v12.12
Benchmarking
Initializing enviroment...
Loading IL program
Found RV610 device at 500 MHz (2 SIMDs, wavefront size=64)
28 MB of cached, 4 MB uncached RAM available
Compiling...
Linking...
Allocating LOCAL buffers
Program info:
Scratch regs needed: 0
Number of shared GPRs: 0
Number of shared GPRs total: 0
Slow mode: no
Number of wavefronts per SIMD: 0
Is max wavefronts per SIMD?: no
---Benchmarking core, peak size (no readback)---
Using optimal size (8x16)
Iters: 1024, time=125 ms, 8192 iters/sec, 4 Mkeys/sec
Iters: 2048, time=250 ms, 8192 iters/sec, 4 Mkeys/sec
Iters: 4096, time=515 ms, 7953 iters/sec, 4 Mkeys/sec
Using optimal size (16x8)
Iters: 1024, time=125 ms, 8192 iters/sec, 4 Mkeys/sec
Iters: 2048, time=250 ms, 8192 iters/sec, 4 Mkeys/sec
Iters: 4096, time=500 ms, 8192 iters/sec, 4 Mkeys/sec
Iters: 8192, time=1016 ms, 8062 iters/sec, 4 Mkeys/sec
---Trying grid (24x24)---
Iters: 256, time=109 ms, 2348 iters/sec, 5 Mkeys/sec
Iters: 512, time=219 ms, 2337 iters/sec, 5 Mkeys/sec
Iters: 1024, time=437 ms, 2343 iters/sec, 5 Mkeys/sec
Iters: 2048, time=875 ms, 2340 iters/sec, 5 Mkeys/sec
---Trying grid (32x32)---
Iters: 256, time=187 ms, 1368 iters/sec, 5 Mkeys/sec
Iters: 512, time=375 ms, 1365 iters/sec, 5 Mkeys/sec
Iters: 1024, time=766 ms, 1336 iters/sec, 5 Mkeys/sec
---Trying grid (40x40)---
Iters: 256, time=281 ms, 911 iters/sec, 5 Mkeys/sec
Iters: 512, time=563 ms, 909 iters/sec, 5 Mkeys/sec
---Trying grid (48x48)---
Iters: 256, time=421 ms, 608 iters/sec, 5 Mkeys/sec
Iters: 512, time=813 ms, 629 iters/sec, 5 Mkeys/sec
---Trying grid (56x56)---
Iters: 256, time=578 ms, 442 iters/sec, 5 Mkeys/sec
---Trying grid (64x64)---
Iters: 256, time=734 ms, 348 iters/sec, 5 Mkeys/sec
---Trying grid (72x72)---
Iters: 256, time=938 ms, 272 iters/sec, 5 Mkeys/sec
---Trying grid (80x80)---
Iters: 256, time=1156 ms, 221 iters/sec, 5 Mkeys/sec
****Calculating readback speed*****
Using optimal size (8x16)
Iters: 2048, time=1984 ms, 1032 iters/sec
Using optimal size (16x8)
Iters: 2048, time=1985 ms, 1031 iters/sec
---Trying grid (24x24)---
Iters: 1024, time=1016 ms, 1007 iters/sec
---Trying grid (32x32)---
Iters: 1024, time=1063 ms, 963 iters/sec
---Trying grid (40x40)---
Iters: 1024, time=1125 ms, 910 iters/sec
---Trying grid (48x48)---
Iters: 1024, time=1203 ms, 851 iters/sec
---Trying grid (56x56)---
Iters: 1024, time=1250 ms, 819 iters/sec
---Trying grid (64x64)---
Iters: 1024, time=1328 ms, 771 iters/sec
---Trying grid (72x72)---
Iters: 1024, time=1407 ms, 727 iters/sec
---Trying grid (80x80)---
Iters: 1024, time=1500 ms, 682 iters/sec
****Benchmarking full cycle (1b4******
Using optimal size (8x16)
Iters: 1024, time=125 ms, 8192 iters/sec, 4 Mkeys/sec
Iters: 2048, time=250 ms, 8192 iters/sec, 4 Mkeys/sec
Iters: 4096, time=515 ms, 7953 iters/sec, 4 Mkeys/sec
Using optimal size (16x8)
Iters: 1024, time=125 ms, 8192 iters/sec, 4 Mkeys/sec
Iters: 2048, time=250 ms, 8192 iters/sec, 4 Mkeys/sec
Iters: 4096, time=516 ms, 7937 iters/sec, 4 Mkeys/sec
---Trying grid (24x24)---
Iters: 256, time=109 ms, 2348 iters/sec, 5 Mkeys/sec
Iters: 512, time=219 ms, 2337 iters/sec, 5 Mkeys/sec
Iters: 1024, time=453 ms, 2260 iters/sec, 5 Mkeys/sec
Iters: 2048, time=875 ms, 2340 iters/sec, 5 Mkeys/sec
---Trying grid (32x32)---
Iters: 256, time=203 ms, 1261 iters/sec, 5 Mkeys/sec
Iters: 512, time=375 ms, 1365 iters/sec, 5 Mkeys/sec
Iters: 1024, time=766 ms, 1336 iters/sec, 5 Mkeys/sec
---Trying grid (40x40)---
Iters: 256, time=281 ms, 911 iters/sec, 5 Mkeys/sec
Iters: 512, time=578 ms, 885 iters/sec, 5 Mkeys/sec
---Trying grid (48x48)---
Iters: 256, time=406 ms, 630 iters/sec, 5 Mkeys/sec
Iters: 512, time=828 ms, 618 iters/sec, 5 Mkeys/sec
---Trying grid (56x56)---
Iters: 256, time=578 ms, 442 iters/sec, 5 Mkeys/sec
---Trying grid (64x64)---
Iters: 256, time=750 ms, 341 iters/sec, 5 Mkeys/sec
---Trying grid (72x72)---
Iters: 256, time=938 ms, 272 iters/sec, 5 Mkeys/sec
---Trying grid (80x80)---
Iters: 256, time=1156 ms, 221 iters/sec, 5 Mkeys/sec
Deallocating resources
|