From ead1a57457c0324a167f3f9e3a70e26c2d75fb12 Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Mon, 3 Oct 2016 11:34:31 +0100 Subject: perf probe: Check if *ptr2 is zero and not ptr2 Static anaylsis with cppcheck[1] detected an incorrect comparison: [tools/perf/util/probe-event.c:216]: (warning) Char literal compared with pointer 'ptr2'. Did you intend to dereference it? Dereference ptr2 for the comparison to fix this. 1: https://sourceforge.net/p/cppcheck/wiki/Home/ Signed-off-by: Colin King Cc: Alexander Shishkin Cc: Masami Hiramatsu Cc: Peter Zijlstra Cc: Ravi Bangoria Cc: Wang Nan Fixes: 35726d3a4ca9 ("perf probe: Fix to cut off incompatible chars from group name") Link: http://lkml.kernel.org/r/20161003103431.18534-1-colin.king@canonical.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/probe-event.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'tools') diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c index fcfbef07b92d..d281ae2b54e8 100644 --- a/tools/perf/util/probe-event.c +++ b/tools/perf/util/probe-event.c @@ -213,7 +213,7 @@ static int convert_exec_to_group(const char *exec, char **result) goto out; } - for (ptr2 = ptr1; ptr2 != '\0'; ptr2++) { + for (ptr2 = ptr1; *ptr2 != '\0'; ptr2++) { if (!isalnum(*ptr2) && *ptr2 != '_') { *ptr2 = '\0'; break; -- cgit v1.2.3 From 18ef15c675a5d5d97f844ebcf340a2a6c7cf3142 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Mon, 3 Oct 2016 11:07:24 -0300 Subject: perf tools: Experiment with cppcheck Experimenting a bit using cppcheck[1], a static checker brought to my attention by Colin, reducing the scope of some variables, reducing the line of source code lines in the process: $ cppcheck --enable=style tools/perf/util/thread.c Checking tools/perf/util/thread.c... [tools/perf/util/thread.c:17]: (style) The scope of the variable 'leader' can be reduced. [tools/perf/util/thread.c:133]: (style) The scope of the variable 'err' can be reduced. [tools/perf/util/thread.c:273]: (style) The scope of the variable 'err' can be reduced. Will continue later, but these are already useful, keep them. 1: https://sourceforge.net/p/cppcheck/wiki/Home/ Cc: Adrian Hunter Cc: Colin Ian King Cc: David Ahern Cc: Jiri Olsa Cc: Namhyung Kim Cc: Wang Nan Link: http://lkml.kernel.org/n/tip-ixws7lbycihhpmq9cc949ti6@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/evlist.c | 12 +++++------- tools/perf/util/evsel.c | 3 +-- tools/perf/util/machine.c | 6 +++--- tools/perf/util/strbuf.h | 3 +-- tools/perf/util/thread.c | 9 +++------ 5 files changed, 13 insertions(+), 20 deletions(-) (limited to 'tools') diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c index ea34c5a32c11..d92e02006fb8 100644 --- a/tools/perf/util/evlist.c +++ b/tools/perf/util/evlist.c @@ -384,15 +384,14 @@ void perf_evlist__toggle_enable(struct perf_evlist *evlist) static int perf_evlist__enable_event_cpu(struct perf_evlist *evlist, struct perf_evsel *evsel, int cpu) { - int thread, err; + int thread; int nr_threads = perf_evlist__nr_threads(evlist, evsel); if (!evsel->fd) return -EINVAL; for (thread = 0; thread < nr_threads; thread++) { - err = ioctl(FD(evsel, cpu, thread), - PERF_EVENT_IOC_ENABLE, 0); + int err = ioctl(FD(evsel, cpu, thread), PERF_EVENT_IOC_ENABLE, 0); if (err) return err; } @@ -403,14 +402,14 @@ static int perf_evlist__enable_event_thread(struct perf_evlist *evlist, struct perf_evsel *evsel, int thread) { - int cpu, err; + int cpu; int nr_cpus = cpu_map__nr(evlist->cpus); if (!evsel->fd) return -EINVAL; for (cpu = 0; cpu < nr_cpus; cpu++) { - err = ioctl(FD(evsel, cpu, thread), PERF_EVENT_IOC_ENABLE, 0); + int err = ioctl(FD(evsel, cpu, thread), PERF_EVENT_IOC_ENABLE, 0); if (err) return err; } @@ -1606,10 +1605,9 @@ void perf_evlist__close(struct perf_evlist *evlist) struct perf_evsel *evsel; int ncpus = cpu_map__nr(evlist->cpus); int nthreads = thread_map__nr(evlist->threads); - int n; evlist__for_each_entry_reverse(evlist, evsel) { - n = evsel->cpus ? evsel->cpus->nr : ncpus; + int n = evsel->cpus ? evsel->cpus->nr : ncpus; perf_evsel__close(evsel, n, nthreads); } } diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c index 380e84c3af3d..8bc271141d9d 100644 --- a/tools/perf/util/evsel.c +++ b/tools/perf/util/evsel.c @@ -985,14 +985,13 @@ void perf_evsel__config(struct perf_evsel *evsel, struct record_opts *opts, static int perf_evsel__alloc_fd(struct perf_evsel *evsel, int ncpus, int nthreads) { - int cpu, thread; - if (evsel->system_wide) nthreads = 1; evsel->fd = xyarray__new(ncpus, nthreads, sizeof(int)); if (evsel->fd) { + int cpu, thread; for (cpu = 0; cpu < ncpus; cpu++) { for (thread = 0; thread < nthreads; thread++) { FD(evsel, cpu, thread) = -1; diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index 18e4519abef2..df85b9efd80f 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -1745,9 +1745,8 @@ static int resolve_lbr_callchain_sample(struct thread *thread, int max_stack) { struct ip_callchain *chain = sample->callchain; - int chain_nr = min(max_stack, (int)chain->nr); + int chain_nr = min(max_stack, (int)chain->nr), i; u8 cpumode = PERF_RECORD_MISC_USER; - int i, j, err; u64 ip; for (i = 0; i < chain_nr; i++) { @@ -1758,7 +1757,7 @@ static int resolve_lbr_callchain_sample(struct thread *thread, /* LBR only affects the user callchain */ if (i != chain_nr) { struct branch_stack *lbr_stack = sample->branch_stack; - int lbr_nr = lbr_stack->nr; + int lbr_nr = lbr_stack->nr, j; /* * LBR callstack can only get user call chain. * The mix_chain_nr is kernel call chain @@ -1772,6 +1771,7 @@ static int resolve_lbr_callchain_sample(struct thread *thread, int mix_chain_nr = i + 1 + lbr_nr + 1; for (j = 0; j < mix_chain_nr; j++) { + int err; if (callchain_param.order == ORDER_CALLEE) { if (j < i + 1) ip = chain->ips[j]; diff --git a/tools/perf/util/strbuf.h b/tools/perf/util/strbuf.h index b268a6648a5d..318424ea561d 100644 --- a/tools/perf/util/strbuf.h +++ b/tools/perf/util/strbuf.h @@ -66,9 +66,8 @@ static inline ssize_t strbuf_avail(const struct strbuf *sb) { int strbuf_grow(struct strbuf *buf, size_t); static inline int strbuf_setlen(struct strbuf *sb, size_t len) { - int ret; if (!sb->alloc) { - ret = strbuf_grow(sb, 0); + int ret = strbuf_grow(sb, 0); if (ret) return ret; } diff --git a/tools/perf/util/thread.c b/tools/perf/util/thread.c index 8b10a55410a2..f5af87f66663 100644 --- a/tools/perf/util/thread.c +++ b/tools/perf/util/thread.c @@ -14,13 +14,12 @@ int thread__init_map_groups(struct thread *thread, struct machine *machine) { - struct thread *leader; pid_t pid = thread->pid_; if (pid == thread->tid || pid == -1) { thread->mg = map_groups__new(machine); } else { - leader = __machine__findnew_thread(machine, pid, pid); + struct thread *leader = __machine__findnew_thread(machine, pid, pid); if (leader) { thread->mg = map_groups__get(leader->mg); thread__put(leader); @@ -130,11 +129,10 @@ int __thread__set_comm(struct thread *thread, const char *str, u64 timestamp, bool exec) { struct comm *new, *curr = thread__comm(thread); - int err; /* Override the default :tid entry */ if (!thread->comm_set) { - err = comm__override(curr, str, timestamp, exec); + int err = comm__override(curr, str, timestamp, exec); if (err) return err; } else { @@ -270,10 +268,9 @@ static int thread__clone_map_groups(struct thread *thread, int thread__fork(struct thread *thread, struct thread *parent, u64 timestamp) { - int err; - if (parent->comm_set) { const char *comm = thread__comm_str(parent); + int err; if (!comm) return -ENOMEM; err = thread__set_comm(thread, comm, timestamp); -- cgit v1.2.3 From 0c3b7e42616f1f6084cfeb0d443cbff0b2c424a9 Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Tue, 27 Sep 2016 16:18:46 +0200 Subject: tools build: Add support for host programs format In some cases, like for fixdep and shortly for jevents, we need to build a tool to run on the host that will be used in building a tool, such as perf, that is being cross compiled, so do like the kernel and provide HOSTCC, HOSTLD and HOSTAR to do that. Signed-off-by: Jiri Olsa Requested-by: Andi Kleen Requested-and-Tested-by: Arnaldo Carvalho de Melo Cc: Madhavan Srinivasan Cc: Peter Zijlstra Cc: Sukadev Bhattiprolu Cc: linuxppc-dev@lists.ozlabs.org Link: http://lkml.kernel.org/r/20160927141846.GA6589@krava Signed-off-by: Arnaldo Carvalho de Melo --- tools/build/Build.include | 5 +++++ tools/build/Makefile | 6 ++++++ tools/build/Makefile.build | 19 +++++++++++++++---- 3 files changed, 26 insertions(+), 4 deletions(-) (limited to 'tools') diff --git a/tools/build/Build.include b/tools/build/Build.include index 4d000bc959b4..02489380d79b 100644 --- a/tools/build/Build.include +++ b/tools/build/Build.include @@ -90,3 +90,8 @@ if_changed = $(if $(strip $(any-prereq) $(arg-check)), \ # - per object C flags # - BUILD_STR macro to allow '-D"$(variable)"' constructs c_flags = -Wp,-MD,$(depfile),-MT,$@ $(CFLAGS) -D"BUILD_STR(s)=\#s" $(CFLAGS_$(basetarget).o) $(CFLAGS_$(obj)) + +### +## HOSTCC C flags + +host_c_flags = -Wp,-MD,$(depfile),-MT,$@ $(CHOSTFLAGS) -D"BUILD_STR(s)=\#s" $(CHOSTFLAGS_$(basetarget).o) $(CHOSTFLAGS_$(obj)) diff --git a/tools/build/Makefile b/tools/build/Makefile index 0d5a0e3a8fa9..653faee2a055 100644 --- a/tools/build/Makefile +++ b/tools/build/Makefile @@ -14,6 +14,12 @@ endef $(call allow-override,CC,$(CROSS_COMPILE)gcc) $(call allow-override,LD,$(CROSS_COMPILE)ld) +HOSTCC ?= gcc +HOSTLD ?= ld +HOSTAR ?= ar + +export HOSTCC HOSTLD HOSTAR + ifeq ($(V),1) Q = else diff --git a/tools/build/Makefile.build b/tools/build/Makefile.build index 27f3583193e6..190519a94ce5 100644 --- a/tools/build/Makefile.build +++ b/tools/build/Makefile.build @@ -58,6 +58,9 @@ quiet_cmd_mkdir = MKDIR $(dir $@) quiet_cmd_cc_o_c = CC $@ cmd_cc_o_c = $(CC) $(c_flags) -c -o $@ $< +quiet_cmd_host_cc_o_c = HOSTCC $@ + cmd_host_cc_o_c = $(HOSTCC) $(host_c_flags) -c -o $@ $< + quiet_cmd_cpp_i_c = CPP $@ cmd_cpp_i_c = $(CC) $(c_flags) -E -o $@ $< @@ -70,16 +73,24 @@ quiet_cmd_gen = GEN $@ # If there's nothing to link, create empty $@ object. quiet_cmd_ld_multi = LD $@ cmd_ld_multi = $(if $(strip $(obj-y)),\ - $(LD) -r -o $@ $(filter $(obj-y),$^),rm -f $@; $(AR) rcs $@) + $(LD) -r -o $@ $(filter $(obj-y),$^),rm -f $@; $(AR) rcs $@) + +quiet_cmd_host_ld_multi = HOSTLD $@ + cmd_host_ld_multi = $(if $(strip $(obj-y)),\ + $(HOSTLD) -r -o $@ $(filter $(obj-y),$^),rm -f $@; $(HOSTAR) rcs $@) + +ifneq ($(filter $(obj),$(hostprogs)),) + host = host_ +endif # Build rules $(OUTPUT)%.o: %.c FORCE $(call rule_mkdir) - $(call if_changed_dep,cc_o_c) + $(call if_changed_dep,$(host)cc_o_c) $(OUTPUT)%.o: %.S FORCE $(call rule_mkdir) - $(call if_changed_dep,cc_o_c) + $(call if_changed_dep,$(host)cc_o_c) $(OUTPUT)%.i: %.c FORCE $(call rule_mkdir) @@ -119,7 +130,7 @@ $(sort $(subdir-obj-y)): $(subdir-y) ; $(in-target): $(obj-y) FORCE $(call rule_mkdir) - $(call if_changed,ld_multi) + $(call if_changed,$(host)ld_multi) __build: $(in-target) @: -- cgit v1.2.3 From 6b3db6f9b970001b751a6cd001bc2ab581ce0fb3 Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Tue, 27 Sep 2016 16:18:46 +0200 Subject: tools build: Make fixdep a hostprog It is used in the build process, so stop suppressing its build in tools cross builds. Signed-off-by: Jiri Olsa Tested-by: Arnaldo Carvalho de Melo Cc: Andi Kleen Cc: Madhavan Srinivasan Cc: Peter Zijlstra Cc: Sukadev Bhattiprolu Cc: linuxppc-dev@lists.ozlabs.org Link: http://lkml.kernel.org/r/20160927141846.GA6589@krava [ Use HOSTCC on the $(OUTPUT)fixdep target, it was using the x-compiler to link fixdep-in.o, that was correctly built with HOSTCC and thus failing ] Signed-off-by: Arnaldo Carvalho de Melo --- tools/build/Build | 2 ++ tools/build/Makefile | 2 +- tools/build/Makefile.include | 4 ---- 3 files changed, 3 insertions(+), 5 deletions(-) (limited to 'tools') diff --git a/tools/build/Build b/tools/build/Build index 63a6c34c0c88..76d1a4960973 100644 --- a/tools/build/Build +++ b/tools/build/Build @@ -1 +1,3 @@ +hostprogs := fixdep + fixdep-y := fixdep.o diff --git a/tools/build/Makefile b/tools/build/Makefile index 653faee2a055..8332959fbca4 100644 --- a/tools/build/Makefile +++ b/tools/build/Makefile @@ -42,7 +42,7 @@ $(OUTPUT)fixdep-in.o: FORCE $(Q)$(MAKE) $(build)=fixdep $(OUTPUT)fixdep: $(OUTPUT)fixdep-in.o - $(QUIET_LINK)$(CC) $(LDFLAGS) -o $@ $< + $(QUIET_LINK)$(HOSTCC) $(LDFLAGS) -o $@ $< FORCE: diff --git a/tools/build/Makefile.include b/tools/build/Makefile.include index be630bed66d2..ad22e4e7bc59 100644 --- a/tools/build/Makefile.include +++ b/tools/build/Makefile.include @@ -1,10 +1,6 @@ build := -f $(srctree)/tools/build/Makefile.build dir=. obj -ifdef CROSS_COMPILE -fixdep: -else fixdep: $(Q)$(MAKE) -C $(srctree)/tools/build CFLAGS= LDFLAGS= $(OUTPUT)fixdep -endif .PHONY: fixdep -- cgit v1.2.3 From 867a979a83d38fb82c4e7534a0281a02ef1700a3 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 15 Sep 2016 15:24:38 -0700 Subject: perf tools: Add jsmn `jasmine' JSON parser I need a JSON parser. This adds the simplest JSON parser I could find -- Serge Zaitsev's jsmn `jasmine' -- to the perf library. I merely converted it to (mostly) Linux style and added support for non 0 terminated input. The parser is quite straight forward and does not copy any data, just returns tokens with offsets into the input buffer. So it's relatively efficient and simple to use. The code is not fully checkpatch clean, but I didn't want to completely fork the upstream code. Original source: http://zserge.bitbucket.org/jsmn.html In addition I added a simple wrapper that mmaps a json file and provides some straight forward access functions. Used in follow-on patches to parse event files. Signed-off-by: Andi Kleen Acked-by: Ingo Molnar Acked-by: Jiri Olsa Acked-by: Namhyung Kim Cc: Madhavan Srinivasan Cc: Peter Zijlstra Cc: linuxppc-dev@lists.ozlabs.org Link: http://lkml.kernel.org/r/1473978296-20712-2-git-send-email-sukadev@linux.vnet.ibm.com Signed-off-by: Sukadev Bhattiprolu [ Use fcntl.h instead of sys/fcntl.h to fix the build on Alpine Linux 3.4/musl libc, use stdbool.h to avoid clashing with 'bool' typedef there ] Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/pmu-events/jsmn.c | 313 +++++++++++++++++++++++++++++++++++++++++++ tools/perf/pmu-events/jsmn.h | 67 +++++++++ tools/perf/pmu-events/json.c | 162 ++++++++++++++++++++++ tools/perf/pmu-events/json.h | 32 +++++ 4 files changed, 574 insertions(+) create mode 100644 tools/perf/pmu-events/jsmn.c create mode 100644 tools/perf/pmu-events/jsmn.h create mode 100644 tools/perf/pmu-events/json.c create mode 100644 tools/perf/pmu-events/json.h (limited to 'tools') diff --git a/tools/perf/pmu-events/jsmn.c b/tools/perf/pmu-events/jsmn.c new file mode 100644 index 000000000000..11d1fa18bfa5 --- /dev/null +++ b/tools/perf/pmu-events/jsmn.c @@ -0,0 +1,313 @@ +/* + * Copyright (c) 2010 Serge A. Zaitsev + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN + * THE SOFTWARE. + * + * Slightly modified by AK to not assume 0 terminated input. + */ + +#include +#include "jsmn.h" + +/* + * Allocates a fresh unused token from the token pool. + */ +static jsmntok_t *jsmn_alloc_token(jsmn_parser *parser, + jsmntok_t *tokens, size_t num_tokens) +{ + jsmntok_t *tok; + + if ((unsigned)parser->toknext >= num_tokens) + return NULL; + tok = &tokens[parser->toknext++]; + tok->start = tok->end = -1; + tok->size = 0; + return tok; +} + +/* + * Fills token type and boundaries. + */ +static void jsmn_fill_token(jsmntok_t *token, jsmntype_t type, + int start, int end) +{ + token->type = type; + token->start = start; + token->end = end; + token->size = 0; +} + +/* + * Fills next available token with JSON primitive. + */ +static jsmnerr_t jsmn_parse_primitive(jsmn_parser *parser, const char *js, + size_t len, + jsmntok_t *tokens, size_t num_tokens) +{ + jsmntok_t *token; + int start; + + start = parser->pos; + + for (; parser->pos < len; parser->pos++) { + switch (js[parser->pos]) { +#ifndef JSMN_STRICT + /* + * In strict mode primitive must be followed by "," + * or "}" or "]" + */ + case ':': +#endif + case '\t': + case '\r': + case '\n': + case ' ': + case ',': + case ']': + case '}': + goto found; + default: + break; + } + if (js[parser->pos] < 32 || js[parser->pos] >= 127) { + parser->pos = start; + return JSMN_ERROR_INVAL; + } + } +#ifdef JSMN_STRICT + /* + * In strict mode primitive must be followed by a + * comma/object/array. + */ + parser->pos = start; + return JSMN_ERROR_PART; +#endif + +found: + token = jsmn_alloc_token(parser, tokens, num_tokens); + if (token == NULL) { + parser->pos = start; + return JSMN_ERROR_NOMEM; + } + jsmn_fill_token(token, JSMN_PRIMITIVE, start, parser->pos); + parser->pos--; /* parent