blueprint.lua

-- location of the gpu computing sdk (for common includes, libraries, and such)
-- this is the BASE PATH to the sdk, in other words the directory where
-- the C, doc, OpenCL, and shared directories live
gpu_sdk_path "/opt/NVIDIA_GPU_Computing_SDK"

-- set our default config, the one that's run if we don't specify one on the
-- command line
default "debug"

-- define a debug config, where debugging options are turned on, profiling
-- options are turned off, and optimizing is turned off
config "debug"
    output "build-debug" -- the output name of our executable
    debugging "on"
    defines {"DEBUG"} -- comma separated list of defines, enclosed in curly brackets
    cflags {"-Wall"} -- flags to pass through to the final compiler (gcc)
    libs {"m"} -- link with the math library
    compute_capability "2.0" -- build for CUDA devices of compute capability 2.0+

-- same as above, but define a "profile" config
config "profile"
    output "build-profile"
    debugging "on"
    profiling "on"
    cflags {"-Wall"}
    libs {"m"}
    compute_capability "2.0"

-- same as above, but define a "release" config
config "release"
    output "build-release"
    optimizing "on"
    defines {"RELEASE"}
    cflags {"-Wall"}
    libs {"m"}
    compute_capability "2.0"