diff options
58 files changed, 3581 insertions, 863 deletions
diff --git a/tools/perf/Documentation/db-export.txt b/tools/perf/Documentation/db-export.txt new file mode 100644 index 000000000000..52ffccb02d55 --- /dev/null +++ b/tools/perf/Documentation/db-export.txt @@ -0,0 +1,41 @@ +Database Export +=============== + +perf tool's python scripting engine: + + tools/perf/util/scripting-engines/trace-event-python.c + +supports scripts: + + tools/perf/scripts/python/export-to-sqlite.py + tools/perf/scripts/python/export-to-postgresql.py + +which export data to a SQLite3 or PostgreSQL database. + +The export process provides records with unique sequential ids which allows the +data to be imported directly to a database and provides the relationships +between tables. + +Over time it is possible to continue to expand the export while maintaining +backward and forward compatibility, by following some simple rules: + +1. Because of the nature of SQL, existing tables and columns can continue to be +used so long as the names and meanings (and to some extent data types) remain +the same. + +2. New tables and columns can be added, without affecting existing SQL queries, +so long as the new names are unique. + +3. Scripts that use a database (e.g. exported-sql-viewer.py) can maintain +backward compatibility by testing for the presence of new tables and columns +before using them. e.g. function IsSelectable() in exported-sql-viewer.py + +4. The export scripts themselves maintain forward compatibility (i.e. an existing +script will continue to work with new versions of perf) by accepting a variable +number of arguments (e.g. def call_return_table(*x)) i.e. perf can pass more +arguments which old scripts will ignore. + +5. The scripting engine tests for the existence of script handler functions +before calling them. The scripting engine can also test for the support of new +or optional features by checking for the existence and value of script global +variables. diff --git a/tools/perf/Documentation/intel-pt.txt b/tools/perf/Documentation/intel-pt.txt index 60d99e5e7921..50c5b60101bd 100644 --- a/tools/perf/Documentation/intel-pt.txt +++ b/tools/perf/Documentation/intel-pt.txt @@ -103,6 +103,36 @@ The flags are "bcrosyiABEx" which stand for branch, call, return, conditional, system, asynchronous, interrupt, transaction abort, trace begin, trace end, and in transaction, respectively. +Another interesting field that is not printed by default is 'ipc' which can be +displayed as follows: + + perf script --itrace=be -F+ipc + +There are two ways that instructions-per-cycle (IPC) can be calculated depending +on the recording. + +If the 'cyc' config term (see config terms section below) was used, then IPC is +calculated using the cycle count from CYC packets, otherwise MTC packets are +used - refer to the 'mtc' config term. When MTC is used, however, the values +are less accurate because the timing is less accurate. + +Because Intel PT does not update the cycle count on every branch or instruction, +the values will often be zero. When there are values, they will be the number +of instructions and number of cycles since the last update, and thus represent +the average IPC since the last IPC for that event type. Note IPC for "branches" +events is calculated separately from IPC for "instructions" events. + +Also note that the IPC instruction count may or may not include the current +instruction. If the cycle count is associated with an asynchronous branch +(e.g. page fault or interrupt), then the instruction count does not include the +current instruction, otherwise it does. That is consistent with whether or not +that instruction has retired when the cycle count is updated. + +Another note, in the case of "branches" events, non-taken branches are not +presently sampled, so IPC values for them do not appear e.g. a CYC packet with a +TNT packet that starts with a non-taken branch. To see every possible IPC +value, "instructions" events can be used e.g. --itrace=i0ns + While it is possible to create scripts to analyze the data, an alternative approach is available to export the data to a sqlite or postgresql database. Refer to script export-to-sqlite.py or export-to-postgresql.py for more details, diff --git a/tools/perf/Documentation/perf-config.txt b/tools/perf/Documentation/perf-config.txt index 462b3cde0675..e4aa268d2e38 100644 --- a/tools/perf/Documentation/perf-config.txt +++ b/tools/perf/Documentation/perf-config.txt @@ -564,9 +564,12 @@ llvm.*:: llvm.clang-bpf-cmd-template:: Cmdline template. Below lines show its default value. Environment variable is used to pass options. - "$CLANG_EXEC -D__KERNEL__ $CLANG_OPTIONS $KERNEL_INC_OPTIONS \ - -Wno-unused-value -Wno-pointer-sign -working-directory \ - $WORKING_DIR -c $CLANG_SOURCE -target bpf -O2 -o -" + "$CLANG_EXEC -D__KERNEL__ -D__NR_CPUS__=$NR_CPUS "\ + "-DLINUX_VERSION_CODE=$LINUX_VERSION_CODE " \ + "$CLANG_OPTIONS $PERF_BPF_INC_OPTIONS $KERNEL_INC_OPTIONS " \ + "-Wno-unused-value -Wno-pointer-sign " \ + "-working-directory $WORKING_DIR " \ + "-c \"$CLANG_SOURCE\" -target bpf $CLANG_EMIT_LLVM -O2 -o - $LLVM_OPTIONS_PIPE" llvm.clang-opt:: Options passed to clang. diff --git a/tools/perf/Documentation/perf-diff.txt b/tools/perf/Documentation/perf-diff.txt index da7809b15cc9..facd91e4e945 100644 --- a/tools/perf/Documentation/perf-diff.txt +++ b/tools/perf/Documentation/perf-diff.txt @@ -142,12 +142,14 @@ OPTIONS perf diff --time 0%-10%,30%-40% It also supports analyzing samples within a given time window - <start>,<stop>. Times have the format seconds.microseconds. If 'start' - is not given (i.e., time string is ',x.y') then analysis starts at - the beginning of the file. If stop time is not given (i.e, time - string is 'x.y,') then analysis goes to the end of the file. Time string is - 'a1.b1,c1.d1:a2.b2,c2.d2'. Use ':' to separate timestamps for different - perf.data files. + <start>,<stop>. Times have the format seconds.nanoseconds. If 'start' + is not given (i.e. time string is ',x.y') then analysis starts at + the beginning of the file. If stop time is not given (i.e. time + string is 'x.y,') then analysis goes to the end of the file. + Multiple ranges can be separated by spaces, which requires the argument + to be quoted e.g. --time "1234.567,1234.789 1235," + Time string is'a1.b1,c1.d1:a2.b2,c2.d2'. Use ':' to separate timestamps + for different perf.data files. For example, we get the timestamp information from 'perf script'. diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt index de269430720a..15e0fa87241b 100644 --- a/tools/perf/Documentation/perf-record.txt +++ b/tools/perf/Documentation/perf-record.txt @@ -490,6 +490,17 @@ Configure all used events to run in kernel space. --all-user:: Configure all used events to run in user space. +--kernel-callchains:: +Collect callchains only from kernel space. I.e. this option sets +perf_event_attr.exclude_callchain_user to 1. + +--user-callchains:: +Collect callchains only from user space. I.e. this option sets +perf_event_attr.exclude_callchain_kernel to 1. + +Don't use both --kernel-callchains and --user-callchains at the same time or no +callchains will be collected. + --timestamp-filename Append timestamp to output file name. diff --git a/tools/perf/Documentation/perf-report.txt b/tools/perf/Documentation/perf-report.txt index f441baa794ce..8c4372819e11 100644 --- a/tools/perf/Documentation/perf-report.txt +++ b/tools/perf/Documentation/perf-report.txt @@ -412,12 +412,13 @@ OPTIONS --time:: Only analyze samples within given time window: <start>,<stop>. Times - have the format seconds.microseconds. If start is not given (i.e., time + have the format seconds.nanoseconds. If start is not given (i.e. time string is ',x.y') then analysis starts at the beginning of the file. If - stop time is not given (i.e, time string is 'x.y,') then analysis goes - to end of file. + stop time is not given (i.e. time string is 'x.y,') then analysis goes + to end of file. Multiple ranges can be separated by spaces, which + requires the argument to be quoted e.g. --time "1234.567,1234.789 1235," - Also support time percent with multiple time range. Time string is + Also support time percent with multiple time ranges. Time string is 'a%/n,b%/m,...' or 'a%-b%,c%-%d,...'. For example: diff --git a/tools/perf/Documentation/perf-script.txt b/tools/perf/Documentation/perf-script.txt index af8282782911..d4e2e18a5881 100644 --- a/tools/perf/Documentation/perf-script.txt +++ b/tools/perf/Documentation/perf-script.txt @@ -117,7 +117,7 @@ OPTIONS Comma separated list of fields to print. Options are: comm, tid, pid, time, cpu, event, trace, ip, sym, dso, addr, symoff, srcline, period, iregs, uregs, brstack, brstacksym, flags, bpf-output, brstackinsn, - brstackoff, callindent, insn, insnlen, synth, phys_addr, metric, misc, srccode. + brstackoff, callindent, insn, insnlen, synth, phys_addr, metric, misc, srccode, ipc. Field list can be prepended with the type, trace, sw or hw, to indicate to which event type the field list applies. e.g., -F sw:comm,tid,time,ip,sym and -F trace:time,cpu,trace @@ -203,6 +203,9 @@ OPTIONS The synth field is used by synthesized events which may be created when Instruction Trace decoding. + The ipc (instructions per cycle) field is synthesized and may have a value when + Instruction Trace decoding. + Finally, a user may not set fields to none for all event types. i.e., -F "" is not allowed. @@ -358,12 +361,13 @@ include::itrace.txt[] --time:: Only analyze samples within given time window: <start>,<stop>. Times - have the format seconds.microseconds. If start is not given (i.e., time + have the format seconds.nanoseconds. If start is not given (i.e. time string is ',x.y') then analysis starts at the beginning of the file. If - stop time is not given (i.e, time string is 'x.y,') then analysis goes - to end of file. + stop time is not given (i.e. time string is 'x.y,') then analysis goes + to end of file. Multiple ranges can be separated by spaces, which + requires the argument to be quoted e.g. --time "1234.567,1234.789 1235," - Also support time percent with multipe time range. Time string is + Also support time percent with multiple time ranges. Time string is 'a%/n,b%/m,...' or 'a%-b%,c%-%d,...'. For example: diff --git a/tools/perf/Documentation/perf-stat.txt b/tools/perf/Documentation/perf-stat.txt index 1e312c2672e4..930c51c01201 100644 --- a/tools/perf/Documentation/perf-stat.txt +++ b/tools/perf/Documentation/perf-stat.txt @@ -200,6 +200,13 @@ use --per-socket in addition to -a. (system-wide). The output includes the socket number and the number of online processors on that socket. This is useful to gauge the amount of aggregation. +--per-die:: +Aggregate counts per processor die for system-wide mode measurements. This +is a useful mode to detect imbalance between dies. To enable this mode, +use --per-die in addition to -a. (system-wide). The output includes the +die number and the number of online processors on that die. This is +useful to gauge the amount of aggregation. + --per-core:: Aggregate counts per physical processor for system-wide mode measurements. This is a useful mode to detect imbalance between physical cores. To enable this mode, @@ -239,6 +246,9 @@ Input file name. --per-socket:: Aggregate counts per processor socket for system-wide mode measurements. +--per-die:: +Aggregate counts per processor die for system-wide mode measurements. + --per-core:: Aggregate counts per physical processor for system-wide mode measurements. diff --git a/tools/perf/Documentation/perf.data-file-format.txt b/tools/perf/Documentation/perf.data-file-format.txt index 6967e9b02be5..5f54feb19977 100644 --- a/tools/perf/Documentation/perf.data-file-format.txt +++ b/tools/perf/Documentation/perf.data-file-format.txt @@ -151,25 +151,45 @@ struct { HEADER_CPU_TOPOLOGY = 13, -String lists defining the core and CPU threads topology. -The string lists are followed by a variable length array -which contains core_id and socket_id of each cpu. -The number of entries can be determined by the size of the -section minus the sizes of both string lists. - struct { + /* + * First revision of HEADER_CPU_TOPOLOGY + * + * See 'struct perf_header_string_list' definition earlier + * in this file. + */ + struct perf_header_string_list cores; /* Variable length */ struct perf_header_string_list threads; /* Variable length */ + + /* + * Second revision of HEADER_CPU_TOPOLOGY, older tools + * will not consider what comes next + */ + struct { uint32_t core_id; uint32_t socket_id; } cpus[nr]; /* Variable length records */ + /* 'nr' comes from previously processed HEADER_NRCPUS's nr_cpu_avail */ + + /* + * Third revision of HEADER_CPU_TOPOLOGY, older tools + * will not consider what comes next + */ + + struct perf_header_string_list dies; /* Variable length */ + uint32_t die_id[nr_cpus_avail]; /* from previously processed HEADER_NR_CPUS, VLA */ }; Example: - sibling cores : 0-3 + sibling sockets : 0-8 + sibling dies : 0-3 + sibling dies : 4-7 sibling threads : 0-1 sibling threads : 2-3 + sibling threads : 4-5 + sibling threads : 6-7 HEADER_NUMA_TOPOLOGY = 14, @@ -272,6 +292,69 @@ struct { Two uint64_t for the time of first sample and the time of last sample. + HEADER_SAMPLE_TOPOLOGY = 22, + +Physical memory map and its node assignments. + +The format of data in MEM_TOPOLOGY is as follows: + + 0 - version | for future changes + 8 - block_size_bytes | /sys/devices/system/memory/block_size_bytes + 16 - count | number of nodes + +For each node we store map of physical indexes: + + 32 - node id | node index + 40 - size | size of bitmap + 48 - bitmap | bitmap of memory indexes that belongs to node + | /sys/devices/system/node/node<NODE>/memory<INDEX> + +The MEM_TOPOLOGY can be displayed with following command: + +$ perf report --header-only -I +... +# memory nodes (nr 1, block size 0x8000000): +# 0 [7G]: 0-23,32-69 + + HEADER_CLOCKID = 23, + +One uint64_t for the clockid frequency, specified, for instance, via 'perf +record -k' (see clock_gettime()), to enable timestamps derived metrics +conversion into wall clock time on the reporting stage. + + HEADER_DIR_FORMAT = 24, + +The data files layout is described by HEADER_DIR_FORMAT feature. Currently it +holds only version number (1): + + uint64_t version; + +The current version holds only version value (1) means that data files: + +- Follow the 'data.*' name format. + +- Contain raw events data in standard perf format as read from kernel (and need + to be sorted) + +Future versions are expected to describe different data files layout according +to special needs. + + HEADER_BPF_PROG_INFO = 25, + +struct bpf_prog_info_linear, which contains detailed information about +a BPF program, including type, id, tag, jited/xlated instructions, etc. + + HEADER_BPF_BTF = 26, + +Contains BPF Type Format (BTF). For more information about BTF, please +refer to Documentation/bpf/btf.rst. + +struct { + u32 id; + u32 data_size; + char data[]; +}; + HEADER_COMPRESSED = 27, struct { diff --git a/tools/perf/Makefile.config b/tools/perf/Makefile.config index 85fbcd265351..51dd00f65709 100644 --- a/tools/perf/Makefile.config +++ b/tools/perf/Makefile.config @@ -413,6 +413,9 @@ ifdef CORESIGHT $(call feature_check,libopencsd) ifeq ($(feature-libopencsd), 1) CFLAGS += -DHAVE_CSTRACE_SUPPORT $(LIBOPENCSD_CFLAGS) + ifeq ($(feature-reallocarray), 0) + CFLAGS += -DCOMPAT_NEED_REALLOCARRAY + endif LDFLAGS += $(LIBOPENCSD_LDFLAGS) EXTLIBS += $(OPENCSDLIBS) $(call detected,CONFIG_LIBOPENCSD) diff --git a/tools/perf/arch/arm/util/cs-etm.c b/tools/perf/arch/arm/util/cs-etm.c index 911426721170..279c69caef91 100644 --- a/tools/perf/arch/arm/util/cs-etm.c +++ b/tools/perf/arch/arm/util/cs-etm.c @@ -31,12 +31,159 @@ struct cs_etm_recording { struct auxtrace_record itr; struct perf_pmu *cs_etm_pmu; struct perf_evlist *evlist; + int wrapped_cnt; + bool *wrapped; bool snapshot_mode; size_t snapshot_size; }; +static const char *metadata_etmv3_ro[CS_ETM_PRIV_MAX] = { + [CS_ETM_ETMCCER] = "mgmt/etmccer", + [CS_ETM_ETMIDR] = "mgmt/etmidr", +}; + +static const char *metadata_etmv4_ro[CS_ETMV4_PRIV_MAX] = { + [CS_ETMV4_TRCIDR0] = "trcidr/trcidr0", + [CS_ETMV4_TRCIDR1] = "trcidr/trcidr1", + [CS_ETMV4_TRCIDR2] = "trcidr/trcidr2", + [CS_ETMV4_TRCIDR8] = "trcidr/trcidr8", + [CS_ETMV4_TRCAUTHSTATUS] = "mgmt/trcauthstatus", +}; + static bool cs_etm_is_etmv4(struct auxtrace_record *itr, int cpu); +static int cs_etm_set_context_id(struct auxtrace_record *itr, + struct perf_evsel *evsel, int cpu) +{ + struct cs_etm_recording *ptr; + struct perf_pmu *cs_etm_pmu; + char path[PATH_MAX]; + int err = -EINVAL; + u32 val; + + ptr = container_of(itr, struct cs_etm_recording, itr); + cs_etm_pmu = ptr->cs_etm_pmu; + + if (!cs_etm_is_etmv4(itr, cpu)) + goto out; + + /* Get a handle on TRCIRD2 */ + snprintf(path, PATH_MAX, "cpu%d/%s", + cpu, metadata_etmv4_ro[CS_ETMV4_TRCIDR2]); + err = perf_pmu__scan_file(cs_etm_pmu, path, "%x", &val); + + /* There was a problem reading the file, bailing out */ + if (err != 1) { + pr_err("%s: can't read file %s\n", + CORESIGHT_ETM_PMU_NAME, path); + goto out; + } + + /* + * TRCIDR2.CIDSIZE, bit [9-5], indicates whether contextID tracing + * is supported: + * 0b00000 Context ID tracing is not supported. + * 0b00100 Maximum of 32-bit Context ID size. + * All other values are reserved. + */ + val = BMVAL(val, 5, 9); + if (!val || val != 0x4) { + err = -EINVAL; + goto out; + } + + /* All good, let the kernel know */ + evsel->attr.config |= (1 << ETM_OPT_CTXTID); + err = 0; + +out: + + return err; +} + +static int cs_etm_set_timestamp(struct auxtrace_record *itr, + struct perf_evsel *evsel, int cpu) +{ + struct cs_etm_recording *ptr; + struct perf_pmu *cs_etm_pmu; + char path[PATH_MAX]; + int err = -EINVAL; + u32 val; + + ptr = container_of(itr, struct cs_etm_recording, itr); + cs_etm_pmu = ptr->cs_etm_pmu; + + if (!cs_etm_is_etmv4(itr, cpu)) + goto out; + + /* Get a handle on TRCIRD0 */ + snprintf(path, PATH_MAX, "cpu%d/%s", + cpu, metadata_etmv4_ro[CS_ETMV4_TRCIDR0]); + err = perf_pmu__scan_file(cs_etm_pmu, path, "%x", &val); + + /* There was a problem reading the file, bailing out */ + if (err != 1) { + pr_err("%s: can't read file %s\n", + CORESIGHT_ETM_PMU_NAME, path); + goto out; + } + + /* + * TRCIDR0.TSSIZE, bit [28-24], indicates whether global timestamping + * is supported: + * 0b00000 Global timestamping is not implemented + * 0b00110 Implementation supports a maximum timestamp of 48bits. + * 0b01000 Implementation supports a maximum timestamp of 64bits. + */ + val &= GENMASK(28, 24); + if (!val) { + err = -EINVAL; + goto out; + } + + /* All good, let the kernel know */ + evsel->attr.config |= (1 << ETM_OPT_TS); + err = 0; + +out: + return err; +} + +static int cs_etm_set_option(struct auxtrace_record *itr, + struct perf_evsel *evsel, u32 option) +{ + int i, err = -EINVAL; + struct cpu_map *event_cpus = evsel->evlist->cpus; + struct cpu_map *online_cpus = cpu_map__new(NULL); + + /* Set option of each CPU we have */ + for (i = 0; i < cpu__max_cpu(); i++) { + if (!cpu_map__has(event_cpus, i) || + !cpu_map__has(online_cpus, i)) + continue; + + switch (option) { + case ETM_OPT_CTXTID: + err = cs_etm_set_context_id(itr, evsel, i); + if (err) + goto out; + break; + case ETM_OPT_TS: + err = cs_etm_set_timestamp(itr, evsel, i); + if (err) + goto out; + break; + default: + goto out; + } + } + + err = 0; +out: + cpu_map__put(online_cpus); + return err; +} + static int cs_etm_parse_snapshot_options(struct auxtrace_record *itr, struct record_opts *opts, const char *str) @@ -105,12 +252,16 @@ static int cs_etm_recording_options(struct auxtrace_record *itr, container_of(itr, struct cs_etm_recording, itr); struct perf_pmu *cs_etm_pmu = ptr->cs_etm_pmu; struct perf_evsel *evsel, *cs_etm_evsel = NULL; - const struct cpu_map *cpus = evlist->cpus; + struct cpu_map *cpus = evlist->cpus; bool privileged = (geteuid() == 0 || perf_event_paranoid() < 0); + int err = 0; ptr->evlist = evlist; ptr->snapshot_mode = opts->auxtrace_snapshot_mode; + if (perf_can_record_switch_events()) + opts->record_switch_events = true; + evlist__for_each_entry(evlist, evsel) { if (evsel->attr.type == cs_etm_pmu->type) { if (cs_etm_evsel) { @@ -241,19 +392,28 @@ static int cs_etm_recording_options(struct auxtrace_record *itr, /* * In the case of per-cpu mmaps, we need the CPU on the - * AUX event. + * AUX event. We also need the contextID in order to be notified + * when a context switch happened. */ - if (!cpu_map__empty(cpus)) + if (!cpu_map__empty(cpus)) { perf_evsel__set_sample_bit(cs_etm_evsel, CPU); + err = cs_etm_set_option(itr, cs_etm_evsel, ETM_OPT_CTXTID); + if (err) + goto out; + + err = cs_etm_set_option(itr, cs_etm_evsel, ETM_OPT_TS); + if (err) + goto out; + } + /* Add dummy event to keep tracking */ if (opts->full_auxtrace) { struct perf_evsel *tracking_evsel; - int err; err = parse_events(evlist, "dummy:u", NULL); if (err) - return err; + goto out; tracking_evsel = perf_evlist__last(evlist); perf_evlist__set_tracking_event(evlist, tracking_evsel); @@ -266,7 +426,8 @@ static int cs_etm_recording_options(struct auxtrace_record *itr, perf_evsel__set_sample_bit(tracking_evsel, TIME); } - return 0; +out: + return err; } static u64 cs_etm_get_config(struct auxtrace_record *itr) @@ -314,6 +475,8 @@ static u64 cs_etmv4_get_config(struct auxtrace_record *itr) config_opts = cs_etm_get_config(itr); if (config_opts & BIT(ETM_OPT_CYCACC)) config |= BIT(ETM4_CFG_BIT_CYCACC); + if (config_opts & BIT(ETM_OPT_CTXTID)) + config |= BIT(ETM4_CFG_BIT_CTXTID); if (config_opts & BIT(ETM_OPT_TS)) config |= BIT(ETM4_CFG_BIT_TS); if (config_opts & BIT(ETM_OPT_RETSTK)) @@ -363,19 +526,6 @@ cs_etm_info_priv_size(struct auxtrace_record *itr __maybe_unused, (etmv3 * CS_ETMV3_PRIV_SIZE)); } -static const char *metadata_etmv3_ro[CS_ETM_PRIV_MAX] = { - [CS_ETM_ETMCCER] = "mgmt/etmccer", - [CS_ETM_ETMIDR] = "mgmt/etmidr", -}; - -static const char *metadata_etmv4_ro[CS_ETMV4_PRIV_MAX] = { - [CS_ETMV4_TRCIDR0] = "trcidr/trcidr0", - [CS_ETMV4_TRCIDR1] = "trcidr/trcidr1", - [CS_ETMV4_TRCIDR2] = "trcidr/trcidr2", - [CS_ETMV4_TRCIDR8] = "trcidr/trcidr8", - [CS_ETMV4_TRCAUTHSTATUS] = "mgmt/trcauthstatus", -}; - static bool cs_etm_is_etmv4(struct auxtrace_record *itr, int cpu) { bool ret = false; @@ -536,16 +686,131 @@ static int cs_etm_info_fill(struct auxtrace_record *itr, return 0; } -static int cs_etm_find_snapshot(struct auxtrace_record *itr __maybe_unused, +static int cs_etm_alloc_wrapped_array(struct cs_etm_recording *ptr, int idx) +{ + bool *wrapped; + int cnt = ptr->wrapped_cnt; + + /* Make @ptr->wrapped as big as @idx */ + while (cnt <= idx) + cnt++; + + /* + * Free'ed in cs_etm_recording_free(). Using realloc() to avoid + * cross compilation problems where the host's system supports + * reallocarray() but not the target. + */ + wrapped = realloc(ptr->wrapped, cnt * sizeof(bool)); + if (!wrapped) + return -ENOMEM; + + wrapped[cnt - 1] = false; + ptr->wrapped_cnt = cnt; + ptr->wrapped = wrapped; + + return 0; +} + +static bool cs_etm_buffer_has_wrapped(unsigned char *buffer, + size_t buffer_size, u64 head) +{ + u64 i, watermark; + u64 *buf = (u64 *)buffer; + size_t buf_size = buffer_size; + + /* + * We want to look the very last 512 byte (chosen arbitrarily) in + * the ring buffer. + */ + watermark = buf_size - 512; + + /* + * @head is continuously increasing - if its value is equal or greater + * than the size of the ring buffer, it has wrapped around. + */ + if (head >= buffer_size) + return true; + + /* + * The value of @head is somewhere within the size of the ring buffer. + * This can be that there hasn't been enough data to fill the ring + * buffer yet or the trace time was so long that @head has numerically + * wrapped around. To find we need to check if we have data at the very + * end of the ring buffer. We can reliably do this because mmap'ed + * pages are zeroed out and there is a fresh mapping with every new + * session. + */ + + /* @head is less than 512 byte from the end of the ring buffer */ + if (head > watermark) + watermark = head; + + /* + * Speed things up by using 64 bit transactions (see "u64 *buf" above) + */ + watermark >>= 3; + buf_size >>= 3; + + /* + * If we find trace data at the end of the ring buffer, @head has + * been there and has numerically wrapped around at least once. + */ + for (i = watermark; i < buf_size; i++) + if (buf[i]) + return true; + + return false; +} + +static int cs_etm_find_snapshot(struct auxtrace_record *itr, int idx, struct auxtrace_mmap *mm, - unsigned char *data __maybe_unused, + unsigned char *data, u64 *head, u64 *old) { + int err; + bool wrapped; + struct cs_etm_recording *ptr = + container_of(itr, struct cs_etm_recording, itr); + + /* + * Allocate memory to keep track of wrapping if this is the first + * time we deal with this *mm. + */ + if (idx >= ptr->wrapped_cnt) { + err = cs_etm_alloc_wrapped_array(ptr, idx); + if (err) + return err; + } + + /* + * Check to see if *head has wrapped around. If it hasn't only the + * amount of data between *head and *old is snapshot'ed to avoid + * bloating the perf.data file with zeros. But as soon as *head has + * wrapped around the entire size of the AUX ring buffer it taken. + */ + wrapped = ptr->wrapped[idx]; + if (!wrapped && cs_etm_buffer_has_wrapped(data, mm->len, *head)) { + wrapped = true; + ptr->wrapped[idx] = true; + } + pr_debug3("%s: mmap index %d old head %zu new head %zu size %zu\n", __func__, idx, (size_t)*old, (size_t)*head, mm->len); - *old = *head; - *head += mm->len; + /* No wrap has occurred, we can just use *head and *old. */ + if (!wrapped) + return 0; + + /* + * *head has wrapped around - adjust *head and *old to pickup the + * entire content of the AUX buffer. + */ + if (*head >= mm->len) { + *old = *head - mm->len; + } else { + *head += mm->len; + *old = *head - mm->len; + } return 0; } @@ -586,6 +851,8 @@ static void cs_etm_recording_free(struct auxtrace_record *itr) { struct cs_etm_recording *ptr = container_of(itr, struct cs_etm_recording, itr); + + zfree(&ptr->wrapped); free(ptr); } diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c index e2c3a585a61e..dca55997934e 100644 --- a/tools/perf/builtin-record.c +++ b/tools/perf/builtin-record.c @@ -2191,6 +2191,10 @@ static struct option __record_options[] = { OPT_BOOLEAN_FLAG(0, "all-user", &record.opts.all_user, "Configure all used events to run in user space.", PARSE_OPT_EXCLUSIVE), + OPT_BOOLEAN(0, "kernel-callchains", &record.opts.kernel_callchains, + "collect kernel callchains"), + OPT_BOOLEAN(0, "user-callchains", &record.opts.user_callchains, + "collect user callchains"), OPT_STRING(0, "clang-path", &llvm_param.clang_path, "clang path", "clang binary to use for compiling BPF scriptlets"), OPT_STRING(0, "clang-opt", &llvm_param.clang_opt, "clang options", diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c index 1ca533f06a4c..91c40808380d 100644 --- a/tools/perf/builtin-report.c +++ b/tools/perf/builtin-report.c @@ -1428,6 +1428,10 @@ repeat: &report.range_num); if (ret < 0) goto error; + + itrace_synth_opts__set_time_range(&itrace_synth_opts, + report.ptime_range, + report.range_num); } if (session->tevent.pevent && @@ -1449,8 +1453,10 @@ repeat: ret = 0; error: - if (report.ptime_range) + if (report.ptime_range) { + itrace_synth_opts__clear_time_range(&itrace_synth_opts); zfree(&report.ptime_range); + } zstd_fini(&(session->zstd_data)); perf_session__delete(session); return ret; diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c index 3a48a2627670..61f00055476a 100644 --- a/tools/perf/builtin-script.c +++ b/tools/perf/builtin-script.c @@ -102,6 +102,7 @@ enum perf_output_field { PERF_OUTPUT_METRIC = 1U << 28, PERF_OUTPUT_MISC = 1U << 29, PERF_OUTPUT_SRCCODE = 1U << 30, + PERF_OUTPUT_IPC = 1U << 31, }; struct output_option { @@ -139,6 +140,7 @@ struct output_option { {.str = "metric", .field = PERF_OUTPUT_METRIC}, {.str = "misc", .field = PERF_OUTPUT_MISC}, {.str = "srccode", .field = PERF_OUTPUT_SRCCODE}, + {.str = "ipc", .field = PERF_OUTPUT_IPC}, }; enum { @@ -1268,6 +1270,20 @@ static int perf_sample__fprintf_insn(struct perf_sample *sample, return printed; } +static int perf_sample__fprintf_ipc(struct perf_sample *sample, + struct perf_event_attr *attr, FILE *fp) +{ + unsigned int ipc; + + if (!PRINT_FIELD(IPC) || !sample->cyc_cnt || !sample->insn_cnt) + return 0; + + ipc = (sample->insn_cnt * 100) / sample->cyc_cnt; + + return fprintf(fp, " \t IPC: %u.