Skip to content

mingw.md

Latest commit

 

History

History
165 lines (120 loc) · 7.55 KB

mingw.md

File metadata and controls

165 lines (120 loc) · 7.55 KB
Raw

MinGW report

This file is generated by make render from reports/mingw/*.json. Do not edit by hand - edits will be overwritten on the next render.

How to read this. These numbers measure function-pointer loading overhead and Vulkan context setup across different API loaders, comparing the stock Vulkan loader against the patched one (see the README for what the patch does). Lower is better.

Do not compare across hosts. Hardware, drivers, OS, and compiler flags all differ between the Linux, MinGW, and macOS reports. The meaningful comparisons are within a single report: loader vs. loader and patched vs. unpatched on the same hardware.

At a glance

Lower is better. Winner = best patched average on this host for each task.

Task Winner Patched avg Unpatched avg Patch speedup
Load instance functions Gloam (enabled-list) 0.97µs 5.08µs 5.2×
Load device functions Gloam (enabled-list) 4.28µs 46.21µs 10.8×
Init + load all functions Gloam (enabled-list) 5.83µs 54.20µs 9.3×
Full VK context (libvulkan persistent) Gloam (discover) 23913.73µs 39178.07µs 1.6×
Full VK context (libvulkan transient) Gloam (enabled-list) 35731.30µs 36125.40µs 1.0×

Task detail

Load instance functions

Loader Unpatched Patched Patch speedup vs. fastest
Gloam (enabled-list) 5.08µs 0.97µs 5.2× 1.0×
Volk 75.59µs 3.12µs 24.2× 3.2×
GLAD (tycho) 32.11µs 3.17µs 10.1× 3.3×
Gloam (discover) 1249.52µs 21.08µs 59.3× 21.7×
GLAD (dav1dde) 1691.24µs 106.91µs 15.8× 110×

Load instance functions (mingw)

Load device functions

Loader Unpatched Patched Patch speedup vs. fastest
Gloam (enabled-list) 46.21µs 4.28µs 10.8× 1.0×
Volk 541.21µs 10.15µs 53.3× 2.4×
GLAD (tycho) 433.67µs 11.07µs 39.2× 2.6×
Gloam (discover) 511.23µs 23.67µs 21.6× 5.5×
GLAD (dav1dde) 3634.71µs 764.79µs 4.8× 179×

Load device functions (mingw)

Init + load all functions

Loader Unpatched Patched Patch speedup vs. fastest
Gloam (enabled-list) 54.20µs 5.83µs 9.3× 1.0×
Volk 695.20µs 13.53µs 51.4× 2.3×
Gloam (discover) 3306.63µs 75.27µs 43.9× 12.9×
GLAD (tycho) 3443.70µs 78.00µs 44.1× 13.4×
GLAD (dav1dde) 6920.37µs 1063.97µs 6.5× 182×

Init + load all functions (mingw)

Full VK context (libvulkan persistent)

Loader Unpatched Patched Patch speedup vs. fastest
Gloam (discover) 39178.07µs 23913.73µs 1.6× 1.0×
Volk 36831.83µs 24580.90µs 1.5× 1.0×
Gloam (enabled-list) 36001.67µs 25128.03µs 1.4× 1.1×
GLAD (tycho) 38912.97µs 25175.33µs 1.5× 1.1×
GLAD (dav1dde) 43235.43µs 26652.23µs 1.6× 1.1×

Full VK context (libvulkan persistent) (mingw)

Full VK context (libvulkan transient)

Loader Unpatched Patched Patch speedup vs. fastest
Gloam (enabled-list) 36125.40µs 35731.30µs 1.0× 1.0×
Gloam (discover) 39094.23µs 36321.73µs 1.1× 1.0×
Volk 37866.63µs 37177.77µs 1.0× 1.0×
GLAD (tycho) 39154.10µs 37178.27µs 1.1× 1.0×
GLAD (dav1dde) 44330.80µs 38640.37µs 1.1× 1.1×

Full VK context (libvulkan transient) (mingw)

Binary sizes

All sizes in bytes. Sorted by stripped binary size. Section values come from size; Mach-O binaries report BSS as zero because the Mach-O segment model folds zero-init into __DATA.

Loader Loader .o Binary text data bss
GLAD (tycho) 53,072 67,072 17,366 47,404 0
Gloam (discover) 53,163 70,144 23,286 44,820 0
Gloam (enabled-list) 53,163 71,168 24,326 44,916 0
Volk 229,029 108,032 75,206 31,224 0
GLAD (dav1dde) 192,597 116,736 73,798 40,588 0
Test host details

Host

Field Value
OS MINGW64_NT-10.0-26200 3.6.10-8fbd9808.x86_64
Architecture x86_64
CPU AMD RYZEN AI MAX+ PRO 395 w/ Radeon 8060S

Toolchain

Field Value
CC clang (22.1.8)
CXX clang++ (22.1.8)
OPTFLAGS -O2 -fno-unroll-loops -march=x86-64-v2 -mtune=znver3 -g0
CFLAGS -std=c17
CXXFLAGS -std=c++20

Project versions

Project Version
GLAD (dav1dde) 2.0.8-10-ge3c14f9
GLAD (tycho) 2.0.8-92-g1002eae
gloam 0.7.0
Volk 1.4.357.0-7-g2e19a77
xxHash 0.7.4-1023-gc0b5ea9
Vulkan-Headers 1.4.360

vulkaninfo

Devices:
========
GPU0:
	apiVersion         = 1.4.349
	driverVersion      = 2.0.395
	vendorID           = 0x1002
	deviceID           = 0x1586
	deviceType         = PHYSICAL_DEVICE_TYPE_INTEGRATED_GPU
	deviceName         = AMD Radeon(TM) 8060S Graphics
	driverID           = DRIVER_ID_AMD_PROPRIETARY
	driverName         = AMD proprietary driver
	driverInfo         = 26.7.1 (LLPC)
	conformanceVersion = 1.4.3.3
	deviceUUID         = 00000000-c300-0000-0000-000000000000
	driverUUID         = 414d442d-5749-4e2d-4452-560000000000