From ad7ad6b5ddf63b436a5344fb686887f2d8b7cf3d Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Thu, 20 Oct 2022 18:25:09 +0300 Subject: perf scripts python: intel-pt-events.py: Add ability interleave output Intel PT timestamps are not provided for every branch, let alone every instruction, so there can be many samples with the same timestamp. With per-cpu contexts, decoding is done for each CPU in turn, which can make it difficult to see what is happening on different CPUs at the same time. Currently the interleaving from perf script --itrace=i0ns is quite coarse grained. There are often long stretches executing on one CPU and nothing on another. Some people are interested in seeing what happened on multiple CPUs before a crash to debug races etc. To improve perf script interleaving for parallel execution, the intel-pt-events.py script has been enhanced to enable interleaving the output with the same timestamp from different CPUs. It is understood that interleaving is not perfect or causal. Add parameter --interleave [] to interleave sample output for the same timestamp so that no more than n samples for a CPU are displayed in a row. 'n' defaults to 4. Note this only affects the order of output, and only when the timestamp is the same. Example: $ perf script intel-pt-events.py --insn-trace --interleave 3 ... bash 2267/2267 [004] 9323.692625625 563caa3c86f0 jz 0x563caa3c89c7 run_pending_traps+0x30 (/usr/bin/bash) IPC: 1.52 (38/25) bash 2267/2267 [004] 9323.692625625 563caa3c89c7 movq 0x118(%rsp), %rax run_pending_traps+0x307 (/usr/bin/bash) bash 2267/2267 [004] 9323.692625625 563caa3c89cf subq %fs:0x28, %rax run_pending_traps+0x30f (/usr/bin/bash) bash 2270/2270 [007] 9323.692625625 55dc58cabf02 jz 0x55dc58cabf48 unquoted_glob_pattern_p+0x102 (/usr/bin/bash) IPC: 1.56 (25/16) bash 2270/2270 [007] 9323.692625625 55dc58cabf04 cmp $0x5d, %al unquoted_glob_pattern_p+0x104 (/usr/bin/bash) bash 2270/2270 [007] 9323.692625625 55dc58cabf06 jnz 0x55dc58cabf10 unquoted_glob_pattern_p+0x106 (/usr/bin/bash) bash 2264/2264 [001] 9323.692625625 7fd556a4376c jbe 0x7fd556a43ac8 round_and_return+0x3fc (/usr/lib/x86_64-linux-gnu/libc.so.6) IPC: 4.30 (43/10) bash 2264/2264 [001] 9323.692625625 7fd556a43772 and $0x8, %edx round_and_return+0x402 (/usr/lib/x86_64-linux-gnu/libc.so.6) bash 2264/2264 [001] 9323.692625625 7fd556a43775 jnz 0x7fd556a43ac8 round_and_return+0x405 (/usr/lib/x86_64-linux-gnu/libc.so.6) bash 2267/2267 [004] 9323.692625625 563caa3c89d8 jnz 0x563caa3c8b11 run_pending_traps+0x318 (/usr/bin/bash) bash 2267/2267 [004] 9323.692625625 563caa3c89de add $0x128, %rsp run_pending_traps+0x31e (/usr/bin/bash) bash 2267/2267 [004] 9323.692625625 563caa3c89e5 popq %rbx run_pending_traps+0x325 (/usr/bin/bash) ... Signed-off-by: Adrian Hunter Cc: Andi Kleen Cc: Ian Rogers Cc: Jiri Olsa Cc: Namhyung Kim Link: https://lore.kernel.org/r/20221020152509.5298-1-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/scripts/python/intel-pt-events.py | 65 +++++++++++++++++++++++++++- 1 file changed, 64 insertions(+), 1 deletion(-) (limited to 'tools/perf/scripts/python') diff --git a/tools/perf/scripts/python/intel-pt-events.py b/tools/perf/scripts/python/intel-pt-events.py index 6be7fd8fd615..08862a2582f4 100644 --- a/tools/perf/scripts/python/intel-pt-events.py +++ b/tools/perf/scripts/python/intel-pt-events.py @@ -13,10 +13,12 @@ from __future__ import print_function +import io import os import sys import struct import argparse +import contextlib from libxed import LibXED from ctypes import create_string_buffer, addressof @@ -39,6 +41,11 @@ glb_src = False glb_source_file_name = None glb_line_number = None glb_dso = None +glb_stash_dict = {} +glb_output = None +glb_output_pos = 0 +glb_cpu = -1 +glb_time = 0 def get_optional_null(perf_dict, field): if field in perf_dict: @@ -70,6 +77,7 @@ def trace_begin(): ap.add_argument("--insn-trace", action='store_true') ap.add_argument("--src-trace", action='store_true') ap.add_argument("--all-switch-events", action='store_true') + ap.add_argument("--interleave", type=int, nargs='?', const=4, default=0) global glb_args global glb_insn global glb_src @@ -94,11 +102,39 @@ def trace_begin(): perf_set_itrace_options(perf_script_context, itrace) def trace_end(): + if glb_args.interleave: + flush_stashed_output() print("End") def trace_unhandled(event_name, context, event_fields_dict): print(' '.join(['%s=%s'%(k,str(v))for k,v in sorted(event_fields_dict.items())])) +def stash_output(): + global glb_stash_dict + global glb_output_pos + output_str = glb_output.getvalue()[glb_output_pos:] + n = len(output_str) + if n: + glb_output_pos += n + if glb_cpu not in glb_stash_dict: + glb_stash_dict[glb_cpu] = [] + glb_stash_dict[glb_cpu].append(output_str) + +def flush_stashed_output(): + global glb_stash_dict + while glb_stash_dict: + cpus = list(glb_stash_dict.keys()) + # Output at most glb_args.interleave output strings per cpu + for cpu in cpus: + items = glb_stash_dict[cpu] + countdown = glb_args.interleave + while len(items) and countdown: + sys.stdout.write(items[0]) + del items[0] + countdown -= 1 + if not items: + del glb_stash_dict[cpu] + def print_ptwrite(raw_buf): data = struct.unpack_from("= 2 and x[0]: machine_pid = x[0] vcpu = x[1] @@ -403,6 +464,8 @@ def auxtrace_error(typ, code, cpu, pid, tid, ip, ts, msg, cpumode, *x): sys.exit(1) def context_switch(ts, cpu, pid, tid, np_pid, np_tid, machine_pid, out, out_preempt, *x): + if glb_args.interleave: + flush_stashed_output() if out: out_str = "Switch out " else: -- cgit From 378ef0f5d9d7f4652d7a40e0711e8b845ada1cbd Mon Sep 17 00:00:00 2001 From: Ian Rogers Date: Mon, 5 Dec 2022 14:59:39 -0800 Subject: perf build: Use libtraceevent from the system MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Remove the LIBTRACEEVENT_DYNAMIC and LIBTRACEFS_DYNAMIC make command line variables. If libtraceevent isn't installed or NO_LIBTRACEEVENT=1 is passed to the build, don't compile in libtraceevent and libtracefs support. This also disables CONFIG_TRACE that controls "perf trace". CONFIG_LIBTRACEEVENT is used to control enablement in Build/Makefiles, HAVE_LIBTRACEEVENT is used in C code. Without HAVE_LIBTRACEEVENT tracepoints are disabled and as such the commands kmem, kwork, lock, sched and timechart are removed. The majority of commands continue to work including "perf test". Committer notes: Fixed up a tools/perf/util/Build reject and added: #include to tools/perf/util/scripting-engines/trace-event-perl.c. Committer testing: $ rpm -qi libtraceevent-devel Name : libtraceevent-devel Version : 1.5.3 Release : 2.fc36 Architecture: x86_64 Install Date: Mon 25 Jul 2022 03:20:19 PM -03 Group : Unspecified Size : 27728 License : LGPLv2+ and GPLv2+ Signature : RSA/SHA256, Fri 15 Apr 2022 02:11:58 PM -03, Key ID 999f7cbf38ab71f4 Source RPM : libtraceevent-1.5.3-2.fc36.src.rpm Build Date : Fri 15 Apr 2022 10:57:01 AM -03 Build Host : buildvm-x86-05.iad2.fedoraproject.org Packager : Fedora Project Vendor : Fedora Project URL : https://git.kernel.org/pub/scm/libs/libtrace/libtraceevent.git/ Bug URL : https://bugz.fedoraproject.org/libtraceevent Summary : Development headers of libtraceevent Description : Development headers of libtraceevent-libs $ Default build: $ ldd ~/bin/perf | grep tracee libtraceevent.so.1 => /lib64/libtraceevent.so.1 (0x00007f1dcaf8f000) $ # perf trace -e sched:* --max-events 10 0.000 migration/0/17 sched:sched_migrate_task(comm: "", pid: 1603763 (perf), prio: 120, dest_cpu: 1) 0.005 migration/0/17 sched:sched_wake_idle_without_ipi(cpu: 1) 0.011 migration/0/17 sched:sched_switch(prev_comm: "", prev_pid: 17 (migration/0), prev_state: 1, next_comm: "", next_prio: 120) 1.173 :0/0 sched:sched_wakeup(comm: "", pid: 3138 (gnome-terminal-), prio: 120) 1.180 :0/0 sched:sched_switch(prev_comm: "", prev_prio: 120, next_comm: "", next_pid: 3138 (gnome-terminal-), next_prio: 120) 0.156 migration/1/21 sched:sched_migrate_task(comm: "", pid: 1603763 (perf), prio: 120, orig_cpu: 1, dest_cpu: 2) 0.160 migration/1/21 sched:sched_wake_idle_without_ipi(cpu: 2) 0.166 migration/1/21 sched:sched_switch(prev_comm: "", prev_pid: 21 (migration/1), prev_state: 1, next_comm: "", next_prio: 120) 1.183 :0/0 sched:sched_wakeup(comm: "", pid: 1602985 (kworker/u16:0-f), prio: 120, target_cpu: 1) 1.186 :0/0 sched:sched_switch(prev_comm: "", prev_prio: 120, next_comm: "", next_pid: 1602985 (kworker/u16:0-f), next_prio: 120) # Had to tweak tools/perf/util/setup.py to make sure the python binding shared object links with libtraceevent if -DHAVE_LIBTRACEEVENT is present in CFLAGS. Building with NO_LIBTRACEEVENT=1 uncovered some more build failures: - Make building of data-convert-bt.c to CONFIG_LIBTRACEEVENT=y - perf-$(CONFIG_LIBTRACEEVENT) += scripts/ - bpf_kwork.o needs also to be dependent on CONFIG_LIBTRACEEVENT=y - The python binding needed some fixups and util/trace-event.c can't be built and linked with the python binding shared object, so remove it in tools/perf/util/setup.py and exclude it from the list of dependencies in the python/perf.so Makefile.perf target. Building without libtraceevent-devel installed uncovered more build failures: - The python binding tools/perf/util/python.c was assuming that traceevent/parse-events.h was always available, which was the case when we defaulted to using the in-kernel tools/lib/traceevent/ files, now we need to enclose it under ifdef HAVE_LIBTRACEEVENT, just like the other parts of it that deal with tracepoints. - We have to ifdef the rules in the Build files with CONFIG_LIBTRACEEVENT=y to build builtin-trace.c and tools/perf/trace/beauty/ as we only ifdef setting CONFIG_TRACE=y when setting NO_LIBTRACEEVENT=1 in the make command line, not when we don't detect libtraceevent-devel installed in the system. Simplification here to avoid these two ways of disabling builtin-trace.c and not having CONFIG_TRACE=y when libtraceevent-devel isn't installed is the clean way. From Athira: tools/perf/arch/powerpc/util/Build -perf-y += kvm-stat.o +perf-$(CONFIG_LIBTRACEEVENT) += kvm-stat.o Then, ditto for arm64 and s390, detected by container cross build tests. - s/390 uses test__checkevent_tracepoint() that is now only available if HAVE_LIBTRACEEVENT is defined, enclose the callsite with ifder HAVE_LIBTRACEEVENT. Also from Athira: With this change, I could successfully compile in these environment: - Without libtraceevent-devel installed - With libtraceevent-devel installed - With “make NO_LIBTRACEEVENT=1” Then, finally rename CONFIG_TRACEEVENT to CONFIG_LIBTRACEEVENT for consistency with other libraries detected in tools/perf/. Signed-off-by: Ian Rogers Tested-by: Arnaldo Carvalho de Melo Tested-by: Athira Rajeev Cc: Alexander Shishkin Cc: Jiri Olsa Cc: Mark Rutland Cc: Namhyung Kim Cc: Nick Desaulniers Cc: Peter Zijlstra Cc: Stephane Eranian Cc: bpf@vger.kernel.org Link: http://lore.kernel.org/lkml/20221205225940.3079667-3-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/Build | 20 ++-- tools/perf/Makefile.config | 37 ++++---- tools/perf/Makefile.perf | 104 ++------------------- tools/perf/arch/arm64/util/Build | 2 +- tools/perf/arch/powerpc/util/Build | 2 +- tools/perf/arch/s390/util/Build | 2 +- tools/perf/arch/x86/util/Build | 2 +- tools/perf/arch/x86/util/intel-pt.c | 4 + tools/perf/builtin-annotate.c | 2 + tools/perf/builtin-data.c | 5 +- tools/perf/builtin-inject.c | 8 ++ tools/perf/builtin-kmem.c | 1 + tools/perf/builtin-kvm.c | 12 +-- tools/perf/builtin-kwork.c | 1 + tools/perf/builtin-record.c | 2 + tools/perf/builtin-report.c | 9 +- tools/perf/builtin-script.c | 19 +++- tools/perf/builtin-timechart.c | 1 + tools/perf/builtin-trace.c | 5 +- tools/perf/builtin-version.c | 1 + tools/perf/perf.c | 24 +++-- tools/perf/scripts/python/Perf-Trace-Util/Build | 2 +- tools/perf/tests/Build | 12 +-- tools/perf/tests/builtin-test.c | 6 ++ tools/perf/tests/parse-events.c | 23 ++++- tools/perf/util/Build | 21 +++-- tools/perf/util/data-convert-bt.c | 5 +- tools/perf/util/data-convert-json.c | 9 +- tools/perf/util/evlist.c | 6 +- tools/perf/util/evlist.h | 4 + tools/perf/util/evsel.c | 11 ++- tools/perf/util/evsel.h | 12 ++- tools/perf/util/evsel_fprintf.c | 7 +- tools/perf/util/header.c | 19 ++++ tools/perf/util/header.h | 2 + tools/perf/util/intel-pt.c | 7 +- tools/perf/util/parse-events.c | 15 +++ tools/perf/util/parse-events.h | 1 - tools/perf/util/python.c | 10 ++ tools/perf/util/scripting-engines/Build | 6 +- .../perf/util/scripting-engines/trace-event-perl.c | 1 + .../util/scripting-engines/trace-event-python.c | 1 + tools/perf/util/session.c | 2 + tools/perf/util/session.h | 2 + tools/perf/util/setup.py | 10 +- tools/perf/util/sort.c | 60 ++++++++++-- tools/perf/util/synthetic-events.c | 6 ++ tools/perf/util/trace-event-parse.c | 2 + tools/perf/util/trace-event-read.c | 1 + tools/perf/util/trace-event-scripting.c | 1 + tools/perf/util/trace-event.c | 1 - tools/perf/util/trace-event.h | 11 ++- 52 files changed, 355 insertions(+), 184 deletions(-) (limited to 'tools/perf/scripts/python') diff --git a/tools/perf/Build b/tools/perf/Build index 496b096153bb..6dd67e502295 100644 --- a/tools/perf/Build +++ b/tools/perf/Build @@ -5,7 +5,6 @@ perf-y += builtin-diff.o perf-y += builtin-evlist.o perf-y += builtin-ftrace.o perf-y += builtin-help.o -perf-y += builtin-sched.o perf-y += builtin-buildid-list.o perf-y += builtin-buildid-cache.o perf-y += builtin-kallsyms.o @@ -13,11 +12,8 @@ perf-y += builtin-list.o perf-y += builtin-record.o perf-y += builtin-report.o perf-y += builtin-stat.o -perf-y += builtin-timechart.o perf-y += builtin-top.o perf-y += builtin-script.o -perf-y += builtin-kmem.o -perf-y += builtin-lock.o perf-y += builtin-kvm.o perf-y += builtin-inject.o perf-y += builtin-mem.o @@ -25,9 +21,18 @@ perf-y += builtin-data.o perf-y += builtin-version.o perf-y += builtin-c2c.o perf-y += builtin-daemon.o -perf-y += builtin-kwork.o -perf-$(CONFIG_TRACE) += builtin-trace.o +perf-$(CONFIG_LIBTRACEEVENT) += builtin-kmem.o +perf-$(CONFIG_LIBTRACEEVENT) += builtin-kwork.o +perf-$(CONFIG_LIBTRACEEVENT) += builtin-lock.o +perf-$(CONFIG_LIBTRACEEVENT) += builtin-sched.o +perf-$(CONFIG_LIBTRACEEVENT) += builtin-timechart.o + +ifeq ($(CONFIG_LIBTRACEEVENT),y) + perf-$(CONFIG_TRACE) += builtin-trace.o + perf-$(CONFIG_TRACE) += trace/beauty/ +endif + perf-$(CONFIG_LIBELF) += builtin-probe.o perf-y += bench/ @@ -51,7 +56,6 @@ CFLAGS_builtin-report.o += -DDOCDIR="BUILD_STR($(srcdir_SQ)/Documentation)" perf-y += util/ perf-y += arch/ perf-y += ui/ -perf-y += scripts/ -perf-$(CONFIG_TRACE) += trace/beauty/ +perf-$(CONFIG_LIBTRACEEVENT) += scripts/ gtk-y += ui/gtk/ diff --git a/tools/perf/Makefile.config b/tools/perf/Makefile.config index 9cc3c48f3288..680228e19c1a 100644 --- a/tools/perf/Makefile.config +++ b/tools/perf/Makefile.config @@ -763,18 +763,20 @@ ifndef NO_LIBUNWIND EXTLIBS += $(EXTLIBS_LIBUNWIND) endif -ifeq ($(NO_SYSCALL_TABLE),0) - $(call detected,CONFIG_TRACE) -else - ifndef NO_LIBAUDIT - $(call feature_check,libaudit) - ifneq ($(feature-libaudit), 1) - msg := $(warning No libaudit.h found, disables 'trace' tool, please install audit-libs-devel or libaudit-dev); - NO_LIBAUDIT := 1 - else - CFLAGS += -DHAVE_LIBAUDIT_SUPPORT - EXTLIBS += -laudit - $(call detected,CONFIG_TRACE) +ifneq ($(NO_LIBTRACEEVENT),1) + ifeq ($(NO_SYSCALL_TABLE),0) + $(call detected,CONFIG_TRACE) + else + ifndef NO_LIBAUDIT + $(call feature_check,libaudit) + ifneq ($(feature-libaudit), 1) + msg := $(warning No libaudit.h found, disables 'trace' tool, please install audit-libs-devel or libaudit-dev); + NO_LIBAUDIT := 1 + else + CFLAGS += -DHAVE_LIBAUDIT_SUPPORT + EXTLIBS += -laudit + $(call detected,CONFIG_TRACE) + endif endif endif endif @@ -1182,9 +1184,11 @@ ifdef LIBPFM4 endif endif -ifdef LIBTRACEEVENT_DYNAMIC +# libtraceevent is a recommended dependency picked up from the system. +ifneq ($(NO_LIBTRACEEVENT),1) $(call feature_check,libtraceevent) ifeq ($(feature-libtraceevent), 1) + CFLAGS += -DHAVE_LIBTRACEEVENT EXTLIBS += -ltraceevent LIBTRACEEVENT_VERSION := $(shell $(PKG_CONFIG) --modversion