summaryrefslogtreecommitdiffstats
path: root/tools/perf/Documentation/topdown.txt
diff options
context:
space:
mode:
Diffstat (limited to 'tools/perf/Documentation/topdown.txt')
-rw-r--r--tools/perf/Documentation/topdown.txt256
1 files changed, 256 insertions, 0 deletions
diff --git a/tools/perf/Documentation/topdown.txt b/tools/perf/Documentation/topdown.txt
new file mode 100644
index 000000000..3c39bb3dc
--- /dev/null
+++ b/tools/perf/Documentation/topdown.txt
@@ -0,0 +1,256 @@
+Using TopDown metrics in user space
+-----------------------------------
+
+Intel CPUs (since Sandy Bridge and Silvermont) support a TopDown
+methology to break down CPU pipeline execution into 4 bottlenecks:
+frontend bound, backend bound, bad speculation, retiring.
+
+For more details on Topdown see [1][5]
+
+Traditionally this was implemented by events in generic counters
+and specific formulas to compute the bottlenecks.
+
+perf stat --topdown implements this.
+
+Full Top Down includes more levels that can break down the
+bottlenecks further. This is not directly implemented in perf,
+but available in other tools that can run on top of perf,
+such as toplev[2] or vtune[3]
+
+New Topdown features in Ice Lake
+===============================
+
+With Ice Lake CPUs the TopDown metrics are directly available as
+fixed counters and do not require generic counters. This allows
+to collect TopDown always in addition to other events.
+
+% perf stat -a --topdown -I1000
+# time retiring bad speculation frontend bound backend bound
+ 1.001281330 23.0% 15.3% 29.6% 32.1%
+ 2.003009005 5.0% 6.8% 46.6% 41.6%
+ 3.004646182 6.7% 6.7% 46.0% 40.6%
+ 4.006326375 5.0% 6.4% 47.6% 41.0%
+ 5.007991804 5.1% 6.3% 46.3% 42.3%
+ 6.009626773 6.2% 7.1% 47.3% 39.3%
+ 7.011296356 4.7% 6.7% 46.2% 42.4%
+ 8.012951831 4.7% 6.7% 47.5% 41.1%
+...
+
+This also enables measuring TopDown per thread/process instead
+of only per core.
+
+Using TopDown through RDPMC in applications on Ice Lake
+======================================================
+
+For more fine grained measurements it can be useful to
+access the new directly from user space. This is more complicated,
+but drastically lowers overhead.
+
+On Ice Lake, there is a new fixed counter 3: SLOTS, which reports
+"pipeline SLOTS" (cycles multiplied by core issue width) and a
+metric register that reports slots ratios for the different bottleneck
+categories.
+
+The metrics counter is CPU model specific and is not available on older
+CPUs.
+
+Example code
+============
+
+Library functions to do the functionality described below
+is also available in libjevents [4]
+
+The application opens a group with fixed counter 3 (SLOTS) and any
+metric event, and allow user programs to read the performance counters.
+
+Fixed counter 3 is mapped to a pseudo event event=0x00, umask=04,
+so the perf_event_attr structure should be initialized with
+{ .config = 0x0400, .type = PERF_TYPE_RAW }
+The metric events are mapped to the pseudo event event=0x00, umask=0x8X.
+For example, the perf_event_attr structure can be initialized with
+{ .config = 0x8000, .type = PERF_TYPE_RAW } for Retiring metric event
+The Fixed counter 3 must be the leader of the group.
+
+#include <linux/perf_event.h>
+#include <sys/syscall.h>
+#include <unistd.h>
+
+/* Provide own perf_event_open stub because glibc doesn't */
+__attribute__((weak))
+int perf_event_open(struct perf_event_attr *attr, pid_t pid,
+ int cpu, int group_fd, unsigned long flags)
+{
+ return syscall(__NR_perf_event_open, attr, pid, cpu, group_fd, flags);
+}
+
+/* Open slots counter file descriptor for current task. */
+struct perf_event_attr slots = {
+ .type = PERF_TYPE_RAW,
+ .size = sizeof(struct perf_event_attr),
+ .config = 0x400,
+ .exclude_kernel = 1,
+};
+
+int slots_fd = perf_event_open(&slots, 0, -1, -1, 0);
+if (slots_fd < 0)
+ ... error ...
+
+/*
+ * Open metrics event file descriptor for current task.
+ * Set slots event as the leader of the group.
+ */
+struct perf_event_attr metrics = {
+ .type = PERF_TYPE_RAW,
+ .size = sizeof(struct perf_event_attr),
+ .config = 0x8000,
+ .exclude_kernel = 1,
+};
+
+int metrics_fd = perf_event_open(&metrics, 0, -1, slots_fd, 0);
+if (metrics_fd < 0)
+ ... error ...
+
+
+The RDPMC instruction (or _rdpmc compiler intrinsic) can now be used
+to read slots and the topdown metrics at different points of the program:
+
+#include <stdint.h>
+#include <x86intrin.h>
+
+#define RDPMC_FIXED (1 << 30) /* return fixed counters */
+#define RDPMC_METRIC (1 << 29) /* return metric counters */
+
+#define FIXED_COUNTER_SLOTS 3
+#define METRIC_COUNTER_TOPDOWN_L1 0
+
+static inline uint64_t read_slots(void)
+{
+ return _rdpmc(RDPMC_FIXED | FIXED_COUNTER_SLOTS);
+}
+
+static inline uint64_t read_metrics(void)
+{
+ return _rdpmc(RDPMC_METRIC | METRIC_COUNTER_TOPDOWN_L1);
+}
+
+Then the program can be instrumented to read these metrics at different
+points.
