mirror of
https://github.com/NVIDIA/nvbench.git
synced 2026-04-27 10:11:20 +00:00
Squashed commit of the following:
commit4b309e6ad8Author: Allison Piper <alliepiper16@gmail.com> Date: Sat Apr 6 13:19:14 2024 +0000 Minor cleanups commit476ed2ceaeAuthor: Allison Piper <alliepiper16@gmail.com> Date: Sat Apr 6 12:53:37 2024 +0000 WAR compiler ice in nlohmann json. Only seeing this on GCC 9 + CTK 11.1. Seems to be having trouble with the `[[no_unique_address]]` optimization. commita9bf1d3e42Author: Allison Piper <alliepiper16@gmail.com> Date: Sat Apr 6 00:24:47 2024 +0000 Bump nlohmann json. commit80980fe373Author: Allison Piper <alliepiper16@gmail.com> Date: Sat Apr 6 00:22:07 2024 +0000 Fix llvm filesystem support commitf6099e6311Author: Allison Piper <alliepiper16@gmail.com> Date: Fri Apr 5 23:18:44 2024 +0000 Drop MSVC 2017 testing. commit5ae50a8ef5Author: Allison Piper <alliepiper16@gmail.com> Date: Fri Apr 5 23:02:32 2024 +0000 Add mroe missing headers. commitb2a9ae04d9Author: Allison Piper <alliepiper16@gmail.com> Date: Fri Apr 5 22:37:56 2024 +0000 Remove old CUDA+MSVC builds and make windows build-only. commit5b18c26a28Author: Allison Piper <alliepiper16@gmail.com> Date: Fri Apr 5 22:37:07 2024 +0000 Fix header for std::min/max. Why do I always think it's utility instead of algorithm.... commit6a409efa2dAuthor: Allison Piper <alliepiper16@gmail.com> Date: Fri Apr 5 22:18:18 2024 +0000 Temporarily disable CUPTI on all windows builds. commitf432f88866Author: Allison Piper <alliepiper16@gmail.com> Date: Fri Apr 5 21:42:52 2024 +0000 Fix warnings on MSVC. commit829787649bAuthor: Allison Piper <alliepiper16@gmail.com> Date: Fri Apr 5 21:03:16 2024 +0000 More flailing about in powershell. commit21742e6beaAuthor: Allison Piper <alliepiper16@gmail.com> Date: Fri Apr 5 20:36:08 2024 +0000 Cleanup filesystem header handling. commitde3d202635Author: Allison Piper <alliepiper16@gmail.com> Date: Fri Apr 5 20:09:00 2024 +0000 Windows CI debugging. commita4151667ffAuthor: Allison Piper <alliepiper16@gmail.com> Date: Fri Apr 5 19:45:40 2024 +0000 Quotation mark madness commitdd04f3befeAuthor: Allison Piper <alliepiper16@gmail.com> Date: Fri Apr 5 19:27:27 2024 +0000 Temporarily disable NVML on windows CI until new containers are ready. commitf3952848c4Author: Allison Piper <alliepiper16@gmail.com> Date: Fri Apr 5 19:25:22 2024 +0000 WAR issues on gcc-7. commit198986875eAuthor: Allison Piper <alliepiper16@gmail.com> Date: Fri Apr 5 19:25:04 2024 +0000 More matrix/devcontainer updates. commitb9712f8696Author: Allison Piper <alliepiper16@gmail.com> Date: Fri Apr 5 18:30:35 2024 +0000 Fix windows build scripts. commit943f268280Author: Allison Piper <alliepiper16@gmail.com> Date: Fri Apr 5 18:18:33 2024 +0000 Fix warnings with clang host compiler. commit7063e1d60aAuthor: Allison Piper <alliepiper16@gmail.com> Date: Fri Apr 5 18:14:28 2024 +0000 More devcontainer hijinks. commit06532fde81Author: Allison Piper <alliepiper16@gmail.com> Date: Fri Apr 5 17:51:25 2024 +0000 More matrix updates. commit78a265ea55Author: Allison Piper <alliepiper16@gmail.com> Date: Fri Apr 5 17:34:00 2024 +0000 Support CLI CMake options for windows ci scripts. commit670895c867Author: Allison Piper <alliepiper16@gmail.com> Date: Fri Apr 5 17:31:59 2024 +0000 Add missing devcontainers. commitb121823e74Author: Allison Piper <alliepiper16@gmail.com> Date: Fri Apr 5 17:22:54 2024 +0000 Build for `all-major` architectures in presets. We can get away with this because we require CMake 3.23.1. This was added in 3.23. commitfccfd44685Author: Allison Piper <alliepiper16@gmail.com> Date: Fri Apr 5 17:22:08 2024 +0000 Update matrix file. commite7d43ba90eAuthor: Allison Piper <alliepiper16@gmail.com> Date: Fri Apr 5 16:23:48 2024 +0000 Consolidate build/test jobs. commitc4044056ecAuthor: Allison Piper <alliepiper16@gmail.com> Date: Fri Apr 5 16:04:11 2024 +0000 Add missing build script.