libtraceevent) LIBTRACEEVENT_VERSION_1 := $(word 1, $(subst ., ,$(LIBTRACEEVENT_VERSION))) @@ -1192,12 +1196,11 @@ ifdef LIBTRACEEVENT_DYNAMIC LIBTRACEEVENT_VERSION_3 := $(word 3, $(subst ., ,$(LIBTRACEEVENT_VERSION))) LIBTRACEEVENT_VERSION_CPP := $(shell expr $(LIBTRACEEVENT_VERSION_1) \* 255 \* 255 + $(LIBTRACEEVENT_VERSION_2) \* 255 + $(LIBTRACEEVENT_VERSION_3)) CFLAGS += -DLIBTRACEEVENT_VERSION=$(LIBTRACEEVENT_VERSION_CPP) + $(call detected,CONFIG_LIBTRACEEVENT) else - dummy := $(error Error: No libtraceevent devel library found, please install libtraceevent-devel); + dummy := $(warning Warning: libtraceevent is missing limiting functionality, please install libtraceevent-dev/libtraceevent-devel) endif -endif -ifdef LIBTRACEFS_DYNAMIC $(call feature_check,libtracefs) ifeq ($(feature-libtracefs), 1) EXTLIBS += -ltracefs @@ -1207,8 +1210,6 @@ ifdef LIBTRACEFS_DYNAMIC LIBTRACEFS_VERSION_3 := $(word 3, $(subst ., ,$(LIBTRACEFS_VERSION))) LIBTRACEFS_VERSION_CPP := $(shell expr $(LIBTRACEFS_VERSION_1) \* 255 \* 255 + $(LIBTRACEFS_VERSION_2) \* 255 + $(LIBTRACEFS_VERSION_3)) CFLAGS += -DLIBTRACEFS_VERSION=$(LIBTRACEFS_VERSION_CPP) - else - dummy := $(error Error: No libtracefs devel library found, please install libtracefs-dev); endif endif diff --git a/tools/perf/Makefile.perf b/tools/perf/Makefile.perf index 6689f644782f..98f629bbd1aa 100644 --- a/tools/perf/Makefile.perf +++ b/tools/perf/Makefile.perf @@ -128,10 +128,6 @@ include ../scripts/utilities.mak # # Define BUILD_BPF_SKEL to enable BPF skeletons # -# Define LIBTRACEEVENT_DYNAMIC to enable libtraceevent dynamic linking -# -# Define LIBTRACEFS_DYNAMIC to enable libtracefs dynamic linking -# # As per kernel Makefile, avoid funny character set dependencies unexport LC_ALL @@ -242,10 +238,6 @@ sub-make: fixdep else # force_fixdep LIBAPI_DIR = $(srctree)/tools/lib/api/ -ifndef LIBTRACEEVENT_DYNAMIC -LIBTRACEEVENT_DIR = $(srctree)/tools/lib/traceevent/ -LIBTRACEEVENT_PLUGINS_DIR = $(LIBTRACEEVENT_DIR)/plugins -endif LIBBPF_DIR = $(srctree)/tools/lib/bpf/ LIBSUBCMD_DIR = $(srctree)/tools/lib/subcmd/ LIBSYMBOL_DIR = $(srctree)/tools/lib/symbol/ @@ -295,31 +287,6 @@ SCRIPT_SH += perf-iostat.sh grep-libs = $(filter -l%,$(1)) strip-libs = $(filter-out -l%,$(1)) -ifndef LIBTRACEEVENT_DYNAMIC -ifneq ($(OUTPUT),) - LIBTRACEEVENT_OUTPUT = $(abspath $(OUTPUT))/libtraceevent -else - LIBTRACEEVENT_OUTPUT = $(CURDIR)/libtraceevent -endif -LIBTRACEEVENT_PLUGINS_OUTPUT = $(LIBTRACEEVENT_OUTPUT)_plugins -LIBTRACEEVENT_DESTDIR = $(LIBTRACEEVENT_OUTPUT) -LIBTRACEEVENT_PLUGINS_DESTDIR = $(LIBTRACEEVENT_PLUGINS_OUTPUT) -LIBTRACEEVENT_INCLUDE = $(LIBTRACEEVENT_DESTDIR)/include -LIBTRACEEVENT = $(LIBTRACEEVENT_OUTPUT)/libtraceevent.a -export LIBTRACEEVENT -LIBTRACEEVENT_DYNAMIC_LIST = $(LIBTRACEEVENT_PLUGINS_OUTPUT)/libtraceevent-dynamic-list -CFLAGS += -I$(LIBTRACEEVENT_OUTPUT)/include -# -# The static build has no dynsym table, so this does not work for -# static build. Looks like linker starts to scream about that now -# (in Fedora 26) so we need to switch it off for static build. -DYNAMIC_LIST_LDFLAGS = -Xlinker --dynamic-list=$(LIBTRACEEVENT_DYNAMIC_LIST) -LIBTRACEEVENT_DYNAMIC_LIST_LDFLAGS = $(if $(findstring -static,$(LDFLAGS)),,$(DYNAMIC_LIST_LDFLAGS)) -else -LIBTRACEEVENT_DYNAMIC_LIST = -LIBTRACEEVENT_DYNAMIC_LIST_LDFLAGS = -endif - ifneq ($(OUTPUT),) LIBAPI_OUTPUT = $(abspath $(OUTPUT))/libapi else @@ -380,13 +347,14 @@ export PYTHON_EXTBUILD_LIB PYTHON_EXTBUILD_TMP python-clean := $(call QUIET_CLEAN, python) $(RM) -r $(PYTHON_EXTBUILD) $(OUTPUT)python/perf*.so -PYTHON_EXT_SRCS := $(shell grep -v ^\# util/python-ext-sources) -ifndef LIBTRACEEVENT_DYNAMIC -PYTHON_EXT_DEPS := util/python-ext-sources util/setup.py $(LIBTRACEEVENT) $(LIBAPI) +ifeq ($(CONFIG_LIBTRACEEVENT),y) + PYTHON_EXT_SRCS := $(shell grep -v ^\# util/python-ext-sources) else -PYTHON_EXT_DEPS := util/python-ext-sources util/setup.py $(LIBAPI) + PYTHON_EXT_SRCS := $(shell grep -v '^\#\|util/trace-event.c' util/python-ext-sources) endif +PYTHON_EXT_DEPS := util/python-ext-sources util/setup.py $(LIBAPI) + SCRIPTS = $(patsubst %.sh,%,$(SCRIPT_SH)) PROGRAMS += $(OUTPUT)perf @@ -430,9 +398,6 @@ ifndef NO_LIBBPF PERFLIBS += $(LIBBPF) endif endif -ifndef LIBTRACEEVENT_DYNAMIC - PERFLIBS += $(LIBTRACEEVENT) -endif # We choose to avoid "if .. else if .. else .. endif endif" # because maintaining the nesting to match is a pain. If @@ -682,9 +647,9 @@ all: shell_compatibility_test $(ALL_PROGRAMS) $(LANG_BINDINGS) $(OTHER_PROGRAMS) # Create python binding output directory if not already present _dummy := $(shell [ -d '$(OUTPUT)python' ] || mkdir -p '$(OUTPUT)python') -$(OUTPUT)python/perf.so: $(PYTHON_EXT_SRCS) $(PYTHON_EXT_DEPS) $(LIBTRACEEVENT_DYNAMIC_LIST) $(LIBPERF) +$(OUTPUT)python/perf.so: $(PYTHON_EXT_SRCS) $(PYTHON_EXT_DEPS) $(LIBPERF) $(QUIET_GEN)LDSHARED="$(CC) -pthread -shared" \ - CFLAGS='$(CFLAGS)' LDFLAGS='$(LDFLAGS) $(LIBTRACEEVENT_DYNAMIC_LIST_LDFLAGS)' \ + CFLAGS='$(CFLAGS)' LDFLAGS='$(LDFLAGS)' \ $(PYTHON_WORD) util/setup.py \ --quiet build_ext; \ cp $(PYTHON_EXTBUILD_LIB)perf*.so $(OUTPUT)python/ @@ -710,8 +675,8 @@ $(PERF_IN): prepare FORCE $(PMU_EVENTS_IN): FORCE prepare $(Q)$(MAKE) -f $(srctree)/tools/build/Makefile.build dir=pmu-events obj=pmu-events -$(OUTPUT)perf: $(PERFLIBS) $(PERF_IN) $(PMU_EVENTS_IN) $(LIBTRACEEVENT_DYNAMIC_LIST) - $(QUIET_LINK)$(CC) $(CFLAGS) $(LDFLAGS) $(LIBTRACEEVENT_DYNAMIC_LIST_LDFLAGS) \ +$(OUTPUT)perf: $(PERFLIBS) $(PERF_IN) $(PMU_EVENTS_IN) + $(QUIET_LINK)$(CC) $(CFLAGS) $(LDFLAGS) \ $(PERF_IN) $(PMU_EVENTS_IN) $(LIBS) -o $@ $(GTK_IN): FORCE prepare @@ -797,10 +762,6 @@ prepare: $(OUTPUT)PERF-VERSION-FILE $(OUTPUT)common-cmds.h archheaders $(drm_ioc $(LIBSYMBOL) \ bpf-skel -ifndef LIBTRACEEVENT_DYNAMIC -prepare: $(LIBTRACEEVENT) -endif - $(OUTPUT)%.o: %.c prepare FORCE $(Q)$(MAKE) -f $(srctree)/tools/build/Makefile.build dir=$(build-dir) $@ @@ -856,38 +817,6 @@ endif $(patsubst perf-%,%.o,$(PROGRAMS)): $(wildcard */*.h) -ifndef LIBTRACEEVENT_DYNAMIC -LIBTRACEEVENT_FLAGS += plugin_dir=$(plugindir_SQ) 'EXTRA_CFLAGS=$(EXTRA_CFLAGS)' 'LDFLAGS=$(filter-out -static,$(LDFLAGS))' - -$(LIBTRACEEVENT): FORCE | $(LIBTRACEEVENT_OUTPUT) - $(Q)$(MAKE) -C $(LIBTRACEEVENT_DIR) O=$(LIBTRACEEVENT_OUTPUT) \ - DESTDIR=$(LIBTRACEEVENT_DESTDIR) prefix= \ - $@ install_headers - -$(LIBTRACEEVENT)-clean: - $(call QUIET_CLEAN, libtraceevent) - $(Q)$(RM) -r -- $(LIBTRACEEVENT_OUTPUT) - -libtraceevent_plugins: FORCE | $(LIBTRACEEVENT_PLUGINS_OUTPUT) - $(Q)$(MAKE) -C $(LIBTRACEEVENT_PLUGINS_DIR) O=$(LIBTRACEEVENT_PLUGINS_OUTPUT) \ - DESTDIR=$(LIBTRACEEVENT_PLUGINS_DESTDIR) prefix= \ - plugins - -libtraceevent_plugins-clean: - $(call QUIET_CLEAN, libtraceevent_plugins) - $(Q)$(RM) -r -- $(LIBTRACEEVENT_PLUGINS_OUTPUT) - -$(LIBTRACEEVENT_DYNAMIC_LIST): libtraceevent_plugins - $(Q)$(MAKE) -C $(LIBTRACEEVENT_PLUGINS_DIR) O=$(LIBTRACEEVENT_PLUGINS_OUTPUT) \ - DESTDIR=$(LIBTRACEEVENT_PLUGINS_DESTDIR) prefix= \ - $(LIBTRACEEVENT_FLAGS) $@ - -install-traceevent-plugins: libtraceevent_plugins - $(Q)$(MAKE) -C $(LIBTRACEEVENT_PLUGINS_DIR) O=$(LIBTRACEEVENT_PLUGINS_OUTPUT) \ - DESTDIR=$(DESTDIR_SQ)$(prefix) prefix= \ - $(LIBTRACEEVENT_FLAGS) install -endif - $(LIBAPI): FORCE | $(LIBAPI_OUTPUT) $(Q)$(MAKE) -C $(LIBAPI_DIR) O=$(LIBAPI_OUTPUT) \ DESTDIR=$(LIBAPI_DESTDIR) prefix= \ @@ -1095,10 +1024,6 @@ install-tests: all install-gtk install-bin: install-tools install-tests -ifndef LIBTRACEEVENT_DYNAMIC -install-bin: install-traceevent-plugins -endif - install: install-bin try-install-man install-python_ext: @@ -1124,11 +1049,6 @@ SKELETONS += $(SKEL_OUT)/kwork_trace.skel.h $(SKEL_TMP_OUT) $(LIBAPI_OUTPUT) $(LIBBPF_OUTPUT) $(LIBPERF_OUTPUT) $(LIBSUBCMD_OUTPUT) $(LIBSYMBOL_OUTPUT): $(Q)$(MKDIR) -p $@ -ifndef LIBTRACEEVENT_DYNAMIC -$(LIBTRACEEVENT_OUTPUT) $(LIBTRACEEVENT_PLUGINS_OUTPUT): - $(Q)$(MKDIR) -p $@ -endif - ifdef BUILD_BPF_SKEL BPFTOOL := $(SKEL_TMP_OUT)/bootstrap/bpftool BPF_INCLUDE := -I$(SKEL_TMP_OUT)/.. -I$(LIBBPF_INCLUDE) @@ -1211,10 +1131,6 @@ clean:: $(LIBAPI)-clean $(LIBBPF)-clean $(LIBSUBCMD)-clean $(LIBSYMBOL)-clean $( $(call QUIET_CLEAN, Documentation) \ $(MAKE) -C $(DOC_DIR) O=$(OUTPUT) clean >/dev/null -ifndef LIBTRACEEVENT_DYNAMIC -clean:: $(LIBTRACEEVENT)-clean libtraceevent_plugins-clean -endif - # # To provide FEATURE-DUMP into $(FEATURE_DUMP_COPY) # file if defined, with no further action. @@ -1232,6 +1148,6 @@ FORCE: .PHONY: all install clean config-clean strip install-gtk .PHONY: shell_compatibility_test please_set_SHELL_PATH_to_a_more_modern_shell .PHONY: .FORCE-PERF-VERSION-FILE TAGS tags cscope FORCE prepare -.PHONY: libtraceevent_plugins archheaders +.PHONY: archheaders endif # force_fixdep diff --git a/tools/perf/arch/arm64/util/Build b/tools/perf/arch/arm64/util/Build index 337aa9bdf905..78ef7115be3d 100644 --- a/tools/perf/arch/arm64/util/Build +++ b/tools/perf/arch/arm64/util/Build @@ -3,7 +3,7 @@ perf-y += machine.o perf-y += perf_regs.o perf-y += tsc.o perf-y += pmu.o -perf-y += kvm-stat.o +perf-$(CONFIG_LIBTRACEEVENT) += kvm-stat.o perf-$(CONFIG_DWARF) += dwarf-regs.o perf-$(CONFIG_LOCAL_LIBUNWIND) += unwind-libunwind.o perf-$(CONFIG_LIBDW_DWARF_UNWIND) += unwind-libdw.o diff --git a/tools/perf/arch/powerpc/util/Build b/tools/perf/arch/powerpc/util/Build index 0115f3166568..9889245c555c 100644 --- a/tools/perf/arch/powerpc/util/Build +++ b/tools/perf/arch/powerpc/util/Build @@ -1,5 +1,5 @@ perf-y += header.o -perf-y += kvm-stat.o +perf-$(CONFIG_LIBTRACEEVENT) += kvm-stat.o perf-y += perf_regs.o perf-y += mem-events.o perf-y += sym-handling.o diff --git a/tools/perf/arch/s390/util/Build b/tools/perf/arch/s390/util/Build index 3d9d0f4f72ca..db6884086997 100644 --- a/tools/perf/arch/s390/util/Build +++ b/tools/perf/arch/s390/util/Build @@ -1,5 +1,5 @@ perf-y += header.o -perf-y += kvm-stat.o +perf-$(CONFIG_LIBTRACEEVENT) += kvm-stat.o perf-y += perf_regs.o perf-$(CONFIG_DWARF) += dwarf-regs.o diff --git a/tools/perf/arch/x86/util/Build b/tools/perf/arch/x86/util/Build index dbeb04cb336e..195ccfdef7aa 100644 --- a/tools/perf/arch/x86/util/Build +++ b/tools/perf/arch/x86/util/Build @@ -1,7 +1,7 @@ perf-y += header.o perf-y += tsc.o perf-y += pmu.o -perf-y += kvm-stat.o +perf-$(CONFIG_LIBTRACEEVENT) += kvm-stat.o perf-y += perf_regs.o perf-y += topdown.o perf-y += machine.o diff --git a/tools/perf/arch/x86/util/intel-pt.c b/tools/perf/arch/x86/util/intel-pt.c index af102f471e9f..1e39a034cee9 100644 --- a/tools/perf/arch/x86/util/intel-pt.c +++ b/tools/perf/arch/x86/util/intel-pt.c @@ -418,6 +418,7 @@ static int intel_pt_info_fill(struct auxtrace_record *itr, return 0; } +#ifdef HAVE_LIBTRACEEVENT static int intel_pt_track_switches(struct evlist *evlist) { const char *sched_switch = "sched:sched_switch"; @@ -439,6 +440,7 @@ static int intel_pt_track_switches(struct evlist *evlist) return 0; } +#endif static void intel_pt_valid_str(char *str, size_t len, u64 valid) { @@ -829,6 +831,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr, ptr->have_sched_switch = 2; } } else { +#ifdef HAVE_LIBTRACEEVENT err = intel_pt_track_switches(evlist); if (err == -EPERM) pr_debug2("Unable to select sched:sched_switch\n"); @@ -836,6 +839,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr, return err; else ptr->have_sched_switch = 1; +#endif } } diff --git a/tools/perf/builtin-annotate.c b/tools/perf/builtin-annotate.c index 517d928c00e3..90458ca6933f 100644 --- a/tools/perf/builtin-annotate.c +++ b/tools/perf/builtin-annotate.c @@ -499,7 +499,9 @@ int cmd_annotate(int argc, const char **argv) .namespaces = perf_event__process_namespaces, .attr = perf_event__process_attr, .build_id = perf_event__process_build_id, +#ifdef HAVE_LIBTRACEEVENT .tracing_data = perf_event__process_tracing_data, +#endif .id_index = perf_event__process_id_index, .auxtrace_info = perf_event__process_auxtrace_info, .auxtrace = perf_event__process_auxtrace, diff --git a/tools/perf/builtin-data.c b/tools/perf/builtin-data.c index c22d82d2a73c..b2a9a3b7f68d 100644 --- a/tools/perf/builtin-data.c +++ b/tools/perf/builtin-data.c @@ -78,12 +78,13 @@ static int cmd_data_convert(int argc, const char **argv) return bt_convert__perf2json(input_name, to_json, &opts); if (to_ctf) { -#ifdef HAVE_LIBBABELTRACE_SUPPORT +#if defined(HAVE_LIBBABELTRACE_SUPPORT) && defined(HAVE_LIBTRACEEVENT) return bt_convert__perf2ctf(input_name, to_ctf, &opts); #else pr_err("The libbabeltrace support is not compiled in. perf should be " "compiled with environment variables LIBBABELTRACE=1 and " - "LIBBABELTRACE_DIR=/path/to/libbabeltrace/\n"); + "LIBBABELTRACE_DIR=/path/to/libbabeltrace/.\n" + "Check also if libbtraceevent devel files are available.\n"); return -1; #endif } diff --git a/tools/perf/builtin-inject.c b/tools/perf/builtin-inject.c index e254f18986f7..3f4e4dd5abf3 100644 --- a/tools/perf/builtin-inject.c +++ b/tools/perf/builtin-inject.c @@ -607,6 +607,7 @@ static int perf_event__repipe_exit(struct perf_tool *tool, return err; } +#ifdef HAVE_LIBTRACEEVENT static int perf_event__repipe_tracing_data(struct perf_session *session, union perf_event *event) { @@ -614,6 +615,7 @@ static int perf_event__repipe_tracing_data(struct perf_session *session, return perf_event__process_tracing_data(session, event); } +#endif static int dso__read_build_id(struct dso *dso) { @@ -807,6 +809,7 @@ static int perf_inject__sched_switch(struct perf_tool *tool, return 0; } +#ifdef HAVE_LIBTRACEEVENT static int perf_inject__sched_stat(struct perf_tool *tool, union perf_event *event __maybe_unused, struct perf_sample *sample, @@ -836,6 +839,7 @@ found: build_id__mark_dso_hit(tool, event_sw, &sample_sw, evsel, machine); return perf_event__repipe(tool, event_sw, &sample_sw, machine); } +#endif static struct guest_vcpu *guest_session__vcpu(struct guest_session *gs, u32 vcpu) { @@ -1961,7 +1965,9 @@ static int __cmd_inject(struct perf_inject *inject) inject->tool.mmap = perf_event__repipe_mmap; inject->tool.mmap2 = perf_event__repipe_mmap2; inject->tool.fork = perf_event__repipe_fork; +#ifdef HAVE_LIBTRACEEVENT inject->tool.tracing_data = perf_event__repipe_tracing_data; +#endif } output_data_offset = perf_session__data_offset(session->evlist); @@ -1984,8 +1990,10 @@ static int __cmd_inject(struct perf_inject *inject) evsel->handler = perf_inject__sched_switch; } else if (!strcmp(name, "sched:sched_process_exit")) evsel->handler = perf_inject__sched_process_exit; +#ifdef HAVE_LIBTRACEEVENT else if (!strncmp(name, "sched:sched_stat_", 17)) evsel->handler = perf_inject__sched_stat; +#endif } } else if (inject->itrace_synth_opts.vm_time_correlation) { session->itrace_synth_opts = &inject->itrace_synth_opts; diff --git a/tools/perf/builtin-kmem.c b/tools/perf/builtin-kmem.c index ebfab2ca1702..e20656c431a4 100644 --- a/tools/perf/builtin-kmem.c +++ b/tools/perf/builtin-kmem.c @@ -35,6 +35,7 @@ #include #include +#include static int kmem_slab; static int kmem_page; diff --git a/tools/perf/builtin-kvm.c b/tools/perf/builtin-kvm.c index 7d9ec1bac1a2..641e739c717c 100644 --- a/tools/perf/builtin-kvm.c +++ b/tools/perf/builtin-kvm.c @@ -63,7 +63,7 @@ static const char *get_filename_for_perf_kvm(void) return filename; } -#ifdef HAVE_KVM_STAT_SUPPORT +#if defined(HAVE_KVM_STAT_SUPPORT) && defined(HAVE_LIBTRACEEVENT) void exit_event_get_key(struct evsel *evsel, struct perf_sample *sample, @@ -654,7 +654,7 @@ static void print_result(struct perf_kvm_stat *kvm) pr_info("\nLost events: %" PRIu64 "\n\n", kvm->lost_events); } -#ifdef HAVE_TIMERFD_SUPPORT +#if defined(HAVE_TIMERFD_SUPPORT) && defined(HAVE_LIBTRACEEVENT) static int process_lost_event(struct perf_tool *tool, union perf_event *event __maybe_unused, struct perf_sample *sample __maybe_unused, @@ -742,7 +742,7 @@ static bool verify_vcpu(int vcpu) return true; } -#ifdef HAVE_TIMERFD_SUPPORT +#if defined(HAVE_TIMERFD_SUPPORT) && defined(HAVE_LIBTRACEEVENT) /* keeping the max events to a modest level to keep * the processing of samples per mmap smooth. */ @@ -1290,7 +1290,7 @@ kvm_events_report(struct perf_kvm_stat *kvm, int argc, const char **argv) return