%02u (%" PRIu64 "/%" PRIu64 ") ", + ipc / 100, ipc % 100, sample->insn_cnt, sample->cyc_cnt); +} + static int perf_sample__fprintf_bts(struct perf_sample *sample, struct perf_evsel *evsel, struct thread *thread, @@ -1312,6 +1328,8 @@ static int perf_sample__fprintf_bts(struct perf_sample *sample, printed += perf_sample__fprintf_addr(sample, thread, attr, fp); } + printed += perf_sample__fprintf_ipc(sample, attr, fp); + if (print_srcline_last) printed += map__fprintf_srcline(al->map, al->addr, "\n ", fp); @@ -1859,6 +1877,9 @@ static void process_event(struct perf_script *script, if (PRINT_FIELD(PHYS_ADDR)) fprintf(fp, "%16" PRIx64, sample->phys_addr); + + perf_sample__fprintf_ipc(sample, attr, fp); + fprintf(fp, "\n"); if (PRINT_FIELD(SRCCODE)) { @@ -3433,7 +3454,7 @@ int cmd_script(int argc, const char **argv) "Fields: comm,tid,pid,time,cpu,event,trace,ip,sym,dso," "addr,symoff,srcline,period,iregs,uregs,brstack," "brstacksym,flags,bpf-output,brstackinsn,brstackoff," - "callindent,insn,insnlen,synth,phys_addr,metric,misc", + "callindent,insn,insnlen,synth,phys_addr,metric,misc,ipc", parse_output_fields), OPT_BOOLEAN('a', "all-cpus", &system_wide, "system-wide collection from all CPUs"), @@ -3808,6 +3829,10 @@ int cmd_script(int argc, const char **argv) &script.range_num); if (err < 0) goto out_delete; + + itrace_synth_opts__set_time_range(&itrace_synth_opts, + script.ptime_range, + script.range_num); } err = __cmd_script(&script); @@ -3815,8 +3840,10 @@ int cmd_script(int argc, const char **argv) flush_scripting(); out_delete: - if (script.ptime_range) + if (script.ptime_range) { + itrace_synth_opts__clear_time_range(&itrace_synth_opts); zfree(&script.ptime_range); + } perf_evlist__free_stats(session->evlist); perf_session__delete(session); diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index 1ae66f09dc7d..8a35fc5a7281 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -776,6 +776,8 @@ static struct option stat_options[] = { "stop workload and print counts after a timeout period in ms (>= 10ms)"), OPT_SET_UINT(0, "per-socket", &stat_config.aggr_mode, "aggregate counts per processor socket", AGGR_SOCKET), + OPT_SET_UINT(0, "per-die", &stat_config.aggr_mode, + "aggregate counts per processor die", AGGR_DIE), OPT_SET_UINT(0, "per-core", &stat_config.aggr_mode, "aggregate counts per physical processor core", AGGR_CORE), OPT_SET_UINT(0, "per-thread", &stat_config.aggr_mode, @@ -800,6 +802,12 @@ static int perf_stat__get_socket(struct perf_stat_config *config __maybe_unused, return cpu_map__get_socket(map, cpu, NULL); } +static int perf_stat__get_die(struct perf_stat_config *config __maybe_unused, + struct cpu_map *map, int cpu) +{ + return cpu_map__get_die(map, cpu, NULL); +} + static int perf_stat__get_core(struct perf_stat_config *config __maybe_unused, struct cpu_map *map, int cpu) { @@ -840,6 +848,12 @@ static int perf_stat__get_socket_cached(struct perf_stat_config *config, return perf_stat__get_aggr(config, perf_stat__get_socket, map, idx); } +static int perf_stat__get_die_cached(struct perf_stat_config *config, + struct cpu_map *map, int idx) +{ + return perf_stat__get_aggr(config, perf_stat__get_die, map, idx); +} + static int perf_stat__get_core_cached(struct perf_stat_config *config, struct cpu_map *map, int idx) { @@ -870,6 +884,13 @@ static int perf_stat_init_aggr_mode(void) } stat_config.aggr_get_id = perf_stat__get_socket_cached; break; + case AGGR_DIE: + if (cpu_map__build_die_map(evsel_list->cpus, &stat_config.aggr_map)) { + perror("cannot build die map"); + return -1; + } + stat_config.aggr_get_id = perf_stat__get_die_cached; + break; case AGGR_CORE: if (cpu_map__build_core_map(evsel_list->cpus, &stat_config.aggr_map)) { perror("cannot build core map"); @@ -935,21 +956,55 @@ static int perf_env__get_socket(struct cpu_map *map, int idx, void *data) return cpu == -1 ? -1 : env->cpu[cpu].socket_id; } +static int perf_env__get_die(struct cpu_map *map, int idx, void *data) +{ + struct perf_env *env = data; + int die_id = -1, cpu = perf_env__get_cpu(env, map, idx); + + if (cpu != -1) { + /* + * Encode socket in bit range 15:8 + * die_id is relative to socket, + * we need a global id. So we combine + * socket + die id + */ + if (WARN_ONCE(env->cpu[cpu].socket_id >> 8, "The socket id number is too big.\n")) + return -1; + + if (WARN_ONCE(env->cpu[cpu].die_id >> 8, "The die id number is too big.\n")) + return -1; + + die_id = (env->cpu[cpu].socket_id << 8) | (env->cpu[cpu].die_id & 0xff); + } + + return die_id; +} + static int perf_env__get_core(struct cpu_map *map, int idx, void *data) { struct perf_env *env = data; int core = -1, cpu = perf_env__get_cpu(env, map, idx); if (cpu != -1) { - int socket_id = env->cpu[cpu].socket_id; - /* - * Encode socket in upper 16 bits - * core_id is relative to socket, and + * Encode socket in bit range 31:24 + * encode die id in bit range 23:16 + * core_id is relative to socket and die, * we need a global id. So we combine - * socket + core id. + * socket + die id + core id */ - core = (socket_id << 16) | (env->cpu[cpu].core_id & 0xffff); + if (WARN_ONCE(env->cpu[cpu].socket_id >> 8, "The socket id number is too big.\n")) + return -1; + + if (WARN_ONCE(env->cpu[cpu].die_id >> 8, "The die id number is too big.\n")) + return -1; + + if (WARN_ONCE(env->cpu[cpu].core_id >> 16, "The core id number is too big.\n")) + return -1; + + core = (env->cpu[cpu].socket_id << 24) | + (env->cpu[cpu].die_id << 16) | + (env->cpu[cpu].core_id & 0xffff); } return core; @@ -961,6 +1016,12 @@ static int perf_env__build_socket_map(struct perf_env *env, struct cpu_map *cpus return cpu_map__build_map(cpus, sockp, perf_env__get_socket, env); } +static int perf_env__build_die_map(struct perf_env *env, struct cpu_map *cpus, + struct cpu_map **diep) +{ + return cpu_map__build_map(cpus, diep, perf_env__get_die, env); +} + static int perf_env__build_core_map(struct perf_env *env, struct cpu_map *cpus, struct cpu_map **corep) { @@ -972,6 +1033,11 @@ static int perf_stat__get_socket_file(struct perf_stat_config *config __maybe_un { return perf_env__get_socket(map, idx, &perf_stat.session->header.env); } +static int perf_stat__get_die_file(struct perf_stat_config *config __maybe_unused, + struct cpu_map *map, int idx) +{ + return perf_env__get_die(map, idx, &perf_stat.session->header.env); +} static int perf_stat__get_core_file(struct perf_stat_config *config __maybe_unused, struct cpu_map *map, int idx) @@ -991,6 +1057,13 @@ static int perf_stat_init_aggr_mode_file(struct perf_stat *st) } stat_config.aggr_get_id = perf_stat__get_socket_file; break; + case AGGR_DIE: + if (perf_env__build_die_map(env, evsel_list->cpus, &stat_config.aggr_map)) { + perror("cannot build die map"); + return -1; + } + stat_config.aggr_get_id = perf_stat__get_die_file; + break; case AGGR_CORE: if (perf_env__build_core_map(env, evsel_list->cpus, &stat_config.aggr_map)) { perror("cannot build core map"); @@ -1541,6 +1614,8 @@ static int __cmd_report(int argc, const char **argv) OPT_STRING('i', "input", &input_name, "file", "input file name"), OPT_SET_UINT(0, "per-socket", &perf_stat.aggr_mode, "aggregate counts per processor socket", AGGR_SOCKET), + OPT_SET_UINT(0, "per-die", &perf_stat.aggr_mode, + "aggregate counts per processor die", AGGR_DIE), OPT_SET_UINT(0, "per-core", &perf_stat.aggr_mode, "aggregate counts per physical processor core", AGGR_CORE), OPT_SET_UINT('A', "no-aggr", &perf_stat.aggr_mode, diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c index ff17c4a1b7f8..12f5ad98f8c1 100644 --- a/tools/perf/builtin-trace.c +++ b/tools/perf/builtin-trace.c @@ -971,8 +971,14 @@ struct syscall { struct syscall_arg_fmt *arg_fmt; }; +/* + * Must match what is in the BPF program: + * + * tools/perf/examples/bpf/augmented_raw_syscalls.c + */ struct bpf_map_syscall_entry { bool enabled; + u16 string_args_len[6]; }; /* @@ -1226,8 +1232,17 @@ static void thread__set_filename_pos(struct thread *thread, const char *bf, static size_t syscall_arg__scnprintf_augmented_string(struct syscall_arg *arg, char *bf, size_t size) { struct augmented_arg *augmented_arg = arg->augmented.args; + size_t printed = scnprintf(bf, size, "\"%.*s\"", augmented_arg->size, augmented_arg->value); + /* + * So that the next arg with a payload can consume its augmented arg, i.e. for rename* syscalls + * we would have two strings, each prefixed by its size. + */ + int consumed = sizeof(*augmented_arg) + augmented_arg->size; + + arg->augmented.args += consumed; + arg->augmented.size -= consumed; - return scnprintf(bf, size, "\"%.*s\"", augmented_arg->size, augmented_arg->value); + return printed; } static size_t syscall_arg__scnprintf_filename(char *bf, size_t size, @@ -1415,10 +1430,11 @@ static int syscall__set_arg_fmts(struct syscall *sc) if (sc->fmt && sc->fmt->arg[idx].scnprintf) continue; + len = strlen(field->name); + if (strcmp(field->type, "const char *") == 0 && - (strcmp(field->name, "filename") == 0 || - strcmp(field->name, "path") == 0 || - strcmp(field->name, "pathname") == 0)) + ((len >= 4 && strcmp(field->name + len - 4, "name") == 0) || + strstr(field->name, "path") != NULL)) sc->arg_fmt[idx].scnprintf = SCA_FILENAME; else if ((field->flags & TEP_FIELD_IS_POINTER) || strstr(field->name, "addr")) sc->arg_fmt[idx].scnprintf = SCA_PTR; @@ -1429,8 +1445,7 @@ static int syscall__set_arg_fmts(struct syscall *sc) else if ((strcmp(field->type, "int") == 0 || strcmp(field->type, "unsigned int") == 0 || strcmp(field->type, "long") == 0) && - (len = strlen(field->name)) >= 2 && - strcmp(field->name + len - 2, "fd") == 0) { + len >= 2 && strcmp(field->name + len - 2, "fd") == 0) { /* * /sys/kernel/tracing/events/syscalls/sys_enter* * egrep 'field:.*fd;' .../format|sed -r 's/.*field:([a-z ]+) [a-z_]*fd.+/\1/g'|sort|uniq -c @@ -1513,6 +1528,7 @@ static int trace__read_syscall_info(struct trace *trace, int id) static int trace__validate_ev_qualifier(struct trace *trace) { int err = 0, i; + bool printed_invalid_prefix = false; size_t nr_allocated; struct str_node *pos; @@ -1539,14 +1555,15 @@ static int trace__validate_ev_qualifier(struct trace *trace) if (id >= 0) goto matches; - if (err == 0) { - fputs("Error:\tInvalid syscall ", trace->output); - err = -EINVAL; + if (!printed_invalid_prefix) { + pr_debug("Skipping unknown syscalls: "); + printed_invalid_prefix = true; } else { - fputs(", ", trace->output); + pr_debug(", "); } - fputs(sc, trace->output); + pr_debug("%s", sc); + continue; } matches: trace->ev_qualifier_ids.entries[i++] = id; @@ -1575,15 +1592,14 @@ matches: } } - if (err < 0) { - fputs("\nHint:\ttry 'perf list syscalls:sys_enter_*'" - "\nHint:\tand: 'man syscalls'\n", trace->output); -out_free: - zfree(&trace->ev_qualifier_ids.entries); - trace->ev_qualifier_ids.nr = 0; - } out: + if (printed_invalid_prefix) + pr_debug("\n"); return err; +out_free: + zfree(&trace->ev_qualifier_ids.entries); + trace->ev_qualifier_ids.nr = 0; + goto out; } /* @@ -2710,6 +2726,25 @@ out_enomem: } #ifdef HAVE_LIBBPF_SUPPORT +static void trace__init_bpf_map_syscall_args(struct trace *trace, int id, struct bpf_map_syscall_entry *entry) +{ + struct syscall *sc = trace__syscall_info(trace, NULL, id); + int arg = 0; + + if (sc == NULL) + goto out; + + for (; arg < sc->nr_args; ++arg) { + entry->string_args_len[arg] = 0; + if (sc->arg_fmt[arg].scnprintf == SCA_FILENAME) { + /* Should be set like strace -s strsize */ + entry->string_args_len[arg] = PATH_MAX; + } + } +out: + for (; arg < 6; ++arg) + entry->string_args_len[arg] = 0; +} static int trace__set_ev_qualifier_bpf_filter(struct trace *trace) { int fd = bpf_map__fd(trace->syscalls.map); @@ -2722,6 +2757,9 @@ static int trace__set_ev_qualifier_bpf_filter(struct trace *trace) for (i = 0; i < trace->ev_qualifier_ids.nr; ++i) { int key = trace->ev_qualifier_ids.entries[i]; + if (value.enabled) + trace__init_bpf_map_syscall_args(trace, key, &value); + err = bpf_map_update_elem(fd, &key, &value, BPF_EXIST); if (err) break; @@ -2739,6 +2777,9 @@ static int __trace__init_syscalls_bpf_map(struct trace *trace, bool enabled) int err = 0, key; for (key = 0; key < trace->sctbl->syscalls.nr_entries; ++key) { + if (enabled) + trace__init_bpf_map_syscall_args(trace, key, &value); + err = bpf_map_update_elem(fd, &key, &value, BPF_ANY); if (err) break; @@ -3662,7 +3703,12 @@ static int trace__config(const char *var, const char *value, void *arg) struct option o = OPT_CALLBACK('e', "event", &trace->evlist, "event", "event selector. use 'perf list' to list available events", parse_events_option); - err = parse_events_option(&o, value, 0); + /* + * We can't propagate parse_event_option() return, as it is 1 + * for failure while perf_config() expects -1. + */ + if (parse_events_option(&o, value, 0)) + err = -1; } else if (!strcmp(var, "trace.show_timestamp")) { trace->show_tstamp = perf_config_bool(var, value); } else if (!strcmp(var, "trace.show_duration")) { diff --git a/tools/perf/examples/bpf/augmented_raw_syscalls.c b/tools/perf/examples/bpf/augmented_raw_syscalls.c index 68a3d61752ce..2f822bb51717 100644 --- a/tools/perf/examples/bpf/augmented_raw_syscalls.c +++ b/tools/perf/examples/bpf/augmented_raw_syscalls.c @@ -21,8 +21,14 @@ /* bpf-output associated map */ bpf_map(__augmented_syscalls__, PERF_EVENT_ARRAY, int, u32, __NR_CPUS__); +/* + * string_args_len: one per syscall arg, 0 means not a string or don't copy it, + * PATH_MAX for copying everything, any other value to limit + * it a la 'strace -s strsize'. + */ struct syscall { bool enabled; + u16 string_args_len[6]; }; bpf_map(syscalls, ARRAY, int, struct syscall, 512); @@ -41,85 +47,10 @@ struct syscall_exit_args { struct augmented_filename { unsigned int size; - int reserved; + int err; char value[PATH_MAX]; }; -/* syscalls where the first arg is a string */ -#define SYS_OPEN 2 -#define SYS_STAT 4 -#define SYS_LSTAT 6 -#define SYS_ACCESS 21 -#define SYS_EXECVE 59 -#define SYS_TRUNCATE 76 -#define SYS_CHDIR 80 -#define SYS_RENAME 82 -#define SYS_MKDIR 83 -#define SYS_RMDIR 84 -#define SYS_CREAT 85 -#define SYS_LINK 86 -#define SYS_UNLINK 87 -#define SYS_SYMLINK 88 -#define SYS_READLINK 89 -#define SYS_CHMOD 90 -#define SYS_CHOWN 92 -#define SYS_LCHOWN 94 -#define SYS_MKNOD 133 -#define SYS_STATFS 137 -#define SYS_PIVOT_ROOT 155 -#define SYS_CHROOT 161 -#define SYS_ACCT 163 -#define SYS_SWAPON 167 -#define SYS_SWAPOFF 168 -#define SYS_DELETE_MODULE 176 -#define SYS_SETXATTR 188 -#define SYS_LSETXATTR 189 -#define SYS_GETXATTR 191 -#define SYS_LGETXATTR 192 -#define SYS_LISTXATTR 194 -#define SYS_LLISTXATTR 195 -#define SYS_REMOVEXATTR 197 -#define SYS_LREMOVEXATTR 198 -#define SYS_MQ_OPEN 240 -#define SYS_MQ_UNLINK 241 -#define SYS_ADD_KEY 248 -#define SYS_REQUEST_KEY 249 -#define SYS_SYMLINKAT 266 -#define SYS_MEMFD_CREATE 319 - -/* syscalls where the second arg is a string */ - -#define SYS_PWRITE64 18 -#define SYS_EXECVE 59 -#define SYS_RENAME 82 -#define SYS_QUOTACTL 179 -#define SYS_FSETXATTR 190 -#define SYS_FGETXATTR 193 -#define SYS_FREMOVEXATTR 199 -#define SYS_MQ_TIMEDSEND 242 -#define SYS_REQUEST_KEY 249 -#define SYS_INOTIFY_ADD_WATCH 254 -#define SYS_OPENAT 257 -#define SYS_MKDIRAT 258 -#define SYS_MKNODAT 259 -#define SYS_FCHOWNAT 260 -#define SYS_FUTIMESAT 261 -#define SYS_NEWFSTATAT 262 -#define SYS_UNLINKAT 263 -#define SYS_RENAMEAT 264 -#define SYS_LINKAT 265 -#define SYS_READLINKAT 267 -#define SYS_FCHMODAT 268 -#define SYS_FACCESSAT 269 -#define SYS_UTIMENSAT 280 -#define SYS_NAME_TO_HANDLE_AT 303 -#define SYS_FINIT_MODULE 313 -#define SYS_RENAMEAT2 316 -#define SYS_EXECVEAT 322 -#define SYS_STATX 332 -#define SYS_MOVE_MOUNT 429 -#define SYS_FSPICK 433 - pid_filter(pids_filtered); struct augmented_args_filename { @@ -129,12 +60,48 @@ struct augmented_args_filename { bpf_map(augmented_filename_map, PERCPU_ARRAY, int, struct augmented_args_filename, 1); +static inline +unsigned int augmented_filename__read(struct augmented_filename *augmented_filename, + const void *filename_arg, unsigned int filename_len) +{ + unsigned int len = sizeof(*augmented_filename); + int size = probe_read_str(&augmented_filename->value, filename_len, filename_arg); + + augmented_filename->size = augmented_filename->err = 0; + /* + * probe_read_str may return < 0, e.g. -EFAULT + * So we leave that in the augmented_filename->size that userspace will + */ + if (size > 0) { + len -= sizeof(augmented_filename->value) - size; + len &= sizeof(augmented_filename->value) - 1; + augmented_filename->size = size; + } else { + /* + * So that username notice the error while still being able + * to skip this augmented arg record + */ + augmented_filename->err = size; + len = offsetof(struct augmented_filename, value); + } + + return len; +} + SEC("raw_syscalls:sys_enter") int sys_enter(struct syscall_enter_args *args) { struct augmented_args_filename *augmented_args; - unsigned int len = sizeof(*augmented_args); - const void *filename_arg = NULL; + /* + * We start len, the amount of data that will be in the perf ring + * buffer, if this is not filtered out by one of pid_filter__has(), + * syscall->enabled, etc, with the non-augmented raw syscall payload, + * i.e. sizeof(augmented_args->args). + * + * We'll add to this as we add augmented syscalls right after that + * initial, non-augmented raw_syscalls:sys_enter payload. + */ + unsigned int len = sizeof(augmented_args->args); struct syscall *syscall; int key = 0; @@ -191,113 +158,67 @@ int sys_enter(struct syscall_enter_args *args) * after the ctx memory access to prevent their down stream merging. */ /* - * This table of what args are strings will be provided by userspace, - * in the syscalls map, i.e. we will already have to do the lookup to - * see if this specific syscall is filtered, so we can as well get more - * info about what syscall args are strings or pointers, and how many - * bytes to copy, per arg, etc. + * For now copy just the first string arg, we need to improve the protocol + * and have more than one. * - * For now hard code it, till we have all the basic mechanisms in place - * to automate everything and make the kernel part be completely driven - * by information obtained in userspace for each kernel version and - * processor architecture, making the kernel part the same no matter what - * kernel version or processor architecture it runs on. - */ - switch (augmented_args->args.syscall_nr) { - case SYS_ACCT: - case SYS_ADD_KEY: - case SYS_CHDIR: - case SYS_CHMOD: - case SYS_CHOWN: - case SYS_CHROOT: - case SYS_CREAT: - case SYS_DELETE_MODULE: - case SYS_EXECVE: - case SYS_GETXATTR: - case SYS_LCHOWN: - case SYS_LGETXATTR: - case SYS_LINK: - case SYS_LISTXATTR: - case SYS_LLISTXATTR: - case SYS_LREMOVEXATTR: - case SYS_LSETXATTR: - case SYS_LSTAT: - case SYS_MEMFD_CREATE: - case SYS_MKDIR: - case SYS_MKNOD: - case SYS_MQ_OPEN: - case SYS_MQ_UNLINK: - case SYS_PIVOT_ROOT: - case SYS_READLINK: - case SYS_REMOVEXATTR: - case SYS_RENAME: - case SYS_REQUEST_KEY: - case SYS_RMDIR: - case SYS_SETXATTR: - case SYS_STAT: - case SYS_STATFS: - case SYS_SWAPOFF: - case SYS_SWAPON: - case SYS_SYMLINK: - case SYS_SYMLINKAT: - case SYS_TRUNCATE: - case SYS_UNLINK: - case SYS_ACCESS: - case SYS_OPEN: filename_arg = (const void *)args->args[0]; + * Using the unrolled loop is not working, only when we do it manually, + * check this out later... + + u8 arg; +#pragma clang loop unroll(full) + for (arg = 0; arg < 6; ++arg) { + if (syscall->string_args_len[arg] != 0) { + filename_len = syscall->string_args_len[arg]; + filename_arg = (const void *)args->args[arg]; __asm__ __volatile__("": : :"memory"); - break; - case SYS_EXECVEAT: - case SYS_FACCESSAT: - case SYS_FCHMODAT: - case SYS_FCHOWNAT: - case SYS_FGETXATTR: - case SYS_FINIT_MODULE: - case SYS_FREMOVEXATTR: - case SYS_FSETXATTR: - case SYS_FSPICK: - case SYS_FUTIMESAT: - case SYS_INOTIFY_ADD_WATCH: - case SYS_LINKAT: - case SYS_MKDIRAT: - case SYS_MKNODAT: - // case SYS_MOVE_MOUNT: - // For now don't copy move_mount first string arg, as it has two and - // 'perf trace's syscall_arg__scnprintf_filename() will use the one - // copied here, the first, for both args, duplicating the first and - // ignoring the second. - // - // We need to copy both here and make syscall_arg__scnprintf_filename - // skip the first when reading the second, using the size of the first, etc. - // Shouldn't be difficult, but now its perf/urgent time, lets wait for - // the next devel window. - case SYS_MQ_TIMEDSEND: - case SYS_NAME_TO_HANDLE_AT: - case SYS_NEWFSTATAT: - case SYS_PWRITE64: - case SYS_QUOTACTL: - case SYS_READLINKAT: - case SYS_RENAMEAT: - case SYS_RENAMEAT2: - case SYS_STATX: - case SYS_UNLINKAT: - case SYS_UTIMENSAT: - case SYS_OPENAT: filename_arg = (const void *)args->args[1]; - break; - } - - if (filename_arg != NULL) { - augmented_args->filename.reserved = 0; - augmented_args->filename.size = probe_read_str(&augmented_args->filename.value, - sizeof(augmented_args->filename.value), - filename_arg); - if (augmented_args->filename.size < sizeof(augmented_args->filename.value)) { - len -= sizeof(augmented_args->filename.value) - augmented_args->filename.size; - len &= sizeof(augmented_args->filename.value) - 1; + break; } - } else { - len = sizeof(augmented_args->args); } + verifier log: + +; if (syscall->string_args_len[arg] != 0) { +37: (69) r3 = *(u16 *)(r0 +2) + R0=map_value(id=0,off=0,ks=4,vs=14,imm=0) R1_w=inv0 R2_w=map_value(id=0,off=2,ks=4,vs=14,imm=0) R6=ctx(id=0,off=0,imm=0) R7=map_value(id=0,off=0,ks=4,vs=4168,imm=0) R10=fp0,call_-1 fp-8=mmmmmmmm +; if (syscall->string_args_len[arg] != 0) { +38: (55) if r3 != 0x0 goto pc+5 + R0=map_value(id=0,off=0,ks=4,vs=14,imm=0) R1=inv0 R2=map_value(id=0,off=2,ks=4,vs=14,imm=0) R3=inv0 R6=ctx(id=0,off=0,imm=0) R7=map_value(id=0,off=0,ks=4,vs=4168,imm=0) R10=fp0,call_-1 fp-8=mmmmmmmm +39: (b7) r1 = 1 +; if (syscall->string_args_len[arg] != 0) { +40: (bf) r2 = r0 +41: (07) r2 += 4 +42: (69) r3 = *(u16 *)(r0 +4) + R0=map_value(id=0,off=0,ks=4,vs=14,imm=0) R1_w=inv1 R2_w=map_value(id=0,off=4,ks=4,vs=14,imm=0) R3_w=inv0 R6=ctx(id=0,off=0,imm=0) R7=map_value(id=0,off=0,ks=4,vs=4168,imm=0) R10=fp0,call_-1 fp-8=mmmmmmmm +; if (syscall->string_args_len[arg] != 0) { +43: (15) if r3 == 0x0 goto pc+32 + R0=map_value(id=0,off=0,ks=4,vs=14,imm=0) R1=inv1 R2=map_value(id=0,off=4,ks=4,vs=14,imm=0) R3=inv(id=0,umax_value=65535,var_off=(0x0; 0xffff)) R6=ctx(id=0,off=0,imm=0) R7=map_value(id=0,off=0,ks=4,vs=4168,imm=0) R10=fp0,call_-1 fp-8=mmmmmmmm +; filename_arg = (const void *)args->args[arg]; +44: (67) r1 <<= 3 +45: (bf) r3 = r6 +46: (0f) r3 += r1 +47: (b7) r5 = 64 +48: (79) r3 = *(u64 *)(r3 +16) +dereference of modified ctx ptr R3 off=8 disallowed +processed 46 insns (limit 1000000) max_states_per_insn 0 total_states 12 peak_states 12 mark_read 7 + */ + +#define __loop_iter(arg) \ + if (syscall->string_args_len[arg] != 0) { \ + unsigned int filename_len = syscall->string_args_len[arg]; \ + const void *filename_arg = (const void *)args->args[arg]; \ + if (filename_len <= sizeof(augmented_args->filename.value)) \ + len += augmented_filename__read(&augmented_args->filename, filename_arg, filename_len); +#define loop_iter_first() __loop_iter(0); } +#define loop_iter(arg) else __loop_iter(arg); } +#define loop_iter_last(arg) else __loop_iter(arg); __asm__ __volatile__("": : :"memory"); } + + loop_iter_first() + loop_iter(1) + loop_iter(2) + loop_iter(3) + loop_iter(4) + loop_iter_last(5) + /* If perf_event_output fails, return non-zero so that it gets recorded unaugmented */ return perf_event_output(args, &__augmented_syscalls__, BPF_F_CURRENT_CPU, augmented_args, len); } diff --git a/tools/perf/jvmti/libjvmti.c b/tools/perf/jvmti/libjvmti.c index aea7b1fe85aa..c441a34cb1c0 100644 --- a/tools/perf/jvmti/libjvmti.c +++ b/tools/perf/jvmti/libjvmti.c @@ -1,5 +1,6 @@ // SPDX-License-Identifier: GPL-2.0 #include <linux/compiler.h> +#include <linux/string.h> #include <sys/types.h> #include <stdio.h> #include <string.h> @@ -162,8 +163,7 @@ copy_class_filename(const char * class_sign, const char * file_name, char * resu result[i] = '\0'; } else { /* fallback case */ - size_t file_name_len = strlen(file_name); - strncpy(result, file_name, file_name_len < max_length ? file_name_len : max_length); + strlcpy(result, file_name, max_length); } } diff --git a/tools/perf/perf.h b/tools/perf/perf.h index d59dee61b64d..711e009381ec 100644 --- a/tools/perf/perf.h +++ b/tools/perf/perf.h @@ -61,6 +61,8 @@ struct record_opts { bool record_switch_events; bool all_kernel; bool all_user; + bool kernel_callchains; + bool user_callchains; bool tail_synthesize; bool overwrite; bool ignore_missing_thread; diff --git a/tools/perf/scripts/python/export-to-postgresql.py b/tools/perf/scripts/python/export-to-postgresql.py index b2f481b0d28d..93225c02117e 100644 --- a/tools/perf/scripts/python/export-to-postgresql.py +++ b/tools/perf/scripts/python/export-to-postgresql.py @@ -394,7 +394,9 @@ if branches: 'to_ip bigint,' 'branch_type integer,' 'in_tx boolean,' - 'call_path_id bigint)') + 'call_path_id bigint,' + 'insn_count bigint,' + 'cyc_count bigint)') else: do_query(query, 'CREATE TABLE samples (' 'id bigint NOT NULL,' @@ -418,7 +420,9 @@ else: 'data_src bigint,' 'branch_type integer,' 'in_tx boolean,' - 'call_path_id bigint)') + 'call_path_id bigint,' + 'insn_count bigint,' + 'cyc_count bigint)') if perf_db_export_calls or perf_db_export_callchains: do_query(query, 'CREATE TABLE call_paths (' @@ -439,7 +443,9 @@ if perf_db_export_calls: 'return_id bigint,' 'parent_call_path_id bigint,' 'flags integer,' - 'parent_id bigint)') + 'parent_id bigint,' + 'insn_count bigint,' + 'cyc_count bigint)') do_query(query, 'CREATE VIEW machines_view AS ' 'SELECT ' @@ -521,6 +527,9 @@ if perf_db_export_calls: 'return_time,' 'return_time - call_time AS elapsed_time,' 'branch_count,' + 'insn_count,' + 'cyc_count,' + 'CASE WHEN cyc_count=0 THEN CAST(0 AS NUMERIC(20, 2)) ELSE CAST((CAST(insn_count AS FLOAT) / cyc_count) AS NUMERIC(20, 2)) END AS IPC,' 'call_id,' 'return_id,' 'CASE WHEN flags=0 THEN \'\' WHEN flags=1 THEN \'no call\' WHEN flags=2 THEN \'no return\' WHEN flags=3 THEN \'no call/return\' WHEN flags=6 THEN \'jump\' ELSE CAST ( flags AS VARCHAR(6) ) END AS flags,' @@ -546,7 +555,10 @@ do_query(query, 'CREATE VIEW samples_view AS ' 'to_sym_offset,' '(SELECT short_name FROM dsos WHERE id = to_dso_id) AS to_dso_short_name,' '(SELECT name FROM branch_types WHERE id = branch_type) AS branch_type_name,' - 'in_tx' + 'in_tx,' + 'insn_count,' + 'cyc_count,' + 'CASE WHEN cyc_count=0 THEN CAST(0 AS NUMERIC(20, 2)) ELSE CAST((CAST(insn_count AS FLOAT) / cyc_count) AS NUMERIC(20, 2)) END AS IPC' ' FROM samples') @@ -618,10 +630,10 @@ def trace_begin(): comm_table(0, "unknown") dso_table(0, 0, "unknown", "unknown", "") symbol_table(0, 0, 0, 0, 0, "unknown") - sample_table(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) + sample_table(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) if perf_db_export_calls or perf_db_export_callchains: call_path_table(0, 0, 0, 0) - call_return_table(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) + call_return_table(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) unhandled_count = 0 @@ -772,11 +784,11 @@ def branch_type_table(branch_type, name, *x): value = struct.pack(fmt, 2, 4, branch_type, n, name) branch_type_file.write(value) -def sample_table(sample_id, evsel_id, machine_id, thread_id, comm_id, dso_id, symbol_id, sym_offset, ip, time, cpu, to_dso_id, to_symbol_id, to_sym_offset, to_ip, period, weight, transaction, data_src, branch_type, in_tx, call_path_id, *x): +def sample_table(sample_id, evsel_id, machine_id, thread_id, comm_id, dso_id, symbol_id, sym_offset, ip, time, cpu, to_dso_id, to_symbol_id, to_sym_offset, to_ip, period, weight, transaction, data_src, branch_type, in_tx, call_path_id, insn_cnt, cyc_cnt, *x): if branches: - value = struct.pack("!hiqiqiqiqiqiqiqiqiqiqiiiqiqiqiqiiiBiq", 18, 8, sample_id, 8, evsel_id, 8, machine_id, 8, thread_id, 8, comm_id, 8, dso_id, 8, symbol_id, 8, sym_offset, 8, ip, 8, time, 4, cpu, 8, to_dso_id, 8, to_symbol_id, 8, to_sym_offset, 8, to_ip, 4, branch_type, 1, in_tx, 8, call_path_id) + value = struct.pack("!hiqiqiqiqiqiqiqiqiqiqiiiqiqiqiqiiiBiqiqiq", 20, 8, sample_id, 8, evsel_id, 8, machine_id, 8, thread_id, 8, comm_id, 8, dso_id, 8, symbol_id, 8, sym_offset, 8, ip, 8, time, 4, cpu, 8, to_dso_id, 8, to_symbol_id, 8, to_sym_offset, 8, to_ip, 4, branch_type, 1, in_tx, 8, call_path_id, 8, insn_cnt, 8, cyc_cnt) else: - value = struct.pack("!hiqiqiqiqiqiqiqiqiqiqiiiqiqiqiqiqiqiqiqiiiBiq", 22, 8, sample_id, 8, evsel_id, 8, machine_id, 8, thread_id, 8, comm_id, 8, dso_id, 8, symbol_id, 8, sym_offset, 8, ip, 8, time, 4, cpu, 8, to_dso_id, 8, to_symbol_id, 8, to_sym_offset, 8, to_ip, 8, period, 8, weight, 8, transaction, 8, data_src, 4, branch_type, 1, in_tx, 8, call_path_id) + value = struct.pack("!hiqiqiqiqiqiqiqiqiqiqiiiqiqiqiqiqiqiqiqiiiBiqiqiq", 24, 8, sample_id, 8, evsel_id, 8, machine_id, 8, thread_id, 8, comm_id, 8, dso_id, 8, symbol_id, 8, sym_offset, 8, ip, 8, time, 4, cpu, 8, to_dso_id, 8, to_symbol_id, 8, to_sym_offset, 8, to_ip, 8, period, 8, weight, 