Install problems s870 on openSUSE 10.3

Accelerated Computing CUDA CUDA Programming and Performance
#1

I just recently received my s870 and man is it huge and loud! I am using a quad core 2 machine with 790i sli ultra motherboard and gx2 9800 gpu.

w/ the 9800 and the s870 hardware installed, I installed the latest NVIDIA driver lisedt under the s870 while under init3 with the following command:

./NVIDIA-Linux-x86-171.06.01-pkg1.run

I then modified my GRUB menu.lst as mentioned here on this forum by adding the following lines:

uppermem 524288

vmalloc=256MB

pci=nommconf

I restarted the machine and performed the following checks with the 9800 and s870 installed:

# lspci | grep 3D

03:00.0 3D controller: nVidia Corporation Unknown device 0604 (rev a2)

0b:00.0 3D controller: nVidia Corporation Unknown device 0197 (rev a2)

0d:00.0 3D controller: nVidia Corporation Unknown device 0197 (rev a2)

18:00.0 3D controller: nVidia Corporation Unknown device 0197 (rev a2)

1a:00.0 3D controller: nVidia Corporation Unknown device 0197 (rev a2)

ls -l /dev/nv*

crw-rw---- 1 root video 195,   0 2008-07-24 09:50 /dev/nvidia0

crw-rw---- 1 root video 195,   1 2008-07-24 09:50 /dev/nvidia1

crw-rw---- 1 root video 195,   2 2008-07-24 09:50 /dev/nvidia2

crw-rw---- 1 root video 195,   3 2008-07-24 09:50 /dev/nvidia3

crw-rw---- 1 root video 195,   4 2008-07-24 09:50 /dev/nvidia4

crw-rw---- 1 root video 195,   5 2008-07-24 09:50 /dev/nvidia5

crw-rw---- 1 root video 195, 255 2008-07-24 09:50 /dev/nvidiactl

# ./deviceQuery 

NVIDIA: could not open the device file /dev/nvidia2 (Input/output error).

There is no device supporting CUDA.

Device 0: "Device Emulation (CPU)"

  Major revision number:                         9999

  Minor revision number:                         9999

  Total amount of global memory:                 4294967295 bytes

  Number of multiprocessors:                     16

  Number of cores:                               128

  Total amount of constant memory:               65536 bytes

  Total amount of shared memory per block:       16384 bytes

  Total number of registers available per block: 8192

  Warp size:                                     1

  Maximum number of threads per block:           512

  Maximum sizes of each dimension of a block:    512 x 512 x 64

  Maximum sizes of each dimension of a grid:     65535 x 65535 x 1

  Maximum memory pitch:                          262144 bytes

  Texture alignment:                             256 bytes

  Clock rate:                                    1.35 GHz

  Concurrent copy and execution:                 No

Test PASSED

Press ENTER to exit...

I then powered down the computer and disconnected the s870 to see what results I had with the 9800. I then performed the same actions as above:

# lspci | grep 3D

03:00.0 3D controller: nVidia Corporation Unknown device 0604 (rev a2)

# ls -l /dev/nv*

crw-rw---- 1 root video 195,   0 2008-07-24 09:27 /dev/nvidia0

crw-rw---- 1 root video 195,   1 2008-07-24 09:27 /dev/nvidia1

crw-rw---- 1 root video 195, 255 2008-07-24 09:27 /dev/nvidiactl

# ./deviceQuery 

There are 2 devices supporting CUDA

Device 0: "G92-450"

  Major revision number:                         1

  Minor revision number:                         1

  Total amount of global memory:                 536543232 bytes

  Number of multiprocessors:                     16

  Number of cores:                               128

  Total amount of constant memory:               65536 bytes

  Total amount of shared memory per block:       16384 bytes

  Total number of registers available per block: 8192

  Warp size:                                     32

  Maximum number of threads per block:           512

  Maximum sizes of each dimension of a block:    512 x 512 x 64

  Maximum sizes of each dimension of a grid:     65535 x 65535 x 1

  Maximum memory pitch:                          262144 bytes

  Texture alignment:                             256 bytes

  Clock rate:                                    1.51 GHz

  Concurrent copy and execution:                 Yes

Device 1: "G92-450"

  Major revision number:                         1

  Minor revision number:                         1

  Total amount of global memory:                 536608768 bytes

  Number of multiprocessors:                     16

  Number of cores:                               128

  Total amount of constant memory:               65536 bytes

  Total amount of shared memory per block:       16384 bytes

  Total number of registers available per block: 8192

  Warp size:                                     32

  Maximum number of threads per block:           512

  Maximum sizes of each dimension of a block:    512 x 512 x 64

  Maximum sizes of each dimension of a grid:     65535 x 65535 x 1

  Maximum memory pitch:                          262144 bytes

  Texture alignment:                             256 bytes

  Clock rate:                                    1.51 GHz

  Concurrent copy and execution:                 Yes

Test PASSED

Press ENTER to exit...

Everything worked! Does anybody have any insight on what the problem may be? I should also mention that this is a fresh install of openSUSE 10.3

Thanks for you time!

