From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753081Ab1DBK0V (ORCPT ); Sat, 2 Apr 2011 06:26:21 -0400 Received: from mx3.mail.elte.hu ([157.181.1.138]:55006 "EHLO mx3.mail.elte.hu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751254Ab1DBK0T (ORCPT ); Sat, 2 Apr 2011 06:26:19 -0400 Date: Sat, 2 Apr 2011 12:25:59 +0200 From: Ingo Molnar To: Linus Torvalds Cc: linux-kernel@vger.kernel.org, Peter Zijlstra , Arnaldo Carvalho de Melo , =?iso-8859-1?Q?Fr=E9d=E9ric?= Weisbecker , Steven Rostedt , Thomas Gleixner Subject: [GIT PULL] perf fixes Message-ID: <20110402102559.GA17557@elte.hu> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.5.20 (2009-08-17) X-ELTE-SpamScore: -2.0 X-ELTE-SpamLevel: X-ELTE-SpamCheck: no X-ELTE-SpamVersion: ELTE 2.0 X-ELTE-SpamCheck-Details: score=-2.0 required=5.9 tests=BAYES_00 autolearn=no SpamAssassin version=3.3.1 -2.0 BAYES_00 BODY: Bayes spam probability is 0 to 1% [score: 0.0000] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Linus, Please pull the latest perf-fixes-for-linus git tree from: git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git perf-fixes-for-linus Thanks, Ingo ------------------> Andi Kleen (1): oprofile, x86: Allow setting EDGE/INV/CMASK for counter events Andrew Lutomirski (1): perf symbols: Fix vsyscall symbol lookup Arnaldo Carvalho de Melo (1): perf tools: Fixup exit path when not able to open events David Ahern (1): perf tools: Emit clearer message for sys_perf_event_open ENOENT return David S. Miller (1): perf symbols: Properly align symbol_conf.priv_size Frederic Weisbecker (2): perf: Rebase max unprivileged mlock threshold on top of page size perf: mmap 512 kiB by default Peter Zijlstra (2): perf: Fix task context scheduling perf: Fix task_struct reference leak Robert Richter (1): perf tools: Fix NO_NEWT=1 python build error arch/x86/oprofile/nmi_int.c | 5 ++++ arch/x86/oprofile/op_counter.h | 1 + include/linux/perf_event.h | 2 +- kernel/perf_event.c | 9 ++++++- tools/perf/Makefile | 8 +++++- tools/perf/builtin-record.c | 25 ++++++++++++++++------ tools/perf/builtin-top.c | 44 +++++++++++++++++++++++++++------------ tools/perf/util/debug.c | 10 +++++++++ tools/perf/util/debug.h | 1 + tools/perf/util/event.c | 2 +- tools/perf/util/setup.py | 7 +++++- tools/perf/util/symbol.c | 5 ++++ 12 files changed, 91 insertions(+), 28 deletions(-) diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c index 8dace18..cf97500 100644 --- a/arch/x86/oprofile/nmi_int.c +++ b/arch/x86/oprofile/nmi_int.c @@ -49,6 +49,10 @@ u64 op_x86_get_ctrl(struct op_x86_model_spec const *model, val |= counter_config->user ? ARCH_PERFMON_EVENTSEL_USR : 0; val |= counter_config->kernel ? ARCH_PERFMON_EVENTSEL_OS : 0; val |= (counter_config->unit_mask & 0xFF) << 8; + counter_config->extra &= (ARCH_PERFMON_EVENTSEL_INV | + ARCH_PERFMON_EVENTSEL_EDGE | + ARCH_PERFMON_EVENTSEL_CMASK); + val |= counter_config->extra; event &= model->event_mask ? model->event_mask : 0xFF; val |= event & 0xFF; val |= (event & 0x0F00) << 24; @@ -440,6 +444,7 @@ static int nmi_create_files(struct super_block *sb, struct dentry *root) oprofilefs_create_ulong(sb, dir, "unit_mask", &counter_config[i].unit_mask); oprofilefs_create_ulong(sb, dir, "kernel", &counter_config[i].kernel); oprofilefs_create_ulong(sb, dir, "user", &counter_config[i].user); + oprofilefs_create_ulong(sb, dir, "extra", &counter_config[i].extra); } return 0; diff --git a/arch/x86/oprofile/op_counter.h b/arch/x86/oprofile/op_counter.h index e28398d..0b7b7b1 100644 --- a/arch/x86/oprofile/op_counter.h +++ b/arch/x86/oprofile/op_counter.h @@ -22,6 +22,7 @@ struct op_counter_config { unsigned long kernel; unsigned long user; unsigned long unit_mask; + unsigned long extra; }; extern struct op_counter_config counter_config[]; diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 311b4dc..04d75a8 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -1086,7 +1086,7 @@ void perf_event_task_sched_out(struct task_struct *task, struct task_struct *nex { perf_sw_event(PERF_COUNT_SW_CONTEXT_SWITCHES, 1, 1, NULL, 0); - COND_STMT(&perf_sched_events, __perf_event_task_sched_out(task, next)); + __perf_event_task_sched_out(task, next); } extern