Skip to contentSkip to conversation

net: phy: micrel: Fix warn: passing zero to PTR_ERR

PATCH Superseded Divya Koppera 2 participants 4 messages Jul 19, 2022 – Jul 21, 2022

Subsystems (3)

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 net: phy: micrel: Fix warn: passing zero to PTR_ERR

Handle the NULL pointer case

Fixes New smatch warnings:
drivers/net/phy/micrel.c:2613 lan8814_ptp_probe_once() warn: passing zero to 'PTR_ERR'

vim +/PTR_ERR +2613 drivers/net/phy/micrel.c

Fixes ece19502834d ("net: phy: micrel: 1588 support for LAN8814 phy")

---
v1 -> v2:
- Handled NULL pointer case
- Changed subject line with net-next to net

---
 drivers/net/phy/micrel.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
drivers/net/phy/micrel.c +5 -1
index e78d0bf69bc3..6be6ee156f40 100644
@@ -2812,12 +2812,16 @@ static int lan8814_ptp_probe_once(struct phy_device *phydev)
 
 	shared->ptp_clock = ptp_clock_register(&shared->ptp_clock_info,
 					       &phydev->mdio.dev);
-	if (IS_ERR_OR_NULL(shared->ptp_clock)) {
+	if (IS_ERR(shared->ptp_clock)) {
 		phydev_err(phydev, "ptp_clock_register failed %lu\n",
 			   PTR_ERR(shared->ptp_clock));
 		return -EINVAL;
 	}
 
+	/* Check if PHC support is missing at the configuration level */
+	if (!shared->ptp_clock)
+		return 0;
+
 	phydev_dbg(phydev, "successfully registered ptp clock\n");
 
 	shared->phydev = phydev;
Signature · 1 line
2.17.1
#1 lore raw
Quoted 7 lines · Divya Koppera

Handle the NULL pointer case

Fixes New smatch warnings:
drivers/net/phy/micrel.c:2613 lan8814_ptp_probe_once() warn: passing zero to 'PTR_ERR'

vim +/PTR_ERR +2613 drivers/net/phy/micrel.c

Quoted 8 lines
Fixes ece19502834d ("net: phy: micrel: 1588 support for LAN8814 phy")

---
v1 -> v2:
- Handled NULL pointer case
- Changed subject line with net-next to net

This is not a genuine bug fix, and so it should target next-next.

Quoted 12 lines
---
 drivers/net/phy/micrel.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
drivers/net/phy/micrel.c +0 -0
index e78d0bf69bc3..6be6ee156f40 100644
@@ -2812,12 +2812,16 @@ static int lan8814_ptp_probe_once(struct phy_device *phydev)
 
 	shared->ptp_clock = ptp_clock_register(&shared->ptp_clock_info,
 					       &phydev->mdio.dev);

drivers/net/phy/micrel.c static int lan8814_ptp_probe_once(struct phy_device *phydev) · line 2823

Quoted 10 lines
drivers/net/phy/micrel.c +4 -1
-	if (IS_ERR_OR_NULL(shared->ptp_clock)) {
+	if (IS_ERR(shared->ptp_clock)) {
 		phydev_err(phydev, "ptp_clock_register failed %lu\n",
 			   PTR_ERR(shared->ptp_clock));
 		return -EINVAL;
 	}
 
+	/* Check if PHC support is missing at the configuration level */
+	if (!shared->ptp_clock)
+		return 0;

This is cause a NULL pointer de-reference in lan8814_ts_info() when it calls

	info->phc_index = ptp_clock_index(shared->ptp_clock);
Quoted 7 lines
drivers/net/phy/micrel.c +1 -0
+
 	phydev_dbg(phydev, "successfully registered ptp clock\n");
 
 	shared->phydev = phydev;
Signature · 1 line
2.17.1

Thanks,
Richard

#2 lore raw

Hi Richard,

Quoted 56 lines

-----Original Message-----
From: Richard Cochran <[email protected]>
Sent: Tuesday, July 19, 2022 7:26 PM
To: Divya Koppera - I30481 <[email protected]>
Cc: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; linux- [email protected]; UNGLinuxDriver <[email protected]>; Madhuri Sripada - I34878 <[email protected]>
Subject: Re: [PATCH v2 net] net: phy: micrel: Fix warn: passing zero to PTR_ERR

