From fa1531fdd7b6332aa61bcc9fda495583acba460d Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Mon, 10 Jun 2013 16:31:28 -0400 Subject: perf tools: Remove callchain_cursor_reset call Removing callchain_cursor_reset call as it is called in subsequent machine__resolve_callchain_sample function. Signed-off-by: Jiri Olsa Cc: Andi Kleen Cc: Corey Ashford Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Link: http://lkml.kernel.org/n/tip-ic53wabwmmgvvwve2ymv3yf7@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/machine.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index b2ecad6ec46b..93527afb09d5 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -1242,8 +1242,6 @@ int machine__resolve_callchain(struct machine *machine, { int ret; - callchain_cursor_reset(&callchain_cursor); - ret = machine__resolve_callchain_sample(machine, thread, sample->callchain, parent); if (ret) -- cgit v1.2.2 From 380512345e13c3af64e59627f1b993c4faa94a84 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Thu, 4 Jul 2013 16:20:31 +0300 Subject: perf tools: struct thread has a tid not a pid As evident from 'machine__process_fork_event()' and 'machine__process_exit_event()' the 'pid' member of struct thread is actually the tid. Rename 'pid' to 'tid' in struct thread accordingly. Signed-off-by: Adrian Hunter Acked-by: David Ahern Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1372944040-32690-13-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/machine.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index 93527afb09d5..5dd5026a82ef 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -233,7 +233,7 @@ void machines__set_id_hdr_size(struct machines *machines, u16 id_hdr_size) return; } -static struct thread *__machine__findnew_thread(struct machine *machine, pid_t pid, +static struct thread *__machine__findnew_thread(struct machine *machine, pid_t tid, bool create) { struct rb_node **p = &machine->threads.rb_node; @@ -241,23 +241,23 @@ static struct thread *__machine__findnew_thread(struct machine *machine, pid_t p struct thread *th; /* - * Font-end cache - PID lookups come in blocks, + * Front-end cache - TID lookups come in blocks, * so most of the time we dont have to look up * the full rbtree: */ - if (machine->last_match && machine->last_match->pid == pid) + if (machine->last_match && machine->last_match->tid == tid) return machine->last_match; while (*p != NULL) { parent = *p; th = rb_entry(parent, struct thread, rb_node); - if (th->pid == pid) { + if (th->tid == tid) { machine->last_match = th; return th; } - if (pid < th->pid) + if (tid < th->tid) p = &(*p)->rb_left; else p = &(*p)->rb_right; @@ -266,7 +266,7 @@ static struct thread *__machine__findnew_thread(struct machine *machine, pid_t p if (!create) return NULL; - th = thread__new(pid); + th = thread__new(tid); if (th != NULL) { rb_link_node(&th->rb_node, parent, p); rb_insert_color(&th->rb_node, &machine->threads); @@ -276,14 +276,14 @@ static struct thread *__machine__findnew_thread(struct machine *machine, pid_t p return th; } -struct thread *machine__findnew_thread(struct machine *machine, pid_t pid) +struct thread *machine__findnew_thread(struct machine *machine, pid_t tid) { - return __machine__findnew_thread(machine, pid, true); + return __machine__findnew_thread(machine, tid, true); } -struct thread *machine__find_thread(struct machine *machine, pid_t pid) +struct thread *machine__find_thread(struct machine *machine, pid_t tid) { - return __machine__findnew_thread(machine, pid, false); + return __machine__findnew_thread(machine, tid, false); } int machine__process_comm_event(struct machine *machine, union perf_event *event) -- cgit v1.2.2 From b21484f1a1f300d422cfe5d4f8f50015e22cea24 Mon Sep 17 00:00:00 2001 From: Greg Price Date: Thu, 6 Dec 2012 21:48:05 -0800 Subject: perf report/top: Add option to collapse undesired parts of call graph For example, in an application with an expensive function implemented with deeply nested recursive calls, the default call-graph presentation is dominated by the different callchains within that function. By ignoring these callees, we can collect the callchains leading into the function and compactly identify what to blame for expensive calls. For example, in this report the callers of garbage_collect() are scattered across the tree: $ perf report -d ruby 