netdev - Re: [RFC] net: udp gso error code change

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Thu, 3 Mar 2022 09:57:25 -0500
From:   Willem de Bruijn <willemdebruijn.kernel@...il.com>
To:     Victor Stewart <v@...etag.social>
Cc:     netdev <netdev@...r.kernel.org>
Subject: Re: [RFC] net: udp gso error code change

On Tue, Mar 1, 2022 at 8:00 AM Victor Stewart <v@...etag.social> wrote:
>
> i assume this is might be a no-go because it could(?) break existing
> code, but i recently debugged an issue where i'd changed my linux box
> at home from ethernet to wifi, (with a few months away from the code),
> and suddenly my sendmsg ops were failing with EIO.
>
> turned out the wifi chip doesn't support checksum offloading, and this
> was the cause.
>
> but it seems in this situation EIO is a bit misleading/meaningless,
> when really the error should be something closer to EOPNOTSUPP?
>
> if there is support for this i can submit a simple patch for udp4/6
>
> https://github.com/torvalds/linux/blob/719fce7539cd3e186598e2aed36325fe892150cf/net/ipv6/udp.c#L1217
>
> if (skb->ip_summed != CHECKSUM_PARTIAL || is_udplite ||
> dst_xfrm(skb_dst(skb))) {
>    kfree_skb(skb);
>    return -EIO;
> }

As you point out, there is some risk with changing error numbers. I
would leave as is.