GPGPU: Generate kernel parameter allocation with right size

Before this change we miscounted the number of function parameters.

llvm-svn: 276960

GPGPU: Add basic support for kernel launches

llvm-svn: 276863

GPGPU: Load GPU kernels

We embed the PTX code into the host IR as a global variable and compile it
at run-time into a GPU kernel.

llvm-svn: 276645

GPGPU: Emit data-transfer code

Also factor out getArraySize() to avoid code dupliciation and reorder some
function arguments to indicate the direction into which data is transferred.

llvm-svn: 2766

GPGPU: Emit data-transfer code

Also factor out getArraySize() to avoid code dupliciation and reorder some
function arguments to indicate the direction into which data is transferred.

llvm-svn: 276636

show more ...

GPGPU: Complete code to allocate and free device arrays

At the beginning of each SCoP, we allocate device arrays for all arrays
used on the GPU and we free such arrays after the SCoP has been execut

GPGPU: Complete code to allocate and free device arrays

At the beginning of each SCoP, we allocate device arrays for all arrays
used on the GPU and we free such arrays after the SCoP has been executed.

llvm-svn: 276635

show more ...

GPGPU: initialize GPU context and simplify the corresponding GPURuntime interface.

There is no need to expose the selected device at the moment. We also pass back
pointers as return values, as this

GPGPU: initialize GPU context and simplify the corresponding GPURuntime interface.

There is no need to expose the selected device at the moment. We also pass back
pointers as return values, as this simplifies the interface.

llvm-svn: 276623

show more ...

IslNodeBuilder: Make finalize() virtual

This allows the finalization routine of the IslNodeBuilder to be overwritten
by derived classes. Being here, we also drop the unnecessary 'Scop' postfix
and t

IslNodeBuilder: Make finalize() virtual

This allows the finalization routine of the IslNodeBuilder to be overwritten
by derived classes. Being here, we also drop the unnecessary 'Scop' postfix
and the unnecessary 'Scop' parameter.

llvm-svn: 276622

show more ...

GPGPU: Optimize kernel IR before generating assembly code

We optimize the kernel _after_ dumping the IR we generate to make the IR we
dump easier readable and independent of possible changes in the

GPGPU: Optimize kernel IR before generating assembly code

We optimize the kernel _after_ dumping the IR we generate to make the IR we
dump easier readable and independent of possible changes in the general
purpose LLVM optimizers.

llvm-svn: 276551

show more ...

GPGPU: Verify kernel IR before generating assembly

llvm-svn: 276550

GPGPU: Generate PTX assembly code for the kernel modules

Run the NVPTX backend over the GPUModule IR and write the resulting assembly
code in a string.

To work correctly, it is important to invalid

GPGPU: Generate PTX assembly code for the kernel modules

Run the NVPTX backend over the GPUModule IR and write the resulting assembly
code in a string.

To work correctly, it is important to invalidate analysis results that still
reference the IR in the kernel module. Hence, this change clears all references
to dominators, loop info, and scalar evolution.

Finally, the NVPTX backend has troubles to generate code for various special
floating point types (not surprising), but also for uncommon integer types. This
commit does not resolve these issues, but pulls out problematic test cases into
separate files to XFAIL them individually and resolve them in future (not
immediate) changes one by one.

llvm-svn: 276396

show more ...

GPGPU: generate code for ScopStatements

This change introduces the actual compute code in the GPU kernels. To ensure
all values referenced from the statements in the GPU kernel are indeed available

GPGPU: generate code for ScopStatements

This change introduces the actual compute code in the GPU kernels. To ensure
all values referenced from the statements in the GPU kernel are indeed available
we scan all ScopStmts in the GPU kernel for references to llvm::Values that
are not yet covered by already modeled outer loop iterators, parameters, or
array base pointers and also pass these additional llvm::Values to the
GPU kernel.

For arrays used in the GPU kernel we introduce a new ScopArrayInfo object, which
is referenced by the newly generated access functions within the GPU kernel and
which is used to help with code generation.

llvm-svn: 276270

show more ...

GPGPU: Bail out of scops with hoisted invariant loads

This is currently not supported and will only be added later. Also update the
test cases to ensure no invariant code hoisting is applied.

llvm-

GPGPU: Bail out of scops with hoisted invariant loads

This is currently not supported and will only be added later. Also update the
test cases to ensure no invariant code hoisting is applied.

llvm-svn: 275987

show more ...

