From: Jim Keniston <jkenisto@us.ibm.com>
To: Andi Kleen <andi@firstfloor.org>
Cc: Avi Kivity <avi@redhat.com>,
Pekka Enberg <penberg@cs.helsinki.fi>,
Srikar Dronamraju <srikar@linux.vnet.ibm.com>,
Peter Zijlstra <peterz@infradead.org>,
ananth@in.ibm.com, Ingo Molnar <mingo@elte.hu>,
Arnaldo Carvalho de Melo <acme@infradead.org>,
utrace-devel <utrace-devel@redhat.com>,
Frederic Weisbecker <fweisbec@gmail.com>,
Masami Hiramatsu <mhiramat@redhat.com>,
Maneesh Soni <maneesh@in.ibm.com>, Mark Wielaard <mjw@redhat.com>,
LKML <linux-kernel@vger.kernel.org>
Subject: Re: [RFC] [PATCH 1/7] User Space Breakpoint Assistance Layer (UBP)
Date: Wed, 20 Jan 2010 12:28:45 -0800 [thread overview]
Message-ID: <1264019325.5122.62.camel@localhost.localdomain> (raw)
In-Reply-To: <20100120195826.GB24355@basil.fritz.box>
On Wed, 2010-01-20 at 20:58 +0100, Andi Kleen wrote:
> > Re: rewriting instructions that use rip-relative addressing. We do that
> > now. See handle_riprel_insn() in patch #2. (As far as we can tell, it
> > works, but we'd appreciate your review of it.)
>
> Yes, but how do you get within 2GB of it?
I'm not sure what you're asking.
To jump between the probed instruction stream and the XOL area, I've
proposed
jmpq *(%rip)
.quad next_insn
next_insn is a 64-bit address, which presumably allows you to jump to
anywhere in the address space.
To read/write the memory addressed by a rip-relative instruction, we
convert the rip-relative addressing to indirect addressing through a
64-bit scratch register (whose saved value we restore before returning
to the probed instruction stream).
> Add lots of holes
> in the address space?
No.
>
> > The instruction decoder is used only during instruction analysis, while
> > registering the probe -- i.e., in kernel space.
>
> Registering the user probe? That means if there's a buffer overflow
> in there it would be exploitable.
Certainly a poorly written probe handler would be a problem. Could you
explain further what you mean? Are you talking about a buffer overflow
in the probed program? in the probe handler? in uprobes?
>
> > >
> > > In general the trend has been also to make traps faster in the CPU, make
> > > sure you're not optimizing for some old CPU here.
> >
> > I won't argue with that. What Avi seems to be proposing buys us a
> > speedup, but at the cost of increased complexity -- among other things,
> > splitting the instrumentation code between user space (in the "XOL" area
> > -- which would then be used for much more than XOL instruction slots)
>
> You can't have a single XOL area, at least not if you want to support
> shared libraries on 64bit & rip relative.
I disagree. See above.
>
> > and kernel space. The splitting would presumably be handled by
> > higher-level code -- SystemTap, perf, or whatever. It's a neat idea,
> > but it seems like a v2 kind of feature.
>
> I'm not sure it can even work, unless you severly limited the allowed
> instructions.
I'm not sure it can work, either. But I still believe that we've
addressed the known issues wrt the big x86_64 address space.
>
> -Andi
>
Thanks.
Jim
next prev parent reply other threads:[~2010-01-20 20:29 UTC|newest]
Thread overview: 163+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-01-11 12:25 [RFC] [PATCH 0/7] UBP, XOL and Uprobes Srikar Dronamraju