+
+It's not a good idea to do this with too short code regions,
+as the parallelism and overlap in the CPU program execution will
+cause too much measurement inaccuracy. For example instrumenting
+individual basic blocks is definitely too fine grained.
+
+Decoding metrics values
+=======================
+
+The value reported by read_metrics() contains four 8 bit fields
+that represent a scaled ratio that represent the Level 1 bottleneck.
+All four fields add up to 0xff (= 100%)
+
+The binary ratios in the metric value can be converted to float ratios:
+
+#define GET_METRIC(m, i) (((m) >> (i*8)) & 0xff)
+
+#define TOPDOWN_RETIRING(val) ((float)GET_METRIC(val, 0) / 0xff)
+#define TOPDOWN_BAD_SPEC(val) ((float)GET_METRIC(val, 1) / 0xff)
+#define TOPDOWN_FE_BOUND(val) ((float)GET_METRIC(val, 2) / 0xff)
+#define TOPDOWN_BE_BOUND(val) ((float)GET_METRIC(val, 3) / 0xff)
+
+and then converted to percent for printing.
+
+The ratios in the metric accumulate for the time when the counter
+is enabled. For measuring programs it is often useful to measure
+specific sections. For this it is needed to deltas on metrics.
+
+This can be done by scaling the metrics with the slots counter
+read at the same time.
+
+Then it's possible to take deltas of these slots counts
+measured at different points, and determine the metrics
+for that time period.
+
+ slots_a = read_slots();
+ metric_a = read_metrics();
+
+ ... larger code region ...
+
+ slots_b = read_slots()
+ metric_b = read_metrics()
+
+ # compute scaled metrics for measurement a
+ retiring_slots_a = GET_METRIC(metric_a, 0) * slots_a
+ bad_spec_slots_a = GET_METRIC(metric_a, 1) * slots_a
+ fe_bound_slots_a = GET_METRIC(metric_a, 2) * slots_a
+ be_bound_slots_a = GET_METRIC(metric_a, 3) * slots_a
+
+ # compute delta scaled metrics between b and a
+ retiring_slots = GET_METRIC(metric_b, 0) * slots_b - retiring_slots_a
+ bad_spec_slots = GET_METRIC(metric_b, 1) * slots_b - bad_spec_slots_a
+ fe_bound_slots = GET_METRIC(metric_b, 2) * slots_b - fe_bound_slots_a
+ be_bound_slots = GET_METRIC(metric_b, 3) * slots_b - be_bound_slots_a
+
+Later the individual ratios for the measurement period can be recreated
+from these counts.
+
+ slots_delta = slots_b - slots_a
+ retiring_ratio = (float)retiring_slots / slots_delta
+ bad_spec_ratio = (float)bad_spec_slots / slots_delta
+ fe_bound_ratio = (float)fe_bound_slots / slots_delta
+ be_bound_ratio = (float)be_bound_slots / slota_delta
+
+ printf("Retiring %.2f%% Bad Speculation %.2f%% FE Bound %.2f%% BE Bound %.2f%%\n",
+ retiring_ratio * 100.,
+ bad_spec_ratio * 100.,
+ fe_bound_ratio * 100.,
+ be_bound_ratio * 100.);
+
+Resetting metrics counters
+==========================
+
+Since the individual metrics are only 8bit they lose precision for
+short regions over time because the number of cycles covered by each
+fraction bit shrinks. So the counters need to be reset regularly.
+
+When using the kernel perf API the kernel resets on every read.
+So as long as the reading is at reasonable intervals (every few
+seconds) the precision is good.
+
+When using perf stat it is recommended to always use the -I option,
+with no longer interval than a few seconds
+
+ perf stat -I 1000 --topdown ...
+
+For user programs using RDPMC directly the counter can
+be reset explicitly using ioctl:
+
+ ioctl(perf_fd, PERF_EVENT_IOC_RESET, 0);
+
+This "opens" a new measurement period.
+
+A program using RDPMC for TopDown should schedule such a reset
+regularly, as in every few seconds.
+
+Limits on Ice Lake
+==================
+
+Four pseudo TopDown metric events are exposed for the end-users,
+topdown-retiring, topdown-bad-spec, topdown-fe-bound and topdown-be-bound.
+They can be used to collect the TopDown value under the following
+rules:
+- All the TopDown metric events must be in a group with the SLOTS event.
+- The SLOTS event must be the leader of the group.
+- The PERF_FORMAT_GROUP flag must be applied for each TopDown metric
+ events
+
+The SLOTS event and the TopDown metric events can be counting members of
+a sampling read group. Since the SLOTS event must be the leader of a TopDown
+group, the second event of the group is the sampling event.
+For example, perf record -e '{slots, $sampling_event, topdown-retiring}:S'
+
+
+[1] https://software.intel.com/en-us/top-down-microarchitecture-analysis-method-win
+[2] https://github.com/andikleen/pmu-tools/wiki/toplev-manual
+[3] https://software.intel.com/en-us/intel-vtune-amplifier-xe
+[4] https://github.com/andikleen/pmu-tools/tree/master/jevents
+[5] https://sites.google.com/site/analysismethods/yasin-pubs