kvm_events_report_vcpu(kvm); } -#ifdef HAVE_TIMERFD_SUPPORT +#if defined(HAVE_TIMERFD_SUPPORT) && defined(HAVE_LIBTRACEEVENT) static struct evlist *kvm_live_event_list(void) { struct evlist *evlist; @@ -1507,7 +1507,7 @@ static int kvm_cmd_stat(const char *file_name, int argc, const char **argv) if (strlen(argv[1]) > 2 && strstarts("report", argv[1])) return kvm_events_report(&kvm, argc - 1 , argv + 1); -#ifdef HAVE_TIMERFD_SUPPORT +#if defined(HAVE_TIMERFD_SUPPORT) && defined(HAVE_LIBTRACEEVENT) if (!strncmp(argv[1], "live", 4)) return kvm_events_live(&kvm, argc - 1 , argv + 1); #endif @@ -1644,7 +1644,7 @@ int cmd_kvm(int argc, const char **argv) return cmd_top(argc, argv); else if (strlen(argv[0]) > 2 && strstarts("buildid-list", argv[0])) return __cmd_buildid_list(file_name, argc, argv); -#ifdef HAVE_KVM_STAT_SUPPORT +#if defined(HAVE_KVM_STAT_SUPPORT) && defined(HAVE_LIBTRACEEVENT) else if (strlen(argv[0]) > 2 && strstarts("stat", argv[0])) return kvm_cmd_stat(file_name, argc, argv); #endif diff --git a/tools/perf/builtin-kwork.c b/tools/perf/builtin-kwork.c index 0e02b8098644..dc59d75180d1 100644 --- a/tools/perf/builtin-kwork.c +++ b/tools/perf/builtin-kwork.c @@ -23,6 +23,7 @@ #include #include +#include #include #include diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c index b7fd7ec586fb..7e17374f6c1a 100644 --- a/tools/perf/builtin-record.c +++ b/tools/perf/builtin-record.c @@ -1701,8 +1701,10 @@ static void record__init_features(struct record *rec) if (rec->no_buildid) perf_header__clear_feat(&session->header, HEADER_BUILD_ID); +#ifdef HAVE_LIBTRACEEVENT if (!have_tracepoints(&rec->evlist->core.entries)) perf_header__clear_feat(&session->header, HEADER_TRACING_DATA); +#endif if (!rec->opts.branch_stack) perf_header__clear_feat(&session->header, HEADER_BRANCH_STACK); diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c index b6d77d3da64f..2ee2ecca208e 100644 --- a/tools/perf/builtin-report.c +++ b/tools/perf/builtin-report.c @@ -67,6 +67,10 @@ #include #include +#ifdef HAVE_LIBTRACEEVENT +#include +#endif + struct report { struct perf_tool tool; struct perf_session *session; @@ -1199,7 +1203,9 @@ int cmd_report(int argc, const char **argv) .lost = perf_event__process_lost, .read = process_read_event, .attr = process_attr, +#ifdef HAVE_LIBTRACEEVENT .tracing_data = perf_event__process_tracing_data, +#endif .build_id = perf_event__process_build_id, .id_index = perf_event__process_id_index, .auxtrace_info = perf_event__process_auxtrace_info, @@ -1660,6 +1666,7 @@ repeat: report.range_num); } +#ifdef HAVE_LIBTRACEEVENT if (session->tevent.pevent && tep_set_function_resolver(session->tevent.pevent, machine__resolve_kernel_addr, @@ -1668,7 +1675,7 @@ repeat: __func__); return -1; } - +#endif sort__setup_elide(stdout); ret = __cmd_report(&report); diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c index d7ec8c1af293..88888fb885c8 100644 --- a/tools/perf/builtin-script.c +++ b/tools/perf/builtin-script.c @@ -62,6 +62,9 @@ #include "perf.h" #include +#ifdef HAVE_LIBTRACEEVENT +#include +#endif static char const *script_name; static char const *generate_script_lang; @@ -2154,12 +2157,12 @@ static void process_event(struct perf_script *script, perf_sample__fprintf_bts(sample, evsel, thread, al, addr_al, machine, fp); return; } - +#ifdef HAVE_LIBTRACEEVENT if (PRINT_FIELD(TRACE) && sample->raw_data) { event_format__fprintf(evsel->tp_format, sample->cpu, sample->raw_data, sample->raw_size, fp); } - +#endif if (attr->type == PERF_TYPE_SYNTH && PRINT_FIELD(SYNTH)) perf_sample__fprintf_synth(sample, evsel, fp); @@ -2283,8 +2286,10 @@ static void process_stat_interval(u64 tstamp) static void setup_scripting(void) { +#ifdef HAVE_LIBTRACEEVENT setup_perl_scripting(); setup_python_scripting(); +#endif } static int flush_scripting(void) @@ -3784,7 +3789,9 @@ int cmd_script(int argc, const char **argv) .fork = perf_event__process_fork, .attr = process_attr, .event_update = perf_event__process_event_update, +#ifdef HAVE_LIBTRACEEVENT .tracing_data = perf_event__process_tracing_data, +#endif .feature = process_feature_event, .build_id = perf_event__process_build_id, .id_index = perf_event__process_id_index, @@ -4215,6 +4222,7 @@ script_found: else symbol_conf.use_callchain = false; +#ifdef HAVE_LIBTRACEEVENT if (session->tevent.pevent && tep_set_function_resolver(session->tevent.pevent, machine__resolve_kernel_addr, @@ -4223,7 +4231,7 @@ script_found: err = -1; goto out_delete; } - +#endif if (generate_script_lang) { struct stat perf_stat; int input; @@ -4259,9 +4267,12 @@ script_found: err = -ENOENT; goto out_delete; } - +#ifdef HAVE_LIBTRACEEVENT err = scripting_ops->generate_script(session->tevent.pevent, "perf-script"); +#else + err = scripting_ops->generate_script(NULL, "perf-script"); +#endif goto out_delete; } diff --git a/tools/perf/builtin-timechart.c b/tools/perf/builtin-timechart.c index c36296bb7637..6c629e7d370a 100644 --- a/tools/perf/builtin-timechart.c +++ b/tools/perf/builtin-timechart.c @@ -38,6 +38,7 @@ #include "util/string2.h" #include "util/tracepoint.h" #include +#include #ifdef LACKS_OPEN_MEMSTREAM_PROTOTYPE FILE *open_memstream(char **ptr, size_t *sizeloc); diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c index 543c379d2a57..6909cd9f48d1 100644 --- a/tools/perf/builtin-trace.c +++ b/tools/perf/builtin-trace.c @@ -15,7 +15,6 @@ */ #include "util/record.h" -#include #include #include #include "util/bpf_map.h" @@ -80,6 +79,10 @@ #include #include +#ifdef HAVE_LIBTRACEEVENT +#include +#endif + #ifndef O_CLOEXEC # define O_CLOEXEC 02000000 #endif diff --git a/tools/perf/builtin-version.c b/tools/perf/builtin-version.c index a71f491224da..a886929ec6e5 100644 --- a/tools/perf/builtin-version.c +++ b/tools/perf/builtin-version.c @@ -82,6 +82,7 @@ static void library_status(void) STATUS(HAVE_AIO_SUPPORT, aio); STATUS(HAVE_ZSTD_SUPPORT, zstd); STATUS(HAVE_LIBPFM, libpfm4); + STATUS(HAVE_LIBTRACEEVENT, libtraceevent); } int cmd_version(int argc, const char **argv) diff --git a/tools/perf/perf.c b/tools/perf/perf.c index 7af135dea1cd..82bbe0ca858b 100644 --- a/tools/perf/perf.c +++ b/tools/perf/perf.c @@ -70,20 +70,26 @@ static struct cmd_struct commands[] = { { "report", cmd_report, 0 }, { "bench", cmd_bench, 0 }, { "stat", cmd_stat, 0 }, +#ifdef HAVE_LIBTRACEEVENT { "timechart", cmd_timechart, 0 }, +#endif { "top", cmd_top, 0 }, { "annotate", cmd_annotate, 0 }, { "version", cmd_version, 0 }, { "script", cmd_script, 0 }, +#ifdef HAVE_LIBTRACEEVENT { "sched", cmd_sched, 0 }, +#endif #ifdef HAVE_LIBELF_SUPPORT { "probe", cmd_probe, 0 }, #endif +#ifdef HAVE_LIBTRACEEVENT { "kmem", cmd_kmem, 0 }, { "lock", cmd_lock, 0 }, +#endif { "kvm", cmd_kvm, 0 }, { "test", cmd_test, 0 }, -#if defined(HAVE_LIBAUDIT_SUPPORT) || defined(HAVE_SYSCALL_TABLE_SUPPORT) +#if defined(HAVE_LIBTRACEEVENT) && (defined(HAVE_LIBAUDIT_SUPPORT) || defined(HAVE_SYSCALL_TABLE_SUPPORT)) { "trace", cmd_trace, 0 }, #endif { "inject", cmd_inject, 0 }, @@ -91,7 +97,9 @@ static struct cmd_struct commands[] = { { "data", cmd_data, 0 }, { "ftrace", cmd_ftrace, 0 }, { "daemon", cmd_daemon, 0 }, +#ifdef HAVE_LIBTRACEEVENT { "kwork", cmd_kwork, 0 }, +#endif }; struct pager_config { @@ -500,14 +508,18 @@ int main(int argc, const char **argv) argv[0] = cmd; } if (strstarts(cmd, "trace")) { -#if defined(HAVE_LIBAUDIT_SUPPORT) || defined(HAVE_SYSCALL_TABLE_SUPPORT) - setup_path(); - argv[0] = "trace"; - return cmd_trace(argc, argv); -#else +#ifndef HAVE_LIBTRACEEVENT + fprintf(stderr, + "trace command not available: missing libtraceevent devel package at build time.\n"); + goto out; +#elif !defined(HAVE_LIBAUDIT_SUPPORT) && !defined(HAVE_SYSCALL_TABLE_SUPPORT) fprintf(stderr, "trace command not available: missing audit-libs devel package at build time.\n"); goto out; +#else + setup_path(); + argv[0] = "trace"; + return cmd_trace(argc, argv); #endif } /* Look for flags.. */ diff --git a/tools/perf/scripts/python/Perf-Trace-Util/Build b/tools/perf/scripts/python/Perf-Trace-Util/Build index 7d0e33ce6aba..d5fed4e42617 100644 --- a/tools/perf/scripts/python/Perf-Trace-Util/Build +++ b/tools/perf/scripts/python/Perf-Trace-Util/Build @@ -1,3 +1,3 @@ -perf-y += Context.o +perf-$(CONFIG_LIBTRACEEVENT) += Context.o CFLAGS_Context.o += $(PYTHON_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-nested-externs diff --git a/tools/perf/tests/Build b/tools/perf/tests/Build index 658b5052c24d..90fd1eb317bb 100644 --- a/tools/perf/tests/Build +++ b/tools/perf/tests/Build @@ -6,13 +6,13 @@ perf-y += parse-events.o perf-y += dso-data.o perf-y += attr.o perf-y += vmlinux-kallsyms.o -perf-y += openat-syscall.o -perf-y += openat-syscall-all-cpus.o -perf-y += openat-syscall-tp-fields.o -perf-y += mmap-basic.o +perf-$(CONFIG_LIBTRACEEVENT) += openat-syscall.o +perf-$(CONFIG_LIBTRACEEVENT) += openat-syscall-all-cpus.o +perf-$(CONFIG_LIBTRACEEVENT) += openat-syscall-tp-fields.o +perf-$(CONFIG_LIBTRACEEVENT) += mmap-basic.o perf-y += perf-record.o perf-y += evsel-roundtrip-name.o -perf-y += evsel-tp-sched.o +perf-$(CONFIG_LIBTRACEEVENT) += evsel-tp-sched.o perf-y += fdarray.o perf-y += pmu.o perf-y += pmu-events.o @@ -30,7 +30,7 @@ perf-y += task-exit.o perf-y += sw-clock.o perf-y += mmap-thread-lookup.o perf-y += thread-maps-share.o -perf-y += switch-tracking.o +perf-$(CONFIG_LIBTRACEEVENT) += switch-tracking.o perf-y += keep-tracking.o perf-y += code-reading.o perf-y += sample-parsing.o diff --git a/tools/perf/tests/builtin-test.c b/tools/perf/tests/builtin-test.c index ddd8262bfa26..f6c16ad8ed50 100644 --- a/tools/perf/tests/builtin-test.c +++ b/tools/perf/tests/builtin-test.c @@ -38,9 +38,11 @@ struct test_suite *__weak arch_tests[] = { static struct test_suite *generic_tests[] = { &suite__vmlinux_matches_kallsyms, +#ifdef HAVE_LIBTRACEEVENT &suite__openat_syscall_event, &suite__openat_syscall_event_on_all_cpus, &suite__basic_mmap, +#endif &suite__mem, &suite__parse_events, &suite__expr, @@ -51,8 +53,10 @@ static struct test_suite *generic_tests[] = { &suite__dso_data_cache, &suite__dso_data_reopen, &suite__perf_evsel__roundtrip_name_test, +#ifdef HAVE_LIBTRACEEVENT &suite__perf_evsel__tp_sched_test, &suite__syscall_openat_tp_fields, +#endif &suite__attr, &suite__hists_link, &suite__python_use, @@ -71,7 +75,9 @@ static struct test_suite *generic_tests[] = { &suite__thread_maps_share, &suite__hists_output, &suite__hists_cumulate, +#ifdef HAVE_LIBTRACEEVENT &suite__switch_tracking, +#endif &suite__fdarray__filter, &suite__fdarray__add, &suite__kmod_path__parse, diff --git a/tools/perf/tests/parse-events.c b/tools/perf/tests/parse-events.c index 3440dd2616b0..71a5cb343311 100644 --- a/tools/perf/tests/parse-events.c +++ b/tools/perf/tests/parse-events.c @@ -20,6 +20,8 @@ #define PERF_TP_SAMPLE_TYPE (PERF_SAMPLE_RAW | PERF_SAMPLE_TIME | \ PERF_SAMPLE_CPU | PERF_SAMPLE_PERIOD) +#ifdef HAVE_LIBTRACEEVENT + #if defined(__s390x__) /* Return true if kvm module is available and loaded. Test this * and return success when trace point kvm_s390_create_vm @@ -76,6 +78,7 @@ static int test__checkevent_tracepoint_multi(struct evlist *evlist) } return TEST_OK; } +#endif /* HAVE_LIBTRACEEVENT */ static int test__checkevent_raw(struct evlist *evlist) { @@ -222,6 +225,7 @@ static int test__checkevent_breakpoint_rw(struct evlist *evlist) return TEST_OK; } +#ifdef HAVE_LIBTRACEEVENT static int test__checkevent_tracepoint_modifier(struct evlist *evlist) { struct evsel *evsel = evlist__first(evlist); @@ -252,6 +256,7 @@ test__checkevent_tracepoint_multi_modifier(struct evlist *evlist) return test__checkevent_tracepoint_multi(evlist); } +#endif /* HAVE_LIBTRACEEVENT */ static int test__checkevent_raw_modifier(struct evlist *evlist) { @@ -453,6 +458,7 @@ static int test__checkevent_pmu(struct evlist *evlist) return TEST_OK; } +#ifdef HAVE_LIBTRACEEVENT static int test__checkevent_list(struct evlist *evlist) { struct evsel *evsel = evlist__first(evlist); @@ -491,6 +497,7 @@ static int test__checkevent_list(struct evlist *evlist) return TEST_OK; } +#endif static int test__checkevent_pmu_name(struct evlist *evlist) { @@ -762,6 +769,7 @@ static int test__group2(struct evlist *evlist) return TEST_OK; } +#ifdef HAVE_LIBTRACEEVENT static int test__group3(struct evlist *evlist __maybe_unused) { struct evsel *evsel, *leader; @@ -853,6 +861,7 @@ static int test__group3(struct evlist *evlist __maybe_unused) return TEST_OK; } +#endif static int test__group4(struct evlist *evlist __maybe_unused) { @@ -1460,6 +1469,7 @@ static int test__sym_event_dc(struct evlist *evlist) return TEST_OK; } +#ifdef HAVE_LIBTRACEEVENT static int count_tracepoints(void) { struct dirent *events_ent; @@ -1513,6 +1523,7 @@ static int test__all_tracepoints(struct evlist *evlist) return test__checkevent_tracepoint_multi(evlist); } +#endif /* HAVE_LIBTRACEVENT */ static int test__hybrid_hw_event_with_pmu(struct evlist *evlist) { @@ -1642,6 +1653,7 @@ struct evlist_test { }; static const struct evlist_test test__events[] = { +#ifdef HAVE_LIBTRACEEVENT { .name = "syscalls:sys_enter_openat", .check = test__checkevent_tracepoint, @@ -1652,6 +1664,7 @@ static const struct evlist_test test__events[] = { .check = test__checkevent_tracepoint_multi, /* 1 */ }, +#endif { .name = "r1a", .check = test__checkevent_raw, @@ -1702,6 +1715,7 @@ static const struct evlist_test test__events[] = { .check = test__checkevent_breakpoint_w, /* 1 */ }, +#ifdef HAVE_LIBTRACEEVENT { .name = "syscalls:sys_enter_openat:k", .check = test__checkevent_tracepoint_modifier, @@ -1712,6 +1726,7 @@ static const struct evlist_test test__events[] = { .check = test__checkevent_tracepoint_multi_modifier, /* 3 */ }, +#endif { .name = "r1a:kp", .check = test__checkevent_raw_modifier, @@ -1757,11 +1772,13 @@ static const struct evlist_test test__events[] = { .check = test__checkevent_breakpoint_w_modifier, /* 2 */ }, +#ifdef HAVE_LIBTRACEEVENT { .name = "r1,syscalls:sys_enter_openat:k,1:1:hp", .check = test__checkevent_list, /* 3 */ }, +#endif { .name = "instructions:G", .check = test__checkevent_exclude_host_modifier, @@ -1792,11 +1809,13 @@ static const struct evlist_test test__events[] = { .check = test__group2, /* 9 */ }, +#ifdef HAVE_LIBTRACEEVENT { .name = "group1{syscalls:sys_enter_openat:H,cycles:kppp},group2{cycles,1:3}:G,instructions:u", .check = test__group3, /* 0 */ }, +#endif { .name = "{cycles:u,instructions:kp}:p", .check = test__group4, @@ -1807,11 +1826,13 @@ static const struct evlist_test test__events[] = { .check = test__group5, /* 2 */ }, +#ifdef HAVE_LIBTRACEEVENT { .name = "*:*", .check = test__all_tracepoints, /* 3 */ }, +#endif { .name = "{cycles,cache-misses:G}:H", .check = test__group_gh1, @@ -1867,7 +1888,7 @@ static const struct evlist_test test__events[] = { .check = test__checkevent_breakpoint_len_rw_modifier, /* 4 */ }, -#if defined(__s390x__) +#if defined(__s390x__) && defined(HAVE_LIBTRACEEVENT) { .name = "kvm-s390:kvm_s390_create_vm", .check = test__checkevent_tracepoint, diff --git a/tools/perf/util/Build b/tools/perf/util/Build index d04802bfa23f..8345464d0130 100644 --- a/tools/perf/util/Build +++ b/tools/perf/util/Build @@ -69,7 +69,6 @@ perf-y += namespaces.o perf-y += comm.o perf-y += thread.o perf-y += thread_map.o -perf-y += trace-event-parse.o perf-y += parse-events-flex.o perf-y += parse-events-bison.o perf-y += pmu.o @@ -77,11 +76,12 @@ perf-y += pmus.o perf-y += pmu-flex.o perf-y += pmu-bison.o perf-y += pmu-hybrid.o -perf-y += trace-event-read.o -perf-y += trace-event-info.o -perf-y += trace-event-scripting.o -perf-y += trace-event.o perf-y += svghelper.o +perf-$(CONFIG_LIBTRACEEVENT) += trace-event-info.o +perf-$(CONFIG_LIBTRACEEVENT) += trace-event-scripting.o +perf-$(CONFIG_LIBTRACEEVENT) += trace-event.o +perf-$(CONFIG_LIBTRACEEVENT) += trace-event-parse.o +perf-$(CONFIG_LIBTRACEEVENT) += trace-event-read.o perf-y += sort.o perf-y += hist.o perf-y += util.o @@ -153,8 +153,12 @@ perf-$(CONFIG_PERF_BPF_SKEL) += bpf_counter.o perf-$(CONFIG_PERF_BPF_SKEL) += bpf_counter_cgroup.o perf-$(CONFIG_PERF_BPF_SKEL) += bpf_ftrace.o