Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754806Ab1FGNLR (ORCPT ); Tue, 7 Jun 2011 09:11:17 -0400 Received: from e23smtp04.au.ibm.com ([202.81.31.146]:48461 "EHLO e23smtp04.au.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753771Ab1FGNFJ (ORCPT ); Tue, 7 Jun 2011 09:05:09 -0400 From: Srikar Dronamraju To: Peter Zijlstra , Ingo Molnar Cc: Steven Rostedt , Srikar Dronamraju , Linux-mm , Arnaldo Carvalho de Melo , Linus Torvalds , Jonathan Corbet , Masami Hiramatsu , Hugh Dickins , Christoph Hellwig , Ananth N Mavinakayanahalli , Thomas Gleixner , Andi Kleen , Oleg Nesterov , Andrew Morton , Jim Keniston , Roland McGrath , LKML Date: Tue, 07 Jun 2011 18:28:04 +0530 Message-Id: <20110607125804.28590.92092.sendpatchset@localhost6.localdomain6> Subject: [PATCH v4 3.0-rc2-tip 0/22] 0: Uprobes patchset with perf probe support Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 7340 Lines: 160 This patchset implements Uprobes which enables you to dynamically break into any routine in a user space application and collect information non-disruptively. This patchset resolves most of the comments on the previous posting https://lkml.org/lkml/2011/4/1/176 and inputs I got at LFCS. This patchset applies on top of tip commit 59c5f46fbe01 Uprobes Patches This patchset implements inode based uprobes which are specified as : where offset is the offset from start of the map. The probehit overhead is around 3X times the overhead from pid based patchset. When a uprobe is registered, Uprobes makes a copy of the probed instruction, replaces the first byte(s) of the probed instruction with a breakpoint instruction. (Uprobes uses background page replacement mechanism and ensures that the breakpoint affects only that process.) When a CPU hits the breakpoint instruction, Uprobes gets notified of trap and finds the associated uprobe. It then executes the associated handler. Uprobes single-steps its copy of the probed instruction and resumes execution of the probed process at the instruction following the probepoint. Instruction copies to be single-stepped are stored in a per-mm "execution out of line (XOL) area". Currently XOL area is allocated as one page vma. For previous postings: please refer: http://lkml.org/lkml/2011/3/14/171/ http://lkml.org/lkml/2010/12/16/65 http://lkml.org/lkml/2010/8/25/165 http://lkml.org/lkml/2010/7/27/121 http://lkml.org/lkml/2010/7/12/67 http://lkml.org/lkml/2010/7/8/239 http://lkml.org/lkml/2010/6/29/299 http://lkml.org/lkml/2010/6/14/41 http://lkml.org/lkml/2010/3/20/107 and http://lkml.org/lkml/2010/5/18/307 This patchset is a rework based on suggestions from discussions on lkml in September, March and January 2010 (http://lkml.org/lkml/2010/1/11/92, http://lkml.org/lkml/2010/1/27/19, http://lkml.org/lkml/2010/3/20/107 and http://lkml.org/lkml/2010/3/31/199 ). This implementation of uprobes doesnt depend on utrace. Advantages of uprobes over conventional debugging include: 1. Non-disruptive. Unlike current ptrace based mechanisms, uprobes tracing wouldnt involve signals, stopping threads and context switching between the tracer and tracee. 2. Much better handling of multithreaded programs because of XOL. Current ptrace based mechanisms use single stepping inline, i.e they copy back the original instruction on hitting a breakpoint. In such mechanisms tracers have to stop all the threads on a breakpoint hit or tracers will not be able to handle all hits to the location of interest. Uprobes uses execution out of line, where the instruction to be traced is analysed at the time of breakpoint insertion and a copy of instruction is stored at a different location. On breakpoint hit, uprobes jumps to that copied location and singlesteps the same instruction and does the necessary fixups post singlestepping. 