Skip to content

Instantly share code, notes, and snippets.

@bzm3r
Created February 22, 2020 03:58
Show Gist options
  • Save bzm3r/6e65bc1c2d642586bd6abfbca919a017 to your computer and use it in GitHub Desktop.
Save bzm3r/6e65bc1c2d642586bd6abfbca919a017 to your computer and use it in GitHub Desktop.
compiling kernel transpose-shuffle-WGS=(64,1)...
num bms: 4096, num dispatch groups: 2048
GPU results verified!
task name:Vk-ShuffleAMD-WG=64
device: Radeon RX 570 Series
num BMs: 4096, TG size: 64
CPU loops: 1001, GPU loops: 5001
timestamp stats (N = 1001): 8.53 +/- 0.54 ms
instant stats (N = 1001): 9.13 +/- 0.59 ms
compiling kernel transpose-shuffle-WGS=(128,1)...
num bms: 4096, num dispatch groups: 1024
GPU results verified!
task name:Vk-ShuffleAMD-WG=128
device: Radeon RX 570 Series
num BMs: 4096, TG size: 128
CPU loops: 1001, GPU loops: 5001
timestamp stats (N = 1001): 8.66 +/- 0.23 ms
instant stats (N = 1001): 9.23 +/- 0.27 ms
compiling kernel transpose-shuffle-WGS=(192,1)...
num bms: 4096, num dispatch groups: 683
GPU results verified!
task name:Vk-ShuffleAMD-WG=192
device: Radeon RX 570 Series
num BMs: 4096, TG size: 192
CPU loops: 1001, GPU loops: 5001
timestamp stats (N = 1001): 9.01 +/- 0.00 ms
instant stats (N = 1001): 9.56 +/- 0.15 ms
compiling kernel transpose-shuffle-WGS=(256,1)...
num bms: 4096, num dispatch groups: 512
GPU results verified!
task name:Vk-ShuffleAMD-WG=256
device: Radeon RX 570 Series
num BMs: 4096, TG size: 256
CPU loops: 1001, GPU loops: 5001
timestamp stats (N = 1001): 8.51 +/- 0.00 ms
instant stats (N = 1001): 9.09 +/- 0.15 ms
compiling kernel transpose-shuffle-WGS=(320,1)...
num bms: 4096, num dispatch groups: 410
GPU results verified!
task name:Vk-ShuffleAMD-WG=320
device: Radeon RX 570 Series
num BMs: 4096, TG size: 320
CPU loops: 1001, GPU loops: 5001
timestamp stats (N = 1001): 9.54 +/- 0.15 ms
instant stats (N = 1001): 10.13 +/- 0.22 ms
compiling kernel transpose-shuffle-WGS=(384,1)...
num bms: 4096, num dispatch groups: 342
GPU results verified!
task name:Vk-ShuffleAMD-WG=384
device: Radeon RX 570 Series
num BMs: 4096, TG size: 384
CPU loops: 1001, GPU loops: 5001
timestamp stats (N = 1001): 9.77 +/- 0.25 ms
instant stats (N = 1001): 10.41 +/- 0.42 ms
compiling kernel transpose-shuffle-WGS=(448,1)...
num bms: 4096, num dispatch groups: 293
GPU results verified!
task name:Vk-ShuffleAMD-WG=448
device: Radeon RX 570 Series
num BMs: 4096, TG size: 448
CPU loops: 1001, GPU loops: 5001
timestamp stats (N = 1001): 9.54 +/- 0.13 ms
instant stats (N = 1001): 10.15 +/- 0.47 ms
compiling kernel transpose-shuffle-WGS=(512,1)...
num bms: 4096, num dispatch groups: 256
GPU results verified!
task name:Vk-ShuffleAMD-WG=512
device: Radeon RX 570 Series
num BMs: 4096, TG size: 512
CPU loops: 1001, GPU loops: 5001
timestamp stats (N = 1001): 8.51 +/- 0.01 ms
instant stats (N = 1001): 9.13 +/- 0.13 ms
compiling kernel transpose-shuffle-WGS=(576,1)...