2>- | grep -m10 ^[^#]*[a-z] 22.03% ruby [.] gc_mark --- gc_mark |--59.40%-- mark_keyvalue | st_foreach | gc_mark_children | |--99.75%-- rb_gc_mark | | rb_vm_mark | | gc_mark_children | | gc_marks | | |--99.00%-- garbage_collect If we ignore the callees of garbage_collect(), its callers are coalesced: $ perf report --ignore-callees garbage_collect -d ruby 2>- | grep -m10 ^[^#]*[a-z] 72.92% ruby [.] garbage_collect --- garbage_collect vm_xmalloc |--47.08%-- ruby_xmalloc | st_insert2 | rb_hash_aset | |--98.45%-- features_index_add | | rb_provide_feature | | rb_require_safe | | vm_call_method Signed-off-by: Greg Price Tested-by: Jiri Olsa Cc: David Ahern Cc: Ingo Molnar Cc: Jiri Olsa Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20130623031720.GW22203@biohazard-cafe.mit.edu Link: http://lkml.kernel.org/r/20130708115746.GO22203@biohazard-cafe.mit.edu Cc: Fengguang Wu [ remove spaces at beginning of line, reported by Fengguang Wu ] Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/machine.c | 24 +++++++++++++++--------- 1 file changed, 15 insertions(+), 9 deletions(-) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index 5dd5026a82ef..f9f9d6381b9a 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -1058,11 +1058,10 @@ int machine__process_event(struct machine *machine, union perf_event *event) return ret; } -static bool symbol__match_parent_regex(struct symbol *sym) +static bool symbol__match_regex(struct symbol *sym, regex_t *regex) { - if (sym->name && !regexec(&parent_regex, sym->name, 0, NULL, 0)) + if (sym->name && !regexec(regex, sym->name, 0, NULL, 0)) return 1; - return 0; } @@ -1159,8 +1158,8 @@ struct branch_info *machine__resolve_bstack(struct machine *machine, static int machine__resolve_callchain_sample(struct machine *machine, struct thread *thread, struct ip_callchain *chain, - struct symbol **parent) - + struct symbol **parent, + struct addr_location *root_al) { u8 cpumode = PERF_RECORD_MISC_USER; unsigned int i; @@ -1211,8 +1210,15 @@ static int machine__resolve_callchain_sample(struct machine *machine, MAP__FUNCTION, ip, &al, NULL); if (al.sym != NULL) { if (sort__has_parent && !*parent && - symbol__match_parent_regex(al.sym)) + symbol__match_regex(al.sym, &parent_regex)) *parent = al.sym; + else if (have_ignore_callees && root_al && + symbol__match_regex(al.sym, &ignore_callees_regex)) { + /* Treat this symbol as the root, + forgetting its callees. */ + *root_al = al; + callchain_cursor_reset(&callchain_cursor); + } if (!symbol_conf.use_callchain) break; } @@ -1237,13 +1243,13 @@ int machine__resolve_callchain(struct machine *machine, struct perf_evsel *evsel, struct thread *thread, struct perf_sample *sample, - struct symbol **parent) - + struct symbol **parent, + struct addr_location *root_al) { int ret; ret = machine__resolve_callchain_sample(machine, thread, - sample->callchain, parent); + sample->callchain, parent, root_al); if (ret) return ret; -- cgit v1.2.2 From 39b12f7812710e9a5896805d96812b3ede7491e8 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Wed, 7 Aug 2013 14:38:47 +0300 Subject: perf tools: Make it possible to read object code from vmlinux The new "object code reading" test shows that it is not possible to read object code from vmlinux. That is because the mappings do not map to the dso. This patch fixes that. A side-effect of changing the kernel map is that the "reloc" offset must be taken into account. As a result of that separate map functions for relocation are no longer needed. Also fixing up the maps to match the symbols no longer makes sense and so is not done. The vmlinux dso data_type is now set to either DSO_BINARY_TYPE__VMLINUX or DSO_BINARY_TYPE__GUEST_VMLINUX as approprite, which enables the correct file name to be determined by dso__binary_type_file(). This patch breaks the "vmlinux symtab matches kallsyms" test. That is fixed in a following patch. Signed-off-by: Adrian Hunter Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1375875537-4509-4-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/machine.