8, transaction, 8, data_src, 4, branch_type, 1, in_tx, 8, call_path_id, 8, insn_cnt, 8, cyc_cnt) sample_file.write(value) def call_path_table(cp_id, parent_id, symbol_id, ip, *x): @@ -784,7 +796,7 @@ def call_path_table(cp_id, parent_id, symbol_id, ip, *x): value = struct.pack(fmt, 4, 8, cp_id, 8, parent_id, 8, symbol_id, 8, ip) call_path_file.write(value) -def call_return_table(cr_id, thread_id, comm_id, call_path_id, call_time, return_time, branch_count, call_id, return_id, parent_call_path_id, flags, parent_id, *x): - fmt = "!hiqiqiqiqiqiqiqiqiqiqiiiq" - value = struct.pack(fmt, 12, 8, cr_id, 8, thread_id, 8, comm_id, 8, call_path_id, 8, call_time, 8, return_time, 8, branch_count, 8, call_id, 8, return_id, 8, parent_call_path_id, 4, flags, 8, parent_id) +def call_return_table(cr_id, thread_id, comm_id, call_path_id, call_time, return_time, branch_count, call_id, return_id, parent_call_path_id, flags, parent_id, insn_cnt, cyc_cnt, *x): + fmt = "!hiqiqiqiqiqiqiqiqiqiqiiiqiqiq" + value = struct.pack(fmt, 14, 8, cr_id, 8, thread_id, 8, comm_id, 8, call_path_id, 8, call_time, 8, return_time, 8, branch_count, 8, call_id, 8, return_id, 8, parent_call_path_id, 4, flags, 8, parent_id, 8, insn_cnt, 8, cyc_cnt) call_file.write(value) diff --git a/tools/perf/scripts/python/export-to-sqlite.py b/tools/perf/scripts/python/export-to-sqlite.py index f617e518332f..4542ce89034b 100644 --- a/tools/perf/scripts/python/export-to-sqlite.py +++ b/tools/perf/scripts/python/export-to-sqlite.py @@ -218,7 +218,9 @@ if branches: 'to_ip bigint,' 'branch_type integer,' 'in_tx boolean,' - 'call_path_id bigint)') + 'call_path_id bigint,' + 'insn_count bigint,' + 'cyc_count bigint)') else: do_query(query, 'CREATE TABLE samples (' 'id integer NOT NULL PRIMARY KEY,' @@ -242,7 +244,9 @@ else: 'data_src bigint,' 'branch_type integer,' 'in_tx boolean,' - 'call_path_id bigint)') + 'call_path_id bigint,' + 'insn_count bigint,' + 'cyc_count bigint)') if perf_db_export_calls or perf_db_export_callchains: do_query(query, 'CREATE TABLE call_paths (' @@ -263,7 +267,9 @@ if perf_db_export_calls: 'return_id bigint,' 'parent_call_path_id bigint,' 'flags integer,' - 'parent_id bigint)') + 'parent_id bigint,' + 'insn_count bigint,' + 'cyc_count bigint)') # printf was added to sqlite in version 3.8.3 sqlite_has_printf = False @@ -359,6 +365,9 @@ if perf_db_export_calls: 'return_time,' 'return_time - call_time AS elapsed_time,' 'branch_count,' + 'insn_count,' + 'cyc_count,' + 'CASE WHEN cyc_count=0 THEN CAST(0 AS FLOAT) ELSE ROUND(CAST(insn_count AS FLOAT) / cyc_count, 2) END AS IPC,' 'call_id,' 'return_id,' 'CASE WHEN flags=0 THEN \'\' WHEN flags=1 THEN \'no call\' WHEN flags=2 THEN \'no return\' WHEN flags=3 THEN \'no call/return\' WHEN flags=6 THEN \'jump\' ELSE flags END AS flags,' @@ -384,7 +393,10 @@ do_query(query, 'CREATE VIEW samples_view AS ' 'to_sym_offset,' '(SELECT short_name FROM dsos WHERE id = to_dso_id) AS to_dso_short_name,' '(SELECT name FROM branch_types WHERE id = branch_type) AS branch_type_name,' - 'in_tx' + 'in_tx,' + 'insn_count,' + 'cyc_count,' + 'CASE WHEN cyc_count=0 THEN CAST(0 AS FLOAT) ELSE ROUND(CAST(insn_count AS FLOAT) / cyc_count, 2) END AS IPC' ' FROM samples') do_query(query, 'END TRANSACTION') @@ -407,15 +419,15 @@ branch_type_query = QSqlQuery(db) branch_type_query.prepare("INSERT INTO branch_types VALUES (?, ?)") sample_query = QSqlQuery(db) if branches: - sample_query.prepare("INSERT INTO samples VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)") + sample_query.prepare("INSERT INTO samples VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)") else: - sample_query.prepare("INSERT INTO samples VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)") + sample_query.prepare("INSERT INTO samples VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)") if perf_db_export_calls or perf_db_export_callchains: call_path_query = QSqlQuery(db) call_path_query.prepare("INSERT INTO call_paths VALUES (?, ?, ?, ?)") if perf_db_export_calls: call_query = QSqlQuery(db) - call_query.prepare("INSERT INTO calls VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)") + call_query.prepare("INSERT INTO calls VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)") def trace_begin(): printdate("Writing records...") @@ -427,10 +439,10 @@ def trace_begin(): comm_table(0, "unknown") dso_table(0, 0, "unknown", "unknown", "") symbol_table(0, 0, 0, 0, 0, "unknown") - sample_table(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) + sample_table(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) if perf_db_export_calls or perf_db_export_callchains: call_path_table(0, 0, 0, 0) - call_return_table(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) + call_return_table(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) unhandled_count = 0 @@ -486,14 +498,14 @@ def sample_table(*x): if branches: for xx in x[0:15]: sample_query.addBindValue(str(xx)) - for xx in x[19:22]: + for xx in x[19:24]: sample_query.addBindValue(str(xx)) do_query_(sample_query) else: - bind_exec(sample_query, 22, x) + bind_exec(sample_query, 24, x) def call_path_table(*x): bind_exec(call_path_query, 4, x) def call_return_table(*x): - bind_exec(call_query, 12, x) + bind_exec(call_query, 14, x) diff --git a/tools/perf/scripts/python/exported-sql-viewer.py b/tools/perf/scripts/python/exported-sql-viewer.py index 6fe553258ce5..6e7934f2ac9a 100755 --- a/tools/perf/scripts/python/exported-sql-viewer.py +++ b/tools/perf/scripts/python/exported-sql-viewer.py @@ -200,9 +200,10 @@ class Thread(QThread): class TreeModel(QAbstractItemModel): - def __init__(self, glb, parent=None): + def __init__(self, glb, params, parent=None): super(TreeModel, self).__init__(parent) self.glb = glb + self.params = params self.root = self.GetRoot() self.last_row_read = 0 @@ -399,6 +400,7 @@ class FindBar(): def Activate(self): self.bar.show() + self.textbox.lineEdit().selectAll() self.textbox.setFocus() def Deactivate(self): @@ -463,8 +465,9 @@ class FindBar(): class CallGraphLevelItemBase(object): - def __init__(self, glb, row, parent_item): + def __init__(self, glb, params, row, parent_item): self.glb = glb + self.params = params self.row = row self.parent_item = parent_item self.query_done = False; @@ -503,18 +506,24 @@ class CallGraphLevelItemBase(object): class CallGraphLevelTwoPlusItemBase(CallGraphLevelItemBase): - def __init__(self, glb, row, comm_id, thread_id, call_path_id, time, branch_count, parent_item): - super(CallGraphLevelTwoPlusItemBase, self).__init__(glb, row, parent_item) + def __init__(self, glb, params, row, comm_id, thread_id, call_path_id, time, insn_cnt, cyc_cnt, branch_count, parent_item): + super(CallGraphLevelTwoPlusItemBase, self).__init__(glb, params, row, parent_item) self.comm_id = comm_id self.thread_id = thread_id self.call_path_id = call_path_id + self.insn_cnt = insn_cnt + self.cyc_cnt = cyc_cnt self.branch_count = branch_count self.time = time def Select(self): self.query_done = True; query = QSqlQuery(self.glb.db) - QueryExec(query, "SELECT call_path_id, name, short_name, COUNT(calls.id), SUM(return_time - call_time), SUM(branch_count)" + if self.params.have_ipc: + ipc_str = ", SUM(insn_count), SUM(cyc_count)" + else: + ipc_str = "" + QueryExec(query, "SELECT call_path_id, name, short_name, COUNT(calls.id), SUM(return_time - call_time)" + ipc_str + ", SUM(branch_count)" " FROM calls" " INNER JOIN call_paths ON calls.call_path_id = call_paths.id" " INNER JOIN symbols ON call_paths.symbol_id = symbols.id" @@ -525,7 +534,15 @@ class CallGraphLevelTwoPlusItemBase(CallGraphLevelItemBase): " GROUP BY call_path_id, name, short_name" " ORDER BY call_path_id") while query.next(): - child_item = CallGraphLevelThreeItem(self.glb, self.child_count, self.comm_id, self.thread_id, query.value(0), query.value(1), query.value(2), query.value(3), int(query.value(4)), int(query.value(5)), self) + if self.params.have_ipc: + insn_cnt = int(query.value(5)) + cyc_cnt = int(query.value(6)) + branch_count = int(query.value(7)) + else: + insn_cnt = 0 + cyc_cnt = 0 + branch_count = int(query.value(5)) + child_item = CallGraphLevelThreeItem(self.glb, self.params, self.child_count, self.comm_id, self.thread_id, query.value(0), query.value(1), query.value(2), query.value(3), int(query.value(4)), insn_cnt, cyc_cnt, branch_count, self) self.child_items.append(child_item) self.child_count += 1 @@ -533,37 +550,57 @@ class CallGraphLevelTwoPlusItemBase(CallGraphLevelItemBase): class CallGraphLevelThreeItem(CallGraphLevelTwoPlusItemBase): - def __init__(self, glb, row, comm_id, thread_id, call_path_id, name, dso, count, time, branch_count, parent_item): - super(CallGraphLevelThreeItem, self).__init__(glb, row, comm_id, thread_id, call_path_id, time, branch_count, parent_item) + def __init__(self, glb, params, row, comm_id, thread_id, call_path_id, name, dso, count, time, insn_cnt, cyc_cnt, branch_count, parent_item): + super(CallGraphLevelThreeItem, self).__init__(glb, params, row, comm_id, thread_id, call_path_id, time, insn_cnt, cyc_cnt, branch_count, parent_item) dso = dsoname(dso) - self.data = [ name, dso, str(count), str(time), PercentToOneDP(time, parent_item.time), str(branch_count), PercentToOneDP(branch_count, parent_item.branch_count) ] + if self.params.have_ipc: + insn_pcnt = PercentToOneDP(insn_cnt, parent_item.insn_cnt) + cyc_pcnt = PercentToOneDP(cyc_cnt, parent_item.cyc_cnt) + br_pcnt = PercentToOneDP(branch_count, parent_item.branch_count) + ipc = CalcIPC(cyc_cnt, insn_cnt) + self.data = [ name, dso, str(count), str(time), PercentToOneDP(time, parent_item.time), str(insn_cnt), insn_pcnt, str(cyc_cnt), cyc_pcnt, ipc, str(branch_count), br_pcnt ] + else: + self.data = [ name, dso, str(count), str(time), PercentToOneDP(time, parent_item.time), str(branch_count), PercentToOneDP(branch_count, parent_item.branch_count) ] self.dbid = call_path_id # Context-sensitive call graph data model level two item class CallGraphLevelTwoItem(CallGraphLevelTwoPlusItemBase): - def __init__(self, glb, row, comm_id, thread_id, pid, tid, parent_item): - super(CallGraphLevelTwoItem, self).__init__(glb, row, comm_id, thread_id, 1, 0, 0, parent_item) - self.data = [str(pid) + ":" + str(tid), "", "", "", "", "", ""] + def __init__(self, glb, params, row, comm_id, thread_id, pid, tid, parent_item): + super(CallGraphLevelTwoItem, self).__init__(glb, params, row, comm_id, thread_id, 1, 0, 0, 0, 0, parent_item) + if self.params.have_ipc: + self.data = [str(pid) + ":" + str(tid), "", "", "", "", "", "", "", "", "", "", ""] + else: + self.data = [str(pid) + ":" + str(tid), "", "", "", "", "", ""] self.dbid = thread_id def Select(self): super(CallGraphLevelTwoItem, self).Select() for child_item in self.child_items: self.time += child_item.time + self.insn_cnt += child_item.insn_cnt + self.cyc_cnt += child_item.cyc_cnt self.branch_count += child_item.branch_count for child_item in self.child_items: child_item.data[4] = PercentToOneDP(child_item.time, self.time) - child_item.data[6] = PercentToOneDP(child_item.branch_count, self.branch_count) + if self.params.have_ipc: + child_item.data[6] = PercentToOneDP(child_item.insn_cnt, self.insn_cnt) + child_item.data[8] = PercentToOneDP(child_item.cyc_cnt, self.cyc_cnt) + child_item.data[11] = PercentToOneDP(child_item.branch_count, self.branch_count) + else: + child_item.data[6] = PercentToOneDP(child_item.branch_count, self.branch_count) # Context-sensitive call graph data model level one item class CallGraphLevelOneItem(CallGraphLevelItemBase): - def __init__(self, glb, row, comm_id, comm, parent_item): - super(CallGraphLevelOneItem, self).__init__(glb, row, parent_item) - self.data = [comm, "", "", "", "", "", ""] + def __init__(self, glb, params, row, comm_id, comm, parent_item): + super(CallGraphLevelOneItem, self).__init__(glb, params, row, parent_item) + if self.params.have_ipc: + self.data = [comm, "", "", "", "", "", "", "", "", "", "", ""] + else: + self.data = [comm, "", "", "", "", "", ""] self.dbid = comm_id def Select(self): @@ -574,7 +611,7 @@ class CallGraphLevelOneItem(CallGraphLevelItemBase): " INNER JOIN threads ON thread_id = threads.id" " WHERE comm_id = " + str(self.dbid)) while query.next(): - child_item = CallGraphLevelTwoItem(self.glb, self.child_count, self.dbid, query.value(0), query.value(1), query.value(2), self) + child_item = CallGraphLevelTwoItem(self.glb, self.params, self.child_count, self.dbid, query.value(0), query.value(1), query.value(2), self) self.child_items.append(child_item) self.child_count += 1 @@ -582,8 +619,8 @@ class CallGraphLevelOneItem(CallGraphLevelItemBase): class CallGraphRootItem(CallGraphLevelItemBase): - def __init__(self, glb): - super(CallGraphRootItem, self).__init__(glb, 0, None) + def __init__(self, glb, params): + super(CallGraphRootItem, self).__init__(glb, params, 0, None) self.dbid = 0 self.query_done = True; query = QSqlQuery(glb.db) @@ -591,16 +628,23 @@ class CallGraphRootItem(CallGraphLevelItemBase): while query.next(): if not query.value(0): continue - child_item = CallGraphLevelOneItem(glb, self.child_count, query.value(0), query.value(1), self) + child_item = CallGraphLevelOneItem(glb, params, self.child_count, query.value(0), query.value(1), self) self.child_items.append(child_item) self.child_count += 1 +# Call graph model parameters + +class CallGraphModelParams(): + + def __init__(self, glb, parent=None): + self.have_ipc = IsSelectable(glb.db, "calls", columns = "insn_count, cyc_count") + # Context-sensitive call graph data model base class CallGraphModelBase(TreeModel): def __init__(self, glb, parent=None): - super(CallGraphModelBase, self).__init__(glb, parent) + super(CallGraphModelBase, self).__init__(glb, CallGraphModelParams(glb), parent) def FindSelect(self, value, pattern, query): if pattern: @@ -682,17 +726,26 @@ class CallGraphModel(CallGraphModelBase): super(CallGraphModel, self).__init__(glb, parent) def GetRoot(self): - return CallGraphRootItem(self.glb) + return CallGraphRootItem(self.glb, self.params) def columnCount(self, parent=None): - return 7 + if self.params.have_ipc: + return 12 + else: + return 7 def columnHeader(self, column): - headers = ["Call Path", "Object", "Count ", "Time (ns) ", "Time (%) ", "Branch Count ", "Branch Count (%) "] + if self.params.have_ipc: + headers = ["Call Path", "Object", "Count ", "Time (ns) ", "Time (%) ", "Insn Cnt", "Insn Cnt (%)", "Cyc Cnt", "Cyc Cnt (%)", "IPC", "Branch Count ", "Branch Count (%) "] + else: + headers = ["Call Path", "Object", "Count ", "Time (ns) ", "Time (%) ", "Branch Count ", "Branch Count (%) "] return headers[column] def columnAlignment(self, column): - alignment = [ Qt.AlignLeft, Qt.AlignLeft, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight ] + if self.params.have_ipc: + alignment = [ Qt.AlignLeft, Qt.AlignLeft, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight ] + else: + alignment = [ Qt.AlignLeft, Qt.AlignLeft, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight ] return alignment[column] def DoFindSelect(self, query, match): @@ -729,11 +782,13 @@ class CallGraphModel(CallGraphModelBase): class CallTreeLevelTwoPlusItemBase(CallGraphLevelItemBase): - def __init__(self, glb, row, comm_id, thread_id, calls_id, time, branch_count, parent_item): - super(CallTreeLevelTwoPlusItemBase, self).__init__(glb, row, parent_item) + def __init__(self, glb, params, row, comm_id, thread_id, calls_id, time, insn_cnt, cyc_cnt, branch_count, parent_item): + super(CallTreeLevelTwoPlusItemBase, self).__init__(glb, params, row, parent_item) self.comm_id = comm_id self.thread_id = thread_id self.calls_id = calls_id + self.insn_cnt = insn_cnt + self.cyc_cnt = cyc_cnt self.branch_count = branch_count self.time = time @@ -743,8 +798,12 @@ class CallTreeLevelTwoPlusItemBase(CallGraphLevelItemBase): comm_thread = " AND comm_id = " + str(self.comm_id) + " AND thread_id = " + str(self.thread_id) else: comm_thread = "" + if self.params.have_ipc: + ipc_str = ", insn_count, cyc_count" + else: + ipc_str = "" query = QSqlQuery(self.glb.db) - QueryExec(query, "SELECT calls.id, name, short_name, call_time, return_time - call_time, branch_count" + QueryExec(query, "SELECT calls.id, name, short_name, call_time, return_time - call_time" + ipc_str + ", branch_count" " FROM calls" " INNER JOIN call_paths ON calls.call_path_id = call_paths.id" " INNER JOIN symbols ON call_paths.symbol_id = symbols.id" @@ -752,7 +811,15 @@ class CallTreeLevelTwoPlusItemBase(CallGraphLevelItemBase): " WHERE calls.parent_id = " + str(self.calls_id) + comm_thread + " ORDER BY call_time, calls.id") while query.next(): - child_item = CallTreeLevelThreeItem(self.glb, self.child_count, self.comm_id, self.thread_id, query.value(0), query.value(1), query.value(2), query.value(3), int(query.value(4)), int(query.value(5)), self) + if self.params.have_ipc: + insn_cnt = int(query.value(5)) + cyc_cnt = int(query.value(6)) + branch_count = int(query.value(7)) + else: + insn_cnt = 0 + cyc_cnt = 0 + branch_count = int(query.value(5)) + child_item = CallTreeLevelThreeItem(self.glb, self.params, self.child_count, self.comm_id, self.thread_id, query.value(0), query.value(1), query.value(2), query.value(3), int(query.value(4)), insn_cnt, cyc_cnt, branch_count, self) self.child_items.append(child_item) self.child_count += 1 @@ -760,37 +827,57 @@ class CallTreeLevelTwoPlusItemBase(CallGraphLevelItemBase): class CallTreeLevelThreeItem(CallTreeLevelTwoPlusItemBase): - def __init__(self, glb, row, comm_id, thread_id, calls_id, name, dso, count, time, branch_count, parent_item): - super(CallTreeLevelThreeItem, self).__init__(glb, row, comm_id, thread_id, calls_id, time, branch_count, parent_item) + def __init__(self, glb, params, row, comm_id, thread_id, calls_id, name, dso, count, time, insn_cnt, cyc_cnt, branch_count, parent_item): + super(CallTreeLevelThreeItem, self).__init__(glb, params, row, comm_id, thread_id, calls_id, time, insn_cnt, cyc_cnt, branch_count, parent_item) dso = dsoname(dso) - self.data = [ name, dso, str(count), str(time), PercentToOneDP(time, parent_item.time), str(branch_count), PercentToOneDP(branch_count, parent_item.branch_count) ] + if self.params.have_ipc: + insn_pcnt = PercentToOneDP(insn_cnt, parent_item.insn_cnt) + cyc_pcnt = PercentToOneDP(cyc_cnt, parent_item.cyc_cnt) + br_pcnt = PercentToOneDP(branch_count, parent_item.branch_count) + ipc = CalcIPC(cyc_cnt, insn_cnt) + self.data = [ name, dso, str(count), str(time), PercentToOneDP(time, parent_item.time), str(insn_cnt), insn_pcnt, str(cyc_cnt), cyc_pcnt, ipc, str(branch_count), br_pcnt ] + else: + self.data = [ name, dso, str(count), str(time), PercentToOneDP(time, parent_item.time), str(branch_count), PercentToOneDP(branch_count, parent_item.branch_count) ] self.dbid = calls_id # Call tree data model level two item class CallTreeLevelTwoItem(CallTreeLevelTwoPlusItemBase): - def __init__(self, glb, row, comm_id, thread_id, pid, tid, parent_item): - super(CallTreeLevelTwoItem, self).__init__(glb, row, comm_id, thread_id, 0, 0, 0, parent_item) - self.data = [str(pid) + ":" + str(tid), "", "", "", "", "", ""] + def __init__(self, glb, params, row, comm_id, thread_id, pid, tid, parent_item): + super(CallTreeLevelTwoItem, self).__init__(glb, params, row, comm_id, thread_id, 0, 0, 0, 0, 0, parent_item) + if self.params.have_ipc: + self.data = [str(pid) + ":" + str(tid), "", "", "", "", "", "", "", "", "", "", ""] + else: + self.data = [str(pid) + ":" + str(tid), "", "", "", "", "", ""] self.dbid = thread_id def Select(self): super(CallTreeLevelTwoItem, self).Select() for child_item in self.child_items: self.time += child_item.time + self.insn_cnt += child_item.insn_cnt + self.cyc_cnt += child_item.cyc_cnt self.branch_count += child_item.branch_count for child_item in self.child_items: child_item.data[4] = PercentToOneDP(child_item.time, self.time) - child_item.data[6] = PercentToOneDP(child_item.branch_count, self.branch_count) + if self.params.have_ipc: + child_item.data[6] = PercentToOneDP(child_item.insn_cnt, self.insn_cnt) + child_item.data[8] = PercentToOneDP(child_item.cyc_cnt, self.cyc_cnt) + child_item.data[11] = PercentToOneDP(child_item.branch_count, self.branch_count) + else: + child_item.data[6] = PercentToOneDP(child_item.branch_count, self.branch_count) # Call tree data model level one item class CallTreeLevelOneItem(CallGraphLevelItemBase): - def __init__(self, glb, row, comm_id, comm, parent_item): - super(CallTreeLevelOneItem, self).__init__(glb, row, parent_item) - self.data = [comm, "", "", "", "", "", ""] + def __init__(self, glb, params, row, comm_id, comm, parent_item): + super(CallTreeLevelOneItem, self).__init__(glb, params, row, parent_item) + if self.params.have_ipc: + self.data = [comm, "", "", "", "", "", "", "", "", "", "", ""] + else: + self.data = [comm, "", "", "", "", "", ""] self.dbid = comm_id def Select(self): @@ -801,7 +888,7 @@ class CallTreeLevelOneItem(CallGraphLevelItemBase): " INNER JOIN threads ON thread_id = threads.id" " WHERE comm_id = " + str(self.dbid)) while query.next(): - child_item = CallTreeLevelTwoItem(self.glb, self.child_count, self.dbid, query.value(0), query.value(1), query.value(2), self) + child_item = CallTreeLevelTwoItem(self.glb, self.params, self.child_count, self.dbid, query.value(0), query.value(1), query.value(2), self) self.child_items.append(child_item) self.child_count += 1 @@ -809,8 +896,8 @@ class CallTreeLevelOneItem(CallGraphLevelItemBase): class CallTreeRootItem(CallGraphLevelItemBase): - def __init__(self, glb): - super(CallTreeRootItem, self).__init__(glb, 0, None) + def __init__(self, glb, params): + super(CallTreeRootItem, self).__init__(glb, params, 0, None) self.dbid = 0 self.query_done = True; query = QSqlQuery(glb.db) @@ -818,7 +905,7 @@ class CallTreeRootItem(CallGraphLevelItemBase): while query.next(): if not query.value(0): continue - child_item = CallTreeLevelOneItem(glb, self.child_count, query.value(0), query.value(1), self) + child_item = CallTreeLevelOneItem(glb, params, self.child_count, query.value(0), query.value(1), self) self.child_items.append(child_item) self.child_count += 1 @@ -830,17 +917,26 @@ class CallTreeModel(CallGraphModelBase): super(CallTreeModel, self).__init__(glb, parent) def GetRoot(self): - return CallTreeRootItem(self.glb) + return CallTreeRootItem(self.glb, self.params) def columnCount(self, parent=None): - return 7 + if self.params.have_ipc: + return 12 + else: + return 7 def columnHeader(self, column): - headers = ["Call Path", "Object", "Call Time", "Time (ns) ", "Time (%) ", "Branch Count ", "Branch Count (%) "] + if self.params.have_ipc: + headers = ["Call Path", "Object", "Call Time", "Time (ns) ", "Time (%) ", "Insn Cnt", "Insn Cnt (%)", "Cyc Cnt", "Cyc Cnt (%)", "IPC", "Branch Count ", "Branch Count (%) "] + else: + headers = ["Call Path", "Object", "Call Time", "Time (ns) ", "Time (%) ", "Branch Count ", "Branch Count (%) "] return headers[column] def columnAlignment(self, column): - alignment = [ Qt.AlignLeft, Qt.AlignLeft, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight ] + if self.params.have_ipc: + alignment = [ Qt.AlignLeft, Qt.AlignLeft, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight ] + else: + alignment = [ Qt.AlignLeft, Qt.AlignLeft, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight, Qt.AlignRight ] return alignment[column] def DoFindSelect(self, query, match): @@ -1369,11 +1465,11 @@ class FetchMoreRecordsBar(): class BranchLevelTwoItem(): - def __init__(self, row, text, parent_item): + def __init__(self, row, col, text, parent_item): self.row = row self.parent_item = parent_item - self.data = [""] * 8 - self.data[7] = text + self.data = [""] * (col + 1) + self.data[col] = text self.level = 2 def getParentItem(self): @@ -1405,6 +1501,7 @@ class BranchLevelOneItem(): self.dbid = data[0] self.level = 1 self.query_done = False + self.br_col = len(self.data) - 1 def getChildItem(self, row): return self.child_items[row] @@ -1485,7 +1582,7 @@ class BranchLevelOneItem(): while k < 15: byte_str += " " k += 1 - self.child_items.append(BranchLevelTwoItem(0, byte_str + " " + text, self)) + self.child_items.append(BranchLevelTwoItem(0, self.br_col, byte_str + " " + text, self)) self.child_count += 1 else: return @@ -1536,16 +1633,37 @@ class BranchRootItem(): def getData(self, column): return "" +# Calculate instructions per cycle + +def CalcIPC(cyc_cnt, insn_cnt): + if cyc_cnt and insn_cnt: + ipc = Decimal(float(insn_cnt) / cyc_cnt) + ipc = str(ipc.quantize(Decimal(".01"), rounding=ROUND_HALF_UP)) + else: + ipc = "0" + return ipc + # Branch data preparation -def BranchDataPrep(query): - data = [] - for i in xrange(0, 8): - data.append(query.value(i)) +def BranchDataPrepBr(query, data): data.append(tohex(query.value(8)).rjust(16) + " " + query.value(9) + offstr(query.value(10)) + " (" + dsoname(query.value(11)) + ")" + " -> " + tohex(query.value(12)) + " " + query.value(13) + offstr(query.value(14)) + " (" + dsoname(query.value(15)) + ")") + +def BranchDataPrepIPC(query, data): + insn_cnt = query.value(16) + cyc_cnt = query.value(17) + ipc = CalcIPC(cyc_cnt, insn_cnt) + data.append(insn_cnt) + data.append(cyc_cnt) + data.append(ipc) + +def BranchDataPrep(query): + data = [] + for i in xrange(0, 8): + data.append(query.value(i)) + BranchDataPrepBr(query, data) return data def BranchDataPrepWA(query): @@ -1555,10 +1673,26 @@ def BranchDataPrepWA(query): data.append("{:>19}".format(query.value(1))) for i in xrange(2, 8): data.append(query.value(i)) - data.append(tohex(query.value(8)).rjust(16) + " " + query.value(9) + offstr(query.value(10)) + - " (" + dsoname(query.value(11)) + ")" + " -> " + - tohex(query.value(12)) + " " + query.value(13) + offstr(query.value(14)) + - " (" + dsoname(query.value(15)) + ")") + BranchDataPrepBr(query, data) + return data + +def BranchDataWithIPCPrep(query): + data = [] + for i in xrange(0, 8): + data.append(query.value(i)) + BranchDataPrepIPC(query, data) + BranchDataPrepBr(query, data) + return data + +def BranchDataWithIPCPrepWA(query): + data = [] + data.append(query.value(0)) + # Workaround pyside failing to handle large integers (i.e. time) in python3 by converting to a string + data.append("{:>19}".format(query.value(1))) + for i in xrange(2, 8): + data.append(query.value(i)) + BranchDataPrepIPC(query, data) + BranchDataPrepBr(query, data) return data # Branch data model @@ -1568,14 +1702,24 @@ class BranchModel(TreeModel): progress = Signal(object) def __init__(self, glb, event_id, where_clause, parent=None): - super(BranchModel, self).__init__(glb, parent) + super(BranchModel, self).