Device=NVIDIA GeForce RTX 5090 Laptop GPU, CC=12.0, SMs=82, driver=13010, runtime=12080
L2_MiB=64.0,host_memory=pageable,batches=5,warmup_calls=3,CPU=single_thread_optimized
Times exclude allocation, input generation and validation; event intervals may include submission gaps.
HEADER,N,threads,blocks,repeats,cpu_min_ms,cpu_median_ms,cpu_max_ms,kernel_min_ms,kernel_median_ms,kernel_max_ms,e2e_min_ms,e2e_median_ms,e2e_max_ms,logical_GBps,cpu_over_kernel,cpu_over_e2e,working_set_MiB,max_error
N=0: empty output; launch skipped; PASS
N=1,threads=128,blocks=1,max_error=0,PASS
N=31,threads=128,blocks=1,max_error=0,PASS
N=32,threads=128,blocks=1,max_error=0,PASS
N=33,threads=128,blocks=1,max_error=0,PASS
N=127,threads=128,blocks=1,max_error=0,PASS
N=128,threads=128,blocks=1,max_error=0,PASS
N=129,threads=128,blocks=2,max_error=0,PASS
N=255,threads=128,blocks=2,max_error=0,PASS
N=256,threads=128,blocks=2,max_error=0,PASS
N=257,threads=128,blocks=3,max_error=0,PASS
N=511,threads=128,blocks=4,max_error=0,PASS
N=512,threads=128,blocks=4,max_error=0,PASS
N=513,threads=128,blocks=5,max_error=0,PASS
N=1003,threads=128,blocks=8,max_error=0,PASS
N=0: empty output; launch skipped; PASS
N=1,threads=256,blocks=1,max_error=0,PASS
N=31,threads=256,blocks=1,max_error=0,PASS
N=32,threads=256,blocks=1,max_error=0,PASS
N=33,threads=256,blocks=1,max_error=0,PASS
N=127,threads=256,blocks=1,max_error=0,PASS
N=128,threads=256,blocks=1,max_error=0,PASS
N=129,threads=256,blocks=1,max_error=0,PASS
N=255,threads=256,blocks=1,max_error=0,PASS
N=256,threads=256,blocks=1,max_error=0,PASS
N=257,threads=256,blocks=2,max_error=0,PASS
N=511,threads=256,blocks=2,max_error=0,PASS
N=512,threads=256,blocks=2,max_error=0,PASS
N=513,threads=256,blocks=3,max_error=0,PASS
N=1003,threads=256,blocks=4,max_error=0,PASS
N=0: empty output; launch skipped; PASS
N=1,threads=512,blocks=1,max_error=0,PASS
N=31,threads=512,blocks=1,max_error=0,PASS
N=32,threads=512,blocks=1,max_error=0,PASS
N=33,threads=512,blocks=1,max_error=0,PASS
N=127,threads=512,blocks=1,max_error=0,PASS
N=128,threads=512,blocks=1,max_error=0,PASS
N=129,threads=512,blocks=1,max_error=0,PASS
N=255,threads=512,blocks=1,max_error=0,PASS
N=256,threads=512,blocks=1,max_error=0,PASS
N=257,threads=512,blocks=1,max_error=0,PASS
N=511,threads=512,blocks=1,max_error=0,PASS
N=512,threads=512,blocks=1,max_error=0,PASS
N=513,threads=512,blocks=2,max_error=0,PASS
N=1003,threads=512,blocks=2,max_error=0,PASS
N=1003,threads=128,blocks=8,max_error=0,PASS
kernel_regs_per_thread=14,static_shared_bytes=0,active_blocks_per_SM=12,theoretical_occupancy=1.000
RESULT,1003,128,8,20,0.000040000,0.000040000,0.000045000,0.005219200,0.006051200,0.017140800,0.070900000,0.071175000,0.088710000,1.989,0.007,0.001,0.011,0
N=1003,threads=256,blocks=4,max_error=0,PASS
kernel_regs_per_thread=14,static_shared_bytes=0,active_blocks_per_SM=6,theoretical_occupancy=1.000
RESULT,1003,256,4,20,0.000040000,0.000045000,0.000045000,0.004148800,0.008718399,0.012083201,0.063525000,0.071285000,0.086760000,1.381,0.005,0.001,0.011,0
N=1003,threads=512,blocks=2,max_error=0,PASS
kernel_regs_per_thread=14,static_shared_bytes=0,active_blocks_per_SM=3,theoretical_occupancy=1.000
RESULT,1003,512,2,20,0.000050000,0.000050000,0.000055000,0.005284800,0.005331200,0.006670400,0.064520000,0.071350000,0.084865000,2.258,0.009,0.001,0.011,0
N=65536,threads=256,blocks=256,max_error=0,PASS
kernel_regs_per_thread=14,static_shared_bytes=0,active_blocks_per_SM=6,theoretical_occupancy=1.000
