netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Russell King <rmk+kernel@armlinux.org.uk>
To: Andrew Lunn <andrew@lunn.ch>,
	Florian Fainelli <f.fainelli@gmail.com>,
	Heiner Kallweit <hkallweit1@gmail.com>
Cc: "David S. Miller" <davem@davemloft.net>, netdev@vger.kernel.org
Subject: [PATCH net-next 4/4] net: sfp: re-attempt probing for phy
Date: Mon, 09 Dec 2019 14:16:11 +0000	[thread overview]
Message-ID: <E1ieJpb-0004Uo-73@rmk-PC.armlinux.org.uk> (raw)
In-Reply-To: <20191209141525.GK25745@shell.armlinux.org.uk>

Some 1000BASE-T PHY modules take a while for the PHY to wake up.
Retry the probe a number of times before deciding that the module has
no PHY.

Tested with:
 Sourcephotonics SPGBTXCNFC - PHY takes less than 50ms to respond.
 Champion One 1000SFPT - PHY takes about 200ms to respond.
 Mikrotik S-RJ01 - no PHY

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
---
 drivers/net/phy/sfp.c | 59 ++++++++++++++++++++++++++++++-------------
 1 file changed, 42 insertions(+), 17 deletions(-)

diff --git a/drivers/net/phy/sfp.c b/drivers/net/phy/sfp.c
index 76fa95e54542..e54aef921038 100644
--- a/drivers/net/phy/sfp.c
+++ b/drivers/net/phy/sfp.c
@@ -62,6 +62,7 @@ enum {
 	SFP_S_FAIL,
 	SFP_S_WAIT,
 	SFP_S_INIT,
+	SFP_S_INIT_PHY,
 	SFP_S_INIT_TX_FAULT,
 	SFP_S_WAIT_LOS,
 	SFP_S_LINK_UP,
@@ -126,6 +127,7 @@ static const char * const sm_state_strings[] = {
 	[SFP_S_FAIL] = "fail",
 	[SFP_S_WAIT] = "wait",
 	[SFP_S_INIT] = "init",
+	[SFP_S_INIT_PHY] = "init_phy",
 	[SFP_S_INIT_TX_FAULT] = "init_tx_fault",
 	[SFP_S_WAIT_LOS] = "wait_los",
 	[SFP_S_LINK_UP] = "link_up",
@@ -180,6 +182,12 @@ static const enum gpiod_flags gpio_flags[] = {
 #define N_FAULT_INIT		5
 #define N_FAULT			5
 
+/* T_PHY_RETRY is the time interval between attempts to probe the PHY.
+ * R_PHY_RETRY is the number of attempts.
+ */
+#define T_PHY_RETRY		msecs_to_jiffies(50)
+#define R_PHY_RETRY		12
+
 /* SFP module presence detection is poor: the three MOD DEF signals are
  * the same length on the PCB, which means it's possible for MOD DEF 0 to
  * connect before the I2C bus on MOD DEF 1/2.
@@ -235,6 +243,7 @@ struct sfp {
 	unsigned char sm_dev_state;
 	unsigned short sm_state;
 	unsigned char sm_fault_retries;
+	unsigned char sm_phy_retries;
 
 	struct sfp_eeprom_id id;
 	unsigned int module_power_mW;
@@ -1416,10 +1425,8 @@ static int sfp_sm_probe_phy(struct sfp *sfp, bool is_c45)
 	int err;
 
 	phy = get_phy_device(sfp->i2c_mii, SFP_PHY_ADDR, is_c45);
-	if (phy == ERR_PTR(-ENODEV)) {
-		dev_info(sfp->dev, "no PHY detected\n");
-		return 0;
-	}
+	if (phy == ERR_PTR(-ENODEV))
+		return PTR_ERR(phy);
 	if (IS_ERR(phy)) {
 		dev_err(sfp->dev, "mdiobus scan returned %ld\n", PTR_ERR(phy));
 		return PTR_ERR(phy);
@@ -1867,6 +1874,7 @@ static void sfp_sm_module(struct sfp *sfp, unsigned int event)
 static void sfp_sm_main(struct sfp *sfp, unsigned int event)
 {
 	unsigned long timeout;
+	int ret;
 
