Here it passes with a single GPU
Code: Select all
dunnea@esx3:~/magma-1.6.0/testing$ ./testing_zgetrf -c2 --ngpu 1 --verbose
MAGMA 1.6.0 compiled for CUDA capability >= 3.0
CUDA runtime 6050, driver 6050. OpenMP threads 32.
device 0: Tesla K40m, 745.0 MHz clock, 11519.6 MB memory, capability 3.5
device 1: Tesla K40m, 745.0 MHz clock, 11519.6 MB memory, capability 3.5
device 2: Tesla K40m, 745.0 MHz clock, 11519.6 MB memory, capability 3.5
device 3: Tesla K40m, 745.0 MHz clock, 11519.6 MB memory, capability 3.5
Usage: ./testing_zgetrf [options] [-h|--help]
ngpu 1
M N CPU GFlop/s (sec) GPU GFlop/s (sec) |Ax-b|/(N*|A|*|x|)
=========================================================================
1088 1088 --- ( --- ) 79.54 ( 0.04) 5.00e-19 ok
2112 2112 --- ( --- ) 158.07 ( 0.16) 4.63e-19 ok
3136 3136 --- ( --- ) 238.51 ( 0.34) 4.14e-19 ok
4160 4160 --- ( --- ) 148.58 ( 1.29) 3.80e-19 ok
5184 5184 --- ( --- ) 185.29 ( 2.00) 3.71e-19 ok
6208 6208 --- ( --- ) 230.90 ( 2.76) 3.73e-19 ok
7232 7232 --- ( --- ) 259.02 ( 3.89) 3.49e-19 ok
8256 8256 --- ( --- ) 175.63 ( 8.54) 3.52e-19 ok
9280 9280 --- ( --- ) 195.91 ( 10.88) 3.27e-19 ok
10304 10304 --- ( --- ) 218.88 ( 13.33) 3.32e-19 ok
Code: Select all
dunnea@esx3:~/magma-1.6.0/testing$ ./testing_zgetrf -c2 --ngpu 4 --verbose
MAGMA 1.6.0 compiled for CUDA capability >= 3.0
CUDA runtime 6050, driver 6050. OpenMP threads 32.
device 0: Tesla K40m, 745.0 MHz clock, 11519.6 MB memory, capability 3.5
device 1: Tesla K40m, 745.0 MHz clock, 11519.6 MB memory, capability 3.5
device 2: Tesla K40m, 745.0 MHz clock, 11519.6 MB memory, capability 3.5
device 3: Tesla K40m, 745.0 MHz clock, 11519.6 MB memory, capability 3.5
Usage: ./testing_zgetrf [options] [-h|--help]
ngpu 4
M N CPU GFlop/s (sec) GPU GFlop/s (sec) |Ax-b|/(N*|A|*|x|)
=========================================================================
On entry to magmablas_zlaswp_q, parameter 4 had an illegal value (info = -4)
...
On entry to magmablas_zlaswp_q, parameter 4 had an illegal value (info = -4)
1088 1088 --- ( --- ) 4.24 ( 0.81) 1.01e-05 failed
...
Code: Select all
...
On entry to magmablas_zlaswp_q, parameter 4 had an illegal value (info = -4)
4160 4160 --- ( --- ) 150.77 ( 1.27) 1.40e-06 failed
On entry to magmablas_zlaswp_q, parameter 5 had an illegal value (info = -5)
On entry to magmablas_zlaswp_q, parameter 5 had an illegal value (info = -5)
On entry to magmablas_zlaswp_q, parameter 5 had an illegal value (info = -5)
On entry to magmablas_zlaswp_q, parameter 5 had an illegal value (info = -5)
On entry to magmablas_zlaswp_q, parameter 4 had an illegal value (info = -4)
...
On entry to magmablas_zlaswp_q, parameter 4 had an illegal value (info = -4)
5184 5184 --- ( --- ) 194.03 ( 1.91) 1.02e-06 failed
...
Thanks!