linux-kernel - Re: [PATCH v3 1/1] tracing: Remove "__attribute_

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: <20250723112329.691ed421@batman.local.home>
Date: Wed, 23 Jul 2025 11:23:29 -0400
From: Steven Rostedt <rostedt@...dmis.org>
To: "Masami Hiramatsu (Google)" <mhiramat@...nel.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@...icios.com>,
 linux-kernel@...r.kernel.org, linux-trace-kernel@...r.kernel.org
Subject: Re: [PATCH v3 1/1] tracing: Remove "__attribute__()" from the type
 field of event format

On Thu, 24 Jul 2025 00:18:06 +0900
Masami Hiramatsu (Google) <mhiramat@...nel.org> wrote:

> > So this will add more work during boot up as it is processed on every
> > event regardless if it has an eval map or not. But this is only needed
> > if CONFIG_DEBUG_INFO_BTF=y and PAHOLE_HAS_BTF_TAG=y is enabled.  
> 
> Hmm, In this case can we check it in trace_event_update_all()?
> If we need to sanitize more word, it is easier to add a condition
> there.

If we need to sanitize more words later, then we can open it up to
more. But why do this work when it's not needed. We are going from
calling this function a 100 times to calling it a 1000 times. That's an
order of magnitude, and I'm not sure we want to do that if it's not
needed.

> 
> Or, maybe we can sanitize it while building the kernel as a part
> of post build process.

Ideally, that would be best, but parsing the elf file isn't trivial.

I would say let's just add this conditional for now, and then we can
start working on a way to parse all of this at build time.

-- Steve