perf-$(CONFIG_PERF_BPF_SKEL) += bpf_off_cpu.o -perf-$(CONFIG_PERF_BPF_SKEL) += bpf_kwork.o perf-$(CONFIG_PERF_BPF_SKEL) += bpf_lock_contention.o + +ifeq ($(CONFIG_LIBTRACEEVENT),y) + perf-$(CONFIG_PERF_BPF_SKEL) += bpf_kwork.o +endif + perf-$(CONFIG_BPF_PROLOGUE) += bpf-prologue.o perf-$(CONFIG_LIBELF) += symbol-elf.o perf-$(CONFIG_LIBELF) += probe-file.o @@ -189,7 +193,10 @@ perf-$(CONFIG_LIBUNWIND) += unwind-libunwind.o perf-$(CONFIG_LIBUNWIND_X86) += libunwind/x86_32.o perf-$(CONFIG_LIBUNWIND_AARCH64) += libunwind/arm64.o -perf-$(CONFIG_LIBBABELTRACE) += data-convert-bt.o +ifeq ($(CONFIG_LIBTRACEEVENT),y) + perf-$(CONFIG_LIBBABELTRACE) += data-convert-bt.o +endif + perf-y += data-convert-json.o perf-y += scripting-engines/ diff --git a/tools/perf/util/data-convert-bt.c b/tools/perf/util/data-convert-bt.c index c65cdaf6975e..8031b586e813 100644 --- a/tools/perf/util/data-convert-bt.c +++ b/tools/perf/util/data-convert-bt.c @@ -19,7 +19,6 @@ #include #include #include -#include #include "asm/bug.h" #include "data-convert.h" #include "session.h" @@ -36,6 +35,10 @@ #include "clockid.h" #include "util/sample.h" +#ifdef HAVE_LIBTRACEEVENT +#include +#endif + #define pr_N(n, fmt, ...) \ eprintf(n, debug_data_convert, fmt, ##__VA_ARGS__) diff --git a/tools/perf/util/data-convert-json.c b/tools/perf/util/data-convert-json.c index 57db59068cb6..ba9d93ce9463 100644 --- a/tools/perf/util/data-convert-json.c +++ b/tools/perf/util/data-convert-json.c @@ -27,6 +27,10 @@ #include "util/thread.h" #include "util/tool.h" +#ifdef HAVE_LIBTRACEEVENT +#include +#endif + struct convert_json { struct perf_tool tool; FILE *out; @@ -217,6 +221,7 @@ static int process_sample_event(struct perf_tool *tool, } output_json_format(out, false, 3, "]"); +#ifdef HAVE_LIBTRACEEVENT if (sample->raw_data) { int i; struct tep_format_field **fields; @@ -236,7 +241,7 @@ static int process_sample_event(struct perf_tool *tool, free(fields); } } - +#endif output_json_format(out, false, 2, "}"); return 0; } @@ -313,7 +318,9 @@ int bt_convert__perf2json(const char *input_name, const char *output_name, .exit = perf_event__process_exit, .fork = perf_event__process_fork, .lost = perf_event__process_lost, +#ifdef HAVE_LIBTRACEEVENT .tracing_data = perf_event__process_tracing_data, +#endif .build_id = perf_event__process_build_id, .id_index = perf_event__process_id_index, .auxtrace_info = perf_event__process_auxtrace_info, diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c index fbf3192bced9..590d4e77effc 100644 --- a/tools/perf/util/evlist.c +++ b/tools/perf/util/evlist.c @@ -290,6 +290,7 @@ struct evsel *evlist__add_aux_dummy(struct evlist *evlist, bool system_wide) return evsel; } +#ifdef HAVE_LIBTRACEEVENT struct evsel *evlist__add_sched_switch(struct evlist *evlist, bool system_wide) { struct evsel *evsel = evsel__newtp_idx("sched", "sched_switch", 0); @@ -305,7 +306,8 @@ struct evsel *evlist__add_sched_switch(struct evlist *evlist, bool system_wide) evlist__add(evlist, evsel); return evsel; -}; +} +#endif int evlist__add_attrs(struct evlist *evlist, struct perf_event_attr *attrs, size_t nr_attrs) { @@ -376,6 +378,7 @@ struct evsel *evlist__find_tracepoint_by_name(struct evlist *evlist, const char return NULL; } +#ifdef HAVE_LIBTRACEEVENT int evlist__add_newtp(struct evlist *evlist, const char *sys, const char *name, void *handler) { struct evsel *evsel = evsel__newtp(sys, name); @@ -387,6 +390,7 @@ int evlist__add_newtp(struct evlist *evlist, const char *sys, const char *name, evlist__add(evlist, evsel); return 0; } +#endif struct evlist_cpu_iterator evlist__cpu_begin(struct evlist *evlist, struct affinity *affinity) { diff --git a/tools/perf/util/evlist.h b/tools/perf/util/evlist.h index 16734c6756b3..e5b84ead566c 100644 --- a/tools/perf/util/evlist.h +++ b/tools/perf/util/evlist.h @@ -127,7 +127,9 @@ static inline struct evsel *evlist__add_dummy_on_all_cpus(struct evlist *evlist) { return evlist__add_aux_dummy(evlist, true); } +#ifdef HAVE_LIBTRACEEVENT struct evsel *evlist__add_sched_switch(struct evlist *evlist, bool system_wide); +#endif int evlist__add_sb_event(struct evlist *evlist, struct perf_event_attr *attr, evsel__sb_cb_t cb, void *data); @@ -135,7 +137,9 @@ void evlist__set_cb(struct evlist *evlist, evsel__sb_cb_t cb, void *data); int evlist__start_sb_thread(struct evlist *evlist, struct target *target); void evlist__stop_sb_thread(struct evlist *evlist); +#ifdef HAVE_LIBTRACEEVENT int evlist__add_newtp(struct evlist *evlist, const char *sys, const char *name, void *handler); +#endif int __evlist__set_tracepoints_handlers(struct evlist *evlist, const struct evsel_str_handler *assocs, diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c index 0f617359a82f..ca911856c4b1 100644 --- a/tools/perf/util/evsel.c +++ b/tools/perf/util/evsel.c @@ -12,7 +12,6 @@ #include #include #include -#include #include #include #include @@ -57,6 +56,10 @@ #include +#ifdef HAVE_LIBTRACEEVENT +#include +#endif + struct perf_missing_features perf_missing_features; static clockid_t clockid; @@ -439,7 +442,9 @@ struct evsel *evsel__clone(struct evsel *orig) goto out_err; } evsel->cgrp = cgroup__get(orig->cgrp); +#ifdef HAVE_LIBTRACEEVENT evsel->tp_format = orig->tp_format; +#endif evsel->handler = orig->handler; evsel->core.leader = orig->core.leader; @@ -479,6 +484,7 @@ out_err: /* * Returns pointer with encoded error via interface. */ +#ifdef HAVE_LIBTRACEEVENT struct evsel *evsel__newtp_idx(const char *sys, const char *name, int idx) { struct evsel *evsel = zalloc(perf_evsel__object.size); @@ -516,6 +522,7 @@ out_free: out_err: return ERR_PTR(err); } +#endif const char *const evsel__hw_names[PERF_COUNT_HW_MAX] = { "cycles", @@ -2758,6 +2765,7 @@ u16 evsel__id_hdr_size(struct evsel *evsel) return size; } +#ifdef HAVE_LIBTRACEEVENT struct tep_format_field *evsel__field(struct evsel *evsel, const char *name) { return tep_find_field(evsel->tp_format, name); @@ -2831,6 +2839,7 @@ u64 evsel__intval(struct evsel *evsel, struct perf_sample *sample, const char *n return field ? format_field__intval(field, sample, evsel->needs_swap) : 0; } +#endif bool evsel__fallback(struct evsel *evsel, int err, char *msg, size_t msgsize) { diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h index f3485799ddf9..d572be41b960 100644 --- a/tools/perf/util/evsel.h +++ b/tools/perf/util/evsel.h @@ -72,7 +72,9 @@ struct evsel { char *name; char *group_name; const char *pmu_name; +#ifdef HAVE_LIBTRACEEVENT struct tep_event *tp_format; +#endif char *filter; unsigned long max_events; double scale; @@ -223,11 +225,14 @@ static inline struct evsel *evsel__new(struct perf_event_attr *attr) } struct evsel *evsel__clone(struct evsel *orig); -struct evsel *evsel__newtp_idx(const char *sys, const char *name, int idx); int copy_config_terms(struct list_head *dst, struct list_head *src); void free_config_terms(struct list_head *config_terms); + +#ifdef HAVE_LIBTRACEEVENT +struct evsel *evsel__newtp_idx(const char *sys, const char *name, int idx); + /* * Returns pointer with encoded error via interface. */ @@ -235,10 +240,13 @@ static inline struct evsel *evsel__newtp(const char *sys, const char *name) { return evsel__newtp_idx(sys, name, 0); } +#endif struct evsel *evsel__new_cycles(bool precise, __u32 type, __u64 config); +#ifdef HAVE_LIBTRACEEVENT struct tep_event *event_format__new(const char *sys, const char *name); +#endif void evsel__init(struct evsel *evsel, struct perf_event_attr *attr, int idx); void evsel__exit(struct evsel *evsel); @@ -323,6 +331,7 @@ bool evsel__precise_ip_fallback(struct evsel *evsel); struct perf_sample; +#ifdef HAVE_LIBTRACEEVENT void *evsel__rawptr(struct evsel *evsel, struct perf_sample *sample, const char *name); u64 evsel__intval(struct evsel *evsel, struct perf_sample *sample, const char *name); @@ -330,6 +339,7 @@ static inline char *evsel__strval(struct evsel *evsel, struct perf_sample *sampl { return evsel__rawptr(evsel, sample, name); } +#endif struct tep_format_field; diff --git a/tools/perf/util/evsel_fprintf.c b/tools/perf/util/evsel_fprintf.c index 8c2ea8001329..bd22c4932d10 100644 --- a/tools/perf/util/evsel_fprintf.c +++ b/tools/perf/util/evsel_fprintf.c @@ -2,7 +2,6 @@ #include #include #include -#include #include "evsel.h" #include "util/evsel_fprintf.h" #include "util/event.h" @@ -13,6 +12,10 @@ #include "srcline.h" #include "dso.h" +#ifdef HAVE_LIBTRACEEVENT +#include +#endif + static int comma_fprintf(FILE *fp, bool *first, const char *fmt, ...) { va_list args; @@ -74,6 +77,7 @@ int evsel__fprintf(struct evsel *evsel, struct perf_attr_details *details, FILE term, (u64)evsel->core.attr.sample_freq); } +#ifdef HAVE_LIBTRACEEVENT if (details->trace_fields) { struct tep_format_field *field; @@ -96,6 +100,7 @@ int evsel__fprintf(struct evsel *evsel, struct perf_attr_details *details, FILE field = field->next; } } +#endif out: fputc('\n', fp); return ++printed; diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c index dc2ae397d400..404d816ca124 100644 --- a/tools/perf/util/header.c +++ b/tools/perf/util/header.c @@ -6,6 +6,7 @@ #include #include #include +#include #include #include #include @@ -55,6 +56,10 @@ #include #include +#ifdef HAVE_LIBTRACEEVENT +#include +#endif + /* * magic2 = "PERFILE2" * must be a numerical value to let the endianness @@ -298,6 +303,7 @@ static int do_read_bitmap(struct feat_fd *ff, unsigned long **pset, u64 *psize) return 0; } +#ifdef HAVE_LIBTRACEEVENT static int write_tracing_data(struct feat_fd *ff, struct evlist *evlist) { @@ -306,6 +312,7 @@ static int write_tracing_data(struct feat_fd *ff, return read_tracing_data(ff->fd, &evlist->core.entries); } +#endif static int write_build_id(struct feat_fd *ff, struct evlist *evlist __maybe_unused) @@ -2394,12 +2401,14 @@ FEAT_PROCESS_STR_FUN(arch, arch); FEAT_PROCESS_STR_FUN(cpudesc, cpu_desc); FEAT_PROCESS_STR_FUN(cpuid, cpuid); +#ifdef HAVE_LIBTRACEEVENT static int process_tracing_data(struct feat_fd *ff, void *data) { ssize_t ret = trace_report(ff->fd, data, false); return ret < 0 ? -1 : 0; } +#endif static int process_build_id(struct feat_fd *ff, void *data __maybe_unused) { @@ -3366,7 +3375,9 @@ err: const struct perf_header_feature_ops feat_ops[HEADER_LAST_FEATURE]; const struct perf_header_feature_ops feat_ops[HEADER_LAST_FEATURE] = { +#ifdef HAVE_LIBTRACEEVENT FEAT_OPN(TRACING_DATA, tracing_data, false), +#endif FEAT_OPN(BUILD_ID, build_id, false), FEAT_OPR(HOSTNAME, hostname, false), FEAT_OPR(OSRELEASE, osrelease, false), @@ -4082,6 +4093,7 @@ static int read_attr(int fd, struct perf_header *ph, return ret <= 0 ? -1 : 0; } +#ifdef HAVE_LIBTRACEEVENT static int evsel__prepare_tracepoint_event(struct evsel *evsel, struct tep_handle *pevent) { struct tep_event *event; @@ -4125,6 +4137,7 @@ static int evlist__prepare_tracepoint_events(struct evlist *evlist, struct tep_h return 0; } +#endif int perf_session__read_header(struct perf_session *session, int repipe_fd) { @@ -4230,11 +4243,15 @@ int perf_session__read_header(struct perf_session *session, int repipe_fd) lseek(fd, tmp, SEEK_SET); } +#ifdef HAVE_LIBTRACEEVENT perf_header__process_sections(header, fd, &session->tevent, perf_file_section__process); if (evlist__prepare_tracepoint_events(session->evlist, session->tevent.pevent)) goto out_delete_evlist; +#else + perf_header__process_sections(header, fd, NULL, perf_file_section__process); +#endif return 0; out_errno: @@ -4412,6 +4429,7 @@ int perf_event__process_event_update(struct perf_tool *tool __maybe_unused, return 0; } +#ifdef HAVE_LIBTRACEEVENT int perf_event__process_tracing_data(struct perf_session *session, union perf_event *event) { @@ -4459,6 +4477,7 @@ int perf_event__process_tracing_data(struct perf_session *session, return size_read + padding; } +#endif int perf_event__process_build_id(struct perf_session *session, union perf_event *event) diff --git a/tools/perf/util/header.h b/tools/perf/util/header.h index 2d5e601ba60f..e3861ae62172 100644 --- a/tools/perf/util/header.h +++ b/tools/perf/util/header.h @@ -160,8 +160,10 @@ int perf_event__process_event_update(struct perf_tool *tool, union perf_event *event, struct evlist **pevlist); size_t perf_event__fprintf_event_update(union perf_event *event, FILE *fp); +#ifdef HAVE_LIBTRACEEVENT int perf_event__process_tracing_data(struct perf_session *session, union perf_event *event); +#endif int perf_event__process_build_id(struct perf_session *session, union perf_event *event); bool is_perf_magic(u64 magic); diff --git a/tools/perf/util/intel-pt.c b/tools/perf/util/intel-pt.c index e3548ddef254..6d3921627e33 100644 --- a/tools/perf/util/intel-pt.c +++ b/tools/perf/util/intel-pt.c @@ -3142,6 +3142,7 @@ static int intel_pt_sync_switch(struct intel_pt *pt, int cpu, pid_t tid, return 1; } +#ifdef HAVE_LIBTRACEEVENT static int intel_pt_process_switch(struct intel_pt *pt, struct perf_sample *sample) { @@ -3165,6 +3166,7 @@ static int intel_pt_process_switch(struct intel_pt *pt, return machine__set_current_tid(pt->machine, cpu, -1, tid); } +#endif /* HAVE_LIBTRACEEVENT */ static int intel_pt_context_switch_in(struct intel_pt *pt, struct perf_sample *sample) @@ -3433,9 +3435,12 @@ static int intel_pt_process_event(struct perf_session *session, return err; } +#ifdef HAVE_LIBTRACEEVENT if (pt->switch_evsel && event->header.type == PERF_RECORD_SAMPLE) err = intel_pt_process_switch(pt, sample); - else if (event->header.type == PERF_RECORD_ITRACE_START) + else +#endif + if (event->header.type == PERF_RECORD_ITRACE_START) err = intel_pt_process_itrace_start(pt, event, sample); else if (event->header.type == PERF_RECORD_AUX_OUTPUT_HW_ID) err = intel_pt_process_aux_output_hw_id(pt, event, sample); diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c index 6502cd679f57..21cce83462b3 100644 --- a/tools/perf/util/parse-events.c +++ b/tools/perf/util/parse-events.c @@ -445,6 +445,7 @@ out_free_terms: return ret; } +#ifdef HAVE_LIBTRACEEVENT static void tracepoint_error(struct parse_events_error *e, int err, const char *sys, const char *name) { @@ -593,6 +594,7 @@ static int add_tracepoint_multi_sys(struct list_head *list, int *idx, closedir(events_dir); return ret; } +#endif /* HAVE_LIBTRACEEVENT */ #ifdef HAVE_LIBBPF_SUPPORT struct __add_bpf_event_param { @@ -1143,6 +1145,7 @@ static int config_term_pmu(struct perf_event_attr *attr, return config_term_common(attr, term, err); } +#ifdef HAVE_LIBTRACEEVENT static int config_term_tracepoint(struct perf_event_attr *attr, struct parse_events_term *term, struct parse_events_error *err) @@ -1170,6 +1173,7 @@ static int config_term_tracepoint(struct perf_event_attr *attr, return 0; } +#endif static int config_attr(struct perf_event_attr *attr, struct list_head *head, @@ -1325,6 +1329,7 @@ int parse_events_add_tracepoint(struct list_head *list, int *idx, struct parse_events_error *err, struct list_head *head_config) { +#ifdef HAVE_LIBTRACEEVENT if (head_config) { struct perf_event_attr attr; @@ -1339,6 +1344,16 @@ int parse_events_add_tracepoint(struct list_head *list, int *idx, else return add_tracepoint_event(list, idx, sys, event, err, head_config); +#else + (void)list; + (void)idx; + (void)sys; + (void)event; + (void)head_config; + parse_events_error__handle(err, 0, strdup("unsupported tracepoint"), + strdup("libtraceevent is necessary for tracepoint support")); + return -1; +#endif } int parse_events_add_numeric(struct parse_events_state *parse_state, diff --git a/tools/perf/util/parse-events.h b/tools/perf/util/parse-events.h index 07df7bb7b042..428e72eaafcc 100644 --- a/tools/perf/util/parse-events.h +++ b/tools/perf/util/parse-events.h @@ -18,7 +18,6 @@ struct parse_events_error; struct option; struct perf_pmu; -bool have_tracepoints(struct list_head *evlist); bool is_event_supported(u8 type, u64 config); const char *event_type(int type); diff --git a/tools/perf/util/python.c b/tools/perf/util/python.c index b5941c74a0d6..6fb84b7455b8 100644 --- a/tools/perf/util/python.c +++ b/tools/perf/util/python.c @@ -5,7 +5,9 @@ #include #include #include +#ifdef HAVE_LIBTRACEEVENT #include +#endif #include #include "evlist.h" #include "callchain.h" @@ -417,6 +419,7 @@ static PyObject *pyrf_sample_event__repr(struct pyrf_event *pevent) return ret; } +#ifdef HAVE_LIBTRACEEVENT static bool is_tracepoint(struct pyrf_event *pevent) { return pevent->evsel->core.attr.type == PERF_TYPE_TRACEPOINT; @@ -486,14 +489,17 @@ get_tracepoint_field(struct pyrf_event *pevent, PyObject *attr_name) return tracepoint_field(pevent, field); } +#endif /* HAVE_LIBTRACEEVENT */ static PyObject* pyrf_sample_event__getattro(struct pyrf_event *pevent, PyObject *attr_name) { PyObject *obj = NULL; +#ifdef HAVE_LIBTRACEEVENT if (is_tracepoint(pevent)) obj = get_tracepoint_field(pevent, attr_name); +#endif return obj ?: PyObject_GenericGetAttr((PyObject *) pevent, attr_name); } @@ -1326,6 +1332,9 @@ static struct { static PyObject *pyrf__tracepoint(struct pyrf_evsel *pevsel, PyObject *args, PyObject *kwargs) { +#ifndef HAVE_LIBTRACEEVENT + return NULL; +#else struct tep_event *tp_format; static char *kwlist[] = { "sys", "name", NULL }; char *sys = NULL; @@ -1340,6 +1349,7 @@ static PyObject *pyrf__tracepoint(struct pyrf_evsel *pevsel, return _PyLong_FromLong(-1); return _PyLong_FromLong(tp_format->id); +#endif // HAVE_LIBTRACEEVENT } static PyMethodDef perf__methods[] = { diff --git a/tools/perf/util/scripting-engines/Build b/tools/perf/util/scripting-engines/Build index 0f5ba28339cf..d47820c0b4d4 100644 --- a/tools/perf/util/scripting-engines/Build +++ b/tools/perf/util/scripting-engines/Build @@ -1,5 +1,7 @@ -perf-$(CONFIG_LIBPERL) += trace-event-perl.o -perf-$(CONFIG_LIBPYTHON) += trace-event-python.o +ifeq ($(CONFIG_LIBTRACEEVENT),y) + perf-$(CONFIG_LIBPERL) += trace-event-perl.o + perf-$(CONFIG_LIBPYTHON) += trace-event-python.o +endif CFLAGS_trace-event-perl.o += $(PERL_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-shadow -Wno-nested-externs -Wno-undef -Wno-switch-default -Wno-bad-function-cast -Wno-declaration-after-statement -Wno-switch-enum diff --git a/tools/perf/util/scripting-engines/trace-event-perl.c b/tools/perf/util/scripting-engines/trace-event-perl.c index 5b602b6d4685..0bacb49408f8 100644 --- a/tools/perf/util/scripting-engines/trace-event-perl.c +++ b/tools/perf/util/scripting-engines/trace-event-perl.c @@ -27,6 +27,7 @@ #include #include #include +#include #include /* perl needs the following define, right after including stdbool.h */ diff --git a/tools/perf/util/scripting-engines/trace-event-python.c b/tools/perf/util/scripting-engines/trace-event-python.c index d685a7399ee2..fabba21919b8 100644 --- a/tools/perf/util/scripting-engines/trace-event-python.c +++ b/tools/perf/util/scripting-engines/trace-event-python.c @@ -30,6 +30,7 @@ #include #include #include +#include #include "../build-id.h" #include "../counts.h" diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c index 1facd4616317..7c021c6cedb9 100644 --- a/tools/perf/util/session.c +++ b/tools/perf/util/session.c @@ -314,7 +314,9 @@ void perf_session__delete(struct perf_session *session) evlist__delete(session->evlist); perf_data__close(session->data); } +#ifdef HAVE_LIBTRACEEVENT trace_event__cleanup(&session->tevent); +#endif free(session); } diff --git a/tools/perf/util/session.h b/tools/perf/util/session.h index be5871ea558f..ee3715e8563b 100644 --- a/tools/perf/util/session.h +++ b/tools/perf/util/session.h @@ -33,7 +33,9 @@ struct perf_session { struct auxtrace *auxtrace; struct itrace_synth_opts *itrace_synth_opts; struct list_head auxtrace_index; +#ifdef HAVE_LIBTRACEEVENT struct trace_event tevent; +#endif struct perf_record_time_conv time_conv; bool repipe; bool one_mmap; diff --git a/tools/perf/util/setup.py b/tools/perf/util/setup.py index 43e7ca40b2ec..e80ffbbfacfb 100644 --- a/tools/perf/util/setup.py +++ b/tools/perf/util/setup.py @@ -63,12 +63,18 @@ libperf = getenv('LIBPERF') ext_sources = [f.strip() for f in open('util/python-ext-sources') if len(f.strip()) > 0 and f[0] != '#'] +extra_libraries = [] + +if '-DHAVE_LIBTRACEEVENT' in cflags: + extra_libraries += [ 'traceevent' ] +else: + ext_sources.remove('util/trace-event.c') + # use full paths with source files ext_sources = list(map(lambda x: '%s/%s' % (src_perf, x) , ext_sources)) -extra_libraries = [] if '-DHAVE_LIBNUMA_SUPPORT' in cflags: - extra_libraries = [ 'numa' ] + extra_libraries += [ 'numa' ] if '-DHAVE_LIBCAP_SUPPORT' in cflags: extra_libraries += [ 'cap' ] diff --git a/tools/perf/util/sort.c b/tools/perf/util/sort.c index 2e7330867e2e..c7a97b33e134 100644 --- a/tools/perf/util/sort.c +++ b/tools/perf/util/sort.c @@ -22,7 +22,6 @@ #include "srcline.h" #include "strlist.h" #include "strbuf.h" -#include #include "mem-events.h" #include "annotate.h" #include "event.h" @@ -32,6 +31,10 @@ #include #include +#ifdef HAVE_LIBTRACEEVENT +#include +#endif + regex_t parent_regex; const char default_parent_pattern[] = "^sys_|^do_page_fault"; const char *parent_pattern = default_parent_pattern; @@ -743,6 +746,7 @@ struct sort_entry sort_time = { /* --sort trace */ +#ifdef HAVE_LIBTRACEEVENT static char *get_trace_output(struct hist_entry *he) { struct trace_seq seq; @@ -806,6 +810,7 @@ struct sort_entry sort_trace = { .se_snprintf = hist_entry__trace_snprintf, .se_width_idx = HISTC_TRACE, }; +#endif /* HAVE_LIBTRACEEVENT */ /* sort keys for branch stacks */ @@ -2022,7 +2027,9 @@ static struct sort_dimension common_sort_dimensions[] = { DIM(SORT_LOCAL_WEIGHT, "local_weight", sort_local_weight), DIM(SORT_GLOBAL_WEIGHT, "weight", sort_global_weight), DIM(SORT_TRANSACTION, "transaction", sort_transaction), +#ifdef HAVE_LIBTRACEEVENT DIM(SORT_TRACE, "trace", sort_trace), +#endif DIM(SORT_SYM_SIZE, "symbol_size", sort_sym_size), DIM(SORT_DSO_SIZE, "dso_size", sort_dso_size), DIM(SORT_CGROUP, "cgroup", sort_cgroup), @@ -2206,7 +2213,14 @@ bool perf_hpp__is_ ## key ## _entry(struct perf_hpp_fmt *fmt) \ return hse->se == &sort_ ## key ; \ } +#ifdef HAVE_LIBTRACEEVENT MK_SORT_ENTRY_CHK(trace) +#else +bool perf_hpp__is_trace_entry(struct perf_hpp_fmt *fmt __maybe_unused) +{ + return false; +} +#endif MK_SORT_ENTRY_CHK(srcline) MK_SORT_ENTRY_CHK(srcfile) MK_SORT_ENTRY_CHK(thread) @@ -2347,6 +2361,17 @@ static int __sort_dimension__add_hpp_output(struct sort_dimension *sd, return 0; } +#ifndef HAVE_LIBTRACEEVENT +bool perf_hpp__is_dynamic_entry(struct perf_hpp_fmt *fmt __maybe_unused) +{ + return false; +} +bool perf_hpp__defined_dynamic_entry(struct perf_hpp_fmt *fmt __maybe_unused, + struct hists *hists __maybe_unused) +{ + return false; +} +#else struct hpp_dynamic_entry { struct perf_hpp_fmt hpp; struct evsel *evsel; @@ -2621,6 +2646,7 @@ __alloc_dynamic_entry(struct evsel *evsel, struct tep_format_field *field, return hde; } +#endif /* HAVE_LIBTRACEEVENT */ struct perf_hpp_fmt *perf_hpp_fmt__dup(struct perf_hpp_fmt *fmt) { @@ -2633,6 +2659,7 @@ struct perf_hpp_fmt *perf_hpp_fmt__dup(struct perf_hpp_fmt *fmt) new_hse = memdup(hse, sizeof(*hse)); if (new_hse) new_fmt = &new_hse->hpp; +#ifdef HAVE_LIBTRACEEVENT } else if (perf_hpp__is_dynamic_entry(fmt)) { struct hpp_dynamic_entry *hde, *new_hde; @@ -2640,6 +2667,7 @@ struct perf_hpp_fmt *perf_hpp_fmt__dup(struct perf_hpp_fmt *fmt) new_hde = memdup(hde, sizeof(*hde)); if (new_hde) new_fmt = &new_hde->hpp; +#endif } else { new_fmt = memdup(fmt, sizeof(*fmt)); } @@ -2719,6 +2747,7 @@ static struct evsel *find_evsel(struct evlist *evlist, char *event_name) return evsel; } +#ifdef HAVE_LIBTRACEEVENT static int __dynamic_dimension__add(struct evsel *evsel, struct tep_format_field *field, bool raw_trace, int level) @@ -2789,13 +2818,13 @@ static int add_all_matching_fields(struct evlist *evlist, } return ret; } +#endif /* HAVE_LIBTRACEEVENT */ static int add_dynamic_entry(struct evlist *evlist, const char *tok, int level) { char *str, *event_name, *field_name, *opt_name; struct evsel *evsel; - struct tep_format_field *field; bool raw_trace = symbol_conf.raw_trace; int ret = 0; @@ -2820,6 +2849,7 @@ static int add_dynamic_entry(struct evlist *evlist, const char *tok, raw_trace = true; } +#ifdef HAVE_LIBTRACEEVENT if (!strcmp(field_name, "trace_fields")) { ret = add_all_dynamic_fields(evlist, raw_trace, level); goto out; @@ -2829,6 +2859,7 @@ static int add_dynamic_entry(struct evlist *evlist, const char *tok, ret = add_all_matching_fields(evlist, field_name, raw_trace, level); goto out; } +#endif evsel = find_evsel(evlist, event_name); if (evsel == NULL) { @@ -2843,10 +2874,12 @@ static int add_dynamic_entry(struct evlist *evlist, const char *tok, goto out; } +#ifdef HAVE_LIBTRACEEVENT if (!strcmp(field_name, "*")) { ret = add_evsel_fields(evsel, raw_trace, level); } else { - field = tep_find_any_field(evsel->tp_format, field_name); + struct tep_format_field *field = tep_find_any_field(evsel->tp_format, field_name); + if (field == NULL) { pr_debug("Cannot find event field for %s.%s\n", event_name, field_name); @@ -2855,6 +2888,10 @@ static int add_dynamic_entry(struct evlist *evlist, const char *tok, ret = __dynamic_dimension__add(evsel, field, raw_trace, level); } +#else + (void)level; + (void)raw_trace; +#endif /* HAVE_LIBTRACEEVENT */ out: free(str); @@ -2955,11 +2992,11 @@ int sort_dimension__add(struct perf_hpp_list *list, const char *tok, for (i = 0; i < ARRAY_SIZE(common_sort_dimensions); i++) { struct sort_dimension *sd = &common_sort_dimensions[i]; - if (strncasecmp(tok, sd->name, strlen(tok))) + if (!sd->name || strncasecmp(tok, sd->name, strlen(tok))) continue; for (j = 0; j < ARRAY_SIZE(dynamic_headers); j++) { - if (!strcmp(dynamic_headers[j], sd->name)) + if (sd->name && !strcmp(dynamic_headers[j], sd->name)) sort_dimension_add_dynamic_header(sd); } @@ -3009,7 +3046,7 @@ int sort_dimension__add(struct perf_hpp_list *list, const char *tok, for (i = 0; i < ARRAY_SIZE(bstack_sort_dimensions); i++) { struct sort_dimension *sd = &bstack_sort_dimensions[i]; - if (strncasecmp(tok, sd->name, strlen(tok))) + if (!sd->name || strncasecmp(tok, sd->name, strlen(tok))) continue; if (sort__mode != SORT_MODE__BRANCH) @@ -3025,7 +3062,7 @@ int sort_dimension__add(struct perf_hpp_list *list, const char *tok, for (i = 0; i < ARRAY_SIZE(memory_sort_dimensions); i++) { struct sort_dimension *sd = &memory_sort_dimensions[i]; - if (strncasecmp(tok, sd->name, strlen(tok))) + if (!sd->name || strncasecmp(tok, sd->name, strlen(tok))) continue; if (sort__mode != SORT_MODE__MEMORY) @@ -3339,7 +3376,7 @@ int output_field_add(struct perf_hpp_list *list, char *tok) for (i = 0; i < ARRAY_SIZE(common_sort_dimensions); i++) { struct sort_dimension *sd = &common_sort_dimensions[i]; - if (strncasecmp(tok, sd->name, strlen(tok))) + if (!sd->name || strncasecmp(tok, sd->name, strlen(tok))) continue; return __sort_dimension__add_output(list, sd); @@ -3357,7 +3394,7 @@ int output_field_add(struct perf_hpp_list *list, char *tok) for (i = 0; i < ARRAY_SIZE(bstack_sort_dimensions); i++) { struct sort_dimension *sd = &bstack_sort_dimensions[i]; - if (strncasecmp(tok, sd->name, strlen(tok))) + if (!sd->name || strncasecmp(tok, sd->name, strlen(tok))) continue; if (sort__mode != SORT_MODE__BRANCH) @@ -3369,7 +3406,7 @@ int output_field_add(struct perf_hpp_list *list, char *tok) for (i = 0; i < ARRAY_SIZE(memory_sort_dimensions); i++) { struct sort_dimension *sd = &memory_sort_dimensions[i]; - if (strncasecmp(tok, sd->name, strlen(tok))) + if (!sd->name || strncasecmp(tok, sd->name, strlen(tok))) continue; if (sort__mode != SORT_MODE__MEMORY) @@ -3508,6 +3545,9 @@ void reset_output_field(void) static void add_key(struct strbuf *sb, const char *str, int *llen) { + if (!str) + return; + if (*llen >= 75) { strbuf_addstr(sb, "\n\t\t\t "); *llen = INDENT; diff --git a/tools/perf/util/synthetic-events.c b/tools/perf/util/synthetic-events.c index 0645795ff080..3ab6a92b1a6d 100644 --- a/tools/perf/util/synthetic-events.c +++ b/tools/perf/util/synthetic-events.c @@ -2157,6 +2157,7 @@ int perf_event__synthesize_attr(struct perf_tool *tool, struct perf_event_attr * return err; } +#ifdef HAVE_LIBTRACEEVENT int perf_event__synthesize_tracing_data(struct perf_tool *tool, int fd, struct evlist *evlist, perf_event__handler_t process) { @@ -2203,6 +2204,7 @@ int perf_event__synthesize_tracing_data(struct perf_tool *tool, int fd, struct e return aligned_size; } +#endif int perf_event__synthesize_build_id(struct perf_tool *tool, struct dso *pos, u16 misc, perf_event__handler_t process, struct machine *machine) @@ -2355,6 +2357,7 @@ int perf_event__synthesize_for_pipe(struct perf_tool *tool, } ret += err; +#ifdef HAVE_LIBTRACEEVENT if (have_tracepoints(&evlist->core.entries)) { int fd = perf_data__fd(data); @@ -2374,6 +2377,9 @@ int perf_event__synthesize_for_pipe(struct perf_tool *tool, } ret += err; } +#else + (void)data; +#endif return ret; } diff --git a/tools/perf/util/trace-event-parse.c b/tools/perf/util/trace-event-parse.c index c9c83a40647c..2d3c2576bab7 100644 --- a/tools/perf/util/trace-event-parse.c +++ b/tools/perf/util/trace-event-parse.c @@ -11,6 +11,8 @@ #include "trace-event.h" #include +#include +#include static int get_common_field(struct scripting_context *context, int *offset, int *size, const char *type) diff --git a/tools/perf/util/trace-event-read.c b/tools/perf/util/trace-event-read.c index 43146a4ce2fb..1162c49b8082 100644 --- a/tools/perf/util/trace-event-read.c +++ b/tools/perf/util/trace-event-read.c @@ -11,6 +11,7 @@ #include #include #include +#include #include #include #include diff --git a/tools/perf/util/trace-event-scripting.c b/tools/perf/util/trace-event-scripting.c index 636a010d929b..56175c53f9af 100644 --- a/tools/perf/util/trace-event-scripting.c +++ b/tools/perf/util/trace-event-scripting.c @@ -9,6 +9,7 @@ #include #include #include +#include #include "debug.h" #include "trace-event.h" diff --git a/tools/perf/util/trace-event.c b/tools/perf/util/trace-event.c index b3ee651e3d91..8ad75b31e09b 100644 --- a/tools/perf/util/trace-event.c +++ b/tools/perf/util/trace-event.c @@ -1,5 +1,4 @@ // SPDX-License-Identifier: GPL-2.0 - #include #include #include diff --git a/tools/perf/util/trace-event.h b/tools/perf/util/trace-event.h index 8f39f5bcb2c2..add6c5d9531c 100644 --- a/tools/perf/util/trace-event.h +++ b/tools/perf/util/trace-event.h @@ -2,9 +2,11 @@ #ifndef _PERF_UTIL_TRACE_EVENT_H #define _PERF_UTIL_TRACE_EVENT_H -#include -#include "parse-events.h" +#include +#include +#include +struct evlist; struct machine; struct perf_sample; union perf_event; @@ -18,6 +20,11 @@ struct trace_event { struct tep_plugin_list *plugin_list; }; +typedef char *(tep_func_resolver_t)(void *priv, + unsigned long long *addrp, char **modp); + +bool have_tracepoints(struct list_head *evlist); + int trace_event__init(struct trace_event *t); void trace_event__cleanup(struct trace_event *t); int trace_event__register_resolver(struct machine *machine, -- cgit From e76aff0523f7d3393f967bc13b10bb04b759abfa Mon Sep 17 00:00:00 2001 From: Hagen Paul Pfeifer Date: Tue, 6 Dec 2022 10:44:04 -0500 Subject: perf script: Introduce task analyzer python script Introduce a new 'perf script' to analyze task scheduling behavior. During the task analysis, some data is always needed - which goes beyond the simple time of switching on and off a task (process/thread). This concerns for example the runtime of a process or the frequency with which the process was called. This script serves to simplify this recurring analyze process. It immediately provides the user with helpful task characteristic information about the tasks runtimes. Usage: Recorded can be in two ways: $ perf script record tasks-analyzer -- sleep 10 $ perf record -e sched:sched_switch -a -- sleep 10 The script can parse all perf.data files, most important: sched:sched_switch events are mandatory, other events will be ignored. Most simple report use case is to just call the script without arguments: $ perf script report tasks-analyzer Switched-In Switched-Out CPU PID TID Comm Runtime Time Out-In 15576.658891407 15576.659156086 4 2412 2428 gdbus 265 1949 15576.659111320 15576.659455410 0 2412 2412 gnome-shell 344 2267 15576.659491326 15576.659506173 2 74 74 kworker/2:1 15 13145 15576.659506173 