leavelet/cutlass
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

3.6.0 update (#2005)

Browse Source 
* 3.6.0 update

* doc and swap stuff

---------

Co-authored-by: yuzhai <yuzhai@nvidia.com>
Co-authored-by: Haicheng Wu <haichengw@nvidia.com>
This commit is contained in:
Yujia Zhai
2024-12-24 22:34:40 -08:00
committed by GitHub
parent e1cd8c7866
commit 3d261a5974
258 changed files with 10863 additions and 3883 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
tools/profiler/src/performance_report.cpp
View File

@@ -337,7 +337,15 @@ std::ostream & PerformanceReport::print_csv_header_(
<< ",Bytes"
<< ",Flops"
<< ",Flops/Byte"
<< ",Runtime"
<< ",Runtime";
if (options_.device.devices.size() > 1) {
for (size_t i = 0; i < options_.device.devices.size(); i++) {
out << ",Runtime_" << i;
}
}
out
<< ",GB/s"
<< ",GFLOPs"
;
@@ -376,6 +384,16 @@ std::ostream & PerformanceReport::print_result_csv_(
<< "," << result.flops / result.bytes
<< "," << result.runtime;
if (options_.device.devices.size() > 1) {
if (result.runtime_vector.size() != options_.device.devices.size()) {
throw std::runtime_error("Runtime vector size mismatch");
}
for (const auto runtime : result.runtime_vector) {
out << "," << runtime;
}
}
if (result.good()) {
out