__init__(glb, None, parent) self.event_id = event_id self.more = True self.populated = 0 + self.have_ipc = IsSelectable(glb.db, "samples", columns = "insn_count, cyc_count") + if self.have_ipc: + select_ipc = ", insn_count, cyc_count" + prep_fn = BranchDataWithIPCPrep + prep_wa_fn = BranchDataWithIPCPrepWA + else: + select_ipc = "" + prep_fn = BranchDataPrep + prep_wa_fn = BranchDataPrepWA sql = ("SELECT samples.id, time, cpu, comm, pid, tid, branch_types.name," " CASE WHEN in_tx = '0' THEN 'No' ELSE 'Yes' END," " ip, symbols.name, sym_offset, dsos.short_name," " to_ip, to_symbols.name, to_sym_offset, to_dsos.short_name" + + select_ipc + " FROM samples" " INNER JOIN comms ON comm_id = comms.id" " INNER JOIN threads ON thread_id = threads.id" @@ -1589,9 +1733,9 @@ class BranchModel(TreeModel): " ORDER BY samples.id" " LIMIT " + str(glb_chunk_sz)) if pyside_version_1 and sys.version_info[0] == 3: - prep = BranchDataPrepWA + prep = prep_fn else: - prep = BranchDataPrep + prep = prep_wa_fn self.fetcher = SQLFetcher(glb, sql, prep, self.AddSample) self.fetcher.done.connect(self.Update) self.fetcher.Fetch(glb_chunk_sz) @@ -1600,13 +1744,23 @@ class BranchModel(TreeModel): return BranchRootItem() def columnCount(self, parent=None): - return 8 + if self.have_ipc: + return 11 + else: + return 8 def columnHeader(self, column): - return ("Time", "CPU", "Command", "PID", "TID", "Branch Type", "In Tx", "Branch")[column] + if self.have_ipc: + return ("Time", "CPU", "Command", "PID", "TID", "Branch Type", "In Tx", "Insn Cnt", "Cyc Cnt", "IPC", "Branch")[column] + else: + return ("Time", "CPU", "Command", "PID", "TID", "Branch Type", "In Tx", "Branch")[column] def columnFont(self, column): - if column != 7: + if self.have_ipc: + br_col = 10 + else: + br_col = 7 + if column != br_col: return None return QFont("Monospace") @@ -2114,10 +2268,10 @@ def GetEventList(db): # Is a table selectable -def IsSelectable(db, table, sql = ""): +def IsSelectable(db, table, sql = "", columns = "*"): query = QSqlQuery(db) try: - QueryExec(query, "SELECT * FROM " + table + " " + sql + " LIMIT 1") + QueryExec(query, "SELECT " + columns + " FROM " + table + " " + sql + " LIMIT 1") except: return False return True @@ -2854,6 +3008,12 @@ cd xed sudo ./mfile.py --prefix=/usr/local install sudo ldconfig </pre> +<h3>Instructions per Cycle (IPC)</h3> +If available, IPC information is displayed in columns 'insn_cnt', 'cyc_cnt' and 'IPC'. +<p><b>Intel PT note:</b> The information applies to the blocks of code ending with, and including, that branch. +Due to the granularity of timing information, the number of cycles for some code blocks will not be known. +In that case, 'insn_cnt', 'cyc_cnt' and 'IPC' are zero, but when 'IPC' is displayed it covers the period +since the previous displayed 'IPC'. <h3>Find</h3> Ctrl-F displays a Find bar which finds substrings by either an exact match or a regular expression match. Refer to Python documentation for the regular expression syntax. diff --git a/tools/perf/tests/Build b/tools/perf/tests/Build index 4afb6319ed51..e3ba63cef01e 100644 --- a/tools/perf/tests/Build +++ b/tools/perf/tests/Build @@ -51,6 +51,7 @@ perf-y += clang.o perf-y += unit_number__scnprintf.o perf-y += mem2node.o perf-y += map_groups.o +perf-y += time-utils-test.o $(OUTPUT)tests/llvm-src-base.c: tests/bpf-script-example.c tests/Build $(call rule_mkdir) diff --git a/tools/perf/tests/builtin-test.c b/tools/perf/tests/builtin-test.c index 941c5456d625..cd72ff0f7658 100644 --- a/tools/perf/tests/builtin-test.c +++ b/tools/perf/tests/builtin-test.c @@ -290,6 +290,10 @@ static struct test generic_tests[] = { .func = test__mem2node, }, { + .desc = "time utils", + .func = test__time_utils, + }, + { .desc = "map_groups__merge_in", .func = test__map_groups__merge_in, }, diff --git a/tools/perf/tests/parse-events.c b/tools/perf/tests/parse-events.c index 4a69c07f4101..8f3c80e13584 100644 --- a/tools/perf/tests/parse-events.c +++ b/tools/perf/tests/parse-events.c @@ -18,6 +18,32 @@ #define PERF_TP_SAMPLE_TYPE (PERF_SAMPLE_RAW | PERF_SAMPLE_TIME | \ PERF_SAMPLE_CPU | PERF_SAMPLE_PERIOD) +#if defined(__s390x__) +/* Return true if kvm module is available and loaded. Test this + * and retun success when trace point kvm_s390_create_vm + * exists. Otherwise this test always fails. + */ +static bool kvm_s390_create_vm_valid(void) +{ + char *eventfile; + bool rc = false; + + eventfile = get_events_file("kvm-s390"); + + if (eventfile) { + DIR *mydir = opendir(eventfile); + + if (mydir) { + rc = true; + closedir(mydir); + } + put_events_file(eventfile); + } + + return rc; +} +#endif + static int test__checkevent_tracepoint(struct perf_evlist *evlist) { struct perf_evsel *evsel = perf_evlist__first(evlist); @@ -1642,6 +1668,7 @@ static struct evlist_test test__events[] = { { .name = "kvm-s390:kvm_s390_create_vm", .check = test__checkevent_tracepoint, + .valid = kvm_s390_create_vm_valid, .id = 100, }, #endif diff --git a/tools/perf/tests/tests.h b/tools/perf/tests/tests.h index e5e3a57cd373..72912eb473cb 100644 --- a/tools/perf/tests/tests.h +++ b/tools/perf/tests/tests.h @@ -108,6 +108,7 @@ int test__clang_subtest_get_nr(void); int test__unit_number__scnprint(struct test *test, int subtest); int test__mem2node(struct test *t, int subtest); int test__map_groups__merge_in(struct test *t, int subtest); +int test__time_utils(struct test *t, int subtest); bool test__bp_signal_is_supported(void); bool test__wp_is_supported(void); diff --git a/tools/perf/tests/time-utils-test.c b/tools/perf/tests/time-utils-test.c new file mode 100644 index 000000000000..4f53006233a1 --- /dev/null +++ b/tools/perf/tests/time-utils-test.c @@ -0,0 +1,251 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/compiler.h> +#include <linux/time64.h> +#include <inttypes.h> +#include <string.h> +#include "time-utils.h" +#include "evlist.h" +#include "session.h" +#include "debug.h" +#include "tests.h" + +static bool test__parse_nsec_time(const char *str, u64 expected) +{ + u64 ptime; + int err; + + pr_debug("\nparse_nsec_time(\"%s\")\n", str); + + err = parse_nsec_time(str, &ptime); + if (err) { + pr_debug("error %d\n", err); + return false; + } + + if (ptime != expected) { + pr_debug("Failed. ptime %" PRIu64 " expected %" PRIu64 "\n", + ptime, expected); + return false; + } + + pr_debug("%" PRIu64 "\n", ptime); + + return true; +} + +static bool test__perf_time__parse_str(const char *ostr, u64 start, u64 end) +{ + struct perf_time_interval ptime; + int err; + + pr_debug("\nperf_time__parse_str(\"%s\")\n", ostr); + + err = perf_time__parse_str(&ptime, ostr); + if (err) { + pr_debug("Error %d\n", err); + return false; + } + + if (ptime.start != start || ptime.end != end) { + pr_debug("Failed. Expected %" PRIu64 " to %" PRIu64 "\n", + start, end); + return false; + } + + return true; +} + +#define TEST_MAX 64 + +struct test_data { + const char *str; + u64 first; + u64 last; + struct perf_time_interval ptime[TEST_MAX]; + int num; + u64 skip[TEST_MAX]; + u64 noskip[TEST_MAX]; +}; + +static bool test__perf_time__parse_for_ranges(struct test_data *d) +{ + struct perf_evlist evlist = { + .first_sample_time = d->first, + .last_sample_time = d->last, + }; + struct perf_session session = { .evlist = &evlist }; + struct perf_time_interval *ptime = NULL; + int range_size, range_num; + bool pass = false; + int i, err; + + pr_debug("\nperf_time__parse_for_ranges(\"%s\")\n", d->str); + + if (strchr(d->str, '%')) + pr_debug("first_sample_time %" PRIu64 " last_sample_time %" PRIu64 "\n", + d->first, d->last); + + err = perf_time__parse_for_ranges(d->str, &session, &ptime, &range_size, + &range_num); + if (err) { + pr_debug("error %d\n", err); + goto out; + } + + if (range_size < d->num || range_num != d->num) { + pr_debug("bad size: range_size %d range_num %d expected num %d\n", + range_size, range_num, d->num); + goto out; + } + + for (i = 0; i < d->num; i++) { + if (ptime[i].start != d->ptime[i].start || + ptime[i].end != d->ptime[i].end) { + pr_debug("bad range %d expected %" PRIu64 " to %" PRIu64 "\n", + i, d->ptime[i].start, d->ptime[i].end); + goto out; + } + } + + if (perf_time__ranges_skip_sample(ptime, d->num, 0)) { + pr_debug("failed to keep 0\n"); + goto out; + } + + for (i = 0; i < TEST_MAX; i++) { + if (d->skip[i] && + !perf_time__ranges_skip_sample(ptime, d->num, d->skip[i])) { + pr_debug("failed to skip %" PRIu64 "\n", d->skip[i]); + goto out; + } + if (d->noskip[i] && + perf_time__ranges_skip_sample(ptime, d->num, d->noskip[i])) { + pr_debug("failed to keep %" PRIu64 "\n", d->noskip[i]); + goto out; + } + } + + pass = true; +out: + free(ptime); + return pass; +} + +int test__time_utils(struct test *t __maybe_unused, int subtest __maybe_unused) +{ + bool pass = true; + + pass &= test__parse_nsec_time("0", 0); + pass &= test__parse_nsec_time("1", 1000000000ULL); + pass &= test__parse_nsec_time("0.000000001", 1); + pass &= test__parse_nsec_time("1.000000001", 1000000001ULL); + pass &= test__parse_nsec_time("123456.123456", 123456123456000ULL); + pass &= test__parse_nsec_time("1234567.123456789", 1234567123456789ULL); + pass &= test__parse_nsec_time("18446744073.709551615", + 0xFFFFFFFFFFFFFFFFULL); + + pass &= test__perf_time__parse_str("1234567.123456789,1234567.123456789", + 1234567123456789ULL, 1234567123456789ULL); + pass &= test__perf_time__parse_str("1234567.123456789,1234567.123456790", + 1234567123456789ULL, 1234567123456790ULL); + pass &= test__perf_time__parse_str("1234567.123456789,", + 1234567123456789ULL, 0); + pass &= test__perf_time__parse_str(",1234567.123456789", + 0, 1234567123456789ULL); + pass &= test__perf_time__parse_str("0,1234567.123456789", + 0, 1234567123456789ULL); + + { + u64 b = 1234567123456789ULL; + struct test_data d = { + .str = "1234567.123456789,1234567.123456790", + .ptime = { {b, b + 1}, }, + .num = 1, + .skip = { b - 1, b + 2, }, + .noskip = { b, b + 1, }, + }; + + pass &= test__perf_time__parse_for_ranges(&d); + } + + { + u64 b = 1234567123456789ULL; + u64 c = 7654321987654321ULL; + u64 e = 8000000000000000ULL; + struct test_data d = { + .str = "1234567.123456789,1234567.123456790 " + "7654321.987654321,7654321.987654444 " + "8000000,8000000.000000005", + .ptime = { {b, b + 1}, {c, c + 123}, {e, e + 5}, }, + .num = 3, + .skip = { b - 1, b + 2, c - 1, c + 124, e - 1, e + 6 }, + .noskip = { b, b + 1, c, c + 123, e, e + 5 }, + }; + + pass &= test__perf_time__parse_for_ranges(&d); + } + + { + u64 b = 7654321ULL * NSEC_PER_SEC; + struct test_data d = { + .str = "10%/1", + .first = b, + .last = b + 100, + .ptime = { {b, b + 9}, }, + .num = 1, + .skip = { b - 1, b + 10, }, + .noskip = { b, b + 9, }, + }; + + pass &= test__perf_time__parse_for_ranges(&d); + } + + { + u64 b = 7654321ULL * NSEC_PER_SEC; + struct test_data d = { + .str = "10%/2", + .first = b, + .last = b + 100, + .ptime = { {b + 10, b + 19}, }, + .num = 1, + .skip = { b + 9, b + 20, }, + .noskip = { b + 10, b + 19, }, + }; + + pass &= test__perf_time__parse_for_ranges(&d); + } + + { + u64 b = 11223344ULL * NSEC_PER_SEC; + struct test_data d = { + .str = "10%/1,10%/2", + .first = b, + .last = b + 100, + .ptime = { {b, b + 9}, {b + 10, b + 19}, }, + .num = 2, + .skip = { b - 1, b + 20, }, + .noskip = { b, b + 8, b + 9, b + 10, b + 11, b + 12, b + 19, }, + }; + + pass &= test__perf_time__parse_for_ranges(&d); + } + + { + u64 b = 11223344ULL * NSEC_PER_SEC; + struct test_data d = { + .str = "10%/1,10%/3,10%/10", + .first = b, + .last = b + 100, + .ptime = { {b, b + 9}, {b + 20, b + 29}, { b + 90, b + 100}, }, + .num = 3, + .skip = { b - 1, b + 10, b + 19, b + 30, b + 89, b + 101 }, + .noskip = { b, b + 9, b + 20, b + 29, b + 90, b + 100}, + }; + + pass &= test__perf_time__parse_for_ranges(&d); + } + + pr_debug("\n"); + + return pass ? 0 : TEST_FAIL; +} diff --git a/tools/perf/util/annotate.c b/tools/perf/util/annotate.c index 79db038b56f2..c8ce13419d9b 100644 --- a/tools/perf/util/annotate.c +++ b/tools/perf/util/annotate.c @@ -931,9 +931,8 @@ static int symbol__inc_addr_samples(struct symbol *sym, struct map *map, if (sym == NULL) return 0; src = symbol__hists(sym, evsel->evlist->nr_entries); - if (src == NULL) - return -ENOMEM; - return __symbol__inc_addr_samples(sym, map, src, evsel->idx, addr, sample); + return (src) ? __symbol__inc_addr_samples(sym, map, src, evsel->idx, + addr, sample) : 0; } static int symbol__account_cycles(u64 addr, u64 start, diff --git a/tools/perf/util/auxtrace.h b/tools/perf/util/auxtrace.h index d62f60eb5df4..e9b4c5edf78b 100644 --- a/tools/perf/util/auxtrace.h +++ b/tools/perf/util/auxtrace.h @@ -74,6 +74,8 @@ enum itrace_period_type { * @period_type: 'instructions' events period type * @initial_skip: skip N events at the beginning. * @cpu_bitmap: CPUs for which to synthesize events, or NULL for all + * @ptime_range: time intervals to trace or NULL + * @range_num: number of time intervals to trace */ struct itrace_synth_opts { bool set; @@ -98,6 +100,8 @@ struct itrace_synth_opts { enum itrace_period_type period_type; unsigned long initial_skip; unsigned long *cpu_bitmap; + struct perf_time_interval *ptime_range; + int range_num; }; /** @@ -590,6 +594,21 @@ static inline void auxtrace__free(struct perf_session *session) " PERIOD[ns|us|ms|i|t]: specify period to sample stream\n" \ " concatenate multiple options. Default is ibxwpe or cewp\n" +static inline +void itrace_synth_opts__set_time_range(struct itrace_synth_opts *opts, + struct perf_time_interval *ptime_range, + int range_num) +{ + opts->ptime_range = ptime_range; + opts->range_num = range_num; +} + +static inline +void itrace_synth_opts__clear_time_range(struct itrace_synth_opts *opts) +{ + opts->ptime_range = NULL; + opts->range_num = 0; +} #else @@ -733,6 +752,21 @@ void auxtrace_mmap_params__set_idx(struct auxtrace_mmap_params *mp, #define ITRACE_HELP "" +static inline +void itrace_synth_opts__set_time_range(struct itrace_synth_opts *opts + __maybe_unused, + struct perf_time_interval *ptime_range + __maybe_unused, + int range_num __maybe_unused) +{ +} + +static inline +void itrace_synth_opts__clear_time_range(struct itrace_synth_opts *opts + __maybe_unused) +{ +} + #endif #endif diff --git a/tools/perf/util/config.c b/tools/perf/util/config.c index 7e3c1b60120c..e7d2c08d263a 100644 --- a/tools/perf/util/config.c +++ b/tools/perf/util/config.c @@ -739,11 +739,15 @@ int perf_config(config_fn_t fn, void *data) if (ret < 0) { pr_err("Error: wrong config key-value pair %s=%s\n", key, value); - break; + /* + * Can't be just a 'break', as perf_config_set__for_each_entry() + * expands to two nested for() loops. + */ + goto out; } } } - +out: return ret; } diff --git a/tools/perf/util/cpumap.c b/tools/perf/util/cpumap.c index 0b599229bc7e..c11a459ca582 100644 --- a/tools/perf/util/cpumap.c +++ b/tools/perf/util/cpumap.c @@ -373,6 +373,46 @@ int cpu_map__build_map(struct cpu_map *cpus, struct cpu_map **res, return 0; } +int cpu_map__get_die_id(int cpu) +{ + int value, ret = cpu__get_topology_int(cpu, "die_id", &value); + + return ret ?: value; +} + +int cpu_map__get_die(struct cpu_map *map, int idx, void *data) +{ + int cpu, die_id, s; + + if (idx > map->nr) + return -1; + + cpu = map->map[idx]; + + die_id = cpu_map__get_die_id(cpu); + /* There is no die_id on legacy system. */ + if (die_id == -1) + die_id = 0; + + s = cpu_map__get_socket(map, idx, data); + if (s == -1) + return -1; + + /* + * Encode socket in bit range 15:8 + * die_id is relative to socket, and + * we need a global id. So we combine + * socket + die id + */ + if (WARN_ONCE(die_id >> 8, "The die id number is too big.\n")) + return -1; + + if (WARN_ONCE(s >> 8, "The socket id number is too big.\n")) + return -1; + + return (s << 8) | (die_id & 0xff); +} + int cpu_map__get_core_id(int cpu) { int value, ret = cpu__get_topology_int(cpu, "core_id", &value); @@ -381,7 +421,7 @@ int cpu_map__get_core_id(int cpu) int cpu_map__get_core(struct cpu_map *map, int idx, void *data) { - int cpu, s; + int cpu, s_die; if (idx > map->nr) return -1; @@ -390,17 +430,22 @@ int cpu_map__get_core(struct cpu_map *map, int idx, void *data) cpu = cpu_map__get_core_id(cpu); - s = cpu_map__get_socket(map, idx, data); - if (s == -1) + /* s_die is the combination of socket + die id */ + s_die = cpu_map__get_die(map, idx, data); + if (s_die == -1) return -1; /* - * encode socket in upper 16 bits - * core_id is relative to socket, and + * encode socket in bit range 31:24 + * encode die id in bit range 23:16 + * core_id is relative to socket and die, * we need a global id. So we combine - * socket+ core id + * socket + die id + core id */ - return (s << 16) | (cpu & 0xffff); + if (WARN_ONCE(cpu >> 16, "The core id number is too big.\n")) + return -1; + + return (s_die << 16) | (cpu & 0xffff); } int cpu_map__build_socket_map(struct cpu_map *cpus, struct cpu_map **sockp) @@ -408,6 +453,11 @@ int cpu_map__build_socket_map(struct cpu_map *cpus, struct cpu_map **sockp) return cpu_map__build_map(cpus, sockp, cpu_map__get_socket, NULL); } +int cpu_map__build_die_map(struct cpu_map *cpus, struct cpu_map **diep) +{ + return cpu_map__build_map(cpus, diep, cpu_map__get_die, NULL); +} + int cpu_map__build_core_map(struct cpu_map *cpus, struct cpu_map **corep) { return cpu_map__build_map(cpus, corep, cpu_map__get_core, NULL); diff --git a/tools/perf/util/cpumap.h b/tools/perf/util/cpumap.h index f00ce624b9f7..1265f0e33920 100644 --- a/tools/perf/util/cpumap.h +++ b/tools/perf/util/cpumap.h @@ -25,9 +25,12 @@ size_t cpu_map__snprint_mask(struct cpu_map *map, char *buf, size_t size); size_t cpu_map__fprintf(struct cpu_map *map, FILE *fp); int cpu_map__get_socket_id(int cpu); int cpu_map__get_socket(struct cpu_map *map, int idx, void *data); +int cpu_map__get_die_id(int cpu); +int cpu_map__get_die(struct cpu_map *map, int idx, void *data); int cpu_map__get_core_id(int cpu); int cpu_map__get_core(struct cpu_map *map, int idx, void *data); int cpu_map__build_socket_map(struct cpu_map *cpus, struct cpu_map **sockp); +int cpu_map__build_die_map(struct cpu_map *cpus, struct cpu_map **diep); int cpu_map__build_core_map(struct cpu_map *cpus, struct cpu_map **corep); const struct cpu_map *cpu_map__online(void); /* thread unsafe */ @@ -43,7 +46,12 @@ static inline int cpu_map__socket(struct cpu_map *sock, int s) static inline int cpu_map__id_to_socket(int id) { - return id >> 16; + return id >> 24; +} + +static inline int cpu_map__id_to_die(int id) +{ + return (id >> 16) & 0xff; } static inline int cpu_map__id_to_cpu(int id) diff --git a/tools/perf/util/cputopo.c b/tools/perf/util/cputopo.c index ece0710249d4..26e73a4bd4fe 100644 --- a/tools/perf/util/cputopo.c +++ b/tools/perf/util/cputopo.c @@ -1,5 +1,6 @@ // SPDX-License-Identifier: GPL-2.0 #include <sys/param.h> +#include <sys/utsname.h> #include <inttypes.h> #include <api/fs/fs.h> @@ -8,11 +9,14 @@ #include "util.h" #include "env.h" - #define CORE_SIB_FMT \ "%s/devices/system/cpu/cpu%d/topology/core_siblings_list" +#define DIE_SIB_FMT \ + "%s/devices/system/cpu/cpu%d/topology/die_cpus_list" #define THRD_SIB_FMT \ "%s/devices/system/cpu/cpu%d/topology/thread_siblings_list" +#define THRD_SIB_FMT_NEW \ + "%s/devices/system/cpu/cpu%d/topology/core_cpus_list" #define NODE_ONLINE_FMT \ "%s/devices/system/node/online" #define NODE_MEMINFO_FMT \ @@ -34,12 +38,12 @@ static int build_cpu_topology(struct cpu_topology *tp, int cpu) sysfs__mountpoint(), cpu); fp = fopen(filename, "r"); if (!fp) - goto try_threads; + goto try_dies; sret = getline(&buf, &len, fp); fclose(fp); if (sret <= 0) - goto try_threads; + goto try_dies; p = strchr(buf, '\n'); if (p) @@ -57,9 +61,44 @@ static int build_cpu_topology(struct cpu_topology *tp, int cpu) } ret = 0; +try_dies: + if (!tp->die_siblings) + goto try_threads; + + scnprintf(filename, MAXPATHLEN, DIE_SIB_FMT, + sysfs__mountpoint(), cpu); + fp = fopen(filename, "r"); + if (!fp) + goto try_threads; + + sret = getline(&buf, &len, fp); + fclose(fp); + if (sret <= 0) + goto try_threads; + + p = strchr(buf, '\n'); + if (p) + *p = '\0'; + + for (i = 0; i < tp->die_sib; i++) { + if (!strcmp(buf, tp->die_siblings[i])) + break; + } + if (i == tp->die_sib) { + tp->die_siblings[i] = buf; + tp->die_sib++; + buf = NULL; + len = 0; + } + ret = 0; + try_threads: - scnprintf(filename, MAXPATHLEN, THRD_SIB_FMT, + scnprintf(filename, MAXPATHLEN, THRD_SIB_FMT_NEW, sysfs__mountpoint(), cpu); + if (access(filename, F_OK) == -1) { + scnprintf(filename, MAXPATHLEN, THRD_SIB_FMT, + sysfs__mountpoint(), cpu); + } fp = fopen(filename, "r"); if (!fp) goto done; @@ -98,21 +137,46 @@ void cpu_topology__delete(struct cpu_topology *tp) for (i = 0 ; i < tp->core_sib; i++) zfree(&tp->core_siblings[i]); + if (tp->die_sib) { + for (i = 0 ; i < tp->die_sib; i++) + zfree(&tp->die_siblings[i]); + } + for (i = 0 ; i < tp->thread_sib; i++) zfree(&tp->thread_siblings[i]); free(tp); } +static bool has_die_topology(void) +{ + char filename[MAXPATHLEN]; + struct utsname uts; + + if (uname(&uts) < 0) + return false; + + if (strncmp(uts.machine, "x86_64", 6)) + return false; + + scnprintf(filename, MAXPATHLEN, DIE_SIB_FMT, + sysfs__mountpoint(), 0); + if (access(filename, F_OK) == -1) + return false; + + return true; +} + struct cpu_topology *cpu_topology__new(void) { struct cpu_topology *tp = NULL; void *addr; - u32 nr, i; + u32 nr, i, nr_addr; size_t sz; long ncpus; int ret = -1; struct cpu_map *map; + bool has_die = has_die_topology(); ncpus = cpu__max_present_cpu(); @@ -126,7 +190,11 @@ struct cpu_topology *cpu_topology__new(void) nr = (u32)(ncpus & UINT_MAX); sz = nr * sizeof(char *); - addr = calloc(1, sizeof(*tp) + 2 * sz); + if (has_die) + nr_addr = 3; + else + nr_addr = 2; + addr = calloc(1, sizeof(*tp) + nr_addr * sz); if (!addr) goto out_free; @@ -134,6 +202,10 @@ struct cpu_topology *cpu_topology__new(void) addr += sizeof(*tp); tp->core_siblings = addr; addr += sz; + if (has_die) { + tp->die_siblings = addr; + addr += sz; + } tp->thread_siblings = addr; for (i = 0; i < nr; i++) { diff --git a/tools/perf/util/cputopo.h b/tools/perf/util/cputopo.h index 47a97e71acdf..bae2f1d41856 100644 --- a/tools/perf/util/cputopo.h +++ b/tools/perf/util/cputopo.h @@ -7,8 +7,10 @@ struct cpu_topology { u32 core_sib; + u32 die_sib; u32 thread_sib; char **core_siblings; + char **die_siblings; char **thread_siblings; }; diff --git a/tools/perf/util/cs-etm-decoder/cs-etm-decoder.c b/tools/perf/util/cs-etm-decoder/cs-etm-decoder.c index 39fe21e1cf93..bb45e23018ee 100644 --- a/tools/perf/util/cs-etm-decoder/cs-etm-decoder.c +++ b/tools/perf/util/cs-etm-decoder/cs-etm-decoder.c @@ -18,8 +18,6 @@ #include "intlist.h" #include "util.h" -#define MAX_BUFFER 1024 - /* use raw logging */ #ifdef CS_DEBUG_RAW #define CS_LOG_RAW_FRAMES @@ -31,33 +29,26 @@ #endif #endif -#define CS_ETM_INVAL_ADDR 0xdeadbeefdeadbeefUL - struct cs_etm_decoder { void *data; void (*packet_printer)(const char *msg); dcd_tree_handle_t dcd_tree; cs_etm_mem_cb_type mem_access; ocsd_datapath_resp_t prev_return; - u32 packet_count; - u32 head; - u32 tail; - struct cs_etm_packet packet_buffer[MAX_BUFFER]; }; static u32 cs_etm_decoder__mem_access(const void *context, const ocsd_vaddr_t address, const ocsd_mem_space_acc_t mem_space __maybe_unused, + const u8 trace_chan_id, const u32 req_size, u8 *buffer) { struct cs_etm_decoder *decoder = (struct cs_etm_decoder *) context; - return decoder->mem_access(decoder->data, - address, - req_size, - buffer); + return decoder->mem_access(decoder->data, trace_chan_id, + address, req_size, buffer); } int cs_etm_decoder__add_mem_access_cb(struct cs_etm_decoder *decoder, @@ -66,9 +57,10 @@ int cs_etm_decoder__add_mem_access_cb(struct cs_etm_decoder *decoder, { decoder->mem_access = cb_func; - if (ocsd_dt_add_callback_mem_acc(decoder->dcd_tree, start, end, - OCSD_MEM_SPACE_ANY, - cs_etm_decoder__mem_access, decoder)) + if (ocsd_dt_add_callback_trcid_mem_acc(decoder->dcd_tree, start, end, + OCSD_MEM_SPACE_ANY, + cs_etm_decoder__mem_access, + decoder)) return -1; return 0; @@ -88,14 +80,14 @@ int cs_etm_decoder__reset(struct cs_etm_decoder *decoder) return 0; } -int cs_etm_decoder__get_packet(struct cs_etm_decoder *decoder, +int cs_etm_decoder__get_packet(struct cs_etm_packet_queue *packet_queue, struct cs_etm_packet *packet) { - if (!decoder || !packet) + if (!packet_queue || !packet) return -EINVAL; /* Nothing to do, might as well just return */ - if (decoder->packet_count == 0) + if (packet_queue->packet_count == 0) return 0; /* * The queueing process in function cs_etm_decoder__buffer_packet() @@ -106,11 +98,12 @@ int cs_etm_decoder__get_packet(struct cs_etm_decoder *decoder, * value. Otherwise the first element of the packet queue is not * used. */ - decoder->head = (decoder->head + 1) & (MAX_BUFFER - 1); + packet_queue->head = (packet_queue->head + 1) & + (CS_ETM_PACKET_MAX_BUFFER - 1); - *packet = decoder->packet_buffer[decoder->head]; + *packet = packet_queue->packet_buffer[packet_queue->head]; - decoder->packet_count--; + packet_queue->packet_count--; return 1; } @@ -276,84 +269,130 @@ cs_etm_decoder__create_etm_packet_printer(struct cs_etm_trace_params *t_params, trace_config); } -static void cs_etm_decoder__clear_buffer(struct cs_etm_decoder *decoder) +static ocsd_datapath_resp_t +cs_etm_decoder__do_soft_timestamp(struct cs_etm_queue *etmq, + struct cs_etm_packet_queue *packet_queue, + const uint8_t trace_chan_id) { - int i; - - decoder->head = 0; - decoder->tail = 0; - decoder->packet_count = 0; - for (i = 0; i < MAX_BUFFER; i++) { - decoder->packet_buffer[i].isa = CS_ETM_ISA_UNKNOWN; - decoder->packet_buffer[i].start_addr = CS_ETM_INVAL_ADDR; - decoder->packet_buffer[i].end_addr = CS_ETM_INVAL_ADDR; - decoder->packet_buffer[i].instr_count = 0; - decoder->packet_buffer[i].last_instr_taken_branch = false; - decoder->packet_buffer[i].last_instr_size = 0; - decoder->packet_buffer[i].last_instr_type = 0; - decoder->packet_buffer[i].last_instr_subtype = 0; - decoder->packet_buffer[i].last_instr_cond = 0; - decoder->packet_buffer[i].flags = 0; - decoder->packet_buffer[i].exception_number = UINT32_MAX; - decoder->packet_buffer[i].trace_chan_id = UINT8_MAX; - decoder->packet_buffer[i].cpu = INT_MIN; + /* No timestamp packet has been received, nothing to do */ + if (!packet_queue->timestamp) + return OCSD_RESP_CONT; + + packet_queue->timestamp = packet_queue->next_timestamp; + + /* Estimate the timestamp for the next range packet */ + packet_queue->next_timestamp += packet_queue->instr_count; + packet_queue->instr_count = 0; + + /* Tell the front end which traceid_queue needs attention */ + cs_etm__etmq_set_traceid_queue_timestamp(etmq, trace_chan_id); + + return OCSD_RESP_WAIT; +} + +static ocsd_datapath_resp_t +cs_etm_decoder__do_hard_timestamp(struct cs_etm_queue *etmq, + const ocsd_generic_trace_elem *elem, + const uint8_t trace_chan_id) +{ + struct cs_etm_packet_queue *packet_queue; + + /* First get the packet queue for this traceID */ + packet_queue = cs_etm__etmq_get_packet_queue(etmq, trace_chan_id); + if (!packet_queue) + return OCSD_RESP_FATAL_SYS_ERR; + + /* + * We've seen a timestamp packet before - simply record the new value. + * Function do_soft_timestamp() will report the value to the front end, + * hence asking the decoder to keep decoding rather than stopping. + */ + if (packet_queue->timestamp) { + packet_queue->next_timestamp = elem->timestamp; + return OCSD_RESP_CONT; } + + /* + * This is the first timestamp we've seen since the beginning of traces + * or a discontinuity. Since timestamps packets are generated *after* + * range packets have been generated, we need to estimate the time at + * which instructions started by substracting the number of instructions + * executed to the timestamp. + */ + packet_queue->timestamp = elem->timestamp - packet_queue->instr_count; + packet_queue->next_timestamp = elem->timestamp; + packet_queue->instr_count = 0; + + /* Tell the front end which traceid_queue needs attention */ + cs_etm__etmq_set_traceid_queue_timestamp(etmq, trace_chan_id); + + /* Halt processing until we are being told to proceed */ + return OCSD_RESP_WAIT; +} + +static void +cs_etm_decoder__reset_timestamp(struct cs_etm_packet_queue *packet_queue) +{ + packet_queue->timestamp = 0; + packet_queue->next_timestamp = 0; + packet_queue->instr_count = 0; } static ocsd_datapath_resp_t -cs_etm_decoder__buffer_packet(struct cs_etm_decoder *decoder, +cs_etm_decoder__buffer_packet(struct cs_etm_packet_queue *packet_queue, const u8 trace_chan_id, enum cs_etm_sample_type sample_type) { u32 et = 0; int cpu; - if (decoder->packet_count >= MAX_BUFFER - 1) + if (packet_queue->packet_count >= CS_ETM_PACKET_MAX_BUFFER - 1) return OCSD_RESP_FATAL_SYS_ERR; if (cs_etm__get_cpu(trace_chan_id, &cpu) < 0) return OCSD_RESP_FATAL_SYS_ERR; - et = decoder->tail; - et = (et + 1) & (MAX_BUFFER - 1); - decoder->tail = et; - decoder->packet_count++; - - decoder->packet_buffer[et].sample_type = sample_type; - decoder->packet_buffer[et].isa = CS_ETM_ISA_UNKNOWN; - decoder->packet_buffer[et].cpu = cpu; - decoder->packet_buffer[et].start_addr = CS_ETM_INVAL_ADDR; - decoder->packet_buffer[et].end_addr = CS_ETM_INVAL_ADDR; - decoder->packet_buffer[et].instr_count = 0; - decoder->packet_buffer[et].last_instr_taken_branch = false; - decoder->packet_buffer[et].last_instr_size = 0; - decoder->packet_buffer[et].last_instr_type = 0; - decoder->packet_buffer[et].last_instr_subtype = 0; - decoder->packet_buffer[et].last_instr_cond = 0; - decoder->packet_buffer[et].flags = 0; - decoder->packet_buffer[et].exception_number = UINT32_MAX; - decoder->packet_buffer[et].trace_chan_id = trace_chan_id; - - if (decoder->packet_count == MAX_BUFFER - 1) + et = packet_queue->tail; + et = (et + 1) & (CS_ETM_PACKET_MAX_BUFFER - 1); + packet_queue->tail = et; + packet_queue->packet_count++; + + packet_queue->packet_buffer[et].sample_type = sample_type; + packet_queue->packet_buffer[et].isa = CS_ETM_ISA_UNKNOWN; + packet_queue->packet_buffer[et].cpu = cpu; + packet_queue->packet_buffer[et].start_addr = CS_ETM_INVAL_ADDR; + packet_queue->packet_buffer[et].end_addr = CS_ETM_INVAL_ADDR; + packet_queue->packet_buffer[et].instr_count = 0; + packet_queue->packet_buffer[et].last_instr_taken_branch = false; + packet_queue->packet_buffer[et].last_instr_size = 0; + packet_queue->packet_buffer[et].last_instr_type = 0; + packet_queue->packet_buffer[et].last_instr_subtype = 0; + packet_queue->packet_buffer[et].last_instr_cond = 0; + packet_queue->packet_buffer[et].flags = 0; + packet_queue->packet_buffer[et].exception_number = UINT32_MAX; + packet_queue->packet_buffer[et].trace_chan_id = trace_chan_id; + + if (packet_queue->packet_count == CS_ETM_PACKET_MAX_BUFFER - 1) return OCSD_RESP_WAIT; return OCSD_RESP_CONT; } static ocsd_datapath_resp_t -cs_etm_decoder__buffer_range(struct cs_etm_decoder *decoder, +cs_etm_decoder__buffer_range(struct cs_etm_queue *etmq, + struct cs_etm_packet_queue *packet_queue, const ocsd_generic_trace_elem *elem, const uint8_t trace_chan_id) { int ret = 0; struct cs_etm_packet *packet; - ret = cs_etm_decoder__buffer_packet(decoder, trace_chan_id, + ret = cs_etm_decoder__buffer_packet(packet_queue, trace_chan_id, CS_ETM_RANGE); if (ret != OCSD_RESP_CONT && ret != OCSD_RESP_WAIT) return ret; - packet = &decoder->packet_buffer[decoder->tail]; + packet = &packet_queue->packet_buffer[packet_queue->tail]; switch (elem->isa) { case ocsd_isa_aarch64: @@ -396,43 +435,90 @@ cs_etm_decoder__buffer_range(struct cs_etm_decoder *decoder, packet->last_instr_size = elem->last_instr_sz; + /* per-thread scenario, no need to generate a timestamp */ + if (cs_etm__etmq_is_timeless(etmq)) + goto out; + + /* + * The packet queue is full and we haven't seen a timestamp (had we + * seen one the packet queue wouldn't be full). Let the front end + * deal with it. + */ + if (ret == OCSD_RESP_WAIT) + goto out; + + packet_queue->instr_count += elem->num_instr_range; + /* Tell the front end we have a new timestamp to process */ + ret = cs_etm_decoder__do_soft_timestamp(etmq, packet_queue, + trace_chan_id); +out: return ret; } static ocsd_datapath_resp_t -cs_etm_decoder__buffer_discontinuity(struct cs_etm_decoder *decoder, - const uint8_t trace_chan_id) +cs_etm_decoder__buffer_discontinuity(struct cs_etm_packet_queue *queue, + const uint8_t trace_chan_id) { - return cs_etm_decoder__buffer_packet(decoder, trace_chan_id, + /* + * Something happened and who knows when we'll get new traces so + * reset time statistics. + */ + cs_etm_decoder__reset_timestamp(queue); + return cs_etm_decoder__buffer_packet(queue, trace_chan_id, CS_ETM_DISCONTINUITY); } static ocsd_datapath_resp_t -cs_etm_decoder__buffer_exception(struct cs_etm_decoder *decoder, +cs_etm_decoder__buffer_exception(struct cs_etm_packet_queue *queue, const ocsd_generic_trace_elem *elem, const uint8_t trace_chan_id) { int ret = 0; struct cs_etm_packet *packet; - ret = cs_etm_decoder__buffer_packet(decoder, trace_chan_id, + ret = cs_etm_decoder__buffer_packet(queue, trace_chan_id, CS_ETM_EXCEPTION); if (ret != OCSD_RESP_CONT && ret != OCSD_RESP_WAIT) return ret; - packet = &decoder->packet_buffer[decoder->tail]; + packet = &queue->packet_buffer[queue->tail]; packet->exception_number = elem->exception_number; return ret; } static ocsd_datapath_resp_t -cs_etm_decoder__buffer_exception_ret(struct cs_etm_decoder *decoder, +cs_etm_decoder__buffer_exception_ret(struct cs_etm_packet_queue *queue, const uint8_t trace_chan_id) { - return cs_etm_decoder__buffer_packet(decoder, trace_chan_id, + return cs_etm_decoder__buffer_packet(queue, trace_chan_id, CS_ETM_EXCEPTION_RET); } +static ocsd_datapath_resp_t +cs_etm_decoder__set_tid(struct cs_etm_queue *etmq, + struct cs_etm_packet_queue *packet_queue, + const ocsd_generic_trace_elem *elem, + const uint8_t trace_chan_id) +{ + pid_t tid; + + /* Ignore PE_CONTEXT packets that don't have a valid contextID */ + if (!elem->context.ctxt_id_valid) + return OCSD_RESP_CONT; + + tid = elem->context.context_id; + if (cs_etm__etmq_set_tid(etmq, tid, trace_chan_id)) + return OCSD_RESP_FATAL_SYS_ERR; + + /* + * A timestamp is generated after a PE_CONTEXT element so make sure + * to rely on that coming one. + */ + cs_etm_decoder__reset_timestamp(packet_queue); + + return OCSD_RESP_CONT; +} + static ocsd_datapath_resp_t cs_etm_decoder__gen_trace_elem_printer( const void *context, const ocsd_trc_index_t indx __maybe_unused, @@ -441,6 +527,13 @@ static ocsd_datapath_resp_t cs_etm_decoder__gen_trace_elem_printer( { ocsd_datapath_resp_t resp = OCSD_RESP_CONT; struct cs_etm_decoder *decoder = (struct cs_etm_decoder *) context; + struct cs_etm_queue *etmq = decoder->data; + struct cs_etm_packet_queue *packet_queue; + + /* First get the packet queue for this traceID */ + packet_queue = cs_etm__etmq_get_packet_queue(etmq, trace_chan_id); + if (!packet_queue) + return OCSD_RESP_FATAL_SYS_ERR; switch (elem->elem_type) { case OCSD_GEN_TRC_ELEM_UNKNOWN: @@ -448,24 +541,30 @@ static ocsd_datapath_resp_t cs_etm_decoder__gen_trace_elem_printer( case OCSD_GEN_TRC_ELEM_EO_TRACE: case OCSD_GEN_TRC_ELEM_NO_SYNC: case OCSD_GEN_TRC_ELEM_TRACE_ON: - resp = cs_etm_decoder__buffer_discontinuity(decoder, + resp = cs_etm_decoder__buffer_discontinuity(packet_queue, trace_chan_id); break; case OCSD_GEN_TRC_ELEM_INSTR_RANGE: - resp = cs_etm_decoder__buffer_range(decoder, elem, + resp = cs_etm_decoder__buffer_range(etmq, packet_queue, elem, trace_chan_id); break; case OCSD_GEN_TRC_ELEM_EXCEPTION: - resp = cs_etm_decoder__buffer_exception(decoder, elem, + resp = cs_etm_decoder__buffer_exception(packet_queue, elem, trace_chan_id); break; case OCSD_GEN_TRC_ELEM_EXCEPTION_RET: - resp = cs_etm_decoder__buffer_exception_ret(decoder, + resp = cs_etm_decoder__buffer_exception_ret(packet_queue, trace_chan_id); break; + case OCSD_GEN_TRC_ELEM_TIMESTAMP: + resp = cs_etm_decoder__do_hard_timestamp(etmq, elem, + trace_chan_id); + break; case OCSD_GEN_TRC_ELEM_PE_CONTEXT: + resp = cs_etm_decoder__set_tid(etmq, packet_queue, + elem, trace_chan_id); + break; case OCSD_GEN_TRC_ELEM_ADDR_NACC: - case OCSD_GEN_TRC_ELEM_TIMESTAMP: case OCSD_GEN_TRC_ELEM_CYCLE_COUNT: case OCSD_GEN_TRC_ELEM_ADDR_UNKNOWN: case OCSD_GEN_TRC_ELEM_EVENT: @@ -554,7 +653,6 @@ cs_etm_decoder__new(int num_cpu, struct cs_etm_decoder_params *d_params, decoder->data = d_params->data; decoder->prev_return = OCSD_RESP_CONT; - cs_etm_decoder__clear_buffer(decoder); format = (d_params->formatted ? OCSD_TRC_SRC_FRAME_FORMATTED : OCSD_TRC_SRC_SINGLE); flags = 0; @@ -577,7 +675,7 @@ cs_etm_decoder__new(int num_cpu, struct cs_etm_decoder_params *d_params, /* init library print logging support */ ret = cs_etm_decoder__init_def_logger_printing(d_params, decoder); if (ret != 0) - goto err_free_decoder_tree; + goto err_free_decoder; /* init raw frame logging if required */ cs_etm_decoder__init_raw_frame_logging(d_params, decoder); @@ -587,15 +685,13 @@ cs_etm_decoder__new(int num_cpu, struct cs_etm_decoder_params *d_params, &t_params[i], decoder); if (ret != 0) - goto err_free_decoder_tree; + goto err_free_decoder; } return decoder; -err_free_decoder_tree: - ocsd_destroy_dcd_tree(decoder->dcd_tree); err_free_decoder: - free(decoder); + cs_etm_decoder__free(decoder); return NULL; } diff --git a/tools/perf/util/cs-etm-decoder/cs-etm-decoder.h b/tools/perf/util/cs-etm-decoder/cs-etm-decoder.h index 3ab11dfa92ae..11f3391d06f2 100644 --- a/tools/perf/util/cs-etm-decoder/cs-etm-decoder.h +++ b/tools/perf/util/cs-etm-decoder/cs-etm-decoder.h @@ -14,43 +14,12 @@ #include <stdio.h> struct cs_etm_decoder; - -enum cs_etm_sample_type { - CS_ETM_EMPTY, - CS_ETM_RANGE, - CS_ETM_DISCONTINUITY, - CS_ETM_EXCEPTION, - CS_ETM_EXCEPTION_RET, -}; - -enum cs_etm_isa { - CS_ETM_ISA_UNKNOWN, - CS_ETM_ISA_A64, - CS_ETM_ISA_A32, - CS_ETM_ISA_T32, -}; - -struct cs_etm_packet { - enum cs_etm_sample_type sample_type; - enum cs_etm_isa isa; - u64 start_addr; - u64 end_addr; - u32 instr_count; - u32 last_instr_type; - u32 last_instr_subtype; - u32 flags; - u32 exception_number; - u8 last_instr_cond; - u8 last_instr_taken_branch; - u8 last_instr_size; - u8 trace_chan_id; - int cpu; -}; +struct cs_etm_packet; +struct cs_etm_packet_queue; struct cs_etm_queue; -typedef u32 (*cs_etm_mem_cb_type)(struct cs_etm_queue *, u64, - size_t, u8 *); +typedef u32 (*cs_etm_mem_cb_type)(struct cs_etm_queue *, u8, u64, size_t, u8 *); struct cs_etmv3_trace_params { u32 reg_ctrl; @@ -119,7 +88,7 @@ int cs_etm_decoder__add_mem_access_cb(struct cs_etm_decoder *decoder, u64 start, u64 end, cs_etm_mem_cb_type cb_func); -int cs_etm_decoder__get_packet(struct cs_etm_decoder *decoder, +int cs_etm_decoder__get_packet(struct cs_etm_packet_queue *packet_queue, struct cs_etm_packet *packet); int cs_etm_decoder__reset(struct cs_etm_decoder *decoder); diff --git a/tools/perf/util/cs-etm.c b/tools/perf/util/cs-etm.c index de488b43f440..0c7776b51045 100644 --- a/tools/perf/util/cs-etm.c +++ b/tools/perf/util/cs-etm.c @@ -29,6 +29,7 @@ #include "thread.h" #include "thread_map.h" #include "thread-stack.h" +#include <tools/libc_compat.h> #include "util.h" #define MAX_TIMESTAMP (~0ULL) @@ -60,33 +61,55 @@ struct cs_etm_auxtrace { unsigned int pmu_type; }; -struct cs_etm_queue { - struct cs_etm_auxtrace *etm; - struct thread *thread; - struct cs_etm_decoder *decoder; - struct auxtrace_buffer *buffer; - union perf_event *event_buf; - unsigned int queue_nr; +struct cs_etm_traceid_queue { + u8 trace_chan_id; pid_t pid, tid; - int cpu; - u64 offset; u64 period_instructions; + size_t last_branch_pos; + union perf_event *event_buf; + struct thread *thread; struct branch_stack *last_branch; struct branch_stack *last_branch_rb; - size_t last_branch_pos; struct cs_etm_packet *prev_packet; struct cs_etm_packet *packet; + struct cs_etm_packet_queue packet_queue; +}; + +struct cs_etm_queue { + struct cs_etm_auxtrace *etm; + struct cs_etm_decoder *decoder; + struct auxtrace_buffer *buffer; + unsigned int queue_nr; + u8 pending_timestamp; + u64 offset; const unsigned char *buf; size_t buf_len, buf_used; + /* Conversion between traceID and index in traceid_queues array */ + struct intlist *traceid_queues_list; + struct cs_etm_traceid_queue **traceid_queues; }; static int cs_etm__update_queues(struct cs_etm_auxtrace *etm); +static int cs_etm__process_queues(struct cs_etm_auxtrace *etm); static int cs_etm__process_timeless_queues(struct cs_etm_auxtrace *etm, pid_t tid); +static int cs_etm__get_data_block(struct cs_etm_queue *etmq); +static int cs_etm__decode_data_block(struct cs_etm_queue *etmq); /* PTMs ETMIDR [11:8] set to b0011 */ #define ETMIDR_PTM_VERSION 0x00000300 +/* + * A struct auxtrace_heap_item only has a queue_nr and a timestamp to + * work with. One option is to modify to auxtrace_heap_XYZ() API or simply + * encode the etm queue number as the upper 16 bit and the channel as + * the lower 16 bit. + */ +#define TO_CS_QUEUE_NR(queue_nr, trace_id_chan) \ + (queue_nr << 16 | trace_chan_id) +#define TO_QUEUE_NR(cs_queue_nr) (cs_queue_nr >> 16) +#define TO_TRACE_CHAN_ID(cs_queue_nr) (cs_queue_nr & 0x0000ffff) + static u32 cs_etm__get_v7_protocol_version(u32 etmidr) { etmidr &= ETMIDR_PTM_VERSION; @@ -125,6 +148,216 @@ int cs_etm__get_cpu(u8 trace_chan_id, int *cpu) return 0; } +void cs_etm__etmq_set_traceid_queue_timestamp(struct cs_etm_queue *etmq, + u8 trace_chan_id) +{ + /* + * Wnen a timestamp packet is encountered the backend code + * is stopped so that the front end has time to process packets + * that were accumulated in the traceID queue. Since there can + * be more than one channel per cs_etm_queue, we need to specify + * what traceID queue needs servicing. + */ + etmq->pending_timestamp = trace_chan_id; +} + +static u64 cs_etm__etmq_get_timestamp(struct cs_etm_queue *etmq, + u8 *trace_chan_id) +{ + struct cs_etm_packet_queue *packet_queue; + + if (!etmq->pending_timestamp) + return 0; + + if (trace_chan_id) + *trace_chan_id = etmq->pending_timestamp; + + packet_queue = cs_etm__etmq_get_packet_queue(etmq, + etmq->pending_timestamp); + if (!packet_queue) + return 0; + + /* Acknowledge pending status */ + etmq->pending_timestamp = 0; + + /* See function cs_etm_decoder__do_{hard|soft}_timestamp() */ + return packet_queue->timestamp; +} + +static void cs_etm__clear_packet_queue(struct cs_etm_packet_queue *queue) +{ + int i; + + queue->head = 0; + queue->tail = 0; + queue->packet_count = 0; + for (i = 0; i < CS_ETM_PACKET_MAX_BUFFER; i++) { + queue->packet_buffer[i].isa = CS_ETM_ISA_UNKNOWN; + queue->packet_buffer[i].start_addr = CS_ETM_INVAL_ADDR; + queue->packet_buffer[i].end_addr = CS_ETM_INVAL_ADDR; + queue->packet_buffer[i].instr_count = 0; + queue->packet_buffer[i].last_instr_taken_branch = false; + queue->packet_buffer[i].last_instr_size = 0; + queue->packet_buffer[i].last_instr_type = 0; + queue->packet_buffer[i].last_instr_subtype = 0; + queue->packet_buffer[i].last_instr_cond = 0; + queue->packet_buffer[i].flags = 0; + queue->packet_buffer[i].exception_number = UINT32_MAX; + queue->packet_buffer[i].trace_chan_id = UINT8_MAX; + queue->packet_buffer[i].cpu = INT_MIN; + } +} + +static void cs_etm__clear_all_packet_queues(struct cs_etm_queue *etmq) +{ + int idx; + struct int_node *inode; + struct cs_etm_traceid_queue *tidq; + struct intlist *traceid_queues_list = etmq->traceid_queues_list; + + intlist__for_each_entry(inode, traceid_queues_list) { + idx = (int)(intptr_t)inode->priv; + tidq = etmq->traceid_queues[idx]; + cs_etm__clear_packet_queue(&tidq->packet_queue); + } +} + +static int cs_etm__init_traceid_queue(struct cs_etm_queue *etmq, + struct cs_etm_traceid_queue *tidq, + u8 trace_chan_id) +{ + int rc = -ENOMEM; + struct auxtrace_queue *queue; + struct cs_etm_auxtrace *etm = etmq->etm; + + cs_etm__clear_packet_queue(&tidq->packet_queue); + + queue = &etmq->etm->queues.queue_array[etmq->queue_nr]; + tidq->tid = queue->tid; + tidq->pid = -1; + tidq->trace_chan_id = trace_chan_id; + + tidq->packet = zalloc(sizeof(struct cs_etm_packet)); + if (!tidq->packet) + goto out; + + tidq->prev_packet = zalloc(sizeof(struct cs_etm_packet)); + if (!tidq->prev_packet) + goto out_free; + + if (etm->synth_opts.last_branch) { + size_t sz = sizeof(struct branch_stack); + + sz += etm->synth_opts.last_branch_sz * + sizeof(struct branch_entry); + tidq->last_branch = zalloc(sz); + if (!tidq->last_branch) + goto out_free; + tidq->last_branch_rb = zalloc(sz); + if (!tidq->last_branch_rb) + goto out_free; + } + + tidq->event_buf = malloc(PERF_SAMPLE_MAX_SIZE); + if (!tidq->event_buf) + goto out_free; + + return 0; + +out_free: + zfree(&tidq->last_branch_rb); + zfree(&tidq->last_branch); + zfree(&tidq->prev_packet); + zfree(&tidq->packet); +out: + return rc; +} + +static struct cs_etm_traceid_queue +*cs_etm__etmq_get_traceid_queue(struct cs_etm_queue *etmq, u8 trace_chan_id) +{ + int idx; + struct int_node *inode; + struct intlist *traceid_queues_list; + struct cs_etm_traceid_queue *tidq, **traceid_queues; + struct cs_etm_auxtrace *etm = etmq->etm; + + if (etm->timeless_decoding) + trace_chan_id = CS_ETM_PER_THREAD_TRACEID; + + traceid_queues_list = etmq->traceid_queues_list; + + /* + * Check if the traceid_queue exist for this traceID by looking + * in the queue list. + */ + inode = intlist__find(traceid_queues_list, trace_chan_id); + if (inode) { + idx = (int)(intptr_t)inode->priv; + return etmq->traceid_queues[idx]; + } + + /* We couldn't find a traceid_queue for this traceID, allocate one */ + tidq = malloc(sizeof(*tidq)); + if (!tidq) + return NULL; + + memset(tidq, 0, sizeof(*tidq)); + + /* Get a valid index for the new traceid_queue */ + idx = intlist__nr_entries(traceid_queues_list); + /* Memory for the inode is free'ed in cs_etm_free_traceid_queues () */ + inode = intlist__findnew(traceid_queues_list, trace_chan_id); + if (!inode) + goto out_free; + + /* Associate this traceID with this index */ + inode->priv = (void *)(intptr_t)idx; + + if (cs_etm__init_traceid_queue(etmq, tidq, trace_chan_id)) + goto out_free; + + /* Grow the traceid_queues array by one unit */ + traceid_queues = etmq->traceid_queues; + traceid_queues = reallocarray(traceid_queues, + idx + 1, + sizeof(*traceid_queues)); + + /* + * On failure reallocarray() returns NULL and the original block of + * memory is left untouched. + */ + if (!traceid_queues) + goto out_free; + + traceid_queues[idx] = tidq; + etmq->traceid_queues = traceid_queues; + + return etmq->traceid_queues[idx]; + +out_free: + /* + * Function intlist__remove() removes the inode from the list + * and delete the memory associated to it. + */ + intlist__remove(traceid_queues_list, inode); + free(tidq); + + return NULL; +} + +struct cs_etm_packet_queue +*cs_etm__etmq_get_packet_queue(struct cs_etm_queue *etmq, u8 trace_chan_id) +{ + struct cs_etm_traceid_queue *tidq; + + tidq = cs_etm__etmq_get_traceid_queue(etmq, trace_chan_id); + if (tidq) + return &tidq->packet_queue; + + return NULL; +} + static void cs_etm__packet_dump(const char *pkt_string) { const char *color = PERF_COLOR_BLUE; @@ -276,15 +509,53 @@ static int cs_etm__flush_events(struct perf_session *session, if (!tool->ordered_events) return -EINVAL; - if (!etm->timeless_decoding) - return -EINVAL; - ret = cs_etm__update_queues(etm); if (ret < 0) return ret; - return cs_etm__process_timeless_queues(etm, -1); + if (etm->timeless_decoding) + return cs_etm__process_timeless_queues(etm, -1); + + return cs_etm__process_queues(etm); +} + +static void cs_etm__free_traceid_queues(struct cs_etm_queue *etmq) +{ + int idx; + uintptr_t priv; + struct int_node *inode, *tmp; + struct cs_etm_traceid_queue *tidq; + struct intlist *traceid_queues_list = etmq->traceid_queues_list; + + intlist__for_each_entry_safe(inode, tmp, traceid_queues_list) { + priv = (uintptr_t)inode->priv; + idx = priv; + + /* Free this traceid_queue from the array */ + tidq = etmq->traceid_queues[idx]; + thread__zput(tidq->thread); + zfree(&tidq->event_buf); + zfree(&tidq->last_branch); + zfree(&tidq->last_branch_rb); + zfree(&tidq->prev_packet); + zfree(&tidq->packet); + zfree(&tidq); + + /* + * Function intlist__remove() removes the inode from the list + * and delete the memory associated to it. + */ + intlist__remove(traceid_queues_list, inode); + } + + /* Then the RB tree itself */ + intlist__delete(traceid_queues_list); + etmq->traceid_queues_list = NULL; + + /* finally free the traceid_queues array */ + free(etmq->traceid_queues); + etmq->traceid_queues = NULL; } static void cs_etm__free_queue(void *priv) @@ -294,13 +565,8 @@ static void cs_etm__free_queue(void *priv) if (!etmq) return; - thread__zput(etmq->thread); cs_etm_decoder__free(etmq->decoder); - zfree(&etmq->event_buf); - zfree(&etmq->last_branch); - zfree(&etmq->last_branch_rb); - zfree(&etmq->prev_packet); - zfree(&etmq->packet); + cs_etm__free_traceid_queues(etmq); free(etmq); } @@ -365,23 +631,27 @@ static u8 cs_etm__cpu_mode(struct cs_etm_queue *etmq, u64 address) } } -static u32 cs_etm__mem_access(struct cs_etm_queue *etmq, u64 address, - size_t size, u8 *buffer) +static u32 cs_etm__mem_access(struct cs_etm_queue *etmq, u8 trace_chan_id, + u64 address, size_t size, u8 *buffer) { u8 cpumode; u64 offset; int len; - struct thread *thread; - struct machine *machine; - struct addr_location al; + struct thread *thread; + struct machine *machine; + struct addr_location al; + struct cs_etm_traceid_queue *tidq; if (!etmq) return 0; machine = etmq->etm->machine; cpumode = cs_etm__cpu_mode(etmq, address); + tidq = cs_etm__etmq_get_traceid_queue(etmq, trace_chan_id); + if (!tidq) + return 0; - thread = etmq->thread; + thread = tidq->thread; if (!thread) { if (cpumode != PERF_RECORD_MISC_KERNEL) return 0; @@ -412,35 +682,13 @@ static struct cs_etm_queue *cs_etm__alloc_queue(struct cs_etm_auxtrace *etm) struct cs_etm_decoder_params d_params; struct cs_etm_trace_params *t_params = NULL; struct cs_etm_queue *etmq; - size_t szp = sizeof(struct cs_etm_packet); etmq = zalloc(sizeof(*etmq)); if (!etmq) return NULL; - etmq->packet = zalloc(szp); - if (!etmq->packet) - goto out_free; - - etmq->prev_packet = zalloc(szp); - if (!etmq->prev_packet) - goto out_free; - - if (etm->synth_opts.last_branch) { - size_t sz = sizeof(struct branch_stack); - - sz += etm->synth_opts.last_branch_sz * - sizeof(struct branch_entry); - etmq->last_branch = zalloc(sz); - if (!etmq->last_branch) - goto out_free; - etmq->last_branch_rb = zalloc(sz); - if (!etmq->last_branch_rb) - goto out_free; - } - - etmq->event_buf = malloc(PERF_SAMPLE_MAX_SIZE); - if (!etmq->event_buf) + etmq->traceid_queues_list = intlist__new(NULL); + if (!etmq->traceid_queues_list) goto out_free; /* Use metadata to fill in trace parameters for trace decoder */ @@ -477,12 +725,7 @@ static struct cs_etm_queue *cs_etm__alloc_queue(struct cs_etm_auxtrace *etm) out_free_decoder: cs_etm_decoder__free(etmq->decoder); out_free: - zfree(&t_params); - zfree(&etmq->event_buf); - zfree(&etmq->last_branch); - zfree(&etmq->last_branch_rb); - zfree(&etmq->prev_packet); - zfree(&etmq->packet); + intlist__delete(etmq->traceid_queues_list); free(etmq); return NULL; @@ -493,6 +736,9 @@ static int cs_etm__setup_queue(struct cs_etm_auxtrace *etm, unsigned int queue_nr) { int ret = 0; + unsigned int cs_queue_nr; + u8 trace_chan_id; + u64 timestamp; struct cs_etm_queue *etmq = queue->priv; if (list_empty(&queue->head) || etmq) @@ -508,12 +754,69 @@ static int cs_etm__setup_queue(struct cs_etm_auxtrace *etm, queue->priv = etmq; etmq->etm = etm; etmq->queue_nr = queue_nr; - etmq->cpu = queue->cpu; - etmq->tid = queue->tid; - etmq->pid = -1; etmq->offset = 0; - etmq->period_instructions = 0; + if (etm->timeless_decoding) + goto out; + + /* + * We are under a CPU-wide trace scenario. As such we need to know + * when the code that generated the traces started to execute so that + * it can be correlated with execution on other CPUs. So we get a + * handle on the beginning of traces and decode until we find a + * timestamp. The timestamp is then added to the auxtrace min heap + * in order to know what nibble (of all the etmqs) to decode first. + */ + while (1) { + /* + * Fetch an aux_buffer from this etmq. Bail if no more + * blocks or an error has been encountered. + */ + ret = cs_etm__get_data_block(etmq); + if (ret <= 0) + goto out; + + /* + * Run decoder on the trace block. The decoder will stop when + * encountering a timestamp, a full packet queue or the end of + * trace for that block. + */ + ret = cs_etm__decode_data_block(etmq); + if (ret) + goto out; + + /* + * Function cs_etm_decoder__do_{hard|soft}_timestamp() does all + * the timestamp calculation for us. + */ + timestamp = cs_etm__etmq_get_timestamp(etmq, &trace_chan_id); + + /* We found a timestamp, no need to continue. */ + if (timestamp) + break; + + /* + * We didn't find a timestamp so empty all the traceid packet + * queues before looking for another timestamp packet, either + * in the current data block or a new one. Packets that were + * just decoded are useless since no timestamp has been + * associated with them. As such simply discard them. + */ + cs_etm__clear_all_packet_queues(etmq); + } + + /* + * We have a timestamp. Add it to the min heap to reflect when + * instructions conveyed by the range packets of this traceID queue + * started to execute. Once the same has been done for all the traceID + * queues of each etmq, redenring and decoding can start in + * chronological order. + * + * Note that packets decoded above are still in the traceID's packet + * queue and will be processed in cs_etm__process_queues(). + */ + cs_queue_nr = TO_CS_QUEUE_NR(queue_nr, trace_id_chan); + ret = auxtrace_heap__add(&etm->heap, cs_queue_nr, timestamp); out: return ret; } @@ -545,10 +848,12 @@ static int cs_etm__update_queues(struct cs_etm_auxtrace *etm) return 0; } -static inline void cs_etm__copy_last_branch_rb(struct cs_etm_queue *etmq) +static inline +void cs_etm__copy_last_branch_rb(struct cs_etm_queue *etmq, + struct cs_etm_traceid_queue *tidq) { - struct branch_stack *bs_src = etmq->last_branch_rb; - struct branch_stack *bs_dst = etmq->last_branch; + struct branch_stack *bs_src = tidq->last_branch_rb; + struct branch_stack *bs_dst = tidq->last_branch; size_t nr = 0; /* @@ -568,9 +873,9 @@ static inline void cs_etm__copy_last_branch_rb(struct cs_etm_queue *etmq) * two steps. First, copy the branches from the most recently inserted * branch ->last_branch_pos until the end of bs_src->entries buffer. */ - nr = etmq->etm->synth_opts.last_branch_sz - etmq->last_branch_pos; + nr = etmq->etm->synth_opts.last_branch_sz - tidq->last_branch_pos; memcpy(&bs_dst->entries[0], - &bs_src->entries[etmq->last_branch_pos], + &bs_src->entries[tidq->last_branch_pos], sizeof(struct branch_entry) * nr); /* @@ -583,21 +888,24 @@ static inline void cs_etm__copy_last_branch_rb(struct cs_etm_queue *etmq) if (bs_src->nr >= etmq->etm->synth_opts.last_branch_sz) { memcpy(&bs_dst->entries[nr], &bs_src->entries[0], - sizeof(struct branch_entry) * etmq->last_branch_pos); + sizeof(struct branch_entry) * tidq->last_branch_pos); } } -static inline void cs_etm__reset_last_branch_rb(struct cs_etm_queue *etmq) +static inline +void cs_etm__reset_last_branch_rb(struct cs_etm_traceid_queue *tidq) { - etmq->last_branch_pos = 0; - etmq->last_branch_rb->nr = 0; + tidq->last_branch_pos = 0; + tidq->last_branch_rb->nr = 0; } static inline int cs_etm__t32_instr_size(struct cs_etm_queue *etmq, - u64 addr) { + u8 trace_chan_id, u64 addr) +{ u8 instrBytes[2]; - cs_etm__mem_access(etmq, addr, ARRAY_SIZE(instrBytes), instrBytes); + cs_etm__mem_access(etmq, trace_chan_id, addr, + ARRAY_SIZE(instrBytes), instrBytes); /* * T32 instruction size is indicated by bits[15:11] of the first * 16-bit word of the instruction: 0b11101, 0b11110 and 0b11111 @@ -626,6 +934,7 @@ u64 cs_etm__last_executed_instr(const struct cs_etm_packet *packet) } static inline u64 cs_etm__instr_addr(struct cs_etm_queue *etmq, + u64 trace_chan_id, const struct cs_etm_packet *packet, u64 offset) { @@ -633,7 +942,8 @@ static inline u64 cs_etm__instr_addr(struct cs_etm_queue *etmq, u64 addr = packet->start_addr; while (offset > 0) { - addr += cs_etm__t32_instr_size(etmq, addr); + addr += cs_etm__t32_instr_size(etmq, + trace_chan_id, addr); offset--; } return addr; @@ -643,9 +953,10 @@ static inline u64 cs_etm__instr_addr(struct cs_etm_queue *etmq, return packet->start_addr + offset * 4; } -static void cs_etm__update_last_branch_rb(struct cs_etm_queue *etmq) +static void cs_etm__update_last_branch_rb(struct cs_etm_queue *etmq, + struct cs_etm_traceid_queue *tidq) { - struct branch_stack *bs = etmq->last_branch_rb; + struct branch_stack *bs = tidq->last_branch_rb; struct branch_entry *be; /* @@ -654,14 +965,14 @@ static void cs_etm__update_last_branch_rb(struct cs_etm_queue *etmq) * buffer down. After writing the first element of the stack, move the * insert position back to the end of the buffer. */ - if (!etmq->last_branch_pos) - etmq->last_branch_pos = etmq->etm->synth_opts.last_branch_sz; + if (!tidq->last_branch_pos) + tidq->last_branch_pos = etmq->etm->synth_opts.last_branch_sz; - etmq->last_branch_pos -= 1; + tidq->last_branch_pos -= 1; - be = &bs->entries[etmq->last_branch_pos]; - be->from = cs_etm__last_executed_instr(etmq->prev_packet); - be->to = cs_etm__first_executed_instr(etmq->packet); + be = &bs->entries[tidq->last_branch_pos]; + be->from = cs_etm__last_executed_instr(tidq->prev_packet); + be->to = cs_etm__first_executed_instr(tidq->packet); /* No support for mispredict */ be->flags.mispred = 0; be->flags.predicted = 1; @@ -725,31 +1036,53 @@ cs_etm__get_trace(struct cs_etm_queue *etmq) } static void cs_etm__set_pid_tid_cpu(struct cs_etm_auxtrace *etm, - struct auxtrace_queue *queue) + struct cs_etm_traceid_queue *tidq) { - struct cs_etm_queue *etmq = queue->priv; + if ((!tidq->thread) && (tidq->tid != -1)) + tidq->thread = machine__find_thread(etm->machine, -1, + tidq->tid); - /* CPU-wide tracing isn't supported yet */ - if (queue->tid == -1) - return; + if (tidq->thread) + tidq->pid = tidq->thread->pid_; +} - if ((!etmq->thread) && (etmq->tid != -1)) - etmq->thread = machine__find_thread(etm->machine, -1, - etmq->tid); +int cs_etm__etmq_set_tid(struct cs_etm_queue *etmq, + pid_t tid, u8 trace_chan_id) +{ + int cpu, err = -EINVAL; + struct cs_etm_auxtrace *etm = etmq->etm; + struct cs_etm_traceid_queue *tidq; - if (etmq->thread) { - etmq->pid = etmq->thread->pid_; - if (queue->cpu == -1) - etmq->cpu = etmq->thread->cpu; - } + tidq = cs_etm__etmq_get_traceid_queue(etmq, trace_chan_id); + if (!tidq) + return err; + + if (cs_etm__get_cpu(trace_chan_id, &cpu) < 0) + return err; + + err = machine__set_current_tid(etm->machine, cpu, tid, tid); + if (err) + return err; + + tidq->tid = tid; + thread__zput(tidq->thread); + + cs_etm__set_pid_tid_cpu(etm, tidq); + return 0; +} + +bool cs_etm__etmq_is_timeless(struct cs_etm_queue *etmq) +{ + return !!etmq->etm->timeless_decoding; } static int cs_etm__synth_instruction_sample(struct cs_etm_queue *etmq, + struct cs_etm_traceid_queue *tidq, u64 addr, u64 period) { int ret = 0; struct cs_etm_auxtrace *etm = etmq->etm; - union perf_event *event = etmq->event_buf; + union perf_event *event = tidq->event_buf; struct perf_sample sample = {.ip = 0,}; event->sample.header.type = PERF_RECORD_SAMPLE; @@ -757,19 +1090,19 @@ static int cs_etm__synth_instruction_sample(struct cs_etm_queue *etmq, event->sample.header.size = sizeof(struct perf_event_header); sample.ip = addr; - sample.pid = etmq->pid; - sample.tid = etmq->tid; + sample.pid = tidq->pid; + sample.tid = tidq->tid; sample.id = etmq->etm->instructions_id; sample.stream_id = etmq->etm->instructions_id; sample.period = period; - sample.cpu = etmq->packet->cpu; - sample.flags = etmq->prev_packet->flags; + sample.cpu = tidq->packet->cpu; + sample.flags = tidq->prev_packet->flags; sample.insn_len = 1; sample.cpumode = event->sample.header.misc; if (etm->synth_opts.last_branch) { - cs_etm__copy_last_branch_rb(etmq); - sample.branch_stack = etmq->last_branch; + cs_etm__copy_last_branch_rb(etmq, tidq); + sample.branch_stack = tidq->last_branch; } if (etm->synth_opts.inject) { @@ -787,7 +1120,7 @@ static int cs_etm__synth_instruction_sample(struct cs_etm_queue *etmq, ret); if (etm->synth_opts.last_branch) - cs_etm__reset_last_branch_rb(etmq); + cs_etm__reset_last_branch_rb(tidq); return ret; } @@ -796,33 +1129,34 @@ static int cs_etm__synth_instruction_sample(struct cs_etm_queue *etmq, * The cs etm packet encodes an instruction range between a branch target * and the next taken branch. Generate sample accordingly. */ -static int cs_etm__synth_branch_sample(struct cs_etm_queue *etmq) +static int cs_etm__synth_branch_sample(struct cs_etm_queue *etmq, + struct cs_etm_traceid_queue *tidq) { int ret = 0; struct cs_etm_auxtrace *etm = etmq->etm; struct perf_sample sample = {.ip = 0,}; - union perf_event *event = etmq->event_buf; + union perf_event *event = tidq->event_buf; struct dummy_branch_stack { u64 nr; struct branch_entry entries; } dummy_bs; u64 ip; - ip = cs_etm__last_executed_instr(etmq->prev_packet); + ip = cs_etm__last_executed_instr(tidq->prev_packet); event->sample.header.type = PERF_RECORD_SAMPLE; event->sample.header.misc = cs_etm__cpu_mode(etmq, ip); event->sample.header.size = sizeof(struct perf_event_header); sample.ip = ip; - sample.pid = etmq->pid; - sample.tid = etmq->tid; - sample.addr = cs_etm__first_executed_instr(etmq->packet); + sample.pid = tidq->pid; + sample.tid = tidq->tid; + sample.addr = cs_etm__first_executed_instr(tidq->packet); sample.id = etmq->etm->branches_id; sample.stream_id = etmq->etm->branches_id; sample.period = 1; - sample.cpu = etmq->packet->cpu; - sample.flags = etmq->prev_packet->flags; + sample.cpu = tidq->packet->cpu; + sample.flags = tidq->prev_packet->flags; sample.cpumode = event->sample.header.misc; /* @@ -965,33 +1299,35 @@ static int cs_etm__synth_events(struct cs_etm_auxtrace *etm, return 0; } -static int cs_etm__sample(struct cs_etm_queue *etmq) +static int cs_etm__sample(struct cs_etm_queue *etmq, + struct cs_etm_traceid_queue *tidq) { struct cs_etm_auxtrace *etm = etmq->etm; struct cs_etm_packet *tmp; int ret; - u64 instrs_executed = etmq->packet->instr_count; + u8 trace_chan_id = tidq->trace_chan_id; + u64 instrs_executed = tidq->packet->instr_count; - etmq->period_instructions += instrs_executed; + tidq->period_instructions += instrs_executed; /* * Record a branch when the last instruction in * PREV_PACKET is a branch. */ if (etm->synth_opts.last_branch && - etmq->prev_packet->sample_type == CS_ETM_RANGE && - etmq->prev_packet->last_instr_taken_branch) - cs_etm__update_last_branch_rb(etmq); + tidq->prev_packet->sample_type == CS_ETM_RANGE && + tidq->prev_packet->last_instr_taken_branch) + cs_etm__update_last_branch_rb(etmq, tidq); if (etm->sample_instructions && - etmq->period_instructions >= etm->instructions_sample_period) { + tidq->period_instructions >= etm->instructions_sample_period) { /* * Emit instruction sample periodically * TODO: allow period to be defined in cycles and clock time */ /* Get number of instructions executed after the sample point */ - u64 instrs_over = etmq->period_instructions - + u64 instrs_over = tidq->period_instructions - etm->instructions_sample_period; /* @@ -1000,31 +1336,32 @@ static int cs_etm__sample(struct cs_etm_queue *etmq) * executed, but PC has not advanced to next instruction) */ u64 offset = (instrs_executed - instrs_over - 1); - u64 addr = cs_etm__instr_addr(etmq, etmq->packet, offset); + u64 addr = cs_etm__instr_addr(etmq, trace_chan_id, + tidq->packet, offset); ret = cs_etm__synth_instruction_sample( - etmq, addr, etm->instructions_sample_period); + etmq, tidq, addr, etm->instructions_sample_period); if (ret) return ret; /* Carry remaining instructions into next sample period */ - etmq->period_instructions = instrs_over; + tidq->period_instructions = instrs_over; } if (etm->sample_branches) { bool generate_sample = false; /* Generate sample for tracing on packet */ - if (etmq->prev_packet->sample_type == CS_ETM_DISCONTINUITY) + if (tidq->prev_packet->sample_type == CS_ETM_DISCONTINUITY) generate_sample = true; /* Generate sample for branch taken packet */ - if (etmq->prev_packet->sample_type == CS_ETM_RANGE && - etmq->prev_packet->last_instr_taken_branch) + if (tidq->prev_packet->sample_type == CS_ETM_RANGE && + tidq->prev_packet->last_instr_taken_branch) generate_sample = true; if (generate_sample) { - ret = cs_etm__synth_branch_sample(etmq); + ret = cs_etm__synth_branch_sample(etmq, tidq); if (ret) return ret; } @@ -1035,15 +1372,15 @@ static int cs_etm__sample(struct cs_etm_queue *etmq) * Swap PACKET with PREV_PACKET: PACKET becomes PREV_PACKET for * the next incoming packet. */ - tmp = etmq->packet; - etmq->packet = etmq->prev_packet; - etmq->prev_packet = tmp; + tmp = tidq->packet; + tidq->packet = tidq->prev_packet; + tidq->prev_packet = tmp; } return 0; } -static int cs_etm__exception(struct cs_etm_queue *etmq) +static int cs_etm__exception(struct cs_etm_traceid_queue *tidq) { /* * When the exception packet is inserted, whether the last instruction @@ -1056,24 +1393,25 @@ static int cs_etm__exception(struct cs_etm_queue *etmq) * swap PACKET with PREV_PACKET. This keeps PREV_PACKET to be useful * for generating instruction and branch samples. */ - if (etmq->prev_packet->sample_type == CS_ETM_RANGE) - etmq->prev_packet->last_instr_taken_branch = true; + if (tidq->prev_packet->sample_type == CS_ETM_RANGE) + tidq->prev_packet->last_instr_taken_branch = true; return 0; } -static int cs_etm__flush(struct cs_etm_queue *etmq) +static int cs_etm__flush(struct cs_etm_queue *etmq, + struct cs_etm_traceid_queue *tidq) { int err = 0; struct cs_etm_auxtrace *etm = etmq->etm; struct cs_etm_packet *tmp; /* Handle start tracing packet */ - if (etmq->prev_packet->sample_type == CS_ETM_EMPTY) + if (tidq->prev_packet->sample_type == CS_ETM_EMPTY) goto swap_packet; if (etmq->etm->synth_opts.last_branch && - etmq->prev_packet->sample_type == CS_ETM_RANGE) { + tidq->prev_packet->sample_type == CS_ETM_RANGE) { /* * Generate a last branch event for the branches left in the * circular buffer at the end of the trace. @@ -1081,21 +1419,21 @@ static int cs_etm__flush(struct cs_etm_queue *etmq) * Use the address of the end of the last reported execution * range */ - u64 addr = cs_etm__last_executed_instr(etmq->prev_packet); + u64 addr = cs_etm__last_executed_instr(tidq->prev_packet); err = cs_etm__synth_instruction_sample( - etmq, addr, - etmq->period_instructions); + etmq, tidq, addr, + tidq->period_instructions); if (err) return err; - etmq->period_instructions = 0; + tidq->period_instructions = 0; } if (etm->sample_branches && - etmq->prev_packet->sample_type == CS_ETM_RANGE) { - err = cs_etm__synth_branch_sample(etmq); + tidq->prev_packet->sample_type == CS_ETM_RANGE) { + err = cs_etm__synth_branch_sample(etmq, tidq); if (err) return err; } @@ -1106,15 +1444,16 @@ swap_packet: * Swap PACKET with PREV_PACKET: PACKET becomes PREV_PACKET for * the next incoming packet. */ - tmp = etmq->packet; - etmq->packet = etmq->prev_packet; - etmq->prev_packet = tmp; + tmp = tidq->packet; + tidq->packet = tidq->prev_packet; + tidq->prev_packet = tmp; } return err; } -static int cs_etm__end_block(struct cs_etm_queue *etmq) +static int cs_etm__end_block(struct cs_etm_queue *etmq, + struct cs_etm_traceid_queue *tidq) { int err; @@ -1128,20 +1467,20 @@ static int cs_etm__end_block(struct cs_etm_queue *etmq) * the trace. */ if (etmq->etm->synth_opts.last_branch && - etmq->prev_packet->sample_type == CS_ETM_RANGE) { + tidq->prev_packet->sample_type == CS_ETM_RANGE) { /* * Use the address of the end of the last reported execution * range. */ - u64 addr = cs_etm__last_executed_instr(etmq->prev_packet); + u64 addr = cs_etm__last_executed_instr(tidq->prev_packet); err = cs_etm__synth_instruction_sample( - etmq, addr, - etmq->period_instructions); + etmq, tidq, addr, + tidq->period_instructions); if (err) return err; - etmq->period_instructions = 0; + tidq->period_instructions = 0; } return 0; @@ -1173,12 +1512,13 @@ static int cs_etm__get_data_block(struct cs_etm_queue *etmq) return etmq->buf_len; } -static bool cs_etm__is_svc_instr(struct cs_etm_queue *etmq, +static bool cs_etm__is_svc_instr(struct cs_etm_queue *etmq, u8 trace_chan_id, struct cs_etm_packet *packet, u64 end_addr) { - u16 instr16; - u32 instr32; + /* Initialise to keep compiler happy */ + u16 instr16 = 0; + u32 instr32 = 0; u64 addr; switch (packet->isa) { @@ -1196,7 +1536,8 @@ static bool cs_etm__is_svc_instr(struct cs_etm_queue *etmq, * so below only read 2 bytes as instruction size for T32. */ addr = end_addr - 2; - cs_etm__mem_access(etmq, addr, sizeof(instr16), (u8 *)&instr16); + cs_etm__mem_access(etmq, trace_chan_id, addr, + sizeof(instr16), (u8 *)&instr16); if ((instr16 & 0xFF00) == 0xDF00) return true; @@ -1211,7 +1552,8 @@ static bool cs_etm__is_svc_instr(struct cs_etm_queue *etmq, * +---------+---------+-------------------------+ */ addr = end_addr - 4; - cs_etm__mem_access(etmq, addr, sizeof(instr32), (u8 *)&instr32); + cs_etm__mem_access(etmq, trace_chan_id, addr, + sizeof(instr32), (u8 *)&instr32); if ((instr32 & 0x0F000000) == 0x0F000000 && (instr32 & 0xF0000000) != 0xF0000000) return true; @@ -1227,7 +1569,8 @@ static bool cs_etm__is_svc_instr(struct cs_etm_queue *etmq, * +-----------------------+---------+-----------+ */ addr = end_addr - 4; - cs_etm__mem_access(etmq, addr, sizeof(instr32), (u8 *)&instr32); + cs_etm__mem_access(etmq, trace_chan_id, addr, + sizeof(instr32), (u8 *)&instr32); if ((instr32 & 0xFFE0001F) == 0xd4000001) return true; @@ -1240,10 +1583,12 @@ static bool cs_etm__is_svc_instr(struct cs_etm_queue *etmq, return false; } -static bool cs_etm__is_syscall(struct cs_etm_queue *etmq, u64 magic) +static bool cs_etm__is_syscall(struct cs_etm_queue *etmq, + struct cs_etm_traceid_queue *tidq, u64 magic) { - struct cs_etm_packet *packet = etmq->packet; - struct cs_etm_packet *prev_packet = etmq->prev_packet; + u8 trace_chan_id = tidq->trace_chan_id; + struct cs_etm_packet *packet = tidq->packet; + struct cs_etm_packet *prev_packet = tidq->prev_packet; if (magic == __perf_cs_etmv3_magic) if (packet->exception_number == CS_ETMV3_EXC_SVC) @@ -1256,7 +1601,7 @@ static bool cs_etm__is_syscall(struct cs_etm_queue *etmq, u64 magic) */ if (magic == __perf_cs_etmv4_magic) { if (packet->exception_number == CS_ETMV4_EXC_CALL && - cs_etm__is_svc_instr(etmq, prev_packet, + cs_etm__is_svc_instr(etmq, trace_chan_id, prev_packet, prev_packet->end_addr)) return true; } @@ -1264,9 +1609,10 @@ static bool cs_etm__is_syscall(struct cs_etm_queue *etmq, u64 magic) return false; } -static bool cs_etm__is_async_exception(struct cs_etm_queue *etmq, u64 magic) +static bool cs_etm__is_async_exception(struct cs_etm_traceid_queue *tidq, + u64 magic) { - struct cs_etm_packet *packet = etmq->packet; + struct cs_etm_packet *packet = tidq->packet; if (magic == __perf_cs_etmv3_magic) if (packet->exception_number == CS_ETMV3_EXC_DEBUG_HALT || @@ -1289,10 +1635,13 @@ static bool cs_etm__is_async_exception(struct cs_etm_queue *etmq, u64 magic) return false; } -static bool cs_etm__is_sync_exception(struct cs_etm_queue *etmq, u64 magic) +static bool cs_etm__is_sync_exception(struct cs_etm_queue *etmq, + struct cs_etm_traceid_queue *tidq, + u64 magic) { - struct cs_etm_packet *packet = etmq->packet; - struct cs_etm_packet *prev_packet = etmq->prev_packet; + u8 trace_chan_id = tidq->trace_chan_id; + struct cs_etm_packet *packet = tidq->packet; + struct cs_etm_packet *prev_packet = tidq->prev_packet; if (magic == __perf_cs_etmv3_magic) if (packet->exception_number == CS_ETMV3_EXC_SMC || @@ -1316,7 +1665,7 @@ static bool cs_etm__is_sync_exception(struct cs_etm_queue *etmq, u64 magic) * (SMC, HVC) are taken as sync exceptions. */ if (packet->exception_number == CS_ETMV4_EXC_CALL && - !cs_etm__is_svc_instr(etmq, prev_packet, + !cs_etm__is_svc_instr(etmq, trace_chan_id, prev_packet, prev_packet->end_addr)) return true; @@ -1335,10 +1684,12 @@ static bool cs_etm__is_sync_exception(struct cs_etm_queue *etmq, u64 magic) return false; } -static int cs_etm__set_sample_flags(struct cs_etm_queue *etmq) +static int cs_etm__set_sample_flags(struct cs_etm_queue *etmq, + struct cs_etm_traceid_queue *tidq) { - struct cs_etm_packet *packet = etmq->packet; - struct cs_etm_packet *prev_packet = etmq->prev_packet; + struct cs_etm_packet *packet = tidq->packet; + struct cs_etm_packet *prev_packet = tidq->prev_packet; + u8 trace_chan_id = tidq->trace_chan_id; u64 magic; int ret; @@ -1419,7 +1770,8 @@ static int cs_etm__set_sample_flags(struct cs_etm_queue *etmq) if (prev_packet->flags == (PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_RETURN | PERF_IP_FLAG_INTERRUPT) && - cs_etm__is_svc_instr(etmq, packet, packet->start_addr)) + cs_etm__is_svc_instr(etmq, trace_chan_id, + packet, packet->start_addr)) prev_packet->flags = PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_RETURN | PERF_IP_FLAG_SYSCALLRET; @@ -1440,7 +1792,7 @@ static int cs_etm__set_sample_flags(struct cs_etm_queue *etmq) return ret; /* The exception is for system call. */ - if (cs_etm__is_syscall(etmq, magic)) + if (cs_etm__is_syscall(etmq, tidq, magic)) packet->flags = PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_CALL | PERF_IP_FLAG_SYSCALLRET; @@ -1448,7 +1800,7 @@ static int cs_etm__set_sample_flags(struct cs_etm_queue *etmq) * The exceptions are triggered by external signals from bus, * interrupt controller, debug module, PE reset or halt. */ - else if (cs_etm__is_async_exception(etmq, magic)) + else if (cs_etm__is_async_exception(tidq, magic)) packet->flags = PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_CALL | PERF_IP_FLAG_ASYNC | @@ -1457,7 +1809,7 @@ static int cs_etm__set_sample_flags(struct cs_etm_queue *etmq) * Otherwise, exception is caused by trap, instruction & * data fault, or alignment errors. */ - else if (cs_etm__is_sync_exception(etmq, magic)) + else if (cs_etm__is_sync_exception(etmq, tidq, magic)) packet->flags = PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_CALL | PERF_IP_FLAG_INTERRUPT; @@ -1539,75 +1891,106 @@ out: return ret; } -static int cs_etm__process_decoder_queue(struct cs_etm_queue *etmq) +static int cs_etm__process_traceid_queue(struct cs_etm_queue *etmq, + struct cs_etm_traceid_queue *tidq) { int ret; + struct cs_etm_packet_queue *packet_queue; - /* Process each packet in this chunk */ - while (1) { - ret = cs_etm_decoder__get_packet(etmq->decoder, - etmq->packet); - if (ret <= 0) - /* - * Stop processing this chunk on - * end of data or error - */ - break; + packet_queue = &tidq->packet_queue; + /* Process each packet in this chunk */ + while (1) { + ret = cs_etm_decoder__get_packet(packet_queue, + tidq->packet); + if (ret <= 0) /* - * Since packet addresses are swapped in packet - * handling within below switch() statements, - * thus setting sample flags must be called - * prior to switch() statement to use address - * information before packets swapping. + * Stop processing this chunk on + * end of data or error */ - ret = cs_etm__set_sample_flags(etmq); - if (ret < 0) - break; - - switch (etmq->packet->sample_type) { - case CS_ETM_RANGE: - /* - * If the packet contains an instruction - * range, generate instruction sequence - * events. - */ - cs_etm__sample(etmq); - break; - case CS_ETM_EXCEPTION: - case CS_ETM_EXCEPTION_RET: - /* - * If the exception packet is coming, - * make sure the previous instruction - * range packet to be handled properly. - */ - cs_etm__exception(etmq); - break; - case CS_ETM_DISCONTINUITY: - /* - * Discontinuity in trace, flush - * previous branch stack - */ - cs_etm__flush(etmq); - break; - case CS_ETM_EMPTY: - /* - * Should not receive empty packet, - * report error. - */ - pr_err("CS ETM Trace: empty packet\n"); - return -EINVAL; - default: - break; - } + break; + + /* + * Since packet addresses are swapped in packet + * handling within below switch() statements, + * thus setting sample flags must be called + * prior to switch() statement to use address + * information before packets swapping. + */ + ret = cs_etm__set_sample_flags(etmq, tidq); + if (ret < 0) + break; + + switch (tidq->packet->sample_type) { + case CS_ETM_RANGE: + /* + * If the packet contains an instruction + * range, generate instruction sequence + * events. + */ + cs_etm__sample(etmq, tidq); + break; + case CS_ETM_EXCEPTION: + case CS_ETM_EXCEPTION_RET: + /* + * If the exception packet is coming, + * make sure the previous instruction + * range packet to be handled properly. + */ + cs_etm__exception(tidq); + break; + case CS_ETM_DISCONTINUITY: + /* + * Discontinuity in trace, flush + * previous branch stack + */ + cs_etm__flush(etmq, tidq); + break; + case CS_ETM_EMPTY: + /* + * Should not receive empty packet, + * report error. + */ + pr_err("CS ETM Trace: empty packet\n"); + return -EINVAL; + default: + break; } + } return ret; } +static void cs_etm__clear_all_traceid_queues(struct cs_etm_queue *etmq) +{ + int idx; + struct int_node *inode; + struct cs_etm_traceid_queue *tidq; + struct intlist *traceid_queues_list = etmq->traceid_queues_list; + + intlist__for_each_entry(inode, traceid_queues_list) { + idx = (int)(intptr_t)inode->priv; + tidq = etmq->traceid_queues[idx]; + + /* Ignore return value */ + cs_etm__process_traceid_queue(etmq, tidq); + + /* + * Generate an instruction sample with the remaining + * branchstack entries. + */ + cs_etm__flush(etmq, tidq); + } +} + static int cs_etm__run_decoder(struct cs_etm_queue *etmq) { int err = 0; + struct cs_etm_traceid_queue *tidq; + + tidq = cs_etm__etmq_get_traceid_queue(etmq, CS_ETM_PER_THREAD_TRACEID); + if (!tidq) + return -EINVAL; /* Go through each buffer in the queue and decode them one by one */ while (1) { @@ -1626,13 +2009,13 @@ static int cs_etm__run_decoder(struct cs_etm_queue *etmq) * an error occurs other than hoping the next one will * be better. */ - err = cs_etm__process_decoder_queue(etmq); + err = cs_etm__process_traceid_queue(etmq, tidq); } while (etmq->buf_len); if (err == 0) /* Flush any remaining branch stack entries */ - err = cs_etm__end_block(etmq); + err = cs_etm__end_block(etmq, tidq); } return err; @@ -1647,9 +2030,19 @@ static int cs_etm__process_timeless_queues(struct cs_etm_auxtrace *etm, for (i = 0; i < queues->nr_queues; i++) { struct auxtrace_queue *queue = &etm->queues.queue_array[i]; struct cs_etm_queue *etmq = queue->priv; + struct cs_etm_traceid_queue *tidq; + + if (!etmq) + continue; + + tidq = cs_etm__etmq_get_traceid_queue(etmq, + CS_ETM_PER_THREAD_TRACEID); + + if (!tidq) + continue; - if (etmq && ((tid == -1) || (etmq->tid == tid))) { - cs_etm__set_pid_tid_cpu(etm, queue); + if ((tid == -1) || (tidq->tid == tid)) { + cs_etm__set_pid_tid_cpu(etm, tidq); cs_etm__run_decoder(etmq); } } @@ -1657,6 +2050,164 @@ static int cs_etm__process_timeless_queues(struct cs_etm_auxtrace *etm, return 0; } +static int cs_etm__process_queues(struct cs_etm_auxtrace *etm) +{ + int ret = 0; + unsigned int cs_queue_nr, queue_nr; + u8 trace_chan_id; + u64 timestamp; + struct auxtrace_queue *queue; + struct cs_etm_queue *etmq; + struct cs_etm_traceid_queue *tidq; + + while (1) { + if (!etm->heap.heap_cnt) + goto out; + + /* Take the entry at the top of the min heap */ + cs_queue_nr = etm->heap.heap_array[0].queue_nr; + queue_nr = TO_QUEUE_NR(cs_queue_nr); + trace_chan_id = TO_TRACE_CHAN_ID(cs_queue_nr); + queue = &etm->queues.queue_array[queue_nr]; + etmq = queue->priv; + + /* + * Remove the top entry from the heap since we are about + * to process it. + */ + auxtrace_heap__pop(&etm->heap); + + tidq = cs_etm__etmq_get_traceid_queue(etmq, trace_chan_id); + if (!tidq) { + /* + * No traceID queue has been allocated for this traceID, + * which means something somewhere went very wrong. No + * other choice than simply exit. + */ + ret = -EINVAL; + goto out; + } + + /* + * Packets associated with this timestamp are already in + * the etmq's traceID queue, so process them. + */ + ret = cs_etm__process_traceid_queue(etmq, tidq); + if (ret < 0) + goto out; + + /* + * Packets for this timestamp have been processed, time to + * move on to the next timestamp, fetching a new auxtrace_buffer + * if need be. + */ +refetch: + ret = cs_etm__get_data_block(etmq); + if (ret < 0) + goto out; + + /* + * No more auxtrace_buffers to process in this etmq, simply + * move on to another entry in the auxtrace_heap. + */ + if (!ret) + continue; + + ret = cs_etm__decode_data_block(etmq); + if (ret) + goto out; + + timestamp = cs_etm__etmq_get_timestamp(etmq, &trace_chan_id); + + if (!timestamp) { + /* + * Function cs_etm__decode_data_block() returns when + * there is no more traces to decode in the current + * auxtrace_buffer OR when a timestamp has been + * encountered on any of the traceID queues. Since we + * did not get a timestamp, there is no more traces to + * process in this auxtrace_buffer. As such empty and + * flush all traceID queues. + */ + cs_etm__clear_all_traceid_queues(etmq); + + /* Fetch another auxtrace_buffer for this etmq */ + goto refetch; + } + + /* + * Add to the min heap the timestamp for packets that have + * just been decoded. They will be processed and synthesized + * during the next call to cs_etm__process_traceid_queue() for + * this queue/traceID. + */ + cs_queue_nr = TO_CS_QUEUE_NR(queue_nr, trace_chan_id); + ret = auxtrace_heap__add(&etm->heap, cs_queue_nr, timestamp); + } + +out: + return ret; +} + +static int cs_etm__process_itrace_start(struct cs_etm_auxtrace *etm, + union perf_event *event) +{ + struct thread *th; + + if (etm->timeless_decoding) + return 0; + + /* + * Add the tid/pid to the log so that we can get a match when + * we get a contextID from the decoder. + */ + th = machine__findnew_thread(etm->machine, + event->itrace_start.pid, + event->itrace_start.tid); + if (!th) + return -ENOMEM; + + thread__put(th); + + return 0; +} + +static int cs_etm__process_switch_cpu_wide(struct cs_etm_auxtrace *etm, + union perf_event *event) +{ + struct thread *th; + bool out = event->header.misc & PERF_RECORD_MISC_SWITCH_OUT; + + /* + * Context switch in per-thread mode are irrelevant since perf + * will start/stop tracing as the process is scheduled. + */ + if (etm->timeless_decoding) + return 0; + + /* + * SWITCH_IN events carry the next process to be switched out while + * SWITCH_OUT events carry the process to be switched in. As such + * we don't care about IN events. + */ + if (!out) + return 0; + + /* + * Add the tid/pid to the log so that we can get a match when + * we get a contextID from the decoder. + */ + th = machine__findnew_thread(etm->machine, + event->context_switch.next_prev_pid, + event->context_switch.next_prev_tid); + if (!th) + return -ENOMEM; + + thread__put(th); + + return 0; +} + static int cs_etm__process_event(struct perf_session *session, union perf_event *event, struct perf_sample *sample, @@ -1676,9 +2227,6 @@ static int cs_etm__process_event(struct perf_session *session, return -EINVAL; } - if (!etm->timeless_decoding) - return -EINVAL; - if (sample->time && (sample->time != (u64) -1)) timestamp = sample->time; else @@ -1690,10 +2238,20 @@ static int cs_etm__process_event(struct perf_session *session, return err; } - if (event->header.type == PERF_RECORD_EXIT) + if (etm->timeless_decoding && + event->header.type == PERF_RECORD_EXIT) return cs_etm__process_timeless_queues(etm, event->fork.tid); + if (event->header.type == PERF_RECORD_ITRACE_START) + return cs_etm__process_itrace_start(etm, event); + else if (event->header.type == PERF_RECORD_SWITCH_CPU_WIDE) + return cs_etm__process_switch_cpu_wide(etm, event); + + if (!etm->timeless_decoding && + event->header.type == PERF_RECORD_AUX) + return cs_etm__process_queues(etm); + return 0; } diff --git a/tools/perf/util/cs-etm.h b/tools/perf/util/cs-etm.h index 0e97c196147a..bc848fd095f4 100644 --- a/tools/perf/util/cs-etm.h +++ b/tools/perf/util/cs-etm.h @@ -9,6 +9,7 @@ #include "util/event.h" #include "util/session.h" +#include <linux/bits.h> /* Versionning header in case things need tro change in the future. That way * decoding of old snapshot is still possible. @@ -97,12 +98,72 @@ enum { CS_ETMV4_EXC_END = 31, }; +enum cs_etm_sample_type { + CS_ETM_EMPTY, + CS_ETM_RANGE, + CS_ETM_DISCONTINUITY, + CS_ETM_EXCEPTION, + CS_ETM_EXCEPTION_RET, +}; + +enum cs_etm_isa { + CS_ETM_ISA_UNKNOWN, + CS_ETM_ISA_A64, + CS_ETM_ISA_A32, + CS_ETM_ISA_T32, +}; + /* RB tree for quick conversion between traceID and metadata pointers */ struct intlist *traceid_list; +struct cs_etm_queue; + +struct cs_etm_packet { + enum cs_etm_sample_type sample_type; + enum cs_etm_isa isa; + u64 start_addr; + u64 end_addr; + u32 instr_count; + u32 last_instr_type; + u32 last_instr_subtype; + u32 flags; + u32 exception_number; + u8 last_instr_cond; + u8 last_instr_taken_branch; + u8 last_instr_size; + u8 trace_chan_id; + int cpu; +}; + +#define CS_ETM_PACKET_MAX_BUFFER 1024 + +/* + * When working with per-thread scenarios the process under trace can + * be scheduled on any CPU and as such, more than one traceID may be + * associated with the same process. Since a traceID of '0' is illegal + * as per the CoreSight architecture, use that specific value to + * identify the queue where all packets (with any traceID) are + * aggregated. + */ +#define CS_ETM_PER_THREAD_TRACEID 0 + +struct cs_etm_packet_queue { + u32 packet_count; + u32 head; + u32 tail; + u32 instr_count; + u64 timestamp; + u64 next_timestamp; + struct cs_etm_packet packet_buffer[CS_ETM_PACKET_MAX_BUFFER]; +}; + #define KiB(x) ((x) * 1024) #define MiB(x) ((x) * 1024 * 1024) +#define CS_ETM_INVAL_ADDR 0xdeadbeefdeadbeefUL + +#define BMVAL(val, lsb, msb) ((val & GENMASK(msb, lsb)) >> lsb) + #define CS_ETM_HEADER_SIZE (CS_HEADER_VERSION_0_MAX * sizeof(u64)) #define __perf_cs_etmv3_magic 0x3030303030303030ULL @@ -114,6 +175,13 @@ struct intlist *traceid_list; int cs_etm__process_auxtrace_info(union perf_event *event, struct perf_session *session); int cs_etm__get_cpu(u8 trace_chan_id, int *cpu); +int cs_etm__etmq_set_tid(struct cs_etm_queue *etmq, + pid_t tid, u8 trace_chan_id); +bool cs_etm__etmq_is_timeless(struct cs_etm_queue *etmq); +void cs_etm__etmq_set_traceid_queue_timestamp(struct cs_etm_queue *etmq, + u8 trace_chan_id); +struct cs_etm_packet_queue +*cs_etm__etmq_get_packet_queue(struct cs_etm_queue *etmq, u8 trace_chan_id); #else static inline int cs_etm__process_auxtrace_info(union perf_event *event __maybe_unused, @@ -127,6 +195,32 @@ static inline int cs_etm__get_cpu(u8 trace_chan_id __maybe_unused, { return -1; } + +static inline int cs_etm__etmq_set_tid( + struct cs_etm_queue *etmq __maybe_unused, + pid_t tid __maybe_unused, + u8 trace_chan_id __maybe_unused) +{ + return -1; +} + +static inline bool cs_etm__etmq_is_timeless( + struct cs_etm_queue *etmq __maybe_unused) +{ + /* What else to return? */ + return true; +} + +static inline void cs_etm__etmq_set_traceid_queue_timestamp( + struct cs_etm_queue *etmq __maybe_unused, + u8 trace_chan_id __maybe_unused) {} + +static inline struct cs_etm_packet_queue *cs_etm__etmq_get_packet_queue( + struct cs_etm_queue *etmq __maybe_unused, + u8 trace_chan_id __maybe_unused) +{ + return NULL; +} #endif #endif diff --git a/tools/perf/util/env.c b/tools/perf/util/env.c index 6a3eaf7d9353..1cc7a1837822 100644 --- a/tools/perf/util/env.c +++ b/tools/perf/util/env.c @@ -246,6 +246,7 @@ int perf_env__read_cpu_topology_map(struct perf_env *env) for (cpu = 0; cpu < nr_cpus; ++cpu) { env->cpu[cpu].core_id = cpu_map__get_core_id(cpu); env->cpu[cpu].socket_id = cpu_map__get_socket_id(cpu); + env->cpu[cpu].die_id = cpu_map__get_die_id(cpu); } env->nr_cpus_avail = nr_cpus; diff --git a/tools/perf/util/env.h b/tools/perf/util/env.h index 271a90b326c4..d5d9865aa812 100644 --- a/tools/perf/util/env.h +++ b/tools/perf/util/env.h @@ -9,6 +9,7 @@ struct cpu_topology_map { int socket_id; + int die_id; int core_id; }; @@ -49,6 +50,7 @@ struct perf_env { int nr_cmdline; int nr_sibling_cores; + int nr_sibling_dies; int nr_sibling_threads; int nr_numa_nodes; int nr_memory_nodes; @@ -57,6 +59,7 @@ struct perf_env { char *cmdline; const char **cmdline_argv; char *sibling_cores; + char *sibling_dies; char *sibling_threads; char *pmu_mappings; struct cpu_topology_map *cpu; diff --git a/tools/perf/util/event.h b/tools/perf/util/event.h index 9e999550f247..1f1da6082806 100644 --- a/tools/perf/util/event.h +++ b/tools/perf/util/event.h @@ -204,6 +204,8 @@ struct perf_sample { u64 period; u64 weight; u64 transaction; + u64 insn_cnt; + u64 cyc_cnt; u32 cpu; u32 raw_size; u64 data_src; diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c index 4a5947625c5c..0f506f10ecf0 100644 --- a/tools/perf/util/evsel.c +++ b/tools/perf/util/evsel.c @@ -679,6 +679,10 @@ static void __perf_evsel__config_callchain(struct perf_evsel *evsel, attr->sample_max_stack = param->max_stack; + if (opts->kernel_callchains) + attr->exclude_callchain_user = 1; + if (opts->user_callchains) + attr->exclude_callchain_kernel = 1; if (param->record_mode == CALLCHAIN_LBR) { if (!opts->branch_stack) { if (attr->exclude_user) { @@ -701,7 +705,14 @@ static void __perf_evsel__config_callchain(struct perf_evsel *evsel, if (!function) { perf_evsel__set_sample_bit(evsel, REGS_USER); perf_evsel__set_sample_bit(evsel, STACK_USER); - attr->sample_regs_user |= PERF_REGS_MASK; + if (opts->sample_user_regs && DWARF_MINIMAL_REGS != PERF_REGS_MASK) { + attr->sample_regs_user |= DWARF_MINIMAL_REGS; + pr_warning("WARNING: The use of --call-graph=dwarf may require all the user registers, " + "specifying a subset with --user-regs may render DWARF unwinding unreliable, " + "so the minimal registers set (IP, SP) is explicitly forced.