 	/* Some events are global */
 	if (sfp->sm_state != SFP_S_DOWN &&
@@ -1940,22 +1948,39 @@ static void sfp_sm_main(struct sfp *sfp, unsigned int event)
 			sfp_sm_fault(sfp, SFP_S_INIT_TX_FAULT,
 				     sfp->sm_fault_retries == N_FAULT_INIT);
 		} else if (event == SFP_E_TIMEOUT || event == SFP_E_TX_CLEAR) {
-	init_done:	/* TX_FAULT deasserted or we timed out with TX_FAULT
-			 * clear.  Probe for the PHY and check the LOS state.
-			 */
-			if (sfp_sm_probe_for_phy(sfp)) {
-				sfp_sm_next(sfp, SFP_S_FAIL, 0);
-				break;
-			}
-			if (sfp_module_start(sfp->sfp_bus)) {
-				sfp_sm_next(sfp, SFP_S_FAIL, 0);
+	init_done:
+			sfp->sm_phy_retries = R_PHY_RETRY;
+			goto phy_probe;
+		}
+		break;
+
+	case SFP_S_INIT_PHY:
+		if (event != SFP_E_TIMEOUT)
+			break;
+	phy_probe:
+		/* TX_FAULT deasserted or we timed out with TX_FAULT
+		 * clear.  Probe for the PHY and check the LOS state.
+		 */
+		ret = sfp_sm_probe_for_phy(sfp);
+		if (ret == -ENODEV) {
+			if (--sfp->sm_phy_retries) {
+				sfp_sm_next(sfp, SFP_S_INIT_PHY, T_PHY_RETRY);
 				break;
+			} else {
+				dev_info(sfp->dev, "no PHY detected\n");
 			}
-			sfp_sm_link_check_los(sfp);
-
-			/* Reset the fault retry count */
-			sfp->sm_fault_retries = N_FAULT;
+		} else if (ret) {
+			sfp_sm_next(sfp, SFP_S_FAIL, 0);
+			break;
 		}
+		if (sfp_module_start(sfp->sfp_bus)) {
+			sfp_sm_next(sfp, SFP_S_FAIL, 0);
+			break;
+		}
+		sfp_sm_link_check_los(sfp);
+
+		/* Reset the fault retry count */
+		sfp->sm_fault_retries = N_FAULT;
 		break;
 
 	case SFP_S_INIT_TX_FAULT:
-- 
2.20.1


  parent reply	other threads:[~2019-12-09 14:16 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-12-09 14:15 [PATCH net-next 0/4] Add support for slow-to-probe-PHY copper SFP modules Russell King - ARM Linux admin
2019-12-09 14:15 ` [PATCH net-next 1/4] net: sfp: use a definition for the fault recovery attempts Russell King
2019-12-09 17:43   ` Andrew Lunn
2019-12-09 14:16 ` [PATCH net-next 2/4] net: sfp: rename sm_retries Russell King
2019-12-09 17:44   ` Andrew Lunn
2019-12-09 14:16 ` [PATCH net-next 3/4] net: sfp: error handling for phy probe Russell King
2019-12-09 17:45   ` Andrew Lunn
2019-12-09 14:16 ` Russell King [this message]
2019-12-09 17:47   ` [PATCH net-next 4/4] net: sfp: re-attempt probing for phy Andrew Lunn
2019-12-09 22:34 ` [PATCH net-next 0/4] Add support for slow-to-probe-PHY copper SFP modules David Miller
2019-12-09 22:35   ` David Miller
2019-12-09 22:38     ` Russell King - ARM Linux admin
2019-12-09 22:37   ` Russell King - ARM Linux admin
2019-12-11 20:46 ` David Miller

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=E1ieJpb-0004Uo-73@rmk-PC.armlinux.org.uk \
    --to=rmk+kernel@armlinux.org.uk \
    --cc=andrew@lunn.ch \
    --cc=davem@davemloft.net \
    --cc=f.fainelli@gmail.com \
    --cc=hkallweit1@gmail.com \
    --cc=netdev@vger.kernel.org \
    /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).