2020-05-04 06:29:47

by Maxim Petrov

Subject: [PATCH v2] stmmac: fix pointer check after utilization in stmmac_interrupt

The paranoidal pointer check in IRQ handler looks very strange - it
really protects us only against bogus drivers which request IRQ line
with null pointer dev_id. However, the code fragment is incorrect
because the dev pointer is used before the actual check which leads
to undefined behavior. Remove the check to avoid confusing people
with incorrect code.

Signed-off-by: Maxim Petrov <[email protected]>
---
Changes in V2:
* Remove the incorrect check instead of fixing it (suggested by
David Miller)
* Add clarification to the function description

drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 7 +------
1 file changed, 1 insertion(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 565da6498c84..e2b095d936cf 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -4060,7 +4060,7 @@ static int stmmac_set_features(struct net_device *netdev,
/**
* stmmac_interrupt - main ISR
* @irq: interrupt number.
- * @dev_id: to pass the net device pointer.
+ * @dev_id: to pass the net device pointer (must be valid).
* Description: this is the main driver interrupt service routine.
* It can call:
* o DMA service routine (to manage incoming frame reception and transmission
@@ -4084,11 +4084,6 @@ static irqreturn_t stmmac_interrupt(int irq, void *dev_id)
if (priv->irq_wake)
pm_wakeup_event(priv->device, 0);

- if (unlikely(!dev)) {
- netdev_err(priv->dev, "%s: invalid dev pointer\n", __func__);
- return IRQ_NONE;
- }
-
/* Check if adapter is up */
if (test_bit(STMMAC_DOWN, &priv->state))
return IRQ_HANDLED;

base-commit: 6a8b55ed4056ea5559ebe4f6a4b247f627870d4c
--
2.17.1

2020-05-04 15:46:02

by Markus Elfring

Subject: Re: [PATCH v2] stmmac: fix pointer check after utilization in stmmac_interrupt

> … However, the code fragment is incorrect
> because the dev pointer is used before the actual check

I find such information interesting.


> which leads to undefined behavior. …

I suggest to adjust the wording for this “conclusion”.

Regards,
Markus

2020-05-04 18:46:55

by David Miller

Subject: Re: [PATCH v2] stmmac: fix pointer check after utilization in stmmac_interrupt

From: Maxim Petrov <[email protected]>
Date: Mon, 4 May 2020 09:26:43 +0300

> The paranoidal pointer check in IRQ handler looks very strange - it
> really protects us only against bogus drivers which request IRQ line
> with null pointer dev_id. However, the code fragment is incorrect
> because the dev pointer is used before the actual check which leads
> to undefined behavior. Remove the check to avoid confusing people
> with incorrect code.
>
> Signed-off-by: Maxim Petrov <[email protected]>

Applied, thanks.