linux-kernel - Re: BPF CI update: veristat-scx job

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: <CAEf4BzbWZrg1Aq1p0c2h-s2Ro=Fm2Dk1uE7frFynOd3CwZqFZA@mail.gmail.com>
Date: Mon, 9 Jun 2025 14:43:05 -0700
From: Andrii Nakryiko <andrii.nakryiko@...il.com>
To: Ihor Solodrai <ihor.solodrai@...ux.dev>
Cc: bpf <bpf@...r.kernel.org>, linux-kernel@...r.kernel.org, 
	Kernel Team <kernel-team@...a.com>, kernel-ci@...a.com, 
	Alexei Starovoitov <ast@...nel.org>, tj@...nel.org, mkutsevol@...a.com, scottbpc@...a.com, 
	jakehillion@...a.com, mykolal@...a.com
Subject: Re: BPF CI update: veristat-scx job

On Mon, Jun 9, 2025 at 2:30 PM Ihor Solodrai <ihor.solodrai@...ux.dev> wrote:
>
> Hi everyone.
>
> In effort to improve testing of both BPF subsystem and sched-ext
> project [1], a new testing workflow has been added to BPF CI:
> veristat-scx.
>
> veristat-kernel and veristat-meta jobs have been running for a while
> now, and veristat-scx is basically adding more test cases using
> sched-ext BPF programs as input.
>
> For those who aren't aware, veristat [2] is a command line tool that
> can be used to load BPF object files into the kernel and check
> verification results, among other things. It's source code is at
> ./tools/testing/selftests/bpf/veristat.c
>
> On BPF CI, veristat-${target} job takes BPF object files (determined
> by ${target}) as input, and runs veristat on them against the kernel
> under testing. In addition to checking whether BPF program is accepted
> by the verifier, the CI job also collects verifier performance stats,
> and compares them to the baseline *failing* if a significant
> regression is detected.
>
> See an example of successful job run here:
> https://github.com/kernel-patches/bpf/actions/runs/15543439297/job/43761685117

Unsuccessful veristat runs are actually more interesting :) Do you
have a link to some examples with veristat failures?

>
> [1] https://github.com/sched-ext/scx
> [2] https://github.com/libbpf/veristat
>