GPGPU: Emit in-kernel synchronization statements

We use this opportunity to further classify the different user statements that
can arise and add TODOs for the ones not yet implemented.

llvm-svn: 2

GPGPU: Emit in-kernel synchronization statements

We use this opportunity to further classify the different user statements that
can arise and add TODOs for the ones not yet implemented.

llvm-svn: 275957

show more ...

GPGPU: generate control flow within the kernel

llvm-svn: 275956

GPGPU: add scop parameters to kernel arguments

llvm-svn: 275955

GPGPU: add host iterators to kernel arguments

llvm-svn: 275954

GPGPU: add intrinsic functions to obtain a kernels thread and block ids

llvm-svn: 275953

GPGPU: create kernel function skeleton

Create for each kernel a separate LLVM-IR module containing a single function
marked as kernel function and taking one pointer for each array referenced
by thi

GPGPU: create kernel function skeleton

Create for each kernel a separate LLVM-IR module containing a single function
marked as kernel function and taking one pointer for each array referenced
by this kernel. Add debugging output to verify the kernels are generated
correctly.

llvm-svn: 275952

show more ...

GPGPU: collect array references

Initialize the list of references to a GPU array to ensure that the arrays that
need to be passed to kernel calls are computed correctly. Furthermore, the very
same

GPGPU: collect array references

Initialize the list of references to a GPU array to ensure that the arrays that
need to be passed to kernel calls are computed correctly. Furthermore, the very
same information is also necessary to compute synchronization correctly. As the
functionality to compute these references is already available, what is left for
us to do is only to connect the necessary functionality to compute array
reference information.

llvm-svn: 275798

show more ...

GPGPU: Pull implementation out of class definition

This will allow us to see the full class definition even after we add
non-trivial implementations of the different member functions.

llvm-svn: 275

GPGPU: Pull implementation out of class definition

This will allow us to see the full class definition even after we add
non-trivial implementations of the different member functions.

llvm-svn: 275797

show more ...

GPGPU: Create host control flow

Create LLVM-IR for all host-side control flow of a given GPU AST. We implement
this by introducing a new GPUNodeBuilder class derived from IslNodeBuilder. The
IslNod

GPGPU: Create host control flow

Create LLVM-IR for all host-side control flow of a given GPU AST. We implement
this by introducing a new GPUNodeBuilder class derived from IslNodeBuilder. The
IslNodeBuilder will take care of generating all general-purpose ast nodes, but
we provide our own createUser implementation to handle the different GPU
specific user statements. For now, we just skip any user statement and only
generate a host-code sceleton, but in subsequent commits we will add handling of
normal ScopStmt's performing computations, kernel calls, as well as host-device
data transfers. We will also introduce run-time check generation and LICM in
subsequent commits.

llvm-svn: 275783

show more ...

GPGPU: Format statements scheduled on the host ourselves

Otherwise ppcg would try to call into pet functionality that this not available,
which obviously will cause trouble. As we can easily print t

GPGPU: Format statements scheduled on the host ourselves

Otherwise ppcg would try to call into pet functionality that this not available,
which obviously will cause trouble. As we can easily print these statements
ourselves, we just do so.

llvm-svn: 275579

show more ...

GPGPU: Use schedule whole components for scheduler

This option increases the scalability of the scheduler and allows us to remove
the 'gisting' workaround we introduced in r275565 to handle a more c

GPGPU: Use schedule whole components for scheduler

This option increases the scalability of the scheduler and allows us to remove
the 'gisting' workaround we introduced in r275565 to handle a more complicated
test case. Another benefit of using this option is also that the generated
code looks a lot more streamlined.

Thanks to Sven Verdoolaege for reminding me of this option.

llvm-svn: 275573

show more ...

GPGPU: Drop domain constraints from flow dependences

This works around a shortcoming of the isl scheduler, which even for some
smaller test cases does not terminate in case domain constraints are pa

GPGPU: Drop domain constraints from flow dependences

This works around a shortcoming of the isl scheduler, which even for some
smaller test cases does not terminate in case domain constraints are part
of the flow dependences.

llvm-svn: 275565

show more ...

GPGPU: Add memory reference tag ids to tagged accesses

It seems we forgot to actually add the memory access ids to the tagged accesses,
but instead just tagged the accesses with empty isl_ids. This

GPGPU: Add memory reference tag ids to tagged accesses

It seems we forgot to actually add the memory access ids to the tagged accesses,
but instead just tagged the accesses with empty isl_ids. This issue was found
by inspection and without code generation it is difficult to test just by
itself. We fix it for now without test case and expect our code generation
tests to cover this later on.

llvm-svn: 275557

show more ...