3. Multiple tracers for an application. Multiple uprobes based tracer could work in unison to trace an application. There could one tracer that could be interested in generic events for a particular set of process. While there could be another tracer that is just interested in one specific event of a particular process thats part of the previous set of process. 4. Corelating events from kernels and userspace. Uprobes could be used with other tools like kprobes, tracepoints or as part of higher level tools like perf to give a consolidated set of events from kernel and userspace. In future we could look at a single backtrace showing application, library and kernel calls. Here is the list of TODO Items. - Breakpoint handling should co-exist with singlestep/blockstep from another tracer/debugger. - Queue and dequeue signals delivered from the singlestep till completion of postprocessing. - Prefiltering (i.e filtering at the time of probe insertion) - Return probes. - Support for other architectures. - Uprobes booster. - replace macro W with bits in inat table. To try please fetch using git fetch \ git://git.kernel.org/pub/scm/linux/kernel/git/srikar/linux-uprobes.git \ tip_inode_uprobes_070611:tip_inode_uprobes Please refer "[RFC] [PATCH 3.0-rc2-tip 18/22] tracing: tracing: Uprobe tracer documentation" on how to use uprobe_tracer. Please refer "[RFC] [PATCH 3.0-rc2-tip 22/22] perf: Documentation for perf uprobes" on how to use uprobe_tracer. Please do provide your valuable comments. Thanks in advance. Srikar Srikar Dronamraju (22) 0: Uprobes patchset with perf probe support 1: X86 specific breakpoint definitions. 2: uprobes: Breakground page replacement. 3: uprobes: Adding and remove a uprobe in a rb tree. 4: Uprobes: register/unregister probes. 5: x86: analyze instruction and determine fixups. 6: uprobes: store/restore original instruction. 7: uprobes: mmap and fork hooks. 8: x86: architecture specific task information. 9: uprobes: task specific information. 10: uprobes: slot allocation for uprobes 11: uprobes: get the breakpoint address. 12: x86: x86 specific probe handling 13: uprobes: Handing int3 and singlestep exception. 14: x86: uprobes exception notifier for x86. 15: uprobes: register a notifier for uprobes. 16: tracing: Extract out common code for kprobes/uprobes traceevents. 17: tracing: uprobes trace_event interface 18: tracing: Uprobe tracer documentation 19: perf: rename target_module to target 20: perf: perf interface for uprobes 21: perf: show possible probes in a given executable file or library. 22: perf: Documentation for perf uprobes Documentation/trace/uprobetrace.txt | 94 ++ arch/Kconfig | 4 + arch/x86/Kconfig | 3 + arch/x86/include/asm/thread_info.h | 2 + arch/x86/include/asm/uprobes.h | 53 ++ arch/x86/kernel/Makefile | 1 + arch/x86/kernel/signal.c | 14 + arch/x86/kernel/uprobes.c | 591 +++++++++++++ include/linux/mm_types.h | 9 + include/linux/sched.h | 9 +- include/linux/uprobes.h | 194 ++++ kernel/Makefile | 1 + kernel/fork.c | 10 + kernel/trace/Kconfig | 20 + kernel/trace/Makefile | 2 + kernel/trace/trace.h | 5 + kernel/trace/trace_kprobe.c | 860 +------------------ kernel/trace/trace_probe.c | 752 ++++++++++++++++ kernel/trace/trace_probe.h | 160 ++++ kernel/trace/trace_uprobe.c | 812 +++++++++++++++++ kernel/uprobes.c | 1476 +++++++++++++++++++++++++++++++ mm/mmap.c | 6 + tools/perf/Documentation/perf-probe.txt | 21 +- tools/perf/builtin-probe.c | 77 ++- tools/perf/util/probe-event.c | 431 ++++++++-- tools/perf/util/probe-event.h | 12 +- tools/perf/util/symbol.c | 10 +- tools/perf/util/symbol.h | 1 + 28 files changed, 4686 insertions(+), 944 deletions(-) -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/