\n"); + } else { + attr->sample_regs_user |= PERF_REGS_MASK; + } attr->sample_stack_user = param->dump_size; attr->exclude_callchain_user = 1; } else { @@ -1136,9 +1147,6 @@ 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) { - if (evsel->system_wide) - nthreads = 1; - evsel->fd = xyarray__new(ncpus, nthreads, sizeof(int)); if (evsel->fd) { diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c index 847ae51a524b..06ddb6618ef3 100644 --- a/tools/perf/util/header.c +++ b/tools/perf/util/header.c @@ -599,6 +599,27 @@ static int write_cpu_topology(struct feat_fd *ff, if (ret < 0) return ret; } + + if (!tp->die_sib) + goto done; + + ret = do_write(ff, &tp->die_sib, sizeof(tp->die_sib)); + if (ret < 0) + goto done; + + for (i = 0; i < tp->die_sib; i++) { + ret = do_write_string(ff, tp->die_siblings[i]); + if (ret < 0) + goto done; + } + + for (j = 0; j < perf_env.nr_cpus_avail; j++) { + ret = do_write(ff, &perf_env.cpu[j].die_id, + sizeof(perf_env.cpu[j].die_id)); + if (ret < 0) + return ret; + } + done: cpu_topology__delete(tp); return ret; @@ -1439,10 +1460,20 @@ static void print_cpu_topology(struct feat_fd *ff, FILE *fp) str = ph->env.sibling_cores; for (i = 0; i < nr; i++) { - fprintf(fp, "# sibling cores : %s\n", str); + fprintf(fp, "# sibling sockets : %s\n", str); str += strlen(str) + 1; } + if (ph->env.nr_sibling_dies) { + nr = ph->env.nr_sibling_dies; + str = ph->env.sibling_dies; + + for (i = 0; i < nr; i++) { + fprintf(fp, "# sibling dies : %s\n", str); + str += strlen(str) + 1; + } + } + nr = ph->env.nr_sibling_threads; str = ph->env.sibling_threads; @@ -1451,12 +1482,28 @@ static void print_cpu_topology(struct feat_fd *ff, FILE *fp) str += strlen(str) + 1; } - if (ph->env.cpu != NULL) { - for (i = 0; i < cpu_nr; i++) - fprintf(fp, "# CPU %d: Core ID %d, Socket ID %d\n", i, - ph->env.cpu[i].core_id, ph->env.cpu[i].socket_id); - } else - fprintf(fp, "# Core ID and Socket ID information is not available\n"); + if (ph->env.nr_sibling_dies) { + if (ph->env.cpu != NULL) { + for (i = 0; i < cpu_nr; i++) + fprintf(fp, "# CPU %d: Core ID %d, " + "Die ID %d, Socket ID %d\n", + i, ph->env.cpu[i].core_id, + ph->env.cpu[i].die_id, + ph->env.cpu[i].socket_id); + } else + fprintf(fp, "# Core ID, Die ID and Socket ID " + "information is not available\n"); + } else { + if (ph->env.cpu != NULL) { + for (i = 0; i < cpu_nr; i++) + fprintf(fp, "# CPU %d: Core ID %d, " + "Socket ID %d\n", + i, ph->env.cpu[i].core_id, + ph->env.cpu[i].socket_id); + } else + fprintf(fp, "# Core ID and Socket ID " + "information is not available\n"); + } } static void print_clockid(struct feat_fd *ff, FILE *fp) @@ -2214,6 +2261,7 @@ static int process_cpu_topology(struct feat_fd *ff, void *data __maybe_unused) goto free_cpu; ph->env.cpu[i].core_id = nr; + size += sizeof(u32); if (do_read_u32(ff, &nr)) goto free_cpu; @@ -2225,6 +2273,40 @@ static int process_cpu_topology(struct feat_fd *ff, void *data __maybe_unused) } ph->env.cpu[i].socket_id = nr; + size += sizeof(u32); + } + + /* + * The header may be from old perf, + * which doesn't include die information. + */ + if (ff->size <= size) + return 0; + + if (do_read_u32(ff, &nr)) + return -1; + + ph->env.nr_sibling_dies = nr; + size += sizeof(u32); + + for (i = 0; i < nr; i++) { + str = do_read_string(ff); + if (!str) + goto error; + + /* include a NULL character at the end */ + if (strbuf_add(&sb, str, strlen(str) + 1) < 0) + goto error; + size += string_size(str); + free(str); + } + ph->env.sibling_dies = strbuf_detach(&sb, NULL); + + for (i = 0; i < (u32)cpu_nr; i++) { + if (do_read_u32(ff, &nr)) + goto free_cpu; + + ph->env.cpu[i].die_id = nr; } return 0; 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 9d189e90fbdc..f001f4ec4ddf 100644 --- a/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c +++ b/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c @@ -95,6 +95,7 @@ struct intel_pt_decoder { uint64_t *insn_cnt_ptr, uint64_t *ip, uint64_t to_ip, uint64_t max_insn_cnt, void *data); bool (*pgd_ip)(uint64_t ip, void *data); + int (*lookahead)(void *data, intel_pt_lookahead_cb_t cb, void *cb_data); void *data; struct intel_pt_state state; const unsigned char *buf; @@ -107,6 +108,7 @@ struct intel_pt_decoder { bool have_cyc; bool fixup_last_mtc; bool have_last_ip; + bool in_psb; enum intel_pt_param_flags flags; uint64_t pos; uint64_t last_ip; @@ -115,6 +117,7 @@ struct intel_pt_decoder { uint64_t timestamp; uint64_t tsc_timestamp; uint64_t ref_timestamp; + uint64_t buf_timestamp; uint64_t sample_timestamp; uint64_t ret_addr; uint64_t ctc_timestamp; @@ -151,6 +154,11 @@ struct intel_pt_decoder { uint64_t period_mask; uint64_t period_ticks; uint64_t last_masked_timestamp; + uint64_t tot_cyc_cnt; + uint64_t sample_tot_cyc_cnt; + uint64_t base_cyc_cnt; + uint64_t cyc_cnt_timestamp; + double tsc_to_cyc; bool continuous_period; bool overflow; bool set_fup_tx_flags; @@ -158,6 +166,7 @@ struct intel_pt_decoder { bool set_fup_mwait; bool set_fup_pwre; bool set_fup_exstop; + bool sample_cyc; unsigned int fup_tx_flags; unsigned int tx_flags; uint64_t fup_ptw_payload; @@ -217,6 +226,7 @@ struct intel_pt_decoder *intel_pt_decoder_new(struct intel_pt_params *params) decoder->get_trace = params->get_trace; decoder->walk_insn = params->walk_insn; decoder->pgd_ip = params->pgd_ip; + decoder->lookahead = params->lookahead; decoder->data = params->data; decoder->return_compression = params->return_compression; decoder->branch_enable = params->branch_enable; @@ -470,7 +480,21 @@ static int intel_pt_bad_packet(struct intel_pt_decoder *decoder) return -EBADMSG; } -static int intel_pt_get_data(struct intel_pt_decoder *decoder) +static inline void intel_pt_update_sample_time(struct intel_pt_decoder *decoder) +{ + decoder->sample_timestamp = decoder->timestamp; + decoder->sample_insn_cnt = decoder->timestamp_insn_cnt; +} + +static void intel_pt_reposition(struct intel_pt_decoder *decoder) +{ + decoder->ip = 0; + decoder->pkt_state = INTEL_PT_STATE_NO_PSB; + decoder->timestamp = 0; + decoder->have_tma = false; +} + +static int intel_pt_get_data(struct intel_pt_decoder *decoder, bool reposition) { struct intel_pt_buffer buffer = { .buf = 0, }; int ret; @@ -487,12 +511,10 @@ static int intel_pt_get_data(struct intel_pt_decoder *decoder) intel_pt_log("No more data\n"); return -ENODATA; } - if (!buffer.consecutive) { - decoder->ip = 0; - decoder->pkt_state = INTEL_PT_STATE_NO_PSB; + decoder->buf_timestamp = buffer.ref_timestamp; + if (!buffer.consecutive || reposition) { + intel_pt_reposition(decoder); decoder->ref_timestamp = buffer.ref_timestamp; - decoder->timestamp = 0; - decoder->have_tma = false; decoder->state.trace_nr = buffer.trace_nr; intel_pt_log("Reference timestamp 0x%" PRIx64 "\n", decoder->ref_timestamp); @@ -502,10 +524,11 @@ static int intel_pt_get_data(struct intel_pt_decoder *decoder) return 0; } -static int intel_pt_get_next_data(struct intel_pt_decoder *decoder) +static int intel_pt_get_next_data(struct intel_pt_decoder *decoder, + bool reposition) { if (!decoder->next_buf) - return intel_pt_get_data(decoder); + return intel_pt_get_data(decoder, reposition); decoder->buf = decoder->next_buf; decoder->len = decoder->next_len; @@ -524,7 +547,7 @@ static int intel_pt_get_split_packet(struct intel_pt_decoder *decoder) len = decoder->len; memcpy(buf, decoder->buf, len); - ret = intel_pt_get_data(decoder); + ret = intel_pt_get_data(decoder, false); if (ret) { decoder->pos += old_len; return ret < 0 ? ret : -EINVAL; @@ -850,7 +873,7 @@ static int intel_pt_get_next_packet(struct intel_pt_decoder *decoder) decoder->len -= decoder->pkt_step; if (!decoder->len) { - ret = intel_pt_get_next_data(decoder); + ret = intel_pt_get_next_data(decoder, false); if (ret) return ret; } @@ -1308,10 +1331,10 @@ static int intel_pt_walk_tnt(struct intel_pt_decoder *decoder) decoder->ip += intel_pt_insn.length; return 0; } + decoder->sample_cyc = false; decoder->ip += intel_pt_insn.length; if (!decoder->tnt.count) { - decoder->sample_timestamp = decoder->timestamp; - decoder->sample_insn_cnt = decoder->timestamp_insn_cnt; + intel_pt_update_sample_time(decoder); return -EAGAIN; } decoder->tnt.payload <<= 1; @@ -1345,6 +1368,21 @@ static int intel_pt_mode_tsx(struct intel_pt_decoder *decoder, bool *no_tip) return 0; } +static uint64_t intel_pt_8b_tsc(uint64_t timestamp, uint64_t ref_timestamp) +{ + timestamp |= (ref_timestamp & (0xffULL << 56)); + + if (timestamp < ref_timestamp) { + if (ref_timestamp - timestamp > (1ULL << 55)) + timestamp += (1ULL << 56); + } else { + if (timestamp - ref_timestamp > (1ULL << 55)) + timestamp -= (1ULL << 56); + } + + return timestamp; +} + static void intel_pt_calc_tsc_timestamp(struct intel_pt_decoder *decoder) { uint64_t timestamp; @@ -1352,15 +1390,8 @@ static void intel_pt_calc_tsc_timestamp(struct intel_pt_decoder *decoder) decoder->have_tma = false; if (decoder->ref_timestamp) { - timestamp = decoder->packet.payload | - (decoder->ref_timestamp & (0xffULL << 56)); - if (timestamp < decoder->ref_timestamp) { - if (decoder->ref_timestamp - timestamp > (1ULL << 55)) - timestamp += (1ULL << 56); - } else { - if (timestamp - decoder->ref_timestamp > (1ULL << 55)) - timestamp -= (1ULL << 56); - } + timestamp = intel_pt_8b_tsc(decoder->packet.payload, + decoder->ref_timestamp); decoder->tsc_timestamp = timestamp; decoder->timestamp = timestamp; decoder->ref_timestamp = 0; @@ -1404,6 +1435,42 @@ static int intel_pt_overflow(struct intel_pt_decoder *decoder) return -EOVERFLOW; } +static inline void intel_pt_mtc_cyc_cnt_pge(struct intel_pt_decoder *decoder) +{ + if (decoder->have_cyc) + return; + + decoder->cyc_cnt_timestamp = decoder->timestamp; + decoder->base_cyc_cnt = decoder->tot_cyc_cnt; +} + +static inline void intel_pt_mtc_cyc_cnt_cbr(struct intel_pt_decoder *decoder) +{ + decoder->tsc_to_cyc = decoder->cbr / decoder->max_non_turbo_ratio_fp; + + if (decoder->pge) + intel_pt_mtc_cyc_cnt_pge(decoder); +} + +static inline void intel_pt_mtc_cyc_cnt_upd(struct intel_pt_decoder *decoder) +{ + uint64_t tot_cyc_cnt, tsc_delta; + + if (decoder->have_cyc) + return; + + decoder->sample_cyc = true; + + if (!decoder->pge || decoder->timestamp <= decoder->cyc_cnt_timestamp) + return; + + tsc_delta = decoder->timestamp - decoder->cyc_cnt_timestamp; + tot_cyc_cnt = tsc_delta * decoder->tsc_to_cyc + decoder->base_cyc_cnt; + + if (tot_cyc_cnt > decoder->tot_cyc_cnt) + decoder->tot_cyc_cnt = tot_cyc_cnt; +} + static void intel_pt_calc_tma(struct intel_pt_decoder *decoder) { uint32_t ctc = decoder->packet.payload; @@ -1413,6 +1480,11 @@ static void intel_pt_calc_tma(struct intel_pt_decoder *decoder) if (!decoder->tsc_ctc_ratio_d) return; + if (decoder->pge && !decoder->in_psb) + intel_pt_mtc_cyc_cnt_pge(decoder); + else + intel_pt_mtc_cyc_cnt_upd(decoder); + decoder->last_mtc = (ctc >> decoder->mtc_shift) & 0xff; decoder->ctc_timestamp = decoder->tsc_timestamp - fc; if (decoder->tsc_ctc_mult) { @@ -1468,6 +1540,8 @@ static void intel_pt_calc_mtc_timestamp(struct intel_pt_decoder *decoder) else decoder->timestamp = timestamp; + intel_pt_mtc_cyc_cnt_upd(decoder); + decoder->timestamp_insn_cnt = 0; decoder->last_mtc = mtc; @@ -1492,6 +1566,8 @@ static void intel_pt_calc_cbr(struct intel_pt_decoder *decoder) decoder->cbr = cbr; decoder->cbr_cyc_to_tsc = decoder->max_non_turbo_ratio_fp / cbr; + + intel_pt_mtc_cyc_cnt_cbr(decoder); } static void intel_pt_calc_cyc_timestamp(struct intel_pt_decoder *decoder) @@ -1501,6 +1577,9 @@ static void intel_pt_calc_cyc_timestamp(struct intel_pt_decoder *decoder) decoder->have_cyc = true; decoder->cycle_cnt += decoder->packet.payload; + if (decoder->pge) + decoder->tot_cyc_cnt += decoder->packet.payload; + decoder->sample_cyc = true; if (!decoder->cyc_ref_timestamp) return; @@ -1528,14 +1607,17 @@ static int intel_pt_walk_psbend(struct intel_pt_decoder *decoder) { int err; + decoder->in_psb = true; + while (1) { err = intel_pt_get_next_packet(decoder); if (err) - return err; + goto out; switch (decoder->packet.type) { case INTEL_PT_PSBEND: - return 0; + err = 0; + goto out; case INTEL_PT_TIP_PGD: case INTEL_PT_TIP_PGE: @@ -1553,10 +1635,12 @@ static int intel_pt_walk_psbend(struct intel_pt_decoder *decoder) case INTEL_PT_PWRX: decoder->have_tma = false; intel_pt_log("ERROR: Unexpected packet\n"); - return -EAGAIN; + err = -EAGAIN; + goto out; case INTEL_PT_OVF: - return intel_pt_overflow(decoder); + err = intel_pt_overflow(decoder); + goto out; case INTEL_PT_TSC: intel_pt_calc_tsc_timestamp(decoder); @@ -1602,6 +1686,10 @@ static int intel_pt_walk_psbend(struct intel_pt_decoder *decoder) break; } } +out: + decoder->in_psb = false; + + return err; } static int intel_pt_walk_fup_tip(struct intel_pt_decoder *decoder) @@ -1675,6 +1763,7 @@ static int intel_pt_walk_fup_tip(struct intel_pt_decoder *decoder) decoder->state.to_ip = decoder->ip; } decoder->state.type |= INTEL_PT_TRACE_BEGIN; + intel_pt_mtc_cyc_cnt_pge(decoder); return 0; case INTEL_PT_TIP: @@ -1745,6 +1834,7 @@ next: case INTEL_PT_TIP_PGE: { decoder->pge = true; + intel_pt_mtc_cyc_cnt_pge(decoder); if (decoder->packet.count == 0) { intel_pt_log_at("Skipping zero TIP.PGE", decoder->pos); @@ -1975,10 +2065,12 @@ static int intel_pt_walk_psb(struct intel_pt_decoder *decoder) { int err; + decoder->in_psb = true; + while (1) { err = intel_pt_get_next_packet(decoder); if (err) - return err; + goto out; switch (decoder->packet.type) { case INTEL_PT_TIP_PGD: @@ -1994,7 +2086,8 @@ static int intel_pt_walk_psb(struct intel_pt_decoder *decoder) case INTEL_PT_PWRE: case INTEL_PT_PWRX: intel_pt_log("ERROR: Unexpected packet\n"); - return -ENOENT; + err = -ENOENT; + goto out; case INTEL_PT_FUP: decoder->pge = true; @@ -2053,16 +2146,20 @@ static int intel_pt_walk_psb(struct intel_pt_decoder *decoder) decoder->pkt_state = INTEL_PT_STATE_ERR4; else decoder->pkt_state = INTEL_PT_STATE_ERR3; - return -ENOENT; + err = -ENOENT; + goto out; case INTEL_PT_BAD: /* Does not happen */ - return intel_pt_bug(decoder); + err = intel_pt_bug(decoder); + goto out; case INTEL_PT_OVF: - return intel_pt_overflow(decoder); + err = intel_pt_overflow(decoder); + goto out; case INTEL_PT_PSBEND: - return 0; + err = 0; + goto out; case INTEL_PT_PSB: case INTEL_PT_VMCS: @@ -2072,6 +2169,10 @@ static int intel_pt_walk_psb(struct intel_pt_decoder *decoder) break; } } +out: + decoder->in_psb = false; + + return err; } static int intel_pt_walk_to_ip(struct intel_pt_decoder *decoder) @@ -2086,18 +2187,30 @@ static int intel_pt_walk_to_ip(struct intel_pt_decoder *decoder) switch (decoder->packet.type) { case INTEL_PT_TIP_PGD: decoder->continuous_period = false; - __fallthrough; + decoder->pge = false; + if (intel_pt_have_ip(decoder)) + intel_pt_set_ip(decoder); + if (!decoder->ip) + break; + decoder->state.type |= INTEL_PT_TRACE_END; + return 0; + case INTEL_PT_TIP_PGE: + decoder->pge = true; + intel_pt_mtc_cyc_cnt_pge(decoder); + if (intel_pt_have_ip(decoder)) + intel_pt_set_ip(decoder); + if (!decoder->ip) + break; + decoder->state.type |= INTEL_PT_TRACE_BEGIN; + return 0; + case INTEL_PT_TIP: - decoder->pge = decoder->packet.type != INTEL_PT_TIP_PGD; + decoder->pge = true; if (intel_pt_have_ip(decoder)) intel_pt_set_ip(decoder); if (!decoder->ip) break; - if (decoder->packet.type == INTEL_PT_TIP_PGE) - decoder->state.type |= INTEL_PT_TRACE_BEGIN; - if (decoder->packet.type == INTEL_PT_TIP_PGD) - decoder->state.type |= INTEL_PT_TRACE_END; return 0; case INTEL_PT_FUP: @@ -2250,7 +2363,7 @@ static int intel_pt_get_split_psb(struct intel_pt_decoder *decoder, decoder->pos += decoder->len; decoder->len = 0; - ret = intel_pt_get_next_data(decoder); + ret = intel_pt_get_next_data(decoder, false); if (ret) return ret; @@ -2276,7 +2389,7 @@ static int intel_pt_scan_for_psb(struct intel_pt_decoder *decoder) intel_pt_log("Scanning for PSB\n"); while (1) { if (!decoder->len) { - ret = intel_pt_get_next_data(decoder); + ret = intel_pt_get_next_data(decoder, false); if (ret) return ret; } @@ -2404,8 +2517,8 @@ const struct intel_pt_state *intel_pt_decode(struct intel_pt_decoder *decoder) if (err) { decoder->state.err = intel_pt_ext_err(err); decoder->state.from_ip = decoder->ip; - decoder->sample_timestamp = decoder->timestamp; - decoder->sample_insn_cnt = decoder->timestamp_insn_cnt; + intel_pt_update_sample_time(decoder); + decoder->sample_tot_cyc_cnt = decoder->tot_cyc_cnt; } else { decoder->state.err = 0; if (decoder->cbr != decoder->cbr_seen && decoder->state.type) { @@ -2414,8 +2527,9 @@ const struct intel_pt_state *intel_pt_decode(struct intel_pt_decoder *decoder) decoder->state.cbr_payload = decoder->cbr_payload; } if (intel_pt_sample_time(decoder->pkt_state)) { - decoder->sample_timestamp = decoder->timestamp; - decoder->sample_insn_cnt = decoder->timestamp_insn_cnt; + intel_pt_update_sample_time(decoder); + if (decoder->sample_cyc) + decoder->sample_tot_cyc_cnt = decoder->tot_cyc_cnt; } } @@ -2423,6 +2537,7 @@ const struct intel_pt_state *intel_pt_decode(struct intel_pt_decoder *decoder) decoder->state.est_timestamp = intel_pt_est_timestamp(decoder); decoder->state.cr3 = decoder->cr3; decoder->state.tot_insn_cnt = decoder->tot_insn_cnt; + decoder->state.tot_cyc_cnt = decoder->sample_tot_cyc_cnt; return &decoder->state; } @@ -2732,3 +2847,131 @@ unsigned char *intel_pt_find_overlap(unsigned char *buf_a, size_t len_a, return buf_b; /* No overlap */ } } + +/** + * struct fast_forward_data - data used by intel_pt_ff_cb(). + * @timestamp: timestamp to fast forward towards + * @buf_timestamp: buffer timestamp of last buffer with trace data earlier than + * the fast forward timestamp. + */ +struct fast_forward_data { + uint64_t timestamp; + uint64_t buf_timestamp; +}; + +/** + * intel_pt_ff_cb - fast forward lookahead callback. + * @buffer: Intel PT trace buffer + * @data: opaque pointer to fast forward data (struct fast_forward_data) + * + * Determine if @buffer trace is past the fast forward timestamp. + * + * Return: 1 (stop lookahead) if @buffer trace is past the fast forward + * timestamp, and 0 otherwise. + */ +static int intel_pt_ff_cb(struct intel_pt_buffer *buffer, void *data) +{ + struct fast_forward_data *d = data; + unsigned char *buf; + uint64_t tsc; + size_t rem; + size_t len; + + buf = (unsigned char *)buffer->buf; + len = buffer->len; + + if (!intel_pt_next_psb(&buf, &len) || + !intel_pt_next_tsc(buf, len, &tsc, &rem)) + return 0; + + tsc = intel_pt_8b_tsc(tsc, buffer->ref_timestamp); + + intel_pt_log("Buffer 1st timestamp " x64_fmt " ref timestamp " x64_fmt "\n", + tsc, buffer->ref_timestamp); + + /* + * If the buffer contains a timestamp earlier that the fast forward + * timestamp, then record it, else stop. + */ + if (tsc < d->timestamp) + d->buf_timestamp = buffer->ref_timestamp; + else + return 1; + + return 0; +} + +/** + * intel_pt_fast_forward - reposition decoder forwards. + * @decoder: Intel PT decoder + * @timestamp: timestamp to fast forward towards + * + * Reposition decoder at the last PSB with a timestamp earlier than @timestamp. + * + * Return: 0 on success or negative error code on failure. + */ +int intel_pt_fast_forward(struct intel_pt_decoder *decoder, uint64_t timestamp) +{ + struct fast_forward_data d = { .timestamp = timestamp }; + unsigned char *buf; + size_t len; + int err; + + intel_pt_log("Fast forward towards timestamp " x64_fmt "\n", timestamp); + + /* Find buffer timestamp of buffer to fast forward to */ + err = decoder->lookahead(decoder->data, intel_pt_ff_cb, &d); + if (err < 0) + return err; + + /* Walk to buffer with same buffer timestamp */ + if (d.buf_timestamp) { + do { + decoder->pos += decoder->len; + decoder->len = 0; + err = intel_pt_get_next_data(decoder, true); + /* -ENOLINK means non-consecutive trace */ + if (err && err != -ENOLINK) + return err; + } while (decoder->buf_timestamp != d.buf_timestamp); + } + + if (!decoder->buf) + return 0; + + buf = (unsigned char *)decoder->buf; + len = decoder->len; + + if (!intel_pt_next_psb(&buf, &len)) + return 0; + + /* + * Walk PSBs while the PSB timestamp is less than the fast forward + * timestamp. + */ + do { + uint64_t tsc; + size_t rem; + + if (!intel_pt_next_tsc(buf, len, &tsc, &rem)) + break; + tsc = intel_pt_8b_tsc(tsc, decoder->buf_timestamp); + /* + * A TSC packet can slip past MTC packets but, after fast + * forward, decoding starts at the TSC timestamp. That means + * the timestamps may not be exactly the same as the timestamps + * that would have been decoded without fast forward. + */ + if (tsc < timestamp) { + intel_pt_log("Fast forward to next PSB timestamp " x64_fmt "\n", tsc); + decoder->pos += decoder->len - len; + decoder->buf = buf; + decoder->len = len; + intel_pt_reposition(decoder); + } else { + break; + } + } while (intel_pt_step_psb(&buf, &len)); + + return 0; +} diff --git a/tools/perf/util/intel-pt-decoder/intel-pt-decoder.h b/tools/perf/util/intel-pt-decoder/intel-pt-decoder.h index 1e8cfdc7bfab..754efa8b501f 100644 --- a/tools/perf/util/intel-pt-decoder/intel-pt-decoder.h +++ b/tools/perf/util/intel-pt-decoder/intel-pt-decoder.h @@ -68,6 +68,7 @@ struct intel_pt_state { uint64_t to_ip; uint64_t cr3; uint64_t tot_insn_cnt; + uint64_t tot_cyc_cnt; uint64_t timestamp; uint64_t est_timestamp; uint64_t trace_nr; @@ -92,12 +93,15 @@ struct intel_pt_buffer { uint64_t trace_nr; }; +typedef int (*intel_pt_lookahead_cb_t)(struct intel_pt_buffer *, void *); + struct intel_pt_params { int (*get_trace)(struct intel_pt_buffer *buffer, void *data); int (*walk_insn)(struct intel_pt_insn *intel_pt_insn, uint64_t *insn_cnt_ptr, uint64_t *ip, uint64_t to_ip, uint64_t max_insn_cnt, void *data); bool (*pgd_ip)(uint64_t ip, void *data); + int (*lookahead)(void *data, intel_pt_lookahead_cb_t cb, void *cb_data); void *data; bool return_compression; bool branch_enable; @@ -117,6 +121,8 @@ void intel_pt_decoder_free(struct intel_pt_decoder *decoder); const struct intel_pt_state *intel_pt_decode(struct intel_pt_decoder *decoder); +int intel_pt_fast_forward(struct intel_pt_decoder *decoder, uint64_t timestamp); + unsigned char *intel_pt_find_overlap(unsigned char *buf_a, size_t len_a, unsigned char *buf_b, size_t len_b, bool have_tsc, bool *consecutive); diff --git a/tools/perf/util/intel-pt.c b/tools/perf/util/intel-pt.c index 9276506d07f8..8ed51f4e9e30 100644 --- a/tools/perf/util/intel-pt.c +++ b/tools/perf/util/intel-pt.c @@ -33,6 +33,7 @@ #include "tsc.h" #include "intel-pt.h" #include "config.h" +#include "time-utils.h" #include "intel-pt-decoder/intel-pt-log.h" #include "intel-pt-decoder/intel-pt-decoder.h" @@ -41,6 +42,11 @@ #define MAX_TIMESTAMP (~0ULL) +struct range { + u64 start; + u64 end; +}; + struct intel_pt { struct auxtrace auxtrace; struct auxtrace_queues queues; @@ -109,6 +115,9 @@ struct intel_pt { char *filter; struct addr_filters filts; + + struct range *time_ranges; + unsigned int range_cnt; }; enum switch_state { @@ -145,9 +154,18 @@ struct intel_pt_queue { bool have_sample; u64 time; u64 timestamp; + u64 sel_timestamp; + bool sel_start; + unsigned int sel_idx; u32 flags; u16 insn_len; u64 last_insn_cnt; + u64 ipc_insn_cnt; + u64 ipc_cyc_cnt; + u64 last_in_insn_cnt; + u64 last_in_cyc_cnt; + u64 last_br_insn_cnt; + u64 last_br_cyc_cnt; char insn[INTEL_PT_INSN_BUF_SZ]; }; @@ -224,32 +242,13 @@ static int intel_pt_do_fix_overlap(struct intel_pt *pt, struct auxtrace_buffer * return 0; } -/* This function assumes data is processed sequentially only */ -static int intel_pt_get_trace(struct intel_pt_buffer *b, void *data) +static int intel_pt_get_buffer(struct intel_pt_queue *ptq, + struct auxtrace_buffer *buffer, + struct auxtrace_buffer *old_buffer, + struct intel_pt_buffer *b) { - struct intel_pt_queue *ptq = data; - struct auxtrace_buffer *buffer = ptq->buffer; - struct auxtrace_buffer *old_buffer = ptq->old_buffer; - struct auxtrace_queue *queue; bool might_overlap; - if (ptq->stop) { - b->len = 0; - return 0; - } - - queue = &ptq->pt->queues.queue_array[ptq->queue_nr]; - - buffer = auxtrace_buffer__next(queue, buffer); - if (!buffer) { - if (old_buffer) - auxtrace_buffer__drop_data(old_buffer); - b->len = 0; - return 0; - } - - ptq->buffer = buffer; - if (!buffer->data) { int fd = perf_data__fd(ptq->pt->session->data); @@ -279,6 +278,95 @@ static int intel_pt_get_trace(struct intel_pt_buffer *b, void *data) b->consecutive = true; } + return 0; +} + +/* Do not drop buffers with references - refer intel_pt_get_trace() */ +static void intel_pt_lookahead_drop_buffer(struct intel_pt_queue *ptq, + struct auxtrace_buffer *buffer) +{ + if (!buffer || buffer == ptq->buffer || buffer == ptq->old_buffer) + return; + + auxtrace_buffer__drop_data(buffer); +} + +/* Must be serialized with respect to intel_pt_get_trace() */ +static int intel_pt_lookahead(void *data, intel_pt_lookahead_cb_t cb, + void *cb_data) +{ + struct intel_pt_queue *ptq = data; + struct auxtrace_buffer *buffer = ptq->buffer; + struct auxtrace_buffer *old_buffer = ptq->old_buffer; + struct auxtrace_queue *queue; + int err = 0; + + queue = &ptq->pt->queues.queue_array[ptq->queue_nr]; + + while (1) { + struct intel_pt_buffer b = { .len = 0 }; + + buffer = auxtrace_buffer__next(queue, buffer); + if (!buffer) + break; + + err = intel_pt_get_buffer(ptq, buffer, old_buffer, &b); + if (err) + break; + + if (b.len) { + intel_pt_lookahead_drop_buffer(ptq, old_buffer); + old_buffer = buffer; + } else { + intel_pt_lookahead_drop_buffer(ptq, buffer); + continue; + } + + err = cb(&b, cb_data); + if (err) + break; + } + + if (buffer != old_buffer) + intel_pt_lookahead_drop_buffer(ptq, buffer); + intel_pt_lookahead_drop_buffer(ptq, old_buffer); + + return err; +} + +/* + * This function assumes data is processed sequentially only. + * Must be serialized with respect to intel_pt_lookahead() + */ +static int intel_pt_get_trace(struct intel_pt_buffer *b, void *data) +{ + struct intel_pt_queue *ptq = data; + struct auxtrace_buffer *buffer = ptq->buffer; + struct auxtrace_buffer *old_buffer = ptq->old_buffer; + struct auxtrace_queue *queue; + int err; + + if (ptq->stop) { + b->len = 0; + return 0; + } + + queue = &ptq->pt->queues.queue_array[ptq->queue_nr]; + + buffer = auxtrace_buffer__next(queue, buffer); + if (!buffer) { + if (old_buffer) + auxtrace_buffer__drop_data(old_buffer); + b->len = 0; + return 0; + } + + ptq->buffer = buffer; + + err = intel_pt_get_buffer(ptq, buffer, old_buffer, b); + if (err) + return err; + if (ptq->step_through_buffers) ptq->stop = true; @@ -798,6 +886,7 @@ static struct intel_pt_queue *intel_pt_alloc_queue(struct intel_pt *pt, params.get_trace = intel_pt_get_trace; params.walk_insn = intel_pt_walk_next_insn; + params.lookahead = intel_pt_lookahead; params.data = ptq; params.return_compression = intel_pt_return_compression(pt); params.branch_enable = intel_pt_branch_enable(pt); @@ -921,6 +1010,23 @@ static void intel_pt_sample_flags(struct intel_pt_queue *ptq) ptq->flags |= PERF_IP_FLAG_TRACE_END; } +static void intel_pt_setup_time_range(struct intel_pt *pt, + struct intel_pt_queue *ptq) +{ + if (!pt->range_cnt) + return; + + ptq->sel_timestamp = pt->time_ranges[0].start; + ptq->sel_idx = 0; + + if (ptq->sel_timestamp) { + ptq->sel_start = true; + } else { + ptq->sel_timestamp = pt->time_ranges[0].end; + ptq->sel_start = false; + } +} + static int intel_pt_setup_queue(struct intel_pt *pt, struct auxtrace_queue *queue, unsigned int queue_nr) @@ -945,6 +1051,8 @@ static int intel_pt_setup_queue(struct intel_pt *pt, ptq->step_through_buffers = true; ptq->sync_switch = pt->sync_switch; + + intel_pt_setup_time_range(pt, ptq); } if (!ptq->on_heap && @@ -959,6 +1067,14 @@ static int intel_pt_setup_queue(struct intel_pt *pt, intel_pt_log("queue %u getting timestamp\n", queue_nr); intel_pt_log("queue %u decoding cpu %d pid %d tid %d\n", queue_nr, ptq->cpu, ptq->pid, ptq->tid); + + if (ptq->sel_start && ptq->sel_timestamp) { + ret = intel_pt_fast_forward(ptq->decoder, + ptq->sel_timestamp); + if (ret) + return ret; + } + while (1) { state = intel_pt_decode(ptq->decoder); if (state->err) { @@ -978,6 +1094,9 @@ static int intel_pt_setup_queue(struct intel_pt *pt, queue_nr, ptq->timestamp); ptq->state = state; ptq->have_sample = true; + if (ptq->sel_start && ptq->sel_timestamp && + ptq->timestamp < ptq->sel_timestamp) + ptq->have_sample = false; intel_pt_sample_flags(ptq); ret = auxtrace_heap__add(&pt->heap, queue_nr, ptq->timestamp); if (ret) @@ -1153,6 +1272,13 @@ static int intel_pt_synth_branch_sample(struct intel_pt_queue *ptq) sample.branch_stack = (struct branch_stack *)&dummy_bs; } + sample.cyc_cnt = ptq->ipc_cyc_cnt - ptq->last_br_cyc_cnt; + if (sample.cyc_cnt) { + sample.insn_cnt = ptq->ipc_insn_cnt - ptq->last_br_insn_cnt; + ptq->last_br_insn_cnt = ptq->ipc_insn_cnt; + ptq->last_br_cyc_cnt = ptq->ipc_cyc_cnt; + } + return intel_pt_deliver_synth_b_event(pt, event, &sample, pt->branches_sample_type); } @@ -1208,6 +1334,13 @@ static int intel_pt_synth_instruction_sample(struct intel_pt_queue *ptq) sample.stream_id = ptq->pt->instructions_id; sample.period = ptq->state->tot_insn_cnt - ptq->last_insn_cnt; + sample.cyc_cnt = ptq->ipc_cyc_cnt - ptq->last_in_cyc_cnt; + if (sample.cyc_cnt) { + sample.insn_cnt = ptq->ipc_insn_cnt - ptq->last_in_insn_cnt; + ptq->last_in_insn_cnt = ptq->ipc_insn_cnt; + ptq->last_in_cyc_cnt = ptq->ipc_cyc_cnt; + } + ptq->last_insn_cnt = ptq->state->tot_insn_cnt; return intel_pt_deliver_synth_event(pt, ptq, event, &sample, @@ -1479,6 +1612,15 @@ static int intel_pt_sample(struct intel_pt_queue *ptq) ptq->have_sample = false; + if (ptq->state->tot_cyc_cnt > ptq->ipc_cyc_cnt) { + /* + * Cycle count and instruction count only go together to create + * a valid IPC ratio when the cycle count changes. + */ + ptq->ipc_insn_cnt = ptq->state->tot_insn_cnt; + ptq->ipc_cyc_cnt = ptq->state->tot_cyc_cnt; + } + if (pt->sample_pwr_events && (state->type & INTEL_PT_PWR_EVT)) { if (state->type & INTEL_PT_CBR_CHG) { err = intel_pt_synth_cbr_sample(ptq); @@ -1641,10 +1783,83 @@ static void intel_pt_enable_sync_switch(struct intel_pt *pt) } } +/* + * To filter against time ranges, it is only necessary to look at the next start + * or end time. + */ +static bool intel_pt_next_time(struct intel_pt_queue *ptq) +{ + struct intel_pt *pt = ptq->pt; + + if (ptq->sel_start) { + /* Next time is an end time */ + ptq->sel_start = false; + ptq->sel_timestamp = pt->time_ranges[ptq->sel_idx].end; + return true; + } else if (ptq->sel_idx + 1 < pt->range_cnt) { + /* Next time is a start time */ + ptq->sel_start = true; + ptq->sel_idx += 1; + ptq->sel_timestamp = pt->time_ranges[ptq->sel_idx].start; + return true; + } + + /* No next time */ + return false; +} + +static int intel_pt_time_filter(struct intel_pt_queue *ptq, u64 *ff_timestamp) +{ + int err; + + while (1) { + if (ptq->sel_start) { + if (ptq->timestamp >= ptq->sel_timestamp) { + /* After start time, so consider next time */ + intel_pt_next_time(ptq); + if (!ptq->sel_timestamp) { + /* No end time */ + return 0; + } + /* Check against end time */ + continue; + } + /* Before start time, so fast forward */ + ptq->have_sample = false; + if (ptq->sel_timestamp > *ff_timestamp) { + if (ptq->sync_switch) { + intel_pt_next_tid(ptq->pt, ptq); + ptq->switch_state = INTEL_PT_SS_UNKNOWN; + } + *ff_timestamp = ptq->sel_timestamp; + err = intel_pt_fast_forward(ptq->decoder, + ptq->sel_timestamp); + if (err) + return err; + } + return 0; + } else if (ptq->timestamp > ptq->sel_timestamp) { + /* After end time, so consider next time */ + if (!intel_pt_next_time(ptq)) { + /* No next time range, so stop decoding */ + ptq->have_sample = false; + ptq->switch_state = INTEL_PT_SS_NOT_TRACING; + return 1; + } + /* Check against next start time */ + continue; + } else { + /* Before end time */ + return 0; + } + } +} + static int intel_pt_run_decoder(struct intel_pt_queue *ptq, u64 *timestamp) { const struct intel_pt_state *state = ptq->state; struct intel_pt *pt = ptq->pt; + u64 ff_timestamp = 0; int err; if (!pt->kernel_start) { @@ -1709,6 +1924,12 @@ static int intel_pt_run_decoder(struct intel_pt_queue *ptq, u64 *timestamp) ptq->timestamp = state->timestamp; } + if (ptq->sel_timestamp) { + err = intel_pt_time_filter(ptq, &ff_timestamp); + if (err) + return err; + } + if (!pt->timeless_decoding && ptq->timestamp >= *timestamp) { *timestamp = ptq->timestamp; return 0; @@ -2114,6 +2335,7 @@ static void intel_pt_free(struct perf_session *session) thread__put(pt->unknown_thread); addr_filters__exit(&pt->filts); zfree(&pt->filter); + zfree(&pt->time_ranges); free(pt); } @@ -2411,6 +2633,85 @@ static int intel_pt_perf_config(const char *var, const char *value, void *data) return 0; } +/* Find least TSC which converts to ns or later */ +static u64 intel_pt_tsc_start(u64 ns, struct intel_pt *pt) +{ + u64 tsc, tm; + + tsc = perf_time_to_tsc(ns, &pt->tc); + + while (1) { + tm = tsc_to_perf_time(tsc, &pt->tc); + if (tm < ns) + break; + tsc -= 1; + } + + while (tm < ns) + tm = tsc_to_perf_time(++tsc, &pt->tc); + + return tsc; +} + +/* Find greatest TSC which converts to ns or earlier */ +static u64 intel_pt_tsc_end(u64 ns, struct intel_pt *pt) +{ + u64 tsc, tm; + + tsc = perf_time_to_tsc(ns, &pt->tc); + + while (1) { + tm = tsc_to_perf_time(tsc, &pt->tc); + if (tm > ns) + break; + tsc += 1; + } + + while (tm > ns) + tm = tsc_to_perf_time(--tsc, &pt->tc); + + return tsc; +} + +static int intel_pt_setup_time_ranges(struct intel_pt *pt, + struct itrace_synth_opts *opts) +{ + struct perf_time_interval *p = opts->ptime_range; + int n = opts->range_num; + int i; + + if (!n || !p || pt->timeless_decoding) + return 0; + + pt->time_ranges = calloc(n, sizeof(struct range)); + if (!pt->time_ranges) + return -ENOMEM; + + pt->range_cnt = n; + + intel_pt_log("%s: %u range(s)\n", __func__, n); + + for (i = 0; i < n; i++) { + struct range *r = &pt->time_ranges[i]; + u64 ts = p[i].start; + u64 te = p[i].end; + + /* + * Take care to ensure the TSC range matches the perf-time range + * when converted back to perf-time. + */ + r->start = ts ? intel_pt_tsc_start(ts, pt) : 0; + r->end = te ? intel_pt_tsc_end(te, pt) : 0; + + intel_pt_log("range %d: perf time interval: %"PRIu64" to %"PRIu64"\n", + i, ts, te); + intel_pt_log("range %d: TSC time interval: %#"PRIx64" to %#"PRIx64"\n", + i, r->start, r->end); + } + + return 0; +} + static const char * const intel_pt_info_fmts[] = { [INTEL_PT_PMU_TYPE] = " PMU Type %"PRId64"\n", [INTEL_PT_TIME_SHIFT] = " Time Shift %"PRIu64"\n", @@ -2643,6 +2944,12 @@ int intel_pt_process_auxtrace_info(union perf_event *event, pt->cbr2khz = tsc_freq / pt->max_non_turbo_ratio / 1000; } + if (session->itrace_synth_opts) { + err = intel_pt_setup_time_ranges(pt, session->itrace_synth_opts); + if (err) + goto err_delete_thread; + } + if (pt->synth_opts.calls) pt->branches_filter |= PERF_IP_FLAG_CALL | PERF_IP_FLAG_ASYNC | PERF_IP_FLAG_TRACE_END; @@ -2683,6 +2990,7 @@ err_free_queues: err_free: addr_filters__exit(&pt->filts); zfree(&pt->filter); + zfree(&pt->time_ranges); free(pt); return err; } diff --git a/tools/perf/util/perf_regs.h b/tools/perf/util/perf_regs.h index cb9c246c8962..47fe34e5f7d5 100644 --- a/tools/perf/util/perf_regs.h +++ b/tools/perf/util/perf_regs.h @@ -29,12 +29,16 @@ uint64_t arch__user_reg_mask(void); #ifdef HAVE_PERF_REGS_SUPPORT #include <perf_regs.h> +#define DWARF_MINIMAL_REGS ((1ULL << PERF_REG_IP) | (1ULL << PERF_REG_SP)) + int perf_reg_value(u64 *valp, struct regs_dump *regs, int id); #else #define PERF_REGS_MASK 0 #define PERF_REGS_MAX 0 +#define DWARF_MINIMAL_REGS PERF_REGS_MASK + static inline const char *perf_reg_name(int id __maybe_unused) { return NULL; diff --git a/tools/perf/util/s390-cpumsf.c b/tools/perf/util/s390-cpumsf.c index c215704931dc..10d36d9b7909 100644 --- a/tools/perf/util/s390-cpumsf.c +++ b/tools/perf/util/s390-cpumsf.c @@ -17,8 +17,8 @@ * see Documentation/perf.data-file-format.txt. * PERF_RECORD_AUXTRACE_INFO: * Defines a table of contains for PERF_RECORD_AUXTRACE records. This - * record is generated during 'perf record' command. Each record contains up - * to 256 entries describing offset and size of the AUXTRACE data in the + * record is generated during 'perf record' command. Each record contains + * up to 256 entries describing offset and size of the AUXTRACE data in the * perf.data file. * PERF_RECORD_AUXTRACE_ERROR: * Indicates an error during AUXTRACE collection such as buffer overflow. @@ -237,10 +237,33 @@ static int s390_cpumcf_dumpctr(struct s390_cpumsf *sf, return rc; } -/* Display s390 CPU measurement facility basic-sampling data entry */ +/* Display s390 CPU measurement facility basic-sampling data entry + * Data written on s390 in big endian byte order and contains bit + * fields across byte boundaries. + */ static bool s390_cpumsf_basic_show(const char *color, size_t pos, - struct hws_basic_entry *basic) + struct hws_basic_entry *basicp) { + struct hws_basic_entry *basic = basicp; +#if __BYTE_ORDER == __LITTLE_ENDIAN + struct hws_basic_entry local; + unsigned long long word = be64toh(*(unsigned long long *)basicp); + + memset(&local, 0, sizeof(local)); + local.def = be16toh(basicp->def); + local.prim_asn = word & 0xffff; + local.CL = word >> 30 & 0x3; + local.I = word >> 32 & 0x1; + local.AS = word >> 33 & 0x3; + local.P = word >> 35 & 0x1; + local.W = word >> 36 & 0x1; + local.T = word >> 37 & 0x1; + local.U = word >> 40 & 0xf; + local.ia = be64toh(basicp->ia); + local.gpp = be64toh(basicp->gpp); + local.hpp = be64toh(basicp->hpp); + basic = &local; +#endif if (basic->def != 1) { pr_err("Invalid AUX trace basic entry [%#08zx]\n", pos); return false; @@ -258,10 +281,22 @@ static bool s390_cpumsf_basic_show(const char *color, size_t pos, return true; } -/* Display s390 CPU measurement facility diagnostic-sampling data entry */ +/* Display s390 CPU measurement facility diagnostic-sampling data entry. + * Data written on s390 in big endian byte order and contains bit + * fields across byte boundaries. + */ static bool s390_cpumsf_diag_show(const char *color, size_t pos, - struct hws_diag_entry *diag) + struct hws_diag_entry *diagp) { + struct hws_diag_entry *diag = diagp; +#if __BYTE_ORDER == __LITTLE_ENDIAN + struct hws_diag_entry local; + unsigned long long word = be64toh(*(unsigned long long *)diagp); + + local.def = be16toh(diagp->def); + local.I = word >> 32 & 0x1; + diag = &local; +#endif if (diag->def < S390_CPUMSF_DIAG_DEF_FIRST) { pr_err("Invalid AUX trace diagnostic entry [%#08zx]\n", pos); return false; @@ -272,35 +307,52 @@ static bool s390_cpumsf_diag_show(const char *color, size_t pos, } /* Return TOD timestamp contained in an trailer entry */ -static unsigned long long trailer_timestamp(struct hws_trailer_entry *te) +static unsigned long long trailer_timestamp(struct hws_trailer_entry *te, + int idx) { /* te->t set: TOD in STCKE format, bytes 8-15 * to->t not set: TOD in STCK format, bytes 0-7 */ unsigned long long ts; - memcpy(&ts, &te->timestamp[te->t], sizeof(ts)); - return ts; + memcpy(&ts, &te->timestamp[idx], sizeof(ts)); + return be64toh(ts); } /* Display s390 CPU measurement facility trailer entry */ static bool s390_cpumsf_trailer_show(const char *color, size_t pos, struct hws_trailer_entry *te) { +#if __BYTE_ORDER == __LITTLE_ENDIAN + struct hws_trailer_entry local; + const unsigned long long flags = be64toh(te->flags); + + memset(&local, 0, sizeof(local)); + local.f = flags >> 63 & 0x1; + local.a = flags >> 62 & 0x1; + local.t = flags >> 61 & 0x1; + local.bsdes = be16toh((flags >> 16 & 0xffff)); + local.dsdes = be16toh((flags & 0xffff)); + memcpy(&local.timestamp, te->timestamp, sizeof(te->timestamp)); + local.overflow = be64toh(te->overflow); + local.clock_base = be64toh(te->progusage[0]) >> 63 & 1; + local.progusage2 = be64toh(te->progusage2); + te = &local; +#endif if (te->bsdes != sizeof(struct hws_basic_entry)) { pr_err("Invalid AUX trace trailer entry [%#08zx]\n", pos); return false; } color_fprintf(stdout, color, " [%#08zx] Trailer %c%c%c bsdes:%d" " dsdes:%d Overflow:%lld Time:%#llx\n" - "\t\tC:%d TOD:%#lx 1:%#llx 2:%#llx\n", + "\t\tC:%d TOD:%#lx\n", pos, te->f ? 'F' : ' ', te->a ? 'A' : ' ', te->t ? 'T' : ' ', te->bsdes, te->dsdes, te->overflow, - trailer_timestamp(te), te->clock_base, te->progusage2, - te->progusage[0], te->progusage[1]); + trailer_timestamp(te, te->clock_base), + te->clock_base, te->progusage2); return true; } @@ -327,13 +379,13 @@ static bool s390_cpumsf_validate(int machine_type, *dsdes = *bsdes = 0; if (len & (S390_CPUMSF_PAGESZ - 1)) /* Illegal size */ return false; - if (basic->def != 1) /* No basic set entry, must be first */ + if (be16toh(basic->def) != 1) /* No basic set entry, must be first */ return false; /* Check for trailer entry at end of SDB */ te = (struct hws_trailer_entry *)(buf + S390_CPUMSF_PAGESZ - sizeof(*te)); - *bsdes = te->bsdes; - *dsdes = te->dsdes; + *bsdes = be16toh(te->bsdes); + *dsdes = be16toh(te->dsdes); if (!te->bsdes && !te->dsdes) { /* Very old hardware, use CPUID */ switch (machine_type) { @@ -495,19 +547,27 @@ static bool s390_cpumsf_make_event(size_t pos, static unsigned long long get_trailer_time(const unsigned char *buf) { struct hws_trailer_entry *te; - unsigned long long aux_time; + unsigned long long aux_time, progusage2; + bool clock_base; te = (struct hws_trailer_entry *)(buf + S390_CPUMSF_PAGESZ - sizeof(*te)); - if (!te->clock_base) /* TOD_CLOCK_BASE value missing */ +#if __BYTE_ORDER == __LITTLE_ENDIAN + clock_base = be64toh(te->progusage[0]) >> 63 & 0x1; + progusage2 = be64toh(te->progusage[1]); +#else + clock_base = te->clock_base; + progusage2 = te->progusage2; +#endif + if (!clock_base) /* TOD_CLOCK_BASE value missing */ return 0; /* Correct calculation to convert time stamp in trailer entry to * nano seconds (taken from arch/s390 function tod_to_ns()). * TOD_CLOCK_BASE is stored in trailer entry member progusage2. */ - aux_time = trailer_timestamp(te) - te->progusage2; + aux_time = trailer_timestamp(te, clock_base) - progusage2; aux_time = (aux_time >> 9) * 125 + (((aux_time & 0x1ff) * 125) >> 9); return aux_time; } diff --git a/tools/perf/util/scripting-engines/trace-event-python.c b/tools/perf/util/scripting-engines/trace-event-python.c index 22f52b669871..6acb379b53ec 100644 --- a/tools/perf/util/scripting-engines/trace-event-python.c +++ b/tools/perf/util/scripting-engines/trace-event-python.c @@ -1111,7 +1111,7 @@ static int python_export_sample(struct db_export *dbe, struct tables *tables = container_of(dbe, struct tables, dbe); PyObject *t; - t = tuple_new(22); + t = tuple_new(24); tuple_set_u64(t, 0, es->db_id); tuple_set_u64(t, 1, es->evsel->db_id); @@ -1135,6 +1135,8 @@ static int python_export_sample(struct db_export *dbe, tuple_set_s32(t, 19, es->sample->flags & PERF_BRANCH_MASK); tuple_set_s32(t, 20, !!(es->sample->flags & PERF_IP_FLAG_IN_TX)); tuple_set_u64(t, 21, es->call_path_id); + tuple_set_u64(t, 22, es->sample->insn_cnt); + tuple_set_u64(t, 23, es->sample->cyc_cnt); call_object(tables->sample_handler, t, "sample_table"); @@ -1173,7 +1175,7 @@ static int python_export_call_return(struct db_export *dbe, u64 comm_db_id = cr->comm ? cr->comm->db_id : 0; PyObject *t; - t = tuple_new(12); + t = tuple_new(14); tuple_set_u64(t, 0, cr->db_id); tuple_set_u64(t, 1, cr->thread->db_id); @@ -1187,6 +1189,8 @@ static int python_export_call_return(struct db_export *dbe, tuple_set_u64(t, 9, cr->cp->parent->db_id); tuple_set_s32(t, 10, cr->flags); tuple_set_u64(t, 11, cr->parent_db_id); + tuple_set_u64(t, 12, cr->insn_count); + tuple_set_u64(t, 13, cr->cyc_count); call_object(tables->call_return_handler, t, "call_return_table"); diff --git a/tools/perf/util/smt.c b/tools/perf/util/smt.c index 453f6f6f29f3..3b791ef2cd50 100644 --- a/tools/perf/util/smt.c +++ b/tools/perf/util/smt.c @@ -23,8 +23,12 @@ int smt_on(void) char fn[256]; snprintf(fn, sizeof fn, - "devices/system/cpu/cpu%d/topology/thread_siblings", - cpu); + "devices/system/cpu/cpu%d/topology/core_cpus", cpu); + if (access(fn, F_OK) == -1) { + snprintf(fn, sizeof fn, + "devices/system/cpu/cpu%d/topology/thread_siblings", + cpu); + } if (sysfs__read_str(fn, &str, &strlen) < 0) continue; /* Entry is hex, but does not have 0x, so need custom parser */ diff --git a/tools/perf/util/stat-display.c b/tools/perf/util/stat-display.c index 4c53bae5644b..a6b9de3e83fc 100644 --- a/tools/perf/util/stat-display.c +++ b/tools/perf/util/stat-display.c @@ -69,8 +69,9 @@ static void aggr_printout(struct perf_stat_config *config, { switch (config->aggr_mode) { case AGGR_CORE: - fprintf(config->output, "S%d-C%*d%s%*d%s", + fprintf(config->output, "S%d-D%d-C%*d%s%*d%s", cpu_map__id_to_socket(id), + cpu_map__id_to_die(id), config->csv_output ? 0 : -8, cpu_map__id_to_cpu(id), config->csv_sep, @@ -78,6 +79,16 @@ static void aggr_printout(struct perf_stat_config *config, nr, config->csv_sep); break; + case AGGR_DIE: + fprintf(config->output, "S%d-D%*d%s%*d%s", + cpu_map__id_to_socket(id << 16), + config->csv_output ? 0 : -8, + cpu_map__id_to_die(id << 16), + config->csv_sep, + config->csv_output ? 0 : 4, + nr, + config->csv_sep); + break; case AGGR_SOCKET: fprintf(config->output, "S%*d%s%*d%s", config->csv_output ? 0 : -5, @@ -89,8 +100,9 @@ static void aggr_printout(struct perf_stat_config *config, break; case AGGR_NONE: if (evsel->percore) { - fprintf(config->output, "S%d-C%*d%s", + fprintf(config->output, "S%d-D%d-C%*d%s", cpu_map__id_to_socket(id), + cpu_map__id_to_die(id), config->csv_output ? 0 : -5, cpu_map__id_to_cpu(id), config->csv_sep); } else { @@ -407,6 +419,7 @@ static void printout(struct perf_stat_config *config, int id, int nr, [AGGR_THREAD] = 1, [AGGR_NONE] = 1, [AGGR_SOCKET] = 2, + [AGGR_DIE] = 2, [AGGR_CORE] = 2, }; @@ -879,7 +892,8 @@ static void print_no_aggr_metric(struct perf_stat_config *config, } static int aggr_header_lens[] = { - [AGGR_CORE] = 18, + [AGGR_CORE] = 24, + [AGGR_DIE] = 18, [AGGR_SOCKET] = 12, [AGGR_NONE] = 6, [AGGR_THREAD] = 24, @@ -888,6 +902,7 @@ static int aggr_header_lens[] = { static const char *aggr_header_csv[] = { [AGGR_CORE] = "core,cpus,", + [AGGR_DIE] = "die,cpus", [AGGR_SOCKET] = "socket,cpus", [AGGR_NONE] = "cpu,", [AGGR_THREAD] = "comm-pid,", @@ -954,8 +969,13 @@ static void print_interval(struct perf_stat_config *config, if (!metric_only) fprintf(output, " counts %*s events\n", unit_width, "unit"); break; + case AGGR_DIE: + fprintf(output, "# time die cpus"); + if (!metric_only) + fprintf(output, " counts %*s events\n", unit_width, "unit"); + break; case AGGR_CORE: - fprintf(output, "# time core cpus"); + fprintf(output, "# time core cpus"); if (!metric_only) fprintf(output, " counts %*s events\n", unit_width, "unit"); break; @@ -1165,6 +1185,7 @@ perf_evlist__print_counters(struct perf_evlist *evlist, switch (config->aggr_mode) { case AGGR_CORE: + case AGGR_DIE: case AGGR_SOCKET: print_aggr(config, evlist, prefix); break; diff --git a/tools/perf/util/stat-shadow.c b/tools/perf/util/stat-shadow.c index 83d8094be4fe..027b09aaa4cf 100644 --- a/tools/perf/util/stat-shadow.c +++ b/tools/perf/util/stat-shadow.c @@ -12,6 +12,7 @@ /* * AGGR_GLOBAL: Use CPU 0 * AGGR_SOCKET: Use first CPU of socket + * AGGR_DIE: Use first CPU of die * AGGR_CORE: Use first CPU of core * AGGR_NONE: Use matching CPU * AGGR_THREAD: Not supported? diff --git a/tools/perf/util/stat.c b/tools/perf/util/stat.c index c3115d939b0b..d91fe754b6d2 100644 --- a/tools/perf/util/stat.c +++ b/tools/perf/util/stat.c @@ -272,6 +272,7 @@ process_counter_values(struct perf_stat_config *config, struct perf_evsel *evsel switch (config->aggr_mode) { case AGGR_THREAD: case AGGR_CORE: + case AGGR_DIE: case AGGR_SOCKET: case AGGR_NONE: if (!evsel->snapshot) diff --git a/tools/perf/util/stat.h b/tools/perf/util/stat.h index 2f9c9159a364..7032dd1eeac2 100644 --- a/tools/perf/util/stat.h +++ b/tools/perf/util/stat.h @@ -44,6 +44,7 @@ enum aggr_mode { AGGR_NONE, AGGR_GLOBAL, AGGR_SOCKET, + AGGR_DIE, AGGR_CORE, AGGR_THREAD, AGGR_UNSET, diff --git a/tools/perf/util/symbol-elf.c b/tools/perf/util/symbol-elf.c index 4ad106a5f2c0..fdc5bd7dbb90 100644 --- a/tools/perf/util/symbol-elf.c +++ b/tools/perf/util/symbol-elf.c @@ -699,7 +699,6 @@ bool __weak elf__needs_adjust_symbols(GElf_Ehdr ehdr) int symsrc__init(struct symsrc *ss, struct dso *dso, const char *name, enum dso_binary_type type) { - int err = -1; GElf_Ehdr ehdr; Elf *elf; int fd; @@ -793,7 +792,7 @@ out_elf_end: elf_end(elf); out_close: close(fd); - return err; + return -1; } /** diff --git a/tools/perf/util/thread-stack.c b/tools/perf/util/thread-stack.c index 4ba9e866b076..c485186a8b6d 100644 --- a/tools/perf/util/thread-stack.c +++ b/tools/perf/util/thread-stack.c @@ -40,6 +40,8 @@ enum retpoline_state_t { * @timestamp: timestamp (if known) * @ref: external reference (e.g. db_id of sample) * @branch_count: the branch count when the entry was created + * @insn_count: the instruction count when the entry was created + * @cyc_count the cycle count when the entry was created * @db_id: id used for db-export * @cp: call path * @no_call: a 'call' was not seen @@ -51,6 +53,8 @@ struct thread_stack_entry { u64 timestamp; u64 ref; u64 branch_count; + u64 insn_count; + u64 cyc_count; u64 db_id; struct call_path *cp; bool no_call; @@ -66,6 +70,8 @@ struct thread_stack_entry { * @sz: current maximum stack size * @trace_nr: current trace number * @branch_count: running branch count + * @insn_count: running instruction count + * @cyc_count running cycle count * @kernel_start: kernel start address * @last_time: last timestamp * @crp: call/return processor @@ -79,6 +85,8 @@ struct thread_stack { size_t sz; u64 trace_nr; u64 branch_count; + u64 insn_count; + u64 cyc_count; u64 kernel_start; u64 last_time; struct call_return_processor *crp; @@ -280,6 +288,8 @@ static int thread_stack__call_return(struct thread *thread, cr.call_time = tse->timestamp; cr.return_time = timestamp; cr.branch_count = ts->branch_count - tse->branch_count; + cr.insn_count = ts->insn_count - tse->insn_count; + cr.cyc_count = ts->cyc_count - tse->cyc_count; cr.db_id = tse->db_id; cr.call_ref = tse->ref; cr.return_ref = ref; @@ -535,6 +545,8 @@ static int thread_stack__push_cp(struct thread_stack *ts, u64 ret_addr, tse->timestamp = timestamp; tse->ref = ref; tse->branch_count = ts->branch_count; + tse->insn_count = ts->insn_count; + tse->cyc_count = ts->cyc_count; tse->cp = cp; tse->no_call = no_call; tse->trace_end = trace_end; @@ -865,6 +877,8 @@ int thread_stack__process(struct thread *thread, struct comm *comm, } ts->branch_count += 1; + ts->insn_count += sample->insn_cnt; + ts->cyc_count += sample->cyc_cnt; ts->last_time = sample->time; if (sample->flags & PERF_IP_FLAG_CALL) { diff --git a/tools/perf/util/thread-stack.h b/tools/perf/util/thread-stack.h index 71e15d4ec533..e1ec5a58f1b2 100644 --- a/tools/perf/util/thread-stack.h +++ b/tools/perf/util/thread-stack.h @@ -43,6 +43,8 @@ enum { * @call_time: timestamp of call (if known) * @return_time: timestamp of return (if known) * @branch_count: number of branches seen between call and return + * @insn_count: approx. number of instructions between call and return + * @cyc_count: approx. number of cycles between call and return * @call_ref: external reference to 'call' sample (e.g. db_id) * @return_ref: external reference to 'return' sample (e.g. db_id) * @db_id: id used for db-export @@ -56,6 +58,8 @@ struct call_return { u64 call_time; u64 return_time; u64 branch_count; + u64 insn_count; + u64 cyc_count; u64 call_ref; u64 return_ref; u64 db_id; diff --git a/tools/perf/util/time-utils.c b/tools/perf/util/time-utils.c index 20663a460df3..2b48816a2d2e 100644 --- a/tools/perf/util/time-utils.c +++ b/tools/perf/util/time-utils.c @@ -7,6 +7,7 @@ #include <errno.h> #include <inttypes.h> #include <math.h> +#include <ctype.h> #include "perf.h" #include "debug.h" @@ -116,6 +117,69 @@ int perf_time__parse_str(struct perf_time_interval *ptime, const char *ostr) return rc; } +static int perf_time__parse_strs(struct perf_time_interval *ptime, + const char *ostr, int size) +{ + const char *cp; + char *str, *arg, *p; + int i, num = 0, rc = 0; + + /* Count the commas */ + for (cp = ostr; *cp; cp++) + num += !!(*cp == ','); + + if (!num) + return -EINVAL; + + BUG_ON(num > size); + + str = strdup(ostr); + if (!str) + return -ENOMEM; + + /* Split the string and parse each piece, except the last */ + for (i = 0, p = str; i < num - 1; i++) { + arg = p; + /* Find next comma, there must be one */ + p = strchr(p, ',') + 1; + /* Skip white space */ + while (isspace(*p)) + p++; + /* Skip the value, must not contain space or comma */ + while (*p && !isspace(*p)) { + if (*p++ == ',') { + rc = -EINVAL; + goto out; + } + } + /* Split and parse */ + if (*p) + *p++ = 0; + rc = perf_time__parse_str(ptime + i, arg); + if (rc < 0) + goto out; + } + + /* Parse the last piece */ + rc = perf_time__parse_str(ptime + i, p); + if (rc < 0) + goto out; + + /* Check there is no overlap */ + for (i = 0; i < num - 1; i++) { + if (ptime[i].end >= ptime[i + 1].start) { + rc = -EINVAL; + goto out; + } + } + + rc = num; +out: + free(str); + + return rc; +} + static int parse_percent(double *pcnt, char *str) { char *c, *endptr; @@ -135,12 +199,30 @@ static int parse_percent(double *pcnt, char *str) return 0; } +static int set_percent_time(struct perf_time_interval *ptime, double start_pcnt, + double end_pcnt, u64 start, u64 end) +{ + u64 total = end - start; + + if (start_pcnt < 0.0 || start_pcnt > 1.0 || + end_pcnt < 0.0 || end_pcnt > 1.0) { + return -1; + } + + ptime->start = start + round(start_pcnt * total); + ptime->end = start + round(end_pcnt * total); + + if (ptime->end > ptime->start && ptime->end != end) + ptime->end -= 1; + + return 0; +} + static int percent_slash_split(char *str, struct perf_time_interval *ptime, u64 start, u64 end) { char *p, *end_str; double pcnt, start_pcnt, end_pcnt; - u64 total = end - start; int i; /* @@ -168,15 +250,7 @@ static int percent_slash_split(char *str, struct perf_time_interval *ptime, start_pcnt = pcnt * (i - 1); end_pcnt = pcnt * i; - if (start_pcnt < 0.0 || start_pcnt > 1.0 || - end_pcnt < 0.0 || end_pcnt > 1.0) { - return -1; - } - - ptime->start = start + round(start_pcnt * total); - ptime->end = start + round(end_pcnt * total); - - return 0; + return set_percent_time(ptime, start_pcnt, end_pcnt, start, end); } static int percent_dash_split(char *str, struct perf_time_interval *ptime, @@ -184,7 +258,6 @@ static int percent_dash_split(char *str, struct perf_time_interval *ptime, { char *start_str = NULL, *end_str; double start_pcnt, end_pcnt; - u64 total = end - start; int ret; /* @@ -203,16 +276,7 @@ static int percent_dash_split(char *str, struct perf_time_interval *ptime, free(start_str); - if (start_pcnt < 0.0 || start_pcnt > 1.0 || - end_pcnt < 0.0 || end_pcnt > 1.0 || - start_pcnt > end_pcnt) { - return -1; - } - - ptime->start = start + round(start_pcnt * total); - ptime->end = start + round(end_pcnt * total); - - return 0; + return set_percent_time(ptime, start_pcnt, end_pcnt, start, end); } typedef int (*time_pecent_split)(char *, struct perf_time_interval *, @@ -389,13 +453,12 @@ bool perf_time__ranges_skip_sample(struct perf_time_interval *ptime_buf, ptime = &ptime_buf[i]; if (timestamp >= ptime->start && - ((timestamp < ptime->end && i < num - 1) || - (timestamp <= ptime->end && i == num - 1))) { - break; + (timestamp <= ptime->end || !ptime->end)) { + return false; } } - return (i == num) ? true : false; + return true; } int perf_time__parse_for_ranges(const char *time_str, @@ -403,20 +466,20 @@ int perf_time__parse_for_ranges(const char *time_str, struct perf_time_interval **ranges, int *range_size, int *range_num) { + bool has_percent = strchr(time_str, '%'); struct perf_time_interval *ptime_range; - int size, num, ret; + int size, num, ret = -EINVAL; ptime_range = perf_time__range_alloc(time_str, &size); if (!ptime_range) return -ENOMEM; - if (perf_time__parse_str(ptime_range, time_str) != 0) { + if (has_percent) { if (session->evlist->first_sample_time == 0 && session->evlist->last_sample_time == 0) { pr_err("HINT: no first/last sample time found in perf data.\n" "Please use latest perf binary to execute 'perf record'\n" "(if '--buildid-all' is enabled, please set '--timestamp-boundary').\n"); - ret = -EINVAL; goto error; } @@ -425,21 +488,20 @@ int perf_time__parse_for_ranges(const char *time_str, time_str, session->evlist->first_sample_time, session->evlist->last_sample_time); - - if (num < 0) { - pr_err("Invalid time string\n"); - ret = -EINVAL; - goto error; - } } else { - num = 1; + num = perf_time__parse_strs(ptime_range, time_str, size); } + if (num < 0) + goto error_invalid; + *range_size = size; *range_num = num; *ranges = ptime_range; return 0; +error_invalid: + pr_err("Invalid time string\n"); error: free(ptime_range); return ret; |