[PATCH v2] ipmi: Fix error code in try_smi_init()

[PATCH v2] ipmi: Fix error code in try_smi_init()

[Dan Carpenter]

If platform_device_alloc() fails, then we should return -ENOMEM instead
of returning success.  I've also removed the "rv" initialization so that
GCC can maybe detect if we forget to set it in the future.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
---
v2: Don't initialize "rv" and fix a typo in the commit message

diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c
index f2a294f78892..c09b279cd55e 100644
--- a/drivers/char/ipmi/ipmi_si_intf.c
+++ b/drivers/char/ipmi/ipmi_si_intf.c
@@ -2026,7 +2026,7 @@ int ipmi_si_add_smi(struct si_sm_io *io)
  */
 static int try_smi_init(struct smi_info *new_smi)
 {
-	int rv = 0;
+	int rv;
 	int i;
 	char *init_name = NULL;
 
@@ -2071,6 +2071,7 @@ static int try_smi_init(struct smi_info *new_smi)
 						      new_smi->intf_num);
 		if (!new_smi->pdev) {
 			pr_err(PFX "Unable to allocate platform device\n");
+			rv = -ENOMEM;
 			goto out_err;
 		}
 		new_smi->io.dev = &new_smi->pdev->dev;

Re: [PATCH v2] ipmi: Fix error code in try_smi_init()

[Arnd Bergmann]

On Tue, Mar 6, 2018 at 11:40 AM, Dan Carpenter <dan.carpenter@oracle.com> wrote:
> If platform_device_alloc() fails, then we should return -ENOMEM instead
> of returning success.  I've also removed the "rv" initialization so that
> GCC can maybe detect if we forget to set it in the future.
>
> Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
> ---
> v2: Don't initialize "rv" and fix a typo in the commit message

This addresses my comment, so

Acked-by: Arnd Bergmann <arnd@arndb.de>