EXTERNAL EMAIL: Do not click links or open attachments unless you know the content is safe

Quoted 4 lines · Divya Koppera

Handle the NULL pointer case

Fixes New smatch warnings:
drivers/net/phy/micrel.c:2613 lan8814_ptp_probe_once() warn: passing

zero to 'PTR_ERR'

Quoted 11 lines

vim +/PTR_ERR +2613 drivers/net/phy/micrel.c

Fixes ece19502834d ("net: phy: micrel: 1588 support for LAN8814 phy")

---
v1 -> v2:
- Handled NULL pointer case
- Changed subject line with net-next to net

This is not a genuine bug fix, and so it should target next-next.

Quoted 20 lines
---
 drivers/net/phy/micrel.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
drivers/net/phy/micrel.c index renamed from drivers/net/phy/micrel.c +0 -0

e78d0bf69bc3..6be6ee156f40 100644

drivers/net/phy/micrel.c +1 -1
@@ -2812,12 +2812,16 @@ static int lan8814_ptp_probe_once(struct
 phy_device *phydev)
 
      shared->ptp_clock = ptp_clock_register(&shared->ptp_clock_info,
                                             &phydev->mdio.dev);
-     if (IS_ERR_OR_NULL(shared->ptp_clock)) {
+     if (IS_ERR(shared->ptp_clock)) {
              phydev_err(phydev, "ptp_clock_register failed %lu\n",
                         PTR_ERR(shared->ptp_clock));
              return -EINVAL;
      }
 
Quoted 8 lines

drivers/net/phy/micrel.c static int lan8814_ptp_probe_once(struct phy_device *phydev) · line 2823

Quoted 3 lines
drivers/net/phy/micrel.c +3 -0
+     /* Check if PHC support is missing at the configuration level */
+     if (!shared->ptp_clock)
+             return 0;

This is cause a NULL pointer de-reference in lan8814_ts_info() when it calls

        info->phc_index = ptp_clock_index(shared->ptp_clock);

NULL case handling seems to be redundant here because at starting of the function itself there is check for config support of ptp as below

static int lan8814_ptp_probe_once(struct phy_device *phydev)

{
        struct lan8814_shared_priv *shared = phydev->shared->priv;

        if (!IS_ENABLED(CONFIG_PTP_1588_CLOCK) ||
            !IS_ENABLED(CONFIG_NETWORK_PHY_TIMESTAMPING))
                return 0;

So chances of getting shared-> ptp_clock to be NULL is 0.

Let me know your thoughts. I'll remove this check in next revision if it is redundant.

Quoted 10 lines
Quoted 7 lines
drivers/net/phy/micrel.c +1 -0
+
      phydev_dbg(phydev, "successfully registered ptp clock\n");
 
      shared->phydev = phydev;

--
2.17.1

Thanks,
Richard

#3 lore raw
Quoted 7 lines · On Wed, Jul 20, 2022 at 04:32:55AM +0000, [email protected] wrote:

static int lan8814_ptp_probe_once(struct phy_device *phydev)

{
        struct lan8814_shared_priv *shared = phydev->shared->priv;

        if (!IS_ENABLED(CONFIG_PTP_1588_CLOCK) ||
            !IS_ENABLED(CONFIG_NETWORK_PHY_TIMESTAMPING))
                return 0;

It is weird to use macros here, but not before calling ptp_clock_register. Make it consistent by checking shared->ptp_clock instead.
That is also better form.

Thanks,
Richard