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index f9f9d6381b9a..dc35dcffbd05 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -628,10 +628,8 @@ int machine__load_vmlinux_path(struct machine *machine, enum map_type type, struct map *map = machine->vmlinux_maps[type]; int ret = dso__load_vmlinux_path(map->dso, map, filter); - if (ret > 0) { + if (ret > 0) dso__set_loaded(map->dso, type); - map__reloc_vmlinux(map); - } return ret; } -- cgit v1.2.2 From 8e0cf965f95edd41df11cca50b92b4cb6ea8d80a Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Wed, 7 Aug 2013 14:38:51 +0300 Subject: perf symbols: Add support for reading from /proc/kcore In the absence of vmlinux, perf tools uses kallsyms for symbols. If the user has access, now also map to /proc/kcore. The dso data_type is now set to either DSO_BINARY_TYPE__KCORE or DSO_BINARY_TYPE__GUEST_KCORE as approprite. This patch breaks the "vmlinux symtab matches kallsyms" test. That is fixed in a following patch. Signed-off-by: Adrian Hunter Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1375875537-4509-8-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/machine.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index dc35dcffbd05..ef3b49cde04e 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -856,6 +856,18 @@ static void machine__set_kernel_mmap_len(struct machine *machine, } } +static bool machine__uses_kcore(struct machine *machine) +{ + struct dso *dso; + + list_for_each_entry(dso, &machine->kernel_dsos, node) { + if (dso__is_kcore(dso)) + return true; + } + + return false; +} + static int machine__process_kernel_mmap_event(struct machine *machine, union perf_event *event) { @@ -864,6 +876,10 @@ static int machine__process_kernel_mmap_event(struct machine *machine, enum dso_kernel_type kernel_type; bool is_kernel_mmap; + /* If we have maps from kcore then we do not need or want any others */ + if (machine__uses_kcore(machine)) + return 0; + machine__mmap_name(machine, kmmap_prefix, sizeof(kmmap_prefix)); if (machine__is_host(machine)) kernel_type = DSO_TYPE_KERNEL; -- cgit v1.2.2 From 8f76fcd902e3b3a7d6f6c695cc8bc053579eb179 Mon Sep 17 00:00:00 2001 From: Jason Wessel Date: Mon, 15 Jul 2013 15:27:53 -0500 Subject: perf machine: Do not require /lib/modules/* on a guest For some types of work loads and special guest environments, you might have a kernel that has no kernel modules. The perf kvm record tool fails instantiate vmlinux maps when the kernel modules directory cannot be opened, even though the kallsyms has been properly processed. This leads to a perf kvm report that has no guest symbols resolved. This patch changes the failure to locate kernel modules to be non-fatal. Signed-off-by: Jason Wessel Acked-by: David Ahern Cc: Paul Mackerras Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/1373920073-4874-1-git-send-email-jason.wessel@windriver.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/machine.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index ef3b49cde04e..6fcc358138ae 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -806,7 +806,10 @@ static int machine__create_modules(struct machine *machine) free(line); fclose(file); - return machine__set_modules_path(machine); + if (machine__set_modules_path(machine) < 0) { + pr_debug("Problems setting modules path maps, continuing anyway...\n"); + } + return 0; out_delete_line: free(line); -- cgit v1.2.2 From 611a5ce8aa4cdb2daefbd9bed77ec3b3e9bd00ea Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Thu, 8 Aug 2013 14:32:20 +0300 Subject: perf machine: Add symbol filter to struct machine The symbol filter needs to be applied machine-wide, so add it to struct machine. Currently tools pass the symbol filter as a parameter to various map-related functions. However a need to load a map can occur anywhere in the code, at which point the filter is needed. Signed-off-by: Adrian Hunter Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1375961547-30267-2-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/machine.