linux-kernel - Re: [PATCH bpf-next v2 5/7] mm: introduce BPF kfunc to access memory events

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]

Message-ID: <23c71e42-7da8-42ed-a93b-0d81dca99f3e@meta.com>
Date: Tue, 23 Dec 2025 09:09:35 -0500
From: Chris Mason <clm@...a.com>
To: Roman Gushchin <roman.gushchin@...ux.dev>
Cc: bot+bpf-ci@...nel.org, bpf@...r.kernel.org, linux-mm@...ck.org,
        linux-kernel@...r.kernel.org, inwardvessel@...il.com, ast@...nel.org,
        daniel@...earbox.net, shakeel.butt@...ux.dev, mhocko@...nel.org,
        hannes@...xchg.org, andrii@...nel.org, martin.lau@...nel.org,
        eddyz87@...il.com, yonghong.song@...ux.dev, ihor.solodrai@...ux.dev
Subject: Re: [PATCH bpf-next v2 5/7] mm: introduce BPF kfunc to access memory
 events

On 12/22/25 5:23 PM, Roman Gushchin wrote:
> Chris Mason <clm@...a.com> writes:
> 
>> On 12/20/25 1:43 PM, Roman Gushchin wrote:
>>>>> No, the bpf verifier enforces event > 0.
>>>>> It's a false positive.
>>>>
>>>> I'll add some words here to the bpf prompts, thanks Roman.
>>>
>>> I'll try to play with it too, hopefully we can fix it.
>>>
>>
>> https://github.com/masoncl/review-prompts/commit/fcc3bf704798f6be64cbb2e28b05a5c91eee9c7b 
> 
> Hi Chris!
> 
> I'm sorry, apparently I was dead wrong and overestimated the bpf
> verifier  (and ai was correct, lol). Somebody told me that enums
> are fully covered as a feedback to an earlier version and I didn't
> check.
> 
> In reality the verifier doesn't guarantee the correctness of the value
> passed as an enum, only that it's a u32. So we need to check the value.
> I've added necessarily checks in v3 of my patchset. It passes the local
> ai review without your latest change. Please, revert it.
> 
> Thanks and sorry for the hassle

Thanks Roman, I adjusted the prompt changes and looked harder for proof
of exactly what checks are done.

-chris