Skip to contentSkip to conversation

staging: r8188eu: remove unnecessary NULL check

PATCH In review Vihas Mak 4 participants 7 messages Nov 22, 2021 – Nov 25, 2021 Staging

lore
Share & tools
lore

Vihas Mak

author
#0 lore raw

PATCH 1/1 staging: r8188eu: remove unnecessary NULL check

remove unnecessary NULL check surrounding rtw_free_netdev(), as the check is already performed inside rtw_free_netdev() in drivers/staging/r8188eu/os_dep/osdep_service.c.

---
 drivers/staging/r8188eu/os_dep/usb_intf.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/staging/r8188eu/os_dep/usb_intf.c +1 -2
index 5a35d9fe3fc9..392bd7868519 100644
@@ -466,8 +466,7 @@ static void rtw_usb_if1_deinit(struct adapter *if1)
 		if1->hw_init_completed);
 	rtw_handle_dualmac(if1, 0);
 	rtw_free_drv_sw(if1);
-	if (pnetdev)
-		rtw_free_netdev(pnetdev);
+	rtw_free_netdev(pnetdev);
 }
 
 static int rtw_drv_init(struct usb_interface *pusb_intf, const struct usb_device_id *pdid)
Signature · 1 line
2.25.1
author
#2 lore raw
Quoted 1 line
Quoted 1 line

BTW, same can be done in rtw_usb_if1_init().

Yea, I noticed that too. But the NULL check in rtw_usb_if1_init() has a follow up "else if", so I didn't change it.

Thanks,
Vihas

Quoted 16 lines · Pavel Skripkin
Quoted 5 lines · Vihas Mak

remove unnecessary NULL check surrounding rtw_free_netdev(), as the check is already performed inside rtw_free_netdev() in drivers/staging/r8188eu/os_dep/osdep_service.c.

BTW, same can be done in rtw_usb_if1_init().

With regards,
Pavel Skripkin

#3 lore raw
In reply to Vihas Mak
Quoted 14 lines · Vihas Mak

remove unnecessary NULL check surrounding rtw_free_netdev(), as the check is already performed inside rtw_free_netdev() in drivers/staging/r8188eu/os_dep/osdep_service.c.

---
 drivers/staging/r8188eu/os_dep/usb_intf.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/staging/r8188eu/os_dep/usb_intf.c +0 -0
index 5a35d9fe3fc9..392bd7868519 100644

drivers/staging/r8188eu/os_dep/usb_intf.c static void rtw_usb_if1_deinit(struct adapter *if1) · line 470

Quoted 7 lines
drivers/staging/r8188eu/os_dep/usb_intf.c +1 -2
@@ -466,8 +466,7 @@ static void rtw_usb_if1_deinit(struct adapter *if1)
 		if1->hw_init_completed);
 	rtw_handle_dualmac(if1, 0);
 	rtw_free_drv_sw(if1);
-	if (pnetdev)
-		rtw_free_netdev(pnetdev);
+	rtw_free_netdev(pnetdev);
 }

I'm not a huge fan of these sorts of patches. They don't make the code more readable because they hide the complexity.

Occasionally we will get a forest cobra in our yard and everyone is screaming and panicking. I'm like, "Calm down. Once you've spotted the snake, even a deadly snake, then the danger has passed." You can just stay two or three meters away and you're fine. Call a snake catcher.

What you're doing here is you've got a potential NULL dereference which is the snake. And this patch is saying, "Snakes are so messy! Let's hide it in the bushes next to the sidewalk where no one can see it."

Hash tag, folksy wisdom. #snakes

On the other hand, it might be worth checking if "pnetdev" can even be NULL at this point, and then deleting both of the NULL checks. That would be a very good clean up.

regards,
dan carpenter

maintainer
#4 lore raw
In reply to Vihas Mak
Quoted 14 lines · Vihas Mak

remove unnecessary NULL check surrounding rtw_free_netdev(), as the check is already performed inside rtw_free_netdev() in drivers/staging/r8188eu/os_dep/osdep_service.c.

---
 drivers/staging/r8188eu/os_dep/usb_intf.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/staging/r8188eu/os_dep/usb_intf.c +0 -0
index 5a35d9fe3fc9..392bd7868519 100644

drivers/staging/r8188eu/os_dep/usb_intf.c static void rtw_usb_if1_deinit(struct adapter *if1) · line 469

Quoted 6 lines
drivers/staging/r8188eu/os_dep/usb_intf.c +1 -2
@@ -466,8 +466,7 @@ static void rtw_usb_if1_deinit(struct adapter *if1)
 		if1->hw_init_completed);
 	rtw_handle_dualmac(if1, 0);
 	rtw_free_drv_sw(if1);
-	if (pnetdev)
-		rtw_free_netdev(pnetdev);
+	rtw_free_netdev(pnetdev);

As Dan said, this isn't usually a good idea to hide this.

And are you sure this ever could be NULL?

thanks,

greg k-h

author
#5 lore raw
Quoted 1 line
Quoted 1 line

And are you sure this ever could be NULL?

Yes.

The function rtw_free_netdev() performs a NULL check before actually freeing the structure, so the "if (pnetdev)" check isn't really necessary before calling rtw_free_netdev().
That's the reason why I removed that check.

Quoted 30 lines · Greg KH
Quoted 14 lines · Vihas Mak

remove unnecessary NULL check surrounding rtw_free_netdev(), as the check is already performed inside rtw_free_netdev() in drivers/staging/r8188eu/os_dep/osdep_service.c.

---
 drivers/staging/r8188eu/os_dep/usb_intf.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/staging/r8188eu/os_dep/usb_intf.c +0 -0
index 5a35d9fe3fc9..392bd7868519 100644

drivers/staging/r8188eu/os_dep/usb_intf.c static void rtw_usb_if1_deinit(struct adapter *if1) · line 469

Quoted 6 lines
drivers/staging/r8188eu/os_dep/usb_intf.c +1 -2
@@ -466,8 +466,7 @@ static void rtw_usb_if1_deinit(struct adapter *if1)
              if1->hw_init_completed);
      rtw_handle_dualmac(if1, 0);
      rtw_free_drv_sw(if1);
-     if (pnetdev)
-             rtw_free_netdev(pnetdev);
+     rtw_free_netdev(pnetdev);

As Dan said, this isn't usually a good idea to hide this.

And are you sure this ever could be NULL?

thanks,

greg k-h

Signature · 2 lines
Thanks,
Vihas
#6 lore raw
Quoted 8 lines · Vihas Mak
Quoted 1 line
Quoted 1 line

And are you sure this ever could be NULL?

Yes.

The function rtw_free_netdev() performs a NULL check before actually freeing the structure, so the "if (pnetdev)" check isn't really necessary before calling rtw_free_netdev().
That's the reason why I removed that check.

That's not the question Greg and I asked. Re-read the emails.

regards,
dan carpenter