blog

Personal blog
git clone git://git.laack.co/blog.git
Log | Files | Refs

commit b379f344ef613be308750ed514ce30493e875220
parent e36b236a4a8057302158fc395c8cf9988507cc72
Author: Andrew Laack <andrew@laack.co>
Date:   Sun, 11 Oct 2026 16:19:17 -0500

Links

Diffstat:
Mposts/site_wip/basic-software-profiling-cpu-usage.html | 33++++++++++++++++++---------------
Mposts/wip/basic-software-profiling-cpu-usage.md | 8+++-----
2 files changed, 21 insertions(+), 20 deletions(-)

diff --git a/posts/site_wip/basic-software-profiling-cpu-usage.html b/posts/site_wip/basic-software-profiling-cpu-usage.html @@ -98,7 +98,10 @@ <link rel="stylesheet" href="style.css" /> </head> <body> -<p>TODOS: - since we support http(s) we mustn’t hardcode file paths. - review the ending stuff - formatting for all the preformatted sections</p> +<p>TODOS:</p> +<ul> +<li>review the ending stuff</li> +</ul> <h1 id="basic-software-profiling-cpu-usage">Basic Software Profiling: CPU Usage</h1> <p>Collectively, we waste trillions of CPU cycles every day because software isn’t profiled frequently. I believe this is an educational problem.</p> <p>This post will explain how you can find the parts of a program hogging the most CPU time. The post will focus on compiled programs, but the core ideas can be applied to interpreted programs.</p> @@ -107,7 +110,7 @@ <p><a href="https://github.com/brendangregg/FlameGraph">Flamegraph</a> is a suite of tools for generating flamegraphs, a type of graph designed for the visualization of performance data.</p> <h2 id="reading-a-flamegraph">Reading a Flamegraph</h2> <p>Take a look at the flamegraph below and try to understand the performance characteristics of the program:</p> -<object data="https://blog.laack.co/images/ex1-fg.svg" type="image/svg+xml" width="100%"> +<object data="/images/ex1-fg.svg" type="image/svg+xml" width="100%"> </object> <details> <summary> @@ -195,18 +198,18 @@ Source code for the program that generated this flamegraph <pre><code>perf stat ./ex1</code></pre> <pre><code> Performance counter stats for &#39;./ex1&#39;: - 0 context-switches:u # 0.0 cs/sec cs_per_second - 0 cpu-migrations:u # 0.0 migrations/sec migrations_per_second - 132 page-faults:u # 4.1 faults/sec page_faults_per_second - 32,436.48 msec task-clock:u # nan CPUs CPUs_utilized - 356,775 branch-misses:u # 0.0 % branch_miss_rate (88.89%) - 63,706,083,988 branches:u # 1964.0 M/sec branch_frequency (88.89%) - 134,126,037,336 cpu-cycles:u # 4.1 GHz cycles_frequency (88.89%) - 300,951,617,557 instructions:u # 2.2 instructions insn_per_cycle (88.89%) + 0 context-switches:u # 0.0 cs/sec cs_per_second + 0 cpu-migrations:u # 0.0 migrations/sec migrations_per_second + 132 page-faults:u # 4.1 faults/sec page_faults_per_second + 32,436.48 msec task-clock:u # nan CPUs CPUs_utilized + 356,775 branch-misses:u # 0.0 % branch_miss_rate (88.89%) + 63,706,083,988 branches:u # 1964.0 M/sec branch_frequency (88.89%) + 134,126,037,336 cpu-cycles:u # 4.1 GHz cycles_frequency (88.89%) + 300,951,617,557 instructions:u # 2.2 instructions insn_per_cycle (88.89%) TopdownL1 # 4.4 % tma_backend_bound - # 61.4 % tma_bad_speculation (88.89%) - # 0.0 % tma_frontend_bound (77.78%) - # 34.2 % tma_retiring (88.89%) + # 61.4 % tma_bad_speculation (88.89%) + # 0.0 % tma_frontend_bound (77.78%) + # 34.2 % tma_retiring (88.89%) 32.440411262 seconds time elapsed @@ -320,7 +323,7 @@ Performance counter stats for &#39;amfora lf.txt&#39;: 63.894262000 seconds user 8.207151000 seconds sys</code></pre> <p>That’s a lot slower for Amfora. Let’s look at a flamegraph for this.</p> -<object data="https://blog.laack.co/images/amfora-before.svg" type="image/svg+xml" width="100%"> +<object data="/images/amfora-before.svg" type="image/svg+xml" width="100%"> </object> <p>So 68.34% samples occurred while executing the GC, that’s a bit surprising, even given how awful garbage collectors are. We also see that 25.47% of the time, most of the remaining time, is spent in the RenderGemini function. Of this, most of it is spent on concatstring3, a function under the hood that’s used to concatenate strings.</p> <p>Here’s where that’d be used in the function:</p> @@ -363,7 +366,7 @@ buf += strings.TrimSuffix(lines[i], &quot;\r&quot;) + &quot;\r\n&quot;</code></p 1.193871000 seconds user 0.039062000 seconds sys</code></pre> <p>Much better, that’s now using 32x less cycles, and is far faster than before.</p> -<object data="https://blog.laack.co/images/amfora-after.svg" type="image/svg+xml" width="100%"> +<object data="/images/amfora-after.svg" type="image/svg+xml" width="100%"> </object> <p>Looking again at the flamegraph, we see the RenderGemini function is now mostly performing word wrapping, and the GC is only utilizing the CPU during 16.48% of our samples, much better.</p> <p>We can go deeper into this if we’d like, but there are now 3 ~equally costly groupings, and as performance becomes more diffuse, optimizations will generally have a smaller amount of impact on perf, so we can leave this where it is.</p> diff --git a/posts/wip/basic-software-profiling-cpu-usage.md b/posts/wip/basic-software-profiling-cpu-usage.md @@ -1,8 +1,6 @@ TODOS: -- since we support http(s) we mustn't hardcode file paths. - review the ending stuff -- formatting for all the preformatted sections # Basic Software Profiling: CPU Usage @@ -20,7 +18,7 @@ This post will explain how you can find the parts of a program hogging the most Take a look at the flamegraph below and try to understand the performance characteristics of the program: -<object data="https://blog.laack.co/images/ex1-fg.svg" type="image/svg+xml" width="100%"></object> +<object data="/images/ex1-fg.svg" type="image/svg+xml" width="100%"></object> <details> <summary>Source code for the program that generated this flamegraph</summary> @@ -299,7 +297,7 @@ perf stat amfora lf.gmi That's a lot slower for Amfora. Let's look at a flamegraph for this. -<object data="https://blog.laack.co/images/amfora-before.svg" type="image/svg+xml" width="100%"></object> +<object data="/images/amfora-before.svg" type="image/svg+xml" width="100%"></object> So 68.34% samples occurred while executing the GC, that's a bit surprising, even given how awful garbage collectors are. We also see that 25.47% of the time, most of the remaining time, is spent in the RenderGemini function. Of this, most of it is spent on concatstring3, a function under the hood that's used to concatenate strings. @@ -355,7 +353,7 @@ perf stat amfora lf.gmi Much better, that's now using 32x less cycles, and is far faster than before. -<object data="https://blog.laack.co/images/amfora-after.svg" type="image/svg+xml" width="100%"></object> +<object data="/images/amfora-after.svg" type="image/svg+xml" width="100%"></object> Looking again at the flamegraph, we see the RenderGemini function is now mostly performing word wrapping, and the GC is only utilizing the CPU during 16.48% of our samples, much better.