sees closing brackets */ + return JSMN_SUCCESS; +} + +/* + * Fills next token with JSON string. + */ +static jsmnerr_t jsmn_parse_string(jsmn_parser *parser, const char *js, + size_t len, + jsmntok_t *tokens, size_t num_tokens) +{ + jsmntok_t *token; + int start = parser->pos; + + /* Skip starting quote */ + parser->pos++; + + for (; parser->pos < len; parser->pos++) { + char c = js[parser->pos]; + + /* Quote: end of string */ + if (c == '\"') { + token = jsmn_alloc_token(parser, tokens, num_tokens); + if (token == NULL) { + parser->pos = start; + return JSMN_ERROR_NOMEM; + } + jsmn_fill_token(token, JSMN_STRING, start+1, + parser->pos); + return JSMN_SUCCESS; + } + + /* Backslash: Quoted symbol expected */ + if (c == '\\') { + parser->pos++; + switch (js[parser->pos]) { + /* Allowed escaped symbols */ + case '\"': + case '/': + case '\\': + case 'b': + case 'f': + case 'r': + case 'n': + case 't': + break; + /* Allows escaped symbol \uXXXX */ + case 'u': + /* TODO */ + break; + /* Unexpected symbol */ + default: + parser->pos = start; + return JSMN_ERROR_INVAL; + } + } + } + parser->pos = start; + return JSMN_ERROR_PART; +} + +/* + * Parse JSON string and fill tokens. + */ +jsmnerr_t jsmn_parse(jsmn_parser *parser, const char *js, size_t len, + jsmntok_t *tokens, unsigned int num_tokens) +{ + jsmnerr_t r; + int i; + jsmntok_t *token; + + for (; parser->pos < len; parser->pos++) { + char c; + jsmntype_t type; + + c = js[parser->pos]; + switch (c) { + case '{': + case '[': + token = jsmn_alloc_token(parser, tokens, num_tokens); + if (token == NULL) + return JSMN_ERROR_NOMEM; + if (parser->toksuper != -1) + tokens[parser->toksuper].size++; + token->type = (c == '{' ? JSMN_OBJECT : JSMN_ARRAY); + token->start = parser->pos; + parser->toksuper = parser->toknext - 1; + break; + case '}': + case ']': + type = (c == '}' ? JSMN_OBJECT : JSMN_ARRAY); + for (i = parser->toknext - 1; i >= 0; i--) { + token = &tokens[i]; + if (token->start != -1 && token->end == -1) { + if (token->type != type) + return JSMN_ERROR_INVAL; + parser->toksuper = -1; + token->end = parser->pos + 1; + break; + } + } + /* Error if unmatched closing bracket */ + if (i == -1) + return JSMN_ERROR_INVAL; + for (; i >= 0; i--) { + token = &tokens[i]; + if (token->start != -1 && token->end == -1) { + parser->toksuper = i; + break; + } + } + break; + case '\"': + r = jsmn_parse_string(parser, js, len, tokens, + num_tokens); + if (r < 0) + return r; + if (parser->toksuper != -1) + tokens[parser->toksuper].size++; + break; + case '\t': + case '\r': + case '\n': + case ':': + case ',': + case ' ': + break; +#ifdef JSMN_STRICT + /* + * In strict mode primitives are: + * numbers and booleans. + */ + case '-': + case '0': + case '1': + case '2': + case '3': + case '4': + case '5': + case '6': + case '7': + case '8': + case '9': + case 't': + case 'f': + case 'n': +#else + /* + * In non-strict mode every unquoted value + * is a primitive. + */ + /*FALL THROUGH */ + default: +#endif + r = jsmn_parse_primitive(parser, js, len, tokens, + num_tokens); + if (r < 0) + return r; + if (parser->toksuper != -1) + tokens[parser->toksuper].size++; + break; + +#ifdef JSMN_STRICT + /* Unexpected char in strict mode */ + default: + return JSMN_ERROR_INVAL; +#endif + } + } + + for (i = parser->toknext - 1; i >= 0; i--) { + /* Unmatched opened object or array */ + if (tokens[i].start != -1 && tokens[i].end == -1) + return JSMN_ERROR_PART; + } + + return JSMN_SUCCESS; +} + +/* + * Creates a new parser based over a given buffer with an array of tokens + * available. + */ +void jsmn_init(jsmn_parser *parser) +{ + parser->pos = 0; + parser->toknext = 0; + parser->toksuper = -1; +} + +const char *jsmn_strerror(jsmnerr_t err) +{ + switch (err) { + case JSMN_ERROR_NOMEM: + return "No enough tokens"; + case JSMN_ERROR_INVAL: + return "Invalid character inside JSON string"; + case JSMN_ERROR_PART: + return "The string is not a full JSON packet, more bytes expected"; + case JSMN_SUCCESS: + return "Success"; + default: + return "Unknown json error"; + } +} diff --git a/tools/perf/pmu-events/jsmn.h b/tools/perf/pmu-events/jsmn.h new file mode 100644 index 000000000000..d666b10cf25b --- /dev/null +++ b/tools/perf/pmu-events/jsmn.h @@ -0,0 +1,67 @@ +#ifndef __JSMN_H_ +#define __JSMN_H_ + +/* + * JSON type identifier. Basic types are: + * o Object + * o Array + * o String + * o Other primitive: number, boolean (true/false) or null + */ +typedef enum { + JSMN_PRIMITIVE = 0, + JSMN_OBJECT = 1, + JSMN_ARRAY = 2, + JSMN_STRING = 3 +} jsmntype_t; + +typedef enum { + /* Not enough tokens were provided */ + JSMN_ERROR_NOMEM = -1, + /* Invalid character inside JSON string */ + JSMN_ERROR_INVAL = -2, + /* The string is not a full JSON packet, more bytes expected */ + JSMN_ERROR_PART = -3, + /* Everything was fine */ + JSMN_SUCCESS = 0 +} jsmnerr_t; + +/* + * JSON token description. + * @param type type (object, array, string etc.) + * @param start start position in JSON data string + * @param end end position in JSON data string + */ +typedef struct { + jsmntype_t type; + int start; + int end; + int size; +} jsmntok_t; + +/* + * JSON parser. Contains an array of token blocks available. Also stores + * the string being parsed now and current position in that string + */ +typedef struct { + unsigned int pos; /* offset in the JSON string */ + int toknext; /* next token to allocate */ + int toksuper; /* superior token node, e.g parent object or array */ +} jsmn_parser; + +/* + * Create JSON parser over an array of tokens + */ +void jsmn_init(jsmn_parser *parser); + +/* + * Run JSON parser. It parses a JSON data string into and array of tokens, + * each describing a single JSON object. + */ +jsmnerr_t jsmn_parse(jsmn_parser *parser, const char *js, + size_t len, + jsmntok_t *tokens, unsigned int num_tokens); + +const char *jsmn_strerror(jsmnerr_t err); + +#endif /* __JSMN_H_ */ diff --git a/tools/perf/pmu-events/json.c b/tools/perf/pmu-events/json.c new file mode 100644 index 000000000000..f67bbb0aa36e --- /dev/null +++ b/tools/perf/pmu-events/json.c @@ -0,0 +1,162 @@ +/* Parse JSON files using the JSMN parser. */ + +/* + * Copyright (c) 2014, Intel Corporation + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright notice, + * this list of conditions and the following disclaimer. + * + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE + * COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. +*/ + +#include +#include +#include +#include +#include +#include +#include +#include +#include "jsmn.h" +#include "json.h" +#include + + +static char *mapfile(const char *fn, size_t *size) +{ + unsigned ps = sysconf(_SC_PAGESIZE); + struct stat st; + char *map = NULL; + int err; + int fd = open(fn, O_RDONLY); + + if (fd < 0 && verbose && fn) { + pr_err("Error opening events file '%s': %s\n", fn, + strerror(errno)); + } + + if (fd < 0) + return NULL; + err = fstat(fd, &st); + if (err < 0) + goto out; + *size = st.st_size; + map = mmap(NULL, + (st.st_size + ps - 1) & ~(ps - 1), + PROT_READ|PROT_WRITE, MAP_PRIVATE, fd, 0); + if (map == MAP_FAILED) + map = NULL; +out: + close(fd); + return map; +} + +static void unmapfile(char *map, size_t size) +{ + unsigned ps = sysconf(_SC_PAGESIZE); + munmap(map, roundup(size, ps)); +} + +/* + * Parse json file using jsmn. Return array of tokens, + * and mapped file. Caller needs to free array. + */ +jsmntok_t *parse_json(const char *fn, char **map, size_t *size, int *len) +{ + jsmn_parser parser; + jsmntok_t *tokens; + jsmnerr_t res; + unsigned sz; + + *map = mapfile(fn, size); + if (!*map) + return NULL; + /* Heuristic */ + sz = *size * 16; + tokens = malloc(sz); + if (!tokens) + goto error; + jsmn_init(&parser); + res = jsmn_parse(&parser, *map, *size, tokens, + sz / sizeof(jsmntok_t)); + if (res != JSMN_SUCCESS) { + pr_err("%s: json error %s\n", fn, jsmn_strerror(res)); + goto error_free; + } + if (len) + *len = parser.toknext; + return tokens; +error_free: + free(tokens); +error: + unmapfile(*map, *size); + return NULL; +} + +void free_json(char *map, size_t size, jsmntok_t *tokens) +{ + free(tokens); + unmapfile(map, size); +} + +static int countchar(char *map, char c, int end) +{ + int i; + int count = 0; + for (i = 0; i < end; i++) + if (map[i] == c) + count++; + return count; +} + +/* Return line number of a jsmn token */ +int json_line(char *map, jsmntok_t *t) +{ + return countchar(map, '\n', t->start) + 1; +} + +static const char * const jsmn_types[] = { + [JSMN_PRIMITIVE] = "primitive", + [JSMN_ARRAY] = "array", + [JSMN_OBJECT] = "object", + [JSMN_STRING] = "string" +}; + +#define LOOKUP(a, i) ((i) < (sizeof(a)/sizeof(*(a))) ? ((a)[i]) : "?") + +/* Return type name of a jsmn token */ +const char *json_name(jsmntok_t *t) +{ + return LOOKUP(jsmn_types, t->type); +} + +int json_len(jsmntok_t *t) +{ + return t->end - t->start; +} + +/* Is string t equal to s? */ +int json_streq(char *map, jsmntok_t *t, const char *s) +{ + unsigned len = json_len(t); + return len == strlen(s) && !strncasecmp(map + t->start, s, len); +} diff --git a/tools/perf/pmu-events/json.h b/tools/perf/pmu-events/json.h new file mode 100644 index 000000000000..f2745c7a87bc --- /dev/null +++ b/tools/perf/pmu-events/json.h @@ -0,0 +1,32 @@ +#ifndef JSON_H +#define JSON_H 1 + +#include "jsmn.h" + +jsmntok_t *parse_json(const char *fn, char **map, size_t *size, int *len); +void free_json(char *map, size_t size, jsmntok_t *tokens); +int json_line(char *map, jsmntok_t *t); +const char *json_name(jsmntok_t *t); +int json_streq(char *map, jsmntok_t *t, const char *s); +int json_len(jsmntok_t *t); + +extern int verbose; + +#include + +extern int eprintf(int level, int var, const char *fmt, ...); +#define pr_fmt(fmt) fmt + +#define pr_err(fmt, ...) \ + eprintf(0, verbose, pr_fmt(fmt), ##__VA_ARGS__) + +#ifndef roundup +#define roundup(x, y) ( \ +{ \ + const typeof(y) __y = y; \ + (((x) + (__y - 1)) / __y) * __y; \ +} \ +) +#endif + +#endif -- cgit v1.2.3 From 80eeb67fe577aa76b2d1bb5b029bca097f0f25bc Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Mon, 19 Sep 2016 17:39:33 -0300 Subject: perf jevents: Program to convert JSON file This is a modified version of an earlier patch by Andi Kleen. We expect architectures to create JSON files describing the performance monitoring (PMU) events that each CPU model/family of the architecture supports. Following is an example of the JSON file entry for an x86 event: [ ... { "EventCode": "0x00", "UMask": "0x01", "EventName": "INST_RETIRED.ANY", "BriefDescription": "Instructions retired from execution.", "PublicDescription": "Instructions retired from execution.", "Counter": "Fixed counter 1", "CounterHTOff": "Fixed counter 1", "SampleAfterValue": "2000003", "SampleAfterValue": "2000003", "MSRIndex": "0", "MSRValue": "0", "TakenAlone": "0", "CounterMask": "0", "Invert": "0", "AnyThread": "0", "EdgeDetect": "0", "PEBS": "0", "PRECISE_STORE": "0", "Errata": "null", "Offcore": "0" }, ... ] All the PMU events supported by a CPU model/family must be grouped into "topics" such as "Pipelining", "Floating-point", "Virtual-memory" etc. All events belonging to a topic must be placed in a separate JSON file (eg: "Pipelining.json") and all the topic JSON files for a CPU model must be in a separate directory. Eg: for the CPU model "Silvermont_core": $ ls tools/perf/pmu-events/arch/x86/Silvermont_core Floating-point.json Memory.json Other.json Pipelining.json Virtualmemory.json Finally, to allow multiple CPU models to share a single set of JSON files, architectures must provide a mapping between a model and its set of events: $ grep Silvermont tools/perf/pmu-events/arch/x86/mapfile.csv GenuineIntel-6-4D,V13,Silvermont_core,core GenuineIntel-6-4C,V13,Silvermont_core,core which maps each CPU, identified by [vendor, family, model, version, type] to a directory of JSON files. Thus two (or more) CPU models support the set of PMU events listed in the directory. tools/perf/pmu-events/arch/x86/Silvermont_core/ Given this organization of files, the program, jevents: - locates all JSON files for each CPU-model of the architecture, - parses all JSON files for the CPU-model and generates a C-style "PMU-events table" (pmu-events.c) for the model - locates a mapfile for the architecture - builds a global table, mapping each model of CPU to the corresponding PMU-events table. The 'pmu-events.c' is generated when building perf and added to libperf.a. The global table pmu_events_map[] table in this pmu-events.c will be used in perf in a follow-on patch. If the architecture does not have any JSON files or there is an error in processing them, an empty mapping file is created. This would allow the build of perf to proceed even if we are not able to provide aliases for events. The parser for JSON files allows parsing Intel style JSON event files. This allows to use an Intel event list directly with perf. The Intel event lists can be quite large and are too big to store in unswappable kernel memory. The conversion from JSON to C-style is straight forward. The parser knows (very little) Intel specific information, and can be easily extended to handle fields for other CPUs. The parser code is partially shared with an independent parsing library, which is 2-clause BSD licensed. To avoid any conflicts I marked those files as BSD licensed too. As part of perf they become GPLv2. Committer notes: Fixes: 1) Limit maxfds to 512 to avoid nftd() segfaulting on alloca() with a big rlim_max, as in docker containers - acme 2) Make jevents a hostprog, supporting cross compilation - jolsa 3) Use HOSTCC for jevents final step - acme 4) Define _GNU_SOURCE for asprintf, as we can't use CC's EXTRA_CFLAGS, that has to have --sysroot on the Android NDK 24 - acme 5) Removed $(srctree)/tools/perf/pmu-events/pmu-events.c from the 'clean' target, it is generated on $(OUTPUT)pmu-events/pmu-events.c, which is already taken care of in the original patch - acme Signed-off-by: Andi Kleen Signed-off-by: Jiri Olsa Signed-off-by: Sukadev Bhattiprolu Tested-by: Arnaldo Carvalho de Melo Acked-by: Ingo Molnar Cc: Peter Zijlstra Cc: Madhavan Srinivasan Cc: linuxppc-dev@lists.ozlabs.org Link: http://lkml.kernel.org/r/1473978296-20712-3-git-send-email-sukadev@linux.vnet.ibm.com Link: http://lkml.kernel.org/r/20160927141846.GA6589@krava Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/Makefile.perf | 34 +- tools/perf/pmu-events/Build | 13 + tools/perf/pmu-events/jevents.c | 767 +++++++++++++++++++++++++++++++++++++ tools/perf/pmu-events/jevents.h | 17 + tools/perf/pmu-events/json.h | 6 + tools/perf/pmu-events/pmu-events.h | 36 ++ 6 files changed, 869 insertions(+), 4 deletions(-) create mode 100644 tools/perf/pmu-events/Build create mode 100644 tools/perf/pmu-events/jevents.c create mode 100644 tools/perf/pmu-events/jevents.h create mode 100644 tools/perf/pmu-events/pmu-events.h (limited to 'tools') diff --git a/tools/perf/Makefile.perf b/tools/perf/Makefile.perf index d710db16b963..982d6439bb07 100644 --- a/tools/perf/Makefile.perf +++ b/tools/perf/Makefile.perf @@ -144,6 +144,10 @@ $(call allow-override,LD,$(CROSS_COMPILE)ld) LD += $(EXTRA_LDFLAGS) +HOSTCC ?= gcc +HOSTLD ?= ld +HOSTAR ?= ar + PKG_CONFIG = $(CROSS_COMPILE)pkg-config RM = rm -f @@ -345,8 +349,18 @@ strip: $(PROGRAMS) $(OUTPUT)perf PERF_IN := $(OUTPUT)perf-in.o export srctree OUTPUT RM CC LD AR CFLAGS V BISON FLEX AWK +export HOSTCC HOSTLD HOSTAR include $(srctree)/tools/build/Makefile.include +JEVENTS := $(OUTPUT)pmu-events/jevents +JEVENTS_IN := $(OUTPUT)pmu-events/jevents-in.o + +PMU_EVENTS_IN := $(OUTPUT)pmu-events/pmu-events-in.o + +export JEVENTS + +build := -f $(srctree)/tools/build/Makefile.build dir=. obj + $(PERF_IN): prepare FORCE @(test -f ../../include/uapi/linux/perf_event.h && ( \ (diff -B ../include/uapi/linux/perf_event.h ../../include/uapi/linux/perf_event.h >/dev/null) \ @@ -443,9 +457,18 @@ $(PERF_IN): prepare FORCE || echo "Warning: tools/include/uapi/linux/mman.h differs from kernel" >&2 )) || true $(Q)$(MAKE) $(build)=perf -$(OUTPUT)perf: $(PERFLIBS) $(PERF_IN) $(LIBTRACEEVENT_DYNAMIC_LIST) +$(JEVENTS_IN): FORCE + $(Q)$(MAKE) -f $(srctree)/tools/build/Makefile.build dir=pmu-events obj=jevents + +$(JEVENTS): $(JEVENTS_IN) + $(QUIET_LINK)$(HOSTCC) $(JEVENTS_IN) -o $@ + +$(PMU_EVENTS_IN): $(JEVENTS) FORCE + $(Q)$(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) \ - $(PERF_IN) $(LIBS) -o $@ + $(PERF_IN) $(PMU_EVENTS_IN) $(LIBS) -o $@ $(GTK_IN): fixdep FORCE $(Q)$(MAKE) $(build)=gtk @@ -474,6 +497,8 @@ perf.spec $(SCRIPTS) \ ifneq ($(OUTPUT),) %.o: $(OUTPUT)%.o @echo " # Redirected target $@ => $(OUTPUT)$@" +pmu-events/%.o: $(OUTPUT)pmu-events/%.o + @echo " # Redirected target $@ => $(OUTPUT)$@" util/%.o: $(OUTPUT)util/%.o @echo " # Redirected target $@ => $(OUTPUT)$@" bench/%.o: $(OUTPUT)bench/%.o @@ -729,10 +754,11 @@ clean:: $(LIBTRACEEVENT)-clean $(LIBAPI)-clean $(LIBBPF)-clean $(LIBSUBCMD)-clea $(call QUIET_CLEAN, core-objs) $(RM) $(LIB_FILE) $(OUTPUT)perf-archive $(OUTPUT)perf-with-kcore $(LANG_BINDINGS) $(Q)find $(if $(OUTPUT),$(OUTPUT),.) -name '*.o' -delete -o -name '\.