linux-kernel - Re: [PATCH v4 05/27] x86: Replace ist_enter() with nmi

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: <20200224170231.3807931d@gandalf.local.home>
Date:   Mon, 24 Feb 2020 17:02:31 -0500
From:   Steven Rostedt <rostedt@...dmis.org>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     Andy Lutomirski <luto@...nel.org>,
        LKML <linux-kernel@...r.kernel.org>,
        linux-arch <linux-arch@...r.kernel.org>,
        Ingo Molnar <mingo@...nel.org>,
        Joel Fernandes <joel@...lfernandes.org>,
        Greg KH <gregkh@...uxfoundation.org>, gustavo@...eddedor.com,
        Thomas Gleixner <tglx@...utronix.de>, paulmck@...nel.org,
        Josh Triplett <josh@...htriplett.org>,
        Mathieu Desnoyers <mathieu.desnoyers@...icios.com>,
        Lai Jiangshan <jiangshanlai@...il.com>,
        Tony Luck <tony.luck@...el.com>,
        Frederic Weisbecker <frederic@...nel.org>,
        Dan Carpenter <dan.carpenter@...cle.com>,
        Masami Hiramatsu <mhiramat@...nel.org>
Subject: Re: [PATCH v4 05/27] x86: Replace ist_enter() with nmi_enter()

On Mon, 24 Feb 2020 22:31:39 +0100
Peter Zijlstra <peterz@...radead.org> wrote:

> > Just want to confirm that printk_nmi_enter(), lockdep_off(),
> > and ftrace_nmi_enter() are all marked fully with NOKPROBE.  
> 
> *sigh*, right you are, I only looked at notrace, not nokprobe.
> 
> In particular the ftrace one is a bit off a mess, let me sort through
> that.

ftrace_nmi_enter() has two purposes. One, for archs that need to deal with
NMIs while they still use stop machine. Although, it appears sh is the only
arch that does that. I can look to see if it can be ripped out.

The other is for the hwlat detector that measures the time it was in an
NMI, as NMIs appear as a hardware latency too.

-- Steve