RESULT,65536,256,256,20,0.006160000,0.006280000,0.006455000,0.006368000,0.007553600,0.014649600,0.147380000,0.166510000,0.205345000,104.114,0.831,0.038,0.750,0
N=65536,threads=512,blocks=128,max_error=0,PASS
kernel_regs_per_thread=14,static_shared_bytes=0,active_blocks_per_SM=3,theoretical_occupancy=1.000
RESULT,65536,512,128,20,0.006155000,0.006205000,0.006525000,0.005385600,0.007521600,0.016267199,0.146905000,0.157780000,0.161675000,104.556,0.825,0.039,0.750,0
N=65536,threads=128,blocks=512,max_error=0,PASS
kernel_regs_per_thread=14,static_shared_bytes=0,active_blocks_per_SM=12,theoretical_occupancy=1.000
RESULT,65536,128,512,20,0.006160000,0.006170000,0.020220000,0.006774400,0.007180800,0.007540800,0.156195000,0.181750000,0.268515000,109.519,0.859,0.034,0.750,0
N=1048576,threads=512,blocks=2048,max_error=0,PASS
kernel_regs_per_thread=14,static_shared_bytes=0,active_blocks_per_SM=3,theoretical_occupancy=1.000
RESULT,1048576,512,2048,20,0.246630000,0.253170000,0.272555000,0.008276800,0.008824000,0.016588800,0.982605000,0.999685000,1.010750000,1425.987,28.691,0.253,12.000,0
N=1048576,threads=128,blocks=2624,max_error=0,PASS
kernel_regs_per_thread=14,static_shared_bytes=0,active_blocks_per_SM=12,theoretical_occupancy=1.000
RESULT,1048576,128,2624,20,0.246885000,0.255795000,0.266955000,0.007364800,0.007976000,0.010032000,0.948420000,0.957045000,1.039365000,1577.597,32.071,0.267,12.000,0
N=1048576,threads=256,blocks=2624,max_error=0,PASS
kernel_regs_per_thread=14,static_shared_bytes=0,active_blocks_per_SM=6,theoretical_occupancy=1.000
RESULT,1048576,256,2624,20,0.267120000,0.267235000,0.267985000,0.007897600,0.008910400,0.009384001,0.884280000,0.920800000,0.983165000,1412.160,29.991,0.290,12.000,0
N=4194304,threads=128,blocks=2624,max_error=0,PASS
kernel_regs_per_thread=14,static_shared_bytes=0,active_blocks_per_SM=12,theoretical_occupancy=1.000
RESULT,4194304,128,2624,20,1.419630000,1.446090000,1.469700000,0.013312000,0.024516800,0.026072001,4.182455000,4.255225000,4.336030000,2052.945,58.984,0.340,48.000,0
N=4194304,threads=256,blocks=2624,max_error=0,PASS
kernel_regs_per_thread=14,static_shared_bytes=0,active_blocks_per_SM=6,theoretical_occupancy=1.000
RESULT,4194304,256,2624,20,1.441000000,1.456290000,1.491155000,0.013577600,0.016201600,0.016931200,3.864660000,3.932075000,4.190460000,3106.585,89.886,0.370,48.000,0
N=4194304,threads=512,blocks=2624,max_error=0,PASS
kernel_regs_per_thread=14,static_shared_bytes=0,active_blocks_per_SM=3,theoretical_occupancy=1.000
RESULT,4194304,512,2624,20,1.409475000,1.462575000,1.501880000,0.014108799,0.015785601,0.018033599,4.125555000,4.492280000,4.614530000,3188.453,92.652,0.326,48.000,0
N=16777216,threads=256,blocks=2624,max_error=0,PASS
kernel_regs_per_thread=14,static_shared_bytes=0,active_blocks_per_SM=6,theoretical_occupancy=1.000
RESULT,16777216,256,2624,20,7.084990000,7.286815000,7.646960000,0.254166394,0.254499197,0.269208014,16.153880000,17.357860000,17.836280000,791.070,28.632,0.420,192.000,0
N=16777216,threads=512,blocks=2624,max_error=0,PASS
kernel_regs_per_thread=14,static_shared_bytes=0,active_blocks_per_SM=3,theoretical_occupancy=1.000
RESULT,16777216,512,2624,20,7.008965000,7.146635000,7.299500000,0.398871988,0.400236785,0.413740784,16.535785000,17.174485000,18.165700000,503.019,17.856,0.416,192.000,0
N=16777216,threads=128,blocks=2624,max_error=0,PASS
kernel_regs_per_thread=14,static_shared_bytes=0,active_blocks_per_SM=12,theoretical_occupancy=1.000
RESULT,16777216,128,2624,20,7.606050000,7.760900000,7.830505000,0.401273578,0.404896021,0.432051182,17.415900000,17.889200000,20.493345000,497.230,19.168,0.434,192.000,0