*.cmd' -delete -o -name '\.*.d' -delete $(Q)$(RM) $(OUTPUT).config-detected - $(call QUIET_CLEAN, core-progs) $(RM) $(ALL_PROGRAMS) perf perf-read-vdso32 perf-read-vdsox32 + $(call QUIET_CLEAN, core-progs) $(RM) $(ALL_PROGRAMS) perf perf-read-vdso32 perf-read-vdsox32 $(OUTPUT)pmu-events/jevents $(call QUIET_CLEAN, core-gen) $(RM) *.spec *.pyc *.pyo */*.pyc */*.pyo $(OUTPUT)common-cmds.h TAGS tags cscope* $(OUTPUT)PERF-VERSION-FILE $(OUTPUT)FEATURE-DUMP $(OUTPUT)util/*-bison* $(OUTPUT)util/*-flex* \ $(OUTPUT)util/intel-pt-decoder/inat-tables.c $(OUTPUT)fixdep \ - $(OUTPUT)tests/llvm-src-{base,kbuild,prologue,relocation}.c + $(OUTPUT)tests/llvm-src-{base,kbuild,prologue,relocation}.c \ + $(OUTPUT)pmu-events/pmu-events.c $(QUIET_SUBDIR0)Documentation $(QUIET_SUBDIR1) clean $(python-clean) diff --git a/tools/perf/pmu-events/Build b/tools/perf/pmu-events/Build new file mode 100644 index 000000000000..9213a1273697 --- /dev/null +++ b/tools/perf/pmu-events/Build @@ -0,0 +1,13 @@ +hostprogs := jevents + +jevents-y += json.o jsmn.o jevents.o +pmu-events-y += pmu-events.o +JDIR = pmu-events/arch/$(ARCH) +JSON = $(shell [ -d $(JDIR) ] && \ + find $(JDIR) -name '*.json' -o -name 'mapfile.csv') +# +# Locate/process JSON files in pmu-events/arch/ +# directory and create tables in pmu-events.c. +# +$(OUTPUT)pmu-events/pmu-events.c: $(JSON) $(JEVENTS) + $(Q)$(call echo-cmd,gen)$(JEVENTS) $(ARCH) pmu-events/arch $(OUTPUT)pmu-events/pmu-events.c $(V) diff --git a/tools/perf/pmu-events/jevents.c b/tools/perf/pmu-events/jevents.c new file mode 100644 index 000000000000..c4c074a49b6e --- /dev/null +++ b/tools/perf/pmu-events/jevents.c @@ -0,0 +1,767 @@ +#define _XOPEN_SOURCE 500 /* needed for nftw() */ +#define _GNU_SOURCE /* needed for asprintf() */ + +/* Parse event JSON files */ + +/* + * Copyright (c) 2014, Intel Corporation + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright notice, + * this list of conditions and the following disclaimer. + * + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE + * COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. +*/ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include /* getrlimit */ +#include /* getrlimit */ +#include +#include +#include "jsmn.h" +#include "json.h" +#include "jevents.h" + +#ifndef __maybe_unused +#define __maybe_unused __attribute__((unused)) +#endif + +int verbose; +char *prog; + +int eprintf(int level, int var, const char *fmt, ...) +{ + + int ret; + va_list args; + + if (var < level) + return 0; + + va_start(args, fmt); + + ret = vfprintf(stderr, fmt, args); + + va_end(args); + + return ret; +} + +__attribute__((weak)) char *get_cpu_str(void) +{ + return NULL; +} + +static void addfield(char *map, char **dst, const char *sep, + const char *a, jsmntok_t *bt) +{ + unsigned int len = strlen(a) + 1 + strlen(sep); + int olen = *dst ? strlen(*dst) : 0; + int blen = bt ? json_len(bt) : 0; + char *out; + + out = realloc(*dst, len + olen + blen); + if (!out) { + /* Don't add field in this case */ + return; + } + *dst = out; + + if (!olen) + *(*dst) = 0; + else + strcat(*dst, sep); + strcat(*dst, a); + if (bt) + strncat(*dst, map + bt->start, blen); +} + +static void fixname(char *s) +{ + for (; *s; s++) + *s = tolower(*s); +} + +static void fixdesc(char *s) +{ + char *e = s + strlen(s); + + /* Remove trailing dots that look ugly in perf list */ + --e; + while (e >= s && isspace(*e)) + --e; + if (*e == '.') + *e = 0; +} + +static struct msrmap { + const char *num; + const char *pname; +} msrmap[] = { + { "0x3F6", "ldlat=" }, + { "0x1A6", "offcore_rsp=" }, + { "0x1A7", "offcore_rsp=" }, + { NULL, NULL } +}; + +static struct field { + const char *field; + const char *kernel; +} fields[] = { + { "EventCode", "event=" }, + { "UMask", "umask=" }, + { "CounterMask", "cmask=" }, + { "Invert", "inv=" }, + { "AnyThread", "any=" }, + { "EdgeDetect", "edge=" }, + { "SampleAfterValue", "period=" }, + { NULL, NULL } +}; + +static void cut_comma(char *map, jsmntok_t *newval) +{ + int i; + + /* Cut off everything after comma */ + for (i = newval->start; i < newval->end; i++) { + if (map[i] == ',') + newval->end = i; + } +} + +static int match_field(char *map, jsmntok_t *field, int nz, + char **event, jsmntok_t *val) +{ + struct field *f; + jsmntok_t newval = *val; + + for (f = fields; f->field; f++) + if (json_streq(map, field, f->field) && nz) { + cut_comma(map, &newval); + addfield(map, event, ",", f->kernel, &newval); + return 1; + } + return 0; +} + +static struct msrmap *lookup_msr(char *map, jsmntok_t *val) +{ + jsmntok_t newval = *val; + static bool warned; + int i; + + cut_comma(map, &newval); + for (i = 0; msrmap[i].num; i++) + if (json_streq(map, &newval, msrmap[i].num)) + return &msrmap[i]; + if (!warned) { + warned = true; + pr_err("%s: Unknown MSR in event file %.*s\n", prog, + json_len(val), map + val->start); + } + return NULL; +} + +#define EXPECT(e, t, m) do { if (!(e)) { \ + jsmntok_t *loc = (t); \ + if (!(t)->start && (t) > tokens) \ + loc = (t) - 1; \ + pr_err("%s:%d: " m ", got %s\n", fn, \ + json_line(map, loc), \ + json_name(t)); \ + goto out_free; \ +} } while (0) + +#define TOPIC_DEPTH 256 +static char *topic_array[TOPIC_DEPTH]; +static int topic_level; + +static char *get_topic(void) +{ + char *tp_old, *tp = NULL; + int i; + + for (i = 0; i < topic_level + 1; i++) { + int n; + + tp_old = tp; + n = asprintf(&tp, "%s%s", tp ?: "", topic_array[i]); + if (n < 0) { + pr_info("%s: asprintf() error %s\n", prog); + return NULL; + } + free(tp_old); + } + + for (i = 0; i < (int) strlen(tp); i++) { + char c = tp[i]; + + if (c == '-') + tp[i] = ' '; + else if (c == '.') { + tp[i] = '\0'; + break; + } + } + + return tp; +} + +static int add_topic(int level, char *bname) +{ + char *topic; + + level -= 2; + + if (level >= TOPIC_DEPTH) + return -EINVAL; + + topic = strdup(bname); + if (!topic) { + pr_info("%s: strdup() error %s for file %s\n", prog, + strerror(errno), bname); + return -ENOMEM; + } + + free(topic_array[topic_level]); + topic_array[topic_level] = topic; + topic_level = level; + return 0; +} + +struct perf_entry_data { + FILE *outfp; + char *topic; +}; + +static int close_table; + +static void print_events_table_prefix(FILE *fp, const char *tblname) +{ + fprintf(fp, "struct pmu_event %s[] = {\n", tblname); + close_table = 1; +} + +static int print_events_table_entry(void *data, char *name, char *event, + char *desc) +{ + struct perf_entry_data *pd = data; + FILE *outfp = pd->outfp; + char *topic = pd->topic; + + /* + * TODO: Remove formatting chars after debugging to reduce + * string lengths. + */ + fprintf(outfp, "{\n"); + + fprintf(outfp, "\t.name = \"%s\",\n", name); + fprintf(outfp, "\t.event = \"%s\",\n", event); + fprintf(outfp, "\t.desc = \"%s\",\n", desc); + fprintf(outfp, "\t.topic = \"%s\",\n", topic); + + fprintf(outfp, "},\n"); + + return 0; +} + +static void print_events_table_suffix(FILE *outfp) +{ + fprintf(outfp, "{\n"); + + fprintf(outfp, "\t.name = 0,\n"); + fprintf(outfp, "\t.event = 0,\n"); + fprintf(outfp, "\t.desc = 0,\n"); + + fprintf(outfp, "},\n"); + fprintf(outfp, "};\n"); + close_table = 0; +} + +/* Call func with each event in the json file */ +int json_events(const char *fn, + int (*func)(void *data, char *name, char *event, char *desc), + void *data) +{ + int err = -EIO; + size_t size; + jsmntok_t *tokens, *tok; + int i, j, len; + char *map; + + if (!fn) + return -ENOENT; + + tokens = parse_json(fn, &map, &size, &len); + if (!tokens) + return -EIO; + EXPECT(tokens->type == JSMN_ARRAY, tokens, "expected top level array"); + tok = tokens + 1; + for (i = 0; i < tokens->size; i++) { + char *event = NULL, *desc = NULL, *name = NULL; + struct msrmap *msr = NULL; + jsmntok_t *msrval = NULL; + jsmntok_t *precise = NULL; + jsmntok_t *obj = tok++; + + EXPECT(obj->type == JSMN_OBJECT, obj, "expected object"); + for (j = 0; j < obj->size; j += 2) { + jsmntok_t *field, *val; + int nz; + + field = tok + j; + EXPECT(field->type == JSMN_STRING, tok + j, + "Expected field name"); + val = tok + j + 1; + EXPECT(val->type == JSMN_STRING, tok + j + 1, + "Expected string value"); + + nz = !json_streq(map, val, "0"); + if (match_field(map, field, nz, &event, val)) { + /* ok */ + } else if (json_streq(map, field, "EventName")) { + addfield(map, &name, "", "", val); + } else if (json_streq(map, field, "BriefDescription")) { + addfield(map, &desc, "", "", val); + fixdesc(desc); + } else if (json_streq(map, field, "PEBS") && nz) { + precise = val; + } else if (json_streq(map, field, "MSRIndex") && nz) { + msr = lookup_msr(map, val); + } else if (json_streq(map, field, "MSRValue")) { + msrval = val; + } else if (json_streq(map, field, "Errata") && + !json_streq(map, val, "null")) { + addfield(map, &desc, ". ", + " Spec update: ", val); + } else if (json_streq(map, field, "Data_LA") && nz) { + addfield(map, &desc, ". ", + " Supports address when precise", + NULL); + } + /* ignore unknown fields */ + } + if (precise && desc && !strstr(desc, "(Precise Event)")) { + if (json_streq(map, precise, "2")) + addfield(map, &desc, " ", "(Must be precise)", + NULL); + else + addfield(map, &desc, " ", + "(Precise event)", NULL); + } + if (msr != NULL) + addfield(map, &event, ",", msr->pname, msrval); + fixname(name); + err = func(data, name, event, desc); + free(event); + free(desc); + free(name); + if (err) + break; + tok += j; + } + EXPECT(tok - tokens == len, tok, "unexpected objects at end"); + err = 0; +out_free: + free_json(map, size, tokens); + return err; +} + +static char *file_name_to_table_name(char *fname) +{ + unsigned int i; + int n; + int c; + char *tblname; + + /* + * Ensure tablename starts with alphabetic character. + * Derive rest of table name from basename of the JSON file, + * replacing hyphens and stripping out .json suffix. + */ + n = asprintf(&tblname, "pme_%s", basename(fname)); + if (n < 0) { + pr_info("%s: asprintf() error %s for file %s\n", prog, + strerror(errno), fname); + return NULL; + } + + for (i = 0; i < strlen(tblname); i++) { + c = tblname[i]; + + if (c == '-') + tblname[i] = '_'; + else if (c == '.') { + tblname[i] = '\0'; + break; + } else if (!isalnum(c) && c != '_') { + pr_err("%s: Invalid character '%c' in file name %s\n", + prog, c, basename(fname)); + free(tblname); + tblname = NULL; + break; + } + } + + return tblname; +} + +static void print_mapping_table_prefix(FILE *outfp) +{ + fprintf(outfp, "struct pmu_events_map pmu_events_map[] = {\n"); +} + +static void print_mapping_table_suffix(FILE *outfp) +{ + /* + * Print the terminating, NULL entry. + */ + fprintf(outfp, "{\n"); + fprintf(outfp, "\t.cpuid = 0,\n"); + fprintf(outfp, "\t.version = 0,\n"); + fprintf(outfp, "\t.type = 0,\n"); + fprintf(outfp, "\t.table = 0,\n"); + fprintf(outfp, "},\n"); + + /* and finally, the closing curly bracket for the struct */ + fprintf(outfp, "};\n"); +} + +static int process_mapfile(FILE *outfp, char *fpath) +{ + int n = 16384; + FILE *mapfp; + char *save = NULL; + char *line, *p; + int line_num; + char *tblname; + + pr_info("%s: Processing mapfile %s\n", prog, fpath); + + line = malloc(n); + if (!line) + return -1; + + mapfp = fopen(fpath, "r"); + if (!mapfp) { + pr_info("%s: Error %s opening %s\n", prog, strerror(errno), + fpath); + return -1; + } + + print_mapping_table_prefix(outfp); + + line_num = 0; + while (1) { + char *cpuid, *version, *type, *fname; + + line_num++; + p = fgets(line, n, mapfp); + if (!p) + break; + + if (line[0] == '#' || line[0] == '\n') + continue; + + if (line[strlen(line)-1] != '\n') { + /* TODO Deal with lines longer than 16K */ + pr_info("%s: Mapfile %s: line %d too long, aborting\n", + prog, fpath, line_num); + return -1; + } + line[strlen(line)-1] = '\0'; + + cpuid = strtok_r(p, ",", &save); + version = strtok_r(NULL, ",", &save); + fname = strtok_r(NULL, ",", &save); + type = strtok_r(NULL, ",", &save); + + tblname = file_name_to_table_name(fname); + fprintf(outfp, "{\n"); + fprintf(outfp, "\t.cpuid = \"%s\",\n", cpuid); + fprintf(outfp, "\t.version = \"%s\",\n", version); + fprintf(outfp, "\t.type = \"%s\",\n", type); + + /* + * CHECK: We can't use the type (eg "core") field in the + * table name. For us to do that, we need to somehow tweak + * the other caller of file_name_to_table(), process_json() + * to determine the type. process_json() file has no way + * of knowing these are "core" events unless file name has + * core in it. If filename has core in it, we can safely + * ignore the type field here also. + */ + fprintf(outfp, "\t.table = %s\n", tblname); + fprintf(outfp, "},\n"); + } + + print_mapping_table_suffix(outfp); + + return 0; +} + +/* + * If we fail to locate/process JSON and map files, create a NULL mapping + * table. This would at least allow perf to build even if we can't find/use + * the aliases. + */ +static void create_empty_mapping(const char *output_file) +{ + FILE *outfp; + + pr_info("%s: Creating empty pmu_events_map[] table\n", prog); + + /* Truncate file to clear any partial writes to it */ + outfp = fopen(output_file, "w"); + if (!outfp) { + perror("fopen()"); + _Exit(1); + } + + fprintf(outfp, "#include \"../../pmu-events/pmu-events.h\"\n"); + print_mapping_table_prefix(outfp); + print_mapping_table_suffix(outfp); + fclose(outfp); +} + +static int get_maxfds(void) +{ + struct rlimit rlim; + + if (getrlimit(RLIMIT_NOFILE, &rlim) == 0) + return min((int)rlim.rlim_max / 2, 512); + + return 512; +} + +/* + * nftw() doesn't let us pass an argument to the processing function, + * so use a global variables. + */ +static FILE *eventsfp; +static char *mapfile; + +static int process_one_file(const char *fpath, const struct stat *sb, + int typeflag, struct FTW *ftwbuf) +{ + char *tblname, *bname = (char *) fpath + ftwbuf->base; + int is_dir = typeflag == FTW_D; + int is_file = typeflag == FTW_F; + int level = ftwbuf->level; + int err = 0; + + pr_debug("%s %d %7jd %-20s %s\n", + is_file ? "f" : is_dir ? "d" : "x", + level, sb->st_size, bname, fpath); + + /* base dir */ + if (level == 0) + return 0; + + /* model directory, reset topic */ + if (level == 1 && is_dir) { + if (close_table) + print_events_table_suffix(eventsfp); + + /* + * Drop file name suffix. Replace hyphens with underscores. + * Fail if file name contains any alphanum characters besides + * underscores. + */ + tblname = file_name_to_table_name(bname); + if (!tblname) { + pr_info("%s: Error determining table name for %s\n", prog, + bname); + return -1; + } + + print_events_table_prefix(eventsfp, tblname); + return 0; + } + + /* + * Save the mapfile name for now. We will process mapfile + * after processing all JSON files (so we can write out the + * mapping table after all PMU events tables). + * + * TODO: Allow for multiple mapfiles? Punt for now. + */ + if (level == 1 && is_file) { + if (!strncmp(bname, "mapfile.csv", 11)) { + if (mapfile) { + pr_info("%s: Many mapfiles? Using %s, ignoring %s\n", + prog, mapfile, fpath); + } else { + mapfile = strdup(fpath); + } + return 0; + } + + pr_info("%s: Ignoring file %s\n", prog, fpath); + return 0; + } + + /* + * If the file name does not have a .json extension, + * ignore it. It could be a readme.txt for instance. + */ + if (is_file) { + char *suffix = bname + strlen(bname) - 5; + + if (strncmp(suffix, ".json", 5)) { + pr_info("%s: Ignoring file without .json suffix %s\n", prog, + fpath); + return 0; + } + } + + if (level > 1 && add_topic(level, bname)) + return -ENOMEM; + + /* + * Assume all other files are JSON files. + * + * If mapfile refers to 'power7_core.json', we create a table + * named 'power7_core'. Any inconsistencies between the mapfile + * and directory tree could result in build failure due to table + * names not being found. + * + * Atleast for now, be strict with processing JSON file names. + * i.e. if JSON file name cannot be mapped to C-style table name, + * fail. + */ + if (is_file) { + struct perf_entry_data data = { + .topic = get_topic(), + .outfp = eventsfp, + }; + + err = json_events(fpath, print_events_table_entry, &data); + + free(data.topic); + } + + return err; +} + +#ifndef PATH_MAX +#define PATH_MAX 4096 +#endif + +/* + * Starting in directory 'start_dirname', find the "mapfile.csv" and + * the set of JSON files for the architecture 'arch'. + * + * From each JSON file, create a C-style "PMU events table" from the + * JSON file (see struct pmu_event). + * + * From the mapfile, create a mapping between the CPU revisions and + * PMU event tables (see struct pmu_events_map). + * + * Write out the PMU events tables and the mapping table to pmu-event.c. + * + * If unable to process the JSON or arch files, create an empty mapping + * table so we can continue to build/use perf even if we cannot use the + * PMU event aliases. + */ +int main(int argc, char *argv[]) +{ + int rc; + int maxfds; + char ldirname[PATH_MAX]; + + const char *arch; + const char *output_file; + const char *start_dirname; + + prog = basename(argv[0]); + if (argc < 4) { + pr_err("Usage: %s \n", prog); + return 1; + } + + arch = argv[1]; + start_dirname = argv[2]; + output_file = argv[3]; + + if (argc > 4) + verbose = atoi(argv[4]); + + eventsfp = fopen(output_file, "w"); + if (!eventsfp) { + pr_err("%s Unable to create required file %s (%s)\n", + prog, output_file, strerror(errno)); + return 2; + } + + /* Include pmu-events.h first */ + fprintf(eventsfp, "#include \"../../pmu-events/pmu-events.h\"\n"); + + sprintf(ldirname, "%s/%s", start_dirname, arch); + + /* + * The mapfile allows multiple CPUids to point to the same JSON file, + * so, not sure if there is a need for symlinks within the pmu-events + * directory. + * + * For now, treat symlinks of JSON files as regular files and create + * separate tables for each symlink (presumably, each symlink refers + * to specific version of the CPU). + */ + + maxfds = get_maxfds(); + mapfile = NULL; + rc = nftw(ldirname, process_one_file, maxfds, 0); + if (rc && verbose) { + pr_info("%s: Error walking file tree %s\n", prog, ldirname); + goto empty_map; + } else if (rc) { + goto empty_map; + } + + if (close_table) + print_events_table_suffix(eventsfp); + + if (!mapfile) { + pr_info("%s: No CPU->JSON mapping?