Benchmark measurement details

Historical measurement. Not a measurement of the latest release.

View the results · Run benchmarks

Measurement identity	Value
Source revision	e62f7eedfa2d1dd27e02730df129be31123e6d31
Measured at	2026-09-21T09:20:19.835397+00:00
Result status	completed
Source document SHA-256	29a46b798a21c1354a867ac4195929158d067973f7a62e4bd3a0d03795f41a57
Policy	collecting_baseline
platform	macOS-14.8.9-arm64-arm-64bit
machine	arm64
cpu_model	arm
runner_image	macos14
rustc_version	rustc 1.96.1 (31fca3adb 2026-06-26)
c_compiler_version	Apple clang version 15.0.0 (clang-1500.3.9.4)

Download the public measurement data. This is a presentation snapshot, not the full source receipt. It preserves the original report hash and numerical observations; hostnames, local paths, and internal traces are omitted.

• timing_only rows do not establish output parity. Values apply only to this operation matrix and host.
• No reviewed regression thresholds are active while the policy is collecting_baseline.

Full measurements

Workload	Subject	Median µs	P90 µs	P95 µs	Samples	Result / correctness	Requested → actual	Terminal receipt
dejavusans_load_20	fontdone	10.414	18.214	Not measured	10	completed; timing_only	native → native	Not applicable
dejavusans_load_20	FreeType	17.440	21.350	Not measured	10	completed; timing_only	native → native	Not applicable
dejavusans_getname_20	fontdone	0.012	0.017	Not measured	10	completed; timing_only	native → native	Not applicable
dejavusans_getname_20	FreeType	0.028	0.044	Not measured	10	completed; timing_only	native → native	Not applicable
dejavusans_getmetrics_20	fontdone	0.005	0.007	Not measured	10	completed; timing_only	native → native	Not applicable
dejavusans_getmetrics_20	FreeType	0.021	0.042	Not measured	10	completed; timing_only	native → native	Not applicable
dejavusans_getlength_av_20	fontdone	6.439	7.884	Not measured	10	completed; timing_only	native → native	Not applicable
dejavusans_getlength_av_20	FreeType	2.204	2.798	Not measured	10	completed; timing_only	native → native	Not applicable
dejavusans_getbbox_hello_20	fontdone	2.166	2.393	Not measured	10	completed; timing_only	native → native	Not applicable
dejavusans_getbbox_hello_20	FreeType	5.340	6.045	Not measured	10	completed; timing_only	native → native	Not applicable
dejavusans_getmask_a_20	fontdone	3.977	4.499	Not measured	10	completed; timing_only	native → native	Not applicable
dejavusans_getmask_a_20	FreeType	1.964	2.412	Not measured	10	completed; timing_only	native → native	Not applicable
dejavusans_getmask_force_autohint_a_20	fontdone	98.785	109.243	Not measured	10	completed; timing_only	native → native	Not applicable
dejavusans_getmask_force_autohint_a_20	FreeType	1.480	1.946	Not measured	10	completed; timing_only	native → native	Not applicable
dejavusans_glyph_metrics_a_20	fontdone	3.978	4.597	Not measured	10	completed; timing_only	native → native	Not applicable
dejavusans_glyph_metrics_a_20	FreeType	2.219	2.683	Not measured	10	completed; timing_only	native → native	Not applicable
dejavusans_render_mono_a_20	fontdone	6.041	7.900	Not measured	10	completed; timing_only	native → native	Not applicable
dejavusans_render_mono_a_20	FreeType	1.866	3.084	Not measured	10	completed; timing_only	native → native	Not applicable
dejavusans_render_lcd_a_20	fontdone	9.502	11.091	Not measured	10	completed; timing_only	native → native	Not applicable
dejavusans_render_lcd_a_20	FreeType	2.026	2.557	Not measured	10	completed; timing_only	native → native	Not applicable
generated_cjk_getmask_hani_20	fontdone	3.441	6.228	Not measured	10	completed; timing_only	native → native	Not applicable
generated_cjk_getmask_hani_20	FreeType	1.015	1.986	Not measured	10	completed; timing_only	native → native	Not applicable

Workload boundaries

Repeat counts, dimensions, modes, and cache states are retained per workload:

dejavusans_load_20

Sample unit: benchmark sample.

{
  "context": {},
  "measurement_policy": {
    "profile": "default",
    "boundary": "public operation; process memory is separate"
  }
}

dejavusans_getname_20

Sample unit: benchmark sample.

{
  "context": {},
  "measurement_policy": {
    "profile": "default",
    "boundary": "public operation; process memory is separate"
  }
}

dejavusans_getmetrics_20

Sample unit: benchmark sample.

{
  "context": {},
  "measurement_policy": {
    "profile": "default",
    "boundary": "public operation; process memory is separate"
  }
}

dejavusans_getlength_av_20

Sample unit: benchmark sample.

{
  "context": {},
  "measurement_policy": {
    "profile": "default",
    "boundary": "public operation; process memory is separate"
  }
}

dejavusans_getbbox_hello_20

Sample unit: benchmark sample.

{
  "context": {},
  "measurement_policy": {
    "profile": "default",
    "boundary": "public operation; process memory is separate"
  }
}

dejavusans_getmask_a_20

Sample unit: benchmark sample.

{
  "context": {},
  "measurement_policy": {
    "profile": "default",
    "boundary": "public operation; process memory is separate"
  }
}

dejavusans_getmask_force_autohint_a_20

Sample unit: benchmark sample.

{
  "context": {},
  "measurement_policy": {
    "profile": "default",
    "boundary": "public operation; process memory is separate"
  }
}

dejavusans_glyph_metrics_a_20

Sample unit: benchmark sample.

{
  "context": {},
  "measurement_policy": {
    "profile": "default",
    "boundary": "public operation; process memory is separate"
  }
}

dejavusans_render_mono_a_20

Sample unit: benchmark sample.

{
  "context": {},
  "measurement_policy": {
    "profile": "default",
    "boundary": "public operation; process memory is separate"
  }
}

dejavusans_render_lcd_a_20

Sample unit: benchmark sample.

{
  "context": {},
  "measurement_policy": {
    "profile": "default",
    "boundary": "public operation; process memory is separate"
  }
}

generated_cjk_getmask_hani_20

Sample unit: benchmark sample.

{
  "context": {},
  "measurement_policy": {
    "profile": "default",
    "boundary": "public operation; process memory is separate"
  }
}

Reproduce and interpret

Use the benchmark protocol. Correctness, native dispatch, timing regressions, process memory, and artifact size are separate claims.