCH net-next v2 12/17] net: dsa: qca8k: dsa: qca8k: protect MASTER busy_wait with mdio mutex
Date: Mon,  3 May 2021 01:07:04 +0200	[thread overview]
Message-ID: <20210502230710.30676-12-ansuelsmth@gmail.com> (raw)
In-Reply-To: <20210502230710.30676-1-ansuelsmth@gmail.com>

MDIO_MASTER operation have a dedicated busy wait that is not protected
by the mdio mutex. This can cause situation where the MASTER operation
is done and a normal operation is executed between the MASTER read/write
and the MASTER busy_wait. Rework the qca8k_mdio_read/write function to
address this issue by binding the lock for the whole MASTER operation
and not only the mdio read/write common operation.

Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com>
---
 drivers/net/dsa/qca8k.c | 69 ++++++++++++++++++++++++++++++++++-------
 1 file changed, 57 insertions(+), 12 deletions(-)

diff --git a/drivers/net/dsa/qca8k.c b/drivers/net/dsa/qca8k.c
index f64e3215a515..9a1b28bcaff0 100644
--- a/drivers/net/dsa/qca8k.c
+++ b/drivers/net/dsa/qca8k.c
@@ -617,9 +617,33 @@ qca8k_port_to_phy(int port)
 	return port - 1;
 }
 
+static int
+qca8k_mdio_busy_wait(struct qca8k_priv *priv, u32 reg, u32 mask)
+{
+	unsigned long timeout;
+	u16 r1, r2, page;
+
+	qca8k_split_addr(reg, &r1, &r2, &page);
+
+	timeout = jiffies + msecs_to_jiffies(20);
+
+	/* loop until the busy flag has cleared */
+	do {
+		u32 val = qca8k_mii_read32(priv->bus, 0x10 | r2, r1);
+		int busy = val & mask;
+
+		if (!busy)
+			break;
+		cond_resched();
+	} while (!time_after_eq(jiffies, timeout));
+
+	return time_after_eq(jiffies, timeout);
+}
+
 static int
 qca8k_mdio_write(struct qca8k_priv *priv, int port, u32 regnum, u16 data)
 {
+	u16 r1, r2, page;
 	u32 phy, val;
 	int ret;
 
@@ -635,12 +659,22 @@ qca8k_mdio_write(struct qca8k_priv *priv, int port, u32 regnum, u16 data)
 	      QCA8K_MDIO_MASTER_REG_ADDR(regnum) |
 	      QCA8K_MDIO_MASTER_DATA(data);
 
-	ret = qca8k_write(priv, QCA8K_MDIO_MASTER_CTRL, val);
+	qca8k_split_addr(QCA8K_MDIO_MASTER_CTRL, &r1, &r2, &page);
+
+	mutex_lock_nested(&priv->bus->mdio_lock, MDIO_MUTEX_NESTED);
+
+	ret = qca8k_set_page(priv->bus, page);
 	if (ret)
-		return ret;
+		goto exit;
+
+	qca8k_mii_write32(priv->bus, 0x10 | r2, r1, val);
 
-	ret = qca8k_busy_wait(priv, QCA8K_MDIO_MASTER_CTRL,
-			      QCA8K_MDIO_MASTER_BUSY);
+	if (qca8k_mdio_busy_wait(priv, QCA8K_MDIO_MASTER_CTRL,
+				 QCA8K_MDIO_MASTER_BUSY))
+		ret = -ETIMEDOUT;
+
+exit:
+	mutex_unlock(&priv->bus->mdio_lock);
 
 	/* even if the busy_wait timeouts try to clear the MASTER_EN */
 	qca8k_reg_clear(priv, QCA8K_MDIO_MASTER_CTRL,
@@ -652,6 +686,7 @@ qca8k_mdio_write(struct qca8k_priv *priv, int port, u32 regnum, u16 data)
 static int
 qca8k_mdio_read(struct qca8k_priv *priv, int port, u32 regnum)
 {
+	u16 r1, r2, page;
 	u32 phy, val;
 	int ret;
 
@@ -666,20 +701,30 @@ qca8k_mdio_read(struct qca8k_priv *priv, int port, u32 regnum)
 	      QCA8K_MDIO_MASTER_READ | QCA8K_MDIO_MASTER_PHY_ADDR(phy) |
 	      QCA8K_MDIO_MASTER_REG_ADDR(regnum);
 
-	ret = qca8k_write(priv, QCA8K_MDIO_MASTER_CTRL, val);
-	if (ret)
-		return ret;
+	qca8k_split_addr(QCA8K_MDIO_MASTER_CTRL, &r1, &r2, &page);
 
-	if (qca8k_busy_wait(priv, QCA8K_MDIO_MASTER_CTRL,
-			    QCA8K_MDIO_MASTER_BUSY))
-		return -ETIMEDOUT;
+	mutex_lock_nested(&priv->bus->mdio_lock, MDIO_MUTEX_NESTED);
 