\n", prog); + goto empty_map; + } + + if (process_mapfile(eventsfp, mapfile)) { + pr_info("%s: Error processing mapfile %s\n", prog, mapfile); + goto empty_map; + } + + return 0; + +empty_map: + fclose(eventsfp); + create_empty_mapping(output_file); + return 0; +} diff --git a/tools/perf/pmu-events/jevents.h b/tools/perf/pmu-events/jevents.h new file mode 100644 index 000000000000..996601f828b6 --- /dev/null +++ b/tools/perf/pmu-events/jevents.h @@ -0,0 +1,17 @@ +#ifndef JEVENTS_H +#define JEVENTS_H 1 + +int json_events(const char *fn, + int (*func)(void *data, char *name, char *event, char *desc), + void *data); +char *get_cpu_str(void); + +#ifndef min +#define min(x, y) ({ \ + typeof(x) _min1 = (x); \ + typeof(y) _min2 = (y); \ + (void) (&_min1 == &_min2); \ + _min1 < _min2 ? _min1 : _min2; }) +#endif + +#endif diff --git a/tools/perf/pmu-events/json.h b/tools/perf/pmu-events/json.h index f2745c7a87bc..278ebd32cfb6 100644 --- a/tools/perf/pmu-events/json.h +++ b/tools/perf/pmu-events/json.h @@ -20,6 +20,12 @@ extern int eprintf(int level, int var, const char *fmt, ...); #define pr_err(fmt, ...) \ eprintf(0, verbose, pr_fmt(fmt), ##__VA_ARGS__) +#define pr_info(fmt, ...) \ + eprintf(1, verbose, pr_fmt(fmt), ##__VA_ARGS__) + +#define pr_debug(fmt, ...) \ + eprintf(2, verbose, pr_fmt(fmt), ##__VA_ARGS__) + #ifndef roundup #define roundup(x, y) ( \ { \ diff --git a/tools/perf/pmu-events/pmu-events.h b/tools/perf/pmu-events/pmu-events.h new file mode 100644 index 000000000000..70d54794e3cb --- /dev/null +++ b/tools/perf/pmu-events/pmu-events.h @@ -0,0 +1,36 @@ +#ifndef PMU_EVENTS_H +#define PMU_EVENTS_H + +/* + * Describe each PMU event. Each CPU has a table of PMU events. + */ +struct pmu_event { + const char *name; + const char *event; + const char *desc; + const char *topic; +}; + +/* + * + * Map a CPU to its table of PMU events. The CPU is identified by the + * cpuid field, which is an arch-specific identifier for the CPU. + * The identifier specified in tools/perf/pmu-events/arch/xxx/mapfile + * must match the get_cpustr() in tools/perf/arch/xxx/util/header.c) + * + * The cpuid can contain any character other than the comma. + */ +struct pmu_events_map { + const char *cpuid; + const char *version; + const char *type; /* core, uncore etc */ + struct pmu_event *table; +}; + +/* + * Global table mapping each known CPU for the architecture to its + * table of PMU events. + */ +extern struct pmu_events_map pmu_events_map[]; + +#endif -- cgit v1.2.3 From 933f82ff72d7d1641663462f61f3056ee1fe3f8b Mon Sep 17 00:00:00 2001 From: Sukadev Bhattiprolu Date: Thu, 15 Sep 2016 15:24:40 -0700 Subject: perf pmu: Use pmu_events table to create aliases At run time (when 'perf' is starting up), locate the specific table of PMU events that corresponds to the current CPU. Using that table, create aliases for the each of the PMU events in the CPU. The use these aliases to parse the user specified perf event. In short this would allow the user to specify events using their aliases rather than raw event codes. Based on input and some earlier patches from Andi Kleen, Jiri Olsa. Signed-off-by: Sukadev Bhattiprolu Acked-by: Ingo Molnar Acked-by: Jiri Olsa Cc: Madhavan Srinivasan Cc: Peter Zijlstra Cc: linuxppc-dev@lists.ozlabs.org Link: http://lkml.kernel.org/r/1473978296-20712-4-git-send-email-sukadev@linux.vnet.ibm.com [ Make pmu_add_cpu_aliases() return void, since it was returning just '0' and furthermore, even that was being discarded via an explicit (void) cast ] Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/header.h | 1 + tools/perf/util/pmu.c | 60 ++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 61 insertions(+) (limited to 'tools') diff --git a/tools/perf/util/header.h b/tools/perf/util/header.h index d306ca118449..d30109b421ee 100644 --- a/tools/perf/util/header.h +++ b/tools/perf/util/header.h @@ -151,4 +151,5 @@ int write_padded(int fd, const void *bf, size_t count, size_t count_aligned); */ int get_cpuid(char *buffer, size_t sz); +char *get_cpuid_str(void); #endif /* __PERF_HEADER_H */ diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c index 2babcdf62839..10668b7f5272 100644 --- a/tools/perf/util/pmu.c +++ b/tools/perf/util/pmu.c @@ -12,6 +12,8 @@ #include "pmu.h" #include "parse-events.h" #include "cpumap.h" +#include "header.h" +#include "pmu-events/pmu-events.h" struct perf_pmu_format { char *name; @@ -473,6 +475,61 @@ static struct cpu_map *pmu_cpumask(const char *name) return cpus; } +/* + * Return the CPU id as a raw string. + * + * Each architecture should provide a more precise id string that + * can be use to match the architecture's "mapfile". + */ +char * __weak get_cpuid_str(void) +{ + return NULL; +} + +/* + * From the pmu_events_map, find the table of PMU events that corresponds + * to the current running CPU. Then, add all PMU events from that table + * as aliases. + */ +static void pmu_add_cpu_aliases(struct list_head *head) +{ + int i; + struct pmu_events_map *map; + struct pmu_event *pe; + char *cpuid; + + cpuid = get_cpuid_str(); + if (!cpuid) + return; + + i = 0; + while (1) { + map = &pmu_events_map[i++]; + if (!map->table) + goto out; + + if (!strcmp(map->cpuid, cpuid)) + break; + } + + /* + * Found a matching PMU events table. Create aliases + */ + i = 0; + while (1) { + pe = &map->table[i++]; + if (!pe->name) + break; + + /* need type casts to override 'const' */ + __perf_pmu__new_alias(head, NULL, (char *)pe->name, + (char *)pe->desc, (char *)pe->event); + } + +out: + free(cpuid); +} + struct perf_event_attr * __weak perf_pmu__get_default_config(struct perf_pmu *pmu __maybe_unused) { @@ -497,6 +554,9 @@ static struct perf_pmu *pmu_lookup(const char *name) if (pmu_aliases(name, &aliases)) return NULL; + if (!strcmp(name, "cpu")) + pmu_add_cpu_aliases(&aliases); + if (pmu_type(name, &type)) return NULL; -- cgit v1.2.3 From ce88f27ccc724fa8e6766fdd4140768931f42f88 Mon Sep 17 00:00:00 2001 From: Sukadev Bhattiprolu Date: Thu, 15 Sep 2016 15:24:41 -0700 Subject: perf powerpc: Support CPU ID matching for Powerpc Implement code that returns the generic CPU ID string for Powerpc. This will be used to identify the specific table of PMU events to parse/compare user specified events against. Signed-off-by: Sukadev Bhattiprolu Acked-by: Ingo Molnar Acked-by: Jiri Olsa Cc: Madhavan Srinivasan Cc: Peter Zijlstra Cc: linuxppc-dev@lists.ozlabs.org Link: http://lkml.kernel.org/r/1473978296-20712-5-git-send-email-sukadev@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/arch/powerpc/util/header.c | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'tools') diff --git a/tools/perf/arch/powerpc/util/header.c b/tools/perf/arch/powerpc/util/header.c index f8ccee132867..9aaa6f5a9347 100644 --- a/tools/perf/arch/powerpc/util/header.c +++ b/tools/perf/arch/powerpc/util/header.c @@ -32,3 +32,14 @@ get_cpuid(char *buffer, size_t sz) } return -1; } + +char * +get_cpuid_str(void) +{ + char *bufp; + + if (asprintf(&bufp, "%.8lx", mfspr(SPRN_PVR)) < 0) + bufp = NULL; + + return bufp; +} -- cgit v1.2.3 From f33d1227197e0fe9d3682c5a766fdc74559da78e Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 15 Sep 2016 15:24:42 -0700 Subject: perf tools: Support CPU id matching for x86 v2 Implement the code to match CPU types to mapfile types for x86 based on CPUID. This extends an existing similar function, but changes it to use the x86 mapfile cpu description. This allows to resolve event lists generated by jevents. Signed-off-by: Andi Kleen Signed-off-by: Sukadev Bhattiprolu Acked-by: Ingo Molnar Acked-by: Jiri Olsa Cc: Madhavan Srinivasan Cc: Peter Zijlstra Cc: linuxppc-dev@lists.ozlabs.org Link: http://lkml.kernel.org/r/1473978296-20712-6-git-send-email-sukadev@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/arch/x86/util/header.c | 24 +++++++++++++++++++++--- 1 file changed, 21 insertions(+), 3 deletions(-) (limited to 'tools') diff --git a/tools/perf/arch/x86/util/header.c b/tools/perf/arch/x86/util/header.c index 146d12a1cec0..a74a48db26f5 100644 --- a/tools/perf/arch/x86/util/header.c +++ b/tools/perf/arch/x86/util/header.c @@ -19,8 +19,8 @@ cpuid(unsigned int op, unsigned int *a, unsigned int *b, unsigned int *c, : "a" (op)); } -int -get_cpuid(char *buffer, size_t sz) +static int +__get_cpuid(char *buffer, size_t sz, const char *fmt) { unsigned int a, b, c, d, lvl; int family = -1, model = -1, step = -1; @@ -48,7 +48,7 @@ get_cpuid(char *buffer, size_t sz) if (family >= 0x6) model += ((a >> 16) & 0xf) << 4; } - nb = scnprintf(buffer, sz, "%s,%u,%u,%u$", vendor, family, model, step); + nb = scnprintf(buffer, sz, fmt, vendor, family, model, step); /* look for end marker to ensure the entire data fit */ if (strchr(buffer, '$')) { @@ -57,3 +57,21 @@ get_cpuid(char *buffer, size_t sz) } return -1; } + +int +get_cpuid(char *buffer, size_t sz) +{ + return __get_cpuid(buffer, sz, "%s,%u,%u,%u$"); +} + +char * +get_cpuid_str(void) +{ + char *buf = malloc(128); + + if (__get_cpuid(buf, 128, "%s-%u-%X$") < 0) { + free(buf); + return NULL; + } + return buf; +} -- cgit v1.2.3 From dc720ffc9863b618a192f5949234c54aad00ed24 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 15 Sep 2016 15:24:51 -0700 Subject: perf jevents: Handle header line in mapfile To work with existing mapfiles, assume that the first line in 'mapfile.csv' is a header line and skip over it. Signed-off-by: Sukadev Bhattiprolu Acked-by: Ingo Molnar Acked-by: Jiri Olsa Cc: Madhavan Srinivasan Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/1473978296-20712-15-git-send-email-sukadev@linux.vnet.ibm.com Cc: linuxppc-dev@lists.ozlabs.org Signed-off-by: Andi Kleen Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/pmu-events/jevents.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'tools') diff --git a/tools/perf/pmu-events/jevents.c b/tools/perf/pmu-events/jevents.c index c4c074a49b6e..c9bf9a7dc7b2 100644 --- a/tools/perf/pmu-events/jevents.c +++ b/tools/perf/pmu-events/jevents.c @@ -477,7 +477,12 @@ static int process_mapfile(FILE *outfp, char *fpath) print_mapping_table_prefix(outfp); - line_num = 0; + /* Skip first line (header) */ + p = fgets(line, n, mapfp); + if (!p) + goto out; + + line_num = 1; while (1) { char *cpuid, *version, *type, *fname; @@ -521,8 +526,8 @@ static int process_mapfile(FILE *outfp, char *fpath) fprintf(outfp, "},\n"); } +out: print_mapping_table_suffix(outfp); - return 0; } -- cgit v1.2.3 From 08e60ed15d0483be38a87d17538ccf02acff5b1f Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 15 Sep 2016 15:24:43 -0700 Subject: perf pmu: Support alias descriptions Add support to print alias descriptions in perf list, which are taken from the generated event files. The sorting code is changed to put the events with descriptions at the end. The descriptions are printed as possibly multiple word wrapped lines. Example output: % perf list ... arith.fpu_div [Divide operations executed] arith.fpu_div_active [Cycles when divider is busy executing divide operations] Committer notes: Further testing on a Broadwell machine (ThinkPad t450s), using these files: $ find tools/perf/pmu-events/arch/x86/ tools/perf/pmu-events/arch/x86/ tools/perf/pmu-events/arch/x86/Broadwell tools/perf/pmu-events/arch/x86/Broadwell/Cache.json tools/perf/pmu-events/arch/x86/Broadwell/Other.json tools/perf/pmu-events/arch/x86/Broadwell/Frontend.json tools/perf/pmu-events/arch/x86/Broadwell/Virtual-Memory.json tools/perf/pmu-events/arch/x86/Broadwell/Pipeline.json tools/perf/pmu-events/arch/x86/Broadwell/Floating-point.json tools/perf/pmu-events/arch/x86/Broadwell/Memory.json tools/perf/pmu-events/arch/x86/mapfile.csv $ Taken from: https://github.com/sukadev/linux/tree/json-code+data-v21/tools/perf/pmu-events/arch/x86/ to get this machinery to actually parse JSON files, generate $(OUTPUT)pmu-events/pmu-events.c, compile it and link it with perf, that will then use the table it contains, these files will be submitted right after this patchkit. [acme@jouet linux]$ perf list page_walker List of pre-defined events (to be used in -e): page_walker_loads.dtlb_l1 [Number of DTLB page walker hits in the L1+FB] page_walker_loads.dtlb_l2 [Number of DTLB page walker hits in the L2] page_walker_loads.dtlb_l3 [Number of DTLB page walker hits in the L3 + XSNP] page_walker_loads.dtlb_memory [Number of DTLB page walker hits in Memory] page_walker_loads.itlb_l1 [Number of ITLB page walker hits in the L1+FB] page_walker_loads.itlb_l2 [Number of ITLB page walker hits in the L2] page_walker_loads.itlb_l3 [Number of ITLB page walker hits in the L3 + XSNP] [acme@jouet linux]$ Signed-off-by: Andi Kleen Signed-off-by: Sukadev Bhattiprolu Acked-by: Ingo Molnar Acked-by: Jiri Olsa Tested-by: Arnaldo Carvalho de Melo Cc: Madhavan Srinivasan Cc: Peter Zijlstra Cc: linuxppc-dev@lists.ozlabs.org Link: http://lkml.kernel.org/r/1473978296-20712-7-git-send-email-sukadev@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/pmu.c | 83 +++++++++++++++++++++++++++++++++++++++++---------- tools/perf/util/pmu.h | 1 + 2 files changed, 68 insertions(+), 16 deletions(-) (limited to 'tools') diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c index 10668b7f5272..9857fb14ea86 100644 --- a/tools/perf/util/pmu.c +++ b/tools/perf/util/pmu.c @@ -222,7 +222,7 @@ static int perf_pmu__parse_snapshot(struct perf_pmu_alias *alias, } static int __perf_pmu__new_alias(struct list_head *list, char *dir, char *name, - char *desc __maybe_unused, char *val) + char *desc, char *val) { struct perf_pmu_alias *alias; int ret; @@ -255,6 +255,8 @@ static int __perf_pmu__new_alias(struct list_head *list, char *dir, char *name, perf_pmu__parse_snapshot(alias, dir, name); } + alias->desc = desc ? strdup(desc) : NULL; + list_add_tail(&alias->list, list); return 0; @@ -1043,11 +1045,42 @@ static char *format_alias_or(char *buf, int len, struct perf_pmu *pmu, return buf; } -static int cmp_string(const void *a, const void *b) +struct pair { + char *name; + char *desc; +}; + +static int cmp_pair(const void *a, const void *b) +{ + const struct pair *as = a; + const struct pair *bs = b; + + /* Put extra events last */ + if (!!as->desc != !!bs->desc) + return !!as->desc - !!bs->desc; + return strcmp(as->name, bs->name); +} + +static void wordwrap(char *s, int start, int max, int corr) { - const char * const *as = a; - const char * const *bs = b; - return strcmp(*as, *bs); + int column = start; + int n; + + while (*s) { + int wlen = strcspn(s, " \t"); + + if (column + wlen >= max && column > start) { + printf("\n%*s", start, ""); + column = start + corr; + } + n = printf("%s%.