Re: [PATCH] include:linux:Optimizations to __skb_push

2015-01-23 Thread David Miller
From: Mohammad Jamal Date: Thu, 22 Jan 2015 23:32:40 +0530 > This patch optimizes __skb_push function > > Signed-off-by: Mohammad Jamal A patch with a commit message that doesn't even explain the optimization, or why it is worthwhile to make that optimization, will not be applied, sorry. -- To

Re: [PATCH] include:linux:Optimizations to __skb_push

2015-01-22 Thread Jamal Mohammad
On Thu, Jan 22, 2015 at 11:54 PM, Eric Dumazet wrote: > On Thu, Jan 22, 2015 at 10:21 AM, Tom Herbert wrote: > >> Hmm, this seems less readable to me. What is the effect of this patch? >> Does the compiler even produce different assembly? > > No change at all in generated assembly, this looks not

Re: [PATCH] include:linux:Optimizations to __skb_push

2015-01-22 Thread Eric Dumazet
On Thu, Jan 22, 2015 at 10:21 AM, Tom Herbert wrote: > Hmm, this seems less readable to me. What is the effect of this patch? > Does the compiler even produce different assembly? No change at all in generated assembly, this looks not worth it. BTW, most __skb_push() callers ignore return value.

Re: [PATCH] include:linux:Optimizations to __skb_push

2015-01-22 Thread Tom Herbert
On Thu, Jan 22, 2015 at 10:02 AM, Mohammad Jamal wrote: > This patch optimizes __skb_push function > > Signed-off-by: Mohammad Jamal > --- > include/linux/skbuff.h |3 +-- > 1 file changed, 1 insertion(+), 2 deletions(-) > > diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h > inde