From: Willy Tarreau <w@1wt.eu>
To: Andy Lutomirski <luto@amacapital.net>
Cc: Peter Zijlstra <peterz@infradead.org>,
Linus Torvalds <torvalds@linux-foundation.org>,
Steven Rostedt <rostedt@goodmis.org>, X86 ML <x86@kernel.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
Borislav Petkov <bp@alien8.de>,
Thomas Gleixner <tglx@linutronix.de>,
Brian Gerst <brgerst@gmail.com>
Subject: Re: Dealing with the NMI mess
Date: Fri, 24 Jul 2015 19:21:00 +0200 [thread overview]
Message-ID: <20150724172100.GA3725@1wt.eu> (raw)
In-Reply-To: <20150724171018.GH3612@1wt.eu>
On Fri, Jul 24, 2015 at 07:10:18PM +0200, Willy Tarreau wrote:
> The OS has to set the RSP by itself before doing SYSRET, which opens a
> race between "mov rsp" and "sysret", but if we only take that path once
> we figure we come from NMI (using just IF+RSP), we know that IRQs and
> NMIs are still disabled and cannot strike at this instant. Maybe MCEs
> can, but they would execute within the NMI's stack just as if they were
> triggered inside the NMI as well so I don't see a problem here.
OK too bad I just found the response here in the code :-(
* SYSRET can't restore RF. SYSRET can restore TF, but unlike IRET,
* restoring TF results in a trap from userspace immediately after
* SYSRET. This would cause an infinite loop whenever #DB happens
* with register state that satisfies the opportunistic SYSRET
* conditions.
Willy
next prev parent reply other threads:[~2015-07-24 17:21 UTC|newest]
Thread overview: 85+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-07-23 20:21 Dealing with the NMI mess Andy Lutomirski
2015-07-23 20:38 ` Linus Torvalds
2015-07-23 20:49 ` Andy Lutomirski
2015-07-23 21:08 ` Linus Torvalds
2015-07-23 21:31 ` Steven Rostedt
2015-07-23 21:46 ` Willy Tarreau
2015-07-23 21:46 ` Andy Lutomirski
2015-07-23 21:50 ` Willy Tarreau
2015-07-23 21:48 ` Linus Torvalds
2015-07-23 21:50 ` Andy Lutomirski
2015-07-23 21:59 ` Linus Torvalds
2015-07-24 8:13 ` Peter Zijlstra
2015-07-24 9:02 ` Willy Tarreau
2015-07-24 11:58 ` Steven Rostedt
2015-07-24 12:43 ` Peter Zijlstra
2015-07-24 13:03 ` Steven Rostedt
2015-07-24 13:21 ` Willy Tarreau
2015-07-24 13:30 ` Peter Zijlstra
2015-07-24 13:33 ` Peter Zijlstra
2015-07-24 14:31 ` Steven Rostedt
2015-07-24 14:59 ` Willy Tarreau
2015-07-24 15:16 ` Steven Rostedt
2015-07-24 15:26 ` Willy Tarreau
2015-07-24 15:30 ` Peter Zijlstra
2015-07-24 15:33 ` Willy Tarreau
2015-07-24 18:29 ` Linus Torvalds
2015-07-24 18:41 ` Linus Torvalds
2015-07-24 19:05 ` Steven Rostedt
2015-07-24 19:55 ` Peter Zijlstra
2015-07-24 20:22 ` Linus Torvalds
2015-07-24 20:51 ` Peter Zijlstra
2015-07-24 21:07 ` Steven Rostedt
2015-07-24 21:08 ` Andy Lutomirski
2015-07-30 15:41 ` Paolo Bonzini
2015-07-30 21:22 ` Andy Lutomirski
2015-07-30 21:58 ` Brian Gerst
2015-07-30 22:59 ` Thomas Gleixner
2015-07-31 4:22 ` Borislav Petkov
2015-07-31 5:11 ` Andy Lutomirski
2015-07-31 7:51 ` Paolo Bonzini
2015-07-31 8:03 ` Borislav Petkov
2015-07-31 9:27 ` Paolo Bonzini
2015-07-31 10:25 ` Borislav Petkov
2015-07-31 10:26 ` Paolo Bonzini
2015-07-31 10:32 ` Borislav Petkov
2015-09-07 5:39 ` Maciej W. Rozycki
2015-09-07 7:42 ` Ingo Molnar
2015-09-07 8:19 ` Maciej W. Rozycki
2015-09-07 10:19 ` Paolo Bonzini
2015-09-07 17:01 ` Maciej W. Rozycki
2015-09-07 17:22 ` Andy Lutomirski
2015-09-07 19:30 ` Maciej W. Rozycki
2015-09-07 21:56 ` Andy Lutomirski
2015-09-08 16:21 ` Maciej W. Rozycki
2015-07-24 23:53 ` Linus Torvalds
2015-07-24 15:34 ` Steven Rostedt
2015-07-24 15:49 ` Willy Tarreau
2015-07-24 15:48 ` Andy Lutomirski
2015-07-24 16:02 ` Steven Rostedt
2015-07-24 16:08 ` Willy Tarreau
2015-07-24 16:31 ` Steven Rostedt
2015-07-24 16:06 ` Steven Rostedt
2015-07-24 16:25 ` Willy Tarreau
2015-07-24 17:21 ` Andy Lutomirski
2015-07-24 17:10 ` Willy Tarreau
2015-07-24 17:20 ` Andy Lutomirski
2015-07-30 15:54 ` Paolo Bonzini
2015-07-24 17:21 ` Willy Tarreau [this message]
2015-07-23 20:52 ` Willy Tarreau
2015-07-23 20:53 ` Andy Lutomirski
2015-07-23 21:07 ` Willy Tarreau
2015-07-23 21:13 ` Linus Torvalds
2015-07-23 21:18 ` Willy Tarreau
2015-07-23 21:20 ` Peter Zijlstra
2015-07-23 21:35 ` Linus Torvalds
2015-07-23 21:45 ` Andy Lutomirski
2015-07-23 21:54 ` Linus Torvalds
2015-07-23 21:59 ` Andy Lutomirski
2015-07-23 22:03 ` Linus Torvalds
2015-07-24 10:28 ` Peter Zijlstra
2015-07-24 11:06 ` Peter Zijlstra
2015-07-23 21:17 ` Peter Zijlstra
2015-07-23 21:20 ` Steven Rostedt
2015-07-23 21:46 ` Andy Lutomirski
2015-07-24 16:33 ` Raymond Jennings
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20150724172100.GA3725@1wt.eu \
--to=w@1wt.eu \
--cc=bp@alien8.de \
--cc=brgerst@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=luto@amacapital.net \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.org \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
--cc=x86@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).