*s", column > start ? " " : "", wlen, s); + if (n <= 0) + break; + s += wlen; + column += n; + while (isspace(*s)) + s++; + } } void print_pmu_events(const char *event_glob, bool name_only) @@ -1057,7 +1090,9 @@ void print_pmu_events(const char *event_glob, bool name_only) char buf[1024]; int printed = 0; int len, j; - char **aliases; + struct pair *aliases; + int numdesc = 0; + int columns = 78; pmu = NULL; len = 0; @@ -1067,14 +1102,15 @@ void print_pmu_events(const char *event_glob, bool name_only) if (pmu->selectable) len++; } - aliases = zalloc(sizeof(char *) * len); + aliases = zalloc(sizeof(struct pair) * len); if (!aliases) goto out_enomem; pmu = NULL; j = 0; while ((pmu = perf_pmu__scan(pmu)) != NULL) { list_for_each_entry(alias, &pmu->aliases, list) { - char *name = format_alias(buf, sizeof(buf), pmu, alias); + char *name = alias->desc ? alias->name : + format_alias(buf, sizeof(buf), pmu, alias); bool is_cpu = !strcmp(pmu->name, "cpu"); if (event_glob != NULL && @@ -1083,12 +1119,19 @@ void print_pmu_events(const char *event_glob, bool name_only) event_glob)))) continue; - if (is_cpu && !name_only) + if (is_cpu && !name_only && !alias->desc) name = format_alias_or(buf, sizeof(buf), pmu, alias); - aliases[j] = strdup(name); - if (aliases[j] == NULL) + aliases[j].name = name; + if (is_cpu && !name_only && !alias->desc) + aliases[j].name = format_alias_or(buf, + sizeof(buf), + pmu, alias); + aliases[j].name = strdup(aliases[j].name); + if (!aliases[j].name) goto out_enomem; + + aliases[j].desc = alias->desc; j++; } if (pmu->selectable && @@ -1096,25 +1139,33 @@ void print_pmu_events(const char *event_glob, bool name_only) char *s; if (asprintf(&s, "%s//", pmu->name) < 0) goto out_enomem; - aliases[j] = s; + aliases[j].name = s; j++; } } len = j; - qsort(aliases, len, sizeof(char *), cmp_string); + qsort(aliases, len, sizeof(struct pair), cmp_pair); for (j = 0; j < len; j++) { if (name_only) { - printf("%s ", aliases[j]); + printf("%s ", aliases[j].name); continue; } - printf(" %-50s [Kernel PMU event]\n", aliases[j]); + if (aliases[j].desc) { + if (numdesc++ == 0) + printf("\n"); + printf(" %-50s\n", aliases[j].name); + printf("%*s", 8, "["); + wordwrap(aliases[j].desc, 8, columns, 0); + printf("]\n"); + } else + printf(" %-50s [Kernel PMU event]\n", aliases[j].name); printed++; } if (printed && pager_in_use()) printf("\n"); out_free: for (j = 0; j < len; j++) - zfree(&aliases[j]); + zfree(&aliases[j].name); zfree(&aliases); return; diff --git a/tools/perf/util/pmu.h b/tools/perf/util/pmu.h index 743422ad900b..51d8d0d35e63 100644 --- a/tools/perf/util/pmu.h +++ b/tools/perf/util/pmu.h @@ -40,6 +40,7 @@ struct perf_pmu_info { struct perf_pmu_alias { char *name; + char *desc; struct list_head terms; /* HEAD struct parse_events_term -> list */ struct list_head list; /* ELEM */ char unit[UNIT_MAX_LEN+1]; -- cgit v1.2.3 From 61eb2eb434b3430c6ef70536eb3d16b616b5ee52 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 15 Sep 2016 15:24:44 -0700 Subject: perf tools: Query terminal width and use in perf list Automatically adapt the now wider and word wrapped perf list output to wider terminals. This requires querying the terminal before the auto pager takes over, and exporting this information from the pager subsystem. Signed-off-by: Andi Kleen Signed-off-by: Sukadev Bhattiprolu Acked-by: Ingo Molnar Acked-by: Jiri Olsa Acked-by: Namhyung Kim Tested-by: Arnaldo Carvalho de Melo Cc: Madhavan Srinivasan Cc: Peter Zijlstra Cc: linuxppc-dev@lists.ozlabs.org Link: http://lkml.kernel.org/r/1473978296-20712-8-git-send-email-sukadev@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/lib/subcmd/pager.c | 16 ++++++++++++++++ tools/lib/subcmd/pager.h | 1 + tools/perf/util/pmu.c | 3 ++- 3 files changed, 19 insertions(+), 1 deletion(-) (limited to 'tools') diff --git a/tools/lib/subcmd/pager.c b/tools/lib/subcmd/pager.c index d50f3b58606b..6518bea926d6 100644 --- a/tools/lib/subcmd/pager.c +++ b/tools/lib/subcmd/pager.c @@ -3,6 +3,7 @@ #include #include #include +#include #include "pager.h" #include "run-command.h" #include "sigchain.h" @@ -14,6 +15,7 @@ */ static int spawned_pager; +static int pager_columns; void pager_init(const char *pager_env) { @@ -58,9 +60,12 @@ static void wait_for_pager_signal(int signo) void setup_pager(void) { const char *pager = getenv(subcmd_config.pager_env); + struct winsize sz; if (!isatty(1)) return; + if (ioctl(1, TIOCGWINSZ, &sz) == 0) + pager_columns = sz.ws_col; if (!pager) pager = getenv("PAGER"); if (!(pager || access("/usr/bin/pager", X_OK))) @@ -98,3 +103,14 @@ int pager_in_use(void) { return spawned_pager; } + +int pager_get_columns(void) +{ + char *s; + + s = getenv("COLUMNS"); + if (s) + return atoi(s); + + return (pager_columns ? pager_columns : 80) - 2; +} diff --git a/tools/lib/subcmd/pager.h b/tools/lib/subcmd/pager.h index 8b83714ecf73..623f5542d05d 100644 --- a/tools/lib/subcmd/pager.h +++ b/tools/lib/subcmd/pager.h @@ -5,5 +5,6 @@ extern void pager_init(const char *pager_env); extern void setup_pager(void); extern int pager_in_use(void); +extern int pager_get_columns(void); #endif /* __SUBCMD_PAGER_H */ diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c index 9857fb14ea86..7b46e772f5f9 100644 --- a/tools/perf/util/pmu.c +++ b/tools/perf/util/pmu.c @@ -14,6 +14,7 @@ #include "cpumap.h" #include "header.h" #include "pmu-events/pmu-events.h" +#include "cache.h" struct perf_pmu_format { char *name; @@ -1092,7 +1093,7 @@ void print_pmu_events(const char *event_glob, bool name_only) int len, j; struct pair *aliases; int numdesc = 0; - int columns = 78; + int columns = pager_get_columns(); pmu = NULL; len = 0; -- cgit v1.2.3 From 1c5f01fe8660fc48625a94a5ed10e4dbaea95d5f Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 15 Sep 2016 15:24:45 -0700 Subject: perf list: Add a --no-desc flag Add a --no-desc flag to 'perf list' to not print the event descriptions that were earlier added for JSON events. This may be useful to get a less crowded listing. It's still default to print descriptions as that is the more useful default for most users. Signed-off-by: Andi Kleen Signed-off-by: Sukadev Bhattiprolu Acked-by: Ingo Molnar Acked-by: Jiri Olsa Tested-by: Arnaldo Carvalho de Melo Cc: linuxppc-dev@lists.ozlabs.org Cc: Madhavan Srinivasan Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/1473978296-20712-9-git-send-email-sukadev@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/Documentation/perf-list.txt | 8 +++++++- tools/perf/builtin-list.c | 14 +++++++++----- tools/perf/util/parse-events.c | 4 ++-- tools/perf/util/parse-events.h | 2 +- tools/perf/util/pmu.c | 4 ++-- tools/perf/util/pmu.h | 2 +- 6 files changed, 22 insertions(+), 12 deletions(-) (limited to 'tools') diff --git a/tools/perf/Documentation/perf-list.txt b/tools/perf/Documentation/perf-list.txt index a126e97a8114..72209bc0f523 100644 --- a/tools/perf/Documentation/perf-list.txt +++ b/tools/perf/Documentation/perf-list.txt @@ -8,13 +8,19 @@ perf-list - List all symbolic event types SYNOPSIS -------- [verse] -'perf list' [hw|sw|cache|tracepoint|pmu|event_glob] +'perf list' [--no-desc] [hw|sw|cache|tracepoint|pmu|event_glob] DESCRIPTION ----------- This command displays the symbolic event types which can be selected in the various perf commands with the -e option. +OPTIONS +------- +--no-desc:: +Don't print descriptions. + + [[EVENT_MODIFIERS]] EVENT MODIFIERS --------------- diff --git a/tools/perf/builtin-list.c b/tools/perf/builtin-list.c index 88ee419e5189..b14cb162f841 100644 --- a/tools/perf/builtin-list.c +++ b/tools/perf/builtin-list.c @@ -16,16 +16,20 @@ #include "util/pmu.h" #include +static bool desc_flag = true; + int cmd_list(int argc, const char **argv, const char *prefix __maybe_unused) { int i; bool raw_dump = false; struct option list_options[] = { OPT_BOOLEAN(0, "raw-dump", &raw_dump, "Dump raw events"), + OPT_BOOLEAN('d', "desc", &desc_flag, + "Print extra event descriptions. --no-desc to not print."), OPT_END() }; const char * const list_usage[] = { - "perf list [hw|sw|cache|tracepoint|pmu|sdt|event_glob]", + "perf list [--no-desc] [hw|sw|cache|tracepoint|pmu|sdt|event_glob]", NULL }; @@ -40,7 +44,7 @@ int cmd_list(int argc, const char **argv, const char *prefix __maybe_unused) printf("\nList of pre-defined events (to be used in -e):\n\n"); if (argc == 0) { - print_events(NULL, raw_dump); + print_events(NULL, raw_dump, !desc_flag); return 0; } @@ -61,14 +65,14 @@ int cmd_list(int argc, const char **argv, const char *prefix __maybe_unused) strcmp(argv[i], "hwcache") == 0) print_hwcache_events(NULL, raw_dump); else if (strcmp(argv[i], "pmu") == 0) - print_pmu_events(NULL, raw_dump); + print_pmu_events(NULL, raw_dump, !desc_flag); else if (strcmp(argv[i], "sdt") == 0) print_sdt_events(NULL, NULL, raw_dump); else if ((sep = strchr(argv[i], ':')) != NULL) { int sep_idx; if (sep == NULL) { - print_events(argv[i], raw_dump); + print_events(argv[i], raw_dump, !desc_flag); continue; } sep_idx = sep - argv[i]; @@ -90,7 +94,7 @@ int cmd_list(int argc, const char **argv, const char *prefix __maybe_unused) print_symbol_events(s, PERF_TYPE_SOFTWARE, event_symbols_sw, PERF_COUNT_SW_MAX, raw_dump); print_hwcache_events(s, raw_dump); - print_pmu_events(s, raw_dump); + print_pmu_events(s, raw_dump, !desc_flag); print_tracepoint_events(NULL, s, raw_dump); print_sdt_events(NULL, s, raw_dump); free(s); diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c index 33546c3ac1fe..3966ad79ee8d 100644 --- a/tools/perf/util/parse-events.c +++ b/tools/perf/util/parse-events.c @@ -2263,7 +2263,7 @@ out_enomem: /* * Print the help text for the event symbols: */ -void print_events(const char *event_glob, bool name_only) +void print_events(const char *event_glob, bool name_only, bool quiet_flag) { print_symbol_events(event_glob, PERF_TYPE_HARDWARE, event_symbols_hw, PERF_COUNT_HW_MAX, name_only); @@ -2273,7 +2273,7 @@ void print_events(const char *event_glob, bool name_only) print_hwcache_events(event_glob, name_only); - print_pmu_events(event_glob, name_only); + print_pmu_events(event_glob, name_only, quiet_flag); if (event_glob != NULL) return; diff --git a/tools/perf/util/parse-events.h b/tools/perf/util/parse-events.h index 8d09a976fca8..3bf376b42bf9 100644 --- a/tools/perf/util/parse-events.h +++ b/tools/perf/util/parse-events.h @@ -172,7 +172,7 @@ void parse_events_update_lists(struct list_head *list_event, void parse_events_evlist_error(struct parse_events_evlist *data, int idx, const char *str); -void print_events(const char *event_glob, bool name_only); +void print_events(const char *event_glob, bool name_only, bool quiet); struct event_symbol { const char *symbol; diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c index 7b46e772f5f9..9dc3506d23fc 100644 --- a/tools/perf/util/pmu.c +++ b/tools/perf/util/pmu.c @@ -1084,7 +1084,7 @@ static void wordwrap(char *s, int start, int max, int corr) } } -void print_pmu_events(const char *event_glob, bool name_only) +void print_pmu_events(const char *event_glob, bool name_only, bool quiet_flag) { struct perf_pmu *pmu; struct perf_pmu_alias *alias; @@ -1151,7 +1151,7 @@ void print_pmu_events(const char *event_glob, bool name_only) printf("%s ", aliases[j].name); continue; } - if (aliases[j].desc) { + if (aliases[j].desc && !quiet_flag) { if (numdesc++ == 0) printf("\n"); printf(" %-50s\n", aliases[j].name); diff --git a/tools/perf/util/pmu.h b/tools/perf/util/pmu.h index 51d8d0d35e63..7b47192e03b5 100644 --- a/tools/perf/util/pmu.h +++ b/tools/perf/util/pmu.h @@ -72,7 +72,7 @@ int perf_pmu__format_parse(char *dir, struct list_head *head); struct perf_pmu *perf_pmu__scan(struct perf_pmu *pmu); -void print_pmu_events(const char *event_glob, bool name_only); +void print_pmu_events(const char *event_glob, bool name_only, bool quiet); bool pmu_have_event(const char *pname, const char *name); int perf_pmu__scan_file(struct perf_pmu *pmu, const char *name, const char *fmt, -- cgit v1.2.3 From fc06e2a5aad9fcf7efaabd7550ac31f648d2f2bc Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 15 Sep 2016 15:24:46 -0700 Subject: perf pmu: Add override support for event list CPUID Add a PERF_CPUID variable to override the CPUID of the current CPU (within the current architecture). This is useful for testing, so that all event lists can be tested on a single system. Signed-off-by: Andi Kleen Signed-off-by: Sukadev Bhattiprolu Acked-by: Ingo Molnar Acked-by: Jiri Olsa Cc: Madhavan Srinivasan Cc: Peter Zijlstra Cc: linuxppc-dev@lists.ozlabs.org Link: http://lkml.kernel.org/r/1473978296-20712-10-git-send-email-sukadev@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/pmu.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'tools') diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c index 9dc3506d23fc..79242cf9bb79 100644 --- a/tools/perf/util/pmu.c +++ b/tools/perf/util/pmu.c @@ -501,10 +501,16 @@ static void pmu_add_cpu_aliases(struct list_head *head) struct pmu_event *pe; char *cpuid; - cpuid = get_cpuid_str(); + cpuid = getenv("PERF_CPUID"); + if (cpuid) + cpuid = strdup(cpuid); + if (!cpuid) + cpuid = get_cpuid_str(); if (!cpuid) return; + pr_debug("Using CPUID %s\n", cpuid); + i = 0; while (1) { map = &pmu_events_map[i++]; -- cgit v1.2.3 From 794ba54a8393456d503958d6217874e455b6a771 Mon Sep 17 00:00:00 2001 From: Sukadev Bhattiprolu Date: Thu, 15 Sep 2016 15:24:47 -0700 Subject: perf jevents: Add support for long descriptions Implement support in jevents to parse long descriptions for events that may have them in the JSON files. A follow on patch will make this long description available to user through the 'perf list' command. Signed-off-by: Andi Kleen Signed-off-by: Sukadev Bhattiprolu Acked-by: Ingo Molnar Acked-by: Jiri Olsa Cc: Madhavan Srinivasan Cc: Peter Zijlstra Cc: linuxppc-dev@lists.ozlabs.org Link: http://lkml.kernel.org/r/1473978296-20712-11-git-send-email-sukadev@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/pmu-events/jevents.c | 32 ++++++++++++++++++++++++-------- tools/perf/pmu-events/jevents.h | 3 ++- tools/perf/pmu-events/pmu-events.h | 1 + 3 files changed, 27 insertions(+), 9 deletions(-) (limited to 'tools') diff --git a/tools/perf/pmu-events/jevents.c b/tools/perf/pmu-events/jevents.c index c9bf9a7dc7b2..13f4284721d5 100644 --- a/tools/perf/pmu-events/jevents.c +++ b/tools/perf/pmu-events/jevents.c @@ -269,7 +269,7 @@ static void print_events_table_prefix(FILE *fp, const char *tblname) } static int print_events_table_entry(void *data, char *name, char *event, - char *desc) + char *desc, char *long_desc) { struct perf_entry_data *pd = data; FILE *outfp = pd->outfp; @@ -285,6 +285,8 @@ static int print_events_table_entry(void *data, char *name, char *event, fprintf(outfp, "\t.event = \"%s\",\n", event); fprintf(outfp, "\t.desc = \"%s\",\n", desc); fprintf(outfp, "\t.topic = \"%s\",\n", topic); + if (long_desc && long_desc[0]) + fprintf(outfp, "\t.long_desc = \"%s\",\n", long_desc); fprintf(outfp, "},\n"); @@ -306,7 +308,8 @@ static void print_events_table_suffix(FILE *outfp) /* Call func with each event in the json file */ int json_events(const char *fn, - int (*func)(void *data, char *name, char *event, char *desc), + int (*func)(void *data, char *name, char *event, char *desc, + char *long_desc), void *data) { int err = -EIO; @@ -325,6 +328,8 @@ int json_events(const char *fn, tok = tokens + 1; for (i = 0; i < tokens->size; i++) { char *event = NULL, *desc = NULL, *name = NULL; + char *long_desc = NULL; + char *extra_desc = NULL; struct msrmap *msr = NULL; jsmntok_t *msrval = NULL; jsmntok_t *precise = NULL; @@ -350,6 +355,10 @@ int json_events(const char *fn, } else if (json_streq(map, field, "BriefDescription")) { addfield(map, &desc, "", "", val); fixdesc(desc); + } else if (json_streq(map, field, + "PublicDescription")) { + addfield(map, &long_desc, "", "", val); + fixdesc(long_desc); } else if (json_streq(map, field, "PEBS") && nz) { precise = val; } else if (json_streq(map, field, "MSRIndex") && nz) { @@ -358,10 +367,10 @@ int json_events(const char *fn, msrval = val; } else if (json_streq(map, field, "Errata") && !json_streq(map, val, "null")) { - addfield(map, &desc, ". ", + addfield(map, &extra_desc, ". ", " Spec update: ", val); } else if (json_streq(map, field, "Data_LA") && nz) { - addfield(map, &desc, ". ", + addfield(map, &extra_desc, ". ", " Supports address when precise", NULL); } @@ -369,19 +378,26 @@ int json_events(const char *fn, } if (precise && desc && !strstr(desc, "(Precise Event)")) { if (json_streq(map, precise, "2")) - addfield(map, &desc, " ", "(Must be precise)", - NULL); + addfield(map, &extra_desc, " ", + "(Must be precise)", NULL); else - addfield(map, &desc, " ", + addfield(map, &extra_desc, " ", "(Precise event)", NULL); } + if (desc && extra_desc) + addfield(map, &desc, " ", extra_desc, NULL); + if (long_desc && extra_desc) + addfield(map, &long_desc, " ", extra_desc, NULL); if (msr != NULL) addfield(map, &event, ",", msr->pname, msrval); fixname(name); - err = func(data, name, event, desc); + + err = func(data, name, event, desc, long_desc); free(event); free(desc); free(name); + free(long_desc); + free(extra_desc); if (err) break; tok += j; diff --git a/tools/perf/pmu-events/jevents.h b/tools/perf/pmu-events/jevents.h index 996601f828b6..b0eb2744b498 100644 --- a/tools/perf/pmu-events/jevents.h +++ b/tools/perf/pmu-events/jevents.h @@ -2,7 +2,8 @@ #define JEVENTS_H 1 int json_events(const char *fn, - int (*func)(void *data, char *name, char *event, char *desc), + int (*func)(void *data, char *name, char *event, char *desc, + char *long_desc), void *data); char *get_cpu_str(void); diff --git a/tools/perf/pmu-events/pmu-events.h b/tools/perf/pmu-events/pmu-events.h index 70d54794e3cb..2eaef595d8a0 100644 --- a/tools/perf/pmu-events/pmu-events.h +++ b/tools/perf/pmu-events/pmu-events.h @@ -9,6 +9,7 @@ struct pmu_event { const char *event; const char *desc; const char *topic; + const char *long_desc; }; /* -- cgit v1.2.3 From c8d6828a6510c9363180ebf662b51d032e0eb85a Mon Sep 17 00:00:00 2001 From: Sukadev Bhattiprolu Date: Thu, 15 Sep 2016 15:24:48 -0700 Subject: perf list: Support long jevents descriptions Previously we were dropping the useful longer descriptions that some events have in the event list completely. This patch makes them appear with perf list. Old perf list: baclears: baclears.all [Counts the number of baclears] vs new: perf list -v: ... baclears: baclears.all [The BACLEARS event counts the number of times the front end is resteered, mainly when the Branch Prediction Unit cannot provide a correct prediction and this is corrected by the Branch Address Calculator at the front end. The BACLEARS.ANY event counts the number of baclears for any type of branch] Signed-off-by: Andi Kleen Signed-off-by: Sukadev Bhattiprolu Acked-by: Ingo Molnar Acked-by: Jiri Olsa Cc: Madhavan Srinivasan Cc: Peter Zijlstra Cc: linuxppc-dev@lists.ozlabs.org Link: http://lkml.kernel.org/r/1473978296-20712-13-git-send-email-sukadev@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/Documentation/perf-list.txt | 6 +++++- tools/perf/builtin-list.c | 16 +++++++++++----- tools/perf/util/parse-events.c | 5 +++-- tools/perf/util/parse-events.h | 3 ++- tools/perf/util/pmu.c | 15 ++++++++++----- tools/perf/util/pmu.h | 4 +++- 6 files changed, 34 insertions(+), 15 deletions(-) (limited to 'tools') diff --git a/tools/perf/Documentation/perf-list.txt b/tools/perf/Documentation/perf-list.txt index 72209bc0f523..41857cce5e86 100644 --- a/tools/perf/Documentation/perf-list.txt +++ b/tools/perf/Documentation/perf-list.txt @@ -8,7 +8,7 @@ perf-list - List all symbolic event types SYNOPSIS -------- [verse] -'perf list' [--no-desc] [hw|sw|cache|tracepoint|pmu|event_glob] +'perf list' [--no-desc] [--long-desc] [hw|sw|cache|tracepoint|pmu|event_glob] DESCRIPTION ----------- @@ -20,6 +20,10 @@ OPTIONS --no-desc:: Don't print descriptions. +-v:: +--long-desc:: +Print longer event descriptions. + [[EVENT_MODIFIERS]] EVENT MODIFIERS diff --git a/tools/perf/builtin-list.c b/tools/perf/builtin-list.c index b14cb162f841..ba9322ff858b 100644 --- a/tools/perf/builtin-list.c +++ b/tools/perf/builtin-list.c @@ -22,14 +22,17 @@ int cmd_list(int argc, const char **argv, const char *prefix __maybe_unused) { int i; bool raw_dump = false; + bool long_desc_flag = false; struct option list_options[] = { OPT_BOOLEAN(0, "raw-dump", &raw_dump, "Dump raw events"), OPT_BOOLEAN('d', "desc", &desc_flag, "Print extra event descriptions. --no-desc to not print."), + OPT_BOOLEAN('v', "long-desc", &long_desc_flag, + "Print longer event descriptions."), OPT_END() }; const char * const list_usage[] = { - "perf list [--no-desc] [hw|sw|cache|tracepoint|pmu|sdt|event_glob]", + "perf list [] [hw|sw|cache|tracepoint|pmu|sdt|event_glob]", NULL }; @@ -44,7 +47,7 @@ int cmd_list(int argc, const char **argv, const char *prefix __maybe_unused) printf("\nList of pre-defined events (to be used in -e):\n\n"); if (argc == 0) { - print_events(NULL, raw_dump, !desc_flag); + print_events(NULL, raw_dump, !desc_flag, long_desc_flag); return 0; } @@ -65,14 +68,16 @@ int cmd_list(int argc, const char **argv, const char *prefix __maybe_unused) strcmp(argv[i], "hwcache") == 0) print_hwcache_events(NULL, raw_dump); else if (strcmp(argv[i], "pmu") == 0) - print_pmu_events(NULL, raw_dump, !desc_flag); + print_pmu_events(NULL, raw_dump, !desc_flag, + long_desc_flag); else if (strcmp(argv[i], "sdt") == 0) print_sdt_events(NULL, NULL, raw_dump); else if ((sep = strchr(argv[i], ':')) != NULL) { int sep_idx; if (sep == NULL) { - print_events(argv[i], raw_dump, !desc_flag); + print_events(argv[i], raw_dump, !desc_flag, + long_desc_flag); continue; } sep_idx = sep - argv[i]; @@ -94,7 +99,8 @@ int cmd_list(int argc, const char **argv, const char *prefix __maybe_unused) print_symbol_events(s, PERF_TYPE_SOFTWARE, event_symbols_sw, PERF_COUNT_SW_MAX, raw_dump); print_hwcache_events(s, raw_dump); - print_pmu_events(s, raw_dump, !desc_flag); + print_pmu_events(s, raw_dump, !desc_flag, + long_desc_flag); print_tracepoint_events(NULL, s, raw_dump); print_sdt_events(NULL, s, raw_dump); free(s); diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c index 3966ad79ee8d..8f88f63bf96c 100644 --- a/tools/perf/util/parse-events.c +++ b/tools/perf/util/parse-events.c @@ -2263,7 +2263,8 @@ out_enomem: /* * Print the help text for the event symbols: */ -void print_events(const char *event_glob, bool name_only, bool quiet_flag) +void print_events(const char *event_glob, bool name_only, bool quiet_flag, + bool long_desc) { print_symbol_events(event_glob, PERF_TYPE_HARDWARE, event_symbols_hw, PERF_COUNT_HW_MAX, name_only); @@ -2273,7 +2274,7 @@ void print_events(const char *event_glob, bool name_only, bool quiet_flag) print_hwcache_events(event_glob, name_only); - print_pmu_events(event_glob, name_only, quiet_flag); + print_pmu_events(event_glob, name_only, quiet_flag, long_desc); if (event_glob != NULL) return; diff --git a/tools/perf/util/parse-events.h b/tools/perf/util/parse-events.h index 3bf376b42bf9..da246a3ddb69 100644 --- a/tools/perf/util/parse-events.h +++ b/tools/perf/util/parse-events.h @@ -172,7 +172,8 @@ void parse_events_update_lists(struct list_head *list_event, void parse_events_evlist_error(struct parse_events_evlist *data, int idx, const char *str); -void print_events(const char *event_glob, bool name_only, bool quiet); +void print_events(const char *event_glob, bool name_only, bool quiet, + bool long_desc); struct event_symbol { const char *symbol; diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c index 79242cf9bb79..8ff382cb161d 100644 --- a/tools/perf/util/pmu.c +++ b/tools/perf/util/pmu.c @@ -223,7 +223,7 @@ static int perf_pmu__parse_snapshot(struct perf_pmu_alias *alias, } static int __perf_pmu__new_alias(struct list_head *list, char *dir, char *name, - char *desc, char *val) + char *desc, char *val, char *long_desc) { struct perf_pmu_alias *alias; int ret; @@ -257,6 +257,8 @@ static int __perf_pmu__new_alias(struct list_head *list, char *dir, char *name, } alias->desc = desc ? strdup(desc) : NULL; + alias->long_desc = long_desc ? strdup(long_desc) : + desc ? strdup(desc) : NULL; list_add_tail(&alias->list, list); @@ -274,7 +276,7 @@ static int perf_pmu__new_alias(struct list_head *list, char *dir, char *name, FI buf[ret] = 0; - return __perf_pmu__new_alias(list, dir, name, NULL, buf); + return __perf_pmu__new_alias(list, dir, name, NULL, buf, NULL); } static inline bool pmu_alias_info_file(char *name) @@ -532,7 +534,8 @@ static void pmu_add_cpu_aliases(struct list_head *head) /* need type casts to override 'const' */ __perf_pmu__new_alias(head, NULL, (char *)pe->name, - (char *)pe->desc, (char *)pe->event); + (char *)pe->desc, (char *)pe->event, + (char *)pe->long_desc); } out: @@ -1090,7 +1093,8 @@ static void wordwrap(char *s, int start, int max, int corr) } } -void print_pmu_events(const char *event_glob, bool name_only, bool quiet_flag) +void print_pmu_events(const char *event_glob, bool name_only, bool quiet_flag, + bool long_desc) { struct perf_pmu *pmu; struct perf_pmu_alias *alias; @@ -1138,7 +1142,8 @@ void print_pmu_events(const char *event_glob, bool name_only, bool quiet_flag) if (!aliases[j].name) goto out_enomem; - aliases[j].desc = alias->desc; + aliases[j].desc = long_desc ? alias->long_desc : + alias->desc; j++; } if (pmu->selectable && diff --git a/tools/perf/util/pmu.h b/tools/perf/util/pmu.h index 7b47192e03b5..2fb8aa0fb47f 100644 --- a/tools/perf/util/pmu.h +++ b/tools/perf/util/pmu.h @@ -41,6 +41,7 @@ struct perf_pmu_info { struct perf_pmu_alias { char *name; char *desc; + char *long_desc; struct list_head terms; /* HEAD struct parse_events_term -> list */ struct list_head list; /* ELEM */ char unit[UNIT_MAX_LEN+1]; @@ -72,7 +73,8 @@ int perf_pmu__format_parse(char *dir, struct list_head *head); struct perf_pmu *perf_pmu__scan(struct perf_pmu *pmu); -void print_pmu_events(const char *event_glob, bool name_only, bool quiet); +void print_pmu_events(const char *event_glob, bool name_only, bool quiet, + bool long_desc); bool pmu_have_event(const char *pname, const char *name); int perf_pmu__scan_file(struct perf_pmu *pmu, const char *name, const char *fmt, -- cgit v1.2.3 From dd5f10368a5f27ec926395056d959513ca4578cc Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 15 Sep 2016 15:24:50 -0700 Subject: perf list jevents: Add support for event list topics Add support to group the output of perf list by the Topic field in the JSON file. Example output: % perf list ... Cache: l1d.replacement [L1D data line replacements] l1d_pend_miss.pending [L1D miss oustandings duration in cycles] l1d_pend_miss.pending_cycles [Cycles with L1D load Misses outstanding] l2_l1d_wb_rqsts.all [Not rejected writebacks from L1D to L2 cache lines in any state] l2_l1d_wb_rqsts.hit_e [Not rejected writebacks from L1D to L2 cache lines in E state] l2_l1d_wb_rqsts.hit_m [Not rejected writebacks from L1D to L2 cache lines in M state] ... Pipeline: arith.fpu_div [Divide operations executed] arith.fpu_div_active [Cycles when divider is busy executing divide operations] baclears.any [Counts the total number when the front end is resteered, mainly when the BPU cannot provide a correct prediction and this is corrected by other branch handling mechanisms at the front end] br_inst_exec.all_branches [Speculative and retired branches] br_inst_exec.all_conditional [Speculative and retired macro-conditional branches] br_inst_exec.all_direct_jmp [Speculative and retired macro-unconditional branches excluding calls and indirects] br_inst_exec.all_direct_near_call [Speculative and retired direct near calls] br_inst_exec.all_indirect_jump_non_call_ret Signed-off-by: Andi Kleen Signed-off-by: Sukadev Bhattiprolu Acked-by: Ingo Molnar Acked-by: Jiri Olsa Tested-by: Arnaldo Carvalho de Melo Cc: Madhavan Srinivasan Cc: Peter Zijlstra Cc: linuxppc-dev@lists.ozlabs.org Link: http://lkml.kernel.org/r/1473978296-20712-14-git-send-email-sukadev@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/pmu.c | 37 +++++++++++++++++++++++++++---------- tools/perf/util/pmu.h | 1 + 2 files changed, 28 insertions(+), 10 deletions(-) (limited to 'tools') diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c index 8ff382cb161d..b1474dcadfa2 100644 --- a/tools/perf/util/pmu.c +++ b/tools/perf/util/pmu.c @@ -223,7 +223,8 @@ static int perf_pmu__parse_snapshot(struct perf_pmu_alias *alias, } static int __perf_pmu__new_alias(struct list_head *list, char *dir, char *name, - char *desc, char *val, char *long_desc) + char *desc, char *val, char *long_desc, + char *topic) { struct perf_pmu_alias *alias; int ret; @@ -259,6 +260,7 @@ static int __perf_pmu__new_alias(struct list_head *list, char *dir, char *name, alias->desc = desc ? strdup(desc) : NULL; alias->long_desc = long_desc ? strdup(long_desc) : desc ? strdup(desc) : NULL; + alias->topic = topic ? strdup(topic) : NULL; list_add_tail(&alias->list, list); @@ -276,7 +278,7 @@ static int perf_pmu__new_alias(struct list_head *list, char *dir, char *name, FI buf[ret] = 0; - return __perf_pmu__new_alias(list, dir, name, NULL, buf, NULL); + return __perf_pmu__new_alias(list, dir, name, NULL, buf, NULL, NULL); } static inline bool pmu_alias_info_file(char *name) @@ -535,7 +537,7 @@ static void pmu_add_cpu_aliases(struct list_head *head) /* need type casts to override 'const' */ __perf_pmu__new_alias(head, NULL, (char *)pe->name, (char *)pe->desc, (char *)pe->event, - (char *)pe->long_desc); + (char *)pe->long_desc, (char *)pe->topic); } out: @@ -1055,19 +1057,26 @@ static char *format_alias_or(char *buf, int len, struct perf_pmu *pmu, return buf; } -struct pair { +struct sevent { char *name; char *desc; + char *topic; }; -static int cmp_pair(const void *a, const void *b) +static int cmp_sevent(const void *a, const void *b) { - const struct pair *as = a; - const struct pair *bs = b; + const struct sevent *as = a; + const struct sevent *bs = b; /* Put extra events last */ if (!!as->desc != !!bs->desc) return !!as->desc - !!bs->desc; + if (as->topic && bs->topic) { + int n = strcmp(as->topic, bs->topic); + + if (n) + return n; + } return strcmp(as->name, bs->name); } @@ -1101,9 +1110,10 @@ void print_pmu_events(const char *event_glob, bool name_only, bool quiet_flag, char buf[1024]; int printed = 0; int len, j; - struct pair *aliases; + struct sevent *aliases; int numdesc = 0; int columns = pager_get_columns(); + char *topic = NULL; pmu = NULL; len = 0; @@ -1113,7 +1123,7 @@ void print_pmu_events(const char *event_glob, bool name_only, bool quiet_flag, if (pmu->selectable) len++; } - aliases = zalloc(sizeof(struct pair) * len); + aliases = zalloc(sizeof(struct sevent) * len); if (!aliases) goto out_enomem; pmu = NULL; @@ -1144,6 +1154,7 @@ void print_pmu_events(const char *event_glob, bool name_only, bool quiet_flag, aliases[j].desc = long_desc ? alias->long_desc : alias->desc; + aliases[j].topic = alias->topic; j++; } if (pmu->selectable && @@ -1156,7 +1167,7 @@ void print_pmu_events(const char *event_glob, bool name_only, bool quiet_flag, } } len = j; - qsort(aliases, len, sizeof(struct pair), cmp_pair); + qsort(aliases, len, sizeof(struct sevent), cmp_sevent); for (j = 0; j < len; j++) { if (name_only) { printf("%s ", aliases[j].name); @@ -1165,6 +1176,12 @@ void print_pmu_events(const char *event_glob, bool name_only, bool quiet_flag, if (aliases[j].desc && !quiet_flag) { if (numdesc++ == 0) printf("\n"); + if (aliases[j].topic && (!topic || + strcmp(topic, aliases[j].topic))) { + printf("%s%s:\n", topic ? "\n" : "", + aliases[j].topic); + topic = aliases[j].topic; + } printf(" %-50s\n", aliases[j].name); printf("%*s", 8, "["); wordwrap(aliases[j].desc, 8, columns, 0); diff --git a/tools/perf/util/pmu.h b/tools/perf/util/pmu.h index 2fb8aa0fb47f..25712034c815 100644 --- a/tools/perf/util/pmu.h +++ b/tools/perf/util/pmu.h @@ -42,6 +42,7 @@ struct perf_pmu_alias { char *name; char *desc; char *long_desc; + char *topic; struct list_head terms; /* HEAD struct parse_events_term -> list */ struct list_head list; /* ELEM */ char unit[UNIT_MAX_LEN+1]; -- cgit v1.2.3 From 06d839b4f712065d76cba92e7e10e9d771771c5a Mon Sep 17 00:00:00 2001 From: Sukadev Bhattiprolu Date: Thu, 15 Sep 2016 15:24:52 -0700 Subject: perf tools: Add README for info on parsing JSON/map files Signed-off-by: Sukadev Bhattiprolu Acked-by: Ingo Molnar Acked-by: Jiri Olsa Cc: Madhavan Srinivasan Cc: Peter Zijlstra Cc: linuxppc-dev@lists.ozlabs.org Link: http://lkml.kernel.org/r/1473978296-20712-16-git-send-email-sukadev@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/pmu-events/README | 147 +++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 147 insertions(+) create mode 100644 tools/perf/pmu-events/README (limited to 'tools') diff --git a/tools/perf/pmu-events/README b/tools/perf/pmu-events/README new file mode 100644 index 000000000000..1408ade0d773 --- /dev/null +++ b/tools/perf/pmu-events/README @@ -0,0 +1,147 @@ + +The contents of this directory allow users to specify PMU events in their +CPUs by their symbolic names rather than raw event codes (see example below). + +The main program in this directory, is the 'jevents', which is built and +executed _BEFORE_ the perf binary itself is built. + +The 'jevents' program tries to locate and process JSON files in the directory +tree tools/perf/pmu-events/arch/foo. + + - Regular files with '.json' extension in the name are assumed to be + JSON files, each of which describes a set of PMU events. + + - Regular files with