void perf_event_mmap(struct vm_area_struct *vma); diff --git a/kernel/perf_event.c b/kernel/perf_event.c index c75925c..27960f1 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -145,8 +145,8 @@ static struct srcu_struct pmus_srcu; */ int sysctl_perf_event_paranoid __read_mostly = 1; -/* Minimum for 128 pages + 1 for the user control page */ -int sysctl_perf_event_mlock __read_mostly = 516; /* 'free' kb per user */ +/* Minimum for 512 kiB + 1 user control page */ +int sysctl_perf_event_mlock __read_mostly = 512 + (PAGE_SIZE / 1024); /* 'free' kiB per user */ /* * max perf event sample rate @@ -6531,6 +6531,11 @@ SYSCALL_DEFINE5(perf_event_open, goto err_alloc; } + if (task) { + put_task_struct(task); + task = NULL; + } + /* * Look up the group leader (we will attach this event to it): */ diff --git a/tools/perf/Makefile b/tools/perf/Makefile index 158c30e..207dee5 100644 --- a/tools/perf/Makefile +++ b/tools/perf/Makefile @@ -165,8 +165,12 @@ grep-libs = $(filter -l%,$(1)) strip-libs = $(filter-out -l%,$(1)) $(OUTPUT)python/perf.so: $(PYRF_OBJS) - $(QUIET_GEN)python util/setup.py --quiet build_ext --build-lib='$(OUTPUT)python' \ - --build-temp='$(OUTPUT)python/temp' + $(QUIET_GEN)( \ + export CFLAGS="$(BASIC_CFLAGS)"; \ + python util/setup.py --quiet build_ext --build-lib='$(OUTPUT)python' \ + --build-temp='$(OUTPUT)python/temp' \ + ) + # # No Perl scripts right now: # diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c index 6febcc1..17d1dcb 100644 --- a/tools/perf/builtin-record.c +++ b/tools/perf/builtin-record.c @@ -41,7 +41,7 @@ static u64 user_interval = ULLONG_MAX; static u64 default_interval = 0; static unsigned int page_size; -static unsigned int mmap_pages = 128; +static unsigned int mmap_pages = UINT_MAX; static unsigned int user_freq = UINT_MAX; static int freq = 1000; static int output; @@ -275,11 +275,10 @@ try_again: !no_inherit) < 0) { int err = errno; - if (err == EPERM || err == EACCES) - die("Permission error - are you root?\n" - "\t Consider tweaking" - " /proc/sys/kernel/perf_event_paranoid.\n"); - else if (err == ENODEV && cpu_list) { + if (err == EPERM || err == EACCES) { + ui__warning_paranoid(); + exit(EXIT_FAILURE); + } else if (err == ENODEV && cpu_list) { die("No such device - did you specify" " an out-of-range profile CPU?\n"); } else if (err == EINVAL && sample_id_all_avail) { @@ -302,11 +301,19 @@ try_again: && attr->config == PERF_COUNT_HW_CPU_CYCLES) { if (verbose) - warning(" ... trying to fall back to cpu-clock-ticks\n"); + ui__warning("The cycles event is not supported, " + "trying to fall back to cpu-clock-ticks\n"); attr->type = PERF_TYPE_SOFTWARE; attr->config = PERF_COUNT_SW_CPU_CLOCK; goto try_again; } + + if (err == ENOENT) { + ui__warning("The %s event is not supported.\n", + event_name(pos)); + exit(EXIT_FAILURE); + } + printf("\n"); error("sys_perf_event_open() syscall returned with %d (%s). /bin/dmesg may provide additional information.\n", err, strerror(err)); @@ -506,6 +513,10 @@ static int __cmd_record(int argc, const char **argv) if (have_tracepoints(&evsel_list->entries)) perf_header__set_feat(&session->header, HEADER_TRACE_INFO); + /* 512 kiB: default amount of unprivileged mlocked memory */ + if (mmap_pages == UINT_MAX) + mmap_pages = (512 * 1024) / page_size; + if (forks) { child_pid = fork(); if (child_pid < 0) { diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c index 676b4fb..fc1273e 100644 --- a/tools/perf/builtin-top.c +++ b/tools/perf/builtin-top.c @@ -850,10 +850,10 @@ try_again: top.evlist->threads, group, inherit) < 0) { int err = errno; - if (err == EPERM || err == EACCES) - die("Permission error - are you root?\n" - "\t Consider tweaking" - " /proc/sys/kernel/perf_event_paranoid.\n"); + if (err == EPERM || err == EACCES) { + ui__warning_paranoid(); + goto out_err; + } /* * If it's cycles then fall back to hrtimer * based cpu-clock-tick sw counter, which @@ -861,25 +861,41 @@ try_again: */ if (attr->type == PERF_TYPE_HARDWARE && attr->config == PERF_COUNT_HW_CPU_CYCLES) { - if (verbose) - warning(" ... trying to fall back to cpu-clock-ticks\n"); + ui__warning("Cycles event not supported,\n" + "trying to fall back to cpu-clock-ticks\n"); attr->type = PERF_TYPE_SOFTWARE; attr->config = PERF_COUNT_SW_CPU_CLOCK; goto try_again; } - printf("\n"); - error("sys_perf_event_open() syscall returned with %d " - "(%s). /bin/dmesg may provide additional information.