[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20201026092403.5e0634f3@kicinski-fedora-PC1C0HJN.hsd1.ca.comcast.net>
Date: Mon, 26 Oct 2020 09:24:03 -0700
From: Jakub Kicinski <kuba@...nel.org>
To: Yi Li <yili@...hong.com>
Cc: davem@...emloft.net, netdev@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2] net/core/dev.c : Use skb_is_gso
On Mon, 26 Oct 2020 09:34:35 +0800 Yi Li wrote:
> Trivial fix to use func skb_is_gso in place of
> test for skb_shinfo(skb)->gso_size.
>
> Signed-off-by: Yi Li <yili@...hong.com>
So you gave up on all the drivers now?
Please replace the word "fix" in the commit message with "refactoring",
the resulting code is identical.
The subject should be something like
net: core: Use skb_is_gso() in skb_checksum_help()
> diff --git a/net/core/dev.c b/net/core/dev.c
> index 9499a414d67e..55f66e108059 100644
> --- a/net/core/dev.c
> +++ b/net/core/dev.c
> @@ -3205,7 +3205,7 @@ int skb_checksum_help(struct sk_buff *skb)
> if (skb->ip_summed == CHECKSUM_COMPLETE)
> goto out_set_summed;
>
> - if (unlikely(skb_shinfo(skb)->gso_size)) {
> + if (unlikely(skb_is_gso(skb))) {
> skb_warn_bad_offload(skb);
> return -EINVAL;
> }
Powered by blists - more mailing lists