basename starting with 'mapfile.csv' are assumed + to be a CSV file that maps a specific CPU to its set of PMU events. + (see below for mapfile format) + + - Directories are traversed, but all other files are ignored. + +The PMU events supported by a CPU model are expected to grouped into topics +such as Pipelining, Cache, Memory, Floating-point etc. All events for a topic +should be placed in a separate JSON file - where the file name identifies +the topic. Eg: "Floating-point.json". + +All the topic JSON files for a CPU model/family should be in a separate +sub directory. Thus for the Silvermont X86 CPU: + + $ ls tools/perf/pmu-events/arch/x86/Silvermont_core + Cache.json Memory.json Virtual-Memory.json + Frontend.json Pipeline.json + +Using the JSON files and the mapfile, 'jevents' generates the C source file, +'pmu-events.c', which encodes the two sets of tables: + + - Set of 'PMU events tables' for all known CPUs in the architecture, + (one table like the following, per JSON file; table name 'pme_power8' + is derived from JSON file name, 'power8.json'). + + struct pmu_event pme_power8[] = { + + ... + + { + .name = "pm_1plus_ppc_cmpl", + .event = "event=0x100f2", + .desc = "1 or more ppc insts finished,", + }, + + ... + } + + - A 'mapping table' that maps each CPU of the architecture, to its + 'PMU events table' + + struct pmu_events_map pmu_events_map[] = { + { + .cpuid = "004b0000", + .version = "1", + .type = "core", + .table = pme_power8 + }, + ... + + }; + +After the 'pmu-events.c' is generated, it is compiled and the resulting +'pmu-events.o' is added to 'libperf.a' which is then used to build perf. + +NOTES: + 1. Several CPUs can support same set of events and hence use a common + JSON file. Hence several entries in the pmu_events_map[] could map + to a single 'PMU events table'. + + 2. The 'pmu-events.h' has an extern declaration for the mapping table + and the generated 'pmu-events.c' defines this table. + + 3. _All_ known CPU tables for architecture are included in the perf + binary. + +At run time, perf determines the actual CPU it is running on, finds the +matching events table and builds aliases for those events. This allows +users to specify events by their name: + + $ perf stat -e pm_1plus_ppc_cmpl sleep 1 + +where 'pm_1plus_ppc_cmpl' is a Power8 PMU event. + +In case of errors when processing files in the tools/perf/pmu-events/arch +directory, 'jevents' tries to create an empty mapping file to allow the perf +build to succeed even if the PMU event aliases cannot be used. + +However some errors in processing may cause the perf build to fail. + +Mapfile format +=============== + +The mapfile enables multiple CPU models to share a single set of PMU events. +It is required even if such mapping is 1:1. + +The mapfile.csv format is expected to be: + + Header line + CPUID,Version,Dir/path/name,Type + +where: + + Comma: + is the required field delimiter (i.e other fields cannot + have commas within them). + + Comments: + Lines in which the first character is either '\n' or '#' + are ignored. + + Header line + The header line is the first line in the file, which is + always _IGNORED_. It can empty. + + CPUID: + CPUID is an arch-specific char string, that can be used + to identify CPU (and associate it with a set of PMU events + it supports). Multiple CPUIDS can point to the same + File/path/name.json. + + Example: + CPUID == 'GenuineIntel-6-2E' (on x86). + CPUID == '004b0100' (PVR value in Powerpc) + Version: + is the Version of the mapfile. + + Dir/path/name: + is the pathname to the directory containing the CPU's JSON + files, relative to the directory containing the mapfile.csv + + Type: + indicates whether the events or "core" or "uncore" events. + + + Eg: + + $ grep Silvermont tools/perf/pmu-events/arch/x86/mapfile.csv + GenuineIntel-6-37,V13,Silvermont_core,core + GenuineIntel-6-4D,V13,Silvermont_core,core + GenuineIntel-6-4C,V13,Silvermont_core,core + + i.e the three CPU models use the JSON files (i.e PMU events) listed + in the directory 'tools/perf/pmu-events/arch/x86/Silvermont_core'. -- cgit v1.2.3 From 06835545b144cab6e7748748576145ea15f68dff Mon Sep 17 00:00:00 2001 From: Sukadev Bhattiprolu Date: Thu, 15 Sep 2016 15:24:56 -0700 Subject: perf tools: Allow period= in perf stat CPU event descriptions. This avoids the JSON PMU events parser having to know whether its aliases are for perf stat or perf record. Signed-off-by: Andi Kleen Signed-off-by: Sukadev Bhattiprolu Acked-by: Ingo Molnar Cc: Madhavan Srinivasan Cc: Peter Zijlstra Cc: linuxppc-dev@lists.ozlabs.org Link: http://lkml.kernel.org/r/1473978296-20712-20-git-send-email-sukadev@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/parse-events.c | 1 + 1 file changed, 1 insertion(+) (limited to 'tools') diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c index 8f88f63bf96c..d445b1144c87 100644 --- a/tools/perf/util/parse-events.c +++ b/tools/perf/util/parse-events.c @@ -924,6 +924,7 @@ config_term_avail(int term_type, struct parse_events_error *err) case PARSE_EVENTS__TERM_TYPE_CONFIG1: case PARSE_EVENTS__TERM_TYPE_CONFIG2: case PARSE_EVENTS__TERM_TYPE_NAME: + case PARSE_EVENTS__TERM_TYPE_SAMPLE_PERIOD: return true; default: if (!err) -- cgit v1.2.3 From e312bcf13053970c0f07ec2c02d7d9be1a036ce0 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 15 Sep 2016 15:24:53 -0700 Subject: perf tools: Make alias matching case-insensitive Make alias matching the events parser case-insensitive. This is useful with the JSON events. perf uses lower case events, but the CPU manuals generally use upper case event names. The JSON files use lower case by default too. But if we search case insensitively then users can cut-n-paste the upper case event names. So the following works: % perf stat -e BR_INST_EXEC.TAKEN_INDIRECT_NEAR_CALL true Performance counter stats for 'true': 305 BR_INST_EXEC.TAKEN_INDIRECT_NEAR_CALL 0.000492799 seconds time elapsed Signed-off-by: Andi Kleen Acked-by: Ingo Molnar Tested-by: Arnaldo Carvalho de Melo Cc: Madhavan Srinivasan Cc: Peter Zijlstra Cc: linuxppc-dev@lists.ozlabs.org Link: http://lkml.kernel.org/r/1473978296-20712-17-git-send-email-sukadev@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/parse-events.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'tools') diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c index d445b1144c87..4e778eae1510 100644 --- a/tools/perf/util/parse-events.c +++ b/tools/perf/util/parse-events.c @@ -1459,7 +1459,7 @@ comp_pmu(const void *p1, const void *p2) struct perf_pmu_event_symbol *pmu1 = (struct perf_pmu_event_symbol *) p1; struct perf_pmu_event_symbol *pmu2 = (struct perf_pmu_event_symbol *) p2; - return strcmp(pmu1->symbol, pmu2->symbol); + return strcasecmp(pmu1->symbol, pmu2->symbol); } static void perf_pmu__parse_cleanup(void) -- cgit v1.2.3 From 0b1db474314c883d0bfc6720278667a6155e995a Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 15 Sep 2016 15:24:54 -0700 Subject: perf pmu-events: Fix fixed counters on Intel The JSON event lists use a different encoding for fixed counters than perf for instructions and cycles (ref-cycles is ok) This lead to some common events like inst_retired.any or cpu_clk_unhalted.thread not counting, when specified with their JSON name. Special case these events in the jevents conversion process. I prefer to not touch the JSON files for this, as it's intended that standard JSON files can be just dropped into the perf build without changes. Signed-off-by: Andi Kleen Signed-off-by: Sukadev Bhattiprolu [Fix minor compile error] Acked-by: Ingo Molnar Cc: Madhavan Srinivasan Cc: Peter Zijlstra Cc: linuxppc-dev@lists.ozlabs.org Link: http://lkml.kernel.org/r/1473978296-20712-18-git-send-email-sukadev@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/pmu-events/jevents.c | 25 ++++++++++++++++++++++++- 1 file changed, 24 insertions(+), 1 deletion(-) (limited to 'tools') diff --git a/tools/perf/pmu-events/jevents.c b/tools/perf/pmu-events/jevents.c index 13f4284721d5..04e106e799ac 100644 --- a/tools/perf/pmu-events/jevents.c +++ b/tools/perf/pmu-events/jevents.c @@ -306,6 +306,29 @@ static void print_events_table_suffix(FILE *outfp) close_table = 0; } +static struct fixed { + const char *name; + const char *event; +} fixed[] = { + { "inst_retired.any", "event=0xc0" }, + { "cpu_clk_unhalted.thread", "event=0x3c" }, + { "cpu_clk_unhalted.thread_any", "event=0x3c,any=1" }, + { NULL, NULL}, +}; + +/* + * Handle different fixed counter encodings between JSON and perf. + */ +static char *real_event(const char *name, char *event) +{ + int i; + + for (i = 0; fixed[i].name; i++) + if (!strcasecmp(name, fixed[i].name)) + return (char *)fixed[i].event; + return event; +} + /* Call func with each event in the json file */ int json_events(const char *fn, int (*func)(void *data, char *name, char *event, char *desc, @@ -392,7 +415,7 @@ int json_events(const char *fn, addfield(map, &event, ",", msr->pname, msrval); fixname(name); - err = func(data, name, event, desc, long_desc); + err = func(data, name, real_event(name, event), desc, long_desc); free(event); free(desc); free(name); -- cgit v1.2.3 From b42c7369e3f451e22c2b0be5d193955498d37546 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 15 Sep 2016 15:24:55 -0700 Subject: perf pmu-events: Add Skylake frontend MSR support Add support for the "frontend" extra MSR on Skylake in the JSON conversion. Signed-off-by: Andi Kleen Acked-by: Ingo Molnar Cc: Madhavan Srinivasan Cc: Peter Zijlstra Cc: linuxppc-dev@lists.ozlabs.org Link: http://lkml.kernel.org/r/1473978296-20712-19-git-send-email-sukadev@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/pmu-events/jevents.c | 1 + 1 file changed, 1 insertion(+) (limited to 'tools') diff --git a/tools/perf/pmu-events/jevents.c b/tools/perf/pmu-events/jevents.c index 04e106e799ac..79c2133bc534 100644 --- a/tools/perf/pmu-events/jevents.c +++ b/tools/perf/pmu-events/jevents.c @@ -127,6 +127,7 @@ static struct msrmap { { "0x3F6", "ldlat=" }, { "0x1A6", "offcore_rsp=" }, { "0x1A7", "offcore_rsp=" }, + { "0x3F7", "frontend=" }, { NULL, NULL } }; -- cgit v1.2.3 From a1d9972e59e15538458c030affbfc298c32a47f8 Mon Sep 17 00:00:00 2001 From: Ravi Bangoria Date: Tue, 4 Oct 2016 15:57:45 +0530 Subject: perf uretprobe ppc64le: Fix probe location Perf uretprobe probes on GEP(Global Entry Point) which fails to record all function calls via LEP(Local Entry Point). Fix that by probing on LEP. Objdump: 00000000100005f0 : 100005f0: 02 10 40 3c lis r2,4098 100005f4: 00 7f 42 38 addi r2,r2,32512 100005f8: a6 02 08 7c mflr r0 100005fc: 10 00 01 f8 std r0,16(r1) 10000600: f8 ff e1 fb std r31,-8(r1) Before applying patch: $ cat /sys/kernel/debug/tracing/uprobe_events r:probe_uprobe_test/doit /home/ravi/uprobe_test:0x00000000000005f0 After applying patch: $ cat /sys/kernel/debug/tracing/uprobe_events r:probe_uprobe_test/doit /home/ravi/uprobe_test:0x00000000000005f8 This is not the case with kretprobes because the kernel itself finds LEP and probes on it. Signed-off-by: Ravi Bangoria Cc: Alexander Shishkin Cc: Balbir Singh Cc: Masami Hiramatsu Cc: Naveen N. Rao Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/1475576865-6562-1-git-send-email-ravi.bangoria@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/arch/powerpc/util/sym-handling.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'tools') diff --git a/tools/perf/arch/powerpc/util/sym-handling.c b/tools/perf/arch/powerpc/util/sym-handling.c index ed9d5d15d5b6..1030a6e504bb 100644 --- a/tools/perf/arch/powerpc/util/sym-handling.c +++ b/tools/perf/arch/powerpc/util/sym-handling.c @@ -82,7 +82,8 @@ void arch__fix_tev_from_maps(struct perf_probe_event *pev, * * In addition, we shouldn't specify an offset for kretprobes. */ - if (pev->point.offset || pev->point.retprobe || !map || !sym) + if (pev->point.offset || (!pev->uprobes && pev->point.retprobe) || + !map || !sym) return; lep_offset = PPC64_LOCAL_ENTRY_OFFSET(sym->arch_sym); -- cgit v1.2.3 From 51ee6481fa8e879cc942bcc1b0af713e158b7a98 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Wed, 28 Sep 2016 14:41:35 +0300 Subject: perf intel-pt: Fix estimated timestamps for cycle-accurate mode In cycle-accurate mode, timestamps can be calculated from CYC packets. The decoder also estimates timestamps based on the number of instructions since the last timestamp. For that to work in cycle-accurate mode, the instruction count needs to be reset to zero when a timestamp is calculated from a CYC packet, but that wasn't happening, so fix it. Signed-off-by: Adrian Hunter Cc: Jiri Olsa Cc: stable@vger.kernel.org # v4.3+ Link: http://lkml.kernel.org/r/1475062896-22274-1-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/intel-pt-decoder/intel-pt-decoder.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'tools') diff --git a/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c b/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c index 7591a0c37473..3d1d446f037f 100644 --- a/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c +++ b/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c @@ -1353,6 +1353,8 @@ static void intel_pt_calc_cyc_timestamp(struct intel_pt_decoder *decoder) timestamp, decoder->timestamp); else decoder->timestamp = timestamp; + + decoder->timestamp_insn_cnt = 0; } /* Walk PSB+ packets when already in sync. */ -- cgit v1.2.3 From 3bccbe20f6d188ce7b00326e776b745cfd35b10a Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Wed, 28 Sep 2016 14:41:36 +0300 Subject: perf intel-pt: Fix MTC timestamp calculation for large MTC periods The MTC packet provides a 8-bit slice of CTC which is related to TSC by the TMA packet, however the TMA packet only provides the lower 16 bits of CTC. If mtc_shift > 8 then some of the MTC bits are not in the CTC provided by the TMA packet. Fix-up the last_mtc calculated from the TMA packet by copying the missing bits from the current MTC assuming the least difference between the two, and that the current MTC comes after last_mtc. Signed-off-by: Adrian Hunter Cc: Jiri Olsa Cc: stable@vger.kernel.org # v4.3+ Link: http://lkml.kernel.org/r/1475062896-22274-2-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- .../perf/util/intel-pt-decoder/intel-pt-decoder.c | 36 ++++++++++++++++++++++ 1 file changed, 36 insertions(+) (limited to 'tools') diff --git a/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c b/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c index 3d1d446f037f..16c06d3ae577 100644 --- a/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c +++ b/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c @@ -90,6 +90,7 @@ struct intel_pt_decoder { bool pge; bool have_tma; bool have_cyc; + bool fixup_last_mtc; uint64_t pos; uint64_t last_ip; uint64_t ip; @@ -586,10 +587,31 @@ struct intel_pt_calc_cyc_to_tsc_info { uint64_t tsc_timestamp; uint64_t timestamp; bool have_tma; + bool fixup_last_mtc; bool from_mtc; double cbr_cyc_to_tsc; }; +/* + * MTC provides a 8-bit slice of CTC but the TMA packet only provides the lower + * 16 bits of CTC. If mtc_shift > 8 then some of the MTC bits are not in the CTC + * provided by the TMA packet. Fix-up the last_mtc calculated from the TMA + * packet by copying the missing bits from the current MTC assuming the least + * difference between the two, and that the current MTC comes after last_mtc. + */ +static void intel_pt_fixup_last_mtc(uint32_t mtc, int mtc_shift, + uint32_t *last_mtc) +{ + uint32_t first_missing_bit = 1U << (16 - mtc_shift); + uint32_t mask = ~(first_missing_bit - 1); + + *last_mtc |= mtc & mask; + if (*last_mtc >= mtc) { + *last_mtc -= first_missing_bit; + *last_mtc &= 0xff; + } +} + static int intel_pt_calc_cyc_cb(struct intel_pt_pkt_info *pkt_info) { struct intel_pt_decoder *decoder = pkt_info->decoder; @@ -619,6 +641,11 @@ static int intel_pt_calc_cyc_cb(struct intel_pt_pkt_info *pkt_info) return 0; mtc = pkt_info->packet.payload; + if (decoder->mtc_shift > 8 && data->fixup_last_mtc) { + data->fixup_last_mtc = false; + intel_pt_fixup_last_mtc(mtc, decoder->mtc_shift, + &data->last_mtc); + } if (mtc > data->last_mtc) mtc_delta = mtc - data->last_mtc; else @@ -687,6 +714,7 @@ static int intel_pt_calc_cyc_cb(struct intel_pt_pkt_info *pkt_info) data->ctc_delta = 0; data->have_tma = true; + data->fixup_last_mtc = true; return 0; @@ -753,6 +781,7 @@ static void intel_pt_calc_cyc_to_tsc(struct intel_pt_decoder *decoder, .tsc_timestamp = decoder->tsc_timestamp, .timestamp = decoder->timestamp, .have_tma = decoder->have_tma, + .fixup_last_mtc = decoder->fixup_last_mtc, .from_mtc = from_mtc, .cbr_cyc_to_tsc = 0, }; @@ -1271,6 +1300,7 @@ static void intel_pt_calc_tma(struct intel_pt_decoder *decoder) } decoder->ctc_delta = 0; decoder->have_tma = true; + decoder->fixup_last_mtc = true; intel_pt_log("CTC timestamp " x64_fmt " last MTC %#x CTC rem %#x\n", decoder->ctc_timestamp, decoder->last_mtc, ctc_rem); } @@ -1285,6 +1315,12 @@ static void intel_pt_calc_mtc_timestamp(struct intel_pt_decoder *decoder) mtc = decoder->packet.payload; + if (decoder->mtc_shift > 8 && decoder->fixup_last_mtc) { + decoder->fixup_last_mtc = false; + intel_pt_fixup_last_mtc(mtc, decoder->mtc_shift, + &decoder->last_mtc); + } + if (mtc > decoder->last_mtc) mtc_delta = mtc - decoder->last_mtc; else -- cgit v1.2.3 From a130347973c408c0e0017a52c74cbc3226c1f0ef Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Sat, 1 Oct 2016 19:17:00 +0900 Subject: tools lib traceevent: Fix kbuffer_read_at_offset() When it's called with an offset less than or equal to the first event, it'll return a garbage value since the data is not initialized. Signed-off-by: Namhyung Kim Acked-by: Steven Rostedt Cc: Jiri Olsa Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20161001101700.29146-1-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/lib/traceevent/kbuffer-parse.c | 1 + 1 file changed, 1 insertion(+) (limited to 'tools') diff --git a/tools/lib/traceevent/kbuffer-parse.c b/tools/lib/traceevent/kbuffer-parse.c index 3bcada3ae05a..65984f1c2974 100644 --- a/tools/lib/traceevent/kbuffer-parse.c +++ b/tools/lib/traceevent/kbuffer-parse.c @@ -622,6 +622,7 @@ void *kbuffer_read_at_offset(struct kbuffer *kbuf, int offset, /* Reset the buffer */ kbuffer_load_subbuffer(kbuf, kbuf->subbuffer); + data = kbuffer_read_event(kbuf, ts); while (kbuf->curr < offset) { data = kbuffer_next_event(kbuf, ts); -- cgit v1.2.3 From 72c6ff2583fba824dc38c0ce87b838631cdb8294 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 5 Oct 2016 12:47:12 -0700 Subject: perf jevents: Fix Intel JSON fixed counter conversions Intel fixed counters are special cases in the JSON conversion process because their decoding differs between perf and the event files. Add some missing entries in the conversion table. Signed-off-by: Andi Kleen Cc: Jiri Olsa Cc: Sukadev Bhattiprolu Link: http://lkml.kernel.org/r/1475696832-9188-4-git-send-email-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/pmu-events/jevents.