Merge pull request #73 from JuliaGPU/sd/documentation

documentation

Author: SimonDanisch

docs/src/index.md

@@ -67,14 +67,14 @@ you want to test.
 
 You can run the test suite like this:
 
-```@example
+```Julia
 using GPUArrays, GPUArrays.TestSuite
 TestSuite.run_tests(MyGPUArrayType)
 ```
 If you don't want to run the whole suite, you can also run parts of it:
 
 
-```@example
+```Julia
 Typ = JLArray
 GPUArrays.allowslow(false) # fail tests when slow indexing path into Array type is used.
 

src/abstract_gpu_interface.jl

@@ -38,7 +38,7 @@ end
 
 Macro form of `linear_index`, which calls return when out of bounds.
 So it can be used like this:
-    ```
+    ```jldoctest
     function kernel(state, A)
         idx = @linear_index A state
         # from here on it's save to index into A with idx
@@ -61,7 +61,7 @@ end
 """
     cartesianidx(A, statesym = :state)
 
-Like `@linearidx`, but returns an N-dimensional `NTuple{ndim(A), Cuint}` as index
+Like [`@linearidx(A, statesym = :state)`](@ref), but returns an N-dimensional `NTuple{ndim(A), Cuint}` as index
 """
 macro cartesianidx(A, statesym = :state)
     quote
@@ -109,9 +109,9 @@ end
 
 
 """
-    gpu_call(f, A::GPUArray, args::Tuple, configuration = length(A))
+    gpu_call(kernel::Function, A::GPUArray, args::Tuple, configuration = length(A))
 
-Calls function `f` on the GPU.
+Calls function `kernel` on the GPU.
 `A` must be an GPUArray and will help to dispatch to the correct GPU backend
 and supplies queues and contexts.
 Calls the kernel function with `kernel(state, args...)`, where state is dependant on the backend
@@ -123,7 +123,7 @@ Optionally, a launch configuration can be supplied in the following way:
     2) Pass a tuple of integer tuples to define blocks and threads per blocks!
 
 """
-function gpu_call(f, A::GPUArray, args::Tuple, configuration = length(A))
+function gpu_call(kernel, A::GPUArray, args::Tuple, configuration = length(A))
     ITuple = NTuple{N, Integer} where N
     # If is a single integer, we assume it to be the global size / total number of threads one wants to launch
     thread_blocks = if isa(configuration, Integer)
@@ -147,7 +147,7 @@ function gpu_call(f, A::GPUArray, args::Tuple, configuration = length(A))
                 `linear_index` will be inbetween 1:prod((blocks..., threads...))
         """)
     end
-    _gpu_call(f, A, args, thread_blocks)
+    _gpu_call(kernel, A, args, thread_blocks)
 end
 
 # Internal GPU call function, that needs to be overloaded by the backends.