What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
rtla timerlat investigates periodic timer wakeup latency by measuring two stages: when the timer interrupt handler runs and when the awakened timerlat thread runs. Use top for a live summary or hist to examine the distribution of samples; thresholds can stop tracing and preserve a trace or stack context around a delay. These are observations from a particular machine, kernel, workload, and run—not a universal performance guarantee.
How does rtla timerlat measure Linux scheduling latency?
rtla timerlat is the command-line interface to the Linux kernel timerlat tracer. The tracer runs periodic timer threads on CPUs and reports latency at two points in the wakeup path. The kernel tracer compares the current time when the timer wakes with the timer’s previously set absolute expiry time. See the kernel timerlat tracer documentation for Linux 6.12 and the rtla timerlat command manual.
As an Amazon Associate I earn from qualifying purchases.
IRQ latency: when the timer interrupt handler runs
The IRQ-stage measurement records latency at the timer IRQ handler. It indicates how late the interrupt handling is relative to the timer’s intended expiry.
Thread latency: when the timerlat thread runs
The thread-stage measurement records latency when the awakened timerlat thread handler runs. This is a later stage than the interrupt handler, so the measurements are related but not interchangeable. If thread latency is notably greater than IRQ latency, the difference can help focus investigation on delay after the interrupt stage. A capture still needs to be interpreted alongside the machine, workload, kernel configuration, and trace context; the difference alone does not identify a cause.
#1 Best Overall
Should you use rtla timerlat top or hist?
The two modes provide different views of the tracer output. Neither is inherently more accurate; choose according to whether you need a summary or a distribution.
| Mode | What it shows | Useful when |
|---|---|---|
top |
A summary of periodic tracer output | You want to watch the current summary while the tracer runs. |
hist |
A histogram of sample occurrences | You want to examine a distribution over a chosen period or duration. |
If you omit the mode, the command defaults to top. For exact syntax and options, consult the current rtla timerlat manual and the installed manual on the target system: available behavior can vary by kernel and tool version.
Rank #2
How do you capture a trace when timerlat exceeds a threshold?
In hist mode, the current manual documents options to select CPUs and duration, set the sampling period and histogram resolution, stop tracing at IRQ or thread thresholds, and request stack and trace capture. Set thresholds to match the latency event you are investigating; the documentation does not establish a universally correct threshold.
The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →-p: set the timer period.-i: set the IRQ latency stop threshold.-T: set the timerlat thread latency stop threshold.-s: request stack capture.-t: save trace output.-c: select CPUs.-d: set the duration.-b: set the histogram bucket size.-E: set the number of histogram entries.
For example, the manual shows a ten-minute histogram on CPUs 0–4. That is an illustration of command use, not a recommendation for every system or a benchmark result. Choose the CPU set, period, duration, thresholds, and histogram settings for the workload and question at hand.
A practical investigation sequence
- Choose CPUs and a period: Select the processors relevant to the workload and a timer period that suits the observation. Use
-cand-pwhere supported. - Pick the view: Run
rtla timerlat topfor a summary orrtla timerlat histfor a histogram. - Set stopping thresholds: Use
-iand/or-Tto stop tracing when the IRQ or thread latency reaches the threshold relevant to the investigation. - Preserve diagnostic context: Request stack capture with
-sand trace output with-tso a threshold event can leave material to inspect. - Bound or shape histogram collection: For
hist, use-d,-b, and-Eas needed. Confirm the exact option syntax with the installed manual before relying on it.
What can osnoise and tracepoints tell you?
Timerlat is built on the osnoise tracer. The kernel tracing documentation describes osnoise tracepoints that can provide context about operating-system noise around a latency event. They can help establish what was happening near an outlier, but a tracepoint is evidence to interpret—not an automatic diagnosis. Review the osnoise and timerlat tracer documentation alongside the saved trace.
How does rtla collect results, and what should you verify?
The current command manual documents synchronous in-kernel processing through a dedicated BPF program attached to the timerlat tracer. If BPF support or the osnoise:timerlat_sample tracepoint is unavailable, rtla falls back to asynchronous user-space processing through tracefs. This is documented behavior, not a promise that every distribution build has identical capabilities. Check the target system’s installed manual and available kernel tracing support before comparing runs.
The tool reports latency for a specific configuration and run. The cited kernel documentation does not establish a typical latency statistic or a universal acceptable target, so do not treat an example histogram or one machine’s result as a general Linux performance claim.
When would an application use the timerlat user-space interface?
The kernel tracer also documents a per-CPU timerlat_fd interface for direct user-space instrumentation. It is separate from ordinary use of rtla timerlat. The documented requirements are:
- Enable the timerlat tracer.
- Set the osnoise workload option to
NO_OSNOISE_WORKLOAD. - Pin the user-space thread to one processor.
- Allow only one thread at a time to access that CPU’s file.
Consult the kernel tracer documentation before building an application around this interface.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