2010-01-11 12:25 ` [RFC] [PATCH 1/7] User Space Breakpoint Assistance Layer (UBP) Srikar Dronamraju
2010-01-14 11:08 ` Peter Zijlstra
2010-01-14 19:46 ` Jim Keniston
2010-01-15 9:02 ` Peter Zijlstra
2010-01-15 21:07 ` Jim Keniston
2010-01-15 21:49 ` Peter Zijlstra
2010-01-16 0:58 ` Jim Keniston
2010-01-16 10:33 ` Peter Zijlstra
2010-01-17 0:12 ` Bryan Donlan
2010-01-18 7:37 ` Peter Zijlstra
2010-01-17 14:37 ` Avi Kivity
2010-01-15 9:03 ` Peter Zijlstra
2010-01-15 9:38 ` Ananth N Mavinakayanahalli
2010-01-15 9:50 ` Peter Zijlstra
2010-01-15 10:10 ` Ananth N Mavinakayanahalli
2010-01-15 10:13 ` Peter Zijlstra
2010-01-15 10:22 ` Ananth N Mavinakayanahalli
2010-01-15 10:56 ` Peter Zijlstra
2010-01-15 11:02 ` Peter Zijlstra
2010-01-15 21:19 ` Jim Keniston
2010-01-17 14:39 ` Avi Kivity
2010-01-17 14:52 ` Peter Zijlstra
2010-01-17 14:56 ` Avi Kivity
2010-01-17 15:01 ` Peter Zijlstra
2010-01-20 12:55 ` Pavel Machek
2010-01-17 14:59 ` Avi Kivity
2010-01-17 15:03 ` Peter Zijlstra
2010-01-17 19:33 ` Avi Kivity
2010-01-18 7:45 ` Peter Zijlstra
2010-01-18 11:01 ` Avi Kivity
2010-01-18 11:44 ` Peter Zijlstra
2010-01-18 12:01 ` Avi Kivity
2010-01-18 12:06 ` Peter Zijlstra
2010-01-18 12:09 ` Avi Kivity
2010-01-18 12:13 ` Pekka Enberg
2010-01-18 12:17 ` Avi Kivity
2010-01-18 12:24 ` Peter Zijlstra
2010-01-18 12:24 ` Pekka Enberg
2010-01-18 12:44 ` Srikar Dronamraju
2010-01-18 12:51 ` Pekka Enberg
2010-01-18 12:53 ` Avi Kivity
2010-01-18 12:57 ` Pekka Enberg
2010-01-18 13:06 ` Avi Kivity
2010-01-18 22:15 ` Jim Keniston
2010-01-19 8:07 ` Avi Kivity
2010-01-19 17:47 ` Jim Keniston
2010-01-19 18:06 ` Frederic Weisbecker
2010-01-20 6:36 ` Srikar Dronamraju
2010-01-20 10:51 ` Frederic Weisbecker
2010-01-20 19:31 ` Masami Hiramatsu
2010-01-20 9:43 ` Avi Kivity
2010-01-20 9:57 ` Peter Zijlstra
2010-01-20 12:22 ` Avi Kivity
2010-01-27 8:24 ` Ingo Molnar
2010-01-27 8:35 ` Avi Kivity
2010-01-27 9:08 ` Ingo Molnar
2010-01-27 9:25 ` Avi Kivity
2010-01-27 10:23 ` Ingo Molnar
2010-02-07 13:47 ` Avi Kivity
2010-01-20 10:45 ` Srikar Dronamraju
2010-01-20 12:23 ` Avi Kivity
2010-01-20 18:31 ` Andi Kleen
2010-01-20 19:34 ` Jim Keniston
2010-01-20 19:58 ` Andi Kleen
2010-01-20 20:28 ` Jim Keniston [this message]
2010-01-18 13:05 ` Peter Zijlstra
2010-01-18 13:34 ` Mark Wielaard
2010-01-18 19:49 ` Jim Keniston
2010-01-18 15:43 ` Ananth N Mavinakayanahalli
2010-01-18 16:52 ` Avi Kivity
2010-01-18 17:10 ` Ananth N Mavinakayanahalli
2010-01-18 12:14 ` Peter Zijlstra
2010-01-18 12:37 ` Avi Kivity
2010-01-18 13:15 ` Peter Zijlstra
2010-01-18 13:33 ` Avi Kivity
2010-01-18 13:34 ` K.Prasad
2010-01-20 15:57 ` Mel Gorman
2010-01-20 18:32 ` Andi Kleen
2010-01-18 11:45 ` Peter Zijlstra
2010-01-11 12:25 ` [RFC] [PATCH 2/7] x86 support for UBP Srikar Dronamraju
2010-01-11 12:25 ` [RFC] [PATCH 3/7] Execution out of line (XOL) Srikar Dronamraju
2010-01-14 11:08 ` Peter Zijlstra
2010-01-14 22:43 ` Jim Keniston
2010-01-15 9:07 ` Peter Zijlstra
2010-01-15 11:12 ` Srikar Dronamraju
2010-01-15 20:18 ` Jim Keniston
2010-01-11 12:25 ` [RFC] [PATCH 4/7] Uprobes Implementation Srikar Dronamraju
2010-01-12 2:01 ` Paul E. McKenney
2010-01-12 8:21 ` Srikar Dronamraju
2010-01-12 5:36 ` Frederic Weisbecker
2010-01-12 8:14 ` Ananth N Mavinakayanahalli
2010-01-13 0:53 ` Jim Keniston