-	ret = qca8k_read(priv, QCA8K_MDIO_MASTER_CTRL, &val);
+	ret = qca8k_set_page(priv->bus, page);
 	if (ret)
-		return ret;
+		goto exit;
+
+	qca8k_mii_write32(priv->bus, 0x10 | r2, r1, val);
+
+	if (qca8k_mdio_busy_wait(priv, QCA8K_MDIO_MASTER_CTRL,
+				 QCA8K_MDIO_MASTER_BUSY))
+		val = -ETIMEDOUT;
+	else
+		val = qca8k_mii_read32(priv->bus, 0x10 | r2, r1);
 
 	val &= QCA8K_MDIO_MASTER_DATA_MASK;
 
+exit:
+	mutex_unlock(&priv->bus->mdio_lock);
+
+	if (val >= 0)
+		val &= QCA8K_MDIO_MASTER_DATA_MASK;
+
 	/* even if the busy_wait timeouts try to clear the MASTER_EN */
 	qca8k_reg_clear(priv, QCA8K_MDIO_MASTER_CTRL,
 			QCA8K_MDIO_MASTER_EN);
-- 
2.30.2


  parent reply	other threads:[~2021-05-02 23:07 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-02 23:06 [RFC PATCH net-next v2 01/17] net: mdio: ipq8064: clean whitespaces in define Ansuel Smith
2021-05-02 23:06 ` [RFC PATCH net-next v2 02/17] net: mdio: ipq8064: switch to write/readl function Ansuel Smith
2021-05-03 12:49   ` Andrew Lunn
2021-05-03 14:24     ` Ansuel Smith
2021-05-03 14:44       ` Andrew Lunn
2021-05-02 23:06 ` [RFC PATCH net-next v2 03/17] net: mdio: ipq8064: enlarge sleep after read/write operation Ansuel Smith
2021-05-02 23:06 ` [RFC PATCH net-next v2 04/17] net: dsa: qca8k: rework read/write/set_page to provide error Ansuel Smith
2021-05-02 23:06 ` [RFC PATCH net-next v2 05/17] net: dsa: qca8k: add support for qca8327 switch Ansuel Smith
2021-05-02 23:06 ` [RFC PATCH net-next v2 06/17] devicetree: net: dsa: qca8k: Document new compatible qca8327 Ansuel Smith
2021-05-02 23:06 ` [RFC PATCH net-next v2 07/17] net: dsa: qca8k: add priority tweak to qca8337 switch Ansuel Smith
2021-05-02 23:07 ` [RFC PATCH net-next v2 08/17] net: dsa: qca8k: add GLOBAL_FC settings needed for qca8327 Ansuel Smith
2021-05-02 23:07 ` [RFC PATCH net-next v2 09/17] net: dsa: qca8k: add support for switch rev Ansuel Smith
2021-05-02 23:07 ` [RFC PATCH net-next v2 10/17] net: dsa: qca8k: make rgmii delay configurable Ansuel Smith
2021-05-03  0:54   ` Ansuel Smith
2021-05-02 23:07 ` [RFC PATCH net-next v2 11/17] net: dsa: qca8k: clear MASTER_EN after phy read/write Ansuel Smith
2021-05-02 23:07 ` Ansuel Smith [this message]
2021-05-02 23:07 ` [RFC PATCH net-next v2 13/17] net: dsa: qca8k: enlarge mdio delay and timeout Ansuel Smith
2021-05-02 23:07 ` [RFC PATCH net-next v2 14/17] net: phy: phylink: permit to pass dev_flags to phylink_connect_phy Ansuel Smith
2021-05-02 23:07 ` [RFC PATCH net-next v2 15/17] net: dsa: slave: pass dev_flags also to internal PHY Ansuel Smith
2021-05-02 23:07 ` [RFC PATCH net-next v2 16/17] net: dsa: qca8k: pass switch_revision info to phy dev_flags Ansuel Smith
2021-05-02 23:07 ` [RFC PATCH net-next v2 17/17] net: phy: add qca8k driver for qca8k switch internal PHY Ansuel Smith
2021-05-02 23:07 ` [RFC PATCH net-next v2 00/17] Multiple improvement to qca8k stability Ansuel Smith

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20210502230710.30676-12-ansuelsmth@gmail.com \
    --to=ansuelsmth@gmail.com \
    --cc=andrew@lunn.ch \
    --cc=davem@davemloft.net \
    --cc=f.fainelli@gmail.com \
    --cc=kuba@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux@armlinux.org.uk \
    --cc=netdev@vger.kernel.org \
    --cc=olteanv@gmail.com \
    --cc=vivien.didelot@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).