Re: [PATCH] net: eliminate meaningless memcpy to data in pskb_carve_inside_nonlinear()

2020-08-11 Thread Eric Dumazet 



On 8/11/20 5:10 AM, linmiaohe wrote:
> Eric Dumazet  wrote:
>> On 8/10/20 5:28 AM, Miaohe Lin wrote:
>>> The skb_shared_info part of the data is assigned in the following 
>>> loop. It is meaningless to do a memcpy here.
>>>
>>
>> Reminder : net-next is CLOSED.
>>
> 
> Thanks for your remind. I would wait for it open.
> 
>> This is not correct. We still have to copy _something_
>>
>> Something like :
>>
>> diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 
>> 2828f6d5ba898a5e50ccce45589bf1370e474b0f..1c0519426c7ba4b04377fc8054c4223c135879ab
>>  100644
>> --- a/net/core/skbuff.c
>> +++ b/net/core/skbuff.c
>> @@ -5953,8 +5953,8 @@ static int pskb_carve_inside_nonlinear(struct sk_buff 
>> *skb, const u32 off,
>>size = SKB_WITH_OVERHEAD(ksize(data));
>>
>>memcpy((struct skb_shared_info *)(data + size),
>> -  skb_shinfo(skb), offsetof(struct skb_shared_info,
>> -frags[skb_shinfo(skb)->nr_frags]));
>> +  skb_shinfo(skb), offsetof(struct skb_shared_info, 
>> + frags[0]));
>> +
>>if (skb_orphan_frags(skb, gfp_mask)) {
>>kfree(data);
>>return -ENOMEM;
>>
> 
> This looks good. Will send a patch v2 soon. May I add a suggested-by tag of 
> you ?

I would advise not using Suggested-by, as this would imply I suggested the idea 
of changing
this function in the first place.

I will add a Reviewed-by:  eventually if your v2 submission looks fine to me.

Thanks.


> Many thanks.
>

Re: [PATCH] net: eliminate meaningless memcpy to data in pskb_carve_inside_nonlinear()

2020-08-11 Thread linmiaohe 
Eric Dumazet  wrote:
> On 8/10/20 5:28 AM, Miaohe Lin wrote:
>> The skb_shared_info part of the data is assigned in the following 
>> loop. It is meaningless to do a memcpy here.
>> 
>
>Reminder : net-next is CLOSED.
>

Thanks for your remind. I would wait for it open.

>This is not correct. We still have to copy _something_
>
>Something like :
>
>diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 
>2828f6d5ba898a5e50ccce45589bf1370e474b0f..1c0519426c7ba4b04377fc8054c4223c135879ab
> 100644
>--- a/net/core/skbuff.c
>+++ b/net/core/skbuff.c
>@@ -5953,8 +5953,8 @@ static int pskb_carve_inside_nonlinear(struct sk_buff 
>*skb, const u32 off,
>size = SKB_WITH_OVERHEAD(ksize(data));
> 
>memcpy((struct skb_shared_info *)(data + size),
>-  skb_shinfo(skb), offsetof(struct skb_shared_info,
>-frags[skb_shinfo(skb)->nr_frags]));
>+  skb_shinfo(skb), offsetof(struct skb_shared_info, 
>+ frags[0]));
>+
>if (skb_orphan_frags(skb, gfp_mask)) {
>kfree(data);
>return -ENOMEM;
>

This looks good. Will send a patch v2 soon. May I add a suggested-by tag of you 
?
Many thanks.

Re: [PATCH] net: eliminate meaningless memcpy to data in pskb_carve_inside_nonlinear()

2020-08-11 Thread linmiaohe 
Florian Westphal  wrote:
>Miaohe Lin  wrote:
>> The skb_shared_info part of the data is assigned in the following loop.
>
>Where?
>

It's at the below for (i = 0; i < nfrags; i++) loop. But I missed something as 
Eric Dumazet pointed out.
Sorry about it.

Re: [PATCH] net: eliminate meaningless memcpy to data in pskb_carve_inside_nonlinear()

2020-08-10 Thread Eric Dumazet 



On 8/10/20 5:28 AM, Miaohe Lin wrote:
> The skb_shared_info part of the data is assigned in the following loop. It
> is meaningless to do a memcpy here.
> 
> Signed-off-by: Miaohe Lin 
> ---
>  net/core/skbuff.c | 3 ---
>  1 file changed, 3 deletions(-)
> 
> diff --git a/net/core/skbuff.c b/net/core/skbuff.c
> index 7e2e502ef519..5b983c9472f5 100644
> --- a/net/core/skbuff.c
> +++ b/net/core/skbuff.c
> @@ -5952,9 +5952,6 @@ static int pskb_carve_inside_nonlinear(struct sk_buff 
> *skb, const u32 off,
>  
>   size = SKB_WITH_OVERHEAD(ksize(data));
>  
> - memcpy((struct skb_shared_info *)(data + size),
> -skb_shinfo(skb), offsetof(struct skb_shared_info,
> -  frags[skb_shinfo(skb)->nr_frags]));
>   if (skb_orphan_frags(skb, gfp_mask)) {
>   kfree(data);
>   return -ENOMEM;
> 

Reminder : net-next is CLOSED.

This is not correct. We still have to copy _something_

Something like :

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 
2828f6d5ba898a5e50ccce45589bf1370e474b0f..1c0519426c7ba4b04377fc8054c4223c135879ab
 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -5953,8 +5953,8 @@ static int pskb_carve_inside_nonlinear(struct sk_buff 
*skb, const u32 off,
size = SKB_WITH_OVERHEAD(ksize(data));
 
memcpy((struct skb_shared_info *)(data + size),
-  skb_shinfo(skb), offsetof(struct skb_shared_info,
-frags[skb_shinfo(skb)->nr_frags]));
+  skb_shinfo(skb), offsetof(struct skb_shared_info, frags[0]));
+
if (skb_orphan_frags(skb, gfp_mask)) {
kfree(data);
return -ENOMEM;

Re: [PATCH] net: eliminate meaningless memcpy to data in pskb_carve_inside_nonlinear()

2020-08-10 Thread Florian Westphal 
Miaohe Lin  wrote:
> The skb_shared_info part of the data is assigned in the following loop.

Where?

[PATCH] net: eliminate meaningless memcpy to data in pskb_carve_inside_nonlinear()

2020-08-10 Thread Miaohe Lin 
The skb_shared_info part of the data is assigned in the following loop. It
is meaningless to do a memcpy here.

Signed-off-by: Miaohe Lin 
---
 net/core/skbuff.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 7e2e502ef519..5b983c9472f5 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -5952,9 +5952,6 @@ static int pskb_carve_inside_nonlinear(struct sk_buff 
*skb, const u32 off,
 
size = SKB_WITH_OVERHEAD(ksize(data));
 
-   memcpy((struct skb_shared_info *)(data + size),
-  skb_shinfo(skb), offsetof(struct skb_shared_info,
-frags[skb_shinfo(skb)->nr_frags]));
if (skb_orphan_frags(skb, gfp_mask)) {
kfree(data);
return -ENOMEM;
-- 
2.19.1