This commit is contained in:
@@ -56,8 +56,8 @@ NVBENCH_BENCH(single_float64_axis)
|
||||
void copy_sweep_grid_shape(nvbench::state &state)
|
||||
{
|
||||
// Get current parameters:
|
||||
const int block_size = static_cast<int>(state.get_int64("BlockSize"));
|
||||
const int num_blocks = static_cast<int>(state.get_int64("NumBlocks"));
|
||||
const auto block_size = static_cast<unsigned int>(state.get_int64("BlockSize"));
|
||||
const auto num_blocks = static_cast<unsigned int>(state.get_int64("NumBlocks"));
|
||||
|
||||
// Number of int32s in 256 MiB:
|
||||
const std::size_t num_values = 256 * 1024 * 1024 / sizeof(nvbench::int32_t);
|
||||
@@ -77,6 +77,7 @@ void copy_sweep_grid_shape(nvbench::state &state)
|
||||
num_values,
|
||||
in_ptr = thrust::raw_pointer_cast(in.data()),
|
||||
out_ptr = thrust::raw_pointer_cast(out.data())](nvbench::launch &launch) {
|
||||
(void) num_values; // clang thinks this is unused...
|
||||
nvbench::copy_kernel<<<num_blocks, block_size, 0, launch.get_stream()>>>(
|
||||
in_ptr,
|
||||
out_ptr,
|
||||
@@ -110,6 +111,7 @@ void copy_type_sweep(nvbench::state &state, nvbench::type_list<ValueType>)
|
||||
[num_values,
|
||||
in_ptr = thrust::raw_pointer_cast(in.data()),
|
||||
out_ptr = thrust::raw_pointer_cast(out.data())](nvbench::launch &launch) {
|
||||
(void) num_values; // clang thinks this is unused...
|
||||
nvbench::copy_kernel<<<256, 256, 0, launch.get_stream()>>>(in_ptr,
|
||||
out_ptr,
|
||||
num_values);
|
||||
@@ -156,6 +158,7 @@ void copy_type_conversion_sweep(nvbench::state &state,
|
||||
[num_values,
|
||||
in_ptr = thrust::raw_pointer_cast(in.data()),
|
||||
out_ptr = thrust::raw_pointer_cast(out.data())](nvbench::launch &launch) {
|
||||
(void) num_values; // clang thinks this is unused...
|
||||
nvbench::copy_kernel<<<256, 256, 0, launch.get_stream()>>>(in_ptr,
|
||||
out_ptr,
|
||||
num_values);
|
||||
|
||||
@@ -36,7 +36,7 @@ public:
|
||||
|
||||
protected:
|
||||
// Setup the criterion in the `do_initialize()` method:
|
||||
virtual void do_initialize() override
|
||||
virtual void do_initialize() override
|
||||
{
|
||||
m_num_samples = 0;
|
||||
}
|
||||
@@ -71,6 +71,7 @@ void throughput_bench(nvbench::state &state)
|
||||
state.add_global_memory_writes<nvbench::int32_t>(num_values);
|
||||
|
||||
state.exec(nvbench::exec_tag::no_batch, [&input, &output, num_values](nvbench::launch &launch) {
|
||||
(void) num_values; // clang thinks this is unused...
|
||||
nvbench::copy_kernel<<<256, 256, 0, launch.get_stream()>>>(
|
||||
thrust::raw_pointer_cast(input.data()),
|
||||
thrust::raw_pointer_cast(output.data()),
|
||||
|
||||
@@ -54,6 +54,8 @@ void mod2_inplace(nvbench::state &state)
|
||||
state.exec(nvbench::exec_tag::timer,
|
||||
// Lambda now takes a `timer` argument:
|
||||
[&input, &data, num_values](nvbench::launch &launch, auto &timer) {
|
||||
(void) num_values; // clang thinks this is unused...
|
||||
|
||||
// Reset working data:
|
||||
thrust::copy(thrust::device.on(launch.get_stream()),
|
||||
input.cbegin(),
|
||||
|
||||
@@ -52,6 +52,7 @@ void stream_bench(nvbench::state &state)
|
||||
state.set_cuda_stream(nvbench::make_cuda_stream_view(default_stream));
|
||||
|
||||
state.exec([&input, &output, num_values](nvbench::launch &) {
|
||||
(void) num_values; // clang thinks this is unused...
|
||||
copy(thrust::raw_pointer_cast(input.data()),
|
||||
thrust::raw_pointer_cast(output.data()),
|
||||
num_values);
|
||||
|
||||
@@ -51,6 +51,7 @@ void throughput_bench(nvbench::state &state)
|
||||
state.add_global_memory_writes<nvbench::int32_t>(num_values);
|
||||
|
||||
state.exec([&input, &output, num_values](nvbench::launch &launch) {
|
||||
(void) num_values; // clang thinks this is unused...
|
||||
nvbench::copy_kernel<<<256, 256, 0, launch.get_stream()>>>(
|
||||
thrust::raw_pointer_cast(input.data()),
|
||||
thrust::raw_pointer_cast(output.data()),
|
||||
|
||||
Reference in New Issue
Block a user