[ Oops. I sent the last one too quickly. Here is another I should have
included. ]
Hello Florian Fainelli,
This is a semi-automatic email about new static checker warnings.
The patch 1c1008c793fa: "net: bcmgenet: add main driver file" from
Feb 13, 2014, leads to the following Smatch complaint:
drivers/net/ethernet/broadcom/genet/bcmgenet.c:2171 bcmgenet_close()
error: we previously assumed 'priv->phydev' could be null (see line
2137)
drivers/net/ethernet/broadcom/genet/bcmgenet.c
2136
2137 if (priv->phydev)
^^^^^^^^^^^^
Check.
2138 phy_stop(priv->phydev);
2139
2140 /* Disable MAC receive */
2141 reg = bcmgenet_umac_readl(priv, UMAC_CMD);
2142 reg &= ~CMD_RX_EN;
2143 bcmgenet_umac_writel(priv, reg, UMAC_CMD);
2144
2145 netif_tx_stop_all_queues(dev);
2146
2147 ret = bcmgenet_dma_teardown(priv);
2148 if (ret)
2149 return ret;
2150
2151 /* Disable MAC transmit. TX DMA disabled have to done before
this */
2152 reg = bcmgenet_umac_readl(priv, UMAC_CMD);
2153 reg &= ~CMD_TX_EN;
2154 bcmgenet_umac_writel(priv, reg, UMAC_CMD);
2155
2156 napi_disable(&priv->napi);
2157
2158 /* tx reclaim */
2159 bcmgenet_tx_reclaim_all(dev);
2160 bcmgenet_fini_dma(priv);
2161
2162 free_irq(priv->irq0, priv);
2163 free_irq(priv->irq1, priv);
2164
2165 /* Wait for pending work items to complete - we are stopping
2166 * the clock now. Since interrupts are disabled, no new work
2167 * will be scheduled.
2168 */
2169 cancel_work_sync(&priv->bcmgenet_irq_work);
2170
2171 if (phy_is_internal(priv->phydev))
^^^^^^^^^^^^
Unchecked dereference.
2172 bcmgenet_power_down(priv, GENET_POWER_PASSIVE);
2173
regards,
dan carpenter
_______________________________________________
kbuild mailing list
kbuild@lists.01.org
https://lists.01.org/mailman/listinfo/kbuild
