Differences

This shows you the differences between two versions of the page.

--- perf [March 21, 2026 at 18:33] – yanevskiv
+++ perf [August 22, 2026 at 15:22] (current) – created - external edit 127.0.0.1
@@ Line 1: / Line 1: @@
+# perf
+**perf** is a Linux profiling tool that reads hardware performance counters from the CPU. Measure cycles, instructions, cache misses, branches—everything the CPU tracks natively. Answer the fundamental question: where is the program spending time, and why is it slow?
+Unlike guessing, perf gives ground truth from the hardware. Low overhead (1-5%), no recompilation needed, and the data is reliable. In HPC work, perf is the first tool to reach for when something is slower than expected.
+```bash
+$ perf stat ./program
+Performance counter stats for './program':
+,102.38 msec task-clock              # 3.98 CPUs utilized
+,847,203,412 cycles                  # 4.107 GHz
+,334,901,088 instructions            # 0.73 insn per cycle
+,292,771 cache-misses            # 33.33% of cache refs
+```
+IPC (instructions per cycle) of 0.73 on a modern superscalar CPU means the pipeline is stalled. A 33% cache miss rate means it's waiting for memory. Two numbers tell you everything you need to know to start optimizing.
+## Concepts
+. [[perf-basics|Basics]]
+. [[perf-stat|perf stat]]
+. [[perf-record|perf record]]
+. [[perf-report|perf report]]
+. [[perf-annotate|perf annotate]]
+. [[perf-events|Events]]
+. [[perf-metrics|Metrics]]
+. [[perf-flame-graphs|Flame graphs]]
+. [[perf-sampling|Sampling]]
+. [[perf-scripting|Scripting]]