From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752604AbbCZTRQ (ORCPT ); Thu, 26 Mar 2015 15:17:16 -0400 Received: from mail-pa0-f50.google.com ([209.85.220.50]:35269 "EHLO mail-pa0-f50.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751018AbbCZTRN (ORCPT ); Thu, 26 Mar 2015 15:17:13 -0400 From: Stephane Eranian To: linux-kernel@vger.kernel.org Cc: acme@redhat.com, peterz@infradead.org, mingo@elte.hu, ak@linux.intel.com, jolsa@redhat.com, namhyung@kernel.org, cel@us.ibm.com, sukadev@linux.vnet.ibm.com, sonnyrao@chromium.org, johnmccutchan@google.com, dsahern@gmail.com, adrian.hunter@intel.com, pawel.moll@arm.com Subject: [PATCH v5 0/4] perf: add support for profiling jitted code Date: Thu, 26 Mar 2015 20:17:05 +0100 Message-Id: <1427397429-14808-1-git-send-email-eranian@google.com> X-Mailer: git-send-email 1.9.1 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org This patch series extends perf record/report/annotate to enable profiling of jitted (just-in-time compiled) code. The current perf tool provides very limited support for profiling jitted code for some runtime environments. But the support is experimental and cannot be used in complex environments. It relies on files in /tmp, for instance. It does not support annotate mode or rejitted code. This patch series adds a better way of profiling jitted code with the following advantages: - support any jitted code environment (some with modifications) - support Java runtime with JVMTI interface with no modifications - provides a portable JVMTI agent library - known to support V8 runtime - known to support DART runtime - supports code rejitting and movements - no files in /tmp - meta-data file is unique to each run - no changes to perf report/annotate - support per-thread and system-wide profiling - support monitoring of multiple simultaneous Jit runtimes The support is based on cooperation with the runtime. For Java runtimes, supporting the JVMTI interface, there is no change necessary. For other runtimes, modifications are necessary to emit the meta-data necessary to support symbolization and annotation of the samples. Those modifications are fairly straighforward and already tested on V8 and DART. The jit environment emits a binary dump file which contains the jitted code (in raw format) and meta-data describing the mapping of functions. The binary format is documented in the jitdump.h header file. It is adapted from the OProfile jitdump format. To enable synchronization of the runtime MMAPs with those recorded by the kernel on behalf of the perf tool, the runtime needs to timestamp any record in the dump file using the same time source. This third version is using Pawel Moll's CLOCK_MONOTONIC patch posted https://lkml.org/lkml/2015/1/21/711 The patch is packaged in this series for convenience. Without this patch installed, records emitted by the runtime cannot be properly synchronized (inserted in the flow of MMAPS) and symbolization is not possible. The current support only works when the runtime is monitored from start to finish: perf record java --agentpath:libpfmjvmti.so my_class. Once the run is completed, the jitdump file needs to be injected into the perf.data file. This is accomplished by using the perf inject command. This will also generate an ELF image for each jitted function. The inject MMAP records will point to those ELF images. The reasoning behind using ELF images is that it makes processing for perf report and annotate automatic and transparent. It also makes it easier to package and analyze on a remote machine. The reporting is unchanged, simply invoke perf report or perf annotate on the modified perf.data file. The jitted code will appear symbolized and the assembly view will display the instruction level profile! As an added bonus, the series includes support for demangling function signature from OpenJDK. The current patch series does not include support for source line information. Therefore, the source code cannot yet be displayed in perf annotate. This will come in a later update. Furthermore, we believe there is a way to skip the perf inject phase and have perf report/annotate directly inject the MMAP records on the fly during processing of the perf.data file. Perf report would also generate the ELF files if necessary. Such optimization, would make using this extension seamless in system-wide mode and larger environments. This will be added in a later update as well. In V2, we have switchde to Pawell Moll and David Ahern posix clock kernel module instead. We have dropped the patch which modified the arguments to map_init() because the change was not used. We are not printing the return type of Java methods anymore and have made the Java demangler a separate module. We also rebased to 3.19.0+ from tip.git. In V3, we switched to Pawel Moll's CLOCK_MONOTONIC perf clock patches. This patch switch perf_events from sched_clock to CLOCK_MONOTONIC, a clock source which is available to users. In V4, we rebased to 4.0-rc5. We also simplified the process by getting rid of the requirement to pass the jitdump file name to perf inject. Now, perf injects automtically detects if jitdumps were generated and it merges the relevant meta-data. This is accomplished by having the jit runtime mmap the jitdump file for the purpose of creating a MMAP record in the perf.data file. That MMAP contains all the info to locate the jitdump file and generate the ELF images for jitted functions. In V5, we rebase to acme's perf/core branch (instead of tip.git). We fixed some bswap issues, switched to using scnprintf() and fixed formatting issues. Also made sure all the files were included in the patches. We also fix one error message in the JVMTI agent. To use the new feature: - compile and install Pawel's CLOCK_MONOTONIC patch: - compile the patched kernel - install the patched (and modules) - reboot the machine to this patched kernel - check that you have /proc/sys/kerne/perf_sample_time_clk_id - compile perf - cd tools/perf/jvmti; make; install wherever is appropriate Example using openJDK: $ perf record java -agentpath:libjvmti.so my_class $ perf inject -i perf.data -jit -o perf.data.jitted $ perf report -i perf.data.jitted Thanks to all the contributors and testers. Enjoy, Pawel Moll (1): perf: Use monotonic clock as a source for timestamps Stephane Eranian (3): perf tools: add Java demangling support perf inject: add jitdump mmap injection support perf tools: add JVMTI agent library Documentation/kernel-parameters.txt | 9 + kernel/events/core.c | 43 ++- tools/build/Makefile.feature | 2 + tools/build/feature/Makefile | 4 + tools/build/feature/test-libcrypto.c | 9 + tools/perf/Documentation/perf-inject.txt | 6 + tools/perf/builtin-inject.c | 60 +++- tools/perf/config/Makefile | 11 + tools/perf/jvmti/Makefile | 70 ++++ tools/perf/jvmti/jvmti_agent.c | 394 +++++++++++++++++++++ tools/perf/jvmti/jvmti_agent.h | 23 ++ tools/perf/jvmti/libjvmti.c | 149 ++++++++ tools/perf/util/Build | 3 + tools/perf/util/demangle-java.c | 199 +++++++++++ tools/perf/util/demangle-java.h | 10 + tools/perf/util/genelf.c | 479 +++++++++++++++++++++++++ tools/perf/util/genelf.h | 6 + tools/perf/util/jit.h | 15 + tools/perf/util/jitdump.c | 588 +++++++++++++++++++++++++++++++ tools/perf/util/jitdump.h | 92 +++++ tools/perf/util/symbol-elf.c | 3 + 21 files changed, 2161 insertions(+), 14 deletions(-) create mode 100644 tools/build/feature/test-libcrypto.c create mode 100644 tools/perf/jvmti/Makefile create mode 100644 tools/perf/jvmti/jvmti_agent.c create mode 100644 tools/perf/jvmti/jvmti_agent.h create mode 100644 tools/perf/jvmti/libjvmti.c create mode 100644 tools/perf/util/demangle-java.c create mode 100644 tools/perf/util/demangle-java.h create mode 100644 tools/perf/util/genelf.c create mode 100644 tools/perf/util/genelf.h create mode 100644 tools/perf/util/jit.h create mode 100644 tools/perf/util/jitdump.c create mode 100644 tools/perf/util/jitdump.h -- 1.9.1