Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Toke =?utf-8?Q?H=C3=B8iland-J=C3=B8rgensen?= <toke@redhat.com>
To:     Jesper Dangaard Brouer <brouer@redhat.com>, bpf@vger.kernel.org,
        Stanislav Fomichev <sdf@google.com>
Cc:     Jesper Dangaard Brouer <brouer@redhat.com>, netdev@vger.kernel.org,
        linux-kernel@vger.kernel.org, martin.lau@kernel.org,
        ast@kernel.org, daniel@iogearbox.net, alexandr.lobakin@intel.com,
        larysa.zaremba@intel.com, xdp-hints@xdp-project.net,
        anthony.l.nguyen@intel.com, yoong.siang.song@intel.com,
        boon.leong.ong@intel.com, intel-wired-lan@lists.osuosl.org,
        pabeni@redhat.com, jesse.brandeburg@intel.com, kuba@kernel.org,
        edumazet@google.com, john.fastabend@gmail.com, hawk@kernel.org,
        davem@davemloft.net
Subject: Re: [xdp-hints] [PATCH bpf RFC-V2 1/5] xdp: rss hash types
 representation
In-Reply-To: <168010734324.3039990.16454026957159811204.stgit@firesoul>
References: <168010726310.3039990.2753040700813178259.stgit@firesoul>
 <168010734324.3039990.16454026957159811204.stgit@firesoul>
Date:   Wed, 29 Mar 2023 23:46:08 +0200
Message-ID: <87355nnsdb.fsf@toke.dk>
MIME-Version: 1.0
Content-Type: text/plain
Precedence: bulk

Jesper Dangaard Brouer <brouer@redhat.com> writes:

> diff --git a/net/core/xdp.c b/net/core/xdp.c
> index 7133017bcd74..81d41df30695 100644
> --- a/net/core/xdp.c
> +++ b/net/core/xdp.c
> @@ -721,12 +721,14 @@ __bpf_kfunc int bpf_xdp_metadata_rx_timestamp(const struct xdp_md *ctx, u64 *tim
>   * @hash: Return value pointer.
>   *
>   * Return:
> - * * Returns 0 on success or ``-errno`` on error.
> + * * Returns (positive) RSS hash **type** on success or ``-errno`` on error.

This change is going to break any BPF program that does:

if (!bpf_xdp_metadata_rx_hash(ctx, &hash))
  /* do something with hash */


so I think adding a second argument is better; that way, at least
breakage will be explicit instead of being a hidden change in semantics
(and the CO-RE style checking for kfuncs Alexei introduced should
trigger correctly).

But really, what we should do anyway is merge this during the -rc phase
to minimise any breakage :)

-Toke