Hi. I am working with CC on linux (Centos 7), and I am able to build the Makefile using cmake (2.8.12) and compile (gcc) without a problem. However, CC does not take advantage of my multiple CPUs -- all processes max out a single processor at a time (so there are 31 other CPUs sitting idle). As I understand it, there are "debug" and "release" modes, and I guess that I am running in "debug" mode by default. My problem is that I am seeing where to set the build type or mode. Is this an option in cmake? Or is it a flag invoked at compilation? Or is it a mode at launch? I have looked at the compilation instructions, and I may be staring right at the answer, but I am just not seeing it. Or, perhaps, I'm just misunderstanding something. In any event, thanks in advance for the clarification.
Paco
Compiling for linux -- release vs debug mode?
Re: Compiling for linux -- release vs debug mode?
Hi,
I don't play a lot with Linux but on my side gcc compiles the project in release mode by default.
If it was debug mode, the window title would be something like 'CloudCompare v2.6.X [64 bits] [DEBUG]'.
And beware that not all algorithms are running in parallel mode. Which one are you using?
I don't play a lot with Linux but on my side gcc compiles the project in release mode by default.
If it was debug mode, the window title would be something like 'CloudCompare v2.6.X [64 bits] [DEBUG]'.
And beware that not all algorithms are running in parallel mode. Which one are you using?
Daniel, CloudCompare admin
Re: Compiling for linux -- release vs debug mode?
Thanks, Daniel.
I appreciate the reply and the clarification. It looks like I was mistaken, as some functions (e.g., computing normals) are running in parallel, so it must be compiling as intended. It looks like I had been using some functions (e.g., subsampling) that are not parallelized, and did not realize that they weren't. No problem.
Thanks for your continued efforts with CloudCompare. My students and I really appreciate it.
Paco
I appreciate the reply and the clarification. It looks like I was mistaken, as some functions (e.g., computing normals) are running in parallel, so it must be compiling as intended. It looks like I had been using some functions (e.g., subsampling) that are not parallelized, and did not realize that they weren't. No problem.
Thanks for your continued efforts with CloudCompare. My students and I really appreciate it.
Paco