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index 6fcc358138ae..4c7e0a28bf3d 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -25,6 +25,8 @@ int machine__init(struct machine *machine, const char *root_dir, pid_t pid) machine->kmaps.machine = machine; machine->pid = pid; + machine->symbol_filter = NULL; + machine->root_dir = strdup(root_dir); if (machine->root_dir == NULL) return -ENOMEM; @@ -95,6 +97,7 @@ void machines__init(struct machines *machines) { machine__init(&machines->host, "", HOST_KERNEL_ID); machines->guests = RB_ROOT; + machines->symbol_filter = NULL; } void machines__exit(struct machines *machines) @@ -118,6 +121,8 @@ struct machine *machines__add(struct machines *machines, pid_t pid, return NULL; } + machine->symbol_filter = machines->symbol_filter; + while (*p != NULL) { parent = *p; pos = rb_entry(parent, struct machine, rb_node); @@ -133,6 +138,21 @@ struct machine *machines__add(struct machines *machines, pid_t pid, return machine; } +void machines__set_symbol_filter(struct machines *machines, + symbol_filter_t symbol_filter) +{ + struct rb_node *nd; + + machines->symbol_filter = symbol_filter; + machines->host.symbol_filter = symbol_filter; + + for (nd = rb_first(&machines->guests); nd; nd = rb_next(nd)) { + struct machine *machine = rb_entry(nd, struct machine, rb_node); + + machine->symbol_filter = symbol_filter; + } +} + struct machine *machines__find(struct machines *machines, pid_t pid) { struct rb_node **p = &machines->guests.rb_node; -- cgit v1.2.2 From 61710bdee324aab1c148c8573ee49cea59d05874 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Thu, 8 Aug 2013 14:32:26 +0300 Subject: perf tools: Remove filter parameter of thread__find_addr_location() Now that the symbol filter is recorded on the machine there is no need to pass it to thread__find_addr_location(). So remove it. Signed-off-by: Adrian Hunter Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1375961547-30267-8-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/machine.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index 4c7e0a28bf3d..4514e7e9b659 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -1130,7 +1130,7 @@ static void ip__resolve_ams(struct machine *machine, struct thread *thread, * or else, the symbol is unknown */ thread__find_addr_location(thread, machine, m, MAP__FUNCTION, - ip, &al, NULL); + ip, &al); if (al.sym) goto found; } @@ -1148,8 +1148,8 @@ static void ip__resolve_data(struct machine *machine, struct thread *thread, memset(&al, 0, sizeof(al)); - thread__find_addr_location(thread, machine, m, MAP__VARIABLE, addr, &al, - NULL); + thread__find_addr_location(thread, machine, m, MAP__VARIABLE, addr, + &al); ams->addr = addr; ams->al_addr = al.addr; ams->sym = al.sym; @@ -1244,7 +1244,7 @@ static int machine__resolve_callchain_sample(struct machine *machine, al.filtered = false; thread__find_addr_location(thread, machine, cpumode, - MAP__FUNCTION, ip, &al, NULL); + MAP__FUNCTION, ip, &al); if (al.sym != NULL) { if (sort__has_parent && !*parent && symbol__match_regex(al.sym, &parent_regex)) -- cgit v1.2.2 From 236a3bbd5cb51edbf9550f5a7df885665d18a271 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Wed, 14 Aug 2013 08:49:27 -0600 Subject: perf tools: Sample after exit loses thread correlation Occassionally events (e.g., context-switch, sched tracepoints) are losing the conversion of sample data associated with a thread. For example: $ perf record -e sched:sched_switch -c 1 -a -- sleep 5 $ perf script ls 30482 [000] 1379727.583037: sched:sched_switch: prev_comm=ls prev_pid=30482 ... ls 30482 [000] 1379727.586339: sched:sched_switch: prev_comm=ls prev_pid=30482 ... :30482 30482 [000] 1379727.589462: sched:sched_switch: prev_comm=ls prev_pid=30482 ... The last line lost the conversion from tid to comm. If you look at the events (perf script -D) you see why - a SAMPLE event is generated after the EXIT: 0 1379727589449774 0x1540b0 [0x38]: PERF_RECORD_EXIT(30482:30482):(30482:30482) 0 1379727589462497 0x1540e8 [0x80]: PERF_RECORD_SAMPLE(IP, 1): 30482/30482: 0xffffffff816416f1 period: 1 addr: 0 ... thread: :30482:30482 When perf processes the EXIT event the thread is moved to the dead_threads list. When the SAMPLE event is processed no thread exists for the pid so a new one is created by machine__findnew_thread. This patch address the problem by delaying the move to the dead_threads list until the tid is re-used (per Adrian's suggestion). With this patch we get the previous example shows: ls 30482 [000] 1379727.583037: sched:sched_switch: prev_comm=ls prev_pid=30482 ... ls 30482 [000] 1379727.586339: sched:sched_switch: prev_comm=ls prev_pid=30482 ... ls 30482 [000] 1379727.589462: sched:sched_switch: prev_comm=ls prev_pid=30482 ... and 0 1379727589449774 0x1540b0 [0x38]: PERF_RECORD_EXIT(30482:30482):(30482:30482) 0 1379727589462497 0x1540e8 [0x80]: PERF_RECORD_SAMPLE(IP, 1): 30482/30482: 0xffffffff816416f1 period: 1 addr: 0 ... thread: ls:30482 v4: per Arnaldo's request add dead flag to thread struct and set when task exits v3: re-do from a time based check to a delayed move to dead_threads list v2: Rebased to latest perf/core branch. Changed time comparison to use a macro which explicitly shows the time basis Signed-off-by: David Ahern Acked-by: Adrian Hunter Cc: Adrian Hunter Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1376491767-84171-1-git-send-email-dsahern@gmail.