\n", - err, strerror(err)); - die("No CONFIG_PERF_EVENTS=y kernel support configured?\n"); - exit(-1); + + if (err == ENOENT) { + ui__warning("The %s event is not supported.\n", + event_name(counter)); + goto out_err; + } + + ui__warning("The sys_perf_event_open() syscall " + "returned with %d (%s). /bin/dmesg " + "may provide additional information.\n" + "No CONFIG_PERF_EVENTS=y kernel support " + "configured?\n", err, strerror(err)); + goto out_err; } } - if (perf_evlist__mmap(evlist, mmap_pages, false) < 0) - die("failed to mmap with %d (%s)\n", errno, strerror(errno)); + if (perf_evlist__mmap(evlist, mmap_pages, false) < 0) { + ui__warning("Failed to mmap with %d (%s)\n", + errno, strerror(errno)); + goto out_err; + } + + return; + +out_err: + exit_browser(0); + exit(0); } static int __cmd_top(void) diff --git a/tools/perf/util/debug.c b/tools/perf/util/debug.c index d4536a9..155749d 100644 --- a/tools/perf/util/debug.c +++ b/tools/perf/util/debug.c @@ -57,6 +57,16 @@ void ui__warning(const char *format, ...) } #endif +void ui__warning_paranoid(void) +{ + ui__warning("Permission error - are you root?\n" + "Consider tweaking /proc/sys/kernel/perf_event_paranoid:\n" + " -1 - Not paranoid at all\n" + " 0 - Disallow raw tracepoint access for unpriv\n" + " 1 - Disallow cpu events for unpriv\n" + " 2 - Disallow kernel profiling for unpriv\n"); +} + void trace_event(union perf_event *event) { unsigned char *raw_event = (void *)event; diff --git a/tools/perf/util/debug.h b/tools/perf/util/debug.h index 93516cf4..fd53db4 100644 --- a/tools/perf/util/debug.h +++ b/tools/perf/util/debug.h @@ -36,5 +36,6 @@ int ui_helpline__show_help(const char *format, va_list ap); #endif void ui__warning(const char *format, ...) __attribute__((format(printf, 1, 2))); +void ui__warning_paranoid(void); #endif /* __PERF_DEBUG_H */ diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c index 2b15c36..1023f67 100644 --- a/tools/perf/util/event.c +++ b/tools/perf/util/event.c @@ -710,7 +710,7 @@ try_again: * in the whole kernel symbol list. */ if ((long long)al->addr < 0 && - cpumode == PERF_RECORD_MISC_KERNEL && + cpumode == PERF_RECORD_MISC_USER && machine && mg != &machine->kmaps) { mg = &machine->kmaps; goto try_again; diff --git a/tools/perf/util/setup.py b/tools/perf/util/setup.py index e24ffad..bbc982f 100644 --- a/tools/perf/util/setup.py +++ b/tools/perf/util/setup.py @@ -1,13 +1,18 @@ #!/usr/bin/python2 from distutils.core import setup, Extension +from os import getenv + +cflags = ['-fno-strict-aliasing', '-Wno-write-strings'] +cflags += getenv('CFLAGS', '').split() perf = Extension('perf', sources = ['util/python.c', 'util/ctype.c', 'util/evlist.c', 'util/evsel.c', 'util/cpumap.c', 'util/thread_map.c', 'util/util.c', 'util/xyarray.c', 'util/cgroup.c'], include_dirs = ['util/include'], - extra_compile_args = ['-fno-strict-aliasing', '-Wno-write-strings']) + extra_compile_args = cflags, + ) setup(name='perf', version='0.1', diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c index 17df793..f06c10f 100644 --- a/tools/perf/util/symbol.c +++ b/tools/perf/util/symbol.c @@ -1196,6 +1196,8 @@ static int dso__load_sym(struct dso *self, struct map *map, const char *name, if (curr_dso == NULL) goto out_elf_end; curr_dso->kernel = self->kernel; + curr_dso->long_name = self->long_name; + curr_dso->long_name_len = self->long_name_len; curr_map = map__new2(start, curr_dso, map->type); if (curr_map == NULL) { @@ -1842,6 +1844,7 @@ int dso__load_vmlinux(struct dso *self, struct map *map, if (fd < 0) return -1; + dso__set_long_name(self, (char *)vmlinux); dso__set_loaded(self, map->type); err = dso__load_sym(self, map, symfs_vmlinux, fd, filter, 0, 0); close(fd); @@ -2403,6 +2406,8 @@ int symbol__init(void) if (symbol_conf.initialized) return 0; + symbol_conf.priv_size = ALIGN(symbol_conf.priv_size, sizeof(u64)); + elf_version(EV_CURRENT); if (symbol_conf.sort_by_name) symbol_conf.priv_size += (sizeof(struct symbol_name_rb_node) -