2010-01-14 11:12 ` Peter Zijlstra
2010-01-12 8:54 ` Srikar Dronamraju
2010-01-14 11:09 ` Peter Zijlstra
2010-01-14 22:49 ` Jim Keniston
2010-01-15 9:10 ` Peter Zijlstra
2010-01-15 9:26 ` Frank Ch. Eigler
2010-01-15 9:35 ` Peter Zijlstra
2010-01-15 13:10 ` Frank Ch. Eigler
2010-01-15 13:25 ` Peter Zijlstra
2010-01-15 13:38 ` Frank Ch. Eigler
2010-01-15 13:47 ` Peter Zijlstra
2010-01-15 14:00 ` Frank Ch. Eigler
2010-01-15 14:06 ` Peter Zijlstra
2010-01-15 14:22 ` Frank Ch. Eigler
2010-01-15 14:40 ` Peter Zijlstra
2010-01-15 14:20 ` Srikar Dronamraju
2010-01-15 14:25 ` Peter Zijlstra
2010-01-15 23:11 ` Jim Keniston
2010-01-16 15:50 ` Frank Ch. Eigler
2010-01-15 10:26 ` Srikar Dronamraju
2010-01-15 10:33 ` Peter Zijlstra
2010-01-15 11:05 ` Maneesh Soni
2010-01-15 11:12 ` Peter Zijlstra
2010-01-15 11:18 ` Peter Zijlstra
2010-01-15 22:27 ` Jim Keniston
2010-01-15 23:44 ` Jim Keniston
2010-01-16 10:04 ` Peter Zijlstra
2010-01-15 13:08 ` Srikar Dronamraju
2010-01-15 13:16 ` Peter Zijlstra
2010-01-15 13:38 ` Peter Zijlstra
2010-01-11 12:25 ` [RFC] [PATCH 5/7] X86 Support for Uprobes Srikar Dronamraju
2010-01-14 11:13 ` Peter Zijlstra
2010-01-14 23:07 ` Jim Keniston
2010-01-11 12:26 ` [RFC] [PATCH 6/7] Uprobes Documentation Srikar Dronamraju
2010-01-11 12:26 ` [RFC] [PATCH 7/7] Ftrace plugin for Uprobes Srikar Dronamraju
2010-01-12 4:54 ` Frederic Weisbecker
2010-01-12 5:08 ` Steven Rostedt
2010-01-12 5:44 ` Frederic Weisbecker
2010-01-12 19:12 ` Tim Bird
2010-01-13 21:58 ` Masami Hiramatsu
2010-01-13 22:12 ` Masami Hiramatsu
2010-01-13 23:36 ` Steven Rostedt
2010-01-12 18:54 ` Frank Ch. Eigler
2010-01-12 22:00 ` Masami Hiramatsu
2010-01-12 22:15 ` Frank Ch. Eigler
2010-01-12 22:30 ` Masami Hiramatsu
2010-01-14 11:23 ` Peter Zijlstra
2010-01-14 11:29 ` Peter Zijlstra
2010-01-14 12:16 ` Mark Wielaard
2010-01-14 12:19 ` Peter Zijlstra
2010-01-14 11:35 ` Frederic Weisbecker
2010-01-14 11:43 ` Peter Zijlstra
2010-01-14 12:23 ` Frederic Weisbecker
2010-01-14 12:29 ` Peter Zijlstra
2010-01-18 13:00 ` Frederic Weisbecker
2010-01-11 14:35 ` [RFC] [PATCH 0/7] UBP, XOL and Uprobes Masami Hiramatsu
2010-01-11 22:59 ` Jim Keniston
2010-01-22 7:02 ` [RFC] [PATCH 0/7] UBP, XOL and Uprobes [ Summary of Comments and actions to be taken ] Srikar Dronamraju
2010-01-22 7:24 ` Ananth N Mavinakayanahalli
2010-01-22 10:47 ` Peter Zijlstra
2010-01-27 6:53 ` Peter Zijlstra
2010-01-27 8:24 ` Peter Zijlstra
2010-01-22 18:06 ` Peter Zijlstra
2010-01-22 18:36 ` Masami Hiramatsu
2010-01-22 23:55 ` Jim Keniston
2010-01-16 23:48 [RFC] [PATCH 1/7] User Space Breakpoint Assistance Layer (UBP) Jim Keniston
2010-01-18 7:23 ` Peter Zijlstra
2010-01-18 15:58 ` Masami Hiramatsu
2010-01-18 19:21 ` Jim Keniston
2010-01-18 21:20 ` Masami Hiramatsu
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=1264019325.5122.62.camel@localhost.localdomain \
--to=jkenisto@us.ibm.com \
--cc=acme@infradead.org \
--cc=ananth@in.ibm.com \
--cc=andi@firstfloor.org \
--cc=avi@redhat.com \
--cc=fweisbec@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=maneesh@in.ibm.com \
--cc=mhiramat@redhat.com \
--cc=mingo@elte.hu \
--cc=mjw@redhat.com \
--cc=penberg@cs.helsinki.fi \
--cc=peterz@infradead.org \
--cc=srikar@linux.vnet.ibm.com \
--cc=utrace-devel@redhat.com \
/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).