llvm-project/polly/test/GPGPU
Tobias Grosser d58acf866a [GPGPU] Ensure arrays where only parts are modified are copied to GPU
To do so we change the way array exents are computed. Instead of the precise
set of memory locations accessed, we now compute the extent as the range between
minimal and maximal address in the first dimension and the full extent defined
by the sizes of the inner array dimensions.

We also move the computation of the may_persist region after the construction
of the arrays, as it relies on array information. Without arrays being
constructed no useful information is computed at all.

llvm-svn: 278212
2016-08-10 10:58:19 +00:00
..
cuda-annotations.ll Add missing 'REQUIRES' line 2016-08-05 07:08:45 +00:00
double-parallel-loop.ll [GPGPU] Use separate basic block for GPU initialization code 2016-08-09 15:35:03 +00:00
host-control-flow.ll [BlockGenerator] Also eliminate dead code not originating from BB 2016-08-09 08:59:05 +00:00
host-statement.ll [GPGPU] Ensure arrays where only parts are modified are copied to GPU 2016-08-10 10:58:19 +00:00
kernel-params-only-some-arrays.ll GPGPU: Mark kernel functions as polly.skip 2016-08-03 12:00:07 +00:00
kernel-params-scop-parameter.ll GPGPU: Mark kernel functions as polly.skip 2016-08-03 12:00:07 +00:00
only-part-of-array-modified.ll [GPGPU] Ensure arrays where only parts are modified are copied to GPU 2016-08-10 10:58:19 +00:00
parametric-loop-bound.ll Add missing REQUIRES line 2016-07-28 07:08:34 +00:00
phi-nodes-in-kernel.ll [GPGPU] Support PHI nodes used in GPU kernel 2016-08-09 15:35:06 +00:00
private-memory.ll GPGPU: Add private memory support 2016-08-04 12:39:03 +00:00
remove-dead-instructions-in-stmt-2.ll [tests] Add two missing 'REQUIRES' lines 2016-08-09 09:11:39 +00:00
remove-dead-instructions-in-stmt.ll [tests] Add two missing 'REQUIRES' lines 2016-08-09 09:11:39 +00:00
run-time-check.ll [GPGPU] Create code to verify run-time conditions 2016-08-08 17:35:55 +00:00
scalar-param-and-value-32-bit.ll [GPGPU] Pass parameters always by using their own type 2016-08-09 07:22:08 +00:00
scalar-param-and-value-use.ll [GPGPU] Support Values referenced from both isl expr and llvm instructions 2016-08-08 19:22:19 +00:00
scalar-parameter-fp128.ll GPGPU: Generate PTX assembly code for the kernel modules 2016-07-22 07:11:12 +00:00
scalar-parameter-i80.ll GPGPU: Generate PTX assembly code for the kernel modules 2016-07-22 07:11:12 +00:00
scalar-parameter-i120.ll GPGPU: Generate PTX assembly code for the kernel modules 2016-07-22 07:11:12 +00:00
scalar-parameter-i128.ll GPGPU: Generate PTX assembly code for the kernel modules 2016-07-22 07:11:12 +00:00
scalar-parameter-i3000.ll GPGPU: Generate PTX assembly code for the kernel modules 2016-07-22 07:11:12 +00:00
scalar-parameter-ppc_fp128.ll GPGPU: Generate PTX assembly code for the kernel modules 2016-07-22 07:11:12 +00:00
scalar-parameter-x86_fp80.ll GPGPU: Generate PTX assembly code for the kernel modules 2016-07-22 07:11:12 +00:00
scalar-parameter.ll GPGPU: Handle scalar array references 2016-08-04 06:55:59 +00:00
scheduler-timeout.ll GPGPU: collect array references 2016-07-18 15:44:32 +00:00
shared-memory-scalar.ll GPGPU: Support scalars that are mapped to shared memory 2016-08-04 13:57:29 +00:00
shared-memory-two-dimensional.ll GPGPU: Sort dimension sizes of multi-dimensional shared memory arrays correctly 2016-08-05 08:27:24 +00:00
shared-memory.ll GPGPU: Add support for shared memory 2016-08-04 12:18:14 +00:00