Visual Profiler Install on Centos 5 Final Error Message: [1] 7348
I ran deviceQuery and bandwidthTest, both passed showing the Tesla M2050 operational.

While in the correct directory, I entered "computeprof &" to begin the installation process. I received [1] 7348 instead.
I ran deviceQuery and bandwidthTest, both passed showing the Tesla M2050 operational.



While in the correct directory, I entered "computeprof &" to begin the installation process. I received [1] 7348 instead.

#1
Posted 01/06/2012 10:14 PM   
Type "[font="Courier New"]fg[/font]" on the same console (window) where you started the profiler to bring it to the foreground.
Or remove the ampersand ("&") from the command the next time you launch it, so it doesn't start in the background.

BTW., this is basic unix shell use, not specific to CUDA. Who told you to add the ampersand?
Type "fg" on the same console (window) where you started the profiler to bring it to the foreground.

Or remove the ampersand ("&") from the command the next time you launch it, so it doesn't start in the background.



BTW., this is basic unix shell use, not specific to CUDA. Who told you to add the ampersand?

Always check return codes of CUDA calls for errors. Do not use __syncthreads() in conditional code unless the condition is guaranteed to evaluate identically for all threads of each block. Run your program under cuda-memcheck to detect stray memory accesses. If your kernel dies for larger problem sizes, it might exceed the runtime limit and trigger the watchdog timer.

#2
Posted 01/07/2012 12:13 AM   
[quote name='tera' date='06 January 2012 - 05:13 PM' timestamp='1325895239' post='1352613']
Type "[font="Courier New"]fg[/font]" on the same console (window) where you started the profiler to bring it to the foreground.
Or remove the ampersand ("&") from the command the next time you launch it, so it doesn't start in the background.

BTW., this is basic unix shell use, not specific to CUDA. Who told you to add the ampersand?
[/quote]


The Getting Started with Visual Profiler Guide (in the linux section) directs one to use the ampersand. I thought this curious as well and tried it also w/o the ampersand before I posted. I am also getting an error message on reboot that the Xconfig file needs to be fixed. I am using Centos 5 final which should be binary compatible with Red Hat version 5. Could it be an issue with Centos 5? I reinstalled the driver (latest version) with no avail. However, I ran the deviceQuery and bandwidthTest and both passed, identifying the Tesla M2050 card as operational.
[quote name='tera' date='06 January 2012 - 05:13 PM' timestamp='1325895239' post='1352613']

Type "fg" on the same console (window) where you started the profiler to bring it to the foreground.

Or remove the ampersand ("&") from the command the next time you launch it, so it doesn't start in the background.



BTW., this is basic unix shell use, not specific to CUDA. Who told you to add the ampersand?







The Getting Started with Visual Profiler Guide (in the linux section) directs one to use the ampersand. I thought this curious as well and tried it also w/o the ampersand before I posted. I am also getting an error message on reboot that the Xconfig file needs to be fixed. I am using Centos 5 final which should be binary compatible with Red Hat version 5. Could it be an issue with Centos 5? I reinstalled the driver (latest version) with no avail. However, I ran the deviceQuery and bandwidthTest and both passed, identifying the Tesla M2050 card as operational.

#3
Posted 01/09/2012 03:35 PM   
Thank you!
Thank you!

#4
Posted 01/09/2012 04:38 PM   
Scroll To Top