| Message ID | 4b78c848826fde1b8a3ccd53f32b80674812cb12.1684182962.git.christophe.jaillet@wanadoo.fr (mailing list archive) |
|---|---|
| State | Awaiting Upstream |
| Delegated to: | Netdev Maintainers |
| Headers | show |
| Series | can: ctucanfd: Fix an error handling path in ctucan_probe_common() | expand |
| Context | Check | Description |
|---|---|---|
| netdev/tree_selection | success | Series ignored based on subject |
On 15.05.2023 22:36:28, Christophe JAILLET wrote: > If register_candev() fails, a previous netif_napi_add() needs to be undone. > Add the missing netif_napi_del() in the error handling path. What about this path: free_candev(ndev) -> free_netdev() -> netif_napi_del() | https://elixir.bootlin.com/linux/v6.3.2/source/net/core/dev.c#L10714 Marc
Le 15/05/2023 à 22:51, Marc Kleine-Budde a écrit : > On 15.05.2023 22:36:28, Christophe JAILLET wrote: >> If register_candev() fails, a previous netif_napi_add() needs to be undone. >> Add the missing netif_napi_del() in the error handling path. > > What about this path: > free_candev(ndev) -> free_netdev() -> netif_napi_del() > > | https://elixir.bootlin.com/linux/v6.3.2/source/net/core/dev.c#L10714 > > Marc > Ok, thanks for the review, so in fact this is the netif_napi_del() call in ctucan_platform_remove() that can be removed instead. Harmless, but would be more consistent. I'll send a patch for that. CJ
On 16.05.2023 18:47:17, Christophe JAILLET wrote: > Le 15/05/2023 à 22:51, Marc Kleine-Budde a écrit : > > On 15.05.2023 22:36:28, Christophe JAILLET wrote: > > > If register_candev() fails, a previous netif_napi_add() needs to be undone. > > > Add the missing netif_napi_del() in the error handling path. > > > > What about this path: > > free_candev(ndev) -> free_netdev() -> netif_napi_del() > > > > | https://elixir.bootlin.com/linux/v6.3.2/source/net/core/dev.c#L10714 > > > > Marc > > > > Ok, thanks for the review, > > so in fact this is the netif_napi_del() call in ctucan_platform_remove() > that can be removed instead. > > Harmless, but would be more consistent. > I'll send a patch for that. Make it so! regards, Marc
diff --git a/drivers/net/can/ctucanfd/ctucanfd_base.c b/drivers/net/can/ctucanfd/ctucanfd_base.c index 64c349fd4600..59e1708e43d6 100644 --- a/drivers/net/can/ctucanfd/ctucanfd_base.c +++ b/drivers/net/can/ctucanfd/ctucanfd_base.c @@ -1429,7 +1429,7 @@ int ctucan_probe_common(struct device *dev, void __iomem *addr, int irq, unsigne ret = register_candev(ndev); if (ret) { dev_err(dev, "fail to register failed (err=%d)\n", ret); - goto err_deviceoff; + goto err_del_napi; } pm_runtime_put(dev); @@ -1439,6 +1439,8 @@ int ctucan_probe_common(struct device *dev, void __iomem *addr, int irq, unsigne return 0; +err_del_napi: + netif_napi_del(&priv->napi); err_deviceoff: pm_runtime_put(priv->dev); err_pmdisable:
If register_candev() fails, a previous netif_napi_add() needs to be undone. Add the missing netif_napi_del() in the error handling path. Fixes: 2dcb8e8782d8 ("can: ctucanfd: add support for CTU CAN FD open-source IP core - bus independent part.") Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr> --- drivers/net/can/ctucanfd/ctucanfd_base.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)