netdev - Re: [PATCH net-next V6 2/5] selftests: drv-net: Test XDP

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

Message-ID: <eaca90db-897c-45a0-8eed-92c36dbec825@nvidia.com>
Date: Mon, 21 Jul 2025 21:34:05 +0300
From: Gal Pressman <gal@...dia.com>
To: Jakub Kicinski <kuba@...nel.org>, Nimrod Oren <noren@...dia.com>
Cc: Mohsin Bashir <mohsin.bashr@...il.com>, netdev@...r.kernel.org,
 andrew+netdev@...n.ch, davem@...emloft.net, edumazet@...gle.com,
 pabeni@...hat.com, shuah@...nel.org, horms@...nel.org, cratiu@...dia.com,
 cjubran@...dia.com, mbloch@...dia.com, jdamato@...tly.com, sdf@...ichev.me,
 ast@...nel.org, daniel@...earbox.net, hawk@...nel.org,
 john.fastabend@...il.com, nathan@...nel.org,
 nick.desaulniers+lkml@...il.com, morbo@...gle.com, justinstitt@...gle.com,
 bpf@...r.kernel.org, linux-kselftest@...r.kernel.org, llvm@...ts.linux.dev,
 tariqt@...dia.com, thoiland@...hat.com
Subject: Re: [PATCH net-next V6 2/5] selftests: drv-net: Test XDP_PASS/DROP
 support

On 21/07/2025 18:40, Jakub Kicinski wrote:
> On Mon, 21 Jul 2025 14:43:15 +0300 Nimrod Oren wrote:
>>> +static struct udphdr *filter_udphdr(struct xdp_md *ctx, __u16 port)
>>> +{
>>> +	void *data_end = (void *)(long)ctx->data_end;
>>> +	void *data = (void *)(long)ctx->data;
>>> +	struct udphdr *udph = NULL;
>>> +	struct ethhdr *eth = data;
>>> +
>>> +	if (data + sizeof(*eth) > data_end)
>>> +		return NULL;
>>> +  
>>
>> This check assumes that the packet headers reside in the linear part of
>> the xdp_buff. However, this assumption does not hold across all drivers.
>> For example, in mlx5, the linear part is empty when using multi-buffer
>> mode with striding rq configuration. This causes all multi-buffer test
>> cases to fail over mlx5.
>>
>> To ensure correctness across all drivers, all direct accesses to packet
>> data should use these safer helper functions instead:
>> bpf_xdp_load_bytes() and bpf_xdp_store_bytes().
>>
>> Related discussion and context can be found here:
>> https://github.com/xdp-project/xdp-tools/pull/409
> 
> That's a reasonable way to modify the test. But I'm not sure it's
> something that should be blocking merging the patches.
> Or for that matter whether it's Mohsin's responsibility to make the
> test cater to quirks of mlx5, 

Definitely not a quirk, you cannot assume the headers are in the linear
part, especially if you're going to put this program as reference in the
kernel tree.

This issue has nothing to do with mlx5, but a buggy XDP program.

> which is not even part of NIPA testing -
> we have no way of knowing what passes for mlx5, what regresses it etc.

People have been developing XDP code that runs on mlx5 long before NIPA
even existed 🤷‍♂️..
And as you know, we run these selftests on mlx5 hardware, as evident by
Nimrod's mail, and others you've seen on the list. You know what regresses.