15576.659825748 2 2858 2858 gnome-terminal- 320 63263 15576.659871270 15576.659902872 6 20932 20932 kworker/u16:0 32 2314582 15576.659909951 15576.659945501 3 27264 27264 sh 36 -1 15576.659853285 15576.659971052 7 27265 27265 perf 118 5050741 [...] What is not shown here are the ASCII color sequences. For example, if the task consists of only one thread, the TID is grayed out. Runtime is the time the task was running on the CPU, Time Out-In is the time between the process being scheduled *out* and scheduled back *in*. So the last time span between two executions. If -1 is printed, then the task simply ran the first time in the measurements - a Out-In delta could not be calculated. In addition to the chronological representation, there is a summary on task level. This output can be additionally switched on via the --summary option and provides information such as max, min & average runtime per process. The maximum runtime is often important for debugging. The call looks like this: $ perf script report tasks-analyzer --summary Summary Task Information Runtime Information PID TID Comm Runs Accumulated Mean Median Min Max Max At 14 14 ksoftirqd/0 13 334 26 15 9 127 15571.621211956 15 15 rcu_preempt 133 1778 13 13 2 33 15572.581176024 16 16 migration/0 3 49 16 13 12 24 15571.608915425 20 20 migration/1 3 34 11 13 8 13 15571.639101555 25 25 migration/2 3 32 11 12 9 12 15575.639239896 [...] Besides these two options, there are a number of other options that change the output and behavior. This can be queried via --help. Options worth mentioning include: - filter-tasks - filter out unneeded tasks, --filter-task 1337,/sbin/init - highlight-tasks - more pleasant focusing, --highlight-tasks 1:red,mutt:yellow - extended-times - show combinations of elapsed times between schedule in/schedule out - summary-extended - summary with additional information, like maximum delta time statistics - rename-comms-by-tids - handy for inexpressive processnames like python, --rename 1337:my-python-app - ms - show timestamps in milliseconds, nanoseconds is also possible (--ns) - time-limit - limit the analyzer to a time range, --time-limit 15576.0:15576.1 Script is tested and prime time ready for python2 & python3: - make PYTHON=python3 prefix=/usr/local install - make PYTHON=python2 prefix=/usr/local install Signed-off-by: Hagen Paul Pfeifer Cc: Andi Kleen Cc: Ian Rogers Cc: Jiri Olsa Cc: Namhyung Kim Link: https://lore.kernel.org/r/20221206154406.41941-2-petar.gligor@gmail.com Signed-off-by: Petar Gligoric Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/scripts/python/bin/task-analyzer-record | 2 + tools/perf/scripts/python/bin/task-analyzer-report | 3 + tools/perf/scripts/python/task-analyzer.py | 838 +++++++++++++++++++++ 3 files changed, 843 insertions(+) create mode 100755 tools/perf/scripts/python/bin/task-analyzer-record create mode 100755 tools/perf/scripts/python/bin/task-analyzer-report create mode 100755 tools/perf/scripts/python/task-analyzer.py (limited to 'tools/perf/scripts/python') diff --git a/tools/perf/scripts/python/bin/task-analyzer-record b/tools/perf/scripts/python/bin/task-analyzer-record new file mode 100755 index 000000000000..0f6b51bb2767 --- /dev/null +++ b/tools/perf/scripts/python/bin/task-analyzer-record @@ -0,0 +1,2 @@ +#!/bin/bash +perf record -e sched:sched_switch -e sched:sched_migrate_task "$@" diff --git a/tools/perf/scripts/python/bin/task-analyzer-report b/tools/perf/scripts/python/bin/task-analyzer-report new file mode 100755 index 000000000000..4b16a8cc40a0 --- /dev/null +++ b/tools/perf/scripts/python/bin/task-analyzer-report @@ -0,0 +1,3 @@ +#!/bin/bash +# description: analyze timings of tasks +perf script -s "$PERF_EXEC_PATH"/scripts/python/task-analyzer.py -- "$@" diff --git a/tools/perf/scripts/python/task-analyzer.py b/tools/perf/scripts/python/task-analyzer.py new file mode 100755 index 000000000000..f74abe50f3b2 --- /dev/null +++ b/tools/perf/scripts/python/task-analyzer.py @@ -0,0 +1,838 @@ +# task-analyzer.py - comprehensive perf tasks analysis +# SPDX-License-Identifier: GPL-2.0 +# Copyright (c) 2022, Hagen Paul Pfeifer +# Licensed under the terms of the GNU GPL License version 2 +# +# Usage: +# +# perf record -e sched:sched_switch -a -- sleep 10 +# perf script report task-analyzer +# + +from __future__ import print_function +import sys +import os +import string +import argparse +import decimal + + +sys.path.append( + os.environ["PERF_EXEC_PATH"] + "/scripts/python/Perf-Trace-Util/lib/Perf/Trace" +) +from perf_trace_context import * +from Core import * + +# Definition of possible ASCII color codes +_COLORS = { + "grey": "\033[90m", + "red": "\033[91m", + "green": "\033[92m", + "yellow": "\033[93m", + "blue": "\033[94m", + "violet": "\033[95m", + "reset": "\033[0m", +} + +# Columns will have a static size to align everything properly +# Support of 116 days of active update with nano precision +LEN_SWITCHED_IN = len("9999999.999999999") # 17 +LEN_SWITCHED_OUT = len("9999999.999999999") # 17 +LEN_CPU = len("000") +LEN_PID = len("maxvalue") # 8 +LEN_TID = len("maxvalue") # 8 +LEN_COMM = len("max-comms-length") # 16 +LEN_RUNTIME = len("999999.999") # 10 +# Support of 3.45 hours of timespans +LEN_OUT_IN = len("99999999999.999") # 15 +LEN_OUT_OUT = len("99999999999.999") # 15 +LEN_IN_IN = len("99999999999.999") # 15 +LEN_IN_OUT = len("99999999999.999") # 15 + + +# py2/py3 compatibility layer, see PEP469 +try: + dict.iteritems +except AttributeError: + # py3 + def itervalues(d): + return iter(d.values()) + + def iteritems(d): + return iter(d.items()) + +else: + # py2 + def itervalues(d): + return d.itervalues() + + def iteritems(d): + return d.iteritems() + + +def _check_color(): + global _COLORS + """user enforced no-color or if stdout is no tty we disable colors""" + if sys.stdout.isatty() and args.stdio_color != "never": + return + _COLORS = { + "grey": "", + "red": "", + "green": "", + "yellow": "", + "blue": "", + "violet": "", + "reset": "", + } + + +def _parse_args(): + global args + parser = argparse.ArgumentParser(description="Analyze tasks behavior") + parser.add_argument( + "--time-limit", + default=[], + help= + "print tasks only in time[s] window e.g" + " --time-limit 123.111:789.222(print all between 123.111 and 789.222)" + " --time-limit 123: (print all from 123)" + " --time-limit :456 (print all until incl. 456)", + ) + parser.add_argument( + "--summary", action="store_true", help="print addtional runtime information" + ) + parser.add_argument( + "--summary-only", action="store_true", help="print only summary without traces" + ) + parser.add_argument( + "--summary-extended", + action="store_true", + help="print the summary with additional information of max inter task times" + " relative to the prev task", + ) + parser.add_argument( + "--ns", action="store_true", help="show timestamps in nanoseconds" + ) + parser.add_argument( + "--ms", action="store_true", help="show timestamps in miliseconds" + ) + parser.add_argument( + "--extended-times", + action="store_true", + help="Show the elapsed times between schedule in/schedule out" + " of this task and the schedule in/schedule out of previous occurrence" + " of the same task", + ) + parser.add_argument( + "--filter-tasks", + default=[], + help="filter out unneeded tasks by tid, pid or processname." + " E.g --filter-task 1337,/sbin/init ", + ) + parser.add_argument( + "--limit-to-tasks", + default=[], + help="limit output to selected task by tid, pid, processname." + " E.g --limit-to-tasks 1337,/sbin/init", + ) + parser.add_argument( + "--highlight-tasks", + default="", + help="colorize special tasks by their pid/tid/comm." + " E.g. --highlight-tasks 1:red,mutt:yellow" + " Colors available: red,grey,yellow,blue,violet,green", + ) + parser.add_argument( + "--rename-comms-by-tids", + default="", + help="rename task names by using tid (:,:)" + " This option is handy for inexpressive processnames like python interpreted" + " process. E.g --rename 1337:my-python-app", + ) + parser.add_argument( + "--stdio-color", + default="auto", + choices=["always", "never", "auto"], + help="always, never or auto, allowing configuring color output" + " via the command line", + ) + args = parser.parse_args() + args.tid_renames = dict() + + _argument_filter_sanity_check() + _argument_prepare_check() + + +def time_uniter(unit): + picker = { + "s": 1, + "ms": 1e3, + "us": 1e6, + "ns": 1e9, + } + return picker[unit] + + +def _init_db(): + global db + db = dict() + db["running"] = dict() + db["cpu"] = dict() + db["tid"] = dict() + db["global"] = [] + if args.summary or args.summary_extended or args.summary_only: + db["task_info"] = dict() + db["runtime_info"] = dict() + # min values for summary depending on the header + db["task_info"]["pid"] = len("PID") + db["task_info"]["tid"] = len("TID") + db["task_info"]["comm"] = len("Comm") + db["runtime_info"]["runs"] = len("Runs") + db["runtime_info"]["acc"] = len("Accumulated") + db["runtime_info"]["max"] = len("Max") + db["runtime_info"]["max_at"] = len("Max At") + db["runtime_info"]["min"] = len("Min") + db["runtime_info"]["mean"] = len("Mean") + db["runtime_info"]["median"] = len("Median") + if args.summary_extended: + db["inter_times"] = dict() + db["inter_times"]["out_in"] = len("Out-In") + db["inter_times"]["inter_at"] = len("At") + db["inter_times"]["out_out"] = len("Out-Out") + db["inter_times"]["in_in"] = len("In-In") + db["inter_times"]["in_out"] = len("In-Out") + + +def _median(numbers): + """phython3 hat statistics module - we have nothing""" + n = len(numbers) + index = n // 2 + if n % 2: + return sorted(numbers)[index] + return sum(sorted(numbers)[index - 1 : index + 1]) / 2 + + +def _mean(numbers): + return sum(numbers) / len(numbers) + + +class Timespans(object): + """ + The elapsed time between two occurrences of the same task is being tracked with the + help of this class. There are 4 of those Timespans Out-Out, In-Out, Out-In and + In-In. + The first half of the name signals the first time point of the + first task. The second half of the name represents the second + timepoint of the second task. + """ + + def __init__(self): + self._last_start = None + self._last_finish = None + self.out_out = -1 + self.in_out = -1 + self.out_in = -1 + self.in_in = -1 + if args.summary_extended: + self._time_in = -1 + self.max_out_in = -1 + self.max_at = -1 + self.max_in_out = -1 + self.max_in_in = -1 + self.max_out_out = -1 + + def feed(self, task): + """ + Called for every recorded trace event to find process pair and calculate the + task timespans. Chronological ordering, feed does not do reordering + """ + if not self._last_finish: + self._last_start = task.time_in(time_unit) + self._last_finish = task.time_out(time_unit) + return + self._time_in = task.time_in() + time_in = task.time_in(time_unit) + time_out = task.time_out(time_unit) + self.in_in = time_in - self._last_start + self.out_in = time_in - self._last_finish + self.in_out = time_out - self._last_start + self.out_out = time_out - self._last_finish + if args.summary_extended: + self._update_max_entries() + self._last_finish = task.time_out(time_unit) + self._last_start = task.time_in(time_unit) + + def _update_max_entries(self): + if self.in_in > self.max_in_in: + self.max_in_in = self.in_in + if self.out_out > self.max_out_out: + self.max_out_out = self.out_out + if self.in_out > self.max_in_out: + self.max_in_out = self.in_out + if self.out_in > self.max_out_in: + self.max_out_in = self.out_in + self.max_at = self._time_in + + + + +class Summary(object): + """ + Primary instance for calculating the summary output. Processes the whole trace to + find and memorize relevant data such as mean, max et cetera. This instance handles + dynamic alignment aspects for summary output. + """ + + def __init__(self): + self._body = [] + + class AlignmentHelper: + """ + Used to calculated the alignment for the output of the summary. + """ + def __init__(self, pid, tid, comm, runs, acc, mean, + median, min, max, max_at): + self.pid = pid + self.tid = tid + self.comm = comm + self.runs = runs + self.acc = acc + self.mean = mean + self.median = median + self.min = min + self.max = max + self.max_at = max_at + if args.summary_extended: + self.out_in = None + self.inter_at = None + self.out_out = None + self.in_in = None + self.in_out = None + + def _print_header(self): + ''' + Output is trimmed in _format_stats thus additional adjustment in the header + is needed, depending on the choice of timeunit. The adjustment corresponds + to the amount of column titles being adjusted in _column_titles. + ''' + decimal_precision = 6 if not args.ns else 9 + fmt = " {{:^{}}}".format(sum(db["task_info"].values())) + fmt += " {{:^{}}}".format( + sum(db["runtime_info"].values()) - 2 * decimal_precision + ) + _header = ("Task Information", "Runtime Information") + + if args.summary_extended: + fmt += " {{:^{}}}".format( + sum(db["inter_times"].values()) - 4 * decimal_precision + ) + _header += ("Max Inter Task Times",) + print(fmt.format(*_header)) + + def _column_titles(self): + """ + Cells are being processed and displayed in different way so an alignment adjust + is implemented depeding on the choice of the timeunit. The positions of the max + values are being displayed in grey. Thus in their format two additional {}, + are placed for color set and reset. + """ + decimal_precision, time_precision = _prepare_fmt_precision() + fmt = " {{:>{}}}".format(db["task_info"]["pid"]) + fmt += " {{:>{}}}".format(db["task_info"]["tid"]) + fmt += " {{:>{}}}".format(db["task_info"]["comm"]) + fmt += " {{:>{}}}".format(db["runtime_info"]["runs"]) + fmt += " {{:>{}}}".format(db["runtime_info"]["acc"]) + fmt += " {{:>{}}}".format(db["runtime_info"]["mean"]) + fmt += " {{:>{}}}".format(db["runtime_info"]["median"]) + fmt += " {{:>{}}}".format(db["runtime_info"]["min"] - decimal_precision) + fmt += " {{:>{}}}".format(db["runtime_info"]["max"] - decimal_precision) + fmt += " {{}}{{:>{}}}{{}}".format(db["runtime_info"]["max_at"] - time_precision) + + column_titles = ("PID", "TID", "Comm") + column_titles += ("Runs", "Accumulated", "Mean", "Median", "Min", "Max") + column_titles += (_COLORS["grey"], "At", _COLORS["reset"]) + + if args.summary_extended: + fmt += " {{:>{}}}".format(db["inter_times"]["out_in"] - decimal_precision) + fmt += " {{}}{{:>{}}}{{}}".format( + db["inter_times"]["inter_at"] - time_precision + ) + fmt += " {{:>{}}}".format(db["inter_times"]["out_out"] - decimal_precision) + fmt += " {{:>{}}}".format(db["inter_times"]["in_in"] - decimal_precision) + fmt += " {{:>{}}}".format(db["inter_times"]["in_out"] - decimal_precision) + + column_titles += ("Out-In", _COLORS["grey"], "Max At", _COLORS["reset"], + "Out-Out", "In-In", "In-Out") + print(fmt.format(*column_titles)) + + def _task_stats(self): + """calculates the stats of every task and constructs the printable summary""" + for tid in sorted(db["tid"]): + color_one_sample = _COLORS["grey"] + color_reset = _COLORS["reset"] + no_executed = 0 + runtimes = [] + time_in = [] + timespans = Timespans() + for task in db["tid"][tid]: + pid = task.pid + comm = task.comm + no_executed += 1 + runtimes.append(task.runtime(time_unit)) + time_in.append(task.time_in()) + timespans.feed(task) + if len(runtimes) > 1: + color_one_sample = "" + color_reset = "" + time_max = max(runtimes) + time_min = min(runtimes) + max_at = time_in[runtimes.index(max(runtimes))] + + # The size of the decimal after sum,mean and median varies, thus we cut + # the decimal number, by rounding it. It has no impact on the output, + # because we have a precision of the decimal points at the output. + time_sum = round(sum(runtimes), 3) + time_mean = round(_mean(runtimes), 3) + time_median = round(_median(runtimes), 3) + + align_helper = self.AlignmentHelper(pid, tid, comm, no_executed, time_sum, + time_mean, time_median, time_min, time_max, max_at) + self._body.append([pid, tid, comm, no_executed, time_sum, color_one_sample, + time_mean, time_median, time_min, time_max, + _COLORS["grey"], max_at, _COLORS["reset"], color_reset]) + if args.summary_extended: + self._body[-1].extend([timespans.max_out_in, + _COLORS["grey"], timespans.max_at, + _COLORS["reset"], timespans.max_out_out, + timespans.max_in_in, + timespans.max_in_out]) + align_helper.out_in = timespans.max_out_in + align_helper.inter_at = timespans.max_at + align_helper.out_out = timespans.max_out_out + align_helper.in_in = timespans.max_in_in + align_helper.in_out = timespans.max_in_out + self._calc_alignments_summary(align_helper) + + def _format_stats(self): + decimal_precision, time_precision = _prepare_fmt_precision() + fmt = " {{:>{}d}}".format(db["task_info"]["pid"]) + fmt += " {{:>{}d}}".format(db["task_info"]["tid"]) + fmt += " {{:>{}}}".format(db["task_info"]["comm"]) + fmt += " {{:>{}d}}".format(db["runtime_info"]["runs"]) + fmt += " {{:>{}.