linux-kernel - Re: [PATCH bpf-next v2 3/3] selftests/bpf: add benchmark testing for kprobe-multi-all

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: <aLGugoKpnZbVPYtC@krava>
Date: Fri, 29 Aug 2025 15:43:30 +0200
From: Jiri Olsa <olsajiri@...il.com>
To: Menglong Dong <menglong.dong@...ux.dev>
Cc: Menglong Dong <menglong8.dong@...il.com>,
	Jiri Olsa <olsajiri@...il.com>, andrii@...nel.org,
	eddyz87@...il.com, mykolal@...com, ast@...nel.org,
	daniel@...earbox.net, martin.lau@...ux.dev, song@...nel.org,
	yonghong.song@...ux.dev, john.fastabend@...il.com,
	kpsingh@...nel.org, sdf@...ichev.me, haoluo@...gle.com,
	shuah@...nel.org, yikai.lin@...o.com, memxor@...il.com,
	bpf@...r.kernel.org, linux-kselftest@...r.kernel.org,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH bpf-next v2 3/3] selftests/bpf: add benchmark testing for
 kprobe-multi-all

On Fri, Aug 29, 2025 at 03:01:41PM +0800, Menglong Dong wrote:
> On 2025/8/27 02:48 Jiri Olsa <olsajiri@...il.com> write:
> > On Tue, Aug 26, 2025 at 04:04:30PM +0800, Menglong Dong wrote:
> > > For now, the benchmark for kprobe-multi is single, which means there is
> > > only 1 function is hooked during testing. Add the testing
> > > "kprobe-multi-all", which will hook all the kernel functions during
> > > the benchmark. And the "kretprobe-multi-all" is added too.
> > 
> > hi,
> > fyi this bench causes panic on my setup.. very silent, so not sure
> > yet which function we should blacklist next, attaching my .config
> 
> According to my testing, the panic is due to the task hang.
> With so many debug config enabled(KASAN, LOCKDEP, etc),
> the system will become quite slow. What's wrose, there are
> many debug function is called during fprobe_entry(), which
> makes the BPF much slower. If you wait long enough, the
> testing can still run.
> 
> Add more symbol to the blacklist can mitigate this problem,
> but we need to maintain many symbols for this purpose. So
> I suggest that we only run this testing in product kernel.
> 
> What do you think?

yes I see the same.. I think we can keep it and perhaps make
some comment about that

jirka