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'tools') diff --git a/tools/perf/pmu-events/jevents.c b/tools/perf/pmu-events/jevents.c index 79c2133bc534..41611d7f9873 100644 --- a/tools/perf/pmu-events/jevents.c +++ b/tools/perf/pmu-events/jevents.c @@ -312,6 +312,8 @@ static struct fixed { const char *event; } fixed[] = { { "inst_retired.any", "event=0xc0" }, + { "inst_retired.any_p", "event=0xc0" }, + { "cpu_clk_unhalted.ref", "event=0x0,umask=0x03" }, { "cpu_clk_unhalted.thread", "event=0x3c" }, { "cpu_clk_unhalted.thread_any", "event=0x3c,any=1" }, { NULL, NULL}, -- cgit v1.2.3 From d23e354fe58aada6d1cdeeb7e8463b75d44bc687 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Wed, 5 Oct 2016 19:12:46 -0300 Subject: perf bench mem: Sync memcpy assembly sources with the kernel Commit 9a6fb28a355d ("x86/mce: Improve memcpy_mcsafe()") renames memcpy_mcsafe() to memcpy_mcsafe_unrolled(), making tools/arch/x86/lib/memcpy_64.S drift from the its kernel counterpart, triggering this warning in the perf build: Warning: tools/arch/x86/lib/memcpy_64.S differs from kernel Sync that copy to acknowledge that, no changes to 'perf bench' are needed, as this function is not used there. Cc: Adrian Hunter Cc: David Ahern Cc: Jiri Olsa Cc: Namhyung Kim Cc: Thomas Gleixner Cc: Tony Luck Cc: Wang Nan Link: http://lkml.kernel.org/n/tip-xfwc1raw8obyrctxerwt1bbb@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/arch/x86/lib/memcpy_64.S | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'tools') diff --git a/tools/arch/x86/lib/memcpy_64.S b/tools/arch/x86/lib/memcpy_64.S index 2ec0b0abbfaa..49e6ebac7e73 100644 --- a/tools/arch/x86/lib/memcpy_64.S +++ b/tools/arch/x86/lib/memcpy_64.S @@ -181,11 +181,11 @@ ENDPROC(memcpy_orig) #ifndef CONFIG_UML /* - * memcpy_mcsafe - memory copy with machine check exception handling + * memcpy_mcsafe_unrolled - memory copy with machine check exception handling * Note that we only catch machine checks when reading the source addresses. * Writes to target are posted and don't generate machine checks. */ -ENTRY(memcpy_mcsafe) +ENTRY(memcpy_mcsafe_unrolled) cmpl $8, %edx /* Less than 8 bytes? Go to byte copy loop */ jb .L_no_whole_words @@ -273,7 +273,7 @@ ENTRY(memcpy_mcsafe) .L_done_memcpy_trap: xorq %rax, %rax ret -ENDPROC(memcpy_mcsafe) +ENDPROC(memcpy_mcsafe_unrolled) .section .fixup, "ax" /* Return -EFAULT for any failure */ -- cgit v1.2.3 From a5c4c74564f000950902d0f1e64926cce2280af2 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Wed, 5 Oct 2016 19:18:34 -0300 Subject: tools: Synchronize tools/arch/x86/include/asm/cpufeatures.h Due to ffb173e657fa ("x86/mce: Drop X86_FEATURE_MCE_RECOVERY and the related model string test"), no changes needed in any other place as no tool uses X86_FEATURE_MCE_RECOVERY. Silences this detected drift when building tools/perf: Warning: tools/arch/x86/include/asm/cpufeatures.h differs from kernel Cc: Adrian Hunter Cc: David Ahern Cc: Jiri Olsa Cc: Namhyung Kim Cc: Thomas Gleixner Cc: Tony Luck Cc: Wang Nan Link: http://lkml.kernel.org/n/tip-f3sfimg58t3cycbbl8f5cwxf@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/arch/x86/include/asm/cpufeatures.h | 1 - 1 file changed, 1 deletion(-) (limited to 'tools') diff --git a/tools/arch/x86/include/asm/cpufeatures.h b/tools/arch/x86/include/asm/cpufeatures.h index 92a8308b96f6..1188bc849ee3 100644 --- a/tools/arch/x86/include/asm/cpufeatures.h +++ b/tools/arch/x86/include/asm/cpufeatures.h @@ -106,7 +106,6 @@ #define X86_FEATURE_APERFMPERF ( 3*32+28) /* APERFMPERF */ #define X86_FEATURE_EAGER_FPU ( 3*32+29) /* "eagerfpu" Non lazy FPU restore */ #define X86_FEATURE_NONSTOP_TSC_S3 ( 3*32+30) /* TSC doesn't stop in S3 state */ -#define X86_FEATURE_MCE_RECOVERY ( 3*32+31) /* cpu has recoverable machine checks */ /* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */ #define X86_FEATURE_XMM3 ( 4*32+ 0) /* "pni" SSE-3 */ -- cgit v1.2.3 From dcfc3a6740d372ebe203dae2c5852ff45f9ef793 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Wed, 5 Oct 2016 19:31:48 -0300 Subject: tools: Synchronize tools/include/uapi/linux/bpf.h Commit 747ea55e4f78 ("bpf: fix bpf_skb_in_cgroup helper naming") renames BPF_FUNC_skb_in_cgroup to bpf_skb_under_cgroup, triggering this warning while building perf: Warning: tools/include/uapi/linux/bpf.h differs from kernel Update the copy to ack that, no changes needed, as BPF_FUNC_skb_in_cgroup isn't used so far. Cc: Adrian Hunter Cc: Daniel Borkmann Cc: David Ahern Cc: David S. Miller Cc: Jiri Olsa Cc: Namhyung Kim Cc: Wang Nan Link: http://lkml.kernel.org/n/tip-x67d2gq8ct6ko12ex14q8bbx@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/include/uapi/linux/bpf.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'tools') diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h index da218fec6056..9e5fc168c8a3 100644 --- a/tools/include/uapi/linux/bpf.h +++ b/tools/include/uapi/linux/bpf.h @@ -339,7 +339,7 @@ enum bpf_func_id { BPF_FUNC_skb_change_type, /** - * bpf_skb_in_cgroup(skb, map, index) - Check cgroup2 membership of skb + * bpf_skb_under_cgroup(skb, map, index) - Check cgroup2 membership of skb * @skb: pointer to skb * @map: pointer to bpf_map in BPF_MAP_TYPE_CGROUP_ARRAY type * @index: index of the cgroup in the bpf_map @@ -348,7 +348,7 @@ enum bpf_func_id { * == 1 skb succeeded the cgroup2 descendant test * < 0 error */ - BPF_FUNC_skb_in_cgroup, + BPF_FUNC_skb_under_cgroup, /** * bpf_get_hash_recalc(skb) -- cgit v1.2.3 From 8649b6434ebc184e632a510c4121899481409c15 Mon Sep 17 00:00:00 2001 From: Kim SeonYoung Date: Thu, 29 Sep 2016 17:16:42 -0700 Subject: perf report/top: Add a tip about source line numbers with overhead There is a existing tip as below. If you have debuginfo enabled, try: perf report -s sym,srcline However this tip only describe a condition to use --sort sym,scrline options. So there is lack of explanation in the tip. I think that it would be better to add a tip that exactly explains the feature of --sort srcline. Signed-off-by: Seonyoung Kim Cc: Jiri Olsa Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Taeung Song Link: http://lkml.kernel.org/r/1475194602-5596-1-git-send-email-adamas0414@gmail.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/Documentation/tips.txt | 1 + 1 file changed, 1 insertion(+) (limited to 'tools') diff --git a/tools/perf/Documentation/tips.txt b/tools/perf/Documentation/tips.txt index 5950b5a24efd..b0311248407f 100644 --- a/tools/perf/Documentation/tips.txt +++ b/tools/perf/Documentation/tips.txt @@ -28,3 +28,4 @@ To change sampling frequency to 100 Hz: perf record -F 100 See assembly instructions with percentage: perf annotate If you prefer Intel style assembly, try: perf annotate -M intel For hierarchical output, try: perf report --hierarchy +Order by the overhead of source file name and line number: perf report -s srcline -- cgit v1.2.3 From 49343235d08a228c7fe84d0606e69253cc0bc751 Mon Sep 17 00:00:00 2001 From: Donghyun Kim Date: Fri, 30 Sep 2016 07:15:57 +0900 Subject: perf report/top: Add a tip about system-wide collection from all CPUs Signed-off-by: Donghyun Kim Cc: Jiri Olsa Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Taeung Song Link: http://lkml.kernel.org/r/1475187357-21882-1-git-send-email-dongdong9335@gmail.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/Documentation/tips.txt | 1 + 1 file changed, 1 insertion(+) (limited to 'tools') diff --git a/tools/perf/Documentation/tips.txt b/tools/perf/Documentation/tips.txt index b0311248407f..9d3e1ee99e83 100644 --- a/tools/perf/Documentation/tips.txt +++ b/tools/perf/Documentation/tips.txt @@ -29,3 +29,4 @@ See assembly instructions with percentage: perf annotate If you prefer Intel style assembly, try: perf annotate -M intel For hierarchical output, try: perf report --hierarchy Order by the overhead of source file name and line number: perf report -s srcline +System-wide collection from all CPUs: perf record -a -- cgit v1.2.3 From 2ad8327fd0013ee07f056d28faddc2e7f01dd127 Mon Sep 17 00:00:00 2001 From: Nambong Ha Date: Fri, 30 Sep 2016 08:26:02 +0900 Subject: perf top/report: Add tips about a list option Add two tips that describe --list option of config sub-command and explain how to choose particular config file location. Signed-off-by: Nambong Ha Cc: Jiri Olsa Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Taeung Song Link: http://lkml.kernel.org/r/1475191562-3240-1-git-send-email-over3025@gmail.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/Documentation/tips.txt | 2 ++ 1 file changed, 2 insertions(+) (limited to 'tools') diff --git a/tools/perf/Documentation/tips.txt b/tools/perf/Documentation/tips.txt index 9d3e1ee99e83..8a6479c0eac9 100644 --- a/tools/perf/Documentation/tips.txt +++ b/tools/perf/Documentation/tips.txt @@ -30,3 +30,5 @@ If you prefer Intel style assembly, try: perf annotate -M intel For hierarchical output, try: perf report --hierarchy Order by the overhead of source file name and line number: perf report -s srcline System-wide collection from all CPUs: perf record -a +Show current config key-value pairs: perf config --list +Show user configuration overrides: perf config --user --list -- cgit v1.2.3 From f61bdc3304281e9ff9eec4f1ed9fb15bcf4d8a5e Mon Sep 17 00:00:00 2001 From: Wang Nan Date: Mon, 26 Sep 2016 07:26:55 +0000 Subject: tools build: Support compiling C++ source file Add new rule to compile .cpp file to .o use g++. C++ support is required for built-in clang and LLVM support. Linker side support will be introduced by following commits. Signed-off-by: Wang Nan Cc: Alexei Starovoitov Cc: He Kuang Cc: Jiri Olsa Cc: Zefan Li Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1474874832-134786-2-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/build/Build.include | 1 + tools/build/Makefile.build | 7 +++++++ 2 files changed, 8 insertions(+) (limited to 'tools') diff --git a/tools/build/Build.include b/tools/build/Build.include index 02489380d79b..1dcb95e76f70 100644 --- a/tools/build/Build.include +++ b/tools/build/Build.include @@ -90,6 +90,7 @@ if_changed = $(if $(strip $(any-prereq) $(arg-check)), \ # - per object C flags # - BUILD_STR macro to allow '-D"$(variable)"' constructs c_flags = -Wp,-MD,$(depfile),-MT,$@ $(CFLAGS) -D"BUILD_STR(s)=\#s" $(CFLAGS_$(basetarget).o) $(CFLAGS_$(obj)) +cxx_flags = -Wp,-MD,$(depfile),-MT,$@ $(CXXFLAGS) -D"BUILD_STR(s)=\#s" $(CXXFLAGS_$(basetarget).o) $(CXXFLAGS_$(obj)) ### ## HOSTCC C flags diff --git a/tools/build/Makefile.build b/tools/build/Makefile.build index 190519a94ce5..99c0ccd2f176 100644 --- a/tools/build/Makefile.build +++ b/tools/build/Makefile.build @@ -61,6 +61,9 @@ quiet_cmd_cc_o_c = CC $@ quiet_cmd_host_cc_o_c = HOSTCC $@ cmd_host_cc_o_c = $(HOSTCC) $(host_c_flags) -c -o $@ $< +quiet_cmd_cxx_o_c = CXX $@ + cmd_cxx_o_c = $(CXX) $(cxx_flags) -c -o $@ $< + quiet_cmd_cpp_i_c = CPP $@ cmd_cpp_i_c = $(CC) $(c_flags) -E -o $@ $< @@ -88,6 +91,10 @@ $(OUTPUT)%.o: %.c FORCE $(call rule_mkdir) $(call if_changed_dep,$(host)cc_o_c) +$(OUTPUT)%.o: %.cpp FORCE + $(call rule_mkdir) + $(call if_changed_dep,cxx_o_c) + $(OUTPUT)%.o: %.S FORCE $(call rule_mkdir) $(call if_changed_dep,$(host)cc_o_c) -- cgit v1.2.3 From 87095f7ddeff3038a0cf8e6574922f9c11688619 Mon Sep 17 00:00:00 2001 From: Wang Nan Date: Mon, 26 Sep 2016 07:26:56 +0000 Subject: tools build: Add feature detection for g++ Check if g++ is available. The result will be used by builtin clang and LLVM support. Since LLVM requires C++11, this feature detector checks std::move(). Signed-off-by: Wang Nan Cc: Alexei Starovoitov Cc: He Kuang Cc: Jiri Olsa Cc: Zefan Li Cc: pi3orama@163.com Link: http://lkml.kernel.org/r/1474874832-134786-3-git-send-email-wangnan0@huawei.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/build/Makefile.feature | 2 +- tools/build/feature/Makefile | 10 +++++++++- tools/build/feature/test-cxx.cpp | 15 +++++++++++++++ 3 files changed, 25 insertions(+), 2 deletions(-) create mode 100644 tools/build/feature/test-cxx.cpp (limited to 'tools') diff --git a/tools/build/Makefile.feature b/tools/build/Makefile.feature index a120c6b755a9..ae52e029dd22 100644 --- a/tools/build/Makefile.feature +++ b/tools/build/Makefile.feature @@ -7,7 +7,7 @@ endif feature_check = $(eval $(feature_check_code)) define feature_check_code - feature-$(1) := $(shell $(MAKE) OUTPUT=$(OUTPUT_FEATURES) CFLAGS="$(EXTRA_CFLAGS) $(FEATURE_CHECK_CFLAGS-$(1))" LDFLAGS="$(LDFLAGS) $(FEATURE_CHECK_LDFLAGS-$(1))" -C $(feature_dir) $(OUTPUT_FEATURES)test-$1.bin >/dev/null 2>/dev/null && echo 1 || echo 0) + feature-$(1) := $(shell $(MAKE) OUTPUT=$(OUTPUT_FEATURES) CFLAGS="$(EXTRA_CFLAGS) $(FEATURE_CHECK_CFLAGS-$(1))" CXXFLAGS="$(EXTRA_CXXFLAGS) $(FEATURE_CHECK_CXXFLAGS-$(1))" LDFLAGS="$(LDFLAGS) $(FEATURE_CHECK_LDFLAGS-$(1))" -C $(feature_dir) $(OUTPUT_FEATURES)test-$1.bin >/dev/null 2>/dev/null && echo 1 || echo 0) endef feature_set = $(eval $(feature_set_code)) diff --git a/tools/build/feature/Makefile b/tools/build/feature/Makefile index a0b29a311816..ac9c477a2a48 100644 --- a/tools/build/feature/Makefile +++ b/tools/build/feature/Makefile @@ -46,11 +46,13 @@ FILES= \ test-lzma.bin \ test-bpf.bin \ test-get_cpuid.bin \ - test-sdt.bin + test-sdt.bin \ + test-cxx.bin FILES := $(addprefix $(OUTPUT),$(FILES)) CC := $(CROSS_COMPILE)gcc -MD +CXX := $(CROSS_COMPILE)g++ -MD PKG_CONFIG := $(CROSS_COMPILE)pkg-config all: $(FILES) @@ -58,6 +60,9 @@ all: $(FILES) __BUILD = $(CC) $(CFLAGS) -Wall -Werror -o $@ $(patsubst %.bin,%.c,$(@F)) $(LDFLAGS) BUILD = $(__BUILD) > $(@:.bin=.make.output) 2>&1 +__BUILDXX = $(CXX) $(CXXFLAGS) -Wall -Werror -o $@ $(patsubst %.bin,%.cpp,$(@F)) $(LDFLAGS) + BUILDXX = $(__BUILDXX) > $(@:.bin=.make.output) 2>&1 + ############################### $(OUTPUT)test-all.bin: @@ -217,6 +222,9 @@ $(OUTPUT)test-bpf.bin: $(OUTPUT)test-sdt.bin: $(BUILD) +$(OUTPUT)test-cxx.bin: + $(BUILDXX) -std=gnu++11 + -include $(OUTPUT)*.d ############################### diff --git a/tools/build/feature/test-cxx.cpp b/tools/build/feature/test-cxx.cpp new file mode 100644 index 000000000000..b1dee9a31d6c --- /dev/null +++ b/tools/build/feature/test-cxx.cpp @@ -0,0 +1,15 @@ +#include +#include + +static void print_str(std::string s) +{ + std::cout << s << std::endl; +} + +int main() +{ + std::string s("Hello World!"); + print_str(std::move(s)); + std::cout << "|" << s << "|" << std::endl; + return 0; +} -- cgit v1.2.3