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/machine.c | 34 ++++++++++++++++++++-------------- 1 file changed, 20 insertions(+), 14 deletions(-) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index 4514e7e9b659..574feb7003ab 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -1031,11 +1031,27 @@ out_problem: return 0; } +static void machine__remove_thread(struct machine *machine, struct thread *th) +{ + machine->last_match = NULL; + rb_erase(&th->rb_node, &machine->threads); + /* + * We may have references to this thread, for instance in some hist_entry + * instances, so just move them to a separate list. + */ + list_add_tail(&th->node, &machine->dead_threads); +} + int machine__process_fork_event(struct machine *machine, union perf_event *event) { - struct thread *thread = machine__findnew_thread(machine, event->fork.tid); + struct thread *thread = machine__find_thread(machine, event->fork.tid); struct thread *parent = machine__findnew_thread(machine, event->fork.ptid); + /* if a thread currently exists for the thread id remove it */ + if (thread != NULL) + machine__remove_thread(machine, thread); + + thread = machine__findnew_thread(machine, event->fork.tid); if (dump_trace) perf_event__fprintf_task(event, stdout); @@ -1048,18 +1064,8 @@ int machine__process_fork_event(struct machine *machine, union perf_event *event return 0; } -static void machine__remove_thread(struct machine *machine, struct thread *th) -{ - machine->last_match = NULL; - rb_erase(&th->rb_node, &machine->threads); - /* - * We may have references to this thread, for instance in some hist_entry - * instances, so just move them to a separate list. - */ - list_add_tail(&th->node, &machine->dead_threads); -} - -int machine__process_exit_event(struct machine *machine, union perf_event *event) +int machine__process_exit_event(struct machine *machine __maybe_unused, + union perf_event *event) { struct thread *thread = machine__find_thread(machine, event->fork.tid); @@ -1067,7 +1073,7 @@ int machine__process_exit_event(struct machine *machine, union perf_event *event perf_event__fprintf_task(event, stdout); if (thread != NULL) - machine__remove_thread(machine, thread); + thread__exited(thread); return 0; } -- cgit v1.2.2 From 99d725fc65563a85d4290342c81b00a673c6be66 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Mon, 26 Aug 2013 16:00:19 +0300 Subject: perf tools: Add pid to struct thread Record pid on struct thread. The member is named 'pid_' to avoid confusion with the 'tid' member which was previously named 'pid'. Signed-off-by: Adrian Hunter Acked-by: David Ahern Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1377522030-27870-3-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/machine.c | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index 574feb7003ab..59486c180626 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -253,7 +253,8 @@ void machines__set_id_hdr_size(struct machines *machines, u16 id_hdr_size) return; } -static struct thread *__machine__findnew_thread(struct machine *machine, pid_t tid, +static struct thread *__machine__findnew_thread(struct machine *machine, + pid_t pid, pid_t tid, bool create) { struct rb_node **p = &machine->threads.rb_node; @@ -265,8 +266,11 @@ static struct thread *__machine__findnew_thread(struct machine *machine, pid_t t * so most of the time we dont have to look up * the full rbtree: */ - if (machine->last_match && machine->last_match->tid == tid) + if (machine->last_match && machine->last_match->tid == tid) { + if (pid && pid != machine->last_match->pid_) + machine->last_match->pid_ = pid; return machine->last_match; + } while (*p != NULL) { parent = *p; @@ -274,6 +278,8 @@ static struct thread *__machine__findnew_thread(struct machine *machine, pid_t t if (th->tid == tid) { machine->last_match = th; + if (pid && pid != th->pid_) + th->pid_ = pid; return th; } @@ -286,7 +292,7 @@ static struct thread *__machine__findnew_thread(struct machine *machine, pid_t t if (!create) return NULL; - th = thread__new(tid); + th = thread__new(pid, tid); if (th != NULL) { rb_link_node(&th->rb_node, parent, p); rb_insert_color(&th->rb_node, &machine->threads); @@ -298,12 +304,12 @@ static struct thread *__machine__findnew_thread(struct machine *machine, pid_t t struct thread *machine__findnew_thread(struct machine *machine, pid_t tid) { - return __machine__findnew_thread(machine, tid, true); + return __machine__findnew_thread(machine, 0, tid, true); } struct thread *machine__find_thread(struct machine *machine, pid_t tid) { - return __machine__findnew_thread(machine, tid, false); + return __machine__findnew_thread(machine, 0, tid, false); } int machine__process_comm_event(struct machine *machine, union perf_event *event) -- cgit v1.2.2 From 314add6b1f045b59ca39683bd0cbc5310cd203f2 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Tue, 27 Aug 2013 11:23:03 +0300 Subject: perf tools: change machine__findnew_thread() to set thread pid Add a new parameter for 'pid' to machine__findnew_thread(). Change callers to pass 'pid' when it is known. Note that callers sometimes want to find the main thread which has the memory maps. The main thread has tid == pid so the usage in that case is: machine__findnew_thread(machine, pid, pid) whereas the usage to find the specific thread is: machine__findnew_thread(machine, pid, tid) Signed-off-by: Adrian Hunter Acked-by: David Ahern Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1377591794-30553-2-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/machine.c | 22 +++++++++++++++------- 1 file changed, 15 insertions(+), 7 deletions(-) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index 59486c180626..1dca61f0512d 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -32,7 +32,8 @@ int machine__init(struct machine *machine, const char *root_dir, pid_t pid) return -ENOMEM; if (pid != HOST_KERNEL_ID) { - struct thread *thread = machine__findnew_thread(machine, pid); + struct thread *thread = machine__findnew_thread(machine, 0, + pid); char comm[64]; if (thread == NULL) @@ -302,9 +303,10 @@ static struct thread *__machine__findnew_thread(struct machine *machine, return th; } -struct thread *machine__findnew_thread(struct machine *machine, pid_t tid) +struct thread *machine__findnew_thread(struct machine *machine, pid_t pid, + pid_t tid) { - return __machine__findnew_thread(machine, 0, tid, true); + return __machine__findnew_thread(machine, pid, tid, true); } struct thread *machine__find_thread(struct machine *machine, pid_t tid) @@ -314,7 +316,9 @@ struct thread *machine__find_thread(struct machine *machine, pid_t tid) int machine__process_comm_event(struct machine *machine, union perf_event *event) { - struct thread *thread = machine__findnew_thread(machine, event->comm.tid); + struct thread *thread = machine__findnew_thread(machine, + event->comm.pid, + event->comm.tid); if (dump_trace) perf_event__fprintf_comm(event, stdout); @@ -1012,7 +1016,8 @@ int machine__process_mmap_event(struct machine *machine, union perf_event *event return 0; } - thread = machine__findnew_thread(machine, event->mmap.pid); + thread = machine__findnew_thread(machine, event->mmap.pid, + event->mmap.pid); if (thread == NULL) goto out_problem; @@ -1051,13 +1056,16 @@ static void machine__remove_thread(struct machine *machine, struct thread *th) int machine__process_fork_event(struct machine *machine, union perf_event *event) { struct thread *thread = machine__find_thread(machine, event->fork.tid); - struct thread *parent = machine__findnew_thread(machine, event->fork.ptid); + struct thread *parent = machine__findnew_thread(machine, + event->fork.ppid, + event->fork.ptid); /* if a thread currently exists for the thread id remove it */ if (thread != NULL) machine__remove_thread(machine, thread); - thread = machine__findnew_thread(machine, event->fork.tid); + thread = machine__findnew_thread(machine, event->fork.pid, + event->fork.tid); if (dump_trace) perf_event__fprintf_task(event, stdout); -- cgit v1.2.2 From 5c5e854bc760a2e2c878df3cfcf2afa4febcd511 Mon Sep 17 00:00:00 2001 From: Stephane Eranian Date: Wed, 21 Aug 2013 12:10:25 +0200 Subject: perf tools: Add attr->mmap2 support This patch adds support for the new PERF_RECORD_MMAP2 record