This is the mail archive of the
systemtap@sourceware.org
mailing list for the systemtap project.
Re: [RFC] [PATCH 2.6.37-rc5-tip 10/20] 10: uprobes: task specific information.
- From: Peter Zijlstra <peterz at infradead dot org>
- To: Josh Stone <jistone at redhat dot com>
- Cc: Srikar Dronamraju <srikar at linux dot vnet dot ibm dot com>, Ingo Molnar <mingo at elte dot hu>, Steven Rostedt <rostedt at goodmis dot org>, Arnaldo Carvalho de Melo <acme at infradead dot org>, Linus Torvalds <torvalds at linux-foundation dot org>, Masami Hiramatsu <masami dot hiramatsu dot pt at hitachi dot com>, Christoph Hellwig <hch at infradead dot org>, Andi Kleen <andi at firstfloor dot org>, Oleg Nesterov <oleg at redhat dot com>, LKML <linux-kernel at vger dot kernel dot org>, SystemTap <systemtap at sources dot redhat dot com>, Linux-mm <linux-mm at vger dot kernel dot org>, Jim Keniston <jkenisto at linux dot vnet dot ibm dot com>, Frederic Weisbecker <fweisbec at gmail dot com>, Ananth N Mavinakayanahalli <ananth at in dot ibm dot com>, Andrew Morton <akpm at linux-foundation dot org>, "Paul E. McKenney" <paulmck at linux dot vnet dot ibm dot com>
- Date: Tue, 25 Jan 2011 20:56:35 +0100
- Subject: Re: [RFC] [PATCH 2.6.37-rc5-tip 10/20] 10: uprobes: task specific information.
- References: <20101216095714.23751.52601.sendpatchset@localhost6.localdomain6> <20101216095912.23751.63180.sendpatchset@localhost6.localdomain6> <1295963775.28776.1056.camel@laptop> <4D3F1897.60300@redhat.com>
On Tue, 2011-01-25 at 10:38 -0800, Josh Stone wrote:
> On 01/25/2011 05:56 AM, Peter Zijlstra wrote:
> > Ah, I think I found it while reading patch 13, you need the pre/post_xol
> > callbacks, can't you simply synthesize their effect into the replacement
> > sequence?
> >
> > push %rax
> > mov $vaddr, %rax
> > $INSN
> > pop %rax
> > jmp $next_insn
> >
> > like replacements would obviate the need for the pre/post callbacks and
> > allow you to run straight through.
>
> For this particular example, you'd better be sure that $INSN doesn't
> need %rsp intact.
Well, either that of fix up the %rsp offset, but yes I had not
considered this.
> Control flow in general also makes this challenging. If $INSN is a
> call, then any inline fixups won't get a chance until after return. If
> $INSN is a jump, then its target must be modified so that both taken and
> not-taken paths land in respective fixup locations. I'm sure there are
> more cases that I'm not thinking of.
Right.
> > It would also remove the whole single-step need since they're proper
> > boosted probes.
>
> Kprobes has boosting, but it doesn't apply to all opcodes. I would
> guess that the same could be done for uprobes, where certain opcodes get
> a fixup sequence like you suggest, but the pre/post_xol mechanism is
> still needed in general.
Bummer..