[PATCH] staging: rtl8192e: check return value of dev_skb_alloc

[PATCH] staging: rtl8192e: check return value of dev_skb_alloc

[Nicholas Krause]

Checks if dev_skb_alloc returns Null in function, fw_download_code.
If the return value of dev_skb_alloc is NULL return false and exit
this function.
Signed-off-by: Nicholas Krause <xerofoify@gmail.com>
---
 drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c b/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c
index 1a95d1f..38ae236 100644
--- a/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c
+++ b/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c
@@ -62,6 +62,8 @@ static bool fw_download_code(struct net_device *dev, u8 *code_virtual_address,
 
 		skb  = dev_alloc_skb(frag_length + 4);
 		memcpy((unsigned char *)(skb->cb), &dev, sizeof(dev));
+		if (!skb)
+			return !rt_status;
 		tcb_desc = (struct cb_desc *)(skb->cb + MAX_DEV_ADDR_SIZE);
 		tcb_desc->queue_index = TXCMD_QUEUE;
 		tcb_desc->bCmdOrInit = DESC_PACKET_TYPE_INIT;
-- 
1.9.1

Re: [PATCH] staging: rtl8192e: check return value of dev_skb_alloc

[Greg KH]

On Wed, Jun 18, 2014 at 10:24:41PM -0400, Nicholas Krause wrote:
> Checks if dev_skb_alloc returns Null in function, fw_download_code.
> If the return value of dev_skb_alloc is NULL return false and exit
> this function.
> Signed-off-by: Nicholas Krause <xerofoify@gmail.com>
> ---
>  drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c b/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c
> index 1a95d1f..38ae236 100644
> --- a/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c
> +++ b/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c
> @@ -62,6 +62,8 @@ static bool fw_download_code(struct net_device *dev, u8 *code_virtual_address,
>  
>  		skb  = dev_alloc_skb(frag_length + 4);
>  		memcpy((unsigned char *)(skb->cb), &dev, sizeof(dev));
> +		if (!skb)
> +			return !rt_status;
>  		tcb_desc = (struct cb_desc *)(skb->cb + MAX_DEV_ADDR_SIZE);
>  		tcb_desc->queue_index = TXCMD_QUEUE;
>  		tcb_desc->bCmdOrInit = DESC_PACKET_TYPE_INIT;

As writing emails to you seems like dumping stuff into a vacuum, I'm
going to just delete this message and start ignoring your mail like you
seem to be ignoring mine...

Best of luck,

greg k-h

Re: [PATCH] staging: rtl8192e: check return value of dev_skb_alloc

[Joe Perches]

On Wed, 2014-06-18 at 22:24 -0400, Nicholas Krause wrote:
> Checks if dev_skb_alloc returns Null in function, fw_download_code.
> If the return value of dev_skb_alloc is NULL return false and exit
> this function.

Hello Nicholas.

If you're going to try add these checks, please look at
the code a bit more before submitting defective patches.

> diff --git a/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c b/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c
[]
> @@ -62,6 +62,8 @@ static bool fw_download_code(struct net_device *dev, u8 *code_virtual_address,
>  
>  		skb  = dev_alloc_skb(frag_length + 4);
>  		memcpy((unsigned char *)(skb->cb), &dev, sizeof(dev));
> +		if (!skb)
> +			return !rt_status;

Umm, if you're going to check whether or not skb is NULL,
do it before it's dereferenced  as skb->cb.

Oh, and get rid of rt_status and use

			return false;

directly.

>  		tcb_desc = (struct cb_desc *)(skb->cb + MAX_DEV_ADDR_SIZE);
>  		tcb_desc->queue_index = TXCMD_QUEUE;
>  		tcb_desc->bCmdOrInit = DESC_PACKET_TYPE_INIT;

and change the end of function

from:
	return rt_status;
to:
	return true;

Re: [PATCH] staging: rtl8192e: check return value of dev_skb_alloc

[Dan Carpenter]

On Wed, Jun 18, 2014 at 10:24:41PM -0400, Nicholas Krause wrote:
> Checks if dev_skb_alloc returns Null in function, fw_download_code.
> If the return value of dev_skb_alloc is NULL return false and exit
> this function.
> Signed-off-by: Nicholas Krause <xerofoify@gmail.com>
> ---
>  drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c b/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c
> index 1a95d1f..38ae236 100644
> --- a/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c
> +++ b/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c
> @@ -62,6 +62,8 @@ static bool fw_download_code(struct net_device *dev, u8 *code_virtual_address,
>  
>  		skb  = dev_alloc_skb(frag_length + 4);
>  		memcpy((unsigned char *)(skb->cb), &dev, sizeof(dev));
> +		if (!skb)
> +			return !rt_status;

What's this !rt_status?  Just return false.

regards,
dan carpenter

Re: [PATCH] staging: rtl8192e: check return value of dev_skb_alloc

[Nick Krause]

Hey  guys,
Sorry about no reply,
my power supply died last night,
have to get a new one. I will be RMAing
the old old in order to get a packup in
case this PSU dies.
Cheers Nick