type exposed by the kernel. This is an extended PERF_RECORD_MMAP record. It adds for each file-backed mapping the device major, minor number and the inode number and generation. This triplet uniquely identifies the source of a file-backed mapping. It can be used to detect identical virtual mappings between processes, for instance. The patch will prefer MMAP2 over MMAP. Signed-off-by: Stephane Eranian Cc: Andi Kleen Cc: David Ahern Cc: Ingo Molnar Cc: Jiri Olsa Cc: Namhyung Kim Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/1377079825-19057-3-git-send-email-eranian@google.com [ Cope with 314add6 "Change machine__findnew_thread() to set thread pid", fix 'perf test' regression test entry affected, use perf_missing_features.mmap2 to fallback to not using .mmap2 in older kernels, so that new tools can work with kernels where this feature is not present ] Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/machine.c | 53 ++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 52 insertions(+), 1 deletion(-) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index 1dca61f0512d..933d14f287ca 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -997,6 +997,54 @@ out_problem: return -1; } +int machine__process_mmap2_event(struct machine *machine, + union perf_event *event) +{ + u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK; + struct thread *thread; + struct map *map; + enum map_type type; + int ret = 0; + + if (dump_trace) + perf_event__fprintf_mmap2(event, stdout); + + if (cpumode == PERF_RECORD_MISC_GUEST_KERNEL || + cpumode == PERF_RECORD_MISC_KERNEL) { + ret = machine__process_kernel_mmap_event(machine, event); + if (ret < 0) + goto out_problem; + return 0; + } + + thread = machine__findnew_thread(machine, event->mmap2.pid, + event->mmap2.pid); + if (thread == NULL) + goto out_problem; + + if (event->header.misc & PERF_RECORD_MISC_MMAP_DATA) + type = MAP__VARIABLE; + else + type = MAP__FUNCTION; + + map = map__new(&machine->user_dsos, event->mmap2.start, + event->mmap2.len, event->mmap2.pgoff, + event->mmap2.pid, event->mmap2.maj, + event->mmap2.min, event->mmap2.ino, + event->mmap2.ino_generation, + event->mmap2.filename, type); + + if (map == NULL) + goto out_problem; + + thread__insert_map(thread, map); + return 0; + +out_problem: + dump_printf("problem processing PERF_RECORD_MMAP2, skipping event.\n"); + return 0; +} + int machine__process_mmap_event(struct machine *machine, union perf_event *event) { u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK; @@ -1028,7 +1076,8 @@ int machine__process_mmap_event(struct machine *machine, union perf_event *event map = map__new(&machine->user_dsos, event->mmap.start, event->mmap.len, event->mmap.pgoff, - event->mmap.pid, event->mmap.filename, + event->mmap.pid, 0, 0, 0, 0, + event->mmap.filename, type); if (map == NULL) @@ -1101,6 +1150,8 @@ int machine__process_event(struct machine *machine, union perf_event *event) ret = machine__process_comm_event(machine, event); break; case PERF_RECORD_MMAP: ret = machine__process_mmap_event(machine, event); break; + case PERF_RECORD_MMAP2: + ret = machine__process_mmap2_event(machine, event); break; case PERF_RECORD_FORK: ret = machine__process_fork_event(machine, event); break; case PERF_RECORD_EXIT: -- cgit v1.2.2 From aa7fe3b0c499fb7987245ac40295af03546f2bd2 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Sun, 22 Sep 2013 13:22:09 +0300 Subject: perf machine: Fix path unpopulated in machine__create_modules() In machine__create_modules() the 'path' char array was used in a call to symbol__restricted_filename() without always being populated. Signed-off-by: Adrian Hunter Cc: David Ahern Cc: Frederic Weisbecker Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1379845338-29637-2-git-send-email-adrian.hunter@intel.com [ Split patch removing unrelated conversion of sprintf to snprintf to perf/core ] Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/machine.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index 933d14f287ca..6188d2876a71 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -792,7 +792,7 @@ static int machine__create_modules(struct machine *machine) modules = path; } - if (symbol__restricted_filename(path, "/proc/modules")) + if (symbol__restricted_filename(modules, "/proc/modules")) return -1; file = fopen(modules, "r"); -- cgit v1.2.2