#2

Well I seem to have solved my problem. My issue was in my grub menu.lst. I had the vmalloc=256MB

pci=nommconf in the wrong location and they were not getting passed to the kernel at boot time.

My output from deviceQuery now look like:

# ./deviceQuery 

There are 6 devices supporting CUDA

Device 0: "G92-450"

  Major revision number:                         1

  Minor revision number:                         1

  Total amount of global memory:                 536543232 bytes

  Number of multiprocessors:                     16

  Number of cores:                               128

  Total amount of constant memory:               65536 bytes

  Total amount of shared memory per block:       16384 bytes

  Total number of registers available per block: 8192

  Warp size:                                     32

  Maximum number of threads per block:           512

  Maximum sizes of each dimension of a block:    512 x 512 x 64

  Maximum sizes of each dimension of a grid:     65535 x 65535 x 1

  Maximum memory pitch:                          262144 bytes

  Texture alignment:                             256 bytes

  Clock rate:                                    1.51 GHz

  Concurrent copy and execution:                 Yes

Device 1: "G92-450"

  Major revision number:                         1

  Minor revision number:                         1

  Total amount of global memory:                 536608768 bytes

  Number of multiprocessors:                     16

  Number of cores:                               128

  Total amount of constant memory:               65536 bytes

  Total amount of shared memory per block:       16384 bytes

  Total number of registers available per block: 8192

  Warp size:                                     32

  Maximum number of threads per block:           512

  Maximum sizes of each dimension of a block:    512 x 512 x 64

  Maximum sizes of each dimension of a grid:     65535 x 65535 x 1

  Maximum memory pitch:                          262144 bytes

  Texture alignment:                             256 bytes

  Clock rate:                                    1.51 GHz

  Concurrent copy and execution:                 Yes

Device 2: "Tesla C870"

  Major revision number:                         1

  Minor revision number:                         0

  Total amount of global memory:                 1610350592 bytes

  Number of multiprocessors:                     16

  Number of cores:                               128

  Total amount of constant memory:               65536 bytes

  Total amount of shared memory per block:       16384 bytes

  Total number of registers available per block: 8192

  Warp size:                                     32

  Maximum number of threads per block:           512

  Maximum sizes of each dimension of a block:    512 x 512 x 64

  Maximum sizes of each dimension of a grid:     65535 x 65535 x 1

  Maximum memory pitch:                          262144 bytes

  Texture alignment:                             256 bytes

  Clock rate:                                    1.35 GHz

  Concurrent copy and execution:                 Yes

Device 3: "Tesla C870"

  Major revision number:                         1

  Minor revision number:                         0

  Total amount of global memory:                 1610350592 bytes

  Number of multiprocessors:                     16

  Number of cores:                               128

  Total amount of constant memory:               65536 bytes

  Total amount of shared memory per block:       16384 bytes

  Total number of registers available per block: 8192

  Warp size:                                     32

  Maximum number of threads per block:           512

  Maximum sizes of each dimension of a block:    512 x 512 x 64

  Maximum sizes of each dimension of a grid:     65535 x 65535 x 1

  Maximum memory pitch:                          262144 bytes

  Texture alignment:                             256 bytes

  Clock rate:                                    1.35 GHz

  Concurrent copy and execution:                 Yes

Device 4: "Tesla C870"

  Major revision number:                         1

  Minor revision number:                         0

  Total amount of global memory:                 1610350592 bytes

  Number of multiprocessors:                     16

  Number of cores:                               128

  Total amount of constant memory:               65536 bytes

  Total amount of shared memory per block:       16384 bytes

  Total number of registers available per block: 8192

  Warp size:                                     32

  Maximum number of threads per block:           512

  Maximum sizes of each dimension of a block:    512 x 512 x 64

  Maximum sizes of each dimension of a grid:     65535 x 65535 x 1

  Maximum memory pitch:                          262144 bytes

  Texture alignment:                             256 bytes

  Clock rate:                                    1.35 GHz

  Concurrent copy and execution:                 Yes

Device 5: "Tesla C870"

  Major revision number:                         1

  Minor revision number:                         0

  Total amount of global memory:                 1610350592 bytes

  Number of multiprocessors:                     16

  Number of cores:                               128

  Total amount of constant memory:               65536 bytes

  Total amount of shared memory per block:       16384 bytes

  Total number of registers available per block: 8192

  Warp size:                                     32

  Maximum number of threads per block:           512

  Maximum sizes of each dimension of a block:    512 x 512 x 64

  Maximum sizes of each dimension of a grid:     65535 x 65535 x 1

  Maximum memory pitch:                          262144 bytes

  Texture alignment:                             256 bytes

  Clock rate:                                    1.35 GHz

  Concurrent copy and execution:                 Yes

Test PASSED

Press ENTER to exit...
#3

That is an obscene amount of GPU power for one machine. Glad to see it’s working.

#4

Hi, everyone,

I also installed s870 on openSUSE 10.3 but when I run deviceQuery, it reports that machine has only one c870 (It should show 4 c870).
when I run “lspci | grep 3D”, it shows only one nvidia 3D controller.
What is the problem? Does anybody have any suggestions?
What can I try?
Thanks,