On Thu, Jun 19, 2014 at 1:29 AM, Dan Carpenter <dan.carpenter@oracle.com> wrote:
> On Wed, Jun 18, 2014 at 10:24:41PM -0400, Nicholas Krause wrote:
>> Checks if dev_skb_alloc returns Null in function, fw_download_code.
>> If the return value of dev_skb_alloc is NULL return false and exit
>> this function.
>> Signed-off-by: Nicholas Krause <xerofoify@gmail.com>
>> ---
>>  drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c | 2 ++
>>  1 file changed, 2 insertions(+)
>>
>> diff --git a/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c b/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c
>> index 1a95d1f..38ae236 100644
>> --- a/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c
>> +++ b/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c
>> @@ -62,6 +62,8 @@ static bool fw_download_code(struct net_device *dev, u8 *code_virtual_address,
>>
>>               skb  = dev_alloc_skb(frag_length + 4);
>>               memcpy((unsigned char *)(skb->cb), &dev, sizeof(dev));
>> +             if (!skb)
>> +                     return !rt_status;
>
> What's this !rt_status?  Just return false.
>
> regards,
> dan carpenter
>

[PATCH] staging: rtl8192e: check return value of dev_skb_alloc

[Nicholas Krause]

---
 drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c b/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c
index 1a95d1f..d8a9427 100644
--- a/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c
+++ b/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c
@@ -36,7 +36,6 @@ static bool fw_download_code(struct net_device *dev, u8 *code_virtual_address,
 			     u32 buffer_len)
 {
 	struct r8192_priv *priv = rtllib_priv(dev);
-	bool		    rt_status = true;
 	u16		    frag_threshold;
 	u16		    frag_length, frag_offset = 0;
 	int		    i;
@@ -59,7 +58,8 @@ static bool fw_download_code(struct net_device *dev, u8 *code_virtual_address,
 			bLastIniPkt = 1;
 
 		}
-
+		if (!skb)
+			return false;
 		skb  = dev_alloc_skb(frag_length + 4);
 		memcpy((unsigned char *)(skb->cb), &dev, sizeof(dev));
 		tcb_desc = (struct cb_desc *)(skb->cb + MAX_DEV_ADDR_SIZE);
@@ -99,7 +99,7 @@ static bool fw_download_code(struct net_device *dev, u8 *code_virtual_address,
 
 	write_nic_byte(dev, TPPoll, TPPoll_CQ);
 
-	return rt_status;
+	return true;
 }
 
 static bool CPUcheck_maincodeok_turnonCPU(struct net_device *dev)
-- 
1.9.1

Re: [PATCH] staging: rtl8192e: check return value of dev_skb_alloc

[Joe Perches]

On Thu, 2014-06-19 at 15:29 -0400, Nicholas Krause wrote:

> diff --git a/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c b/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c
[]
> @@ -59,7 +58,8 @@ static bool fw_download_code(struct net_device *dev, u8 *code_virtual_address,
>  			bLastIniPkt = 1;
>  
>  		}
> -
> +		if (!skb)
> +			return false;
>  		skb  = dev_alloc_skb(frag_length + 4);

Nick.

You may have compiled this, but it's obvious you
haven't tested it.

Please stop sending patches until you can verify what
you're doing with the code.

Please train a few more of your neurons for coding.

It takes some time and some practice but please don't
practice on linux-kernel.

Use google to find some other suitable forums like
http://cboard.cprogramming.com/c-programming/

Back to your patch:

	if (!skb)
		return false;

must go _after_

	skb = dev_alloc_skb(frag_length + 4);

Re: [PATCH] staging: rtl8192e: check return value of dev_skb_alloc

[Nick Krause]

This patch is old I thought I fixed it and resent.
Cheers Nick

On Thu, Jun 19, 2014 at 5:25 PM, Joe Perches <joe@perches.com> wrote:
> On Thu, 2014-06-19 at 15:29 -0400, Nicholas Krause wrote:
>
>> diff --git a/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c b/drivers/staging/rtl8192e/rtl8192e/r8192E_firmware.c
> []
>> @@ -59,7 +58,8 @@ static bool fw_download_code(struct net_device *dev, u8 *code_virtual_address,
>>                       bLastIniPkt = 1;
>>
>>               }
>> -
>> +             if (!skb)
>> +                     return false;
>>               skb  = dev_alloc_skb(frag_length + 4);
>
> Nick.
>
> You may have compiled this, but it's obvious you
> haven't tested it.
>
> Please stop sending patches until you can verify what
> you're doing with the code.
>
> Please train a few more of your neurons for coding.
>
> It takes some time and some practice but please don't
> practice on linux-kernel.
>
> Use google to find some other suitable forums like
> http://cboard.cprogramming.com/c-programming/
>
> Back to your patch:
>
>         if (!skb)
>                 return false;
>
> must go _after_
>
>         skb = dev_alloc_skb(frag_length + 4);
>
>
>