{}f}}".format(db["runtime_info"]["acc"], time_precision) + fmt += " {{}}{{:>{}.{}f}}".format(db["runtime_info"]["mean"], time_precision) + fmt += " {{:>{}.{}f}}".format(db["runtime_info"]["median"], time_precision) + fmt += " {{:>{}.{}f}}".format( + db["runtime_info"]["min"] - decimal_precision, time_precision + ) + fmt += " {{:>{}.{}f}}".format( + db["runtime_info"]["max"] - decimal_precision, time_precision + ) + fmt += " {{}}{{:>{}.{}f}}{{}}{{}}".format( + db["runtime_info"]["max_at"] - time_precision, decimal_precision + ) + if args.summary_extended: + fmt += " {{:>{}.{}f}}".format( + db["inter_times"]["out_in"] - decimal_precision, time_precision + ) + fmt += " {{}}{{:>{}.{}f}}{{}}".format( + db["inter_times"]["inter_at"] - time_precision, decimal_precision + ) + fmt += " {{:>{}.{}f}}".format( + db["inter_times"]["out_out"] - decimal_precision, time_precision + ) + fmt += " {{:>{}.{}f}}".format( + db["inter_times"]["in_in"] - decimal_precision, time_precision + ) + fmt += " {{:>{}.{}f}}".format( + db["inter_times"]["in_out"] - decimal_precision, time_precision + ) + return fmt + + + def _calc_alignments_summary(self, align_helper): + # Length is being cut in 3 groups so that further addition is easier to handle. + # The length of every argument from the alignment helper is being checked if it + # is longer than the longest until now. In that case the length is being saved. + for key in db["task_info"]: + if len(str(getattr(align_helper, key))) > db["task_info"][key]: + db["task_info"][key] = len(str(getattr(align_helper, key))) + for key in db["runtime_info"]: + if len(str(getattr(align_helper, key))) > db["runtime_info"][key]: + db["runtime_info"][key] = len(str(getattr(align_helper, key))) + if args.summary_extended: + for key in db["inter_times"]: + if len(str(getattr(align_helper, key))) > db["inter_times"][key]: + db["inter_times"][key] = len(str(getattr(align_helper, key))) + + + def print(self): + print("\nSummary") + self._task_stats() + self._print_header() + self._column_titles() + fmt = self._format_stats() + for i in range(len(self._body)): + print(fmt.format(*tuple(self._body[i]))) + + + +class Task(object): + """ The class is used to handle the information of a given task.""" + + def __init__(self, id, tid, cpu, comm): + self.id = id + self.tid = tid + self.cpu = cpu + self.comm = comm + self.pid = None + self._time_in = None + self._time_out = None + + def schedule_in_at(self, time): + """set the time where the task was scheduled in""" + self._time_in = time + + def schedule_out_at(self, time): + """set the time where the task was scheduled out""" + self._time_out = time + + def time_out(self, unit="s"): + """return time where a given task was scheduled out""" + factor = time_uniter(unit) + return self._time_out * decimal.Decimal(factor) + + def time_in(self, unit="s"): + """return time where a given task was scheduled in""" + factor = time_uniter(unit) + return self._time_in * decimal.Decimal(factor) + + def runtime(self, unit="us"): + factor = time_uniter(unit) + return (self._time_out - self._time_in) * decimal.Decimal(factor) + + def update_pid(self, pid): + self.pid = pid + + +def _task_id(pid, cpu): + """returns a "unique-enough" identifier, please do not change""" + return "{}-{}".format(pid, cpu) + + +def _filter_non_printable(unfiltered): + """comm names may contain loony chars like '\x00000'""" + filtered = "" + for char in unfiltered: + if char not in string.printable: + continue + filtered += char + return filtered + + +def _fmt_header(): + fmt = "{{:>{}}}".format(LEN_SWITCHED_IN) + fmt += " {{:>{}}}".format(LEN_SWITCHED_OUT) + fmt += " {{:>{}}}".format(LEN_CPU) + fmt += " {{:>{}}}".format(LEN_PID) + fmt += " {{:>{}}}".format(LEN_TID) + fmt += " {{:>{}}}".format(LEN_COMM) + fmt += " {{:>{}}}".format(LEN_RUNTIME) + fmt += " {{:>{}}}".format(LEN_OUT_IN) + if args.extended_times: + fmt += " {{:>{}}}".format(LEN_OUT_OUT) + fmt += " {{:>{}}}".format(LEN_IN_IN) + fmt += " {{:>{}}}".format(LEN_IN_OUT) + return fmt + + +def _fmt_body(): + decimal_precision, time_precision = _prepare_fmt_precision() + fmt = "{{}}{{:{}.{}f}}".format(LEN_SWITCHED_IN, decimal_precision) + fmt += " {{:{}.{}f}}".format(LEN_SWITCHED_OUT, decimal_precision) + fmt += " {{:{}d}}".format(LEN_CPU) + fmt += " {{:{}d}}".format(LEN_PID) + fmt += " {{}}{{:{}d}}{{}}".format(LEN_TID) + fmt += " {{}}{{:>{}}}".format(LEN_COMM) + fmt += " {{:{}.{}f}}".format(LEN_RUNTIME, time_precision) + if args.extended_times: + fmt += " {{:{}.{}f}}".format(LEN_OUT_IN, time_precision) + fmt += " {{:{}.{}f}}".format(LEN_OUT_OUT, time_precision) + fmt += " {{:{}.{}f}}".format(LEN_IN_IN, time_precision) + fmt += " {{:{}.{}f}}{{}}".format(LEN_IN_OUT, time_precision) + else: + fmt += " {{:{}.{}f}}{{}}".format(LEN_OUT_IN, time_precision) + return fmt + + +def _print_header(): + fmt = _fmt_header() + header = ("Switched-In", "Switched-Out", "CPU", "PID", "TID", "Comm", "Runtime", + "Time Out-In") + if args.extended_times: + header += ("Time Out-Out", "Time In-In", "Time In-Out") + print(fmt.format(*header)) + + +def _print_task_finish(task): + """calculating every entry of a row and printing it immediately""" + c_row_set = "" + c_row_reset = "" + out_in = -1 + out_out = -1 + in_in = -1 + in_out = -1 + fmt = _fmt_body() + + # depending on user provided highlight option we change the color + # for particular tasks + if str(task.tid) in args.highlight_tasks_map: + c_row_set = _COLORS[args.highlight_tasks_map[str(task.tid)]] + c_row_reset = _COLORS["reset"] + if task.comm in args.highlight_tasks_map: + c_row_set = _COLORS[args.highlight_tasks_map[task.comm]] + c_row_reset = _COLORS["reset"] + # grey-out entries if PID == TID, they + # are identical, no threaded model so the + # thread id (tid) do not matter + c_tid_set = "" + c_tid_reset = "" + if task.pid == task.tid: + c_tid_set = _COLORS["grey"] + c_tid_reset = _COLORS["reset"] + if task.tid in db["tid"]: + # get last task of tid + last_tid_task = db["tid"][task.tid][-1] + # feed the timespan calculate, last in tid db + # and second the current one + timespan_gap_tid = Timespans() + timespan_gap_tid.feed(last_tid_task) + timespan_gap_tid.feed(task) + out_in = timespan_gap_tid.out_in + out_out = timespan_gap_tid.out_out + in_in = timespan_gap_tid.in_in + in_out = timespan_gap_tid.in_out + if args.extended_times: + print(fmt.format(c_row_set, task.time_in(), task.time_out(), task.cpu, task.pid, + c_tid_set, task.tid, c_tid_reset, c_row_set, task.comm, + task.runtime(time_unit), out_in, out_out, in_in, in_out, + c_row_reset)) + else: + print(fmt.format(c_row_set, task.time_in(), task.time_out(), task.cpu, task.pid, + c_tid_set, task.tid, c_tid_reset, c_row_set, task.comm, + task.runtime(time_unit), out_in, c_row_reset)) + + +def _record_cleanup(_list): + """ + no need to store more then one element if --summarize + is not enabled + """ + if not args.summary and len(_list) > 1: + _list = _list[len(_list) - 1 :] + + +def _record_by_tid(task): + tid = task.tid + if tid not in db["tid"]: + db["tid"][tid] = [] + db["tid"][tid].append(task) + _record_cleanup(db["tid"][tid]) + + +def _record_by_cpu(task): + cpu = task.cpu + if cpu not in db["cpu"]: + db["cpu"][cpu] = [] + db["cpu"][cpu].append(task) + _record_cleanup(db["cpu"][cpu]) + + +def _record_global(task): + """record all executed task, ordered by finish chronological""" + db["global"].append(task) + _record_cleanup(db["global"]) + + +def _handle_task_finish(tid, cpu, time, perf_sample_dict): + if tid == 0: + return + _id = _task_id(tid, cpu) + if _id not in db["running"]: + # may happen, if we missed the switch to + # event. Seen in combination with --exclude-perf + # where the start is filtered out, but not the + # switched in. Probably a bug in exclude-perf + # option. + return + task = db["running"][_id] + task.schedule_out_at(time) + + # record tid, during schedule in the tid + # is not available, update now + pid = int(perf_sample_dict["sample"]["pid"]) + + task.update_pid(pid) + del db["running"][_id] + + # print only tasks which are not being filtered and no print of trace + # for summary only, but record every task. + if not _limit_filtered(tid, pid, task.comm) and not args.summary_only: + _print_task_finish(task) + _record_by_tid(task) + _record_by_cpu(task) + _record_global(task) + + +def _handle_task_start(tid, cpu, comm, time): + if tid == 0: + return + if tid in args.tid_renames: + comm = args.tid_renames[tid] + _id = _task_id(tid, cpu) + if _id in db["running"]: + # handle corner cases where already running tasks + # are switched-to again - saw this via --exclude-perf + # recorded traces. We simple ignore this "second start" + # event. + return + assert _id not in db["running"] + task = Task(_id, tid, cpu, comm) + task.schedule_in_at(time) + db["running"][_id] = task + + +def _time_to_internal(time_ns): + """ + To prevent float rounding errors we use Decimal internally + """ + return decimal.Decimal(time_ns) / decimal.Decimal(1e9) + + +def _limit_filtered(tid, pid, comm): + if args.filter_tasks: + if str(tid) in args.filter_tasks or comm in args.filter_tasks: + return True + else: + return False + if args.limit_to_tasks: + if str(tid) in args.limit_to_tasks or comm in args.limit_to_tasks: + return False + else: + return True + + +def _argument_filter_sanity_check(): + if args.limit_to_tasks and args.filter_tasks: + sys.exit("Error: Filter and Limit at the same time active.") + if args.extended_times and args.summary_only: + sys.exit("Error: Summary only and extended times active.") + if args.time_limit and ":" not in args.time_limit: + sys.exit( + "Error: No bound set for time limit. Please set bound by ':' e.g :123." + ) + if args.time_limit and (args.summary or args.summary_only or args.summary_extended): + sys.exit("Error: Cannot set time limit and print summary") + + +def _argument_prepare_check(): + global time_unit + if args.filter_tasks: + args.filter_tasks = args.filter_tasks.split(",") + if args.limit_to_tasks: + args.limit_to_tasks = args.limit_to_tasks.split(",") + if args.time_limit: + args.time_limit = args.time_limit.split(":") + for rename_tuple in args.rename_comms_by_tids.split(","): + tid_name = rename_tuple.split(":") + if len(tid_name) != 2: + continue + args.tid_renames[int(tid_name[0])] = tid_name[1] + args.highlight_tasks_map = dict() + for highlight_tasks_tuple in args.highlight_tasks.split(","): + tasks_color_map = highlight_tasks_tuple.split(":") + # default highlight color to red if no color set by user + if len(tasks_color_map) == 1: + tasks_color_map.append("red") + if args.highlight_tasks and tasks_color_map[1].lower() not in _COLORS: + sys.exit( + "Error: Color not defined, please choose from grey,red,green,yellow,blue," + "violet" + ) + if len(tasks_color_map) != 2: + continue + args.highlight_tasks_map[tasks_color_map[0]] = tasks_color_map[1] + time_unit = "us" + if args.ns: + time_unit = "ns" + elif args.ms: + time_unit = "ms" + + +def _is_within_timelimit(time): + """ + Check if a time limit was given by parameter, if so ignore the rest. If not, + process the recorded trace in its entirety. + """ + if not args.time_limit: + return True + lower_time_limit = args.time_limit[0] + upper_time_limit = args.time_limit[1] + # check for upper limit + if upper_time_limit == "": + if time >= decimal.Decimal(lower_time_limit): + return True + # check for lower limit + if lower_time_limit == "": + if time <= decimal.Decimal(upper_time_limit): + return True + # quit if time exceeds upper limit. Good for big datasets + else: + quit() + if lower_time_limit != "" and upper_time_limit != "": + if (time >= decimal.Decimal(lower_time_limit) and + time <= decimal.Decimal(upper_time_limit)): + return True + # quit if time exceeds upper limit. Good for big datasets + elif time > decimal.Decimal(upper_time_limit): + quit() + +def _prepare_fmt_precision(): + decimal_precision = 6 + time_precision = 3 + if args.ns: + decimal_precision = 9 + time_precision = 0 + return decimal_precision, time_precision + + +def trace_unhandled(event_name, context, event_fields_dict, perf_sample_dict): + pass + + +def trace_begin(): + _parse_args() + _check_color() + _init_db() + if not args.summary_only: + _print_header() + +def trace_end(): + if args.summary or args.summary_extended or args.summary_only: + Summary().print() + +def sched__sched_switch(event_name, context, common_cpu, common_secs, common_nsecs, + common_pid, common_comm, common_callchain, prev_comm, + prev_pid, prev_prio, prev_state, next_comm, next_pid, + next_prio, perf_sample_dict): + # ignore common_secs & common_nsecs cause we need + # high res timestamp anyway, using the raw value is + # faster + time = _time_to_internal(perf_sample_dict["sample"]["time"]) + if not _is_within_timelimit(time): + # user specific --time-limit a:b set + return + + next_comm = _filter_non_printable(next_comm) + _handle_task_finish(prev_pid, common_cpu, time, perf_sample_dict) + _handle_task_start(next_pid, common_cpu, next_comm, time) -- cgit From fdd0f81f0528d55d4362d072c6d6ecc7ecd61def Mon Sep 17 00:00:00 2001 From: Petar Gligoric Date: Tue, 6 Dec 2022 10:44:05 -0500 Subject: perf script: task-analyzer add csv support This patch adds the possibility to write the trace and the summary as csv files to a user specified file. A format as such simplifies further data processing. This is achieved by having ";" as separators instead of spaces and solely one header per file. Additional parameters are being considered, like in the normal usage of the script. Colors are turned off in the case of a csv output, thus the highlight option is also being ignored. Usage: Write standard task to csv file: $ perf script report tasks-analyzer --csv write limited output to csv file in nanoseconds: $ perf script report tasks-analyzer --csv --ns --limit-to-tasks 1337 Write summary to a csv file: $ perf script report tasks-analyzer --csv-summary Write summary to csv file with additional schedule information: $ perf script report tasks-analyzer --csv-summary --summary-extended Write both summary and standard task to a csv file: $ perf script report tasks-analyzer --csv --csv-summary The following examples illustrate what is possible with the CSV output. The first command sequence will record all scheduler switch events for 10 seconds, the task-analyzer calculates task information like runtimes as CSV. A small python snippet using pandas and matplotlib will visualize the most frequent task (e.g. kworker/1:1) runtimes - each runtime as a bar in a bar chart: $ perf record -e sched:sched_switch -a -- sleep 10 $ perf script report tasks-analyzer --ns --csv tasks.csv $ cat << EOF > /tmp/freq-comm-runtimes-bar.py import pandas as pd import matplotlib.pyplot as plt df = pd.read_csv("tasks.csv", sep=';') most_freq_comm = df["COMM"].value_counts().idxmax() most_freq_runtimes = df[df["COMM"]==most_freq_comm]["Runtime"] plt.title(f"Runtimes for Task {most_freq_comm} in Nanoseconds") plt.bar(range(len(most_freq_runtimes)), most_freq_runtimes) plt.show() $ python3 /tmp/freq-comm-runtimes-bar.py As a seconds example, the subsequent script generates a pie chart of all accumulated tasks runtimes for 10 seconds of system recordings: $ perf record -e sched:sched_switch -a -- sleep 10 $ perf script report tasks-analyzer --csv-summary task-summary.csv $ cat << EOF > /tmp/accumulated-task-pie.py import pandas as pd from matplotlib.pyplot import pie, axis, show df = pd.read_csv("task-summary.csv", sep=';') sums = df.groupby(df["Comm"])["Accumulated"].sum() axis("equal") pie(sums, labels=sums.index); show() EOF $ python3 /tmp/accumulated-task-pie.py A variety of other visualizations are possible in matplotlib and other environments. Of course, pandas, numpy and co. also allow easy statistical analysis of the data! Signed-off-by: Petar Gligoric Cc: Andi Kleen Cc: Ian Rogers Cc: Jiri Olsa Cc: Namhyung Kim Link: https://lore.kernel.org/r/20221206154406.41941-3-petar.gligor@gmail.com Signed-off-by: Hagen Paul Pfeifer Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/scripts/python/task-analyzer.py | 274 +++++++++++++++++++---------- 1 file changed, 185 insertions(+), 89 deletions(-) (limited to 'tools/perf/scripts/python') diff --git a/tools/perf/scripts/python/task-analyzer.py b/tools/perf/scripts/python/task-analyzer.py index f74abe50f3b2..52e8dae9b1f0 100755 --- a/tools/perf/scripts/python/task-analyzer.py +++ b/tools/perf/scripts/python/task-analyzer.py @@ -156,6 +156,18 @@ def _parse_args(): help="always, never or auto, allowing configuring color output" " via the command line", ) + parser.add_argument( + "--csv", + default="", + help="Write trace to file selected by user. Options, like --ns or --extended" + "-times are used.", + ) + parser.add_argument( + "--csv-summary", + default="", + help="Write summary to file selected by user. Options, like --ns or" + " --summary-extended are used.", + ) args = parser.parse_args() args.tid_renames = dict() @@ -275,7 +287,6 @@ class Timespans(object): - class Summary(object): """ Primary instance for calculating the summary output. Processes the whole trace to @@ -327,7 +338,7 @@ class Summary(object): sum(db["inter_times"].values()) - 4 * decimal_precision ) _header += ("Max Inter Task Times",) - print(fmt.format(*_header)) + fd_sum.write(fmt.format(*_header) + "\n") def _column_titles(self): """ @@ -336,34 +347,58 @@ class Summary(object): values are being displayed in grey. Thus in their format two additional {}, are placed for color set and reset. """ + separator, fix_csv_align = _prepare_fmt_sep() decimal_precision, time_precision = _prepare_fmt_precision() - fmt = " {{:>{}}}".format(db["task_info"]["pid"]) - fmt += " {{:>{}}}".format(db["task_info"]["tid"]) - fmt += " {{:>{}}}".format(db["task_info"]["comm"]) - fmt += " {{:>{}}}".format(db["runtime_info"]["runs"]) - fmt += " {{:>{}}}".format(db["runtime_info"]["acc"]) - fmt += " {{:>{}}}".format(db["runtime_info"]["mean"]) - fmt += " {{:>{}}}".format(db["runtime_info"]["median"]) - fmt += " {{:>{}}}".format(db["runtime_info"]["min"] - decimal_precision) - fmt += " {{:>{}}}".format(db["runtime_info"]["max"] - decimal_precision) - fmt += " {{}}{{:>{}}}{{}}".format(db["runtime_info"]["max_at"] - time_precision) + fmt = "{{:>{}}}".format(db["task_info"]["pid"] * fix_csv_align) + fmt += "{}{{:>{}}}".format(separator, db["task_info"]["tid"] * fix_csv_align) + fmt += "{}{{:>{}}}".format(separator, db["task_info"]["comm"] * fix_csv_align) + fmt += "{}{{:>{}}}".format(separator, db["runtime_info"]["runs"] * fix_csv_align) + fmt += "{}{{:>{}}}".format(separator, db["runtime_info"]["acc"] * fix_csv_align) + fmt += "{}{{:>{}}}".format(separator, db["runtime_info"]["mean"] * fix_csv_align) + fmt += "{}{{:>{}}}".format( + separator, db["runtime_info"]["median"] * fix_csv_align + ) + fmt += "{}{{:>{}}}".format( + separator, (db["runtime_info"]["min"] - decimal_precision) * fix_csv_align + ) + fmt += "{}{{:>{}}}".format( + separator, (db["runtime_info"]["max"] - decimal_precision) * fix_csv_align + ) + fmt += "{}{{}}{{:>{}}}{{}}".format( + separator, (db["runtime_info"]["max_at"] - time_precision) * fix_csv_align + ) column_titles = ("PID", "TID", "Comm") column_titles += ("Runs", "Accumulated", "Mean", "Median", "Min", "Max") - column_titles += (_COLORS["grey"], "At", _COLORS["reset"]) + column_titles += (_COLORS["grey"], "Max At", _COLORS["reset"]) if args.summary_extended: - fmt += " {{:>{}}}".format(db["inter_times"]["out_in"] - decimal_precision) - fmt += " {{}}{{:>{}}}{{}}".format( - db["inter_times"]["inter_at"] - time_precision + fmt += "{}{{:>{}}}".format( + separator, + (db["inter_times"]["out_in"] - decimal_precision) * fix_csv_align + ) + fmt += "{}{{}}{{:>{}}}{{}}".format( + separator, + (db["inter_times"]["inter_at"] - time_precision) * fix_csv_align + ) + fmt += "{}{{:>{}}}".format( + separator, + (db["inter_times"]["out_out"] - decimal_precision) * fix_csv_align + ) + fmt += "{}{{:>{}}}".format( + separator, + (db["inter_times"]["in_in"] - decimal_precision) * fix_csv_align + ) + fmt += "{}{{:>{}}}".format( + separator, + (db["inter_times"]["in_out"] - decimal_precision) * fix_csv_align ) - fmt += " {{:>{}}}".format(db["inter_times"]["out_out"] - decimal_precision) - fmt += " {{:>{}}}".format(db["inter_times"]["in_in"] - decimal_precision) - fmt += " {{:>{}}}".format(db["inter_times"]["in_out"] - decimal_precision) column_titles += ("Out-In", _COLORS["grey"], "Max At", _COLORS["reset"], "Out-Out", "In-In", "In-Out") - print(fmt.format(*column_titles)) + + fd_sum.write(fmt.format(*column_titles) + "\n") + def _task_stats(self): """calculates the stats of every task and constructs the printable summary""" @@ -414,39 +449,53 @@ class Summary(object): self._calc_alignments_summary(align_helper) def _format_stats(self): + separator, fix_csv_align = _prepare_fmt_sep() decimal_precision, time_precision = _prepare_fmt_precision() - fmt = " {{:>{}d}}".format(db["task_info"]["pid"]) - fmt += " {{:>{}d}}".format(db["task_info"]["tid"]) - fmt += " {{:>{}}}".format(db["task_info"]["comm"]) - fmt += " {{:>{}d}}".format(db["runtime_info"]["runs"]) - fmt += " {{:>{}.{}f}}".format(db["runtime_info"]["acc"], time_precision) - fmt += " {{}}{{:>{}.{}f}}".format(db["runtime_info"]["mean"], time_precision) - fmt += " {{:>{}.{}f}}".format(db["runtime_info"]["median"], time_precision) - fmt += " {{:>{}.{}f}}".format( - db["runtime_info"]["min"] - decimal_precision, time_precision - ) - fmt += " {{:>{}.{}f}}".format( - db["runtime_info"]["max"] - decimal_precision, time_precision - ) - fmt += " {{}}{{:>{}.{}f}}{{}}{{}}".format( - db["runtime_info"]["max_at"] - time_precision, decimal_precision + len_pid = db["task_info"]["pid"] * fix_csv_align + len_tid = db["task_info"]["tid"] * fix_csv_align + len_comm = db["task_info"]["comm"] * fix_csv_align + len_runs = db["runtime_info"]["runs"] * fix_csv_align + len_acc = db["runtime_info"]["acc"] * fix_csv_align + len_mean = db["runtime_info"]["mean"] * fix_csv_align + len_median = db["runtime_info"]["median"] * fix_csv_align + len_min = (db["runtime_info"]["min"] - decimal_precision) * fix_csv_align + len_max = (db["runtime_info"]["max"] - decimal_precision) * fix_csv_align + len_max_at = (db["runtime_info"]["max_at"] - time_precision) * fix_csv_align + if args.summary_extended: + len_out_in = ( + db["inter_times"]["out_in"] - decimal_precision + ) * fix_csv_align + len_inter_at = ( + db["inter_times"]["inter_at"] - time_precision + ) * fix_csv_align + len_out_out = ( + db["inter_times"]["out_out"] - decimal_precision + ) * fix_csv_align + len_in_in = (db["inter_times"]["in_in"] - decimal_precision) * fix_csv_align + len_in_out = ( + db["inter_times"]["in_out"] - decimal_precision + ) * fix_csv_align + + fmt = "{{:{}d}}".format(len_pid) + fmt += "{}{{:{}d}}".format(separator, len_tid) + fmt += "{}{{:>{}}}".format(separator, len_comm) + fmt += "{}{{:{}d}}".format(separator, len_runs) + fmt += "{}{{:{}.{}f}}".format(separator, len_acc, time_precision) + fmt += "{}{{}}{{:{}.{}f}}".format(separator, len_mean, time_precision) + fmt += "{}{{:{}.{}f}}".format(separator, len_median, time_precision) + fmt += "{}{{:{}.{}f}}".format(separator, len_min, time_precision) + fmt += "{}{{:{}.{}f}}".format(separator, len_max, time_precision) + fmt += "{}{{}}{{:{}.{}f}}{{}}{{}}".format( + separator, len_max_at, decimal_precision ) if args.summary_extended: - fmt += " {{:>{}.{}f}}".format( - db["inter_times"]["out_in"] - decimal_precision, time_precision - ) - fmt += " {{}}{{:>{}.{}f}}{{}}".format( - db["inter_times"]["inter_at"] - time_precision, decimal_precision - ) - fmt += " {{:>{}.{}f}}".format( - db["inter_times"]["out_out"] - decimal_precision, time_precision - ) - fmt += " {{:>{}.{}f}}".format( - db["inter_times"]["in_in"] - decimal_precision, time_precision - ) - fmt += " {{:>{}.{}f}}".format( - db["inter_times"]["in_out"] - decimal_precision, time_precision + fmt += "{}{{:{}.{}f}}".format(separator, len_out_in, time_precision) + fmt += "{}{{}}{{:{}.{}f}}{{}}".format( + separator, len_inter_at, decimal_precision ) + fmt += "{}{{:{}.{}f}}".format(separator, len_out_out, time_precision) + fmt += "{}{{:{}.{}f}}".format(separator, len_in_in, time_precision) + fmt += "{}{{:{}.{}f}}".format(separator, len_in_out, time_precision) return fmt @@ -467,13 +516,15 @@ class Summary(object): def print(self): - print("\nSummary") self._task_stats() - self._print_header() - self._column_titles() fmt = self._format_stats() + + if not args.csv_summary: + print("\nSummary") + self._print_header() + self._column_titles() for i in range(len(self._body)): - print(fmt.format(*tuple(self._body[i]))) + fd_sum.write(fmt.format(*tuple(self._body[i])) + "\n") @@ -531,37 +582,45 @@ def _filter_non_printable(unfiltered): def _fmt_header(): - fmt = "{{:>{}}}".format(LEN_SWITCHED_IN) - fmt += " {{:>{}}}".format(LEN_SWITCHED_OUT) - fmt += " {{:>{}}}".format(LEN_CPU) - fmt += " {{:>{}}}".format(LEN_PID) - fmt += " {{:>{}}}".format(LEN_TID) - fmt += " {{:>{}}}".format(LEN_COMM) - fmt += " {{:>{}}}".format(LEN_RUNTIME) - fmt += " {{:>{}}}".format(LEN_OUT_IN) + separator, fix_csv_align = _prepare_fmt_sep() + fmt = "{{:>{}}}".format(LEN_SWITCHED_IN*fix_csv_align) + fmt += "{}{{:>{}}}".format(separator, LEN_SWITCHED_OUT*fix_csv_align) + fmt += "{}{{:>{}}}".format(separator, LEN_CPU*fix_csv_align) + fmt += "{}{{:>{}}}".format(separator, LEN_PID*fix_csv_align) + fmt += "{}{{:>{}}}".format(separator, LEN_TID*fix_csv_align) + fmt += "{}{{:>{}}}".format(separator, LEN_COMM*fix_csv_align) + fmt += "{}{{:>{}}}".format(separator, LEN_RUNTIME*fix_csv_align) + fmt += "{}{{:>{}}}".format(separator, LEN_OUT_IN*fix_csv_align) if args.extended_times: - fmt += " {{:>{}}}".format(LEN_OUT_OUT) - fmt += " {{:>{}}}".format(LEN_IN_IN) - fmt += " {{:>{}}}".format(LEN_IN_OUT) + fmt += "{}{{:>{}}}".format(separator, LEN_OUT_OUT*fix_csv_align) + fmt += "{}{{:>{}}}".format(separator, LEN_IN_IN*fix_csv_align) + fmt += "{}{{:>{}}}".format(separator, LEN_IN_OUT*fix_csv_align) return fmt def _fmt_body(): + separator, fix_csv_align = _prepare_fmt_sep() decimal_precision, time_precision = _prepare_fmt_precision() - fmt = "{{}}{{:{}.{}f}}".format(LEN_SWITCHED_IN, decimal_precision) - fmt += " {{:{}.{}f}}".format(LEN_SWITCHED_OUT, decimal_precision) - fmt += " {{:{}d}}".format(LEN_CPU) - fmt += " {{:{}d}}".format(LEN_PID) - fmt += " {{}}{{:{}d}}{{}}".format(LEN_TID) - fmt += " {{}}{{:>{}}}".format(LEN_COMM) - fmt += " {{:{}.{}f}}".format(LEN_RUNTIME, time_precision) + fmt = "{{}}{{:{}.{}f}}".format(LEN_SWITCHED_IN*fix_csv_align, decimal_precision) + fmt += "{}{{:{}.{}f}}".format( + separator, LEN_SWITCHED_OUT*fix_csv_align, decimal_precision + ) + fmt += "{}{{:{}d}}".format(separator, LEN_CPU*fix_csv_align) + fmt += "{}{{:{}d}}".format(separator, LEN_PID*fix_csv_align) + fmt += "{}{{}}{{:{}d}}{{}}".format(separator, LEN_TID*fix_csv_align) + fmt += "{}{{}}{{:>{}}}".format(separator, LEN_COMM*fix_csv_align) + fmt += "{}{{:{}.{}f}}".format(separator, LEN_RUNTIME*fix_csv_align, time_precision) if args.extended_times: - fmt += " {{:{}.{}f}}".format(LEN_OUT_IN, time_precision) - fmt += " {{:{}.{}f}}".format(LEN_OUT_OUT, time_precision) - fmt += " {{:{}.{}f}}".format(LEN_IN_IN, time_precision) - fmt += " {{:{}.{}f}}{{}}".format(LEN_IN_OUT, time_precision) + fmt += "{}{{:{}.{}f}}".format(separator, LEN_OUT_IN*fix_csv_align, time_precision) + fmt += "{}{{:{}.{}f}}".format(separator, LEN_OUT_OUT*fix_csv_align, time_precision) + fmt += "{}{{:{}.{}f}}".format(separator, LEN_IN_IN*fix_csv_align, time_precision) + fmt += "{}{{:{}.{}f}}{{}}".format( + separator, LEN_IN_OUT*fix_csv_align, time_precision + ) else: - fmt += " {{:{}.{}f}}{{}}".format(LEN_OUT_IN, time_precision) + fmt += "{}{{:{}.{}f}}{{}}".format( + separator, LEN_OUT_IN*fix_csv_align, time_precision + ) return fmt @@ -571,7 +630,8 @@ def _print_header(): "Time Out-In") if args.extended_times: header += ("Time Out-Out", "Time In-In", "Time In-Out") - print(fmt.format(*header)) + fd_task.write(fmt.format(*header) + "\n") + def _print_task_finish(task): @@ -583,7 +643,6 @@ def _print_task_finish(task): in_in = -1 in_out = -1 fmt = _fmt_body() - # depending on user provided highlight option we change the color # for particular tasks if str(task.tid) in args.highlight_tasks_map: @@ -612,16 +671,22 @@ def _print_task_finish(task): out_out = timespan_gap_tid.out_out in_in = timespan_gap_tid.in_in in_out = timespan_gap_tid.in_out + + if args.extended_times: - print(fmt.format(c_row_set, task.time_in(), task.time_out(), task.cpu, task.pid, - c_tid_set, task.tid, c_tid_reset, c_row_set, task.comm, + line_out = fmt.format(c_row_set, task.time_in(), task.time_out(), task.cpu, + task.pid, c_tid_set, task.tid, c_tid_reset, c_row_set, task.comm, task.runtime(time_unit), out_in, out_out, in_in, in_out, - c_row_reset)) + c_row_reset) + "\n" else: - print(fmt.format(c_row_set, task.time_in(), task.time_out(), task.cpu, task.pid, - c_tid_set, task.tid, c_tid_reset, c_row_set, task.comm, - task.runtime(time_unit), out_in, c_row_reset)) - + line_out = fmt.format(c_row_set, task.time_in(), task.time_out(), task.cpu, + task.pid, c_tid_set, task.tid, c_tid_reset, c_row_set, task.comm, + task.runtime(time_unit), out_in, c_row_reset) + "\n" + try: + fd_task.write(line_out) + except(IOError): + # don't mangle the output if user SIGINT this script + sys.exit() def _record_cleanup(_list): """ @@ -733,10 +798,19 @@ def _argument_filter_sanity_check(): ) if args.time_limit and (args.summary or args.summary_only or args.summary_extended): sys.exit("Error: Cannot set time limit and print summary") - + if args.csv_summary: + args.summary = True + if args.csv == args.csv_summary: + sys.exit("Error: Chosen files for csv and csv summary are the same") + if args.csv and (args.summary_extended or args.summary) and not args.csv_summary: + sys.exit("Error: No file chosen to write summary to. Choose with --csv-summary " + "") + if args.csv and args.summary_only: + sys.exit("Error: --csv chosen and --summary-only. Standard task would not be" + "written to csv file.") def _argument_prepare_check(): - global time_unit + global time_unit, fd_task, fd_sum if args.filter_tasks: args.filter_tasks = args.filter_tasks.split(",") if args.limit_to_tasks: @@ -769,6 +843,21 @@ def _argument_prepare_check(): time_unit = "ms" + fd_task = sys.stdout + if args.csv: + args.stdio_color = "never" + fd_task = open(args.csv, "w") + print("generating csv at",args.csv,) + + fd_sum = sys.stdout + if args.csv_summary: + args.stdio_color = "never" + fd_sum = open(args.csv_summary, "w") + print("generating csv summary at",args.csv_summary) + if not args.csv: + args.summary_only = True + + def _is_within_timelimit(time): """ Check if a time limit was given by parameter, if so ignore the rest. If not, @@ -801,10 +890,17 @@ def _prepare_fmt_precision(): decimal_precision = 6 time_precision = 3 if args.ns: - decimal_precision = 9 - time_precision = 0 + decimal_precision = 9 + time_precision = 0 return decimal_precision, time_precision +def _prepare_fmt_sep(): + separator = " " + fix_csv_align = 1 + if args.csv or args.csv_summary: + separator = ";" + fix_csv_align = 0 + return separator, fix_csv_align def trace_unhandled(event_name, context, event_fields_dict, perf_sample_dict): pass -- cgit