num bms: 4096, num dispatch groups: 228
GPU results verified!
task name:Vk-ShuffleAMD-WG=576
device: Radeon RX 570 Series
num BMs: 4096, TG size: 576
CPU loops: 1001, GPU loops: 5001
timestamp stats (N = 1001): 9.99 +/- 0.10 ms
instant stats (N = 1001): 10.50 +/- 0.30 ms
compiling kernel transpose-shuffle-WGS=(640,1)...
num bms: 4096, num dispatch groups: 205
GPU results verified!
task name:Vk-ShuffleAMD-WG=640
device: Radeon RX 570 Series
num BMs: 4096, TG size: 640
CPU loops: 1001, GPU loops: 5001
timestamp stats (N = 1001): 10.02 +/- 0.13 ms
instant stats (N = 1001): 10.53 +/- 0.19 ms
compiling kernel transpose-shuffle-WGS=(704,1)...
num bms: 4096, num dispatch groups: 187
GPU results verified!
task name:Vk-ShuffleAMD-WG=704
device: Radeon RX 570 Series
num BMs: 4096, TG size: 704
CPU loops: 1001, GPU loops: 5001
timestamp stats (N = 1001): 9.50 +/- 0.04 ms
instant stats (N = 1001): 10.05 +/- 0.15 ms
compiling kernel transpose-shuffle-WGS=(768,1)...
num bms: 4096, num dispatch groups: 171
GPU results verified!
task name:Vk-ShuffleAMD-WG=768
device: Radeon RX 570 Series
num BMs: 4096, TG size: 768
CPU loops: 1001, GPU loops: 5001
timestamp stats (N = 1001): 9.51 +/- 0.01 ms
instant stats (N = 1001): 10.02 +/- 0.14 ms
compiling kernel transpose-shuffle-WGS=(832,1)...
num bms: 4096, num dispatch groups: 158
GPU results verified!
task name:Vk-ShuffleAMD-WG=832
device: Radeon RX 570 Series
num BMs: 4096, TG size: 832
CPU loops: 1001, GPU loops: 5001
timestamp stats (N = 1001): 9.36 +/- 0.22 ms
instant stats (N = 1001): 9.99 +/- 0.26 ms
compiling kernel transpose-shuffle-WGS=(896,1)...
num bms: 4096, num dispatch groups: 147
GPU results verified!
task name:Vk-ShuffleAMD-WG=896
device: Radeon RX 570 Series
num BMs: 4096, TG size: 896
CPU loops: 1001, GPU loops: 5001
timestamp stats (N = 1001): 10.02 +/- 0.10 ms
instant stats (N = 1001): 10.60 +/- 0.19 ms
compiling kernel transpose-shuffle-WGS=(960,1)...
num bms: 4096, num dispatch groups: 137
GPU results verified!
task name:Vk-ShuffleAMD-WG=960
device: Radeon RX 570 Series
num BMs: 4096, TG size: 960
CPU loops: 1001, GPU loops: 5001
timestamp stats (N = 1001): 10.39 +/- 0.21 ms
instant stats (N = 1001): 10.88 +/- 0.25 ms
compiling kernel transpose-shuffle-WGS=(1024,1)...
num bms: 4096, num dispatch groups: 128
GPU results verified!
task name:Vk-ShuffleAMD-WG=1024
device: Radeon RX 570 Series
num BMs: 4096, TG size: 1024
CPU loops: 1001, GPU loops: 5001
timestamp stats (N = 1001): 8.51 +/- 0.00 ms
instant stats (N = 1001): 9.09 +/- 0.16 ms
compiling kernel transpose-shuffle-WGS=(1088,1)...
thread 'main' panicked at 'called `Result::unwrap()` on an `Err` value: CompilationError(1, "transpose-shuffle-WGS=(1088,1).glsl:8: error: \'local_size\' : too large; see gl_MaxComputeWorkGroupSize\n")', src\task.rs:99:32
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
error: process didn't exit successfully: `target\release\transpose-timing-tests.exe` (exit code: 101)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment